15 Abstracts AHPBA

SUPPLEMENT 1
FEBRUARY 2015
R
12 CTO
20 FA
CT 39
PA 1.9
IM
HPB
VOLUME 17
VOLUME 17
SUPPLEMENT 1
FEBRUARY 2015
THE OFFICIAL JOURNAL OF THE
International Hepato-Pancreato-Biliary Association

Americas Hepato-Pancreato-Biliary Association
European-African Hepato-Pancreato-Biliary Association
PAGES 186
IN THIS ISSUE
Abstracts of the Fifteenth Annual Americas Hepato-Pancreato-Biliary Congress

1115 March 2015
Miami Beach, FL, USA
www.hpbjournal.com
HPB Editors and Editorial Board

The Official Journal of the
International Hepato-Pancreato-Biliary Association
Americas Hepato-Pancreato-Biliary Association
European-African Hepato-Pancreato-Biliary Association
Australian and New Zealand Hepatic, Pancreatic & Biliary Association
Editor-in-Chief
O. J. Garden, UK
Editorial Board
P. Allen, USA
C. Bassi, Italy
K. Behrns, USA
J. Belghiti, France
G. Belli, Italy
M. G. H. Besselink, the Netherlands
A. Biankin, UK
M. Buechler, Germany
I. Di Carlo, Italy
R. Carter, UK
W. Chapman, USA
R. Chari, USA
M.-F. Chen, Taiwan
M. Choti, USA
J. Christein, USA
C. Christophi, Australia
Editors
M. P. Callery, USA
S. J. Connor, New Zealand
S. J. Wigmore, UK
Assistant Editors
E. M. Harrison, UK
R. M. Minter, USA
Honorary Regional Editors

Y. Nimura, Japan
H. Obertop, the Netherlands
S. Strasberg, USA
P.-A. Clavien, Switzerland

M. Cox, Australia
J. E. Cunha, Brazil
M. de Boer, the Netherlands
C. Dejong, the Netherlands
E. de Santibanes, Argentina
C. Dervenis, Greece
J. Espat, USA
O. Farges, France
L. Fernndez-Cruz, Spain
A. Frilling, UK
S. Gallinger, Canada
H. Gooszen, the Netherlands
D. Gouma, the Netherlands
E. Hagopian, USA
S. Helton, USA
A. Hemming, USA
T. J. Howard, USA
J. Izbicki, Germany
P. Jagannath, India
W. Jarnagin, USA
N. Kokudo, Japan
M. Krawczyk, Poland
P. Lai, China
J. Lendoire, Argentina
D. Mahvi, USA
M. Makuuchi, Japan
R. Martin, USA
J. McCall, New Zealand
M. Mercado, Mexico
D. Nagorney, USA
R. Padbury, Australia
T. Pappas, USA
R. Parks, UK
T. Pawlik, USA
W. Pinson, USA
H. Pitt, USA
M. Rees, UK
M. Ryska, Czech Republic
R. Schulick, USA
M. Selzner, Canada
M. Smith, South Africa
T. Takada, Japan
J. Tseng, USA
T. van Gulik, the Netherlands
B. Visser, USA
C. Vollmer, USA
J. Windsor, New Zealand
Aims and Scope. HPB is an international forum for clinical, scientific and educational communication. Twelve issues a year bring the reader leading articles, expert
reviews, original articles, images, editorials, and reader correspondence encompassing all aspects of benign and malignant hepatobiliary disease and its management.
HPB features relevant aspects of clinical and translational research and practice.
Specific areas of interest include HPB diseases encountered globally by clinical practitioners in this specialist field of gastrointestinal surgery. The journal addresses the
challenges faced in the management of cancer involving the liver, biliary system and
pancreas. While surgical oncology represents a large part of HPB practice, submission of manuscripts relating to liver and pancreas transplantation, the treatment of
benign conditions such as acute and chronic pancreatitis, and those relating to hepatobiliary infection and inflammation are also welcomed. There will be a focus on
developing a multidisciplinary approach to diagnosis and treatment with endoscopic
and laparoscopic approaches, radiological interventions and surgical techniques
being strongly represented. HPB aims to help its readers surgeons, physicians,
radiologists and basic scientists to develop their knowledge and practice. HPB
will be of interest to specialists involved in the management of hepatobiliary
and pancreatic disease however will also inform those working in related fields.
ISSN 1365-182X (Print); ISSN 1477-2574 (Online)
HPB is the official journal of the Australian and New Zealand Hepatic, Pancreatic
& Biliary Association (ANZHPBA)
Disclaimer. The Publisher, the International Hepato-Pancreato-Biliary Association and Editors cannot be held responsible for errors or any consequences arising
from the use of information contained in this journal; the views and opinions
expressed do not necessarily reflect those of the Publisher, the International
Hepato-Pancreato-Biliary Association and Editors, neither does the publication of
advertisements constitute any endorsement by the Publisher, the International
Hepato-Pancreato-Biliary Association and Editors of the products advertised.
Copyright 2015 International Hepato-Pancreato-Biliary Association. All rights
reserved. No part of this publication may be reproduced, stored or transmitted in
any form or by any means without the prior permission in writing from the copyright holder. Authorization to copy items for internal and personal use is granted
by the copyright holder for libraries and other users registered with their local
Reproduction Rights Organisation (RRO), e.g. Copyright Clearance Center (CCC),
222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright.com), provided the
appropriate fee is paid directly to the RRO. This consent does not extend to other
kinds of copying such as copying for general distribution, for advertising and promotional purposes, for creating new collective works or for resale. Special requests
should be addressed to: permissions@wiley.com.
Information for subscribers. HPB is published in 12 issues per year. Institutional subscription prices for 2015 are:
Institutional Print + Online: US$1838 (US and Rest of World), 941 (UK), #1185
(Europe). Prices are exclusive of tax. Asia-Pacific GST, Canadian GST/HST and
European VAT will be applied at the appropriate rates. For more information on
current tax rates, please go to www.wileyonlinelibrary.com/tax-vat. The price
includes online access to the current and all online back files to January 1st 2011,
where available. For other pricing options, including access information and terms
and conditions, please visit www.wileyonlinelibrary.com/access.
Delivery Terms and Legal Title. Where the subscription price includes print
issues and delivery is to the recipients address, delivery terms are Delivered at Place
(DAP); the recipient is responsible for paying any import duty or taxes. Title to all
issues transfers FOB our shipping point, freight prepaid. We will endeavour to fulfil
claims for missing or damaged copies within six months of publication, within our
reasonable discretion and subject to availability.
Back issues. Single issues from current and recent volumes are available at the
current single issue price from cs-journals@wiley.com. Earlier issues may be obtained
from Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA.
Tel: +1 518 537 4700, Fax: +1 518 537 5899, Email: psc@periodicals.com.
HPB (ISSN 1365-182X), is published monthly. US mailing agent: Mercury Media
Processing, LLC, 1850 Elizabeth Avenue, Suite #C, Rahway, NJ 07065 USA.
Periodical postage paid at Rahway, NJ, USA.
Postmaster: Send all address changes to HPB, JJohn Wiley & Sons Inc., C/O The
Sheridan Press, PO Box 465, Hanover, PA 17331.
HPB is published by John Wiley & Sons Ltd, 9600 Garsington Road, Oxford OX4
2DQ, UK. Tel: +44 1865 776868, Fax: +44 1865 714591.
Journal Customer Services. For ordering information, claims and any enquiry
concerning your journal subscription please go to www.wileycustomerhelp.com/ask

or contact your nearest office:
Americas: Email: cs-journals@wiley.com; Tel: +1 781 388 8598 or +1 800 835 6770
(Toll free in the USA & Canada).
Europe, Middle East and Africa: Email: cs-journals@wiley.com; Tel:
+44(0) 1865 778 315.
Asia Pacific: Email: cs-journals@wiley.com; Tel: +65 6511 8000.
Japan: For Japanese speaking support, Email: cs-japan@wiley.com; Tel: +65 6511
8010 or Tel (toll-free): 005 316 50 480.
Visit our Online Customer Help available in 7 languages at www.
wileycustomerhelp.com.
Production Editor: Valerie Oliveiro (email: hpb@wiley.com).
For submission instructions, subscription and all other information visit: www.
hpbjournal.com/ask.
View this journal online at wileyonlinelibrary.com/journal/hpb.
Wileys Corporate Citizenship initiative seeks to address the environmental, social,
economic, and ethical challenges faced in our business and which are important to
our diverse stakeholder groups. Since launching the initiative, we have focused on
sharing our content with those in need, enhancing community philanthropy, reducing our carbon impact, creating global guidelines and best practices for paper use,
establishing a vendor code of ethics, and engaging our colleagues and other stakeholders in our efforts. Follow our progress at www.wiley.com/go/citizenship.
Printed in Singapore by C.O.S. Printers Pte Ltd.
Abstracting and Indexing Services. The Journal is indexed by EMBASE/
Excerpta Media, Index Medicus/MEDLINE, SIIC database, Science Citation Index

Expanded (Thomson ISI), Web of Science (Thomson ISI).
HPB accepts articles for Open Access publication. Please visit http://olabout.wiley.
com/WileyCDA/Section/id-406241.html for further information about OnlineOpen.
HPB
Volume 17, Supplement 1, February 2015
Abstracts of the Fifteenth Annual Americas Hepato-Pancreato-Biliary Congress

1115 March 2015
Miami Beach, FL, USA
Disclaimer
This abstract book has been produced using authors-supplied copy. Editing has been restricted to some corrections of spelling and
style where appropriate. No responsibility is assumed for any claims, instructions, methods or drug dosages contained in the abstracts:
it is recommended that these are verified independently.
HPB
CONTENTS
Abstracts
Author Index
Volume 17, Supplement 1, February 2015

181
8286
ABSTRACTS
FRIDAY, MARCH 13, 2015,
7:00AM8:00AM
PRESIDENTIAL PLENARY
PP.01 DEFINING THE PRACTICE OF
PANCREATODUODENECTOMY AROUND
THE WORLD
M. T. McMillan1, M. H. Sprys1, G. Malleo2, C. Bassi2,
C. M. Vollmer1
1
University Of Pennsylvania Perelman School Of Medicine,
Philadelphia, PA; 2University Of Verona, Verona, VERONA
Introduction: Pancreatoduodenectomy (PD) is a technically challenging operation characterized by numerous
management decisions. We hypothesize that there is significant variation in the contemporary global practice of
pancreatoduodenectomy.
Methods: A survey with native-language translation was
distributed to members of six international GI surgical societies (including AHPBA and IHPBA). Practice patterns and
surgical decision-making for PD were assessed. To evaluate
global variance, regions were clustered: North America,
South America/Mexico, Asia/Australia, and Europe/Africa.
Results: Surveys were completed by 864 surgeons, representing six continents and seven languages. Median age and
experience were 46 and 14 years. Surgeons performed a
median of 13 PDs in the past calendar year, and reported a
median career experience of 90, with only 54% surpassing
the published learning curve (>60). Significant regional differences were observed for annual and career PD volumes
(P < 0.001). Only 4% of respondents practice pancreas
surgery exclusively, but 57% perform HPB surgery
only greatest in Asia/Australia (70%, P < 0.0001). Worldwide, the preferred form of anastomotic reconstruction was
pancreaticojejunostomy (88%); however, this choice was
more common in North America compared with Europe/
Africa (97 vs. 81%, P < 0.0001). Regional variability was
also evident in terms of suture technique, stent use/type,
drain use/type/number, as well as the use of octreotide, sealants, and autologous patches (P < 0.02 for all). In particular,
there were stark differences in practice between North and
South American surgeons (Table).
Conclusion: Globally, there is significant variability in the
practice of pancreatoduodenectomy. Many of these choices
contrast with established randomized evidence and may contribute to variance in outcomes.
HPB 2015, 17 (Suppl. 1), 181
PP.02 COLORECTAL LIVER

METASTASES: DISAPPEARING LESIONS
IN THE ERA OF EOVIST
HEPATOBILIARY MAGNETIC
RESONANCE IMAGING
J. W. Owen, K. J. Fowler, M. B. Doyle, N. E. Saad,
D. C. Linehan, W. C. Chapman
Washington University In Saint Louis, Saint Louis, MO
Purpose: Prior studies show that the disappearance of
colorectal liver metastases on CT and PET does not imply
complete pathological response. Eovist MRI is known to be
the most sensitive imaging modality for liver lesions. Our
study evaluates Disappearing Lesions on Eovist MRI to
determine if disappearance predicts pathologic response.
Materials and Methods: Retrospective review of hepatic
resections for colorectal metastases between 01/2008 and
01/2014 was performed. Patients with pre-neoadjuvant
imaging and pre-operative Eovist MR were included. Disappearing lesions were lesions on baseline imaging that
were not identifiable on pre-operative Eovist MRI. Complete
pathologic response was defined on pathology or by no
re-development of a lesion with 1 year follow-up if the site
was not resected. Persistent disease was defined as viable
lesions on pathology, or imaging evidence of recurrence
within 1 year.
Results: Baseline imaging and Eovist MRI was available for
24 patients. In 24 patients, 198 colorectal metastases were
identified on baseline imaging. On pre-operative Eovist MR
82 of the 194 lesions (42%) were disappearing lesions. At
2015 The Authors
HPB 2015 Americas Hepato-Pancreato-Biliary Association
Abstracts
surgical pathology or 1 year follow-up imaging, 37 of 82

lesions (45%) demonstrated viable tumor (16) or recurrence
(21). Thirty-five of 82 lesions (43%) were nonviable at
pathology (9) or without evidence of recurrence at 1 year
(26). 10 lesions were resected, but difficult to correlate with
surgical pathology retrospectively and were indeterminate.
Conclusion: Despite the sensitivity of Eovist MR imaging,
disappearing lesions should be considered for surgical resection as 45% of disappearing lesions were viable on surgical
pathology or recurred when not resected.
those with MVI and CD44 or CD133 is 30% and 27.7%;

respectively (p < 0.001). Thirteen patients developed tumor
recurrence. On multivariable analysis, CD44 and CD133
expression in combination with MVI, were found to be
independent predictors of tumor recurrence and overall
patient survival (recurrence p < 0.003, OR = 8.05; p = 0.001,
OR = 9.5, survival p = 0.001, HR 3.7, p = 0.004, HR 3.2
respectively).
Conclusion: The expression of CD44 or CD133, combined
with MVI, were independent factors associated with
poor outcomes in patients undergoing transplantation for
HCC.
PP.03 CANCER STEM CELL MARKER

EXPRESSION IN COMBINATION WITH
MICROVASCULAR INVASION PREDICT
POOR SURVIVAL IN PATIENTS
UNDERGOING LIVER
TRANSPLANTATION WITH
HEPATOCELLULAR CARCINOMA
V. Vilchez1, L. Turcios 1, Y. Zaytseva2, E. Maynard1,
M. Shah1, M. F. Daily1, C. Tzeng1, D. Davenport1,
A. Castellanos1, S. Krohmer 3, R. Gedaly1
1
Department Of Surgery/Transplant Division University
Of Kentucky, Lexington, KENTUCKY; 2Markey Cancer
Center Core Support, Lexington, KENTUCKY;
3
Department Of Radiology University Of Kentucky,
Lexington, KENTUKCY
Objective: We investigated the expression of Liver Cancer
Stem Cells (LCSC) markers, CD44 and CD133, combined
with the presence of microvascular invasion (MVI); as predictors of outcomes in patients undergoing Liver Transplantation (LT) for HCC.
Methods: Explanted livers from 95 patients with HCC who
underwent LT at the University of Kentucky Transplant
Center were analyzed. The expression of CD44 and CD133
was evaluated using immunofluorescence.
Results: Median age was 56 + 7 years, and 77 (81%) were
male. The most common causes of end-stage liver disease
were hepatitis C (50%) and alcoholic liver disease (41%).
Forty-one patients had lab MELD score >15. Median
follow-up was 64 months. Twenty two (23%) tumors had
MVI. Forty three (45%) tumors were positive for CD44, and
33 (34%) for CD133. Overall 1, 3, and 5 year survival rates
were 86%, 75%, and 64%, respectively. The combination of
MVI and either CD44 or CD133 positivity was associated
with significantly worse overall outcomes. The 5-year survival rate of patients with MVI alone is 51% compared to
2015 The Authors
PP.04 INTRAHEPATIC
CHOLANGIOCARCINOMA AND
GALLBLADDER CANCER:
DISTINGUISHING MOLECULAR
PROFILES TO GUIDE
POTENTIAL THERAPY
M. Potkonjak, J. Miura, K. K. Turaga, F. M. Johnston,
S. Tsai, K. Christians, T. C. Gamblin
Division Of Surgical Oncology, Department Of Surgery,
Medical College Of Wisconsin, Milwaukee, WI
Chemotherapy regimens according to National Comprehensive Cancer Network (NCCN) guidelines for intrahepatic
cholangiocarcinoma (IHC) and gallbladder adenocarcinoma
HPB 2015, 17 (Suppl. 1), 181
Abstracts
(GC) are interchangeable; however, the molecular differences driving tumorigenesis for these cancers remain poorly
defined. The present study utilized biomarker analysis of
actionable targets for IHC and GC to distinguish them and
potentially refine current treatment strategies.
217 IHC and 28 GC specimens referred to Caris Life Sciences between 2009 thru 2012 were evaluated. Specific
testing by immunohistochemical analysis for 17 different
biomarkers was performed. In the collective cohort
(n = 245), actionable targets included: 95% low TS, 82% low
RMM1, and 74% low ERCC1, indicating potential susceptibility to fluoropyrimidines/capecitabine, gemcitabine, and
platinum agents, respectively. Additional non-NCCN compendium targets included TOPO1 (53.3% high, irinotecan),
MGMT (50.3% low, temozolomide), TOP2A (33% high,
anthracyclines), and PGP (30.1% low, taxanes). Subgroup
analysis by tumor origin demonstrated a differential
biomarker expression pattern with a higher frequency of IHC
tumors showing low levels of TS (99% vs. 72%, p < 0.0001),
and RRM1 (85% vs. 64%, p = 0.021) when compared to GC.
Conversely a greater frequency of GC demonstrated high
levels of TOPO1 (76% vs. 50%, p = 0.018) versus IHC,
indicating a potential increased benefit from irinotecan.
Biomarker analysis possesses the capacity to identify additional targets for which established agents are available. Differences in molecular profiles of IHC and GC provide
evidence that the two are distinct diseases and require different treatments.

12:30PM1:00PM
LUNCH VIDEO PRESENTATION
VL.01 TRANS-THORACIC MINIMALLY
INVASIVE SEGMENT 8 LIVER
RESECTION GUIDED BY AUGMENTED
REALITY
J. Hallet1,2,4, L. Soler1,2, M. Diana2, D. Mutter1,2,3,
T. Baumert2, F. Habersetzer2, J. Marescaux1,2, P. Pessaux1,2,3
1
Institut De Recherche Sur Les Cancers De LAppareil
Digestif (IRCAD), Strasbourg, ALSACE; 2Institut
Hospitalier Universitaire De Strasbourg, Universitaire
Strasbourg, Strasbourg, ALSACE; 3Service De Chirurgie
Digestive, Nouvel Hal Civil, Strasbourg, ALSACE;
4
Sunnybrook Health Sciences Centre Odette Cancer
Centre, Toronto, ONTARIO
Background: Liver dome tumors are not traditionally amenable to minimally invasive hepatectomy (MIH) due to
superior/central location. In order to increase the number of
lesions amenable to MIH, new approaches are needed. Additional challenges in MIH include loss of 3D visualization and
tactile perception for intraoperative guidance within the
intra-hepatic anatomy.
Methods: This video includes the use of a pre-operative 3D
virtual model and intra-operative augmented reality (AR)
navigation to facilitate trans-thoracic MIH of the liver
dome.
HPB 2015, 17 (Suppl. 1), 181
Results: We present a 52 year-old gentleman with a 3 cm

isolated hepatocellular carcinoma in segment 8. A transthoracic approach was chosen to allow for MIH. The video
begins with presentation of 3D reconstruction and virtual
resection planning. Principles of AR are detailed. The surgical steps include positioning and triangulation of thoracic
ports under AR guidance, followed by trans-diaphragmatic
tumor localization, identification of the phrenotomy site, and
planning of margins using intra-operative ultrasound combined with AR. The parenchyma is transected with bipolar
radiofrequency ablation and ultrasonic scalpel. After closure
of diaphragm, the specimen is extracted through an enlarged
thoracic port, and a chest tube is placed. The procedure was
well tolerated. The chest tube was pulled on day 3 and discharge occurred on day 4.
Conclusion: This video of liver resection for challenging
tumor localization illustrates a different, safe, and valuable
approach to MIH. It highlights how 3D virtual resection
planning and AR can enhance and facilitate complex MIH,
thereby easing the transition into the minimally invasive era
for liver surgery.

3:00PM4:30PM
LONG ORAL A
PANCREAS ONCOLOGY
LO-A.01 UTILITY OF ESTABLISHING
A PANCREAS CANCER SCREENING
PROGRAM WITHIN A HIGH VOLUME
PANCREATIC CANCER PROGRAM
B. A. Krzywda, S. M. Lahiff, D. M. McDowell, B. George,
P. S. Ritch, B. A. Erickson, F. M. Johnston,
K. K. Christians, D. B. Evans, S. Tsai
Background: Approximately 10% of pancreatic cancer
(PC) may be hereditary and screening of high risk individuals
has been recommended. Herein we describe the establishment of a comprehensive multidisciplinary screening
program.
Methods: Screening criteria included the presence of PC
in: 2+ first-degree relatives (FDR), or 3+ any degree relatives
(ADR), or any known hereditary cancer syndrome with
increased PC risk. Lifetime PC risk was calculated using the
CancerGene PancPro software. The clinic provided genetic
counseling and nutrition/wellness education. MRI imaging
was selectively recommended based the lifetime PC risk.
Results: Forty-three patients were screened; 65% were
female and the median age was 54 (IQR:11). Family history
was significant for two FDR in 12 (28%) patients and three
ADR in 11 (26%). Median age of the earliest affected family
member with PC was 59 (IQR:15). Hereditary cancer syndromes were present in 18 (41%) patients: BRCA1(3),
BRCA2(6), MLH1(1), PMS2(1), PALB2(1), ATM(1),
CDKN2A(4) and STK11(1). Median PancPro estimated lifetime risk of screened patients was 7% (IQR:6). Twenty-two
(51%) of 43 patients had a lifetime risk over 10%. Elevated
2015 The Authors
Abstracts
CA19-9 or HbA1c was detected in 2 (5%) and 4 (9%)

patients, respectively. Screening MRI was obtained in 37
(86%) of 43 patients and 10 (27%) of the 37 had pancreatic
cystic lesions. No patient has undergone surgical resection of
a pancreatic lesion.
Conclusions: Initiation of a high risk PC screening clinic
identifies patients with radiographic or biochemical abnormalities for which surveillance is necessary. Guidelines for
the frequency of surveillance and indications for surgery are
needed.
LO-A.02 HAS SURVIVAL IMPROVED

FOLLOWING RESECTION FOR
PANCREATIC ADENOCARCINOMA?
A. S. Rosemurgy, R. Klein, C. Ryan, P. Sukharmwala,
B. Sadowitz, K. Luberice, S. B. Ross
Florida Hospital Tampa, Tampa, FL
Introduction: Billions of dollars have been spent on the
research and treatment of pancreatic cancer. This study was
undertaken to determine if survival after resection of pancreatic adenocarcinoma has been extended over the past two
decades.
Methods: The SEER database was queried for patients who
underwent pancreatectomy for pancreatic adenocarcinoma
from 1992 through 2010. AJCC Stage and survival were
determined for each patient. Data were analyzed using
Mantel-Cox test and linear regression. Significance was
accepted at p < 0.05.
Results: 15,604 patients underwent pancreatectomy from
1992 through 2010. Survival improved from 1992 through
2010 (p < 0.0001), as denoted in Figure 1 with the patients
divided into three cohorts for illustrative purposes (199297,
N = 1,846; 19982003, N = 4,528; 200410, N = 9,230).
Similarly, median survival increased 1992 through 2010 (14
vs. 15 vs. 18 months for the cohorts, p < 0.0001). However,
5-year survival rates did not change 1992 through 2010
(14.4% vs. 15.2% vs. 17.0% for the cohorts; p = 0.07). More
patients (p = 0.007) and relatively more patients (p = 0.004)
underwent resections of Stage I and Stage II cancers
2004 through 2010 with commensurately smaller tumors
(p = 0.01).
Conclusions: From 1992 through 2010, progressively more
patients underwent pancreatectomy for pancreatic adenocarcinoma with progressively smaller tumors and earlier stages.
These patients lived more years (e.g., improved survival
curves and median survival) but without improved 5-year
survival, denoting better early and intermediate survival.
Early detection, better perioperative care, more efficacious
noncurative chemotherapy undoubtedly play a role, but
better solutions for long-term survival must be sought.
2015 The Authors

LO-A.03 HIGH-GRADE INTRADUCTAL

PAPILLARY MUCINOUS NEOPLASM IS
NOT MALIGNANCY
N. Rezaee1, J. He1, B. Salman1, R. H. Hruban2,3,
J. L. Cameron1, N. Ahuja1,2, A. Lennon1,4, M. J. Weiss1,2,
L. D. Wood3, C. L. Wolfgang1,2,3
1
Department Of Surgery, Johns Hopkins Medical
Institutions, Baltimore, MD; 2Department Of Oncology,
Johns Hopkins Medical Institutions, Baltimore, MD;
3
Department Of Pathology, Johns Hopkins Medical
Institutions, Baltimore, MD; 4Department Of
Gastroenterology, Johns Hopkins Medical Institutions,
Baltimore, MD
Background: Since identification of intraductal papillary
mucinous neoplasm (IPMN) in 1996, high-grade dysplasia
and IPMN-associated invasive carcinoma was used frequently under the umbrella term malignancy. We aimed to
compare the pathological features and survival outcomes of
high-grade IPMN to invasive carcinoma.
Patients and Methods: From 1996 to 2013 data of 616
patients who underwent pancreatic resection for an IPMN
were reviewed. IPMNs were classified as low/intermediate-,
high-grade dysplasia (HGD), and invasive carcinoma.
Results: A total of 293 (48%) patients diagnosed with low/
intermediate-grade dysplasia, 140 (23%) with HGD, and 183
(30%) with invasive carcinoma. Actual 5-year survival was
55% for the entire cohort. The median overall survival was
94 months for HGD, which was similar to low/intermediategrade IPMN (118 months, p = 0.07), and superior to invasive
carcinoma (29 months, p < 0.001) (figure). Invasive carcinoma was associated with regional lymph node metastasis in
34%, perineural invasion in 38%, and vascular invasion in
38%. In contrast no lymph node metastasis, perineural or
vascular invasion was observed after resection of HGD.
Compared to invasive carcinoma, HGD was associated with
a lower rate of positive margin (38% vs. 24%, p = 0.007).
Among patients who had more than 6 months follow-up, the
recurrence rate after resection of HGD (16%) was similar to
low/intermediate dysplasia (19%, p = 0.50); and was lower
compared to invasive IPMN (29%, p = 0.03).
Conclusion: IPMN with high-grade dysplasia has a
favorable survival outcome and a lower rate of recurrence
after resection compared to IPMN-associated invasive carcinoma, and thus should not be considered a malignant entity.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
IRE near the portal vein, and plastic stenting should be considered when performing IRE near the common bile duct.
IRE is a potentially crucial tool in the arsenal of surgeons
treating otherwise inoperable pancreatic cancer.
LO-A.04 IRREVERSIBLE
ELECTROPORATION (NANOKNIFE) FOR
PANCREATIC CANCER: A SINGLE
INSTITUTION SERIES OF 50
CONSECUTIVE PATIENTS
K. Mahendraraj, I. Epelboym, B. Schrope, J. A. Chabot,
M. D. Kluger
Department Of Surgery, College Of Physicians And
Surgeons, Columbia University Medical Center, New York,
NEW YORK
Introduction: The NanoKnife irreversible electroporation
system (IRE) uses electrical energy to destroy neoplastic
tissue invading surrounding neurovascular structures. Large
scale IRE for pancreatic cancer has yet to be reported. This
study examines a large cohort of IRE-treated pancreatic
cancer patients to evaluate the safety of this novel surgical
approach.
Methods: Data was abstracted on all T3 and T4 pancreatic
cancer patients who underwent IRE at a tertiary hepatobiliary
unit from 20122014. Standard statistical methodology was
used.
Results: 50 consecutive patients were treated with IRE by 3
pancreatic surgeons, with 36(72%) cases performed by a
single surgeon. Mean patient age was 65.8 7.8 years, with
31(62%) male patients. There were 45(90%) adenocarcinoma cases, most commonly involving the pancreatic head
(n = 16;32%) or body (n = 16;32%). IRE was used for
primary local control in 25(50%) cases and margin ablation
in 21(42%). Median survival was 11.8 6.2 months. Median
follow-up was 7.8 9.6 months, with length of stay
7.34 5.6 days and readmission rate of 20%(n = 10). 30- and
90-day complication rates were 36%(n = 18) and 6%(n = 3),
most commonly portal vein thrombosis(n = 4;8%), intraabdominal collection(n = 3;6%), and anemia requiring
transfusion(n = 3;6%). Overall mortality attributable to IRE
was 6%(n = 3). 3 additional mortalities were related to
disease progression.
Conclusions: IRE offers a feasible technique to manage
advanced pancreatic cancer. To reduce morbidity and mortality, anticoagulation should be considered when performing
HPB 2015, 17 (Suppl. 1), 181
LO-A.05 PRETREATMENT SERUM CA

19-9 LEVELS IN PATIENTS WITH
LOCALIZED PANCREATIC
CANCER TREATED WITH
NEOADJUVANT THERAPY
M. Aldakkak, K. K. Christians, A. N. Krepline, B. George,
P. S. Ritch, B. A. Erickson, F. M. Johnston, D. B. Evans,
S. Tsai
Background: Among pancreatic cancer (PC) pts treated
with a surgery-first approach, normal CA19-9 levels have
been associated with improved survival. The impact of
neoadjuvant therapy on this association is unknown.
Methods: Localized PC pts with a CA19-9 level prior to
neoadjuvant therapy were dichotomized into two groups; low
CA19-9 and elevated CA19-9 based on a cutoff of 36 U/mL.
Results: CA19-9 was evaluable in 230 pts prior to any treatment; 57 (25%) were low and 173 (75%) were elevated. The
median CA19-9 level at diagnosis in low and elevated
CA19-9 pts was 14 (IQR:23) and 267 (IQR:594) respectively. Neoadjuvant therapy including successful surgery was
completed in 164 (71%) of the 230 patients; 41 (72%) of 57
low CA19-9 and 123 (71%) of 173 elevated CA19-9 pts
(p = 0.90). Median survival of all 230 pts was 23.8 months;
36.7 months for the 164 pts who completed all therapy
2015 The Authors
Abstracts
including surgery vs. 11.7 months for the 66 pts not resected.
Among the 164 pts who completed all therapy, no difference
in median survival was observed between low CA199 and
elevated CA19-9 pts; 36.7 months vs. 33.1 months, p = 0.89.
Conclusions: An elevated CA19-9 at diagnosis did not
predict a failure to complete neoadjuvant therapy and was not
associated with inferior survival. These data suggest two
cautionary notes: an elevated CA19-9 at diagnosis should not
be considered synonymous with advanced (non-surgical)
disease; and, a low/normal CA19-9 should not be interpreted
as a predictor of favorable outcome (and used to justify a
surgery first strategy).
Conclusion: This study illustrates criteria for the highest

yield of genetic evaluation for high-risk of PDAC. Insurance
coverage for unaffected relatives is lacking. Identification of
a causative mutation in an affected family member allows for
cost-effective targeted testing in at-risk relatives. Individuals
with apparently idiopathic pancreatitis, onset of pancreatitis < 30 years, and those with a family history of pancreatitis
or PDAC are candidates for genetic evaluation.

3:00PM4:30PM
LONG ORAL B LIVER HCC
LO-B.01 POST-EMBOLIZATION
SYNDROME AS AN EARLY PREDICTOR
OF LONG-TERM OUTCOME
AFTER TRANSARTERIAL
CHEMOEMBOLIZATION FOR
M. C. Mason1,2, N. N. Massarweh2,3, A. Salami2,
M. A. Sultenfuss4, D. A. Anaya2,3
1
Michael E DeBakey Department Of Surgery, Baylor
College Of Medicine, Houston, TX; 2VA HSR&D Center
For Innovations In Quality, Effectiveness And Safety,
Houston, TX; 3Michael E DeBakey Department Of Surgery
Division Of Surgical Oncology, Houston, TX;
4
Department Of Radiology, Baylor College Of Medicine,
Houston, TX
LO-A.06 ONE YEAR EXPERIENCE

OF CHARACTERIZATION OF
GENETIC RISK IN A HIGH-RISK
PANCREATIC CLINIC
K. Flores, K. Dinh, E. Rouleau, W. Wassef, J. LaFemina
University Of Massachusetts, Worcester, MA
Background: Despite the established benefit of early detection in pancreatic cancer (PDAC) prognosis, high-risk pancreatic clinics are less common than their breast or colon
counterparts. One of the difficulties of establishing a highrisk clinic is delineating which individuals are high-risk.
Methods: We retrospectively examined patients referred for
genetic counseling for PDAC from January 2009-June 2014.
Patients were referred for a personal and/or family history of
PDAC or a potential diagnosis of hereditary pancreatitis
(HP).
Results: 75 patients were referred for genetic counseling; 36
underwent testing. Twelve (33%) mutation carriers were
identified, demonstrating a positivity rate higher than in highrisk clinics for other malignancies. The most common reason
to decline testing was lack of insurance. 11% of patients with
a family history of PDAC were found to carry a mutation.
20% of those a personal history of PDAC were found to carry
a mutation. Ten of 43 patients with a personal history of
chronic pancreatitis were found to carry 1 mutations. Of
these, 8 were heterozygous for CFTR mutations, 1 was
CFTR homozygous, and 1 was homozygous for SPINK1
mutations.
2015 The Authors
Introduction: Transarterial chemoembolization (TACE) is

commonly used in the management of hepatocellular carcinoma (HCC) patients not candidates for curative resection or
as a bridge to transplantation. A common post-TACE complication is post-embolization syndrome (PES). The goal of
this work was to evaluate PES as an early, post-procedural
predictor of treatment outcome.
Methods: A retrospective cohort study of HCC patients who
underwent TACE at a tertiary referral center was performed
(20092014). Patients were excluded if they had other therapies in combination with TACE. Patients were categorized
based on whether they presented with PES, defined as fever
with or without abdominal pain within 2 weeks of TACE.
The primary outcome was overall survival (OS) evaluated
using multivariate Cox regression while adjusting for relevant demographic and clinical characteristics.
Results: Among 144 patients treated with TACE alone
(70.1% 65 years and 44.4% Childs B or C cirrhosis),
36.1% experienced PES. Median follow-up for the cohort
was 11.4 (0.649.9) months. Median and 3-year OS were 16
months and 18.0% in the PES group versus 25 months and
41.0% in the no PES group (log-rank, p = 0.027). After multivariate modeling, patients with PES had a higher risk of
death compared to non-PES patients (Hazard Ratio 1.94;
95% Confidence Interval 1.123.34, P = 0.016).
Conclusion: Patients who suffer from PES have significantly worse overall survival even after adjusting for important factors. Future investigation into the pathophysiologic
mechanism underlying PES may help identify patients at
increased risk of death and better select treatment strategies
for HCC patients with different biologic behavior.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
LO-B.04 SURGICAL RESECTION VERSUS

ABLATION FOR HEPATOCELLULAR
CARCINOMA LESS THAN 3CM: A
POPULATION BASED ANALYSIS
LO-B.03 THE MODIFIED RESPONSE

EVALUATION CRITERIA IN SOLID
TUMORS (MRECIST) PREDICTS
SURVIVAL FOLLOWING
TRANSARTERIAL
CHEMOEMBOLIZATION (TACE) FOR
N. S. Haywood, N. S. Haywood, S. Saddekni,
K. H. Gennaro, D. T. Redden, J. A. White, S. H. Gray,
A. K. Aal, D. E. Eckhoff, D. A. DuBay
University Of Alabama At Birmingham, Birmingham, AL
Background: Transarterial chemoembolization (TACE) is
the most common oncologic treatment administered for the
treatment of hepatocellular carcinoma (HCC) in the US. The
impact of TACE-induced HCC tumor necrosis on patient
survival is poorly defined.
Hypothesis: We hypothesize that survival will be superior
in HCC patients with increased TACE-induced tumor
necrosis.
Methods: All first TACE interventions for HCC performed
at a single institution from 20082013 were retrospectively
reviewed (n = 344, Table 1). HCC tumor response to TACE
was quantified via the modified response evaluation criteria
in solid tumors (mRECIST) criteria. Differences in survival
were compared using the log-rank test. A multivariable
analysis of survival predictors was completed with a Cox
proportional hazard model.
Results: The median survival following TACE treatment
for HCC varied according to the mRECIST response
(p = 0.012), with the longest survival observed in patients
with a complete response and shortest survival in patients
with progressive disease (Table 1) Patients with a complete
response had the lowest frequency of repeat TACE, and
highest probability of receiving a liver transplant (Table 1).
The mRECIST score remained significantly associated with
survival (p = 0.0005) in a multivariable model of survival
predictors controlling for age, gender, race, tumor size and
number, and Childs score.
Conclusions: The mRECIST response to TACE in patients
with HCC was predictive of survival, the need for repeat
TACE, and the probability of receiving a liver transplant.
However, the absolute differences in median survival
between mRECIST categories were not as large as predicted.
HPB 2015, 17 (Suppl. 1), 181
J. T. Miura, R. T. Groeschl, F. M. Johnston, S. Tsai,

K. K. Christians, K. K. Turaga, T. C. Gamblin
Medical College Of Wisconsin, Department Of Surgery,
Milwaukee, WI
Background: Ablation for 3 cm hepatocellular carcinoma
(HCC) has been demonstrated to be an effective treatment
strategy. Whether ablation achieves a similar survival benefit
as compared to surgical resection for early stage HCC
remains ill defined. The present study sought to examine the
outcomes of patients with 3 cm HCC following ablation
versus resection.
Methods: Patients treated by ablation or surgical resection
for 3 cm T1 HCC were identified from the National Cancer
Database (20022011). Cox proportional hazards models
were used to assess overall survival (OS) between treatment
types (ablation vs resection) following adjustment for age,
gender, alpha-fetoprotein (AFP), Charlson Comorbidity
Score, and cirrhosis.
Results: A total of 2,855 patients underwent ablation
(n = 1,984) or resection (n = 871) for solitary HCC 3 cm.
The median age of the collective cohort was 61 (IQR: 5570)
with the majority being male (n = 2,007, 70.1%). Patients
treated with ablation as compared to resection had a higher
frequency in AFP elevation (46.5% vs 39%, p < 0.01) and
presence of cirrhosis (22.2% vs 14.9%, p < 0.01). Unadjusted OS at 3 and 5 years was greater following resection
(67%, 55%) versus ablation (52%, 36%, p < 0.01). In
multivariable models, resection was independently associated with improved OS (HR: 0.65, 95% CI: 0.510.83;
p < 0.01).
Conclusion: While more invasive, resection of HCC 3 cm
results in better long-term survival as compared to ablation.
Treatment strategies for small solitary HCC should emphasize a resection first approach, with ablation being reserved
for patients precluded from surgery.
2015 The Authors

Abstracts
LO-B.05 ROLE OF SURGICAL

RESECTION AND LOCOREGIONAL
THERAPY IN PATIENTS WITH STAGE 3
HEPATOCELLULAR CARCINOMA A
RETROSPECTIVE REVIEW FROM THE
NATIONAL CANCER DATABASE
LO-B.06 CLINICAL OUTCOMES OF

RESECTION VS. ABLATION OF SINGLE
NODULE HEPATOCELLULAR
CARCINOMA IN THE ERA OF ORGAN
SHORTAGE. SHOULD WE EXTEND
THE LIMITS?
R. Seshadri, E. H. Baker, S. W. Ross, M. Templin,

R. Z. Swan, J. B. Martinie, D. A. Iannitti
Carolinas Medical Center, Charlotte, NC
Introduction: In advanced stages, hepatocellular carcinoma

(HCC) is often associated with major vascular involvement
(cava, portal vein). Our aim was to analyze the role of surgical resection (SR) and locoregional therapy (LRT) in these
advanced stage patients to determine if there was a survival
benefit.
Methods: The study is a retrospective analysis from the
Commission on Cancers National Cancer Data Base
(NCDB) between 19982011. 148,882 patients with liver
cancer were identified, of which 126,858 had HCC. Of these,
64,227 patients (19982006) had 5-year survival data available and 16,036 patients had Stage 3A disease based on
AJCC classification. Of these patients, 1,346 had SR, 1,003
had LRT and 11,942 patients had neither intervention.
Kaplan-Meier curves and log rank tests were used for statistical analysis.
Results: 14,291 patients met analysis criteria. Mean age
(years) in the SR, LRT and no intervention group were 62.5
64.3 and 64.2 respectively. Most patients were males in all 3
groups (77.5%, 74.5% and 68.1%). Mean tumor size (cm) in
the 3 groups was 9.8, 6.4 and 8.4 respectively. SR and LRT
were primarily performed in major academic and comprehensive cancer programs compared to community cancer
programs and other centers (SR: 93% vs 7%; LRT: 94.6% vs
5.4%). The median 5 year survival (months) was 26.55 in SR;
16.36 in LRT and 4.8 in no intervention group. (p < 0.0001).
Conclusion: Surgical resection and locoregional therapy
offer a survival benefit in select patients diagnosed with
Stage 3 HCC.
2015 The Authors

O. C. Kutlu1, M. Wachtel2, S. Dissanaike1

TTU HSC Dept Of Surgery, Lubbock, TX; 2TTU HSC Dept
Of Pathology, Lubbock, TX
Introduction: Donor shortage is a limitation for transplantation, leaving resection as the modality of choice for HCC.
Radiofrequency ablation(RFA) is an option for early HCC
patients or for bad surgical candidates. Studies are being
published investigating the use of RFA in larger tumors with
smaller ablative margins. We aimed to compare RFA to other
modalities in a large epidemiological database.
Methods: SEER database was queried for patients at least
20 years age, diagnosed between 20048, no metastatic
disease, with tumors less than 3.5 cm. Table 1 displays data
of importance. Binomial logit regression calculated estimates
and standard errors of odds ratios of death at 1 y and 3 y,
using R3.1.1. Null hypotheses were rejected if P < 0.05/
2 = 0.025.
Results: Figure 1 displays the results of binary logit regression. While survival was similar in RFA and Resection at 1 y,
at 3 y RFA fared worse. Elevated AFP and age were not
significant at 1 y, but were associated with increased odds of
death at 3 y. No assumptions of general linear models were
found to be violated. No influential outlier groups were seen.
Conclusion: Researchers are seeking ways to facilitate
resection and RFA in HCC. Studies have been published
discussing RFA and margins <1 cm for HCC up to 5 cms in
size. RFA probes have been shown to produce a cavity size of
4.5 cms in diameter. Although 3.5 cm was the largest tumor
size in our study, three year outcomes of RFA were significantly inferior to resection and transplantation, reaching
almost twice the odds of death.
HPB 2015, 17 (Suppl. 1), 181
Abstracts

3:00PM4:30PM
LONG ORAL C
TRANSPLANT I
LO-C.01 SPLIT-LIVER
TRANSPLANTATION: REPORT OF
RIGHT AND LEFT GRAFTS OUTCOMES
FROM A MULTICENTER ARGENTINEAN
GROUP
E. Halac1, M. Dip1, F. Alvarez2, J. Leiva Espinoza2,
E. Quinonez3,4, P. Romero 7, F. Nievas5, R. Maurette6,
C. Luque8, D. Matus1, P. Surraco1, M. Fauda5,6,
G. Gondolesi7, L. McCormack3, J. Mattera2,4,
O. Imventarza1
1
Hospital Nacional De PediatrJuan P Garrahan, CABA,
CABA; 2Hospital Italiano De Buenos Aires, CABA, CABA;
3
Hospital Aleman, CABA, CABA; 4Hospital El Cruce Dr.
Nestor Carlos Kirchner, Florencio Varela, BUENOS
AIRES; 5Hospital Universitario Austral, Derqui, BUENOS
AIRES; 6Hospital Britanico, CABA, CABA; 7Fundacion
Favaloro, CABA, CABA; 8Hospital General De Ninos Dr
Ricardo Gutierrez, CABA, CABA
Grafts from split livers constitute an accepted approach to
expand the donor pool. There was a significant interest from
most of the Argentinean centers to increase in the use of this
technique over the last 5 years.
Aims: Describe and analyze the outcomes of right (RSG)
and left (LSG) grafts from a multicenter study.
Methods: Multicenter retrospective study including data
from 111 recipients of split liver grafts between 1/1/2009
31/12/2013. Incidence of surgical complications, patient and
graft survival, and factors that affected RSG and LSG survival were analyzed.
Results: Grafts types were 57 LSG and 54 RSG. Median
follow up times for LSG and RSG were 46 and 42 months
respectively. The 36-month patient and graft survivals for
LSG were 83% and 79%, and for RSG was 78% and 69%
respectively. Retrasplantation rates for LSG and RSG were
3.5% and 11% respectively. Artherial complications were the
most common cause of early retransplantation (less than 12
months). Cold ischemia time (CIT) longer than 10 hours and
the use of high risk donors (older than 40 years or BMI 30
or 5 days ICU) were independent factors for diminished
graft survival in RSG. No analyzed variables were associated
with worse graft survival in LSG. Biliary complications were
the most frequent in both groups(57% in LSG and 33% in
RSG).
Conclusions: Partial grafts obtained from liver splitting are
an excellent option for patients in need of liver transplantation, and have the potential to alleviate the organ shortage.
Adequate donor selection and reducing CIT are crucial for
optimizing results.
HPB 2015, 17 (Suppl. 1), 181
LO-C.02 INTRAOPERATIVE
HYPOTENSION DURING LIVER
TRANSPLANTATION IS ASSOCIATED
WITH DECREASED ONE YEAR PATIENT
AND GRAFT SURVIVAL
P. F. Sauer, D. A. DuBay, P. A. MacLennan,
J. H. Crawford, J. A. White, S. H. Gray, D. E. Eckhoff
University Of Alabama At Birmingham, Birmingham, AL
Introduction: Patients with end-stage liver disease experience peripheral vasodilatation and have lower mean arterial
blood pressure (MAP) at baseline. However, it is unclear how
intraoperative hypotension effects post-liver transplant
patient and graft survival. We hypothesized that increased
duration of MAP <60 mmHg (MAP <60) was associated
with decreased patient survival.
Methods: A retrospective study of adult liver transplant
patients from a single center was performed. Anesthesia
records were used to quantify the intraoperative duration of
MAP <60. Patients were stratified into 3 groups based on
cumulative duration MAP <60: <30 minutes, 3059 minutes
and 60+ minutes. Blood product usage, length of hospitalization, vasoactive drug administration, and 1-year patient
and graft survival were measured. MAP <60 duration groups
were compared using ANOVA, Chi-square tests, and LogRank tests for continuous and categorical variables, and survival curves, respectively.
Results: Overall 565 patients were included: <30 minutes
(N = 461, 81.6%), 3059 minutes (N = 42, 7.4%), and 60+
minutes (N = 62, 11.0%). Patients in the 60+ minutes group
had the highest MELD score (27.7, p = 0.0051), received
the most units of pRBCs (5.5, p < 0.0001), FFP (2.96,
p < 0.0001), platelets (1.5, p < 0.0001) and intravenous phenylephrine (14487 g, p < 0.0001), and had the lowest 1-year
patient (72.2%, p = 0.0060) and graft survival (72.2%,
p = 0.0063)(Table).
Conclusion: Intraoperative hypotension, i.e., MAP <60, for
60+ minutes was associated with increased intraoperative
phenylephrine, blood product usage, and decreased 1-year
patient and graft survival. Surgical technique, anesthetic
management, and the avoidance of hypotension are critically
important for patient outcomes.
2015 The Authors

10
Abstracts
LO-C.03 A GLIMPSE INTO THE FUTURE:

DO WE KNOW WHAT INCREASING
NATIONALLY SHARED LIVER
ALLOGRAFTS WILL MEAN?
A. Ertel, K. Wima, R. Hoehn, D. Abbott, S. Shah
University Of Cincinnati Department Of Surgery,
Cincinnati, OHIO
Purpose: Due to the current geographic disparities in liver
allocation, a new policy endorsing broader sharing of liver
allografts has been proposed. In this model there will be a
substantial increase in the number of nationally shared,
deceased donor livers. We aimed to investigate the potential
ramifications this new policy may have on patient outcomes
and resource utilization following liver transplantation (LT).
Methods: The Scientific Registry of Transplant Recipients
(SRTR) was queried for all nationally shared LT from 2007
2011 and linked to the University Health Consortium (UHC)
database (n = 12,445). Univariate analysis was used to determine how donor, recipient characteristics, clinical outcomes
and hospital resource utilization were associated with
national sharing as compared to local and regional sharing.
Results: Nationally shared livers have a higher donor risk
index (>1.8), and are more likely to be classified as expanded
criteria donors compared to regional and locally shared
livers. Recipients of nationally shared livers were more likely
to be independent, have lower MELD scores, and were less
likely to be hospitalized or severely limited at the time of
transplant. Nationally shared LTs were more likely to be
performed at high volume centers, resulted in higher transplant hospitalization costs, in hospital mortality, and 30-day
readmissions (Table 1).
Conclusion: Nationally shared donor livers are of lesser
quality, and are being transplanted into healthier recipients.
Despite this, they have increased mortality, cost, and readmission rates. These data suggest that with the proposed
policy of broader sharing, patient outcomes will be impaired,
though require more hospital resources.
LO-C.04 SOCIETAL REINTEGRATION

FOLLOWING CADAVERIC LIVER
TRANSPLANTATION
R. P. Kelly, M. Molinari
Dalhousie University, Halifax, NS
Background: Data related to social reintegration following
Orthotopic liver transplantation (OLT) is lacking. For the
purpose of this study social re-integration was defined as:
the ability of an individual to return to the same marital,
educational, and financial status they held pre-OLT. The
2015 The Authors
primary aim of this study was to assess the rate of social

re-integration following OLT in Atlantic Canada.
Methods: Between September 2006 and January 2008
patients that were 3 years post OLT were included in this
prospective cohort study. Externally validated Questionnaires were self administered to those who participated. All
data was then prospectively collected. Categorical variables
were analyzed by Chi-Square or Fishers exact test and continuous variables by Students t-test.
Results: Among 157 eligible patients 110 (70%) participated. The mean age was 57 years (SD + 11.4). Prior to OLT,
75% were married and 6% were divorced. Following OLT
there was no significant difference in marital status. Employment rate fell from 72% pre-OLT to 30% post-OLT. Lowskill employees were less likely to return to work post-OLT
compared to than those who were self-employed. Personal
income fell an average of $4,363 (SD + 20,733) (p = 0.03).
Nevertheless, 38% of patients reported improved earning
potential post-OLT. Overall, 80% of participants expressed
satisfaction for their role in society post-OLT.
Conclusion: Following OLT, few patients are able to fully
reintegrate into the positions and status held pre-OLT.
Employment status is most likely to be negatively affected
following OLT. These findings suggest that focused strategies for preserving social status be implemented before and
after OLT.
LO-C.05 DEFINING THE INFLUENCE OF

GENDER DISCORDANCE ON SURVIVAL
FOLLOWING LIVING DONOR LIVER
TRANSPLANTATION
J. T. Miura, A. Fathi, F. M. Johnston, S. Tsai,
K. K. Christians, K. K. Turaga, T. C. Gamblin
Medical College Of Wisconsin, Department Of Surgery,
Milwaukee, WI
Background: Donor-recipient gender discordance has been
suggested to be an independent predictor of poor outcomes
following deceased donor liver transplantation (DDLT).
Whether a similar trend occurs with living donor transplants
(LDLT) remains ill defined. The present study sought to
evaluate the association between gender mismatch and survival following LDLT.
Study Design: Patients that underwent LDLT were identified from the Organ Procurement and Transplant Network
database (20002012), and categorized by the following
donor-recipient gender groups: Female (F)/F, Male (M)/F,
M/M, and F/M. Log rank test and multivariable Cox proportional hazards models were used to assess graft survival (GS)
and overall survival (OS) of the study groups.
Results: A total of 3,143 LDLT were identified: 731 F/F,
794 M/F, 1,015 M/M, and 603 F/M. Ten year GS and OS
were highest in the F/F group, at 62% and 71% respectively,
as compared to M/F (53%, 66%), M/M (55%,62%), and F/M
(55%,66%). Multivariate analysis demonstrated that gender
mismatch was not associated with poorer GS (MM: Reference; MF: HR 1.22, p = 0.06; FF: HR 1.02, p = 0.86; FM:
HR 1.18, p = 0.18). Independent factors associated with
poorer GS included older recipient age (HR: 1.01, 95% CI:
1.001.02; p = 0.01), increasing donor age (HR: 1.01,
95%CI: 1.001.02; p = 0.01), higher Model for End-stage
Liver Disease score (HR: 1.02, 95%CI: 1.011.04;
p = 0.003), and left lobe allografts (HR: 2.08, 95%CI: 1.55
2.79; p < 0.001).
HPB 2015, 17 (Suppl. 1), 181
Abstracts
Conclusions: Donor-recipient gender discordance does not
impact both GS and OS following LDLT. Living donor organ
allocation in an era of supply limitation should not be influenced by donor gender.
LO-C.06 RECCURRENT
HEPATOCELLULAR CANCER AFTER
LIVER TRANSPLANTION: THE ROLE OF
LIVER-DIRECTED THERAPY
V. Donchev, G. Voidonikolas, M. Sheckley, A. Annamalai,
I. Kim, A. Klein, A. Wachsman, M. Friedman,
S. Colquhoun, N. Nissen
Cedars-Sinai Medical Center, Los Angeles, CA
Introduction: Recurrence of hepatocellular cancer (HCC)
after liver transplant (LT) generally carries a poor prognosis.
We reviewed our experience to determine the role of liverdirected treatments.
Methods: Retrospective review of 10-year single center
experience. All patients were within radiographic Milan criteria for HCC at the time of LT. Of 201 patients undergoing
LT for HCC, 29 (14%) were identified with HCC recurrence.
Results: Mean and median time to recurrence after LT was
24 +/ 4 mo. and 15 mo. (range 3160 mo.) respectively.
Explant pathology showed that most patients (26/29) had
pathologic staging exceeding Milan criteria. The initial
recurrence was liver-only in 7 patients (24%), liver-dominant
in 5 patients (17%) and systemic in 17 patients (59%). Time
to recurrence was greatest in the liver-only group at mean
34 mo., followed by mean 24 mo. in the liver-dominant
group and mean 14 mo. in systemic group. Patients with
liver-only or liver-dominant tumor underwent a total of 57
loco-regional treatments (range 114). There were 3 significant treatment-related complications.
Survival: Mean survival after tumor recurrence in all
patients was 15 +/ 4 mo. The 1 and 3 year survival after
recurrence was 62% and 21% respectively. Patients undergoing liver-directed therapy had improved survival compared to
the remainder of patients (25 mo. vs 14 mo., p < .05, figure).
Conclusions: HCC recurrence after LT is heterogeneous.
Almost all recurrences are in patients with poor explant
pathology. Aggressive liver-directed therapies are safe and
effective in patients with liver-dominant recurrence and can
lead to improved survival.
HPB 2015, 17 (Suppl. 1), 181
11

4:30PM6:30PM
LONG ORAL D PANCREAS
PERIOPERATIVE/TECHNIQUES
LO-D.01 IMPROVED PERIOPERATIVE
OUTCOMES WITH EPIDURAL
ANALGESIA IN PATIENTS
UNDERGOING PANCREATECTOMY:
A NATIONWIDE ANALYSIS
D. E. Sanford, W. G. Hawkins, R. C. Fields
Washington University, St. Louis, MISSOURI
Background: Despite scant evidence demonstrating benefit,
epidural analgesia (EA) is often used for patients undergoing pancreatectomy. We sought to examine the impact
of epidural analgesia on postoperative outcomes after
pancreatectomy.
Methods: We used the 20082011 Healthcare Cost and Utilization Project Nationwide Inpatient Sample to examine the
effect of EA on perioperative outcomes. Students t tests and
chi-square tests were used for univariate comparisons.
Multivariable logistic and linear regression with propensity
score matching were utilized for risk-adjusted comparisons.
Results: 12,440 patients underwent pancreatectomy. Of
these, 1,130 (9.1%) patients received epidural analgesia.
Patients who received EA were significantly more likely to
be male, undergo pancreaticoduodenectomy, have cancer,
and undergo surgery at higher volume centers. By univariate
comparison, patients who received EA had significantly
lower rates of pneumonia, blood transfusions, and acute renal
failure, and this was associated with decreased postoperative
length of stay (LOS), lower hospital charges, and decreased
postoperative mortality. In multivariate analyses, EA was
independently associated with decreased postoperative LOS
(adjusted mean difference = 1.2 days, p < 0.01), decreased
hospital charges (adjusted mean difference = $16,814,
p < 0.01), and decreased postoperative mortality (adjusted
OR = 0.42, p < 0.01). Using 1 : 1 propensity score matching,
patients who received EA (n = 1,070) had significantly
decreased postoperative LOS (11.0 days vs 12.1 days,
p = 0.01), lower hospital charges ($112,086 vs $128,939,
p < 0.01), and decreased postoperative mortality (1.5% vs
3.6%, p < 0.01) compared to matched controls without EA
(n = 1,070) (Table).
Conclusion: EA is associated with improved perioperative
outcomes and reduced hospital charges following pancreatectomy. Additional studies are required to fully understand if
this relationship is causative.
2015 The Authors

12
Abstracts
ence in OS according to the presence of IPMN at margin (96
vs 123 months, p = 0.18). However, when present, degree of
IPMN dysplasia at the margin (low vs moderate vs high vs
invasive) impacted OS (84 vs 8 vs 13 vs 9 months, respectively p = 0.002). In addition, topography of positive margin
in the ductal system (main vs branch vs mixed) impacted
DFS (19 vs 65 vs 34 months, respectively p = 0.009).
Conclusion: Positive margin status in main-duct involved
IPMN may have effects on patient survival. Involvement of
the main duct at the margin, and higher degrees of IPMN
dysplasia at the margin predict a worse survival. Total pancreatectomy may need careful consideration in select patients
in these groups.
LO-D.03 INCREASED MORBIDITY AND

MORTALITY OF CONCOMITANT
COLECTOMY DURING
PANCREATICODUODENECTOMY:
A NSQIP PROPENSITY SCORE
MATCHED ANALYSIS
J. W. Harris, J. T. Martin, E. C. Maynard, P. C. McGrath,
C. D. Tzeng
University Of Kentucky, Lexington, KY
LO-D.02 IMPACT OF MARGIN STATUS

ON SURVIVAL IN MAIN DUCT
INVOLVED INTRADUCTAL PAPILLARY
MUCINOUS NEOPLASM
M. T. Welsh1, A. M. Roch1, J. L. Cioffi1, J. J. Easler2,
J. M. DeWitt2, E. P. Ceppa1, M. G. House1,
N. J. Zyromski1, A. Nakeeb1, C. Schmidt1
1
Department Of Surgery, Indiana University School Of
Medecine, Indianapolis, IN; 2Department Of Medicine,
Division Of Gastroenterology, Indiana University School
Of Medicine, Indianapolis, IN
Background: The natural history of Main Duct (MD)involved Intraductal Papillary Mucinous Neoplasm (IPMN)
is poorly understood. We examined the impact of pancreatic
margin status on recurrence and survival, and whether there
is benefit of total pancreatectomy (TP) over partial pancreatectomy (PP).
Methods: A retrospective review of a prospectively maintained database of patients who underwent resection for
IPMN at a single academic center (20002013) was performed.
Results: 152 patients with MD-involved IPMN were
included in this study. Of them, 15 underwent TP and 137 PP.
Median follow-up was 41 months. There was no difference in
surgical (postoperative complications, hospital stay) and
long-term outcomes (malignant recurrence, overall survival
OS and disease-free survival DFS) between TP and PP
groups. In patients who underwent PP, there was no differ 2015 The Authors
Introduction: Select patients with periampullary cancers

require concomitant colon resection (CR) during pancreaticoduodenectomy (PD) for margin-negative resections. This
study analyzed the impact of concomitant CR on post-PD
major morbidity and mortality using a large national dataset.
Methods: National Surgical Quality Improvement Program
(NSQIP) patients undergoing PD for periampullary cancers
(with/without CR) from 20052012, were screened. A 4 : 1
propensity score matched analysis was constructed to isolate
the impact of CR upon PD. Risk factors for 30-day major
morbidity and mortality were analyzed to determine the postoperative sequelae of PD + CR.
Results: Of 10,965 PD and 159 PD + CR patients in total,
624 and 156, respectively, were selected for the 4 : 1 matched
analysis. PD + CR resulted in significantly higher major
morbidity and mortality (50.0% and 9.0%) vs. PD alone
(28.8% and 2.9%, respectively, p < 0.001). Multivariate
analysis identified the following risk factors for major morbidity after PD: concomitant CR (OR-3.19, p < 0.001),
smoking history (OR-1.92, p = 0.005), lack of functional
independence (OR-3.29, p = 0.018), cardiac disease (OR2.39, p = 0.011), decreased albumin (per g/dL, OR-1.38,
p = 0.033), and longer operative time (vs. median time,
OR-1.56, p = 0.029). Independent predictors of mortality
included concomitant CR (OR-3.16, p = 0.010), ventilator
dependence (OR-13.87, p < 0.001), and septic shock (OR6.02, p < 0.001).
Conclusion: Contrary to previous single-institution studies,
this propensity score matched analysis using the NSQIP
dataset showed that adding CR to PD significantly increased
the magnitude of surgery and was an independent predictor
of both major morbidity and mortality. Using high-resolution
imaging, patients who may need PD + CR should be identified preoperatively, maximally optimized, and referred to
expert centers.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
13
LO-D.05 FACTORS INFLUENCING

FAILURE TO RESCUE AFTER
A NSQIP PERSPECTIVE
LO-D.04 LAPAROSCOPIC
PANCREATICODUODENECTOMY DOES
NOT COMPLETELY MITIGATE
INCREASED PERI-OPERATIVE RISKS IN
ELDERLY PATIENTS SEEN WITH OPEN
PANCREATICODUODENECTOMY
M. C. Tee, K. P. Croome, C. R. Shubert, M. B. Farnell,
M. J. Truty, F. G. Que, K. Reid-Lombardo, R. L. Smoot,
D. M. Nagorney, M. L. Kendrick
Mayo Clinic, Rochester, MN
Background: The effect of totally laparoscopic pancreaticoduodenectomy (TLPD) in elderly patients has not been
evaluated. We hypothesize that increased peri-operative risks
associated with open pancreaticoduodenectomy (OPD) in
elderly patients may be mitigated with TLPD.
Methods: A review of prospectively collected data on all
pancreaticoduodenectomy (PD) was conducted from August
2008 to January 2014 (N = 756). Elderly patients (Age 70
years, N = 281) were compared to non-elderly patients (Age
<70 years, N = 475) with respect to risk-adjusted perioperative morbidity and mortality. Differences in outcomes
between TLPD (N = 106) versus OPD (N = 175) were evaluated in the elderly subgroup in an intention-to-treat analysis.
Results: Elderly patients have increased risk of ICU admission, any cardiac event, and pneumonia following PD compared to non-elderly patients. TLPD in elderly patients is
associated with decreased risk of surgical site infection (SSI)
and delayed gastric emptying (DGE) as well as decreased
estimated blood loss (EBL) and transfusion. Only factors that
were statistically significant on multivariate analysis are
reported in the results table. All other outcomes (pancreatic
fistula, hemorrhage, length of stay, etc.) were not statistically
significant on multivariate analysis.
Conclusions: Elderly patients undergoing TLPD experience
a similar risk of 90-day mortality, ICU admission, and cardiopulmonary events compared to patients undergoing OPD.
TLPD does appear to offer benefits to the elderly with respect
to decreased EBL, DGE, and SSI.
HPB 2015, 17 (Suppl. 1), 181
P. Varley, A. Tsung
University Of Pittsburgh, PITTSBURGH, PA
Background: Previous studies have identified structural
factors that may impact the failure to rescue rate after
pancreaticoduodenectomy (PD). The goal of this study is to
identify patient-level factors associated with failure to rescue
in order to guide interventions that prevent progression to
mortality.
Methods: Patients undergoing PD as the primary procedure
were identified from the 20052012 National Surgical
Quality Improvement Project (NSQIP) Participant Data Use
(PUF) files. Since NSQIP only measures mortality for 30
days post-operatively, we treated failure to rescue as time to
event data and analyzed it using Cox proportional hazards
methods.
Results: A total of 14,546 patients were available for analysis. Of these, 1137 (7.8%) experienced only a minor complication while 5321 (36.6%) experienced at least one severe
complication. Failure to rescue rate was 0% vs. 5.9%, respectively (p < 0.001). Failure to rescue rates were 1.2%, 4.2%
and 18.6% in patients experiencing a total of 1, 2 or 3+
serious complications (p < 0.001). Results from univariable
Cox regression were used to build a multivariable Cox model
which was refined by AIC criteria. Factors significantly influencing failure to rescue after serious complication included
number of complications, resident participation, age,
reoperation, and dyspnea (Table 1).
Conclusions: Essentially all patients who experience postoperative mortality after PD first had a serious complication.
Interestingly, our analysis shows that it is not the first postoperative complication, but instead the accumulation of
multiple events that results in failure to rescue. Preventing
this progression should be the focus of future quality
improvement efforts.
2015 The Authors

14
Abstracts
place. In univariate analysis, patients with RPSA had a significantly higher rate of postoperative SSIs, and this was
associated with longer length of postoperative stay, higher
postoperative hospital costs, and increased postoperative
30-day readmission rates (Table). In Multivariate analysis,
RPSA was an independent predictor of postoperative SSI
(OR = 1.68, p = 0.013), and the risk of SSI increased with
increasing RPSA length of stay (OR = 1.07 per day,
p = 0.001).
Conclusions: RPSA is an important risk factor for SSI after
pancreatectomy. Many patients with RPSA are not admitted
preoperatively to the same hospital where pancreatectomy
occurs; in such circumstances, SSI rates may not be a sole
reflection of the care provided by operating hospitals.
LO-D.07 PANCREATECTOMY WITH

VEIN RESECTION/RECONSTRUCTION:
TECHNIQUE MATTERS
M. M. Dua1, T. B. Tran1, J. Q. Klausner2, K. J. Hwa1,
G. A. Poultsides1, J. A. Norton1, B. C. Visser1
1
Stanford University School Of Medicine, Stanford, CA;
2
UCLA Medical Center, Los Angeles, CA
LO-D.06 THE IMPACT OF RECENT

HOSPITALIZATION ON SURGICAL
SITE INFECTION FOLLOWING
PANCREATECTOMY
D. E. Sanford, R. C. Fields, W. G. Hawkins
Washington University In St. Louis, St. Louis, MO
Background: Surgical site infections (SSI) are a major
cause of increased morbidity and cost after pancreatectomy.
Patients undergoing pancreatectomy frequently have had
recent inpatient hospital admissions prior to their surgical
admission (recent presurgical admission, RPSA), which
could increase the risk of SSI. We sought to examine the
impact of RPSA on SSI following pancreatectomy.
Methods: We used the 20092011 Healthcare Cost Utilization Project California State Inpatient Database. RPSA was
defined as hospital stays >48 hrs with a discharge date within
30 days prior to admission for pancreatectomy. We used
Chi-square tests, Students t tests, and multivariable logistic
regression.
Results: 3,376 patients underwent pancreatectomy, and 444
(13.2%) had RPSA. The most common RPSA diagnoses
were pancreatitis (n = 164, 36.9%) and biliary obstruction
(n = 255, 57.4%), and 235 (52.9%) underwent an endoscopic
procedure during RPSA. 180 (40.5%) RPSAs were to different hospitals other than where patients pancreatectomy took
2015 The Authors
Introduction: A variety of techniques have been described

for portal vein (PV) and/or superior mesenteric vein (SMV)
resection/reconstruction during pancreatectomy. The ideal
strategy remains unclear.
Methods: Using a prospective database, we identified all
patients between 20052014 who underwent PV/SMV
resection/reconstruction during pancreatectomy (subtotal/
total/whipple) for all diagnoses. We reviewed the medical
records and imaging for operative details and outcomes, with
special emphasis on patency.
Results: Ninety patients underwent vein resection/
reconstruction with one of five techniques:1) longitudinal
venorrhaphy (n = 17, 19%); 2) transverse venorrhaphy
(n = 9, 10%); 3) primary end-to-end (n = 28, 31%); 4) patch
venoplasty (n = 17, 19%); and 5) interposition graft (n = 19,
21%). With median follow-up (last available imaging to
assess patency) of 282 dys, thrombosis was observed in
16/90 (18%). The rate of thrombosis varied according to
technique. All patients with primary end-to-end or transverse
venorrhaphy remained patent. Longitudinal venorrhaphy,
patch closure, and interposition graft were all associated with
significant rates of thrombosis (25%, 31%, 44%, respectively, p < 0.01 vs no thrombosis). Comparing those
thrombosed to those that remained patent, there were no
differences with respect to pancreatectomy type, preoperative knowledge of vein involvement, and neoadjuvant
therapy. Patients with thrombosis had significantly longer
HPB 2015, 17 (Suppl. 1), 181
Abstracts
operative times (480 vs 401 min,p < 0.01) and increased
blood loss (1150 vs 600 mL,p < 0.05). Post-operative heparin
drip was used in only 7%. Prophylactic aspirin was used in
69% of the total cohort (66% of patent, 81% of thrombosed)
and showed no protective benefit.
Conclusions: Primary end-to-end and transverse venorrhaphy have better patency than the alternatives after
PV/SMV resection and should be the preferred techniques
for short (<3 cm reconstructions).
15
adenocarcinomas (63 and 60%) or neuroendocrine tumors

(13 and 15%). Operating Room (OR) and postoperative outcomes are presented in the table.
Conclusions: Distal pancreatectomy with celiac axis resection is associated with increased operative time, renal failure
and a 10% operative mortality. The decision to offer an
Appleby procedure should be made with full disclosure of
the increased risks.
LO-D.10 SURGICAL MANAGEMENT OF

CHRONIC PANCREATITIS: A THERAPY
IN DECLINE?
L. A. Bliss, C. J. Yang, M. Eskander, S. DeGeus,
M. P. Callery, T. S. Kent, A. J. Moser, S. D. Freedman,
J. F. Tseng
Beth Israel Deaconess Medical Center, Boston, MA
LO-D.09 DISTAL PANCREATECTOMY

WITH CELIAC AXIS RESECTION: WHAT
ARE THE ADDED RISKS?
J. D. Beane1, M. G. House1, S. C. Pitt3, E. M. Kilbane1,
B. L. Hall3, A. Parmar4, T. S. Riall4, H. A. Pitt2
1
Indiana University School Of Medicine, Indianapolis, IN;
2
Temple University School Of Medicine, Philadelphia, PA;
3
Washington University School Of Medicine, St. Louis,
MO; 4University Of Texas Medical Branch, Galveston, TX
Background: Surgeons have become aggressive at operating on tumors of the body of the pancreas which require
resection of the celiac axis (Appleby procedure). Reported
series are small and not adequately controlled. The aim of
this analysis was to report a large series of Appleby procedures with a comparison group to determine the relative
risk.
Methods: Data were gathered through the American
College of Surgeons-National Surgical Quality Improvement
Program, Pancreatectomy Demonstration Project. Over 14
months, 822 patients underwent a distal pancreatectomy
(DP) at 43 institutions. Twenty of these patients (2.4%) also
underwent celiac axis resection (CAR). Appleby procedure
patients were then matched by age, gender, BMI, serum
albumin, ASA class, gland texture, duct size and pathology to
180 patients undergoing DP without CAR. Operative and
postoperative outcomes were compared by Fishers Exact
and Wilcoxon tests.
Results: The median age of the DP and DP + CAR patients
was 65 and 64 years, respectively. Most patients were female
(67 and 70%). The mean BMI of the two groups was
identical (27.1 kg/m2). The majority of patients had
HPB 2015, 17 (Suppl. 1), 181
Background: Surgical intervention is uncommon in chronic

pancreatitis management. Literature largely describes single
institution or international experiences. This study describes
US-based chronic pancreatitis surgical management and its
impact on readmission rates.
Methods: Retrospective analysis of chronic pancreatitis
patients in Healthcare Cost and Utilization Project Florida
State Inpatient Database 20072011 using revisit variables.
Patients with malignancy or congenital abnormalities
excluded. Surgical interventions and complications of
chronic pancreatitis identified using ICD-9 codes. Univariate
analysis of sex, Elixhauser score, race, insurance, complications, receipt of surgery by chi-square. Number of readmissions, time to surgery by Wilcoxon rank sum. Multivariate
analysis of operative management by logistic regression.
Results: 21,448 chronic pancreatitis patients. 6.18% (1,325)
underwent surgery including 629 with drainage procedures,
275 with pancreatectomies, 735 with cholecystectomies.
Procedures decreased from 8.65% in 2007 to 3.10% in 2011
(p < 0.0001). 12.95% (2,778) developed pancreatitis-related
complications: pancreatic cysts or pseudocysts (4.40%), diabetes (10.02%). Pancreatic exocrine insufficiency in <11
patients. Median number of readmissions 1 (IQR 04) and 5
(IQR 29) among non-surgical and surgical patients, respectively (p < 0.001). Median number of admissions prior to
pancreatectomy was 2 (IQR 14) and drainage procedure
was 2 (IQR 16). Predictors of surgical intervention displayed in table.
Conclusions: Chronic pancreatitis leads to numerous inpatient readmissions and difficult to manage complications.
Surgical intervention occurs in a declining minority of cases.
Complicated patients are more likely to undergo surgery;
operative patients experience more admissions than
nonoperative patients. The complexities of chronic pancreatitis management warrant early multidisciplinary evaluation
and ongoing consideration of surgical and nonsurgical
options.
2015 The Authors
16
Abstracts
both groups, the predicted FLR was inversely correlated with
the % in liver regeneration only (p < 0.001).
Conclusion: Neoadjuvant chemotherapy does not seem to
affect the liver regeneration. The predicted FLR only is
inversely correlating with the amount of LR occurring after
major resection or after PVE.
LO-E.02 HIGH-DOSE STEREOTACTIC

BODY RADIOTHERAPY (SBRT) FOR
PRIMARY AND METASTATIC
LIVER TUMORS

4:30PM6:30PM
LONG ORAL E LIVER
ONCOLOGY
LO-E.01 NEOADJUVANT
CHEMOTHERAPY DOES NOT IMPAIR
LIVER REGENERATION FOLLOWING
MAJOR HEPATECTOMY OR PORTAL
VEIN EMBOLIZATION FOR
COLORECTAL LIVER METASTASES
E. Simoneau1, N. Molla1, R. Alanazi2, J. Alshenaifi2,
M. Aljiffry3, A. Medkhaly2, L. Boucher1, P. Metrakos1,
M. Hassanain1,2
1
Department Of Surgery, McGill University, Montreal, QC;
2
Department Of Radiology, McGill University, Montreal,
QC; 3Department Of Surgery, College Of Medicine, King
Saud University, Riyadh, SA; 4Department Of Surgery,
Faculty Of Medicine, King Abdulaziz University, Jeddah,
SA; 5Department Of Oncology, McGill University,
Montreal, QC
Introduction: Treatment strategies for colorectal liver
metastasis (CRCLM) such as major hepatectomy and portal
vein embolization (PVE) rely on the regenerative capacity of
the liver. Neoadjuvant chemotherapy is most often used with
patients undergoing these procedures. We aimed to investigate the effect of neoadjuvant chemotherapy on liver regeneration after PVE and after major hepatectomy.
Methods: All CRCLM patients undergoing PVE or major
resection (without PVE) with 3D liver volumetry measurements were included. Liver regeneration (expressed as future
liver remnant (FLR) and percentage of liver regeneration(%
LR)), total liver volume (TLV) and clinical characteristics
were collected from our CRCLM database.
Results: Between 20032013, 226 patients were included
(85 major resections, 141 PVE). Mean age was 63 12 years
old and median number of cycles was 6(58). In each group,
overall adequate regeneration was observed (+96.5% in FLR
(p < 0.001) post PVE and +45.8% in FLR (p < 0.001) post
resection). In the PVE group, chemotherapy variables did not
show significant association with the amount of liver regeneration (number of cycles (p = 0.435), timing (p = 0.563),
chemotherapy agent (p = 0.116)). Similarly in the major
hepatectomy group, neoadjuvant chemotherapy administration did not show a significant association with %LR
(p = 0.592) or with other treatment variables (number of
cycles, p = 0.114; agent, p = 0.061, timing, p = 0.126). In
2015 The Authors
A. Kirichenko, D. Parda, K. Tom, P. Abrams, J. Oliva,

M. Szramowski, N. Thai
Allegheny Health Network, Pittsburgh, PA
Purpose: SBRT has emerged as an effective non-invasive
radiation therapy to precisely target liver tumors with ablative doses of radiation while avoiding surrounding liver
tissue. We report our single center experience on the efficacy
and tolerability of high-dose SBRT in the management of
inoperable primary and metastatic liver tumors from 2007
2013.
Patients and Methods: Patients with 14 hepatic lesions
and tumor diameter 8 cm received liver SBRT of 4060 Gy
delivered in 46 fractions. The primary end point was local
control with at least 8 months of radiographic follow-up, and
secondary end points were toxicity and survival.
Results: 68 patients (79 lesions) completed high-dose liver
SBRT for HCC (31) or oligo-metastases (37). 22 patients
underwent hepatic resections or liver transplant in combination with SBRT. With median followup 22.5 months (range,
867 months) overall survival was 78% for patients with
hepatic oligometastases and 60% in HCC patients (including
11 patients who completed SBRT prior to liver transplant).
No incidence of >grade 2 treatment toxicity or accelerated
MELD score migration was observed. Overall local control
within radiation field at two years after SBRT was 94% and
for lesions with diameter of 4 cm was 100%.
Conclusion: In this retrospective analysis we demonstrate
that liver SBRT is safe and effective for the treatment of
hepatic malignancies providing local control rates similar to
hepatic resection.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
17
LO-E.03 THE ROLE OF LYMPHOCYTE

TO NEUTROPHIL RATIO (LNR) AND
PLATELET TO LYMPHOCYTE RATIO
(PLR) AS PROGNOSTIC MARKERS IN
METASTATIC COLORECTAL
CARCINOMA: A REVIEW OF
DATA FROM A RANDOMIZED
CONTROLLED STUDY
P. Philips1, C. R. Scoggins1, R. D. Tomalty2,
M. T. Schreeder3, J. Kaugh4, K. Kim4, W. R. Rilling5,
C. J. Laing6, C. M. Tatum7, L. R. Kelly7,
R. D. Garcia-Monaco8, V. R. Sharma1, R. A. Redman1,
T. S. Crocenzi1, S. M. Strasberg1, R. C. Martin1
1
University Of Louisville And James Graham Brown
Cancer Center, Louisville, KY; 2Huntsville Hospital
Interventional Radiology, Huntsville, AL; 3Clearview
Cancer Institute, Huntsville, AL, USA, Huntsville, AL;
4
Winship Cancer Institute Emory University, Atlanta, GA;
5
Froedtert Medical College, Milwaukee, WI; 6Radiological
Associates Of Sacramento (RAS), Sacramento, CA; 7Norton
Radiology Associates, Louisville, KY; 8Hospital Italiano,
Buenos Aires, ARGENTIAN
Background: Recent evidence suggests that elevated
neutrophil-lymphocyte ratio(NLR) and platelet-lymphocyte
ratio(PLR) are adverse prognostic markers for survival in
various cancers including colorectal carcinoma.
Methods: Data from a multicenter, prospective-randomized
DEBIRI(Irinotecan-drug eluting beads) study comparing
LC Bead, loaded with irinotecan plus chemotherapy and
bevacizumab versus chemotherapy with bevacizumab in
unresectable metastatic colorectal cancer was analyzed.
Results: There were 30 in the control arm and 41 patients
(112 treatments) in the DEBIRI test arm. The median NLR
and PLR for the whole cohort was 3.41 (range: 1.113.9,
IQR: 2.9,4.2) and 187.5 (range: 77792, IQR: 146.6,227)
respectively. NLR and PLR correlated well with adjusted R2
of 0.825 (p-0.000, F = 128). On univariate analysis, an NLR
>5(median OS 31.9 vs.14.7 months, p = 0.067) and a PLR
>150 (median 31.6 vs. 24 months, p = 0.046) was associated
with decreased overall survival (OS). Other factors associated with decreased OS were presence of response (CR/PR
or SD) to treatment (16.9 vs. 33.3 months, p0.001), ECOG
score of 1 (33.7 vs. 17.4 months, p = 0.006), lung metastasis (18.5 vs. 33.6 months, p = 0.016), history of heart disease
(16.4 vs. 28.7, p = 0.033), albumin <3.5 (10.5 vs. 31.9,
p = 0.001). On multivariate analysis with Cox proportionalhazards model, only response to treatment was an independent predictor of decreased OS (hazard ratio = 0.25, 95%CI:
0.060.9, p = 0.044).
Conclusion: In this study, response to treatment was the
most important favorable prognostic marker in metastatic
colorectal carcinoma with liver dominant disease. Patients
with high NLR and PLR showed a trend towards worse
prognosis, albeit not independently and needs to be investigated in larger studies.
HPB 2015, 17 (Suppl. 1), 181
LO-E.04 MANAGEMENT AND OUTCOME

OF COLORECTAL CANCER (CRC) LIVER
METASTASES IN THE ELDERLY:
A POPULATION-BASED STUDY
S. Nanji1,2, W. J. Mackillop2,3,4, X. Wei3, C. M. Booth2,3,4
Department Of Surgery, Queens University, Kingston,
ON; 2Department Of Oncology, Queens University,
Kingston, ON; 3Department Of Public Health Sciences,
Queens University, Kingston, ON; 4Division Of Cancer
Care And Epidemiology, Queens University Cancer
Research Institute, Kingston, ON
1
Background: Surgical resection is standard treatment for

patients with CRC liver metastases (LM). Limited data
describe practice and outcomes among elderly patients. We
report management and outcomes of surgical resection of
CRC LM in the elderly in routine practice.
Methods: All cases of CRC in Ontario who underwent surgical resection of LM from 19942009 were identified using
the population-based Ontario Cancer Registry. We linked
electronic records of treatment to the registry to identify
surgical procedures and utilization of chemotherapy. Pathology reports provided details regarding extent of disease and
surgical procedure. Patients were classified as <65, 6569,
7074, and 75 years of age.
Results: We identified 1310 patients: 710 (54%) <65; 220
(17%) 6569; 194 (15%) 7074; and 186 (14%) 75 years of
age. Mean number of lesions (2.3, 2.1, 1.8, 1.6, p < 0.0001)
and mean size of the largest lesion (4.0, 4.3, 4.4, 4.5 cm,
p = 0.031) varied across age groups. Elderly patients were
less likely to undergo a major liver resection (3 segments):
55%, 17%, 15%, 12%, p = 0.10. Peri-operative chemotherapy was used less frequently in the elderly (71%, 63%,
2015 The Authors
18
Abstracts
51%, 41%, p < 0.0001). 90-day mortality (2%, 6%, 4%, 8%,
p < 0.001) was greatest among patients 75 years of age.
Overall survival at 5 years across the age groups was 49%,
40%, 47%, 28% (p < 0.0001).
Conclusions: Resection of CRC liver metastases is associated with greater risk of post-operative mortality among
elderly patients despite less aggressive treatment. Although
the long-term outcomes are inferior to younger patients, a
substantial proportion of elderly patients will have long-term
survival.
LO-E.05 FIBROSES PREDICTS SURVIVAL

IN PATIENTS WITH COLORECTAL
LIVER METASTASES RESECTED AFTER
PREOPERATIVE CHEMOTHERAPY
M. C. Marques, H. S. Ribeiro, R. S. De Souza,
W. L. Da Costa, A. L. Diniz, A. L. De Godoy,
I. C. De Farias, M. F. Begnami, R. Chojniak,
V. H. Fonseca, R. S. Bonachi, F. J. Coimbra
A.C. Camargo Cancer Center, Sao Paulo, Brasil
Pathological response to preoperative chemotherapy in
patients with resected colorectal liver metastases has been
identified as one of the most powerful predictors of outcomes. Fibrosis, necrosis and percentage of viable cells are
recognized as markers of response, but these changes are
also present in liver specimens from patients who didnt
have preoperative treatment. The aim of this study was to
analyze the incidence of these pathological findings in
patients exposed or not to preoperative treatment and, in the
group of preoperative treatment, identify predictive factors
of pathological response and their impact on survival
results. From 2009 to 2012, ninety patients were analyzed,
and twenty of them did not have preoperative chemotherapy.
Survival analysis only included patients who had preoperative chemotherapy. These patients had higher percentages
of 25% fibrosis (p = 0.007) and <50% viable cells
(p = 0.005). With a median follow-up period of 26.5
months, patients who had preoperative treatment and <25%
of fibrosis had a statistically significant worse 3-year overall
survival (96.4 59.9%, p = 0.019). In multivariate analyses,
the only independent prognostic factor for overall survival
was 25% of fibrosis (HR = 5.5, p = 0.045, 95% CI 1.0
29.3) and it was correlated with Kras wild type status
(HR = 4.1, p = 0.032, 95% CI 1.115.4). No pathologic
findings could predict disease-free survival results. Percentage of fibroses seems to be the best predictor of survival
among response variables.
2015 The Authors

LO-E.06 CYTOREDUCTIVE SURGERY

AND HYPERTHERMIC
INTRAPERITONEAL CHEMOTHERAPY
IN PATIENTS WITH LIVER
INVOLVEMENT
Y. Berger1, S. Aycart1, P. Tabrizian1, J. Mandeli2, S. Hiotis1,
U. Sarpel1, D. Labow1
1
Department Of Surgery, Division Of Surgical Oncology,
Mount Sinai Medical Center, New York, NEW YORK;
2
Department Of Preventive Medicine, Mount Sinai School
Of Medicine, New York, NEW YORK
Background: The aim of this study was to examine the
perioperative and long-term results in patients with liver
involvement undergoing cytoreductive surgery and
hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) at
a single tertiary institution.
Methods: Data were obtained from a prospectively collected database maintained between March 2007 and July
2014. All patients undergoing CRS/HIPEC were divided into
patients who underwent synchronous liver resection (group
1) and those who did not (group 2). Perioperative and longterm results were compared between the groups.
Results: Out of 269 CRS/HIPEC procedures, group 1
included 103 procedures (38%) performed in 101 patients
with primary diagnoses of colorectal cancer (CRC, 27%),
appendiceal cancer (33%), pseudomyxoma peritonei (13%)
and other malignancies. Most patients (77%) in group 1
underwent stripping of the liver capsule or minor parenchymal resection, whereas 24 patients (23%) underwent parenchymal resection of at least one liver segment. Group 1 had
longer duration of surgery (p < 0.001), increased
intraoperative blood loss (p = 0.004), higher number of
organs resected (p < 0.001), longer hospital stay (p = 0.002)
and more ICU admissions (p = 0.02). Major complication
and 90-day mortality rates were not significantly different
between the groups. Median overall survival for CRC
patients in group 1 was poorer (21.8 months, vs. 43.8 months
in group 2, p = 0.01).
Conclusions: CRS/HIPEC procedures with synchronous
liver resection are prolonged, challenging and often require
HPB 2015, 17 (Suppl. 1), 181
Abstracts
multivisceral resections in order to achieve optimal
cytoreduction, but are safe even when substantial parenchymal resection is being performed. Liver involvement at the
time of CRS/HIPEC is a marker of poor survival in CRC
patients.
19
age and Charlson comorbidity index (both p < 0.001), both

use of hepatic resection (HR = 0.40, 95%CI: 0.380.42,
p < 0.001) and current treatment after 2002 (HR = 0.93,
95%CI: 0.910.96, p < 0.001) were associated with
improved overall survival.
Conclusions: Despite statistical significance, clinical
improvement in the rate of hepatic resection from 6.1% to
7.9% is minor. As survival continues to improve, ongoing
efforts should focus on improving access to operative treatment for patients with hepatic metastases.
LO-E.08 SURVIVAL OF PATIENTS AFTER

PORTAL EMBOLIZATION FOR LIVER
METASTASES OF COLORECTAL
CANCER
Y. Collin, R. Huang, M. Plasse, R. Letourneau,
M. Dagenais, S. Turcotte, A. Roy, R. Lapointe,
F. Vandenbroucke-Menu
Centre Hospitalier Universitaire de Montreal, Service de
Chirurgie HBP, Montreal, QC
LO-E.07 ACCESS AND SURVIVAL:

A POPULATION-BASED STUDY OF
CURRENT OUTCOMES AFTER HEPATIC
RESECTION IN PATIENTS WITH
METASTATIC COLORECTAL CANCER
V. M. Zaydfudim1, T. L. McMurry2, A. M. Harrigan1,
C. M. Friel1, G. J. Stukenborg2, T. W. Bauer1,
R. B. Adams1, T. L. Hedrick1
1
University Of Virginia, Charlottesville, VIRGINIA;
2
University Of Virginia, Charlottesville, VIRGINIA
Background: Despite compelling institutional data supporting hepatic resection, population-based studies historically
reported underutilization of operative treatment in patients
with colorectal metastases to the liver. The current study
examines trends in hepatic resection and survival among
Medicare recipients with hepatic metastases.
Methods: Medicare recipients with incident colorectal
cancer diagnosed 19912009 were identified from the linked
SEER-Medicare dataset. Patients were stratified into historical control (19912001) and current cohort (20022009).
Analyses compared rates of hepatectomy and peri-operative
morbidity and mortality. Cox proportional hazards model
tested effects of clinically relevant variables on overall
survival.
Results: Of 31,574 patients with metastatic colorectal
cancer to the liver, 14,925 were in the current cohort treated
after 2002 and 16,649 comprised the historical control group.
The rate of hepatic resection increased from 6.1% pre-2002
to 7.9% currently (p < 0.001). The proportion of patients
treated with major (>3 segments) or minor hepatectomies did
not change (p = 0.345). Peri-operative morbidity including
hemorrhage, infections, or gastrointestinal complications did
not differ (all p > 0.089). 30-day mortality did not differ
between historical controls (4.5%) and current cohort
(3.7%), p = 0.329. After adjusting for significant effects of
HPB 2015, 17 (Suppl. 1), 181
Introduction: Portal vein embolization (PVE) is an efficient

method used to induce liver hypertrophy in cases of otherwise unresectable liver tumors, but its impact on survival is
debated. This study compares survival and peri-operative
data between patients resected with prior PVE (group PVE)
and without (group NoPVE).
Methods: This is a prospective study 128 patients with
colorectal liver metastases (CRLM): 71 with PVE and 57
without.
Results: The groups were without significant difference for
age, sex and prior medical history. The PVE group had more
lesions (3.30 VS 2.84; p < 0.001). Within the PVE group,
operative time was longer (221.31 VS 186.54 min; p = 0.01).
There was no statistical difference between the groups for
blood loss (756.25 VS 604.02 mL; p = 0.078). There was no
difference in complication rates between the groups
(p = 0.465) and there was no peri-operative mortality. Hospital stay was shorter in the PVE group (7.84 VS 8.95 d,
p = 0.038). 5-year overall survival rate and disease free survival rate were similar between groups (53.2% VS 54.4%;
p = 0.999 and 38.3% VS 40.6%; p = 0.823) Median overall
survival and disease free survival were 65.4 months and 33.6
months respectively, with a median overall follow-up of 39.3
months.
Conclusion: Our results showed similar survival whether
PVE was used or not. Hence, PVE offers a chance for cure
for patients who could not be operated upfront. Moreover,
PVE patients seem to have the same prognosis even with a
larger extent of disease.
2015 The Authors

20
Abstracts

5:00PM6:30PM
ORAL POSTER I (BILIARY,
EDUCATION, LIVER)
OP-I.01 T2 GALLBLADDER CANCER
STILL A NODAL DISEASE
M. R. Sheikh, H. Osman, S. Cheek, S. Hunter,
D. R. Jeyarajah
Methodist Dallas Medical Center, Dallas, TX
LO-E.09 EVALUATING ATTITUDES

TOWARD AND APPLICATION OF
MECHANISMS TO AUGMENT THE
LIVER IN NORTH AND SOUTH
AMERICA (THE MALINSA SURVEY)
R. W. Day, C. Conrad, J. Vauthey, T. A. Aloia
Department Of Surgical Oncology, The University Of Texas
MD Anderson Cancer Center, Houston, TEXAS
Introduction: Various techniques, including PVE, PVL,
and ALPPS, are being used to augment the future liver
remnant volume in preparation for major hepatectomy. Given
the significant variation between and within these techniques, there is no scientific way to compare their safety and
efficacy.
Purpose: The aim of this study was to survey and document
the availability, variation, utilization and attitudes toward
each of these techniques across high volume HPB centers in
North and South America
Method: A descriptive 20 question survey was developed
and internally validated with expert review. The survey was
IRB approved and distributed to 42 high volume centers in
Canada, US, Mexico, and South America. Data were collected, collated and analyzed.
Results: Complete surveys were returned from 23 institutions, including representatives from each region (Canada,
US, Mexico, and South America). All of the institutions
responding performed PVE with 5 centers (21.7%) also performing ALPPS procedures. In the previous year, the average
PVE and ALPPS procedures performed were 15.75 and 6.2
per instutition respectively. Only 18 (78.3%) reported the
capability to extend PVE to segment 4, and 12 (52.2%)
reported embolization utilizing embolic microspheres.
Twenty respondents (87%) rated PVE the safest option for
liver hypertrophy; however, 12 respondents (52.2%) believe
the ALPPS procedure is most likely to result in adequate
hypertrophy.
Conclusions: There exists extreme variability in utilization
and attitudes toward the available techniques for FLR volume
augmentation. Penetration of best practice techniques for
PVE is lacking and may contribue toward the attraction of the
riskier ALPPS procedure.
2015 The Authors

Introduction: Treatment of gall bladder cancer (GBC) has

traditionally been viewed with pessimism and lymph node
positivity has been associated with worse prognosis. The aim
of this study is to analyze lymph node positivity in patients
with T2 tumors.
Methods: All patients who underwent surgery for GBC
between September 2005 and June 2014 have been identified
retrospectively in our database. Data collected included clinical presentation, operative findings, and histopathological
data.
Results: Charts of 36 patients were reviewed. 26 patients
had incidental GBC diagnosis following cholecystectomy. 10
patients were T2 on initial cholecystectomy and all underwent subsequent radical resection. 2 patients from this group
were N1 on initial cholecystectomy and 4 more patients
became N1 on second surgery. Overall 60% patients with T2
disease had node positivity. 2 patients were found to have
residual disease at the liver margin and were upgraded to T3
following resection, one of them also had N1 disease. Overall
50% patients with T2 disease had stage upgrade after radical
resection. 10 patients were diagnosed on imaging. 3 of these
patients were unresectable and 6 were either stage T3 or
higher or node positive.
Conclusions: 60% of T2 GBC were node positive in our
experience. 50% T2 patients found on initial cholecystectomy have stage upgrade as a result of radical surgery. These
findings support the call for radical resection in patients with
incidental diagnosis of T2 tumor on cholecystectomy. This
study also shows preoperative imaging diagnosis is associated with higher stage and overall worse prognosis.
OP-I.02 ADJUVANT THERAPY

FOR INTRAHEPATIC
CHOLANGIOCARCINOMA: DOES
ANYONE BENEFIT?
R. S. Hoehn, K. Wima, A. Ertel, A. M. Meier, S. Ahmad,
J. J. Sussman, S. A. Shah, D. E. Abbott
Department Of Surgery, University Of Cincinnati School
Of Medicine, Cincinnati, OH
Objectives: There are no level one data to suggest a survival
advantage for patients with hepatobiliary cancers who
receive adjuvant therapy following surgery. We sought to
identify patients with intrahepatic cholagniocarcinoma (IHC)
who may benefit from adjuvant therapy.
Methods: The American College of Surgeons National
Cancer Data Base (NCDB) was queried for all patients with
resected IHC (pathologic stage 13) between 19982006
(n = 920). Three groups were compared: surgery only
HPB 2015, 17 (Suppl. 1), 181
Abstracts
(S, n = 626), surgery with adjuvant chemotherapy (AC,
n = 65), and surgery with both adjuvant chemotherapy and
radiation therapy (ACR, n = 146). Univariate and Cox
regression analysis were used to investigate the influence of
patient demographics, tumor characteristics and operative
details on receipt of adjuvant therapy and overall survival.
Results: Patients who received adjuvant treatment more
likely to have positive lymph nodes (S: 28.9%; AC: 47.5%;
ACR: 48.2%), positive surgical margins (S: 12.3%; AC:
17.0%; ACR: 38.0%), and pathologic stage 3 disease (S:
50.5%%; AC: 60.0%; ACR: 67.1%)(all p < 0.01). Multivariate analysis revealed that adjuvant chemotherapy alone was
not associated with a survival benefit in any patient groups
examined. Adjuvant chemotherapy and radiation did not
affect survival overall, but patients with positive lymph nodes
who received ACR did have improved survival compared to
surgery alone (HR 0.64, 95%CI 0.450.91).
Conclusion: Patients with high-risk IHC (positive lymph
nodes) appear to benefit from adjuvant chemotherapy and
radiation therapy, but lower risk patients do not. In the
absence of data from a randomized trial, these results can
guide application of adjuvant therapy following resection of
IHC.
OP-I.03 SUBTOTAL
CHOLECYSTECTOMY FOR THE
HOSTILE GALLBLADDER
M. E. Lidsky, A. W. Castleberry, A. Perez, T. N. Pappas
Department Of Surgery, Duke University Medical Center,
Durham, NC
Background: Outcomes following the inability to safely
control the cystic duct in the setting of a hostile triangle of
Calot during cholecystectomy remain unknown. The purpose
of this study was to analyze the safety and efficacy of subtotal
cholecystectomy, with specific attention to the necessity and
timing of secondary procedures.
Methods: Medical records of 16,585 cholecystectomies
from January 2002-August 2014 were reviewed, with identification of patients managed with subtotal cholecystectomy,
defined as the inability to isolate and transect the cystic duct.
We investigated surgical indications, intraoperative variables,
and 30-day postoperative mortality and morbidity. We also
analyzed the necessity for ERCP, percutaneous drainage procedures, and completion cholecystectomy.
Results: 69 (0.4%) patients underwent subtotal cholecystectomy, of which 57 (82.6%) were laparoscopic; 30 (43.5%)
required conversion to laparotomy. 1 (1.4%) patient died
postoperatively, and 26 (37.7%) patients suffered 35 complications, most frequently infectious (14 wound/surgical site
infections, 4 UTIs). Indication for cholecystectomy included
acute cholecystitis (69.6%), 10 (14.5%) of which had a
cholecystostomy tube, symptomatic cholelithiasis (23.2%),
chronic cholecystitis (13%), and biliary pancreatitis (10.1%).
Secondary interventions were required in the form of 49
ERCPs in 20 (29%) patients, percutaneous drainage for
biloma or abscess in 6 (8.7%), and completion cholecystectomy on average 13.75 months (527 months) after the index
operation in 4 (5.8%).
Conclusions: The hostile gallbladder represents a complicated disease process for which patient safety is of
paramount. Postoperative morbidity after subtotal cholecysHPB 2015, 17 (Suppl. 1), 181
21
tectomy is significant. Most do not require completion

cholecystectomy; however, these patients demand close
observation and, frequently, secondary interventions.
OP-I.04 MAJOR BILE DUCT INJURY

AFTER LAPAROSCOPIC
CHOLECYSTECTOMY: EXPERIENCE
FROM A TERTIARY REFERRAL CENTER
IN RURAL STATE
A. Greenbaum, E. Alkhalili, I. Nir
University Of New Mexico, Albuquerque, NEW MEXICO
Background: Risk factors for iatrogenic bile duct injuries
(BDI) after laparoscopic cholecystectomy (LC) remain a
topic of controversy. Few studies have examined ethnicity or
body mass index as potential risk factors resulting in the need
for complex biliary reconstruction. We hypothesize that LC
performed in a rural setting may present a higher risk of BDI
in certain patient populations.
Methods: This retrospective cohort study includes all
patients referred to our tertiary center from 20102014 for
biliary reconstruction secondary to major BDI during LC.
Results: A total of 21 patients were analyzed. These patients
were predominately female (76% vs. 24% male). A disproportionate number of patients were Native American (47.6%
vs. 10.4% comprising the New Mexico state population). The
mean BMI of all patients was 33.5, though the Native American population average BMI was 39.2. The majority of referrals came from rural centers (62%). The mean time to BDI
recognition after LC was highest at rural facilities (16.4 days
vs. 3.3 days at metropolitan centers) as well as the timing
of referral (4.7 days vs. 0.8 days). The overall rate of morbidity was 42.8% and one patient died of sepsis prior to
reconstruction.
Conclusion: Major BDI after LC is a highly morbid event.
This retrospective cohort study suggests Native American
patients, specifically those who are morbidly obese, may be
at higher risk of BDI during laparoscopic cholecystectomy in
a rural setting. We prompt rural surgeons to consider early
referral of high risk patients to a tertiary setting for gallbladder surgery.
OP-I.05 RACIAL DISPARITIES IN

PATIENTS WITH GALLBLADDER
CANCER
S. Zenoni1, X. Zhu1, P. Velduis1, S. Eubanks1,2,
P. Arnoletti1,2, S. De La Fuente1,2
1
Florida Hospital Orlando, Orlando, FL; 2University Of
Central Florida, Orlando, FL
Background: Epidemiological studies have shown widely
variable geographic patterns in patients with gallbladder
cancer. The incidence rates are extraordinarily high in Latin
America, Asia and some countries in eastern and central
Europe but relatively low in the United States. In contrast to
other biliary malignancies, there is limited data regarding
racial disparities and oncologic outcomes in patients with
gallbladder cancer. In this study, a prospectively maintained
tumor registry was used to determine overall survival,
2015 The Authors
22
Abstracts
time-to-initial treatment, and time-to-surgery according to

race in patients with gallbladder cancer.
Methods: Patients with gallbladder cancer registered at the
Florida Hospital tumor database from the year 20012013
were included in the analysis. Analyzed variables included
basic demographics as well as grade and pathological ACCJ
stage at diagnosis, date at diagnosis, initial treatment date,
surgery date, last date of contact, patients status (alive/dead)
at the last contact, and overall survival according to race.
Results: A total 133 patients with gallbladder cancer were
identified from the database; 91 of which were identified as
non-Hispanic, 22 as Hispanics, and 20 as African Americans.
There were no differences in age, gender distribution, grade
and pathological ACCJ stage at presentations between the
different races. Five-years overall survival (figure), time-toinitial treatment, and time-to-surgery were not significantly
different between groups.
Conclusions: This study shows similar survival rates and
time to therapy between Caucasian and minorities. Furthermore, as opposed to what it has been observed with other
cancers, no differences were noted in time-to-treatment in
gallbladder cancer according to race.
ences in length of stay, reoperation, readmission, or 30-day

mortality rates. Patients who underwent bile duct resection
had shorter DSS compared with patients not requiring bile
duct resection (9.3 vs 39.9 mo; p = 0.002; Figure). When
accounting for differences between the two groups, the need
for bile duct resection was independently associated with
reduced DSS (HR: 3.06;95%CI: 1.128.34;p = 0.029).
Conclusion: Major hepatectomy with concomitant bile duct
resection is seldom performed in patients undergoing resection of colorectal cancer liver metastases. Bile duct resection
is associated with higher major morbidity and reduced
disease-specific survival. Stringent selection criteria should
be applied when patients may need bile duct resection during
hepatectomy for colorectal cancer liver metastases.
OP-I.06 THE IMPACT OF CONCOMITANT

BILE DUCT RESECTION WITH MAJOR
HEPATECTOMY ON SURVIVAL
OUTCOMES OF PATIENTS
UNDERGOING TREATMENT OF
METASTATIC COLORECTAL CANCER:
A MULTI-INSTITUTIONAL ANALYSIS
OF 429 PATIENTS
L. M. Postlewait1, M. H. Squires1, D. A. Kooby1,
S. M. Weber2, C. R. Scoggins3, K. Cardona1, C. S. Cho2,
R. C. Martin3, E. Winslow2, S. K. Maithel1
1
Division Of Surgical Oncology, Emory University, Atlanta,
GA; 2Division Of Surgical Oncology, University Of
Wisconsin, Madison, WI; 3Division Of Surgical Oncology,
University Of Louisville, Louisville, KY
Background: Data are lacking on the long-term outcomes
of patients undergoing major hepatectomy requiring a bile
duct resection for the treatment of colorectal cancer
metastases.
Methods: All patients who underwent major hepatectomy
(3 segments) for metastatic colorectal cancer from 2000 to
2010 at three US academic institutions were included.
Patients who died from unknown cause were excluded.
Primary outcome was disease-specific survival (DSS).
Results: Of 456 patients, 429 met inclusion criteria.
Median follow-up was 38.7 mos. Bile duct resection was
performed in 9 patients (2.1%) and was associated with preoperative portal vein embolization (25.0%vs4.3%;p =
0.049). There were no significant differences in other clinicopathologic factors between the two groups (age, ASA
class, margin status, number of lesions, tumor size, cirrhosis,
perineural invasion, and lymphovascular invasion). Bile
duct resection was associated with increased postoperative major complications (Clavien III-V) in univariate
(66.7%vs20.7%;p = 0.004) and multivariate analyses (HR:
6.22;95%CI: 1.4426.97;p = 0.015). There were no differ 2015 The Authors
OP-I.08 INCORPORATING TRANSITION

TO PRACTICE INTO A
HEPATOPANCREATICOBILIARY
FELLOWSHIP: A MODEL FOR
GRADUATED AUTONOMY
C. Scally, S. G. Warner, R. M. Minter
University Of Michigan, Ann Arbor, MICHIGAN
Introduction: There is significant concern for graduating
trainees readiness for independent practice. The American
College of Surgeons (ACS) has developed pilot fellowships
to facilitate Transition to Practice. However, these programs currently serve as an alternative to advanced
subspecialty training. We sought to evaluate our institutional
experience incorporating a novel transition to practice
element into an advanced HPB fellowship.
Methods: We obtained complete operative records from
Fiscal Years 201214 through present for our HPB fellows
(201113 fellow A, 201315 fellow B). We then identified all
cases in which the fellows billed as the attending surgeon, as
well as all cases in which they participated as a trainee. We
also analyzed relative value units (RVUs), and total charges
billed by the fellows per financial year.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
Results: The independent practice element expanded in
each year of the fellowship (Table). Independent cases represented 36.6% of the fellows total operative volume from
201214. In the second year of the fellowship, the fellows
have increased the complexity of diagnoses seen in their
personal clinic with explicit recognition of their ability to
progress in this manner. The fellows do not participate in
either emergency or service call as an attending.
Conclusions: Our HPB fellows training included a significant independent practice component, with an expanding
scope of practice in each year of training. This method of
graduated autonomy in an advanced HPB fellowship may
represent a feasible blended model for advanced surgical
training, meeting both the need for specialty expertise and
preparation for independent practice.
HPB 2015, 17 (Suppl. 1), 181
23
OP-I.09 STROKE VOLUME VARIATION

(SVV) CONTINUOUS MONITORING FOR
INTRAOPERATIVE INTRAVASCULAR
FLUID MONITORING IN HPB SURGERY
E. H. Baker, J. Drummond, A. Cochran, R. Seshadri,
J. Martinie, D. Iannitti, R. Swan
Carolinas HealthCare System, Charlotte, NORTH
CAROLINA
Introduction: Previous studies have demonstrated that
stroke volume variation (SVV) represents a more accurate
determinant of intravascular volume status and fluid responsiveness. This has been particularly true for patients who are
intubated in which previously used measurements of intravascular status such as mean arterial pressure (MAP) or
central venous pressure (CVP) may be misrepresentative.
Methods: Continuous intraoperative monitoring was performed for a consecutive series of patients who underwent
laparoscopic and open liver and pancreas surgeries. Data
points collected for analysis included SVV, MAP, CVP,
stroke volume (SV), cardiac index (CI) every 5 minutes.
Estimated blood loss (EBL), hemoglobin (Hgb), pH and
surgeon and anesthesia estimates of patient volume status
were determined every hour.
Results: 23 patients were enrolled and underwent continuous monitoring during the following procedures: 5 open
liver, 6 laparoscopic liver, 6 open pancreas, 6 laparoscopic
pancreas. A weak inverse relationship was seen between
SVV and total fluid status in 14 out of 23 total cases (60.9%)
as indicated by a negative linear regression slope. The correlations were low for all groups, with an average R2 of 0.10
or 10% of the variability in the model explained by the
SVV and total fluid variables.
Conclusions: SVV correlates inversely with total fluid
status and may be a used as a non-invasive tool in determining fluid status in HPB procedures. While the sample size
was small, we plan on using the data to develop a standardized fluid management protocol for HPB surgeries and
examine surgical outcomes compared to matched, retrospective cases.
2015 The Authors

24
Abstracts
OP-I.10 RECURRENCE PATTERN AND

SURVIVAL IN PATIENTS UNDERGOING
SIMULTANEOUS RESECTION FOR
SYNCHRONOUS LIVER METASTASES
FROM PRIMARY COLORECTAL
CANCER : RETROSPECTIVE STUDY OF
286 PATIENTS FROM A SINGLE HIGH
VOLUME HPB CENTRE
R. Kumar1, S. Park2
1
Sir H.N. Reliance Foundation Hospital, Mumbai,
MAHARASHTRA; 2National Cancer Center, Goyang,
GYEONGGI-DO
Background: The optimal combination of available therapies for patients with resectable synchronous liver metastases
from colorectal cancer (SLMCC) is unknown, and the pattern
of recurrence after simultaneous resection has been poorly
investigated. In this study, the authors examined recurrence
patterns and survival after simultaneous resection for
SLMCC.
Methods: Consecutive patients with SLMRC who underwent complete simultaneous resection of both the rectal
primary and liver metastases with curative intend between
May 2001 and December 2010 were identified from a prospective database. Fifteen patients were excluded due to
follow-up loss. Clinicopathological factors were retrospectively analyzed to investigate initial recurrence pattern affecting survival.
Results: In total, 286 patients underwent simultaneous
resection of both primary colorectal cancer and liver metastases with curative intend. The 3, 5 and 10-years recurrencefree survival rate were 35.5%, 31.5% and 27.2%,
respectively, for the entire cohort with a median follow-up of
60.8 months for survivors. 196 patients(68.5%) developed a
recurrence. Most common site of the initial recurrence was
liver(48.5%), followed by the lung (25%). Initial recurrence
pattern correlated with survival (P < 0.001). Analysis demonstrated that a loco-regional recurrence was significant risk
factor for survival.
Conclusions: Of the patients with SLMCC who developed
recurrent disease, systemic sites were overwhelmingly more
common than pelvic recurrences. The current results indicated that initial recurrence patterns in patients undergoing
simultaneous resection with curative intent was important to
predict survival, especially in patients with loco-regional
recurrence. Also, simultaneous resection seems safe and feasible despite lack of evidence, provided an institution specific
protocol is followed.
OP-I.11 IS THERE A ROLE FOR NEAR

INFRARED SPECTOMETRY (NIRS)
OXYMETRY DURING LIVER SURGERY?
Y. Collin1, T. Hu1, R. Allard1, A. Cloutier1, F. Payette1,
A. Denault2,3, R. Lapointe1, F. Vandenbroucke-Menu1
1
Centre Hospitalier Universitaire de Montreal, Service de
Chirurgie HBP, Montreal, QC; 2Centre Hospitalier
Universitaire De Montreal, Soins Intensifs, Montreal, QC;
3
Institut De Cardiologie De Montreal, Soins Intensifs,
Montreal, QC
Introduction: Peri-operative cerebral and somatic oxymetry is used for monitoring during cardiac surgery, but it
2015 The Authors
has never been used in liver surgery. Our goal was to define
the place of NIRS oxymetry in liver resections.
Methods: 90 patients undergoing major hepatectomy were
included (45 men and 45 women). Oxymetry was obtained
by NIRS at 4 sites (cerebral right and left, arm and thigh)
before and during surgery. Baseline oxymetry (BaseO) and
desaturation (Desat) (Threshold = 80% baseline) values
were compared to peri-operative data.
Results: Median ICU stay was 2d and median hospital stay
(LOS) was 7d. Cerebral BaseO correlated significantly with
duration of stay in the intensive care unit (ICU) stay
(p = 0.04), in the hospital length of stay (LOS) (p = 0.01) and
respiratory insufficiency (p = 0.002). Arm BaseO correlated
with blood loss (p = 0.05), blood transfusion (p = 0.03), ICU
stay (p = 0.01) and surgical complications (p = 0.049). Also,
thigh BaseO correlated with surgical complications
(p = 0.0035) and LOS (p = 0.01). Cerebral Desat did not
correlate with any complication but thigh Desat did with
blood loss (p = 0.03), LOS (p = 0.05) and surgical complications (0.0132). Arm Desat inversely correlated with Pringle
duration (p = 0.01).
Conclusion: NIRS is a very simple approach for oxymetry
evaluation during major liver surgery. Baseline and systemic
values are correlated with some operative data and complications. These preliminary results should lead to a more
extensive study to determine whether correction of impaired
level of oxymetry will improve patients outcome.
OP-I.12 USING THE CLINICAL RISK

SCORE TO PREDICT WHICH PATIENTS
WILL BENEFIT FROM PRE-OPERATIVE
CHEMOTHERAPY FOR COLORECTAL
LIVER METASTASES
N. Sela1,2, K. A. Bertens1,2, S. Welch1,2, J. Chung2,
C. S. Yoshy1, R. Hernandez-Alejandro1,2
1
London Health Sciences Centre, London, ONTARIO;
2
Western University, London, ONTARIO
Background: Pre-operative chemotherapy has become a
mainstay in the treatment of colorectal liver metastases
(CRLM) in patients with both resectable and unresectable
disease. We aim to evaluate whether the Clinical Risk Score
(CRS) can be used to predict patients with resectable disease
who will benefit from metastectomy before chemotherapy.
Methods: All patients with CRLM who underwent 4 or
more cycles of chemotherapy at our institution between 2000
and 2013 were retrospectively analyzed. Patients had to have
cross-sectional computed tomography (CT) imaging before
and after the treatment to meet inclusion criteria. Tumor
response was determined using RECIST 1.1 criteria. The
primary outcome was tumor progression, defined as RECIST
1.1 progressive disease. No progression was defined as
RECIST 1.1 stable disease, complete response, or partial
response. Multivariate regression was used to assess if the
CRS, as well as its individual components, were predictive of
disease progression.
Results: Seventy-seven patients were identified (mean
follow-up of 2.19 1.50 years). Twenty-seven patients had
progressive disease (35.1%). On multivariate analysis, a
lower CRS was predictive of disease progression on chemotherapy (OR = 0.370, p = 0.018). Furthermore, patients with
HPB 2015, 17 (Suppl. 1), 181
Abstracts
5 or more metastases (OR 0.122, p = 0.023), and increased
size of the largest lesion (OR 0.647, p = 0.047) were less
likely to progress.
Conclusion: Patients with more aggressive tumors, as
reflected by a higher CRS, are less likely to have tumor
progression on chemotherapy. Conversely, patients with low
CRS are more likely to progress on chemotherapy, and therefore those with resectable disease upfront would benefit from
metastectomy (without pre-operative chemotherapy).
OP-I.13 INTRAOPERATIVE
RADIOFREQUENCY ABLATION VERSUS
SURGICAL RESECTION IN SOLITARY
SMALL HCC
A. M. Elgendi, M. Elshafey, E. Bdeawey
Faculty Of Medicine, Alexandria University,
ALEXANDRIA, SELECT A STATE/PROVINCE
Background: Percutaneous radiofrequency ablation (RFA)
is used for treatment of small HCC however surgeons are
frequently using intraoperative RFA for tumors at locations
difficult for the percutaneous procedure. The aim was to
evaluate the results of intraoperative RFA for small HCCs
(<2 cm) at locations difficult for percutaneous route.
Methods: 420 patients with small solitary HCC (<2 cm)
were treated; 328 via percutaneous RFA while 92 patients
presented at sites not amenable for percutaneous route. 48
out of 92 patients underwent surgical resection, while 44/92
patients underwent intraoperative RFA.
Results: The location and depth of the HCC from the liver
capsule was the only significant factors in the choice of the
surgeon between resection and RFA. RFA group acheived
complete ablation rate of 100% compared to the surgery
group, where all patients achieved R0 resection. Complication rate was comparable (p = 1.0). After a median follow-up
of 46 months (range, 1665 months), no tumors showed
neither local progression nor local recurrence and no significant difference was observed between two groups as regards
early recurrence and number of de novo lesions (p = 0.49).
One-year and 3-year survival rates were 92% and 83%,
respectively, in the resection group comparable to the corresponding rates of 91% and 76% in the RFA group (p = 0.8).
Conclusion: For small HCC in locations difficult for a percutaneous approach, intraoperative RFA can be an alternative
option for deep seated tumors necessitating more than one
segmentectomy achieving similar tumor control, overall and
disease-free survival.
OP-I.14 HEALTH-RELATED QUALITY OF

LIFE FOLLOWING TREATMENT OF
NEUROENDOCRINE LIVER METASTASIS
G. Spolverato, H. Maqsood, Y. Kim, T. Luo, R. Gupta,
T. M. Pawlik
Johns Hopkins Hospital, Baltimore, MARYLAND
Background: A large subset of patients with neuroendocrine liver metastasis(NELM) is symptomatic at the
time of presentation. In addition to improving survival, treatment of NELM seeks to provide palliation of symptoms.
HPB 2015, 17 (Suppl. 1), 181
25
Data on health-related quality of life(QoL) are uncommon.

We sought to define patient-reported QoL following treatment of NELM.
Methods: Patients who underwent treatment of NELM at
Johns Hopkins Hospital between 19982013 and who were
alive as of March 2014 were identified(n = 276). These
patients were invited to complete a QOL survey designed
using validated assessment tools.
Results: The response rate was 27.5%(n = 76); questionnaires were completed at a median of 49.1 months(range,
23.3117.8) following initial treatment. Median patient age
was 55 years and the majority was male(n = 43, 56.6%).
Most patients had a pancreatic(n = 21, 27.6%) or a small
bowel(n = 30, 39.5%) primary tumor; the overwhelming
majority had multiple NELM(88.2%). Prior to the initiation
of any therapy, 84.2% patients reported symptoms, with the
most common symptoms being fatigue(78.9%), diarrhea(67.1%), and flushing(44.7%). Initial treatment of
NELM consisted of resection ablation(64.5%) or intraarterial therapy(IAT)(35.5%). Many patients reported overall
improvement in physical health(41.4%) and mental
health(34.3%). After treatment, the proportion of patients
with severe symptoms decreased from 36.8% to
21.0%(P = 0.03); symptoms such as diarrhea and flushing
improved at the end of the treatment course(both p < 0.05),
whereas fatigue remained unchanged(p = 0.45).
Conclusions: Surgery and IAT management of NELM
provides a reasonable improvement in patient-reported
symptoms and QoL. Liver-directed therapies should be considered in those patients with symptomatic, high-volume
disease even if complete treatment of the NELM is not
feasible.
OP-I.15 IMAGING SURVEILLANCE OF

HEPATOCELLULAR ADENOMAS
Y. Chun, R. Parker, S. Reddy, E. Ehrenwald, M. Hill,
S. Inampudi, T. Sielaff
Virginia Piper Cancer Institute, Minneapolis, MN
Background: A consensus surveillance protocol is lacking
for patients with hepatocellular adenomas.
Methods: Patients with hypervascular hepatic lesions
5 cm that did not meet criteria for focal nodular hyperplasia
or hepatocellular carcinoma were entered into a surveillance
schedule with contrast-enhanced MRI 6, 12, and 24 months
after baseline imaging. Patients with risk factors, including
male gender and active oral contraceptive pill (OCP) use,
were excluded. If lesions remained stable or decreased in
size, then surveillance imaging was discontinued.
Results: Between 20112014, 109 consecutive patients with
benign hypervascular lesions were evaluated at our multidisciplinary liver conference. Median follow-up from date of
diagnosis was 24 months (range, 0168 months). By imaging
criteria or biopsy, 37 patients were diagnosed with focal
nodular hyperplasia, and 44 patients with hepatocellular
adenoma, including 18 with adenomatosis. Twenty-eight
patients had indeterminate lesions. Clinically significant
hemorrhage +/ rupture occurred in 6 patients without prior
imaging and 2 patients with known hypervascular hepatic
lesions. All patients who suffered hemorrhage had adenomas
>5 cm and other risk factors, including OCP use (n = 7) and
anticoagulation (n = 1). In 43 patients eligible for our
2015 The Authors
26
Abstracts
surveillance schedule, all lesions remained stable or

decreased in size, and no patient developed complications.
Conclusions: Patients with hepatocellular adenomas
without risk factors, such as size >5 cm or OCP use, can
safely be observed with serial imaging 6, 12, and 24 months
after initial diagnosis. If lesions remain stable or decrease in
size, then longer-term surveillance is unlikely to identify
patients at risk for complications.
OP-I.17 NATIONAL TRENDS WITH

LAPAROSCOPIC LIVER RESECTION:
RESULTS FROM A POPULATION-BASED
ANALYSIS
J. He, N. Amini, G. Spolverato, K. Hirose, M. A. Makary,
C. L. Wolfgang, M. J. Weiss, T. M. Pawlik
The Johns Hopkins Hospital, Baltimore, MD
Background: Interest in laparoscopic liver resection (LLR)
has grown since the International Louisville Statement was
published in 2009. However, limited population-based data on
LLR utilization patterns and outcomes are available.
Methods: LLR data from the Nationwide Inpatient Sample
(NIS, 20002012) and the National Surgical Quality
Improvement Project (NSQIP, 20052012) were divided into
2 cohorts, before and after the Louisville Statement. Patient
demographics, indications, trends in LLR utilization, and
perioperative outcomes were compared before and after the
2009.
Results: Patients undergoing open versus LLR were comparable with regard to age, sex, and comorbidity status
(Table). 1,131 and 642 LLR were identified from NIS and
NSQIP, respectively. The majority of patients underwent
LLR for a malignant indication (NIS: primary malignancy,
29.9% vs. metastasis, 43.2%; NSQIP: primary malignancy,
25.7% vs. metastasis, 42.2%). The mean annual volume of
LLR increased from 20002008 vs. 20092012 (NIS: 63 vs.
168; NSQIP: 52 vs. 127; both P < 0.01). The perioperative
mortality associated with LLR was low (NIS: 2.8% vs.
NSQIP: 0.9%), while the morbidity was higher (NIS: 38.1%
vs. NSQIP: 30.7%); mortality and morbidity did not change
over time (both P > 0.05). Since 2009, LLR was associated
with a shorter length of stay (NIS: 5 vs. 6 days, P < 0.01) and
more likely to be performed in teaching hospitals (NIS: 93%
vs. 87%, P = 0.02).
Conclusions: Since the Louisville Statement in 2009, utilization of LLR has increased. LLR appears to be safe with
low mortality and reasonable morbidity, as well as be associated with a modest decrease in LOS.
2015 The Authors

OP-I.18 POST-HEPATECTOMY
HYPERBILIRUBINEMIA: THE POINT
OF NO RETURN
J. Y. Liu1, L. M. Postlewait1, J. W. Etra1, M. H. Squires1,
K. Cardona1, J. H. Winer1, J. M. Sarmiento2, C. A. Staley 1,
S. K. Maithel1, D. A. Kooby1, M. C. Russell1
1
Emory Winship Cancer Institute, Atlanta, GA; 2Emory
Department Of Surgery, Atlanta, GA
Background: Post-hepatectomy hyperbilirubinemia is associated with liver insufficiency and failure. The threshold of
the highest survivable total bilirubin (tbili) is not defined. Our
aim was to identify the peak postoperative tbili beyond which
is survival is improbable.
Methods: An institutional database of patients undergoing
major hepatectomy (3 segments), excluding biliary resections, from 20002012 was reviewed. A peak bilirubin of
18 mg/dL in the first 45 days post op was associated with
increasing 90-day mortality (90DM). Clinicopathologic
factors were assessed for association with 90DM. We also
examined predictors of elevated postoperative tbili.
Results: 607 patients were identified with a 90DM of 4.4%.
90DM for a peak tbili 18 (n = 16) was 81%, compared to
2.4% for a bilirubin <18 mg/dL (graph). All patients with a
tbili 30 died (n = 7). On multivariate analysis (MVA) for
90DM, post-operative tbili 18 (HR 24, CI 3.3174;p =
0.002), post-operative FFP (HR 4.8, CI 1.120.2;p = 0.034),
and cirrhosis (HR 5.9, CI 1.131.3;p = 0.038) were significant predictors. Furthermore, predictors of tbili 18 identified on MVA included: older age (HR 1.1, CI 1.01.2;
p = 0.001) and postoperative FFP (HR 10.1, CI 2.540.8;
p = 0.001).
Conclusion: Total bilirubin 18 is significantly associated
with an increase in 90-day mortality after major hepatectomy; there are no survivors for patients whose tbili rises
30.This information can help clinicians advise patients
and families who experience posthepatectomy hyperbilirubinemia; as well, it may be an important marker for intervention as supportive therapies improve.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
27
OP-I.19 EARLY AND LONG-TERM

SURVIVAL OUTCOMES OF PATIENTS
WITH COLORECTAL LIVER
METASTASES RESECTED AFTER
CONVERSION CHEMOTHERAPY
M. C. Marques, H. S. Ribeiro, R. S. De Souza,
I. C. De Farias, C. A. De Mello, F. A. Soares,
M. A. Almeida, F. J. Coimbra
Liver resection is the standard of care for patients with
colorectal liver metastases, but only 15% to 25% are
resectable at the initial diagnosis. Improvements in chemotherapy response rates and surgical technique expanded the
resectability criteria for patients initially considered as
having unresectable disease. These patients have an intermediate prognosis between those who were upfront
resectable and those who did not achieve resectability. The
aim of this study was to analyze survival outcomes and
determine clinicopathological prognostic factors in this scenario. Patients who underwent liver resection for colorectal
liver metastases after conversion therapy between 1998 and
2013 were retrospectively analyzed. Unresectability was
defined according to the Consensus Guidelines recently published. In the study period, 352 liver resections for colorectal
metastases were performed in 268 patients. Fifty-one
patients met the inclusion criteria and were analyzed. Thirty
and 90-day mortality rate was 5.8% and 9.8%. Major morbidity rate was 29.4% according to Clavien-Dindo Classification. Median number of chemotherapy cycles before
hepatectomy was 12. With a median follow-up of 36 months,
the 3-year overall survival was 66.1% versus 76.4% in the
resectable group (p = 0.060) and 3-year disease-free survival
was 10.4% versus 35.8% (p < 0.001). In multivariate analyses, factors that influenced overall survival were the occurrence of Class III and IV postoperative complications and
more than 4 liver nodules. There was no independent predictor of disease free survival on multivariate analyses. Liver
resection after conversion therapy for colorectal liver metastases is a procedure with high morbimortality but also associated with long-term survival in selected patients.
OP-I.20 RESECTED INTRAHEPATIC

CHOLANGIOCARCINOMA: PATTERNS
OF ADJUVANT THERAPY
AND RECURRENCE
I. T. Konstantinidis 1,2, A. X. Zhu1, L. Goyal 1, D. T. Ting 1,
V. Deshpande 1, K. K. Tanabe 1, K. D. Lillemoe 1,
C. R. Ferrone 1
1
Massachusetts General Hospital Department Of Surgery,
Boston, MA; 2University Of Arizona Department Of
Surgery, Tucson, AZ
Background: The majority of patients who undergo liver
resection for intrahepatic cholangiocarcinoma (ICC) suffer
from recurrence and succumb to their disease. The role of
adjuvant treatment remains unknown.
Methods: Clinicopathologic data of resected ICC between
1/200012/2013 were evaluated. Patterns of adjuvant therapy
and recurrence were analyzed.
Results: Of the 76 patients who underwent resection of a
ICC the median age was 65 years and 54% were female. The
majority of patients (72%) underwent a major hepatectomy
HPB 2015, 17 (Suppl. 1), 181
2015 The Authors

28
Abstracts
with 87% being an R0 resection. Median ICC size was

5.5 cm. Excluding patients who died within 60 days or were
lost to follow-up, 44%(30/68) did not receive adjuvant treatment (median overall survival (OS)not reached), 34%(23/68)
received adjuvant therapy within 3 months (OS: 48 months),
and an additional 22%(15/68) received therapy after the first
recurrence (OS: 44 months). Adjuvant chemotherapy was
more likely to be given to patients with an R1 resection
(26%vs7%;p = 0.05) and nodal disease (35%vs11%;p =
0.02). Gemcitabine based regimens were the most common
first line (85%) (gemcitabine: 32%, gemcitabine/cisplatin:
36%, gemcitabine/oxaliplatin: 32%); 5 FU/chemoradiation
was first line in 15%. After median follow-up of 25 months
41% remained free of disease(NED), 12% had a single liver
recurrence, 13% multinodular liver recurrence and 34% extrahepatic recurrence. Of the patients who recurred single
liver recurrences had the best median survival compared to
multinodular recurrences and distant disease (73 vs 23 vs 37
months respectively;p = 0.09).
Conclusions: The majority of patients undergoing resection
for an ICC either do not receive adjuvant treatment or they
receive it after recurrence of their tumor. Single nodule
hepatic recurrences have the best outcome.
Conclusion: Liver resection should be considered only in

selected patients with melanoma liver metastases without
EHD which is associated with poor outcomes. When EHD is
controlled with systemic therapy, indication for combined
liver ablation needs further investigations.
OP-I.21 ROLE OF LIVER RESECTION OR

ABLATION IN METASTATIC
MELANOMA MANAGEMENT
A. Doussot1, C. Nardin1, H. Takaki2, T. Litchman2,
M. I. DAngelica1, W. R. Jarnagin1, M. A. Postow3,
J. P. Erinjeri2, P. Kingham1
1
Department Of Surgery Memorial Sloan Kettering Cancer
Center, New York, NY; 2Department Of Interventional
Radiology Memorial Sloan Kettering Cancer Center, New
York, NY; 3Department Of Medical Oncology Memorial
Sloan Kettering Cancer Center, New York, NY
Introduction: The median survival for patients with metastatic melanoma is usually limited to approximately one year.
Liver ablation and metastasectomy are associated with
improved survival in well-selected patients, but their role in
era of more effective systemic therapies is uncertain.
Methods: Patients undergoing liver ablation or resection for
melanoma liver metastases between 1993 and 2013 were
included. Outcomes and prognostic factors such as medical
therapy impact were evaluated.
Results: Forty eight patients underwent ablation (n = 16) or
resection (n = 32) for metastases from cutaneous (n = 26) or
ocular (n = 22) origin. Median overall survival (OS) was 26
months, with 12 patients (27.3%) alive at 3 years, after resection (n = 9) and after ablation (n = 3) respectively. Patients in
the ablation group harbored more aggressive disease at time of
presentation, with more extrahepatic disease (EHD)
(p = 0.008) and a shorter disease-free interval between
primary tumor and liver metastasis diagnosis (p = 0.01). In
addition, 68.8% of patients received preoperative systemic
therapy before ablation. Median OS in ablation (18 months)
and resection (27 months) groups was not different (p = 0.9).
EHD was a poor prognostic factor for OS in the resection
group (p < 0.05; see Figure). In patients with EHD, 78% of
patients receiving preoperative therapy before ablation
experienced partial response and tended to have longer OS (12
months) compared to resected patients (6 months, p = 0.2).
2015 The Authors
OP-I.22 RADIOFREQUENCY-ASSISTED
LIVER PARTITION AND PORTAL VEIN
LIGATION (RALPP): COMPARATIVE
SERIES OF A MODIFIED ALPPS
TECHNIQUE FOR TWO-STAGE
LIVER RESECTION
M. H. Sodergren, T. M. Gall, M. Nagendran, L. R. Jiao
Imperial College, London, GREATER LONDON
Background: The introduction of portal vein embolization
and recently the ALPPS technique has rendered a greater
proportion of liver tumours surgically resectable by increasing the volume of future liver remnant (FLR) in selected
patients. The RALPP technique involves a laparoscopic first
stage portal vein ligation and in situ liver splitting using
ablation only without complete transection. We hypothesise
that this will rapidly increase the size of the FLR limiting any
associated morbidity from liver transection.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
Methods: Consecutive patients who underwent RALPP
were compared to an age-sex- and liver function-matched
cohort of patients undergoing PVE prior to right hepatectomy. The primary endpoint was the percentage increase in
FLR volume. Secondary endpoints were morbidity, mortality, and postoperative liver function.
Results: There were 12 patients (6M : 6F) in the RALPP
group and 8 (4M : 4F) in the PVE group with a median age
of 62.5 and 65 yrs respectively. The mean % increase in the
FLR volume was 61.5 +/ 16.3 measured after a mean of
20.8 +/ 7.3 days following the first stage for RALPP compared to a % increase of 16.46 +/ 11.7 (p = 0.001) after
52.3 +/ 14.8 days (p < 0.001) following PVE. There was
one mortality in the RALPP group at day 19 following right
hepatectomy from bowel ischaemia and liver failure. There
was no difference in morbidity or post-operative liver
function.
Conclusion: The RALPP technique is feasible and safe in
this limited series, with a greater increase in FLR volume in
a shorter time period compared to PVE.
OP-I.23 TEMPORAL TRENDS IN

SURGICAL RESECTION AND
PERI-OPERATIVE CHEMOTHERAPY
FOR COLORECTAL CANCER LIVER
METASTASES (CRCLM) IN ROUTINE
CLINICAL PRACTICE
1
Background: The treatment of CRCLM continues to

change over time. We report trends in management and
outcome of all patients with resected CRCLM in Ontario,
Canada.
Methods: All cases of CRC in Ontario who underwent
surgical resection of liver metastases in 20022009 were
identified using the population-based Ontario Cancer Registry. Electronic records of treatment and pathology reports
were linked to the registry to identify utilization
of neoadjuvant (NACT) and adjuvant chemotherapy (ACT)
and describe surgical management and pathologic findings.
We describe differences over 2 study periods: 20022005
and 20062009.
Results: During 20022009, 1711 patients underwent resection of CRCLM. Mean age was 63 years. During the study
period there was a 60% increase in patients undergoing
resection of CRCLM. For the 2 study periods, mean number
of liver lesions resected was 2.0 and 2.2 (p = 0.051), mean
size of largest lesion was 4.5 cm and 4.0 cm (p = 0.003),
major hepatic resection (3 Couinaud segments) rate was
66% and 63% (p = 0.264) and R1 resection margin rate was
6% and 9% (p = 0.021), respectively. 90-day mortality rates
for the study periods were 4% and 3% (p = 0.499). Use of
NACT and ACT increased from 19% to 41% (p < 0.001) and
42% to 50% (p < 0.001) between study periods, respectively.
Five year overall survival during the 2 study periods was 43%
(95%CI 4047) and 45% (95%CI 4248) (p = 0.402).
HPB 2015, 17 (Suppl. 1), 181
29
Conclusions: Resection of CRCLM and the use of perioperative chemotherapy increased during the study period.
Survival outcomes among patients treated in routine clinical
practice are comparable to institution-based studies.
OP-I.24 PATIENTS WITH HEPATITIS B

PRESENT WITH MORE ADVANCED
LIVER CANCER THAN PATIENTS WITH
HEPATITIS C
J. L. Pasko1, A. C. Anderton1, C. Costantino1, S. L. Orloff1,
L. L. Wong2, W. E. Naugler3
1
Oregon Health And Sciences University, Department Of
Surgery, Portland, OREGON; 2University Of Hawaii
Cancer Center And School Of Medicine, Department Of
Surgery, Honolulu, HI; 3Oregon Health And Sciences
University, Department Of Hepatology, Portland,
OREGON
Background: Well-described factors affecting the incidence
of Hepatocellular Carcinoma (HCC) include gender, age, and
etiology of liver disease. Factors that may affect the stage of
HCC at presentation are poorly understood. Stage at presentation, however, largely dictates available treatments and ultimately prognosis for patients.
Aims: Using a large cohort of patients with HCC, identify
elements that correlate with HCC stage at initial BCLC stage.
Methods: This is a two-institution retrospective review of
patients with pathologically or radiographically confirmed
HCC from January 1991August 2014.
Results: There were 1134 patients (835 men and 299
women) with HCC in this study. MELD scores ranged from
638. Higher MELD scores correlated with higher stages of
HCC (p = .01). Screening was found to correlate with a lower
BCLC stage (p = .001). Additionally, patients with HBV presented at higher stages (p = 0.03) compared to patients with
HCV who presented at lower stages (p = .001). Thirty-three
percent of patients with HCV were screened, and 19% of
HBV patients were screened. There was no difference in
mean MELD score in HCV vs HBV patients. (10.8 vs 10.3,
p = 0.10) Mean tumor size for HCV was 4.4 cm compared to
6.7 cm in HBV (p = 0.001).
Conclusions: Despite similar MELD scores between
patients with HCV and HBV, patients with HBV infection
presented with higher stages of HCC and larger tumors. It is
unclear if this is biologically driven or is related to a screening disparity between the two groups.
OP-I.25 IMPACT OF POSTOPERATIVE

MORBIDITY AND LIVER FAILURE ON
SURVIVAL OF PATIENTS WITH
RESECTED COLORECTAL LIVER
METASTASES
H. S. Ribeiro, M. C. Marques, R. S. De Souza,
I. C. De Farias, H. C. Freitas, F. J. Coimbra
Liver resection has become the best chance of cure for
patients with colorectal liver metastases. However, studies
have shown that postoperative complications could compromise long-term survival results. The aim of this study was to
determine the incidence and prognostic factors for postop 2015 The Authors
30
Abstracts
erative morbidity, including liver failure, and their impact on

long-term survival. Patients who underwent liver resection
for colorectal liver metastases between 1998 and 2012 were
analyzed. Any deviation from usual postoperative early
outcome was recorded as complications and stratified
according to the Clavien-Dindo classification. Three hundred
forty-six liver resections were performed in 285 patients.
Postoperative morbidity occurred in 57.8% of operations and
90-day mortality was 3.4%. Major complications (grade 3
and 4) occurred in 21.1% of hepatectomies. Twenty-three
(6.6%) patients had postoperative liver failure. After a
median follow-up period of 37.5 months, patients who had
major postoperative complications and/or postoperative liver
failure had statistically significant worse 3-year overall survival. In multivariate analyses, both factors were independent
predictors of survival (Major postoperative complications
HR = 2.1, p = 0.004, 95% CI 1.23.7; postoperative liver
failure HR = 1.9, p = 0.046, 95% IC 1.03.8). However,
these factors had no impact on disease free survival. These
findings may be explained by a decrease in life expectancy
observed in patients who experience life-threatening events.
Detailed clinical preoperative evaluation, careful surgical
techniques and meticulous postoperative care may improve
long-term survival results.
OP-I.26 GADOXETIC ACID (GA)

RELATIVE LIVER ENHANCEMENT
(RLE) ON PREOPERATIVE MAGNETIC
RESONANCE IMAGING (MRI) AS A
MARKER OF LIVER FUNCTION TO
PREDICT THE RISK OF
POST-HEPATECTOMY LIVER
FAILURE (PHLF)
A. Tremblay St-Germain1, A. Costa2, R. Smoot3,
K. Jhaveri2, S. Cleary1
1
Department Of Surgery, Toronto General Hospital,
University Of Toronto, Toronto, ON; 2Joint Department Of
Medical Imaging, Toronto General Hospital, University Of
Toronto, Toronto, ON; 3Department Of Surgery, Mayo
Clinic, Rochester, MN
Preoperative liver function and future liver remnant (FLR)
are typically used to predict the risk of PHLF. The hepatic
physiologic reserve and ability to hypertrophy are difficult to
predict. We investigated the use of contrast-enhanced MRI
with GA as a marker for liver function and potential predictor
of PHLF. We retrospectively analyzed all consecutive
patients who underwent major liver resection (3 segments)
and preoperative MRI-GA at our institution between October
2010 and December 2013. Mean RLE was calculated based
on regions of interest drawn of the liver on the unenhanced
and hepatobiliary phases. The associations between mean
RLE and PHLF according to the 50-50 and ISGLS criterias
were tested with univariate and multivariate logistic regression analysis. 68 patients (44 men; median age 60.5 years)
fulfilled the inclusion criterias: 47 CRLM, 14 HCC, 2
intrahepatic cholangiocarcinoma, 2 NETLM and 3 benign
diseases. 1 patient had PHLF according to the 50-50 criteria
and 13 patients had PHLF according to the ISGLS criterias (4
grade A, 8 grade B, 1 grade C and death). Mean RLE correlated with the presence and stage of fibrosis on histology
(p = 0,032 and p = 0,045). In logistic regression analysis,
mean RLE with a cut-off of 100%, in combination with FLR,
increased the ability to predict the presence of PHLF according to the ISGLS criteria. However, it did not reach statistical
significance as an independent predictor. MRI-GA and mean
RLE, in addition to FLR, can improve risk assessment for
PHLF after major liver resection.
OP-I.27 THE LEARNING CURVE EFFECT

IN LAPAROSCOPIC LIVER RESECTION
V. Villani, F. Sabbatino, R. Torabi, D. L. Berger,
K. K. Tanabe, K. D. Lillemoe, C. R. Ferrone
Massachusetts General Hospital Department Of Surgery,
Boston, MA
Background: The expansion of laparoscopic liver resection
(LLR) has been considerably slower compared to that of
other laparoscopic procedures. In this study we analyzed the
learning curve associated with LLR.
Method: Retrospective database analysis of consecutive
LLR performed between 3/076/14. Procedures were
divided in three chronological groups: A (03/0705/11,
43 pts), B (05/1105/13, 43 pts), C (05/1306/14, 42 pts).
Results: The three groups were comparable for patients
median age (57 yo; 59 yo; 59 yo), gender (25.6, 39.5, 45.2,%
2015 The Authors
HPB 2015, 17 (Suppl. 1), 181
Abstracts
male), percentage of patients who had a resection for a
malignancy (86.0%, 83.7%, 65.6%), and median operative
time (182 min, 190 min, 197 min). Complex procedures
(defined as: trisegmentectomies, left hepatectomies, right
hepatectomies or central hepatectomies) increased significantly over time (P = 0.007). One 90-day mortality occurred
in group A, while mortality in groups B and C was zero.
There was an improving trend in estimated blood loss (568.5,
563.9, 342.0 ml), in number of patients transfused (4, 3, 2
patients), conversion to open procedures (4, 2, 1 conversions), number of complications (8, 3, 4 complications,
Clavien grade III or higher). Median length of stay was
significantly reduced in groups B and C (6.4 days, 3.8 days,
3.8 days; P = 0.006).
Conclusions: Similar to other surgical procedures, LLR is
subject to a learning curve. Despite an increase in surgical
complexity, after the first 43 procedures there was an
improvement in blood loss, rate of conversion, morbidity,
and length of stay.
OP-I.28 SAFETY AND EFFICACY OF A

NEW ARTICULATING BIPOLAR
ENERGY DEVICE FOR PARENCHYMAL
TRANSECTION IN LAPAROSCOPIC
LIVER RESECTION
C. Dural, M. Akyuz, E. Aksoy, P. Yazici, F. Aucejo,
C. Quintini, C. Miller, J. Fung, E. Berber
Cleveland Clinic, Cleveland, OH
Background: The aim of this study is to assess the safety
and efficacy of a new articulating vessel sealer (VS) for
laparoscopic liver resection (LLR).
Methods: A new 5 cm, bipolar VS was used in 28 LLRs
(group 1). A comparison was made to 28 patients who underwent LLR (group 2) using other energy devices. T-test and
Chi square were used for statistics.
Results: Tumor type was malignant in 71% of patients in
group 1 and 89% of the patients in group 2 (p = 0.360).
Number and size of tumors, and resection type were similar
in both groups. In group 1, less number of adjunctive devices
(i.e. energy, clip appliers, staplers) were used (median 2)
versus group 2 (median 3, p = 0.032). Staplers were used in
28% (n = 8) of cases in group 1 and 54% (n = 15) in group 2,
with fewer number of cartridges fired in group 1, when used
(median 1.5 vs 4, respectively, p = 0.005). Parenchymal
transection time (28.2 3.5 minutes vs 55.2 4.1, respectively, p < 0.001) and total operative time (200.1 13.7 vs
242.7 14.4, respectively, p = 0.036) were shorter for group
1. Morbidity was 11% (n = 3) in group 1 and 18% (n = 5) in
group 2 (p = NS). Intraoperative costs were an average of $
3000 less in group 1 versus group 2 (p = 0.0029).
Conclusion: This study demonstrates the safety and efficacy
of a new energy device for LLR. The data suggests a potential benefit of this device to reduce operative time and
decrease costs by facilitating parenchymal transection.
31
OP-I.30 TREATMENT OF RECURRENCE

AFTER RESECTION OF
IN CIRRHOTIC LIVER
Z. Rong, Y. Collin, S. Turcotte, M. Dagenais,
R. Letourneau, M. Plasse, A. Roy, R. Lapointe,
F. Vandenbroucke-Menu
Centre Hospitalier De LUniversitaire Montreal (CHUM),
Montreal, QUEBEC
Background: High recurrence rates after liver resection for
hepatocellular carcinoma (HCC) remain problematic and no
consensus exists on the optimal management of recurrence.
Methods: A retrospective analysis was performed on cirrhotic patients with resected HCC in our center between
1992 and 2013. Survival and recurrence outcomes were
analyzed.
Results: Ninety-four patients with Child A (90 patients;
95.7%) or B (4 patients; 4.3%) cirrhosis were included.
There were 79 males with a mean age of 60.8 11.2 years.
Major hepatectomy was performed in 43 patients (45.7%).
Median HCC size was 3.5 cm (0.816.0 cm) and 79 patients
(84.0%) had a single tumor. Forty-three patients (45.7%)
experienced recurrent disease, mostly intrahepatic (36
patients; 83.7%). Overall and disease-free survivals were
70.0% and 48.4% at 3 years, and 61.6% and 41.6% at 5 years.
Inferior overall and disease-free survivals were significantly
associated with positive margins (p = 0.005 and p < 0.001),
multiple tumors (p = 0.018 and p = 0.03), and multinodularity (p = 0.007 and p < 0.001). Age (<65 years;
p = 0.03), tumor size (<3 cm; p = 0.02) and lower T stage
(p = 0.02) were linked with longer overall survival. Invasion
of adjacent organs (p < 0.001) and major vascular structures
(p = 0.01) were associated with recurrence. Thirty-three
patients with recurrent HCC received treatment, which
included chemotherapy (13 patients), chemoembolization
(12), radiofrequency (8), alcoholization (3), repeat resection
(4), or transplantation (3). Overall survival was significantly
superior in patients with treated recurrence (p = 0.005).
Conclusions: Recurrence is frequent after HCC resection,
particularly in patients with invasion of adjacent organs or
vascular structures. However, satisfactory 5-year survival
rates are achievable in patients with treated recurrent disease.
OP-I.31 MULTIMODAL TREATMENT OF

UNRESECTABLE HEPATOCELLULAR
CARCINOMA TO ACHIEVE COMPLETE
RESPONSE RESULTS IN IMPROVED
SURVIVAL
P. Newell1,2,5, R. Uppal3, Y. Wu1, H. Hoen1,5, J. T. Thiesing4,
K. Sasadeusz4, M. Cassera1, R. Wolf1,2, P. Hansen1,2,
C. Hammill1,2
1
Providence Cancer Center, Portland, OR; 2The Oregon
Clinic, Portland, OR; 3The Portland Clinic, Portland, OR;
4
The Radiology Group, Portland, OR; 5Earle A. Chiles
Research Institute, Portland, OR
Introduction: With technological advances, questions arise
regarding how to best fit newer treatment modalities, such as
transarterial therapies, into the treatment algorithm for
patients with hepatocellular carcinoma (HCC).
HPB 2015, 17 (Suppl. 1), 181
2015 The Authors

32
Abstracts
Methods: Between 20052011, 128 consecutive patients

initially treated with transarterial radioembolization or
chemoembolization using drug-eluting beads were identified. Response was graded retrospectively. Toxicity was
measured at 1,3, and 6 months after first and last treatments
by recording biochemical adverse events in bilirubin,
albumin, and INR.
Results: 53% of the patients were considered to have
advanced stage disease (BCLC stage C). 16% of patients had
an initial complete response, but with additional treatments,
this was increased to 36%. Patients with a complete response
as their best response to treatment had a median survival
(95% confidence interval) of 5.77 (2.58, the upper limit has
not yet been reached) years, significantly longer than those
whose best response was a partial response, 1.22 (0.84, 2.06)
years and those with stable disease as their best response,
0.34 (0.29, 0.67) years. Repeated treatments did not increase
the risk of toxicity.
Discussion: This retrospective review of patients treated for
intermediate and advanced stage HCC revealed a significant
survival advantage in patients who achieved a complete
response. We did not demonstrate superiority of one modality over the other, but did show that the two could be used
sequentially without accumulating significant toxicity. These
data support use of a multi-modality approach to intermediate and advanced stage HCC, combining liver-directed treatments as necessary to achieve a complete response.
OP-I.32 HEPATIC RESECTION FOR

DISAPPEARING LIVER METASTASIS:
A COST-UTILITY ANALYSIS
G. Spolverato1, A. Vitale2, A. Ejaz1, D. Cosgrove1,
D. Cowzer1, U. Cillo2, T. M. Pawlik1
1
Johns Hopkins Hospital, Baltimore, MARYLAND;
2
Universita Di Padova, Chirurgia Epatobiliare E Trapianto
Epatico, Padova, PADOVA
Introduction: Data on cost-effectiveness and efficacy of
hepatic resection(HR) for colorectal liver metastasis that
2015 The Authors
disappear after systemic chemotherapy(sCT) are lacking. We

estimated the cost-effectiveness of HR plus 6 months of sCT
in case of recurrence(strategy A) relative to surveillance and
6 months of sCT in case of recurrence(strategy B) for
patients with colorectal disappearing liver metastasis(DLM).
Methods: Through a Markov model three base cases were
evaluated involving a 65 year-old patient with three lesions in
the right hemi-liver who underwent 6 months of sCT and
1)had DLM based on MRI; 2)had DLM based on CT scan;
3)had also hepatic artery infusion(HAI) with subsequent
DLM based on CT scan.
Results: The NHB of strategy A(HR) versus strategy
B(surveillance) was negative(-1.7 QALMs) for base case 1.
In contrast, the NHB of HR was positive in base case 2(5.4
QALMs); the NHB of HR was positive for base case 3, but
the effect was much more modest(0.15 QALMs). The ICER
of strategy A versus B was highest for base case 1($105,216/
QALY) and lowest for base case 2($-18,768/QALY); the
ICER for HR versus surveillance was intermediate for base
case 3($48,924/QALY). Sensitivity analyses demonstrated
that HR was cost effective when compared with surveillance
when the rate of complete pathological response after 6
months of sCT was estimated to be <60%.
Conclusion: Surveillance of DLM is an acceptable strategy
when the diagnosis of DLM is made through MRI. NHB and
ICER favor HR when the presence of DLM is determined by
CT scan alone.
OP-I.33 SIMULTANEOUS RESECTION OF

PRIMARY COLORECTAL CANCER AND
SYNCHRONOUS LIVER METASTASES:
A POPULATION-BASED STUDY
1
Background: The role of combined resection of primary

colorectal cancer (CRC) and synchronous liver metastases
(LM) is gaining interest. Here we describe management and
HPB 2015, 17 (Suppl. 1), 181
Abstracts
outcomes of patients in the general population managed with
simultaneous or staged resection of the primary tumor and
synchronous CRC LM.
Methods: All cases of CRC in Ontario who underwent
surgical resection of LM in 20022009 were identified using
the population-based Ontario Cancer Registry. Synchronous
disease was defined as having resection of CRC LM
within 12 weeks of surgery for the primary tumor. Pathology
reports were reviewed to identify extent of disease and
surgery.
Results: During 20022009, 1711 patients underwent resection of CRC LM; pathology reports were identified for 1252
cases. 283 patients had synchronous disease; 116 (41%)
patients had simultaneous resections and 167 (59%) had a
staged resection. For the simultaneous and the staged groups,
mean number of liver lesions resected was 1.7 and 2.3
(p < 0.001), mean size of the largest lesion was 3.1 and
4.7 cm (p < 0.001), major hepatic resection (3 Couinaud
segments) rate was 26% and 76% (p < 0.001) and the R1
resection margin rate was 10% and 8% (p = 0.46), respectively. 30- and 90-day post-operative mortality rates for simultaneous and staged groups were 0.9% and 2.4% (p = 0.65)
and 3.5% and 4.2% (p = 1.00), respectively.
Conclusions: Simultaneous resection of synchronous CRC
LM is common in routine clinical practice. Compared to a
staged approach, patients undergoing simultaneous resections had fewer and smaller liver metastases, less aggressive
resections and comparable post-operative mortality.
OP-I.34 COMPARATIVE ANALYSIS OF

LAPAROSCOPIC RESECTIONS OF
POSTEROSUPERIOR SEGMENTS IN
SEMIPRONE POSITION VS.
LAPAROSCOPIC LEFT LATERAL
SECTIONECTOMY IN SUPINE POSITION.
DO WE NEED TO REDEFINE THE
DEFINITION OF A MAJOR
LAPAROSCOPIC LIVER RESECTION?
M. DHondt1, E. Yoshihara1, D. Devriendt1,
F. Vansteenkiste1, H. Pottel2
1
Dept. Of Digestive And HPB Surgery, Groeninge Hospital,
Kortrijk, W-VL; 2Interdisciplinary Research Center,
Catholic University Leuven, Campus Kortrijk, Kortrijk,
W-VL
Introduction: The Louisville-statement defined laparoscopic resections of posterosuperior segments (LPSS) as
major hepatectomies. It has been shown that LPSS, are associated with a good field of view,lower conversion rate and
less blood loss when performed in semiprone position. All
patients whom underwent LPSS at our center were positioned in semiprone since August 2011. The aims of this
study were to assess differences in perioperative outcomes
between laparoscopic left lateral sectionectomies (LLLS)
performed in supine position and LPSS in semiprone.
Methods: We reviewed a prospectively collected singlecenter database of all liver resections performed between
August 2011 and August 2014. LLLS and LPSS were compared with respect to demographics and perioperative
outcomes.
Results: Thirty nine patients underwent LLLS(n = 19) or
LPSS(n = 20). There were no differences in demographics
HPB 2015, 17 (Suppl. 1), 181
33
(table) or maximal tumor diameter (p = 0.7569). There were

no conversions. Pringle manoeuvre was not used in both
groups.There was no difference in peroperative central
venous pressure. Operative time in the LLLS group was 100
(60160) min and 150 (100270) min in the LPSS group
(p = 0.0037) with median intra-operative blood loss in the
LLLS group of 50(0550) ml versus a larger 150(50700) ml
(p = 0.0191) for patients receiving LPSS. No patients
required transfusion. Intraoperative and postoperative complication rate was similar in both groups. Mortality rate was
nil in both groups. Median hospital stay was 6 days in both
groups (p = 0.6382).
Conclusion: LPSS in semiprone can be performed with
similar clinical outcomes as a minor laparoscopic liver resection except for longer operative time and larger intraoperative
blood loss without the need for transfusion.
OP-I.35 ASSOCIATING LIVER

PARTITION AND PORTAL VEIN
LIGATION IN STAGED HEPATECTOMY
(ALPPS) IN SCANDINAVIA. A
TRI-INSTITUTIONAL INTRODUCTORY
FEASIBILITY STUDY
B. I. Rosok1, E. Sparrelid2, B. Bjrnson3, B. A. Bjrnbeth1,
B. Isaksson2, L. Lundgren3, E. Pomianowska1,
T. Gasslander3, P. Sandstrm3
1
Section For HPB Surgery, Oslo University Hospital, Oslo,
NONE; 2Department Of Surgery, Stockholm, NONE;
3
Surgical Department,, Linkg, NONE
Introduction: ALPPS has been introduced as an alternative
to conventional portal vein embolization or ligation (PVE/
PVL) in patients with technically resectable liver tumors but
insufficient future liver remnant. Initial experiences however,
indicated that the complication rate and perioperative mortality following ALPPS exceeded that of PVE/PVL.
Materials and Methods: Thirty patients (19 males 11
females) were operated during a 5 month period at our three
institutions. Underlying diagnoses were colorectal liver
metastases (n = 23), Cholangiocarcinoma (n = 4), HCC
(n = 2) and Carolis syndrom The number of lesions varied
from 120. None of the patients had underlying liver disease.
2015 The Authors
34
Abstracts
Results: All patients completed the 2-stage procedure at a

median of 8 days (715) following procedure 1. Twenty-three
patients had extended right hepatectomies, wheras the remaining 7 had conventional right hemihepatectomies. Median
hospital stay after the second procedure was 9.5 days (250).
No perioperative mortality was observed. Complications
according to Clavien-Dindo grading was 9 grade 1, 9 grade 2,
4 grade 3A and 3 patient with a grade 3b complication.
Discussion/Conclusion: ALPPS may be an alternative to
PVE/PVL in some patients. An acceptable complication rate
can be obtained if certain selection criteria are met. Many
patients have early recurrences following ALPPS, and our
short and intermediate oncological results will be presented.
Randomized controlled trials are necessary to define what
patients may benefit from ALPPS. A multicentre Scandinavian trial, LIGRO (ClinicalTrials.gov NCT02215577) has
therefore been initiated and has been enrolling patients since
May 2014.
OP-I.36 LIVER ABSCESS: A REVIEW

OF MANAGEMENT AND CLINICAL
OUTCOMES AT WESTERN
HEALTH, AUSTRALIA
OP-I.37 FACTORS PREDICITNG

OUTCOMES IN NON-TRAUMATIC
EMERGENCY HEPATECTOMYA
NSQIP ANALYSIS
N. P. Kohli, J. Choi, S. T. Chan, V. Usatoff

Western Health, Footscray, VICTORIA
J. A. Parikh1, S. Anantha Sathyanarayana1, S. Bendix1,

M. J. Jacobs1, R. Kather2, I. S. Rubenfeld1
1
St.John Providence And Providence Park Hospitals,
Southfield, MI; 2Henry Ford Hospital, Detroit, MI
Background: There is no clear consensus in the literature

about definitive liver abscess management. This is the first
study aimed to evaluate management and outcomes in
Melbournes western suburbs.
Study Design: A retrospective review of patients with pyogenic and amoebic liver abscess(s) captured using ICD-10
coding over the past five years at Western Health. Primary
outcomes were success of management, predictive factors for
surgical management, microbial aeitology of abscess and
major complications (Clavien grade III to V).
Results: Sixty-five patients (43 : 22 male to female) with
median age 61 (50.7572.00 Interquartile range; IQR). Fourteen (22%) were successfully managed with antibiotic treatment. Radiologically guided percutaneous drainage was
performed in 41 (63%) cases with median time to drainage of
5.5 (111 IQR) days. It was successful in 32 of 41 cases
(78%). Seven patients (11%) underwent surgery, of which 3
had failed drainage. Surgical predictive factors were found to
be CRP of >100 on admission (Odds ratio 30.750, 95% CI
3.25291.31, p = 0.003). The median length of stay was 15
(10.525.0 IQR) days. Forty-two patients had diagnostic
investigations for the cause of their abscess (see Fig. 1).
There were three deaths during this period of which two were
due to sepsis in context of liver abscess.
Conclusion: This study demonstrates that management with
intravenous antibiotics and radiologically guided percutaneous drainage is successful in majority of cases. There were
seven cases requiring surgery and a CRP >100 was shown to
be predictive for need of surgery.
2015 The Authors

Introduction: Although non-traumatic emergent hepatectomies are rarely indicated, their burden to healthcare system
in terms of utilization of resources is likely to be high, though
has never been quantified.
Methods: Using the ACS-NSQIP participant use files
for 20052012, we identified hepatic resections by Current
Procedural Terminology (CPT) code and segregated all
non-traumatic hepatectomies into 2 groups: Emergent Hepatectomy (EH) and Non-emergent Hepatectomy(NEH).
Preoperative, intraoperative and postoperative factors were
analyzed to identify predictors of complications and
mortality.
Results: Of the 13227 non-traumatic hepatectomies from
the NSQIP data, 137 emergency hepatectomies were identified. African Americans required significantly increased EH
(2.0% vs 0.92%, OR 2.2, p < 0.001). The most common
diagnosis for EH overall, was primary and secondary malignant neoplasm of the liver (38%; n = 33). Preoperative and
perioperative transfusion requirements were higher in the EH
group compared to NEH (17.5% vs 0.49%, OR 42.5,
P < .001; and 52.7% vs 26.4%, OR 3.1, P < .001). Patients in
the EH group were significantly more likely to experience a
Clavien 4 complication (19.7% vs 7.2%, OR 3.2, p < .001).
Mortality rate was higher in the EH group compared to
NEH(8.8% vs 2.5%, OR 3.7, P < .001).A multivariate logistic regression analysis revealed ASA score, ascites, and emergent indication as poor outcome indicators. Surprisingly age
and length of operation were not significant factors.
(Table 1).
Conclusion: Emergent hepatectomy has a significantly
higher perioperative blood transfusion requirement, with
increased morbidity and mortality rate. ASA score, ascites,
and emergent indication as poor outcome indicators, while
age and length of operation were not significant factors.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
OP-I.38 DETECTION OF INVISIBLE

LIVER TUMORS USING REAL-TIME
VIRTUAL SONOGRAPHY
A. Miyata, Y. Mise, T. Aoki, J. Kaneko, Y. Sakamoto,
K. Hasegawa, Y. Sugawara, N. Kokudo
The Univercity Of Tokyo Hospital, Tokyo, TOKYO
Introduction: Real-time virtual sonography is an innovative
imaging technology that synchronizes an intraoperative
ultrasonography (IOUS) with preoperative computed tomography (CT). We validated the effectiveness of the navigation
system in 2 cases to locate liver tumors, which had been
found in preoperative images and were difficult to detect
using conventional intraoperative inspection.
Case 1: A 65-year-old man had a metastatic liver tumor
from renal cell carcinoma. Preoperative CT revealed that the
tumor was 6 mm in size, located in the deep segment 6.
Intraoperatively, the tumor was not detected using plain and
enhanced ultrasonography. Real-time virtual sonography
projected the CT image of the tumor on IOUS image, which
helped us to add mobilization of the liver. After mobilizing
the liver towards the tumor location, we could find and resect
the tumor using additional enhanced IOUS.
Case 2: A 58-year-old man had a 6 mm-hepatocellular carcinoma located in segment 4. He had a previous history of
left lateral sectionectomy for hepatocellular carcinoma. The
recurrent tumor located in the previous cut surface of the
liver was not found by conventional IOUS because of the
severe adhesion. We used real-time virtual sonography,
which navigated us to dissect the adhesion towards the tumor.
Conclusions: A novel navigation system using real-time
virtual sonography is helpful to locate small tumors that are
difficult to find using conventional intraoperative inspection.
HPB 2015, 17 (Suppl. 1), 181
35
OP-I.39 THE NEED FOR

HEPATOPANCREATOBILIARY
SURGEONS: ARE THE COMMUNITY
HOSPITALS UNDERSERVED?
S. Anantha Sathyanarayana, S. Randhawa, P. Annigeri,
G. Marshall, E. Negussie, M. J. Jacobs, J. A. Parikh
St.John Providence And Providence Park Hospitals,
Southfield, MI
Introduction: Surgical educators have recently questioned
if too many Hepato-Pancreato-Biliary (HPB) surgeons are
being trained. While academic centers may be saturated,
many community hospitals may be underserved. Thus, we
sought to determine the need for an HPB surgeon at a tertiary
care community hospital.
Methods: All abdominal computed tomography (CT) scans
from February 2014 to May 2014 performed at a community
teaching hospital were reviewed and scans with pertinent
HPB pathology were isolated.
Results: A total of 389 CT scans having pertinent HPB
pathology were identified from 3500 scans, for which an
HPB surgeon consultation would be appropriate (Table 1).
Out of the 291 patients with liver specific pathology, 17
patients had hepatic cysts >4 cm in size, 76 had a solid mass.
Eighty three patients were found with pancreatic pathology,
out of which 17 patients had cystic lesions >1 cm, 21 of them
had a solid mass, 13 had non-specific main duct dilatation
and 9 had chronic pancreatitis. Fifteen patients had biliary
pathology including 11 with biliary ductal dilatation, one
choledochal cyst and one extrahepatic bile duct stricture. For
the 3 month study period, a total of 178 patients with significant HPB pathology were identified and the projected
volume of patients will be over 700 for a period of 1 year that
would require an HPB surgeon consultation.
Conclusions: The national need for HPB surgeons should
be re-evaluated based on the workload at the community
hospital setting and cannot be based on saturation at the
academic hospitals.
2015 The Authors

36
Abstracts
4 (8.2%) versus 0 in the pre-POH group. Deaths were due to
amioderone related pulmonary fibrosis (1), cardiac arrest (1),
and severe multi-organ failure related to a leak(1) and a
postoperative bleed (1).
Conclusion: The introduction of a hospitalist who specializes in perioperative management of high-risk surgical
patients was associated with a decrease in 30-day readmissions and a similar overall complication rate, even in the
setting of a higher mean ASA. There was a trend toward a
higher mortality rate in the post-POH cohort. More study is
required to understand the overall quality and financial
impact of POH co-management.
OP-I.40 EVALUATING THE IMPACT OF

ADDING PERIOPERATIVE HOSPITALIST
CO-MANAGEMENT ON OUTCOMES
FOR PATIENTS UNDERGOING
OP-I.41 PERSONALITY TRAITS

COMMON AMONG
HEPATO-PANCREATO-BILIARY
SURGEONS AND THEIR RELATIONSHIP
TO JOB SATISFACTION
M. Brown, P. M. Campbell, R. F. Wolf, W. C. Johnston,

M. A. Cassera, C. W. Hammill, P. H. Newell, P. D. Hansen
Providence Portland Medical Center, Cancer Center,
Portland, OREGON
S. N. Osayi1, L. Yu2, J. Drosdeck1, C. E. Ellison1,

M. Bloomston1, C. Schmidt1, M. Dillhoff1, S. Weber3,
P. Muscarella1
1
Department Of Surgery, The Ohio State University Wexner
Medical Center, Columbus, OH; 2Center For Biostatistics,
The Ohio State University, Columbus, OH; 3Department Of
Surgery, University Of Wisconsin, Madison, WI
Introduction: We hypothesize elderly patients, with multiple comorbidities, undergoing high risk surgical procedures
will benefit from pre- and postoperative co-management by a
hospitalist who specializes in this field. We report clinical
outcomes for two cohorts of patients undergoing pancreaticoduodenectomy (PD), pre and post introduction of a
perioperative hospitalist (POH) program.
Methods: Data was collected retrospectively on 89 consecutive patients undergoing PD between 2012 and 2014.
Analysis was performed on 40 patients prior and 49 patients
after the introduction of the POH program. Groups were
compared by chi-square and T-test.
Results: Results are summarized in the table below. 14
patients in the post-POH cohort were not seen by the POH
due to patient selection, distance and transportation issues.
Although the ASA was significantly higher in the post-POH
group, overall complication rates were similar. 30-day readmissions were also significantly lower in the post-POH
group. There were 4 deaths observed in the post-POH group,
2015 The Authors
Background: Personality may influence career choice, and

can predict job satisfaction. The purpose of this study was to
evaluate the personality traits of hepato-pancreato-biliary
(HPB) surgeons and to determine whether they correlate with
job satisfaction.
Methods: A web-based survey of surgical members of the
AHPBA was conducted. Personality traits and job satisfaction were assessed using the Big Five Inventory and the Brief
Index of Affective Job Satisfaction instrument.
Results: One hundred and thirty-six HPB surgeons completed the survey, 22% response rate (86.8% male, 70.7%
>40 years old, 85.3% completed fellowship training, 50%
>10 years in practice, and 76.5% academic practice). HPB
surgeons scored higher on extraversion and conscientiousness, and lower on neuroticism (p < 0.001) compared to a
normal population sample (n = 71,867). High extraversion
HPB 2015, 17 (Suppl. 1), 181
Abstracts
p = 0.02) and low neuroticism (p < 0.001) independently correlated with job satisfaction. Job satisfaction was higher
among females (p = 0.004). While 91.9% of respondents
indicated that they would choose the specialty again, only
53.7% would recommend it to their child/family. Those who
would choose the specialty again were less neurotic
(p = 0.039) and more satisfied with their job (p = 0.003).
Additionally, those who would recommend the specialty
were more agreeable (p = 0.001), more satisfied (p = 0.002),
have been in practice longer (p = 0.006), and were more
likely to choose the specialty again (p < 0.001).
Conclusion: Extraversion and neuroticism correlate with job
satisfaction among HPB surgeons. Furthermore, female HPB
surgeons appear to have higher levels of job satisfaction.
These findings may aid in the recruitment of HPB trainees and
may have implications for job performance and patient care.
OP-I.42 IMPACT OF FOCUSED NURSING

EDUCATION CURRICULUM IN THE
CARE OF HEPATO-PANCREATOBILIARY (HPB) SURGICAL PATIENTS
J. Drummond, C. Aviles, A. Cochran, E. Baker,
R. Seshadri, J. Martinie, D. Iannitti, R. Swan
Carolinas Healthcare System, Charlotte, NC
Nurses provide point of contact care for HPB surgery
patients. To improve patient education, nurse-physician communication, and nursing comfort with complex HPB patient
care, we offered a six hour nursing targeted, educational
course, to all nurses from inpatient units at a large, tertiary
care hospital. Topics included anatomy, pathology, surgical
procedures, nutrition, and pre/post-operative care of the HPB
surgical patient. Two weeks prior to the course, a 30 question
pre-test was distributed to all participants: 20 questions
regarding HPB disease processes, 4 regarding the participants background, and 6 describing comfort level, physician
communication, and experience caring for HPB patients. The
same test was given to participants following the course and
scores were compared. Descriptive statistics were performed,
survey results tallied. 59 nurses participated in the course: 50
completed the pre-test and 25 completed the post-test. 50%
of respondents reported nursing experience of less than 5
years. 68% reported working with HPB surgery patients over
half of their shifts. Average pre-test score was 11.1 (55.6%)
and post-test was 12.5 (62.5%), an increase of 13.5%
(p < 0.01). Nursing confidence, comfort, and communication
in the care of HPB surgery patients increased by 37.1%
(p < 0.01). The percentage who felt completely confident in
answering patient questions regarding HPB diseases more
than doubled (38.9% to 84.1%). Our focused HPB nursing
core curriculum course was associated with improved understanding of HPB anatomy and disease processes and
increased nursing confidence in caring for HPB surgical
patients. We hope this will translate to improved patient care
and nursing-physician communication.
HPB 2015, 17 (Suppl. 1), 181
37
OP-I.43 OPEN DATA FOSTERS QUALITY

OUTCOMES FOR HPB SURGERY:
RESULT OF 1625 PANCREATIC AND
HEPATIC RESECTION
C. K. Chang, S. H. TEH, P. Fuchshuber, J. S. Choi,
P. D. Peng, B. L. Bolinger, C. E. Binkley, R. M. Ramirez,
G. B. Kazantsev, A. L. Spitzer, K. Kojouri, E. E. Rosas,
M. M. Mortenson, C. A. Perez, M. A. Schlieman
Kaiser Permanente Northern California
Hepatico-Biliary-Pancreas Collaborative, Walnut Creek,
CA
Introduction: Healthcare economics is driving hospital to
deliver improved pay for performance for all surgical programs. Much of the data regarding efficiency and performance is not actionable on an individual surgeon level as it is
kept mostly anonymous. We hypothesize that the development of the Center of Excellence initiative for HPB surgery
within KP integrated health care system which comprised of
21 medical centers, can improve surgical outcome by firstly
improve individual surgeon performance.
Methods: All HPB cases from 2008 to 2014 were retrospectively analyzed. Three-time periods were chosen, and data
were given to all HBP surgeons. The first set of data served as
a baseline to provide transparency of operative time (OR) and
length of stay (LOS). The second set of data was prospective
obtained during the initiation of program development The
third set of data showed the result of the implementation of
such program including a bi-weekly conference to discuss
multifaceted best practices.
Results: See Graph
Conclusion: Transparency of surgical data allows surgeons
to self-identify potential surgical outliers among their peers. A
continuous open discussion of best practice in terms of
detail surgical technique, intra-operative management and
post-operative allows a gradual transition of coaching away
from their personal preference. A critical review of data and
discussion of best practice allows for systematic change in
efficient surgical technique as well as discharge criteria.
OP-I.44 DUODENAL NEUROENDOCRINE

TUMORS LOCATION MATTERS
M. R. Sheikh, H. Osman, S. Cheek, S. Hunter,
D. R. Jeyarajah
Objective: Duodenal neuroendocrine tumors are rare. Historically, when feasible a less aggressive surgical approach is
always considered to treat these tumors. The aim of this study
was to identify factors associated with necessity for more
aggressive surgical procedures.
Method: All patients who underwent surgery for duodenal
neuroendocrine tumor between September 2005 and June
2014 have been identified retrospectively in our database.
2015 The Authors
38
Abstracts
Data collected included clinical presentation, operative findings and histopathological data.
Results: 18 patients were identified that underwent surgical
management for duodenal endocrine tumors. This included 2
patient with transduodenal excision (11%), 2 patients with
duodenal resection (11%), 6 patients had antrectomy (33%)
and 8 underwent pancreaticoduodenectomy (44%). On analysis, peri-ampullary location was the most common site of
duodenal endocrine tumors (n = 9, 50%). 77% (n = 7) of
peri-ampullary lesions led to pancreaticoduodenectomy. The
odds of having a pancreaticoduodenectomy is 10 times higher
when the lesion is in peri-ampullary location. 6 patients had
positive lymph nodes. The odds of having a positive lymph
node are almost 9 times higher when the lesion is in ampulla.
83% (n = 5) of tumors with positive lymph nodes were greater
than T1 stage. The odds of having positive lymph node is 3
times higher when lesion is greater than T1.
Conclusions: Ampullary location of neuroendocrine tumor
in duodenum is associated with higher odds of lymph node
positivity and need for treatment with more extensive procedures like pancreaticoduodenectomy.
Conclusions: Early post-operative hypophosphatemia is an

independent predictor of LRC. A simple LRC risk prediction
tool that includes this variable accurately identified low-risk
patients and may help identify those most likely to benefit
from enhanced postoperative recovery pathways.
OP-I.45 EARLY POST-OPERATIVE

HYPOPHOSPHATEMIA AS A NOVEL
PREDICTOR OF ANASTOMOTIC
FAILURE AFTER PANCREATIC
RESECTION: A RISK-PREDICTION TOOL
OP-I.46 SURVIVAL FOLLOWING

FOR STAGE 1A PANCREATIC
ADENOCARCINOMA IS NOT IMPROVED
BY MULTIMODALITY TREATMENT
E. Sadot, L. Roach, C. A. McIntyre, P. J. Allen,

A. A. Eaton, M. I. DAngelica, R. P. DeMatteo,
T. P. Kingham, Y. Fong, W. R. Jarnagin
Memorial Sloan Kettering Cancer Center, New York,
NEW YORK
K. T. Ostapoff, P. Thirunavukarasu, B. W. Kuvshinoff,

S. J. Nurkin, S. N. Hochwald
Roswell Park Cancer Institute, Buffalo, NY
Introduction: Leak-related complications (LRC) remain

serious potential sequela of pancreatic resection. Current
LRC risk assessment is inadequate and rarely affects management algorithms. Hypophosphatemia appears to correlate
with infective complications after some abdominal
operations. This study evaluates early post-pancreatectomy
hypophosphatemia as a predictor of LRC.
Methods: Consecutive patients who underwent pancreaticoduodenectomy or distal pancreatectomy were analyzed.
LRC were defined as pancreatic leak, fistula, or abscess; only
grade 2 or higher LRC were recorded. Postoperative serum
phosphate levels and other recognized LRC risk factors (duct
diameter, soft pancreatic parenchyma, high-risk pathology,
excessive blood loss, procedure type, and preoperative
chemotherapy) were analyzed. Factors significant on multivariate analysis were used to construct an LRC risk prediction
model.
Results: From 2011 through 2012, 465 patients were
included with a median age of 66 years. LRC (grade >2) were
recorded for 85 patients (18%). Univariate analysis identified
the following predictors of LRC: hypophosphatemia on postoperative day 3 (p = 0.006), small duct diameter (p = 0.007),
soft gland consistency (p = 0.002), and intra-operative
blood loss >400 cc (p = 0.01). Hypophosphatemia on postoperative day 3(OR = 2.2, CI: 1.14.5), soft gland
consistency(OR = 3.1, CI: 1.76), and intra-operative blood
loss >400 cc(OR = 2.3, CI: 1.34) remained significant on
multivariate analysis and were used to construct an LRC risk
prediction tool, which had a negative predictive value of 93%
and a c-index of 0.68(Table 1).
2015 The Authors
Introduction: Pancreaticoduodenectomy is an integral part

of pancreatic adenocarcinoma treatment. NCCN guidelines
recommend patients undergo multimodality therapy, but few
studies validate its significance in early stage patients.
Methods: Using NCDB from 19982006, patients who
underwent a pancreaticoduodenectomy were identified.
Patients with invasive histology and stage 1 disease were
included. X2 test was used for categorical variables. Median
survival was estimated using Kaplan-Meier method with logrank comparison and Cox regression.
Results: Over 8 years, 2,801 patients were identified.
Median overall survival for Stage 1A (n = 1047) and Stage
1B (n = 1776) was 31.8 and 23.2 months with a median
overall follow-up of 23.7 months (0172.9). Stage 1B
patients were more likely to have higher grade tumors
(p = 0.002), receive chemotherapy (p = 0.007) or radiation
(p = 0.002) and have positive margins (p < 0.0001). There
were no differences between groups with respect to lymph
node yield (LN) or type of treatment facility. For both stages,
chemotherapy, radiation, LN yield 17 nodes, age <70,
tumor grade and facility type were associated with an
improved overall survival on univariate analysis. However on
multivariate analysis, for Stage 1A patients only LN yield
17 was independently predictive of survival while chemotherapy, radiation, facility type, age and sex had no impact on
survival. For patients with Stage 1B, chemotherapy, age <70
and LN harvest 17 were associated with an improved
overall survival on multivariate analysis.
Conclusion: Despite guidelines recommending multimodality treatment for pancreatic adenocarcinoma, there is no
improvement in survival for patients with Stage 1A disease. A
surgery only approach should be considered for these patients.
HPB 2015, 17 (Suppl. 1), 181
Abstracts

6:30PM7:30PM
COCKTAIL VIDEO
PRESENTATION
VC.01 TOTAL LAPAROSCOPIC
CENTRAL PANCREATECTOMY WITH
PANCREATIGOGASTROSTOMY FOR
HIGH RISK CYSTIC NEOPLASM
L. Schwarz, J. B. Fleming, M. H. Katz, J. E. Lee,
T. A. Aloia, J. Vauthey, C. H. Conrad
UT MD Anderson Cancer Center, Department Of Surgical
Oncology, Houston, TEXAS
Background: Organ-sparing pancreatic resection is important in prophylactic surgery for cystic neoplasms. There is
controversy regarding the optimal surgical approach for pancreatic lesions in the neck or proximal body of the pancreas.
Central compared to distal pancreatectomy is technically
more challenging but preserves more functional pancreatic
tissue. Due to the prophylactic nature of the surgery and long
survival of patients with benign and borderline malignant
lesions, surgeons need to stratify greater importance to surgical morbidity and sparing pancreatic parenchyma.
Patient: The patient is a 59-year-old active woman with a
symptomatic cystic neoplasm of the pancreas exhibiting high
risk imaging features. The cyst of 2.2 1.8 cm in the body of
the pancreas was impinging on the portal venous confluence.
Technique: The patient was positioned in the French Position, the lesser sac was opened and the pancreatic body
exposed. A retropancreatic tunnel was created with staple
division of the neck. The body was mobilized off the
portal vein and splenic vessels transected. A retrogastric
pancreaticogastrostomy was sewn through an anterior
gastrotomy. The stent was delivered past the pylorus to
decrease pancreatic enzymatic activation.
Conclusion: Laparoscopic ultrasound helps in defining cyst
borders and minimal blood loss optimizes visualization
during the dissection. A minimally invasive pancreaticogastrostomy created through an anterior gastrotomy is
technically feasible and safe. This approach can minimize the
morbidity of prophylactic pancreatic surgery for patients with
cystic neoplasms. Nevertheless, it should not compromise
safety, oncologic completeness or an organ-sparing approach.
HPB 2015, 17 (Suppl. 1), 181
39
SATURDAY, MARCH 14, 2015,

7:30AM9:30AM
LONG ORAL F OUTCOMES
LO-F.01 COST VARIATION IN
LAPAROSCOPIC CHOLECYSTECTOMY
AND ASSOCIATION WITH OUTCOMES
ACROSS A SINGLE HEALTH
SYSTEM: IMPLICATIONS FOR
STANDARDIZATION AND IMPROVED
RESOURCE UTILIZATION
R. C. Fields, W. G. Hawkins, S. M. Strasberg, L. M. Brunt,
N. Mercurio, D. P. Jaques, B. L. Hall
Barnes-Jewish Hospital, Washington University School Of
Medicine, St. Louis, MO
Background: Payers and regulatory bodies are increasingly
placing emphasis on cost containment, quality/outcome
measurement, and transparent reporting. Significant cost
variation occurs in many operative procedures without a
clear relationship with outcomes. Clear cost-benefit associations will be necessary to justify increased expenditures in
the era of bundled payment structures.
Hypothesis: Operating room (OR) supply cost variation in
laparoscopic cholecystectomy (LCCK) is not associated with
improved outcomes.
Methods: All LCCKs performed within a single health
system over a one-year period were analyzed for OR supply
cost. Results were obtained for individual surgeons and
system hospitals. Costs were correlated with NSQIP
outcomes.
Results: From July, 2013June, 2014, 2,178 LCCKs were
performed by 55 surgeons at 7 hospitals. The median case
OR supply cost was $513 156. There was significant variation in cost between individual surgeons, hospitals, and
within an individual surgeons practice (Figure). There was
no significant correlation between cost and individual
surgeon volume, hospital, or NSQIP outcomes. The majority
of cost variation was explained by selection of trocar and clip
applier constructs.
Conclusions: Significant case OR cost variation is present in
LCCK across a single health system. Differences in cost are
not associated with individual surgeon volume or hospital and
there is no clear association between increased cost and
NSQIP outcomes. Placed within the larger context of overall
cost, opportunity exists for OR case standardization and
savings with no obvious risk for a reduction in quality of care,
which is critical in the era of bundled payment structures.
2015 The Authors

40
Abstracts
LO-F.02 ANALYSIS OF LYMPH NODE

POSITIVE PATIENTS IN SURGICALLY
TREATED INTRAHEPATIC
CHOLANGIOCARCINOMA: A REVIEW
OF THE NATIONAL CANCER DATABASE
Z. Jutric, C. W. Johnston, H. M. Hoen, P. H. Newell,
M. A. Cassera, C. W. Hammill, R. F. Wolf, P. D. Hansen
Providence Portland Cancer Center, Portland, OR
Introduction: The role of routine lymphadenectomy in the
surgical treatment of intrahepatic cholangiocarcinoma (ICC)
has been poorly defined. Recent studies have recommended
consideration of lymphadenectomy given its prognostic
implications. We aim to define predictive indicators of survival in patients with positive lymph nodes.
Methods: The National Cancer Data Base (NCDB) was
queried for patients who underwent surgical resection for
ICC between 1998 and 2011. Single predictor univariate
analyses were performed on 23 variables including demographics, tumor characteristics, surgery outcomes and adjuvant therapy details. Both single predictor univariate and
multivariate Cox proportional hazards survival analysis were
then performed on 160 patients identified to have positive
lymph nodes.
Results: Of 823 patients with complete data, 57% had at
least one lymph node examined. Median survival for lymph
node negative patients was 37 months versus 15 months for
lymph node positive patients, results shown. Other univariate
factors associated with decreased survival include male sex,
urban location, tumor size, grade and positive margin status.
2015 The Authors
In lymph node positive patients, multivariate analysis

showed poorer survival in the patients not receiving chemotherapy or radiation (HR 1.43, p = .006), tumor size >5 cm
compared to <5 cm (p = .018), and older age (p < .0001).
Lymph node positive patients <age 45 had a median survival
of 27 months.
Conclusions: Overall survival in most ICC patients with
lymph node metastases is limited. Adjuvant therapy should
be considered in these patients. Strategies to improve lymph
node staging deserve further study, as many of these patients
may not benefit from resection.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
41
LO-F.03 NATURAL HISTORY AND

TREATMENT TRENDS IN
SUBTYPES: INSIGHTS FROM A
NATIONAL CANCER REGISTRY
P. L. Jernigan, K. Wima, D. J. Hanseman, R. S. Hoehn,
A. Ertel, E. Midura, I. M. Paquette, S. A. Ahmad,
S. A. Shah, D. E. Abbott
Department Of Sugery, University Of Cincinnati School Of
Medicine, Cicinnati, OH
Introduction: Histopathological
advancements
have
enabled more sophisticated characterization of hepatocellular carcinoma (HCC), but the clinical significance of these
distinctions is incompletely understood. Our aim was to
investigate pathologic and treatment differences between
HCC variants.
Cancer Data Base (19982011) was queried to identify 784
patients with surgical management of six HCC subtypes
(1.8% of all HCC patients): fibrolamellar (FL, n = 206), scirrhous (Sc, n = 29), spindle cell (Sp, n = 20), clear cell (CC,
n = 169), mixed type (M, n = 291), and trabecular (T,
n = 69). Chi-square, Kaplan-Meier and Cox regression
analysis were used to identify associations between demographic, tumor and treatment-specific variables and clinical
outcomes, namely overall survival (OS).
Results: Patients with FL-HCC were younger than other
variants (median age 27 vs. 5461, p < 0.001), more commonly female (56.3%, p < 0.001), and less likely to receive a
transplant (3.66%, p < 0.001). Patients with FL- and Sp-HCC
presented more frequently with larger tumors (>5 cm,
p < 0.001) and node-positive disease (p < 0.001). Median OS
(years) of the six subtypes was: FL 2.99 (CI 2.084.55), Sc
4.12 (1.477.01), Sp 0.56 (0.131.04), CC 2.99 (2.084.55),
M 2.22 (1.612.7), and T 2.91 (1.888.86). On multivariate
analysis, better OS was associated with FL-HCC, lower
pathologic stage, node-negative disease, and liver transplant
(Table 1). Eighty-six patients (11%) received adjuvant
therapy, which was not associated with better OS.
Conclusion: These data represent the largest series of recognizable HCC variants, demonstrating distinct differences
in presentation and natural history. These findings can help
clinicians and patients discuss treatment decisions and prognosis for rare clinical entities.
HPB 2015, 17 (Suppl. 1), 181
LO-F.04 UNDERSTANDING DRIVERS OF

COST VARIATION FOR EPISODES OF
CARE AMONG PATIENTS UNDERGOING
HEPATOPANCREATOBILIARY SURGERY
G. Spolverato, A. Ejaz, N. Ahuja, K. Hirose,
M. A. Makary, C. L. Wolfgang, M. J. Weiss, J. Cameron,
T. Pawlik
Johns Hopkins Hospital, Baltimore, MD
Background: Understanding factors associated with variation in procedure-related costs may help identify means to
increase savings. We sought to define potential variation in
hospital charges associated with hepatopancreatobiliary
(HPB) surgery.
Methods: Patients who underwent a HPB procedure
between 20092013 were identified. Perioperative morbidity
was ascertained through ICD-9 codes. Total hospital
charges were tabulated for room&board, surgical/anesthesia
services, medications, laboratory/radiology services, and
other charges.
Results: 2,545 patients underwent either a pancreas(66.8%)
or liver/biliary(33.2%) resection. Mean total charges
for all patients were $42,357 $33,745(pancreas: $46,352
$34,932 vs. liver: $34,303 $29,639;P < 0.001). Major
morbidity(pancreas, range: 7%-18%; liver, range: 9%18%) and observed : expected(O : E) length-of-stay(LOS)
(pancreas, range: 0.671.64; liver, range: 1.063.35) varied
among providers(both P < 0.001). While a perioperative complication resulted in increased total hospital
charges(complication: $66,401 $55,124 vs. no complication: $39,668 $29,250;P < 0.001), total charges remained
variable even among patients who did not experience a
complication(P < 0.001). For example, mean total charges
for a pancreaticoduodenectomy ranged from $41,413$49,543 vs. $29,376-$43,420 for a partial hepatectomy
(Figure). Surgeons within the lowest quartile of O : E LOS
had lower total charges($33,879 $27,398) versus surgeons
in the highest quartile($49,498 $40,971)(P < 0.001).
Surgeons with the highest O : E LOS had higher across-theboard charges (operating room, highest quartile:
$10,514 $4,496 vs. lowest quartile: $7,842 $3,706;
medication, highest quartile: $1,796 $3,799 vs. lowest
2015 The Authors
42
Abstracts
quartile: $925 $2,211; radiology, highest quartile:

$2,494 $4,683 vs. lowest quartile: $1,424 $3,247;
laboratory, highest quartile: $4,236 $5,991 vs. lowest
quartile: $3,028 $3,804; all P < 0.001).
Conclusions: After accounting for in-hospital complications, total mean hospital charges for HPB surgery remained
variable by case type and provider. While the variation in cost
was associated with LOS, provider-level differences in
across-the-board charges were also noted.
LO-F.05 AN ECONOMIC ANALYSIS OF

SHOULD COSTS DRIVE CONSUMER
DECISIONS?
T. B. Tran, M. M. Dua, D. J. Worhunsky, G. A. Poultsides,
J. A. Norton, B. C. Visser
Stanford University School Of Medicine, Stanford, CA
Background: Consumer groups campaign for cost transparency believing that patients will select hospitals accordingly.
The aim of this study was to determine whether the cost of
a Whipple might be considered in choosing a hospital.
Methods: Using Nationwide Inpatient Sample Database,
we analyzed charges for patients who underwent pancreaticoduodenectomy (PD) from 20002010. Outcomes were
compared between high volume hospitals (HVH, >20/yr) and
2015 The Authors
low volume hospitals (LVH). The relationship of inflationadjusted charges and outcomes were stratified by uncomplicated PD (length of stay <14 days) vs complicated PD (>14
days).
Results: A total of 15,599 PD were performed in 1,186
hospitals at a median cost of $87,444 (IQR $60,015$144,869). While only 94 (8%) hospitals performed >20
PD/year, 57% of all PD were performed in these HVH. HVH
had shorter hospital stay (11 vs 15 days, p < 0.001) and
mortality (3% vs 7.6%, p < 0.001). PD performed at LVH
had higher median charges compared to HVH ($97,923 vs.
$81,581, p < 0.001). The cost of uncomplicated PD was significantly lower than a complicated PD ($67,238 (IQR
51,11291,401) vs $138,325 (IQR 95,206224,919),
p < 0.001). When comparing uncomplicated PD between
HVH and LVH, the median cost was very similar ($67,389 vs
$66,922). Among uncomplicated PD, a multivariate analysis
controlling for demographics and co-morbidities revealed
that cost did not affect the risk of mortality (OR 1).
Conclusions: The cost of a Whipple (even uncomplicated)
remains surprisingly variable. PD at HVH are associated
with better outcomes, which is reflected in lower charges.
But ultimately, patients should choose volume over price.
LO-F.06 THE IMPACT OF

PERIOPERATIVE BLOOD TRANSFUSION
ON SHORT-TERM OUTCOMES
FOLLOWING PANCREATECTOMY:
AN ANALYSIS FROM THE AMERICAN
COLLEGE OF SURGEONS NATIONAL
SURGICAL QUALITY IMPROVEMENT
PROGRAM (ACS-NSQIP)
J. Hallet1,2, A. L. Mahar3, M. Tsang2, C. H. Law1,2,
N. G. Coburn1,2, P. J. Karanicolas1,2
1
Division Of General Surgery, Sunnybrook Health Sciences
Centre Odette Cancer Centre, Toronto, ONTARIO;
2
Department Of Surgery, University Of Toronto, Toronto,
ONTARIO; 3Department Of Public Health Sciences,
Queens University, Kingston, ONTARIO
Background: While perioperative red blood cell transfusions (RBCT) are associated with worse outcomes in
colorectal surgery, their impact following pancreatectomy
HPB 2015, 17 (Suppl. 1), 181
Abstracts
remains unclear. We sought to examine the association
between RBCT and post-operative morbidity following pancreatectomy.
Methods: Using the ACS-NSQIP database, we identified
patients undergoing elective pancreatectomy from 2006 to
2012. Patients missing data on key variables were excluded.
We compared post-operative morbidity and length of stay
based on RBCT status using univariate and multivariate
analyses. A sensitivity analysis was conducted excluding
patients with higher baseline risk for RBCT.
Results: From 21,132 pancreatectomies, we included
14,322 patients of whom 1624 (11.3%) received RBCT.
Major morbidity (34.9% Vs. 21.6%; p < 0.0001) and mortality (15.7% Vs. 11.5%; p < 0.0001) were higher, and median
length of stay was prolonged (15.7 Vs. 11.5 days; p < 0.0001)
with RBCT. After adjustment for baseline characteristics
including comorbidities, malignant diagnosis, procedure,
and operative time, RBCT was independently associated
with increased major morbidity (Relative Risk RR 1.45;
p < 0.0001), post-operative infections (RR 1.30; p < 0.001),
thrombo-embolic events (RR 1.41; p = 0.01), cardiac events
(RR 2.41; p < 0.0001), respiratory failure (RR 2.60;
p < 0.0001), and mortality (RR 2.51; p < 0.0001). Length of
stay was prolonged with RBCT (adjusted mean estimate
1.22; p < 0.0001). Excluding patients with higher baseline
risk of RBCT did not substantially alter the results.
Conclusion: Perioperative RBCT is independently associated with worse short-term outcomes and prolonged length
of stay following pancreatectomy. This observation holds
true in patients with lower baseline risk of RBCT. Comprehensive multidisciplinary strategies to minimize and rationalize the use of RBCT are warranted.
43
study was to evaluate the oncologic outcomes of patients

with specified CRLM-EHD vs. CRLM-only and to determine the effect of PET-CT on disease-free (DFS) and overall
survival (OS).
Methods: This is a sub-study of a randomized trial studying
the effect of PET-CT before liver resection on surgical management of CRLM with or without EHD (limited to specified
resectable metastases: portal lymph nodes/lung/local recurrence) in patients with resected colorectal cancer from 2005
2013. Survival data adjusted for PET-CT was analyzed using
standard statistics.
Results: From 404 patients in the trial, 25 had EHD at
randomization [14 PET-CT group and 11 control (no
PET-CT)]. The most common EHD-site was lung (n = 18).
All EHD sites known at randomization were resected
(PET-CT and control group). After median follow-up of
36-months, median DFS for CRLM-EHD was 5.9-months
(95%CI: 3.611.7) and 16.2-months (95%CI: 13.718.9) for
CRLM-only [unadjusted-HR: 3.03 (95%CI: 2.004.59)]; the
estimated OS was similar between groups [36.7-months
(95%CI: 26.843.4) vs. 40.7-months (95%CI: 40.557)
respectively; unadjusted-HR: 1.68 (95%CI: 0.98, 2.89)].
After adjusting for the use of PET-CT, DFS for CRLM-EHD
remained significantly worse compared to CRLM-only
without significant differences in the adjusted OS.
Conclusions: Metastasectomy for specified and limited
CRLM-EHD is associated with similar OS to CRLM-only
despite a lower DFS. The use of PET-CT prior to complete
metastasectomy did not affect the time to recurrence or the
OS of patients with CRLM-EHD.
LO-F.08 TIMING, INCIDENCE, AND

RISK FACTORS ASSOCIATED WITH
UNPLANNED POSTOPERATIVE
HOSPITAL READMISSIONS IN THE
HEPATO-PANCREATICO-BILIARY
PATIENTS
LO-F.07 RESECTION OF COLORECTAL

CANCER LIVER METASTASES IN THE
SETTING OF EXTRAHEPATIC DISEASE;
RESULTS FROM A RANDOMIZED TRIAL
EVALUATING THE EFFECT OF PET-CT
P. E. Serrano1, C. Moulton2, C. Gu1, C. H. Law3, L. Ruo1,
K. Y. Gulenchyn1, D. Quan4, R. Fairfull Smith5,
D. W. Jalink6, M. Husien7, J. A. Julian1, M. N. Levine1,
S. Gallinger2
1
McMaster University, Hamilton, ON; 2University Health
Network, University Of Toronto, Toronto, ON; 3Sunnybrook
Health Sciences Centre And Odette Cancer Centre,
Toronto, ON; 4London Health Sciences Centre, London,
ON; 5The Ottawa Hospital, Ottawa, ON; 6Cancer Centre
Of Southeastern Ontario, Kingston, ON; 7Grand River
Regional Cancer Centre, Kitchener, ON
Introduction: Selected patients with colorectal cancer liver
metastases (CRLM) and extra-hepatic disease (EHD) are
considered for curative surgery. The main objective of this
HPB 2015, 17 (Suppl. 1), 181
G. C. Edwards1, L. Du1, J. R. Miller2, J. Ehrenfeld1,

H. R. Mir1, Y. Shyr1, A. A. Parikh1, N. B. Merchant1,
K. Idrees1
1
Vanderbilt University Medical Center, Nashville, TN;
2
Meharry Medical College, Nashville, TN
Introduction: As part of the Affordable Care Act, the Hospital Readmissions Reduction Program (HRRP), has started
to impose financial sanctions on hospitals with increased
readmission rates. The purpose of this study is to define the
incidence and identify peri-operative factors associated with
30 and 90-day readmission after hepatic (HR) and pancreatic
resections (PR).
Methods: HR and PR patients were retrospectively
reviewed over an 8 year period. Pre-operative factors [patient
demographics, ASA class, Charlson Comorbidity Index,
Elixhauser Comorbidity Index], intra-operative factors [Surgical Apgar Score (SAS), operative duration] and postoperative factors [postoperative major complications (PMC),
hospital length of stay (LOS), ICU LOS, discharge disposition] were evaluated. Multivariable Cox regression (MVR)
analysis was used to examine associations for hospital
readmission.
Results: 30 and 90-day readmission rates in 878 patients are
shown in the table. By MVR, PMC (HR = 8.9, 95%,
2015 The Authors
44
Abstracts
p = 0.0003), PR (HR = 2.3, p = 0.006), ICU LOS (HR = 1.1,

p = 0.02) and SAS (HR = 0.75, p = 0.02) were associated
with 30-day readmission. PMC (HR = 7.5, p = 0.0001) and
PR (HR = 2.1, p = 0.007) were also associated with 90-day
readmission. In patients without complications, ICU LOS
(HR = 7.0, p = 0.03), PR (HR = 2.4, p = 0.006) and SAS
(HR = 0.7, p = 0.04) were independently associated with
30-day readmission while only PR (HR = 2.1, p = 0.007) was
associated with 90-day readmission.
Conclusions: The 90-day readmission rate in HPB patients
remains high (18.8%) and the majority of readmissions
(88.8%) occur within 30 days of discharge. Identifying
factors associated with an increased risk of readmission is
critical in the development of interventions and resource
utilization to help reduce unplanned readmissions and
decrease costs within this population.
Conclusions: Psychiatric disease and substance abuse are

highly prevalent among veterans with HCC. Most patients
are surviving on very meager income. These profound socioeconomic and psychosocial problems must be recognized
when providing care for HCC to this population to provide
adequate treatment and surveillance.

7:30AM9:30AM
LONG ORAL G BASIC/
TRANSLATIONAL/EDUCATION
LO-G.01 TAUROURSODEOXYCHOLIC
ACID ALLEVIATES ISCHEMIA/
REPERFUSION INJURY IN STEATOTIC
MOUSE LIVER
C. D. Anderson, J. Zhang, N. Singh, W. Dorsett-Martin,
T. M. Earl
University Of Mississippi Medical Center, Jackson, MS
LO-F.09 MISSING THE OBVIOUS:

PSYCHOSOCIAL OBSTACLES IN
VETERANS WITH HCC
K. J. Hwa1,2, M. M. Dua1,2, S. M. Wren1,2, B. C. Visser1,2
1
Stanford University School Of Medicine, Stanford, CA;
2
Palo Alto Veterans Affairs Hospital, Palo Alto, CA
Introduction: Socioeconomic disparities in patients with
hepatocellular carcinoma (HCC) influence medical access
and treatment. However, in addition to socioeconomic barriers, the veteran population suffers from significant psychosocial obstacles. This study identifies the often overlooked
social challenges that veterans face while undergoing treatment for HCC.
Methods: We retrospectively reviewed prospectively gathered data regarding social/behavioral hardships for 100 veterans at the Palo Alto VA who had been treated for HCC
between 20092014 (50 consecutive patients who underwent
resection and 50 treated with intra-arterial therapy).
Results: Substance abuse history was identified in 96%,
with 34% actively abusing alcohol or drugs. Half were unemployed. Most patients survived on very limited income
(median $1340/mo, IQR $9002125); 36% on <$1000/mo,
37% between $10012000/mo, and 27% with >$2000/mo.
Fully, 37% were homeless at time of index treatment, which
was more common in those with the lowest income (62% of
<$1K/mo group, 32% of $12K/mo group, and 6% of
>$2K/mo group, p < 0.05). Psychiatric illness was present in
64/100 (64%) patients; among these the majority (59/64,
92%) received ongoing psychiatric treatment. The top two
diagnoses were depression and PTSD. Fifty-one percent of
all patients had been incarcerated. Transportation was provided to 23% of patients who would otherwise have been
unable to attend medical appointments.
2015 The Authors
Background: Tauroursodeoxycholic
acid
(TUDCA)
decreases endoplasmic reticulum (ER) stress, autophagy, and
cell death in cultured rat hepatocytes. We hypothesized that
TUDCA could reduce the injury caused by total warm
ischemia reperfusion (WIR) in steatotic mouse liver.
Methods: Male ob/ob mice underwent 100% hepatic warm
ischemia by clamping the portal triad for 30 minutes. For the
experiment group, 200 mg/kg TUDCA was injected IP 1
hour before the surgery. Animals were sacrificed at 12 hours
and 48 hours after reperfusion. Quantitative real time PCR
measured ER stress markers such as C/EBP homologous
protein (CHOP), glucose regulated protein 78 (GRP78),
protein kinase dsRNA-dependent-like ER kinase (PERK),
and activating transcription factor-6 (ATF6). Western blot
examined autophagy marker microtubule-associated protein
1 light chain 3 (LC3 II). ELISA determined interleukine-6
(IL6) levels (liver and serum).
Results: Compared to controls, WIR increased ER stress in
the liver [CHOP (3 fold, p = 0.004), GRP78 (4 fold,
p = 0.001), PERK (2 fold, p = 0.005), and ATF6 (1.5 fold,
p = 0.004)] at 12 but not 48 hours. LC3 II protein levels were
increased at both 12 (3 fold, p = 0.019) and 48 hours (4 fold,
p = 0.025). Serum IL6 levels were increased at 12 (40 fold,
p = 0.034) and 48 hours (33 fold, p = 0.034). TUDCA treatment decreased LC3 II at 12 (p = 0.018) and 48 hours
(p = 0.034), decreased serum IL6 at 12 (p = 0.025) and 48
hours (p = 0.025), and improved animal survival (median 26
hours vs 41 hours, p = 0.02). ER stress levels were not
changed.
Conclusion: TUDCA improves survival and reduces the
inflammation following WIR in steatotic liver through a
non-ER stress pathway.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
LO-G.02 CHARACTERIZATION OF A
PORCINE MODEL FOR ASSOCIATING
LIVER PARTITION AND PORTAL VEIN
LIGATION FOR STAGED
HEPATECTOMY (ALPPS)
K. P. Croome, S. A. Mao, J. M. Glorioso, S. L. Nyberg,
D. M. Nagorney
Background: Publications using the ALPPS procedure have
demonstrated a future liver remnant(FLR) growth of
40160% in only 69 days. The present study aimed to
develop and describe the first large animal model of ALPPS
that can be used for future studies.
Methods: A total of 13 female domestic swine were studied.
ALPPS stage 1 (portal vein division and parenchymal
transection) was followed by ALPPS stage 2 (completion left
extended hepatectomy) 7 days later. An abdominal CT scan
was performed immediately prior to ALPPS stage 1 surgery
and again 7 days later to assess hypertrophy immediately
prior to ALPPS stage 2 surgery. Blood samples as well as
tissue analysis were performed.
Results: On CT volumetric analysis mean size of the FLR
prior to ALPPS stage 1 was 21.4 1.8% and 39.8 4.6%
prior to ALPPS stage 2. Median degree of hypertrophy was
74.5% with a median kinetic growth rate of 10.6% per day.
Liver weights at autopsy correlated well with CT volumetric
analysis(p = 0.65). There was no significant difference in
mean lab values (AST,ALT,ammonia,INR or bilirubin) from
baseline until immediately prior to ALPPS stage 2. Post
ALPPS stage 2 there was a significant increase in INR from
baseline 1.1 0.1 and 1.6 0.1 (p = 0.005), respectively. No
post-operative deaths secondary to liver failure were
observed.
Conclusion: The present study describes the first reproducible large animal model of the ALPPS procedure. Degree of
hypertrophy and kinetic growth rate were similar to that which
has been demonstrated in human publications. This model
will be valuable as future laboratory studies are performed.
HPB 2015, 17 (Suppl. 1), 181
45
LO-G.03 PREOPERATIVE ANAEMIA AND

POSTOPERATIVE OUTCOMES AFTER
HEPATECTOMY: A RETROSPECTIVE
COHORT STUDY
S. Tohme, P. Varley, M. Khreiss, A. Tsung
University Of Pittsburgh, Pittsburgh, PA
Background: Preoperative anemia is associated with
adverse outcomes after surgery in general but outcomes after
hepatectomy specifically are not well established. We aimed
to assess the effect of preoperative anemia on 30-day postoperative morbidity and mortality in patients undergoing
major hepatectomies.
Methods: All elective hepatectomies for the period 2005
2012 recorded in the NSQIP database were evaluated. We
selected to study partial lobectomies, total left, total right,
and trisegmentectomies and exclude minor procedures. We
obtained anonymized data for 30-day mortality and morbidity, demographics, and preoperative and perioperative risk
factors. We used multivariate logistic regression to assess the
adjusted and modified effect of anemia, which was defined as
(hematocrit <39% in men and <36% in women), on postoperative outcomes.
Results: We obtained data for 13,198 patients, of whom
4,383(33.2%) had preoperative anemia. Postoperative mortality at 30-days was higher in patients with anemia than
those without anemia (odds ratio[OR]2.15,95%CI 1.70
2.71). Morbidity at 30-days was also higher in patients with
anemia (for any complication1.93,1.792.09; for serious
complications 2.05,1.902.22). After adjustment for predefined clinical and laboratory risk factors, postoperative
morbidity was higher in patients with anemia than in
those without anemia (adjusted OR any complication
148,1.361.61, serious complications 1.54,1.411.68). Postoperative mortality was similar in both groups after adjustment (1.09, 0.8361.433).
Conclusion: Preoperative anemia is independently associated with an increased risk of morbidity in patients undergoing hepatectomy. Therefore, it is crucial to readdress
preoperative blood management in anemic patients prior to
hepatectomy. Anemia was not an independent predictor of
mortality which may be due to the rare event of 30-day
mortality after elective hepatectomy.
2015 The Authors

46
Abstracts
LO-G.05 INTERACTION OF GLYCOGEN

SYNTHASE KINASE-3 AND NOTCH1 IN
PANCREATIC CANCER
M. Kunnimalaiyaan, S. Kunnimalaiyaan, T. Gamblin
Abstract: Glycogen synthase kinase-3 (GSK-3) can act as
either tumor promoter or suppressor by its inactivation
depending on the cell type. There are conflicting reports on
the roles of GSK-3 isoforms and their interaction with
Notch1 in pancreatic cancer. We hypothesize that GSK-3
stabilizes Notch1 in pancreatic cancer cells thereby promoting cellular proliferation.
Methods: Pancreatic cancer cell lines MiaPaCa2, PANC-1,
and BxPC-3 were treated with 020 M of AR-A014418
(AR). Cell growth was determined by MTT assay and LiveCell Imaging. The levels of Notch pathway members
(Notch1, HES-1, survivin, cyclinD1), phosphorylated
GSK-3 isoforms, and apoptotic markers were determined by
Western blot. Immunoprecipitation was performed to identify the binding of GSK-3 specific isoform to Notch1.
2015 The Authors
Results: AR-A014418 treatment had a significant dosedependent growth reduction (p < 0.001) in pancreatic cancer
cells compared to control. The growth suppression effect is
due to apoptosis. Importantly, reduction in GSK-3 phosphorylation leads to a reduction in Notch pathway members.
Over expression of active Notch1 in AR-A014418-treated
cells resulted in negation of growth suppression. Immunoprecipitation analysis revealed that GSK-3 binds to Notch1.
Conclusions: This study demonstrates for the first time
that the growth suppressive effect of AR-A014418 in pancreatic cancer cells is mainly mediated by reduction in
phosphorylation of GSK-3 with concomitant Notch1 reduction. GSK-3appears to stabilize Notch1 by binding and may
represent a target for therapeutic development. Furthermore,
down regulation of GSK-3 and Notch1 may be a viable
strategy for possible chemosensitization of pancreatic cancer
cells to standard therapeutics.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
LO-G.06 NOVEL CONCEPT OF

ELECTROCOAGULATION & TUMOR
CELL IMPLANTATION: CREATION OF
MINIMALLY INVASIVE ORTHOTOPIC
MURINE MODEL OF PANCREATIC
CANCER
J. S. Bhullar1, Y. Cozakov1, N. Varshney2, S. Bindroo1,
S. Chaudhary1, J. Tilak3, M. Decker3, M. Jacobs1,
V. K. Mittal1
1
Department Of Surgery, Southfield, MI; 2Department Of
Pathology, Toledo, OHIO; 3Department Of Patient Care
Research, Southfield, MI
Background: Orthotopic murine models of pancreatic
cancer represent an important tool for evaluating treatment
strategies. Several genetically modified mouse tumors and
xenograft models have been reported. Genetic models have
unpredictable growth & variable waiting period, while
orthotopic models are operative ones, difficult to create and
result in irregular metastasis. There is a constant endeavor to
create an orthotopic model which replicates the human
disease process.
Study Design: Orthotopic pancreatic tumors were induced
in 20 SCID mice using a novel technique. Low dose
electrocoagulation of pancreas under laparoscopic guidance
(using Coloview-mouse colonoscope) with thin electrode,
followed by injection of 0.1 cc BxPC3 pancreatic cancer
cells was done (n = 12, study group). Control mice underwent electrocoagulation alone (n = 4, group 1) and tumor cell
injection alone (n = 4, group 2). Mice were evaluated for
tumor growth and metastasis by necropsy (4 and 8 week for
experimental group; 8 weeks for control group).
Results: Tumors were detected in 11/12 mice in experimental group, 1/4 in control group 2, and none in control group 1.
Over time there was an increase in tumor growth, tumor
volume, lymphovascular invasion of pancreas, with metastasis to lymph nodes and surrounding organs.
Conclusions: We report a novel concept of tumor cell
implantation at site of electrocoagulation of pancreas. Combined with the minimally invasive technique, yields a
replicative orthotopic murine model of pancreatic cancer.
Our model is minimally invasive, easy to create, and overcomes the limitations of the existing models while questions
the possibility free floating tumor cell implantation at resection site.
HPB 2015, 17 (Suppl. 1), 181
47
LO-G.08 ADEQUACY OF HPB TRAINING:

POTENTIAL DISCONNECT BETWEEN
FELLOW AND PROGRAM DIRECTOR
PERCEPTIONS?
A. K. Bressan1, J. P. Edwards1, E. Dixon1, R. M. Minter2,
D. R. Jeyarajah3, S. C. Grondin1, C. G. Ball1
1
Department Of Surgery, Foothills Medical Center And The
University Of Calgary, Calgary, AB; 2Department Of
Surgery, University Of Michigan, Ann Arbor, MI;
3
Background: Hepatopancreatobiliary (HPB) fellowship
programs have undergone recent significant changes with
regard to training standards, case volume thresholds and
multimodality educational platforms. The goals of this study
were to (1) compare perspectives of residents and program
directors on perceptions of readiness to enter practice and (2)
identify core HPB procedures that require increased emphasis during training.
Methods: This survey targeted program directors (PDs) and
trainees participating in the Fellowship Council / AHPBA
pathway. Demographics, education, and career plans were
collected. A comparative analysis of PD and trainee opinions
on their confidence to perform thirteen core HPB procedures
was completed (p < 0.05).
Results: The response rate was 88% for both fellows (21/
24) and PDs (23/26). Amongst fellows, 72% believe there is
an excessive number of trainees, 81% aim to work in
university-based or academic institutions, 90% expect to
have an HPB practice combined with non-HPB cases, and
95% do not plan to pursue additional training. For all thirteen
HPB procedures, the volume of cases during training was
more often considered good or excellent by PDs than by
fellows. This difference reached statistical significance for:
major hepatectomies (PDs: 87% vs. fellows: 57%, p = .042);
pancreaticoduodenectomies (100% vs. 81%, p = .044); and
laparoscopic distal pancreatectomies (78% vs. 43%,
p = .029). Trainees also systematically rated their confidence
to perform HPB procedures lower compared to PDs perception (Figure 1).
Conclusions: This study provides insight into content
domains which may require additional attention during fellowship to achieve an appropriate level of proficiency and
confidence upon completion of training.
2015 The Authors

48
Abstracts
LO-G.09 WHAT TO EXPECT WHEN

YOURE EXPECTING A
HEPATOPANCREATOBILIARY
SURGEON: SELF-REPORTED
EXPERIENCE OF HPB SURGEONS FROM
DIFFERENT TRAINING TRACKS
S. G. Warner1, A. Alseidi2, J. C. Hong3, T. M. Pawlik4,
R. M. Minter1
1
Departments Of Surgery And Medical Education,
University Of Michigan Health System, Ann Arbor, MI;
2
Department Of Surgery, Virginia Mason Medical Center,
Seattle, WA; 3Department Of Surgery, Medical College Of
Wisconsin, Milwaukee, WI; 4Department Of Surgery, Johns
Hopkins Hospital, Baltimore, MD
Background: With a recent increase in fellowships offering
HPB training through multiple routes, prospective trainees
and employers must understand the differences between
available HPB training pathways. This study highlights selfreported fellowship experience and current scope of practice
across 3 different training pathways.
Methods: A survey was disseminated to 654 surgeons
active AHPBA members and recent graduates of HPB,
transplant-HPB, and surgical oncology fellowships using
SurveyGizmo. Descriptive statistics were calculated.
Results: 416 (66%) surgeons responded. Most respondents
were male (89%), and most (83%) practice in an academic
setting. Table 1 demonstrates fellowship operative experience
and current case mix in practice. MIS training was the most
commonly identified training deficiency, with 47% HPB, 49%
transplant, and 52% SSO-trained respondents in agreement.
Ultrasound was also a commonly identified training gap with
34% HPB, 40% transplant, and 25% SSO-trained respondents
in agreement. Non-HPB cases routinely performed in practice
were most commonly GI surgery and general surgery (56% &
49%) for HPB-trained respondents, transplant and general
surgery (87% & 21%) for transplant-trained respondents, and
GI surgery and non-HPB surgical oncology (70% & 28%) for
surgical oncology-trained respondents.
Conclusions: HPB surgery fellowship training experiences
vary by training pathway, though perceived deficiencies in
MIS and US training are common across all pathways.
Despite this variability, the ultimate scope of non-transplant
HPB practice is similar across training pathways. Thus,
selection of a training pathway may best be guided by the
desired training experience and planned focus of other
components of ones future practice.
2015 The Authors

LO-G.10 TEACHING PREOPERATIVE

PLANNING: A NEW EDUCATIONAL
INITIATIVE
N. Zilbert1, T. Lam2, S. Gallinger1, L. St. Martin2,
C. Moulton1,2
1
University Of Toronto Department Of Surgery, Toronto,
ON; 2The Wilson Centre, Toronto, ON
Background: Previous research has identified the importance that expert surgeons place on preoperative planning.
Currently the teaching and assessment of preoperative planning is limited. This video demonstrates a novel education
initiative to teach surgical trainees strategies for preoperative
planning for complex HPB procedures.
Methods/Results: This video reviews one preoperative
planning module for a case of a patient with a colorectal liver
metastasis requiring a left hemihepatectomy. There are five
screens that make up the module. The introductory screen
provides a brief clinical history. The second screen allows the
trainee to review the patients preoperative imaging. The next
screen displays two videos of attending surgeons reviewing
the same preoperative imaging. The fourth screen displays
intraoperative video clips narrated by the operating surgeon.
Each video clip focuses on an issue that one of the attendings
on the preceding screen. The final screen shows a video from
the operating surgeon reviewing the key learning points for
the case to reinforce these for the trainee.
Conclusions: In conclusion this video demonstrates a novel
strategy for teaching preoperative planning for HPB surgery.
Following the completion of this module the trainee should
have a deeper understanding of the issues and considerations
that are relevant for left hemihepatectomies. However the
ultimate strength of this program is that several different left
hemihepatectomy cases will be presented in the same format,
so collectively the trainee will gain a richer and more complete understanding of the procedure. The process will be
repeated for the other index procedures for HPB surgery.
LO-G.11 UNDERSTANDING SURGICAL

ANATOMY OF THE LIVER: THERES AN
APP FOR THAT
L. M. Postlewait, M. Konomos, J. A. Matlock, T. White,
K. A. Delman, S. K. Maithel
Carlos And Davis Center For Surgical Anatomy And
Technique, Emory University, Atlanta, GA
Introduction: Surgical anatomy of the liver is complex and
difficult to visualize. The two-dimensional renderings available for trainees make translation to practical application
challenging. Our aim was to create an interactive App to
teach liver anatomy to improve trainee preparation for
hepatic surgery.
Methods: Liver model and animation storyboards were
created from radiographic images by a certified medicalillustrator under the guidance of a hepatobiliary fellowshiptrained surgeon in the education center for anatomy and
simulation at the authors institution. Animations were
completed in Adobe Photoshop and Illustrator. 3D polygonal
models were completed with detailed attention to liver shape,
vessel placement, and internal divisions. Interactivity of the
App was designed in Unity3D. The animated video models
were UVMapped and brought into Cinema 4D. The App was
designed for use on the Apple iPad.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
49
Results: The App is an interactive model with a narrated

video to teach liver anatomy. The liver rotates in space permitting a better understanding of its 3-D structure. Parenchyma can be removed to reveal vascular and biliary
anatomy. A narrated video provides a detailed overview of
hepatic anatomy with a logical progression from whole liver
topography sequentially down to segmental detail. Initial
feedback via internal assessment is exceptional.
Conclusions: This portable, mobile-device based instrument is a novel educational tool to teach liver anatomy via an
interactive approach. Studies to assess its educational utility
are underway. Integration with cross-sectional imaging is
planned to enhance clinical applicability. The current iteration is applicable for student, resident and fellow-level
trainees.
LO-H.02 RESECTION IS NOT INFERIOR

TO LIVER TRANSPLANTATION IN NODE
NEGATIVE INTRAHEPATIC
CHOLANGIOCARCINOMA
O. C. Kutlu, S. Garcia, M. V. Williams
TTU HSC Dept Of Surgery, Lubbock, TEXAS
Introduction: Intrahepatic cholangiocarcinoma(IHCCC) is
the second most common malignancy of the liver. Despite the
increasing incidence, few studies have been published on
therapeutic options and outcomes. Although survival benefits
of transplantation are well established for HCC, there is little
information on the outcomes between liver transplantation
and resection for IHCCC. In this study we investigated the
survival of IHCCC in a large population database and identified if there was a survival advantage of transplantation over
resection for stage I and II tumors.
Material Methods: SEER database was used to identify
IHCCC patients. Patients diagnosed between 1990 and 2008,
histologically proven IHCCC, T1 and T2 tumors, N zero, no
metastasis, no radiotherapy, and not lost to follow up were
included in the study. Analyses were performed using SPSS
20 with Kaplan-Meier statistics and Cox proportional
hazards regression.
Results: A total of 297 patients, 221 underwent resection
and 76 underwent transplantation met the criteria. Mean survival for resection was 36.8 months and 41.1 months for
transplantation. Survival for transplantation vs surgery is as
follows, 80% and 78% at one year, 62% and 63% two years,
54% and 51% three years, 45% and 36% four years, 28% and
36% at 5 years respectively. Survival between both groups
were similar (P = 0.29).
Conclusion: We evaluated if transplantation offered a survival benefit in patients with early IHCCC. Results showed
no difference in survival between resection and transplantation. This study questions the utility of transplantation for
stage I and II IHCC in the era of organ shortage.
HPB 2015, 17 (Suppl. 1), 181
LO-H.03 A NATIONWIDE ASSESSMENT

OF OUTCOMES AFTER BILE DUCT
RECONSTRUCTION
M. F. Eskander, L. A. Bliss, O. K. Yousafzai,
S. W. De Geus, S. Ng, M. P. Callery, K. Khwaja,
J. F. Tseng
Beth Israel Deaconess Medical Center, Boston, MA
Background: Bile duct reconstruction (BDR) is used to
manage benign and malignant neoplasms, choledochal cysts
and congenital anomalies, trauma and iatrogenic bile duct
injuries, and other non-malignant diseases. We compared
BDR outcomes overall and by indication.
Methods: Retrospective analysis of Nationwide Inpatient
Sample discharges (20042011) including ICD-9 codes for
BDR. All statistical testing performed using survey weighting. Univariate analysis of patient, hospital, admission characteristics, and outcomes by indication using chi square
testing. Multivariate modeling for inpatient complications
and inpatient death by logistic regression.
Results: Identified 67,160 weighted patient discharges in
which BDR was coded: 2.5% for congenital anomaly, 37.4%
for malignant neoplasm, 2.3% for benign neoplasm, 9.9% for
biliary injury, and 48% for other non-malignant disease.
68.4% of BDR discharges for neoplasm were elective vs.
2015 The Authors
50
Abstracts
60.8% for congenital anomaly, 46.1% for other nonmalignant and 37.8% for biliary injury (p < 0.0001.) 79.8%
of neoplasm discharges were from teaching hospitals vs.
62.3% for other non-malignant disease, 65.2% for biliary
injury and 66.1% for congenital anomaly (p < 0.0001.)
33.3% of total BDR discharges involved at least one complication and 84.8% were discharges to home. Median length of
stay was 9 days (IQR 6, 15) and median cost was $22,230
(IQR 14,399, 38,358.) Significant multivariate predictors of
inpatient death include indication of biliary injury or malignancy (figure), and predictors of any complication include
public insurance and non-elective admission.
Conclusion: This is the first national description of BDR
using a large database. In this diverse sampling, both procedure indication and patient characteristics influence morbidity and mortality.
included surgery (n = 256, 57.5%), antibiotics alone

(n = 117, 26.3%), and cholecystostomy tube (n = 72, 16.2%).
For all patients, length of stay (p < 0.001), disposition to
home (p < 0.001), and morbidity (p = 0.003) were related to
increasing TG13 grade. For surgical patients, worsened outcomes with increasing TG13 grade were seen for conversion
to open (p = 0.001), OR duration (p < 0.001), length of stay
(p = 0.009), disposition to home (p < 0.001), and readmission (p = 0.037). On multivariate analysis, TG13 grade was
an independent predictor of increasing length of stay
(p = 0.009) and conversion to open surgery (grade 2 = OR
7.63 (2.2525.90), grade 3 = OR 24.2 (5.0116.37)).
Conclusion: The TG13 criteria for grading acute cholecystitis accurately stratify patient outcomes in a US population.
Wide adoption of TG13 can better inform patients, hospital
systems, and payers of the expected outcomes of acute
cholecystitis.
LO-H.05 TARGETING DEFINITIVE

MANAGEMENT IN PATIENTS WITH
ACUTE GALLSTONE PANCREATITIS
AND CHOLEDOCHOLITHIASIS
L. C. Ewan, H. Jenkins, D. A. Subar
Department Of HPB Surgery, Blackburn, LANCASHIRE
LO-H.04 PREDICTING LENGTH OF STAY

AND CONVERSION TO OPEN SURGERY
FOR ACUTE CHOLECYSTITIS:
VALIDATING THE 2013 TOKYO
GUIDELINES IN A US POPULATION
G. Wright1,2, M. T. Hefty1,2, K. Stilwell2, J. Johnson2,
M. H. Chung1,2,3
1
GRMEP/Michigan State University General Surgery
Residency Program, Grand Rapids, MI; 2Michigan State
University College Of Human Medicine, Grand Rapids,
MI; 3Spectrum Health Medical Group, Grand Rapids, MI
Introduction: Predicting expected patient outcomes based
on disease severity is becoming increasingly important in the
US healthcare system. The 2013 Tokyo Guidelines (TG13)
for the diagnosis and severity of acute cholecystitis were put
forward by a consensus panel.
Methods: A retrospective review of patients presenting with
acute cholecystitis to a single center from 20092013 was
performed. The diagnosis and severity of cholecystitis were
assigned according to the TG13. The primary outcome measures were length of stay and conversion to open surgery.
Regression models were constructed for risk-adjusted
analysis.
Results: A total of 445 patients were eligible for study.
Patients were divided as follows: 137 (30.8%) grade 1, 191
(42.9%) grade 2, and 117 (26.3%) grade 3. Primary treatment
2015 The Authors
Background: Appropriate management of common bile

duct stones in patients with gallstone pancreatitis often varies
from recommended guidelines.
Aim: To determine if patients with gallstone pancreatitis
with common bile duct stones (CBDS) were appropriately
investigated and managed according to guidelines.
Methods: This retrospective study identified 165 patients
from 20092013 with a first episode of gallstone pancreatitis.
Cumulative scoring (05) based on age >55 yrs, CBD
>7 mm, bilirubin >22 mmol/L, ALP >200, ALT >53 was
used to predict probability of CBDS. Investigation with
either MRCP only, ERCP only, or MRCP then ERCP
was determined for each group. The presence of CBDS was
used determine whether the primary investigation was
appropriate.
Results: In the high CBDS probability group (Score
5,n = 18) 83% demonstrated a stone. 72% of this group had
ERCP directly and 28% had MRCP then ERCP. The mean
delay between MRCP and ERCP was 6.8d. For those with 4
risk factors for CBDS (n = 44) 45% were found to have a
CBDS. The majority had an MRCP before ERCP (45%),
20% went straight to ERCP and 16% had MRCP only.
Average delay to ERCP from MRCP was 5.3d (inpatients)
and 99d (outpatients). For those at low risk of CBDS (Score
02,n = 55), 9% had CBDS. Most underwent MRCP only
(34.5%) however 24% had ERCP of which 85% negative for
CBDS.
Conclusion: Acute gallstone pancreatitis cases predicted as
high risk for CBDS should undergo ERCP directly. For those
at moderate risk outpatient delays for MRCP should be
avoided. ERCP should be avoided in low risk groups unless
MRCP demonstrates a CBDS.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
LO-H.06 BILE DUCT RESECTION IN THE

TREATMENT OF HEPATOBILIARY AND
GALLBLADDER MALIGNANCY: EFFECT
OF ASSOCIATED PROCEDURES ON
OUTCOMES
P. Shen, N. Fino, E. Levine, C. Clark
Wake Forest School Of Medicine, Winston Salem, NC
Introduction: Resection of the bile duct is required for the
treatment of cholangiocarcinoma and sometimes indicated
when resecting gallbladder and hepatic tumors.
Methods: The American College of Surgeons National Surgical Quality Improvement Program Participant Use File was
used to analyze surgical outcomes in a database of patients
with hepatobiliary and gallbladder malignancies undergoing
bile duct resection with or without hepatic or vascular resection (n = 787). Patients were divided into three groups based
on type of procedure performed: 1)Bile duct resection only
(n = 289); 2)Bile duct resection with hepatic resection
(n = 454); and 3)Bile duct resection with hepatic resection
and vascular resection (n = 44). Postoperative complications
were compared between groups and regression-adjusted risk
factors were analyzed to produce observed and expected
(O/E) morbidity and mortality rates and indices.
Results: Performing additional procedures significantly
increased rates of organ space surgical site infection
(p < 0.0001), being on ventilator >48 hours (p = 0.0388),
acute renal failure (p = 0.0055), sepsis (p = 0.0292), septic
shock (p = 0.0208) and overall risk of having at least one
complication (p < 0.0001). Thirty-day mortality rates for
Groups 1, 2 and 3 were 6.23%, 8.15% and 18.18%, respectively (p = 0.0242). Risk-adjusted morbidity and mortality
rates also increased when Group 1 (O/E = 1.27 and 1.64) was
compared to Group 2 (O/E = 1.61 and 2.31) and Group 3
(O/E = 1.88 and 8.00).
Conclusion: Hepatic and vascular resection significantly
increase morbidity and mortality when performed with bile
duct resection for malignancy. Patients undergoing all three
procedures experience an 8-fold increase in mortality risk
with a morbidity risk that is 2-fold higher.
LO-H.07 CLINICAL AND

PATHOLOGICAL FEATURES OF
INTRADUCTAL PAPILLARY NEOPLASM
OF THE BILIARY TRACT AND
GALLBLADDER
S. Bennett1, C. Marginean2, M. Paquin-Gobeil1,
J. Wasserman2, J. Weaver1, R. Mimeault1, F. K. Balaa1,
G. Martel1
1
University Of Ottawa, Dept Of Surgery, Ottawa, ON;
2
University Of Ottawa, Dept Of Pathology And Laboratory
Medicine, Ottawa, ON
Background: Intraductal papillary neoplasms of the bile
duct (IPNB) and intracholecystic papillary neoplasms
(ICPN) are rare tumors of biliary epithelium, characterized
by papillary growth within lumen that can be associated with
invasive carcinoma. Their natural history remains poorly
understood. This study examines clinicopathological features
and outcomes.
HPB 2015, 17 (Suppl. 1), 181
51
Methods: Patients who underwent surgery for IPNB/ICPN

between 2009 and 2014 were identified. Descriptive statistics
were generated.
Results: Of 23 patients found to harbor IPNB/ICPN, 43%
were male and average age was 68. Most common presentations were jaundice (43%), abdominal pain (29%), and incidental ultrasound finding (14%). Preoperative ERCP with
brushing/biopsy showed at least cytologic atypia in 8/10
cases. Tumor locations were: 5 intrahepatic, 3 hilar, 8 extrahepatic bile duct, and 7 gallbladder. Mean tumor size
was 3.8 cm, 25% had positive lymph nodes, 47% had
lymphovascular invasion, and 37% had perineural invasion.
The R0 resection rate was 83%. The average number of
lymph nodes sampled was 4.8. Epithelial subtypes included
pancreatobiliary (52%) and intestinal (48%), and 87% demonstrated invasive carcinoma, either tubular type or
mucinous. Median follow-up was 25 months. The 3-year
overall and disease-free survivals were 70% and 61%,
respectively. Of the 6 recurrences, 4 occurred in patients who
had extrahepatic bile duct cancers and 2 in gallbladder
cancers.
Conclusion: IPNB/ICPN are rare tumors that spread along
the entire biliary epithelium, including the gallbladder. At
pathology, the majority of patients demonstrate invasive carcinoma, thus warranting radical resection. Oncologic prognosis may be superior to that of other biliary tract cancers.
LO-H.09 INTRAOPERATIVE
NEAR-INFRARED CHOLANGIOGRAPHY:
OPTIMIZATION OF TIMING AND DOSE
A. Zarrinpar, E. P. Dutson, C. Mobley, R. W. Busuttil,
C. E. Lewis, A. Tillou, A. Cheaito, O. J. Hines,
V. G. Agopian, D. T. Hiyama
Department Of Surgery, David Geffen School Of Medicine,
UCLA, Los Angeles, CA
Introduction: Intraoperative cholangiography is the gold
standard for clear delineation of biliary anatomy. However,
logistical difficulties lead to its low utilization. Near-infrared
fluorescence cholangiography (NIRFC) with indocyanine
green (ICG) has been developed for real-time, intraoperative
biliary imaging. While several studies have shown its feasibility, dosing and timing for its practical use have not been
systematically optimized.
Objective: We undertook a prospective observational study
with varying doses and elapsed times from injection of ICG
to visualization. Image quality of NIRFC and its utility to the
operating surgeon were assessed.
Methods: Adult patients undergoing laparoscopic biliary
and hepatic operations were enrolled. A single intravenous
dose of ICG (0.020.25 mg/kg) was administered at various
times (15180 mins) prior to planned visualization. The
porta hepatis was examined using a dedicated laparoscopic
system. Each operating surgeon evaluated the intraoperative
recognition of biliary structures using a qualitative scoring
system (1-poor to 5-excellent). Quantitation studies were
also performed on the images obtained during the operation.
Results: Thirty-four patients were enrolled. Visualization
scores of the extrahepatic biliary tract improved with increasing doses of ICG up to 0.08 mg/kg. The score also improved
with increased time up to 45 min after ICG administration.
Similarly the CBD-to-liver intensity ratio increased with
2015 The Authors
52
Abstracts
both dose and time. These results suggest that a dose of

0.08 mg/kg administered 45 minutes prior to visualization is
optimal for visualization, with diminishing improvements
with increased dose and time.
Conclusion: NIRFC is safe, practical, and effective in delineating extrahepatic biliary anatomy during laparoscopic
biliary and hepatic operations.
LO-I.02 LONG-TERM OUTCOME OF

PATIENTS UNDERGOING LIVER
TRANSPLANTATION FOR MIXED
HEPATOCELLULAR CARCINOMA AND
CHOLANGIOCARCINOMA: AN
ANALYSIS OF THE UNOS DATABASE
V. Vilchez1, L. Pena2, M. Shah1, M. F. Daily1, C. Tzeng1,
D. Davenport1, R. Gedaly1, E. Maynard1
1
Department Of Surgery University Of Kentucky,
Lexington, KENTUCKY; 2Department Of Internal
Medicine-Gastroenterology University Of Kentucky,
Lexington, KENTUCKY
Objective: To compare long-term outcomes in patients
undergoing liver transplantation (LT) for mixed hepatocellular carcinoma/cholangiocarcinoma (HCC-CC) versus
those with hepatocellular carcinoma (HCC) or cholangiocarcinoma (CC).
Methods: A retrospective analysis of patients undergoing
LT for HCC-CC was performed using the United Network
for Organ Sharing (UNOS) database from 19942013.
Overall and disease-free survival (OS, DFS) in patients with
HCC-CC, HCC, and CC were compared.
Results: Of the 123,167 patients who underwent LT, 4,049
patients had a primary malignancy (94 HCC-CC; 3,515
HCC; 440 CC). Within the HCC-CC cohort 47(50%) had
diagnosis of HCV, compared to 1260 (35%) with HCC and
11(2%) with CC. The mean age of the patients with HCC-CC
was 57 10 years and 77% were male. MELD at time of
listing did not differ among the three groups. Forty-six
percent of the patients with HCC-CC recurred. OS at 1, 3 and
5-years for HCC-CC (82%, 47%, 40%) was similar to CC
(79%, 58%, 47%) but significantly worse compared to HCC
(86%, 72%, and 62% p = 0.002). Similarly, DFS at 1, 3, and
5 years, for HCC-CC (78%, 45%, 38%) was similar to CC
(75%, 55%, 44%) but significantly worse than HCC (82%,
68%, 54%, p = 0.005).
Conclusion: LT for mixed HCC-CC have inferior OS and
DFS compared to those with HCC, suggesting that HCC-CC
outcomes more closely follow the CC phenotype. Attempts
2015 The Authors
should be made to identify HCC-CC patients prior to transplant and if transplanted undergo close surveillance and consideration for immunosuppression modification and/or
adjuvant therapy.
LO-I.03 LIVER TRANSPLANTATION FOR

HEPATOCELLULAR CANCER IN HIV
POSITIVE PATIENTS
N. Nissen1, R. Rogers2, B. Barin4, P. Stock2
Cedars-Sinai Medical Center, Los Angeles, CA;
2
University Of California, San Francisco, San Francisco,
CA; 4EMMES Corporation, Rockville, MD
1
Introduction: Liver transplantation (LT) is an excellent

treatment option for hepatocellular carcinoma (HCC) in
many patients, but whether this applies to the HIV+ patient is
unknown.
Aim: To evaluate the outcomes of LT for HCC in patients
with HIV co-infection by analyzing results of a US
multicenter trial of solid organ transplantation in HIV+
patients. In this trial, 125 patients underwent LT between
2001 and 2007 at 11 US centers.
Methods: Database review.
Patients: Forty-five out of 125 patients (30%) undergoing
LT in this trial had HCC. These patients averaged 50 years of
age and the median native MELD at transplant was 15 (IQR
1115). All patients were within Milan criteria at transplant
and only 1 patient had been previously downstaged. Hepatitis
C (HCV) was present in 30 of 45 patients.
Results: One and 3 year graft survivals for the HCC+/HIV+
patients were 82% and 67%, compared to 75% and 58% in
the HCC-/HIV+ patient group from the same trial (p ns).
Most grafts were lost due to HCV. At a median followup of
4.5 years, 4 patients (11%) developed HCC recurrence,
which occurred at a mean of 14 months after LT. Of the
recurrences one patient survived >24 months.
Conclusion: LT is an effective treatment for HCC in HIV+
patients, with survival rates similar to that of HIV+ patients
without HCC. Patient and graft survival are primarily determined by HCV status. HIV+ patients do not have excessive
risk of tumor progression after LT.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
LO-I.04 PREDICTIVE FACTORS FOR

EXTRAHEPATIC RECURRENCE OF
FOLLOWING ORTHOTOPIC LIVER
TRANSPLANTATION
A. Andreou, J. Pratschke, D. Seehofer
Department Of General, Visceral And Transplant Surgery,
Charite Campus Virchow Klinikum, Berlin,
DEUTSCHLAND
Background: Recurrence of hepatocellular carcinoma
(HCC) in patients treated with orthotopic liver transplantation (oLTX) is associated with diminished survival. Particularly extrahepatic localization of HCC recurrence contributes
to poor prognosis.
Patients and Methods: Clinicopathological data of patients
who underwent oLTX for HCC between 1989 and 2010 in a
high-volume transplant center were retrospectively evaluated
and predictors of extrahepatic recurrence were identified.
Results: Three hundred and sixty-seven patients underwent
oLTX for HCC. After a median follow-up time of 77 months,
93 patients (25%) were diagnosed with a recurrence. Median
time to recurrence was 18.9 months. Recurrence was located
exclusively in the liver in 19 cases (20%) and 74 patients
(80%) had extrahepatic recurrence. Factors associated with
extrahepatic recurrence in multivariate analysis included
HCC beyond the Milan criteria (P < .0001) and the presence
of major vascular tumor invasion (MVI) (P = .035). In
patients with HCC beyond the Milan criteria who developed
a recurrence (n = 73), MVI was the only positive predictor of
extrahepatic recurrence in multivariate analysis (P = .0001).
In patients with HCC within the Milan criteria who recurred
after oLTX, DNA-index >1.5 (P = .04) was the only predictive factor for extrahepatic recurrence.
Conclusions: Advanced HCC beyond the Milan criteria and
the presence of MVI are associated with an increased risk for
extrahepatic recurrence and are currently considered as
contraindications to oLTX. In patients with HCC within the
Milan criteria, the DNA-index represents a valuable prognostic marker for the development of extrahepatic recurrence
and may support the selection of patients for intensive postoperative tumor surveillance.

4:00PM5:30PM
ORAL POSTER II (PANCREAS,
TRANSPLANT)
OP-II.01 PANCREATIC DEBRIDEMENT
FOR NECROTIZING PANCREATITIS:
NATIONAL OUTCOMES AND
PREDICTORS OF MORTALITY
S. W. Ross, E. M. Hanna, R. C. Kirks, R. Seshadri,
J. B. Martinie, R. Z. Swan, D. A. Iannitti
Carolinas Medical Center, Division Of HPB Surgery,
Charlotte, NC
Background: Necrotizing pancreatitis often requires pancreatic debridement, which can result in a high rate of
HPB 2015, 17 (Suppl. 1), 181
53
morbidity and mortality. Risk factors that predispose patients

to post-operative complications or death are not well-defined.
Using a national surgical database, we sought to identify
pre-operative predictors of mortality after debridement.
Methods: The American College of Surgeons NSQIP database was queried from 20052011 by CPT code for pancreatic debridement. Risk factors for complications and death at
30 days were examined using univariate tests and predictors
of mortality were identified using step-wise logistic
regression.
Results: 1,162 patients underwent pancreatic debridement.
On average patients were middle aged(54.9 14.2 years),
male(70.0%), and obese(30.6 8.0 kg/m2). The most
common
comorbidities
included
diabetes(33.2%),
smoking(22.3%), COPD(6.5%), steroid use(4.0%) and
cardiac history (4.0%). Prior to surgery, 21.2% had ascites,
7.6% had acute renal failure, 14.7% had recent >10% body
mass loss, and 63.2% had pre-operative sepsis. Emergent
surgery occurred in 29.3%. Wound, general, and major complications occurred in 7.2%, 56.6% and 31.3% respectively.
Mean length of stay was 34.1 31.0 days. 30 day mortality
was 7.7%, and was higher in patients with COPD, cardiac
history, ascites, acute renal failure, steroid use, sepsis, older
age, higher BMI, and emergent operations(p < 0.05). The
table summarizes independent predictors of mortality identified using multivariate analysis.
Conclusion: Nationally, rates of adverse outcomes following pancreatic debridement are high. Patients who are older,
with higher BMI, are on dialysis, have COPD, with
poor functional status, hypoalbuminemia, azotemia or
hyperbilirubinemia have increased odds of mortality following debridement for necrotizing pancreatitis.
OP-II.02 EXPERIENCE WITH

PANCREAS-SPARING DUODENECTOMY
FOR FAMILIAL ADENOMATOUS
POLYPOSIS
M. Dong, N. Ali, S. Reddy, C. ORourke, G. Morris-Stiff,
R. M. Walsh
Department Of HPB Surgery, Cleveland, OHIO
Introduction: Duodenal adenomas are a common finding in
patients with familial adenomatous polyposis (FAP) and
individuals with Spigelman stage IV adenomas are at high
risk of developing duodenal carcinoma. These patients are
traditionally treated by pancreatoduodenectomy (PD) though
an alternate approach is pancreas-sparing duodenectomy
2015 The Authors
54
Abstracts
(PSD). We report a 22-year experience with PSD for the

treatment of duodenal polyps in FAP.
Methods: A retrospective review was performed of a prospectively maintained database containing all patients undergoing PSD from 1992 to 2013. Phone interviews were
conducted to confirm current status of patient at follow-up.
Results: Fifty-four patients underwent PSD during the
study period, all for Spigelman stage IV polyps. An unsuspected invasive cancer was found in one patient on final
pathology. The mean operative time was 305 70 minutes
with a mean blood loss of 300 170 mL. There was one
peri-operative mortality, unrelated to the operative procedure. Thirteen patients (24%) had an immediate postoperative complication including eight (15%) biliary/
pancreatic leaks, and 1 (2%) enteric anastomotic leak.
Pancreatitis was observed in 4 (10%). 42(78%) of patients
were available for follow-up. Recurrent polyps were found in
16(34%). Of these, only 3(19%) patients required operative
intervention, two proximal jejunal resections and one PD for
development of a polyp at the ampullary anastomosis.
Conclusion: Our experience with PSD reinforces its value
as a definitive prophylactic procedure for duodenal polyposis
in FAP and allows for full preservation of pancreatic
function.
OP-II.03 THE EFFECT OF EARLY

POST-OPERATIVE NON-STEROIDAL
ANTI-INFLAMMATORY DRUGS ON
OUTCOMES FOLLOWING
R. Behman1,2, P. Karanicolas1,2, M. Lemke1,2, S. Hanna1,2,
C. Law1,2, N. Coburn1,2, J. Hallet1,2
1
University Of Toronto, Toronto, ONTARIO; 2Odette
Cancer Centre, Toronto, ONTARIO
Introduction: Non-steroidal
anti-inflammatory
drugs
(NSAIDs) are used commonly for post-operative analgesia,
but can potentially impair healing. Their effect on
pancreaticoduodenectomy (PD) outcomes is unknown. We
sought to examine the impact of early post-operative
NSAIDs on pancreatic fistula (PF) after PD.
Methods: We reviewed our prospective pancreatectomy
database supplemented by medication administration
records, including all PDs from 2002 to 2012. Primary
outcome was occurrence of clinically significant (Grade
B-C) PF. Secondary outcomes included major morbidity
(Clavien grade III-V) and 90-day mortality. Patients were
compared based on early post-operative NSAIDs use (first 3
days following surgery) using univariate and multivariate
analyses. Sub-group analyses were conducted based on
NSAIDs type (ketorolac and COX2-inhibitor).
Results: We identified 251 PDs, of whom 127 (50.6%)
patients received NSAIDs postoperatively (15.1% ketorolac,
32.7%, COX2-inhibitor, 2.8% both). Use of any NSAIDs
was associated with a non-significant increase in PF (16.5%
Vs 11.3%%; p = 0.23), and no difference in major morbidity
and mortality. Use of ketorolac was not associated with an
increase in PF (8.7% Vs. 15.1%; p = 0.256). COX2inhibitors were associated with increased PF (20.2% Vs.
10.5%; p = 0.033), but no difference in major morbidity or
mortality. After adjusting for age, Charlson comorbidity
2015 The Authors
score, diagnosis, and estimated blood loss, use of COX2inhibitors was independently associated with PF (odds ratio
2.32; p = 0.026).
Conclusions: COX2-inhibitors are associated with PF in the
early postoperative period. While ketorolac appears safe in
this setting, caution is warranted with the use of COX2inhibitors.
OP-II.04 NEOADJUVANT THERAPY

WITH ANATOMICAL BORDERLINE
PANCREATIC DUCTAL
ADENOCARCINOMA. DOES IT
MAKE DIFFERENCE?
A. M. Zaki1, N. Rezaee1, J. He1, J. L. Cameron1,
K. Hirose1, T. M. Pawlik1, N. Ahuja1, M. A. Makary1,
H. E. Horeya2, M. J. Weiss1, C. L. Wolfgang1
1
Johns Hopkins University School Of Medicine, Baltimore,
MARYLAND; 2Mansoura University School Of Medicine,
Mansoura, DAKAHLIA
Background: The benefit of neoadjuvant therapy over a
surgery-first approach in patients with borderline pancreatic
ductal adenocarcinoma (PDAC) has not been well defined.
Aim: To compare postoperative outcomes of patients with
borderline PDAC who underwent pancreatectomy after
neoadjuvant treatment with those of patients who underwent
upfront surgery.
Methods: Between 2008 and 2014, 231 patients were identified as anatomical borderline PDAC. 117 of 231 (50.6%)
patients received neoadjuvant therapy and 114 (49.4%)
patients had a surgery-first approach. Univariate, multivariate
and survival analyses were performed.
Results: Compared to surgery first group, neoadjuvant
group was associated with smaller tumor size in the pathological specimen (P < .001), lower incidence of metastatic
lymph nodes (39% vs. 80%; P < .001), less perineural invasion (61% vs. 97%: P < .001), less micro-vascular invasion
(32% vs. 68%: P < .001), less vascular resection rate (31%
vs. 57%; P < .001) and a lower rate of positive resection
margin (32% vs. 44%; P < .055). Univariate analysis identified nodal status, lymph node ratio and tumor size as predictors for survival. Multivariate analysis identified only lymph
node ratio (P = .005) as independent predictor of patient survival. Postoperative mortality and morbidity rates were
similar in the 2 groups. However, survival analysis starting
from date of first dose of neoadjuvant therapy versus date of
upfront surgery showed better median overall survival in
favor of the neoadjuvant group (44 months vs. 20 months;
P = .011).
Conclusion: Neoadjuvant treatment for borderline PDAC is
associated with better pathological outcomes and overall survival. Lymph node ratio can provide significant prognostic
information after pancreatectomy for patients with borderline PDAC.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
OP-II.05 PANCREATODUODENECTOMY
PROVIDES EFFECTIVE LONG-TERM
PAIN RELIEF FOR CHRONIC
PANCREATITIS IN SELECT PATIENTS AT
GREATER THAN 15 YEARS FOLLOW-UP
K. P. Croome, D. M. Nagorney, M. Tee, M. J. Truty,
K. Reid-Lombardo, F. G. Que, M. L. Kendrick,
M. B. Farnell
Background: We
have
employed
pancreaticoduodenectomy (PD) for selected patients with small duct, headdominant chronic pancreatitis (CP) with intractable pain.
Information examining very long term outcomes in patients
undergoing PD for CP is lacking.
Patients and Methods: All patients who underwent PD for
CP from 1976 to 2013 were reviewed. Surviving patients
were contacted for a follow-up questionnaire and SF-12
Quality of Life Survey.
Results: A total of 166 patients were identified (Cohort 1:
19761999(N = 105) and Cohort 2: 20002013(N = 61)).
Median time from presentation until surgery was significantly longer in Cohort 2(2.09 years) compared to Cohort
1(1.13 years)(p = 0.017). A higher proportion of patients in
Cohort 2(98%) had intractable pain prior to surgery than in
Cohort 1 (82%)(p = 0.002). Prior to PD a higher proportion
of patients in Cohort 2 had undergone endoscopic stenting,
67% vs 10%(p < 0.001) and/or celiac plexus block 15% and
5%(p = 0.026). Median follow-up for all survey respondents
was 15 years. On the SF-12, mean physical component
score(PCS) was 43.8 11.8 and mental component
score(MCS) was 54.3 7.9. Patients were significantly
lower on the PCS(p < 0.001) and significantly better on the
MCS(p = 0.001) than the general US population. Mean pain
score out of 10 was significantly lower after surgery 1.6 2.6
than before surgery 7.9 3.5(p < 0.001). Diabetes developed
in 28% of patients who were not diabetic prior to surgery.
Conclusion: Although practice has changed so that patients
have a longer time from presentation until surgery as less
invasive techniques are attempted, PD appears to provide
effective long-term pain relief and acceptable quality of life
in appropriately selected patients with chronic pancreatitis
and intractable pain.
OP-II.06 GETTING THE FULL PICTURE:

SIGNIFICANCE OF INDETERMINATE
LUNG NODULES IN PANCREATIC
ADENOCARCINOMA
K. T. Hemingway1, E. Halpern2, P. V. Pandharipande3,
A. L. Warshaw1, C. Fernandez Del Castillo1,
K. D. Lillemoe1, C. R. Ferrone1
1
Department Of Surgery-Massachusetts General Hospital,
Boston, MA; 2Department Of Biostatistics-Massachusetts
General Hospital, Boston, MA; 3Department Of
Radiology-Massachusetts General Hospital, Boston, MA
Objectives: Patients presenting with resectable pancreatic
adenocarcinoma (PDAC) often have indeterminate lung
nodules on pre-operative chest CT scans. The aim of this
study was to determine whether nodules on chest CT are
predictive of lung first metastases (LFM).
HPB 2015, 17 (Suppl. 1), 181
55
Methods and Materials: Retrospective clinicopathologic

data was recorded for patients undergoing resection of their
PDAC between 19932012 who also underwent a preoperative chest CT scan. Time to metastasis and overall survival (OS) were calculated.
Results: Of 792 patients who underwent surgical resection
of their PDAC, 171 underwent a pre-operative chest CT scan.
Median age was 67.7 years and 54.7% were female. Median
PDAC size was 30 mm. Indeterminate nodules were identified in 111 patients. Indeterminate lung nodules of 03 mm in
35 patients, 36 mm in 44 patients, 610 mm in 12 patients
and >10 mm in 14 patients. Lung-first metastasis (LFM)
developed in 20.7% of patients with nodules vs. 10.8% of
patients with no nodules (p = 0.142). Patients with LFM had
the longest median OS compared to patients developing
lung + liver (6.3%) or lung + local (5.4%) metastases
(32.7 mo vs 22.3 mo vs. 18.6 mo, respectively). The risk of
lung-first metastases did not increase with increasing size,
number, or location of indeterminate nodules, smoking
history or pre-operative Ca19-9. Of the 28 patients with
LFM, 6 underwent surgical removal of one or more nodules.
Conclusions: Neither the location, number, or size of indeterminate lung nodules on pre operative chest CT scans are
predictive of developing lung metastases as the first site of
metastatic disease.
OP-II.07 READABILITY AND ACCURACY

OF ONLINE PATIENT MATERIALS FOR
PANCREATIC CANCER BY TREATMENT
MODALITY AND WEBSITE AFFILIATION
A. Storino1, M. Castillo-Angeles1, A. A. Watkins1,
C. Vargas1, J. Mancias1, A. Bullock1, A. Demirjian2,
A. J. Moser1, T. S. Kent1
1
Beth Israel Deaconess Medical Center Harvard Medical
School, Boston, MA; 2University Of California, Irvine,
Orange, CA
Introduction: Patients search online health information frequently but there is little quality control. Patient/family
understanding of presented information may depend, partly,
upon readability and accuracy. This study was undertaken to
evaluate reading level and accuracy among commonly
searched websites about pancreatic cancer treatment options.
Methods: An online search on 5 pancreatic cancer treatment
modalities was conducted. Readability was measured by 9
standardized tests and accuracy was assessed by an expert
panel. Readability and accuracy were compared by treatment
modality and website affiliation by Kruskal-Wallis test.
Results: Significant differences existed by treatment modality for both readability and accuracy (Table 1), with surgeryrelated websites having the lowest readability level.
Alternative therapy-related websites had the lowest accuracy.
Readability varied by affiliation, with lower readability for
non-profits than media-owned (p = 0.00001) and academic
center sites (p = 0.0001). Privately-owned websites had
lower readability than media sites (p = 0.009). Accuracy was
highest for government websites; government, academic, and
non-profit sites were more accurate than privately-owned or
media-owned websites (p = 0.0001). There was no association between accuracy and readability level.
2015 The Authors
56
Abstracts
Conclusions: Although variation existed in both readability

and accuracy, improvement is needed throughout. The readability level for all treatment modalities is higher than recommended, which may negatively impact patient/family
understanding of treatment options. Accuracy was reasonable, except for alternative therapy websites. Website affiliation impacted both readability and accuracy. Privately-owned
and media sites had lower accuracy. In accordance with
patient-centered care, improvement is needed in the quality
of online resources in order to empower patients in the
shared-decision making setting.
OP-II.08 ENGLISH AND SPANISH

LANGUAGE READABILITY OF ONLINE
PATIENT RESOURCES FOR
PANCREATIC CANCER
M. Castillo-Angeles, A. Storino, A. A. Watkins, C. R.
Vargas, J. F. Tseng, M. P. Callery, A. J. Moser, T. S. Kent
Beth Israel Deaconess Medical Center, Harvard Medical
School, Boston, MA
Introduction: 50% of cancer patients search the Internet for
health information. The NIH and AMA recommend that
patient-oriented health information should be written at a
sixth-grade reading level. We evaluated commonly-searched
English and Spanish-language pancreatic cancer websites in
light of these recommendations.
Methods: The top ten websites for pancreas cancer in
English and Spanish were identified in a patient-simulated
manner using the most-visited internet search engine. 122
English and 76 Spanish articles were assessed using ten
readability analyses. Spanish-language websites native to 3
representative Spanish-language countries with AHPBA
chapters were identified. Mean readability scores were determined for each article and website. T-tests were used to
compare reading levels by website, language, and country.
Results: Overall mean reading level across US websites was
12.4 1.42 (English) and 10.5 0.95 (Spanish). Mean readability by website ranged from 9.5 to 15.2 in English, and
from 8.6 to 12.3 in US Spanish. Argentine, Chilean, and
Mexican websites had mean reading levels 9.7, 10.9, and 11
respectively. Spanish websites in each country were significantly easier to read than those in English (Figure 1).
Conclusion: Commonly searched online pancreatic cancer
resources in English and Spanish exceed the recommended
sixth grade reading level. Spanish-language websites,
US-based or not, were significantly easier to read than
English-language sites, though still above the recommended
6th grade level. With the growing focus on patient-centered
care, attention should be paid to ensuring availability of
2015 The Authors
information at a level understandable by the general public.

More easily understandable materials may facilitate patient
participation in shared decision-making.
OP-II.09 PLASMA CANCER ANTIGEN

19-9 (CA19-9) LEVELS DIFFERENTIATE
PATIENTS WITH MALIGNANT
INTRADUCTAL PAPILLARY MUCINOUS
NEOPLASM (IPMN) FROM THOSE WITH
IPMN ALONE
T. K. Nguyen, D. Joyce, G. A. Falk, S. Chalikonda,
G. Morris-Stiff, R. Walsh
Cleveland Clinic Foundation, Cleveland, OH
Introduction: Malignant transformation is a recognized
complication of IPMNs, in particular when the disease
affects the main pancreatic duct. However they are often only
recognized during histopathological examination of resection specimens. Plasma CA19-9 has potential use in the
diagnostic work-up of pancreatic adenocarcinoma but has
not been well investigated in IPMN. The aim of this study
was to evaluate the role of CA19-9 in differentiating between
malignant IPMN (invasive [IPMN Ca] and high grade dysplasia [HGD]) IPMN from those with IPMN and low/
moderate (LGD & MGD) dysplasia.
Methods: The institutional pancreatic cyst database was
interrogated to identify all patients with a histopathological
diagnosis of IPMN. Patients were sub-divided into three categories based on the degree of neoplastic change: IPMN
carcinoma; IPMN HGD; and IPMN LGD&MGD. Ca19-9
levels were assessed in relation to the 3 categories. The
IPMN Ca and IPMN HGD were then co-assessed.
Results: During the period January 2000 to December 2013,
158 patients underwent resection, with final histopathology
indicating: IPMN [n = 45] carcinoma, IPMN HGD [n = 24];
and IPMN LGD/MGD [n = 89]. There was no difference in
comparing incidental versus pre-operatively diagnosed
cancers or when comparing main duct and side-branch
IPMNs.
Conclusions: Plasma CA19-9 is independently valuable in
the evaluation of HGD and carcinoma arising in the setting of
IPMN.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
57
OP-II.10 PANCREATIC NECROSIS: A

SINGLE INSTITUTIONS REVIEW OF
PRACTICAL ADHERENCE TO A
STEP-UP APPROACH
S. Downs-Canner, B. Boone, J. Steve, A. Zureikat,
K. K. Lee, H. J. Zeh, M. Hogg
University Of Pittsburgh Medical Center, Pittsburgh, PA
Introduction: Level 1 evidence demonstrated the step-up
approach to infected pancreatic necrosis improves outcomes.
We examined our institutions compliance with this approach
and its impact on outcomes.
Methods: We performed a retrospective review of the index
admission of all patients with necrotizing pancreatitis treated
in 2012 at a tertiary referral academic medical center.
Results: 41 patients (median age 52, 75% male) were
treated for necrotizing pancreatitis. 80% were transferred a
median of 3 days from presentation. By the revised Atlanta
Classification, 23 (56%) had severe acute pancreatitis and 19
(46%) had suspected/confirmed infected necrosis. Of those
infected, 18 (95%) required pancreatic drainage procedures;
10 (56%) underwent step-up (Table). Median time from presentation to first pancreatic drainage was 1 day (step-up = 0.5
days versus surgery first = 4.5 days). Five (50%) of step-up
went on to surgery a median of 1 day later. Nine (47%) had
multi-system organ failure (step-up = 30% versus surgery
first = 75%). Median length of stay was 23 days for step-up
versus 21 days for surgery first. ICU admission rate was 90%
for step-up versus 80% for surgery first. In-hospital mortality
was 20% in step-up and 12.5% in surgery first. Overall survival for all necrotizing pancreatitis patients was 78% and
71% of deaths were related to pancreatitis. Pancreatitis
related readmission rate was 59% (step-up = 30% versus
surgery = 88%).
Conclusion: Step-up adherence was 56%; however, half of
those went on to surgery. Further analysis is necessary to
determine if the subset of patients undergoing surgery first
represent deviations from level 1 recommendations or medically appropriate deviations.
HPB 2015, 17 (Suppl. 1), 181
OP-II.11 NATIONAL DISPARITIES IN

MINIMALLY INVASIVE SURGERY FOR
PANCREATIC CANCER
E. M. Gabriel1, P. Thirunavukarasu1, K. Attwood2,
S. Hochwald1, B. Kuvshinoff1, S. Nurkin1
1
Roswell Park Cancer Institute, Buffalo, NY; 2Roswell Park
Cancer Institute, Buffalo, NY
Introduction: Social and racial disparities have been related
to differences in access to care. This study investigated patterns in minimally invasive surgery (MIS) across different
social, racial and geographic populations of patients with
pancreatic cancer.
Methods: We utilized the National Cancer Database, 2004
to 2011, to identify patients with pancreatic cancer who
underwent surgery through either an open, laparoscopic or
robotic approach. Multivariate analysis was performed to
characterize differences in patient demographics in relation
to surgical approach.
Results: A total of 11,464 patients were identified. The
initial surgical approach included 82.5% open (9,461),
15.8% laparoscopic (1,815) and 1.6% robotic (188). Table 1
shows the results of our analysis. Race was not statistically
significant across the different surgical approaches. There
was a trend toward increased MIS in patients with private
insurance. Academic centers performed more MIS compared
to community cancer programs. On multivariate analysis,
only national location was shown to be a statistically significant factor associated with increased rates of MIS. Patients in
the Middle Atlantic region of the US were most likely to have
robotic surgery. Regarding laparoscopic surgery, the Mountain and West South Central states had the lowest rates of
laparoscopic procedures, but among the other national
regions there were no statistically significant differences.
2015 The Authors
58
Abstracts
Conclusions: Minimally invasive approaches for pancreatic

cancer comprise 17.5% of surgical procedures. Race and
insurance status were not statistically significant factors associated with MIS. Although academic centers performed most
of the MIS, specific geographic regions comprised the only
statistically significant factor on multivariate analysis.
whereas only 47 (38.5%) were reported in their pathology

reports. Thirty patients were found to have pathologic evidence of PV/SMV resection, but not based on billing.
Conclusion: Histologic characteristics and margin distances
are frequently missing from PAC pathology reports. High
rates of R1 resections exist, as well as discrepancies regarding reporting of major vein resections between billing and
pathology reports, which holds substantial implications for
utilization of pathology reports for determination of
PV/SMV resection.
OP-II.13 COMPARATIVE ANALYSIS OF

INSTITUTIONAL CLINICAL PATHWAYS
FOR MANAGEMENT OF BORDERLINE
RESECTABLE PANCREATIC CANCER
O. M. Rashid1, J. M. Pimiento1, P. Nguyen1, G. Springett1,
P. Hodul1, S. Hoffe1, R. Shridhar1, B. L. Johnson2, K. Illig2,
P. A. Armstrong2, W. J. Fulp1, M. P. Malafa1
1
H. Lee Moffitt Cancer Center And Research Institute,
Tampa, FL; 2University Of South Florida, Tampa, FL
OP-II.12 WHAT IS THE COMPLETENESS

OF PATHOLOGY REPORTS FOR
RESECTED PANCREATIC
ADENOCARCINOMA SPECIMENS? A
POPULATION-BASED ANALYSIS
A. El-Sedfy1,2, D. J. Kagedan1,3, E. Shin1, R. Raju1,
M. E. Dixon1,4, M. Elmi 1,3, C. Rowsell3, Q. Li1,3,
N. Mittmann1,3, N. G. Coburn1,3
1
Sunnybrook Research Institute, Toronto, ONTARIO; 2Saint
Barnabas Medical Center, Livingston, NJ; 3Sunnybrook
Health Sciences Centre, Toronto, ONTARIO; 4Maimonides
Medical Centre, New York, NY
Introduction: Following oncologic resection, pathology
reports provide critical information for determining prognosis and directing treatment. However, reports are often
incomplete. We assessed pathology report completeness
from pancreatic adenocarcinoma (PAC) resections on a
population level.
Methods: All patients who underwent resection for PAC in
Ontario between 2005 and 2010 were identified using provincial databases. Pathology reports were evaluated for completeness based on the College of American Pathologists
(CAP) 2012 Protocol. Portal (PV) and superior mesenteric
vein (SMV) resections identified from physician billing
codes were compared to the corresponding pathology report.
Results: 475 patients had available pathology reports for
review. 421 were pancreaticoduodenectomy and 54 were
distal pancreatectomy. CAP pathologic variables and the percentage of reports missing each is as follows: T stage 0.8%;
N stage 0.4%; number of lymph nodes examined 3.4%;
margin status 2.7%; tumor grade 1.7%; lymphovascular invasion 20.2%; perineural invasion 13.5%; microscopic tumor
extension 5.5%. 32.4% of pathology reports (n = 156)
revealed R1 resection. 21.4% of reports for specimens with
R0 resection (n = 308) were missing margin distance. Based
on billing, 122 patients underwent PV/SMV resections,
2015 The Authors
Background: Currently there is no consensus combination

therapy clinical pathway (CP) for the management of borderline resectable pancreatic adenocarcinoma (BPA) based on
randomized prospective studies, therefore, BPA CPs vary by
institution. We recently reviewed outcomes of our BPA CP
and now seek to compare results at other centers to further
evaluate optimal therapy.
Methods: We performed a literature review of all BPA CP
reports, excluding case reports and feasibility studies. BPA
diagnostic criteria, pancreatectomy rate, margin status,
pathologic response (grade IIa-IV) rate (PR), disease free
(DFS), disease specific (DSS), and overall survival (OS)
were reviewed and compared to our institutional data.
Standard statistical methods were used for statistical
comparison.
Results: There were three studies which met inclusion criteria (Table). Including the 101 patients treated at our institution, 279 entered 4 CPs. 175(59%) met NCCN/AHPBA
BPA diagnostic criteria, 144(41%) MDACC Type A. The CP
regimens are listed (Table). With 95% confidence, resection
rate at our institution was 44.264.4% versus 27.749.3%
with CP 1, 16.646.5% with CP 2, and 35.861.3% with CP
3. With 95% confidence, the PR at our institution was 69.9
93.7% versus 55.263.5% with CP 1, 77.789.0% with CP2,
and 8.610.7% with CP3. DFS, DSS, and OS are reported
(Table).
Conclusions: A comparative analysis of outcomes demonstrates that many BPA patients benefit from the reported
CPs. However, the differences in reported resection and
pathologic response to treatment rates between institutional
CPs warrant further investigation to determine optimal
therapy.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
59
OP-II.14 DOMAIN-BASED ASSESSMENT

OF THE LEARNING CURVE FOR NEW
SURGICAL TECHNOLOGY:
ROBOT-ASSISTED VS. OPEN DISTAL
PANCREATECTOMY
OP-II.15 TOTAL PANCREATECTOMY

FOR PANCREATIC ADENOCARCINOMA
IS ASSOCIATED WITH EQUIVALENT
PERI-OPERATIVE MORTALITY AND
LONG TERM SURVIVAL AS COMPARED
TO PARTIAL PANCREATECTOMY
S. Klompmaker, A. A. Watkins, W. J. Van Der Vliet,

S. J. Thoolen, A. Storino, M. Castillo-Angeles, J. F. Tseng,
M. P. Callery, T. S. Kent, A. J. Moser
Beth Israel Deaconess Medical Center, Harvard Medical
School, Boston, MA
E. H. Baker1, R. M. Seshadri1, M. A. Templin2,

R. Z. Swan1, J. B. Martinie1, D. A. Iannitti1
1
Department Of General Surgery, Division Of HPB
Surgery, Carolinas Medical Center, Charlotte, NC;
2
Dickson Advanced Analytics Group, Charlotte, NC
Introduction: The Learning Curve for minimally-invasive

surgery monitors operating time, conversion rate, and incremental cost as surrogate markers of proficiency and outcome.
We expanded this concept to include four aggregate domains
for new technology assessment (Table 1) based on Institute
of Medicine principles to evaluate overall risk/benefit. The
initial Learning Curve for robot-assisted distal pancreatectomy (RADP) was compared to unmatched consecutive open
DP (ODP) at an expert center.
Methods: Unmatched comparison between 29 RADP and
169 consecutive ODP performed between 20062012 prior
to implementation of RADP. Cumulative treatment burden at
90 days was assessed. Propensity scoring controlled for
selection bias.
Results: No differences in age, gender, race, Charlson
Comorbidity Index, suspected pathology, tumor location, or
size were observed between the RADP and ODP cohorts.
Within the efficiency domain, RADP patients had a reduced
90-day total hospital stay (6 vs. 7 days, p = .002), but longer
mean operative time as compared to ODP patients (p < .001).
Cumulative morbidity and oncological efficacy for malignancy was similar to ODP within the limits of sample size.
Safety, as measured by blood loss and laparotomy rate
(3.3%), was improved following RADP compared to ODP.
Propensity-scored sensitivity analysis did not alter these
results.
Conclusion: Domain-based evaluation of the initial RADP
learning curve was comparable to the established phase of
ODP in consecutive patients at an expert center. Operating
time and associated costs should be re-evaluated in the
context of reduced total hospital stay and increased patient
eligibility for the minimally-invasive approach. Prospective
validation of these metrics is required.
Introduction: Historically, total pancreatectomy (TP) for

pancreatic adenocarcinoma (PAC) has been performed for
large or advanced tumors but carries inherent significant
postoperative morbidity secondary to endocrine and exocrine
insufficiency. Questions regarding postoperative mortality
and long term survival remain important when comparing
utility of TP to partial pancreatecomy (PP) procedures such
as pancreaticoduodenectomy (PD) or distal pancreatectomy
(DP) for pancreatic cancer.
Methods: Retrospective analysis of the National Cancer
Data Base between the years 19982011 was performed.
224,335 patients were identified with diagnosis code of PAC;
128,368 patients had relevant data from 19982006 and were
eligible for 5 year survival analysis. Chi-square, t-test,
Kaplan-Meier curves and log rank tests were used for statistical analysis.
Results: 16,744 patients underwent surgical resection for
PAC during the study period. 1,954 patients underwent TP
while 16,111 underwent PP (959 DP and 13,357 PD). Rates
of TP increased during the study period (164 in 1998 to 295
in 2006). Mean tumor size was 38 mm for TP and 34.8 mm
for PP (p = 0.001). More lymph nodes were examined following TP (11.8 vs 10.8, p < 0.001). 30 day mortality rates
following TP were similar to PP (6.05% vs 5.35%, p = 0.20).
When compared stage for stage, there were no differences in
30 day mortality between groups. Median overall survival
following TP was 14.98 months vs 15.64 months for PP
(p = 0.16). No difference was seen in long term overall survival when patients were compared stage for stage.
Conclusions: Post-operative mortality and long term overall
survival rates are similar between TP and PP procedures.
HPB 2015, 17 (Suppl. 1), 181
2015 The Authors

60
Abstracts
and lack of at least a 50% decrease in CA19-9 levels (OR
13.2 [2.569.1]) were. On sub-analysis, CA19-9 decrease
<50% remained predictive for any progression while tumor
size predicted distant progression only. A receiver operating
characteristic curve showed that tumor size >3 cm was 87%
sensitive for progression (AUC 0.785).
Conclusion: LAPD patients undergoing neoadjuvant
chemotherapy with tumors >3 cm or that exhibit less than
50% reduction in CA 19-9 maybe at higher risk for progression on chemotherapy. Patients with these risk factors may
benefit from additional treatment prior to an attempt at
resection.
OP-II.17 SURGERY VERSUS

NON-OPERATIVE MANAGEMENT OF
PANCREATIC ADENOCARCINOMA
WITH SMA INVASION
P. Thirunavukarasu, E. Gabriel, B. Kuvshinoff,
S. Hochwald, S. Nurkin
Roswell Park Cancer Institute, Buffalo, NY
OP-II.16 PREDICTORS OF PROGRESSION

OF LOCALLY ADVANCED PANCREATIC
CANCER ON NEOADJUVANT
CHEMOTHERAPY
J. B. Rose1, F. G. Rocha1, A. Alseidi1, T. Biehl1, B. Lin2,
V. Picozzi2, S. Helton1
1
Virginia Mason Medical Center, Section Of Hepatobiliary
Surgery, Seattle, WA; 2Virginia Mason Medical Center,
Department Of Hematology/Oncology, Seattle, WA
Introduction: Neoadjuvant treatment for patients with
locally advanced pancreatic adenocarcinoma (LAPD) is
becoming more widely utilized and may be associated with
improved survival. However, even with careful patient selection, many patients do not undergo curative resection due to
disease progression. The objective of the present study is
to identify predictors of disease progression prohibiting
resection.
Methods: A retrospective review was performed on all
patients with LAPD at a high volume tertiary hospital
between January 2008 and August 2014 who received
extended neoadjuvant gemcitabine/docetaxel chemotherapy.
Clinicopathologic predictors of disease progression prohibiting resection were determined by univariate and multivariate logistic regression analysis.
Results: Eighty-four patients with LAPD were initiated on
neoadjuvant chemotherapy. 16 patients (19%) progressed on
treatment by RECIST criteria (9 distant, 7 local). Multivariate logistic regression analysis found that sex, age by
quartile, McGill-Brisbane score, clinical stage III, or multivessel involvement were not predictive of progression prohibiting resection. However, tumor size (OR 2.6 [1.35.5])
2015 The Authors
Background: Pancreatic adenocarcinoma (PDAC) invading

the Superior Mesenteric Artery (SMA) is traditionally considered unresectable, with surgery offered to select patients.
We evaluated outcomes of surgery versus non-operative
management.
Methods: Data for patients with SMA-invading PDAC
without evidence of distant metastases was extracted from
the National Cancer Database, 19982006.
Results: Of 3,445 patients, 227 (6.9%) had surgery,
specifically pancreaticoduodenectomies (73.2%), distal
pancreatectomies (14.5%) and total pancreatectomies
(12.3%). Overall R0 resection rate was 43.7%. Median hospital stay was 9 days, 30-day mortality 7.0%, and 30-day
readmission rate 8.7%. In patients who underwent surgery
plus systemic therapy, median overall survival (OS) was 21
months when margin-negative and 13.6 months when
margin-positive (p < 0.001). In contrast, among patients who
underwent surgery alone, OS was uniformly poor regardless
of margin status (8.3 vs 6.7 months, p = 0.09). Patients with
R0 resection plus systemic therapy had significantly better
OS compared to patients in whom surgery was not recommended (e.g.: due to comorbidities) or given systemic
therapy alone despite recommendation for surgery (Figure
1). Among patients without comorbidities, R0 resection plus
systemic therapy had significantly better survival than systemic therapy alone (22.2 months vs. 11.4 months, p < 0.01).
The operative mortality of patients with 1 comorbidity
was lower than with 2 comorbidities (6.0% vs. 12.6%,
p = 0.05).
Conclusion: For appropriate patients with SMA-invading
PDAC, a combination of R0 resection with systemic therapy
offers the best outcome. Given the higher mortality, these
operations should be offered selectively in specialized
centers.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
61
OP-II.19 751 CONSECUTIVE

PANCREATICO-DUODENECTOMIES AT
A TERTIARY CARE CENTRE: MAKING A
CASE FOR STANDARDIZATION AND
CENTRALIZATION
E. Pai, A. Mitra, S. Patkar, M. Goel, S. V. Shrikhande
Tata Memorial Hospital, India, Mumbai, MAHARASHTRA
OP-II.18 EARLY DRAIN REMOVAL IS A

BEST PRACTICE IN SELECTED
PANCREATIC SURGERY PATIENTS
H. A. Pitt1, B. L. ZarZaur2, S. W. Behrman3,
E. M. Kilbane2, B. L. Hall4, A. Parmar5, R. S. Riall5
1
Temple University School Of Medicine, Philadelphia, PA;
2
Indiana University School Of Medicine, Indianapolis, IN;
3
University Of Tennessee College Of Medicine, Memphis,
TN; 4Washington University School Of Medicine, St. Louis,
MO; 5University Of Texas Medical Branch, Galveston, TX
Background: The morbidity of pancreatic surgery remains
unacceptably high. Recent reports suggest that drain management may influence postoperative complications. While
early drain removal may improve outcomes, pancreatic surgeons have been slow to adopt this practice. Therefore, the
aim of this analysis was to compare outcomes when drains
were removed early or late after pancreatic surgery.
College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) Pancreatectomy Demonstration Project (PDP). Over a 14-month period, 2,805 patients
underwent a pancreatoduodenectomy, distal pancreatectomy,
total pancreatectomy or pancreatic enucleation at 43 institutions. After exclusion of patients without drains or data on
drain removal, 1,841 patients were available for analysis.
Early drain removal was defined as on or before POD#3.
Early drain patients (n = 148, 8.1%) were younger, had
more women and were less likely to undergo a
pancreatoduodenectomy. Therefore, propensity score matching and sensitivity analyses were performed. Outcomes were
determined by ACS-NSQIP and PDP definitions. Standard
statistical tests were applied.
Results: After propensity score matching early (n = 127)
and late (n = 127) drain removal patients were well-balanced
for age, gender, BMI, serum albumin, ASA class, operation
type, time and approach, vascular resection, gland texture,
duct size, pathology and POD#1 drain amylase. Outcomes
are presented in the table.
Conclusions: This analysis suggests that early drain
removal is associated with reduced morbidity following
pancreatic surgery. Early drain removal should be considered
in selected pancreatectomy patients.
HPB 2015, 17 (Suppl. 1), 181
Background: The
outcome
of
Pancreaticoduodenectomy(PD) has been closely linked to hospital
volume and experience. The low incidence of pancreatic
cancer, coupled with few specialized Hepato-PancreaticoBiliary teams and lack of referral patterns and service centralization, contribute to sparse data from the Indian
subcontinent.
Methods: Prospective database of PDs from 1992 to 2014
was evaluated retrospectively over 4 time periods based on
changing practice trends: A (19922001), B (2003-July
2009), C (August 2009-December 2011) and D (January
2012-August 2014). Peri-operative parameters were compared using SPSS v.21.0.
Results: 751 patients underwent PD. The average resections
increased from 14 to 94, over periods A to D, respectively.
While post-operative pancreatic fistula(POPF) rates increased
from 16% to 21.5% over periods A to D, the incidence of bile
leaks and post-pancreatectomy hemorrhage declined to 0.8%
and 3.9%, respectively. Morbidity and hospital stay was more
in period D compared to B and C, but mortality declined from
6.3% to 2.8% over periods A to D. Overall series morbidity
and mortality was 35.2% and 4.5%, respectively.
Conclusion: This series represents the largest single-centre
experience with PD from India and the surrounding region.
The higher morbidity in period D is likely attributable to a
combination of increasing surgical complexity and reduced
selection bias. The gradually increasing experience should
serve as a benchmark for developing dedicated pancreatic
surgery teams even in regions of low incidence of pancreatic
cancer where the need for training and centralization is
greater compared to high incidence regions, such as USA and
Europe.
2015 The Authors

62
Abstracts
OP-II.20 LEAKAGE OF AN
INVAGINATION
PANCREATICOJEJUNOSTOMY MAY
HAVE LETHAL CONSEQUENCES
H. Lavu1, S. W. Keith 1, E. M. Kilbane3, A. Parmar4,
B. L. Hall 5, H. A. Pitt2
1
Thomas Jefferson University, Philadelphia, PA; 2Temple
University School Of Medicine, Philadelphia, PA; 3Indiana
University Health, Indianapolis, IN; 4University Of Texas
Medical Branch, Galveston, TX; 5Washington University
School Of Medicine, St. Louis, MO
Background: No consensus exists regarding the
most effective form of pancreaticojejunostomy following
pancreatoduodenectomy. The aim of this analysis was to
determine whether the type of pancreaticojejunostomy influences morbidity or mortality.
College of Surgeons-National Surgical Quality Improvement
Program, Pancreatectomy Demonstration Project. Over 14
months, 1,781 patients underwent a pancreatoduodenectomy
(PD) at 43 institutions. After exclusion of patients undergoing minimally invasive PD and those without information on
gland texture or duct size, 890 patients were analyzed.
Patients were divided into duct-to-mucosa (n = 734, 82%)
and invagination (n = 156, 18%). Type of pancreaticojejunostomy (PJ) was then included in eight separate morbidity and mortality multivariable analyses.
Results: Invagination patients had higher serum albumin
(p < 0.01) lower BMIs (p < 0.01) and preoperative serum
bilirubin (p < 0.02), were less likely to have a preoperative
biliary stent (p < 0.01) or chemotherapy (p < 0.04), were
more likely to have a soft gland (p < 0.01) and were
less likely to undergo pylorus preservation (p < 0.01).
Multivariable analyses demonstrated that age, gender, BMI,
preoperative albumin and biliary stents, gland texture and
pancreatic duct size were related (p < 0.05) to multiple postoperative morbidity outcomes. PJ anastomosis type was not
associated with morbidity but did affect mortality (duct-tomucosa vs. invagination Odds Ratio 0.22, p < 0.01). Among
patients who developed a pancreatic fistula, none of the 119
duct-to-mucosa compared to five of 20 invagination patients
died (p < 0.01).
Conclusions: Patients who undergo a pancreaticojejunostomy (PJ) by duct-to-mucosa or invagination differ
with respect to pre- and intra-operative variables. When an
invagination PJ leaks, the consequences may be lethal.
OP-II.21 SHOULD ACUTE PANCREATITIS

BE AN INDICATION TO RESECT IPMN?
J. L. Cioffi, S. J. Lee, J. A. Waters, C. M. Schmidt,
A. Nakeeb, M. G. House, E. P. Ceppa, N. J. Zyromski
Indiana University, Indianapolis, IN
Introduction: Intraductal papillary mucinous neoplasms
(IPMN) cause acute pancreatitis (AP) more commonly than
generally appreciated. The natural history of IPMN with AP
is unclear, and whether an episode of AP should be an indication for surgery in the setting of IPMN is controversial. We
sought to determine the natural history of IPMN causing AP.
Methods: 348 patients with pathologically proved, resected
IPMN were analyzed. Patients with single versus multiple
2015 The Authors
episodes of AP were compared to determine clinical and

pathological differences between these groups.
Results: 114 (33%) IPMN patients had at least 1 episode of
AP. Among IPMN/AP patients, 22 (19%) had more than one
episode of AP. IPMN type (main duct, branch duct, mixed
type), location (head versus body/tail), and size were similar
between patients with single versus multiple AP episodes.
Duration of symptoms related to IPMN was significantly
longer for multiple AP patients (single episode 17 months,
multiple episodes 37 months). Invasive carcinoma was
present in 13 (11%) of all IPMN/AP patients, and was more
common in patients with multiple AP episodes (23%) than
those with a single AP episode (9%).
Conclusions: Acute pancreatitis occurs in 33% of patients
with resected IPMN, 19% of whom had multiple AP episodes
prior to resection. Patients with multiple AP episodes were
more likely to harbor invasive carcinoma compared to those
with a single episode of AP. These data support early resection of IPMN patients who develop acute pancreatitis.
OP-II.22 ACCURACY OF PREOPERATIVE

IMAGING FOR VASCULAR
INVOLVEMENT IN LOCALLY
ADVANCED, BORDERLINE RESECTABLE
FOLLOWING NEOADJUVANT
CHEMOTHERAPY
J. Clanton, J. B. Rose, A. Alseidi, T. Biehl, S. Helton,
F. Rocha
Virginia Mason Medical Center, Seattle, WA
Background: Radiographic imaging with arterial/venous
phase, thin-slice computed tomography (MDCT) and endoscopic ultrasound (EUS) are utilized for assessment and
staging of locally advanced, borderline resectable pancreatic
adenocarcinoma (BR-LAPD). Neoadjuvant therapy followed
by pancreaticoduodenctomy with vascular resection is typically required for BR-LAPD. However, little is known about
the correlation of MDCT and EUS findings with need for
vascular resection and pathologic invasion.
Methods: A retrospective review of a prospectivelymaintained database of consecutive patients with BR-LAPD
treated with neoadjuvant chemotherapy at our institution
between 2011 and 2014 was performed. MDCT and EUS
results were compared to operative findings and the final
pathologic reports regarding vascular involvement.
Results: A cohort of 33 patients underwent staging by
MDCT and EUS followed by neoadjuvant chemotherapy and
successful resection. All patients were reported to have
venous involvement on CT, EUS, or both. Based on preoperative imaging and operative findings, a venous resection
was performed in 15/33 (45.5%) patients, but histological
evidence of vascular invasion was only noted in 6/33 (18.2%)
pathologic specimens. These six patients were noted to have
vascular invasion on both CT and EUS. CT demonstrated
vascular involvement without EUS confirmation in 4/33
(12%) patients, while EUS demonstrated vascular involvement without CT confirmation in 3/33 (9%) patients. None of
these patients had true pathologic vascular invasion.
Conclusion: CT and EUS can be complementary modalities
to detect BR-LAPD, but may overestimate the actual
HPB 2015, 17 (Suppl. 1), 181
Abstracts
incidence of venous involvement. This may be due to inherent limitations of these techniques versus a true downstaging
effect of neoadjuvant chemotherapy.
OP-II.23 DISTAL PANCREATECTOMY

PERFORMED BY HIGH VS LOW
VOLUME SURGEONS IN A LARGE
INTEGRATED HEALTH CARE SYSTEM:
A RISK ADJUSTED ANALYSIS
S. Woo, A. Difronzo
Kaiser Permanente Los Angeles Medical Center, Los
Angeles, CALIFORNIA
Introduction: Improved outcomes for pancreatectomy are
observed when performed at high-volume centers. The study
objective was to compare predicted and actual risk-adjusted
outcomes of distal pancreatectomy (DP) performed by high
vs low volume surgeons in an integrated health care system.
Methods: A retrospective review of patients having DP
from 20052013. A standardized risk calculator was used to
predict risk of complications and mortality. The Clavien
grading system was used for actual complications. High
volume was 7 or more DP per year.
Results: 50 surgeons performed 247 DP. 3 high volume
surgeons performed 54 cases. In the low volume cohort, 193
DP were performed (median 2 per surgeon). The calculated
risk of any complication and risk of mortality was no different between the two groups, but was significantly lower
than actual observed complications. There were more
laparoscopic (56 vs 33%, p = 0.001) and spleen-preserving
cases in the high volume group (59 vs 18%, p < 0.001). There
was a significant difference in overall actual 60-day complications: 33% in the high volume cohort vs 52% (p = 0.008).
There were fewer Clavien grade 1/2 complications (15 vs
37%, p < 0.001), and less postoperative pancreatic fistula
(PPF) (11% vs 33%, p = 0.0009) in the high volume group.
There was no difference in mortality between the groups.
Conclusion: Distal pancreatectomy performed by high
volume surgeons in a large health care system results in
decreased risk-adjusted overall morbidity including PPF.
Some observed differences may be related to greater use of
laparoscopy and splenic preservation.
OP-II.24 PORTAL VEIN THROMBOSIS

AFTER TOTAL PANCREATECTOMY AND
AUTOLOGOUS ISLET CELL
TRANSPLANTATION
W. P. Lancaster, D. B. Adams, K. A. Morgan
Medical University Of South Carolina Department Of
Surgery, Charleston, SC
Introduction: Portal vein thrombosis (PVT) is a rare complication of total pancreatectomy with autologous islet transplantation (TPIAT). Little is reported about the risk factors,
consequences, or treatment for this complication.
Methods: A retrospective review and analysis of a
prospectively-collected database of patients undergoing
TPIAT from March 2009 to August 2014 was conducted.
Two-tailed t-tests were used comparing continuous data and
Fishers exact test comparing categorical data.
Results: 135 patients (102 women,76%) underwent TPIAT;
Nine(7%) had PVT. All patients with PVT were women.
HPB 2015, 17 (Suppl. 1), 181
63
There were no differences in age or islet equivalents transplanted in patients with and without PVT. Mean BMI of
patients with PVT was lower than those without (21.8 vs
26.5 kg/m2,p = 0.03).Mean portal pressure post-islet
infusion was higher in patients with PVT (25.2 vs
16.0,p = 0.0007), with 4/9 having pressures over 30 mmHg.
The median time to diagnosis of PVT was 10.5 days
postoperative(range 7 to 210),with 7/9 having negative
duplex POD1. Eight of 9 patients with PVT were treated
with systemic anticoagulation and 7/8 had resolution on
repeat imaging. One patient died from complications of
anticoagulation. Two patients developed cavernous
transformation(CTPV), one untreated and one diagnosed
after CTPV. All patients with PVT were insulin-requiring at
latest follow-up versus 72/94 patients(77%) without PVT
with at least 1-year follow-up(p = 0.035).
Conclusions: PVT following TPIAT is an uncommon but
serious complication. It occurs late in the postoperative
period in women with a low BMI. A standardized follow-up
imaging protocol is suggested. The treatment for PVT is
anticoagulation. Patients with PVT can expect to be insulindependent.
OP-II.25 READMISSION AFTER

PANCREATECTOMY FOR CANCER;
A DATA-DRIVEN APPROACH TO
QUALITY IMPROVEMENT
S. M. Misustin, K. K. Christians, F. M. Johnston,
E. A. Krzywda, S. L. Lahiff, S. Tsai, D. B. Evans
Background: Hospital readmissions have been proposed as
a hospital quality measure. Analysis of readmissions has
largely focused on post-discharge management. Assessment
of pre-discharge data may reveal additional information.
Methods: Pancreatic cancer (PC) pts who underwent resection were identified. The causes of readmissions within 30
days of discharge were reviewed.
Results: Surgical resection was performed on 210 consecutive pts. The median length of stay of the index operation was
8 days (IQR: 4). Postoperative complications were identified
during the index hospitalization in 45 (22%) pts, with one
perioperative death. Pts were discharged to: home with
family (48; 23%), home with visiting nurse (144; 69%), other
facility (15; 8%). Readmission occurred in 29 (14%) pts.
Post-discharge analysis identified 21 (72%) of 29 readmissions related to the operation; infection (6), GI bleed (3),
delayed gastric emptying (3), pancreatic fistula (6), ileus (1),
SBO (2). An additional 8 (28%) of the 29 readmissions
were due to potentially preventable causes including;
hypoglycemia (1), dehydration/anorexia (5), and pain (2).
However, pre-discharge analysis using data available on the
day of discharge suggested that 9 (31%) of the 29 readmissions should not have been discharged and could have been
identified prior to discharge.
Conclusions: Pre-discharge analysis identified data
elements in 9 (31%) of the 29 readmissions that could have
prevented discharge, but were undetected in post-discharge
analysis. To prospectively identify these pts prior to discharge, the causes of these 29 readmissions were used to
create a checklist and a formal time out discharge procedure has been implemented.
2015 The Authors
64
Abstracts
OP-II.26 TITLE
D. P. Nussbaum, L. M. Youngwirth, R. R. White,
B. M. Clary, J. A. Sosa, D. G. Blazer
Duke University Medical Center, Durham, NORTH
CAROLINA
Introduction: Pancreatic acinar cell carcinoma (pACC) has
cure rates up to 40% following resection, yet many patients
with localized disease do not undergo surgery.
Methods: The 19982011 National Cancer Data Base was
queried for patients with pACC. Among patients with localized disease, multivariable analysis was used to predict the
likelihood of undergoing resection. Cox proportional hazards
modeling was then used to assess variables associated with
survival following resection.
Results: 933 patients were identified. Median age at diagnosis was 64 years. Tumors were most common in men
(66%) and white patients (88%), and occurred most frequently in the pancreatic head (57%). Mean size was 6.6 cm.
While 42% of patients presented with localized disease,
nearly one-quarter of these patients did not undergo resection. Median survival was 55 months following resection,
compared to 23 months without surgery (p < 0.01). Failure to
undergo surgery was associated with older age (OR 1.32,
p = 0.02), male sex (OR 2.30, p < 0.01), black race (OR 2.86,
p = 0.03), higher grade (OR 2.45, p = 0.03), location within
the head (OR 3.33, p < 0.01), and treatment at a nonacademic facility (OR 2.09, p < 0.01). Following adjustment,
only older age (HR 1.17, p = 0.01) and lymph node metastases (HR 2.58, p = 0.04) were associated with increased
mortality following resection (Table 1).
Conclusions: Survival following resection of pACC is
nearly five years, yet specific subsets of patients appear less
likely to undergo surgery. Of these groups, only older age is
independently associated with mortality. Efforts to increase
access to care could result in improvements in survival for
patients with pACC.
OP-II.27 RADIOGRAPHIC RESPONSE

AND RESECTABILITY OF LOCALLY
ADVANCED, BORDERLINE RESECTABLE
AFTER EXTENDED NEOADJUVANT
CHEMOTHERAPY
J. Clanton, J. B. Rose, A. Alseidi, T. Biehl, S. Helton,
F. Rocha
Virginia Mason Medical Center, Seattle, WA
Background: Neoadjuvant therapy is often utilized for
locally advanced, borderline resectable pancreatic cancer
(BR-LAPD) prior to resection, despite previous reports suggesting minimal downstaging. However little is known about
the effect of extended preoperative chemotherapy on disease
response.
Methods: Retrospective review of a prospective database of
consecutive patients with BR-LAPD treated with a 24-week
course of gemcitabine and docetaxel between 2011 and 2014
was performed. Patients with high-quality imaging before
and after treatment, and who completed full course of
therapy with intention to resect were included. RECIST 1.1
criteria were used to assess radiographic response.
Results: Forty-five patients who completed extended
neoadjuvant chemotherapy were included for analysis.
Thirty-two (71.1%) patients underwent pancreaticoduodenectomy and 14 (31.1%) received a vein resection.
Thirteen (28.9%) were not resected due to local or distant
progression or reduced performance status. The majority of
patients demonstrated stable tumor size (24/45, 53.3%) or
partial response (20/45, 44.4%). Suspicious lymph nodes
remained stable (27/45, 60%) or improved (11/45, 24.4%)
while vascular involvement was stable (21/45, 46.7%) or
improved (16/45, 35.6%) after treatment. None of the ten
patients not resected after chemotherapy had an increased
tumor size, three had enlarging lymph nodes, and one had
more extensive vascular involvement by CT. None of these
factors were associated with ability to undergo resection
(p > 0.05)
Conclusions: Extended neoadjuvant chemotherapy for
BR-LAPD commonly results in either stable disease or a
partial response. One-third of patients have demonstrable
downstaging of vascular involvement, however radiographic
response or downstaging is not associated with successful
resection.
OP-II.28 PATIENT SELECTION FOR

ROBOT-ASSISTED
PANCREATODUODENECTOMY
FAVOURS OBESITY
N. Ali, A. T. Stafford, G. Morris-Stiff, J. Wey, K. ElHayek,
S. Chalikonda, R. M. Walsh
Department Of HPB Surgery, Cleveland, OHIO
Introduction: There are potential advantages to the
application of minimally invasive techniques to pancreatoduodenectomy (PD), with current data indicating its technical feasibility. We present a single center experience of
robot-assisted PD (RAPD) focusing on patient selection and
outcomes, in particular infections, comparing results to open
PD.
2015 The Authors
HPB 2015, 17 (Suppl. 1), 181
Abstracts
Methods: A retrospective review of a prospectively maintained database of all open PDs and RAPDs from March
2009 to June 2014 was performed. RAPD has been introduced selectively during this period. As the main outcome
assessment was infection rate (total/deep organ space and
surgical site [SSI]), patients converted from RAPD to open
were included in the open group.
Results: 69 patients underwent completed RAPD, and 372
open PD during the study period. There was a change in
disease etiology from predominantly premalignant lesions
and ampullary carcinomas to include all indications for PD.
There was a significant difference in BMI between groups
with RAPD preferentially used in obese patients (p = 0.004)
that evolved during the study. Comparing RAPD and open
PD, the SSI rate was less in the RAPD group (9% vs.14%)
with no difference in overall infective complication rate
(26% vs. 26%) or intra-abdominal infection (10% vs. 10%).
Estimated blood loss was significantly less (200 versus
400 cc (p < 0.001) in favor of RAPD.
Conclusion: Increasing experience with RAPD has led to a
change in utilization of the procedure and a transition to its
use in all pathologies. To maximize its benefits we have a
positive selection bias towards obese patients.
OP-II.29 PREDICTORS OF ACTUAL

SURVIVAL IN RESECTED
PANCREATIC ADENOCARCINOMA:
A POPULATION-LEVEL ANALYSIS
D. J. Kagedan1,2, R. Raju2, M. Dixon3, E. Shin1, Q. Li5,
N. Liu5, M. Elmi1,2, A. El-Sedfy4, C. Earle2, N. Mittmann2,
N. G. Coburn2
1
University Of Toronto, Toronto, ON; 2Sunnybrook Health
Sciences Centre, Toronto, ON; 3Maimonides Medical
Center, Brooklyn, NY; 4Saint Barnabas Medical Center,
Livingston, NJ; 5Institute For Clinical Evaluative Sciences,
Toronto, ON
Introduction: Among patients diagnosed with pancreatic
adenocarcinoma, numerous clinicopathologic factors have
prognostic value following curative-intent resection. We
sought to assess actual survival following resection and to
determine factors predictive of survival on a population level.
Methods: Patients undergoing resection for pancreatic
adenocarcinoma between 200510 were identified within the
provincial cancer registry and administrative databases that
include actual survival for all patients in Ontario, Canada
(population 13 million). We fully abstracted pathology
reports for 473. Kaplan-Meier survival analysis and Cox
proportional hazards multivariate regression were performed
to determine the clinicopathologic variables associated with
decreased survival.
Results: The actual 1-, 3-, and 5-year survival rates were
65%, 23%, and 15% respectively, with median survival 1.48
years. Follow-up time ranged from 2.077.22 years, and 377
(79.7%) were censored for death before the end of follow-up.
Multivariate regression revealed the following variables to be
negatively associated with survival: age >70 (p = 0.001), T
stage (p < 0.01), nodal metastasis (p < 0.001), tumor grade
(p < 0.001), positive margin status (p < 0.01), lymphovascular invasion (p < 0.001), lymph node positivity ratio
>0.2 (p < 0.001). Patients with multivisceral or major vascular resections, and patients with low socioeconomic status
HPB 2015, 17 (Suppl. 1), 181
65
did not have worse survival. Receiving treatment at a

high-volume hepatopancreatobiliary center was associated
with improved survival (HR = 0.49, 95%CI = 0.360.67,
p < 0.0001).
Conclusion: Advanced age, positive margins, and histopathologic tumor characteristics predict poor prognosis, and
undergoing more extensive resection does not worsen
survival. Receiving treatment at a hepatopancreatobiliary
centre improves survival. In a publicly-funded healthcare system, poor socioeconomic status does not worsen
survival.
OP-II.30 DOES THE MD ANDERSON

(MDA) CRITERIA OF BORDERLINE
RESECTABLE (BLR) PANCREATIC
CANCER IMPACT SUCCESSFUL
COMPLETION OF NEOADJUVANT
THERAPY AND SURVIVAL?
A. N. Krepline, K. K. Christians, B. George, P. S. Ritch,
B. A. Erickson, K. Oshima, P. Tolat, D. B. Evans, S. Tsai
Background: BLR PC has been classified based on tumorvessel abutment/encasement or the radiographic presence of
indeterminate metastatic lesions. The importance of this subclassification on outcome is unclear.
Methods: BLR PC patients were classified as: type A, anatomic criteria; type B, indeterminate CT findings including
suspicious peripancreatic lymphadenopathy (LAD), or a
CA19-9 >2000 (hiCa19-9). Patients meeting both type A and
B criteria were classified as type B.
Results: Of the 110 BLR patients, 56 (51%) were type A and
54 (49%) were type B. Restaging after neoadjuvant therapy
(Neoadj), demonstrated metastases in 8 (14%) of 56 type A
vs. 11 (20%) of 54 type B patients (p = 0.82). Neoadj and
surgery (Surg) was completed in 66 (60%) of the 110
patients; 37 (66%) type A and 29 (54%) type B (p = 0.19). Of
the 54 type B patients, Neoadj and Surg was successfully
completed in: 8 (42%) of 19 with indeterminate CT findings;
15 (56%) of 27 with suspicious LAD; and 6 (86%) of 7 with
hiCa19-9 (p = 0.14). Overall survival (OS) for type A vs. B
patients was 17 vs.19 months (p = 0.93). Of the 66 patients
who completed Neoadj + Surg, type A and B patients had OS
of 31 vs. 42 months respectively (p = 0.38); there were no OS
differences between type B subgroups.
Conclusions: In a contemporary series of 110 pts with BLR
PC, 60% completed neoadjuvant therapy including surgery.
Subgroup categorization of BLR patients into type A and B
had no prognostic significance.
OP-II.31 THE IMPACT OF MINIMALLY

INVASIVE DISTAL PANCREATECTOMY
ON 90-DAY READMISSIONS AND COST:
IS IT ANY BETTER THAN OPEN?
J. Parikh, S. Anantha Sathyanarayana, S. Bendix,
M. J. Jacobs
Providence Hospital Medical Center, Southfield, MI
Introduction: Laparoscopic distal pancreatectomy (LDP) is
commonly performed for lesions of pancreatic body and tail.
2015 The Authors
66
Abstracts
Surprisingly, recent literature suggests an increase in readmission rates after LDP, hence potentially negating any
gained length of stay (LOS) benefit compared to open distal
pancreatectomy (ODP). Therefore, we sought to examine
readmission rates and total cost of LDP versus ODP at a
high-volume community hospital.
Methods: Between January 2003 to December 2013, 81
distal pancreatectomies were performed at a community
teaching hospital. A retrospective analysis on demographics,
90-day outcomes, readmission rates, length of stay (LOS),
and total cost were collected.
Results: Eighty-one patients underwent distal pancreatectomy (41 open and 40 laparoscopic). Median age was 62
years. Two-thirds of patients were female. LDP had significantly shorter mean operative time (150 vs. 183 minutes;
p < 0.01) and decreased blood loss compared to ODP (135
vs. 568 mL; p < 0.001). Table 1 compares tumor characteristics, LOS, readmission rates, and costs. Pancreatic fistula
rates were comparable with no Grade C fistulae in either
group. Overall 90-day morbidity was lower in the LDP group
with no mortalities. The 30-day and 90-day readmission rate
was lower in LDP; hence LDP has lower total hospital days.
The overall costs for both the index admission and the total
hospital stay (including readmission) were lower for LDP
group.
Conclusion: LDP has significantly lower index LOS, fewer
total hospital days and lower overall costs compared to ODP.
LDP should be the standard of care for amenable lesions in
the body or tail of the pancreas.
OP-II.32 SHOULD SMALL PANCREATIC

NEUROENDOCRINE TUMORS BE
OBSERVED OR RESECTED?
R. E. Eldert, N. Valsangkar, M. Kilbane, J. L. Cioffi,
M. G. House, N. J. Zyromski, C. M. Schmidt, A. Nakeeb,
E. P. Ceppa
Indiana University School Of Medicine, Indianapolis, IN
Background: Pancreatic Neuroendocrine Tumors (PNET)
are heterogeneous in behavior and metastatic potential.
Recent series recommend surveillance of PNET less than
2 cm in size; consensus is lacking whether a size cutoff for
PNET can be used to define malignant potential and guide
therapy. The aim of this study was to determine if there were
differences in cancer-specific outcomes following resection
for tumors 2 cm or less versus greater than 2 cm.
Methods: Consecutive cases of resected nonfunctional
PNET (n = 48) were reviewed retrospectively at a high
volume academic center between 2004 and 2009. Data were
gathered through the American College of SurgeonsNational Surgical Quality Improvement Program and electronic medical records. Survival was determined from the
institutional cancer registry and Social Security Death Index.
Results: Patients with PNET 2 cm or less (n = 16) versus
those greater than 2 cm (n = 32) had no significant difference
in demographic, preoperative, or intraoperative variables.
Surgical pathology was reviewed (Table). Patients underwent
distal pancreatectomy (56%), pancreaticoduodenectomy
(29%), enucleation (8%), central pancreatectomy (4%), and
Frey-procedure (2%). Mortality at 30 and 90 days was 0% for
both groups. The 2 cm or less group had a lower 30-day and
90-day morbidity (p < 0.01). There was no significant difference in five-year survival between groups.
Conclusions: Resected nonfunctional PNET 2 cm or less
appear to have a similar rate of lymph node and distant
metastasis and no significant benefit of 5 year survival compared to larger PNETs. The lower rate of postoperative morbidity and equivalent overall malignant nature of PNET
supports resection over observation of PNET 2 cm or less.
OP-II.33 CHOLEDOCHODUODENOSTOMY
IS A SAFE AND EFFICIENT
ALTERNATIVE FOR BILE DUCT
RECONSTRUCTION DURING
LIVER TRANSPLANT
T. L. Nydam
University Of Colorado School Of Medicine, Aurora, CO
In our institution, choledochoduodenostomy (CDD) has
become the bile duct reconstruction of choice in liver transplants when a duct-to-duct choledochocholedocostomy
(D2D) is not possible. We provide evidence that CDD is a
safe option for bile duct reconstruction with significant
advantages during postoperative care.
2015 The Authors
HPB 2015, 17 (Suppl. 1), 181
Abstracts
Methods: All orthotopic liver transplants performed at The
University of Colorado Hospital from July 2006 to July 2013
were retrospectively reviewed. Patient demographics, donor
type, post-transplant complications, ERCP times, and biliary
percutaneous transhepatic interventions (PTC) were collected. Statistical analysis was performed using a paired
students t-test assuming equal variances.
Results: 632 liver transplants were performed. Eighty-two
patients underwent CDD, 28 patients underwent Roux en Y
choledochojejunostomy (CDJ), and 522 patients underwent
D2D. There was no statistical difference in cholangitis, bile
leak, anastomotic stricture, or other complications. However,
there was a statistically significant difference in mean length
of ERCP and number of PTCs between the CDD and CDJ
cohorts. (Table 1).
Conclusions: Contrary to traditional teaching, a CDD
reconstruction appears to have no difference in complications
compared to a CDJ reconstruction. In addition, the length of
time spent during ERCP and the number of PTCs required
were significantly lower in the CDD cohort. In our institution, CDD is a safe option for bile duct reconstruction during
liver transplant that provides improved postoperative access
to the graft biliary system.
67
Methods: Twelve high risk LTx from 2011 to 2014 were

reviewed. Sorafenib started within 2 months if: microvasular
invasion (MVI), beyond Milan, or AFP >500. Dose was escalated from initial 400 mg/d.
Results: 75% were male, average BMI was 31, and 66% had
bilobar tumors. Pathologic staging revealed ave. 2 tumors,
total tumor burden 5.1 cm, 75% either stage B or C (BCLC
staging), and MVI in 42%. AFP was >100 in 17%. Ave. dose
was 265 mg in all patients but 300 mg in those that tolerated
treatment. Four (33%) of patients are still on treatment, but 5
( 42% ) received it for less than 50 days. Side effects included
hand foot syndrome, flushing, anxiety,nausea, diarrhea,
hypertension, neutropenia, arthralgias and abdominal pain.
No recurrences were observed. Eleven (92%) of patients are
alive, all without recurrence.
Conclusion: This pilot study of adjuvant sorafenib post LTx
revealed that 68% of targeted patients received adequate
treatment at tolerable doses of 300 mg daily. Recurrences
were not observed in this high risk group, although follow up
is short. The study gives insight to the current multicenter
sorafenib adjuvant post LTx trial.
OP-II.35 DIPS VS. TIPS IS THERE ANY

DIFFERENCE WHEN IT COMES TO
TRANSPLANTATION?
J. L. Pasko, R. C. Schenning, B. D. Petersen, S. L. Orloff
Oregon Health And Science University, Portland, OR
OP-II.34 A PILOT STUDY OF ADJUVANT

SORAFENIB IN HIGH RISK LIVER
TRANSPLANTATION FOR HCC
A. E. Alsina1,2, E. S. Franco1,2, A. Makris4, J. Shim3,
J. Allison3, R. Claudio3, M. Johnstone3, K. Barber5,
C. Albers1, N. Kemmer1
1
Tampa General Medical Group, Tampa, FL; 2University
Of South Florida, Department Of Surgery, Morsani
College Of Medicine, Tampa, FL; 3Tampa General
Hospital, Office Of Clinical Research, Tampa, FL;
4
University Of South Florida, School Of Business, Tampa,
FL; 5Tampa General Hospital, Tampa, FL
Introduction: Sorafenib can prevent post liver transplant
(LTx) recurrences for hepatocellular carcinoma (HCC) (Yan,
J. Liver Transpl 2013), but data in humans is scarce. We
hypothesized that recurrences in high risk HCC LTx can be
reduced.
Aims: Determine applicability, tolerability, and efficacy.
HPB 2015, 17 (Suppl. 1), 181
Background: Direct intrahepatic portocaval shunt (DIPS)

is a modification of transjugular intrahepatic portosystemic
shunt (TIPS) commonly used for ESLD patients with
refractory ascites, recurrent variceal bleeding, and end-stage
liver disease. DIPS uses ultrasound guidance to stent a
communication between the IVC and the portal vein. While
TIPS is more readily utilized, DIPS have shown to be
helpful for patients with difficulty anatomy and failed
TIPS.
Objective: To assess whether liver transplantation is safe
and feasible in recipients with DIPS. Secondary aims were to
do a matched comparison of DIPS vs.TIPS transplant recipients with respect to autologous and PRBC transfusions,
warm ischemia time, cold ischemia time, LOS, graft survival,
and overall patient survival.
Methods: Retrospective review January 1994December
2011 in a shared two academic institution transplant program
of patients who underwent liver transplantation with an existing DIPS or TIPS. The transplant operation was either
bicaval (before 2007), or onlay cavacavostomy (after 2008).
Results: Nine DIPS and 27 TIPS patients underwent transplantation during the study period. Nine DIPS were matched
to 18 TIPS patients by age, gender, etiologic disease, and
MELD at transplantation. There was was no statistical difference in amount of autologous and PRBC blood transfusion, warm ischemia, cold ischemia, LOS, graft survival, or
overall survival.
Conclusion: DIPS when compared to TIPS is safe, and feasible, and carries no added risk to patients undergoing liver
transplantation. Potential liver transplant candidates who are
referred for DIPS based on unfeasible or failed TIPS should
not be denied this treatment option.
2015 The Authors
68
Abstracts
OP-II.37 INCIDENCE AND LONG-TERM

SURVIVAL OF PATIENTS WITH
DE-NOVO HEAD AND NECK
CARCINOMA AFTER LIVER
TRANSPLANTATION
A. Andreou1, M. Lenarz2, D. Seehofer1, J. Pratschke1,
A. Coordes2
1
Department Of General, Visceral And Transplant Surgery,
Charite Campus Virchow-Klinikum, Berlin, BERLIN;
2
Department Of Otorhinolaryngology, Head And Neck
Surgery, Charite Campus Benjamin Franklin, Berlin,
BERLIN
Background: Liver transplant recipients have an increased
risk for the development of de-novo malignancies.
Methods: Clinicopathological data of patients who developed head and neck cancer after liver transplantation (LT)
were evaluated and predictors for overall survival (OS) were
identified.
Results: Thirty-three of 2040 patients (1.6%) who underwent LT between 1988 and 2010 developed de-novo
squamous cell carcinoma of head and neck (HNSCC). The
incidence of HNSCC in LT recipients with end stage
alcoholic liver disease was 5%. After a median follow-up of
9 years, 1-, 3-, and 5-year OS rates were 74%, 47% and 34%,
respectively. Tumor size, cervical lymph-node metastasis,
tumor site and tumor therapy (surgery only versus surgery
and adjuvant radio/chemoradiotherapy versus radio/
chemoradiotherapy only, P < 0.0001) were significantly
associated with OS in univariate analysis. However, surgery
only independently predicted OS in multivariate analysis.
Conclusions: Early diagnosis and surgical treatment for
de-novo HNSCC are crucial for outcome. Therefore,
HNSCC risk should be taken into close consideration during
post-transplant follow-up examinations especially among
patients with positive history of smoking and alcohol
consumption.
2015 The Authors

OP-II.38 IMPACT OF BODY MASS INDEX

ON LIVER TRANSPLANT PATIENT
SURVIVAL: 19942013
S. Ayloo1, P. Talbot2, M. Molinari1
Dalhousie University, Halifax, NS; 2Diabetes Care
Program Of Nova Scotia, Halifax, NS
1
Back Ground: Obesity is a world-wide epidemic impacting

the field of transplantation. In 2012, more than one-third of
the United States liver transplantations (LT) were performed
on patients with body mass index (BMI) 30.
Objectives: To explore the OPTN/UNOS database for
donor and recipient characteristics and transplant outcomes
among first time, whole organ, cadaveric LT patients stratified by different weight-status.
Methods: Retrospective study of the SRTR database from
January 1994 to September 2013 was conducted to analyze
donor/recipient characteristics and transplant outcomes
stratified by underweight (BMI 18.5), normal (BMI 18.5
24.99), overweight (BMI 2529.99), Class I (BMI
3034.99), class II (BMI 3539.99) and class III obese (BMI
40). Patients with missing BMI were excluded. Transplant
outcomes-Patient survival by Kaplan-Meier, cause of death,
length of hospital stay was analyzed.
Results: Of the 222,000 cases, 66,461 met the inclusion
criteria. The donor characteristics of mean age, median BMI,
cold ischemia time, percent male, Caucasian ethnicity and
leading cause of death of head trauma were 40 years,
25.4(7.473.22), 7 hours(049.5), 60%, 71% and 41%
respectively. Recipient characteristics and outcomes are presented in Table 1. Kaplan-Meier shows lower survival for
underweight and obese. BMI is significantly associated with
gender, ethnicity, education, functional status, primary diagnosis, comorbidities, graft failure (acute/de novo hepatitis),
primary cause of death.
Conclusion: LT in obese patients is increasing, with
increasing proportion of NASH as an indication for transplantation. Cardiovascular events were significantly higher in
higher BMI groups. Patient survival was lower in underweight and obese Class III in comparison to other BMI
grouping.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
OP-II.39 THE EFFECT OF A PORTAL

PRESSURE < 15 ON THE PREVENTION
OF SMALL-FOR-SIZE SYNDROME IN
ADULT-TO-ADULT LIVING-DONOR
LIVER TRANSPLANTATION: AN
OBSERVATIONAL STUDY
A. S. Helmy1, A. Abdelhady1, A. Hosny1, A. Dahaba3,
M. A. Al-Shazly1, M. El-mansy2, S. Uemoto4
1
Cairo University, Cairo, GIZA; 2Department Of Clinical
Pathology, Theodor Bilharz Research Institute, Cairo,
GIZA; 3Department Of Anaesthesiology And Intensive Care
Medecine, Graz, GRAZ; 4Department Of Surgery, Division
Of HPB Surgery And Transplantation, Graduate School Of
Medicine, Kyoto, KYOTO
Background: Small-for-size liver graft injury in Adult-toadult living-donor liver transplantation (A-LDLT) can contribute to severe postoperative graft dysfunction, known as
Small-for-size syndrome (SFSS). Recent evidence implicates
portal hypertension in SFSS pathogenesis. The aim of our
study was to investigate the effect of a portal venous pressure
(PVP) cut-off value of <15 mmHg on the prevention of SFSS
in A-LDLT.
Methods: Thirty-three patients underwent A-LDLT
between October 2009 and June 2013. We aimed to keep the
PVP under 20 mmHg at the end of the operation using graft
inflow modulation (GIM). Patients fell into 2 groups; group
A with final PVP <15 mmHg (n = 16) and group B with final
PVP 15 mmHg (n = 17). We diagnosed postoperative SFSS
according to the Clavien definition.
HPB 2015, 17 (Suppl. 1), 181
69
Results: Final PVP was controlled under 20 mmHg in all

patients. Three patients suffered SFSS in group B (17.6%)
compared with no patients in group A (P = 0.078). There was
a higher proportion of dead patients in group B [4, two of
whom died of SFSS (23.5%)] versus group A [1 (6.2%);
P = 0.166].
Conclusion: A final PVP <15 mmHg correlates with better
outcomes and seems to be a safe target level for the prevention of postoperative SFSS in A-LDLT.
OP-II.40 COMPLETE PATHOLOGIC

TUMOR KILL IS NOT REQUIRED AFTER
BRIDGING THERAPY FOR HCC
E. Beal, S. Black, M. Bloomston, C. Schmidt
The Ohio State University, Columbus, OH
Introduction: In patients with hepatocellular carcinoma
(HCC), bridging therapy is common prior to liver transplant
(LT) using ablation or trans-arterial chemoembolization
(TACE). It is unknown whether complete pathologic tumor
kill is needed for superior outcome.
Methods: The medical records of all patients who underwent LT between 2008 and 2013 at a single hospital were
reviewed. The incidence of viable HCC within the hepatic
explant was determined for patients who underwent LT after
bridging therapy. Outcomes were compared between those
with complete and incomplete tumor kill.
Results: There were 125 patients who underwent LT during
the study period, and 42 (34%) had HCC. Twenty-eight
2015 The Authors
70
Abstracts
(65%) were treated with bridging therapy, specifically 13

(46%) underwent ablation and 15 (54%) underwent TACE.
Viable tumor within the hepatic explant was found in 11
(69%) patients after ablation and 10 (67%) after TACE.
Median follow-up after LT in the HCC group was 17.3
months. Estimated 5-year survival was similar in the ablation
(75%) and TACE (76%) groups. One death occurred in the
perioperative period (3%). There was one patient with recurrent cancer in the cohort initially treated with ablation who
died 13.3 months after LT with multifocal disease.
Conclusions: After bridging therapy with either ablation or
TACE, rates of viable HCC are high in the explant after LT.
Despite this, recurrence is uncommon and overall survival is
comparable to patients transplanted without HCC. Depending on regional waitlist mortality, it may not always be necessary to achieve complete pathologic tumor kill with
bridging therapy.
OP-II.41 EXPANSION OF HPB AND

LIVER TRANSPLANTATION IN AN
UNDERSERVED STATE: CREATING
GROWTH AND MAINTAINING QUALITY
C. D. Anderson, A. Seawright, B. Borg, T. M. Earl
University Of Mississippi Medical Center, Jackson, MS
Significant disparities in HPB/liver transplant care prompted
expansion of Mississippis only transplant program. The strategic plan hypothesized that a systems based approach would
allow growth while maintaining quality.
Methods: Systems were implemented to start the program
in 8/2011 including a streamlined transplant clinic, increased
outreach, a robust QAPI system, and the creation of a liver
tumor program. This study examined institutional wide data
from 1/1/2010 until 3/31/2014. Period 1 (1/1/20107/31/
2011) represented pre-expansion, and period 2 (1/1/20133/
31/2014) was after full implementation. The performance of
the liver tumor, and liver transplant programs was analyzed.
Data is reported by median. Periods are compared using the
student t-test.
Results: From 8/2011 until 3/2014, 222 patients with liver
tumors were evaluated. Loco-regional liver tumor treatments
increased from 5/quarter to 25/quarter (p < 0.001). Major
HPB operations increased from 3/quarter to 14/quarter
(p < 0.001). This increase was within the transplant program
(14 vs 2/quarter). The first liver transplant occurred in
3/2013. 29 liver transplants were performed in 13 months.
1-month allograft and patient survivals were 93% and 100%.
Actuarial 1-year allograft and patient survivals were 89.66%
and 96.55%. MELD at transplant was 22 (range 1540). 35%
of transplant recipients were African American (13% regionally, 10% nationally). LOS was 9 days. 30 day readmission
was 20.69%.
Conclusion: A disparity in HPB and liver transplant access
existed in Mississippi which was addressed via a comprehensive expansion of the abdominal transplant programs. The
approach allowed rapid creation of a high volume transplant/
HPB center and improved patient access while maintaining
expected outcome metrics.
OP-II.43 RACIAL DISPARITY IN LIVER

TRANSPLANTATION IN AN INNER
CITY POPULATION
A. Smith, A. T. Hauch, E. Kandil, A. Paramesh,
M. Killackey, M. Moehlen, L. A. Balart, J. F. Buell
Tulane Transplant Institute, Tulane University And
Louisiana State University Medical System, New Orleans,
LA
Several studies have identified disparity in access to liver
transplantation among African Americans. This study examines the experience of an inner city University transplant
program with a significant proportion of socioeconomically
challenged African American patients.
Methods: A retrospective analysis of all liver transplant
(LT) recipients at our institution from 20072014. Pretransplant demographics and post-transplant outcomes were
analyzed. A multivariate analysis was performed to examine
patient and graft survival as well as risk for complications.
Conclusions: African American patients that were successfully transplanted at our institution were significantly
younger with trends of lower portal vein thrombosis and
acutely decompensated (higher INR, inpatient transplant
event). Despite limited evidence these data suggest a selection bias. To further delineate the origins of this bias we plan
on further analyzing waitlist mortalities and dropout events
along racial lines to identify if African Americans are at
higher risk for dropout under this current allocation system.
OP-II.44 ADDITION OF STEREOTACTIC

BODY RADIATION TO PREVENT LIVER
TRANSPLANT WAITLIST DROPOUT:
IS IT SAFE?
T. J. Smith3, J. E. Hooper6, A. E. Castillo1,2, W. Naugler5,
P. Worth1,2, K. Kolbeck7, K. Farsad7, K. Enestvedt1,2,
M. Fuss4, T. Mitin4, C. Thomas4, S. Orloff1,2
1
Oregon Health And Science University. Department Of
Surgery, Portland, OR; 2Portland VA Medical Center.
Department Of Surgery, Portland, OR; 3Gundersen Health
System. Department Of Surgery, La Crosse, WI; 4Oregon
Health And Science University. Department Of Radiation
Oncology, Portland, OR; 5Oregon Health And Science
University. Department Of Medicine, Portland, OR; 6Johns
Hopkins. Department Of Pathology, Baltimore, MD;
7
Oregon Health And Science University. Department Of
Radiology, Portland, OR
Background: Waitlist dropout is a major consideration in
liver transplantation (LT) for HCC. Loco-regional treatment
is used to prevent dropout. Little is known about Stereotactic
2015 The Authors

HPB 2015, 17 (Suppl. 1), 181
Abstracts
Body Radiation (SBRT) safety and its influence on waitlist
dropout, perioperative LT complications, or recurrence
post-LT.
Aim: Whether adding SBRT to HCC LT waitlist patients having received Trans-Arterial-Chemo-Embolization
(TACE), impacts the safety/efficacy profile as measured by
waitlist dropout, perioperative complications or posttransplant HCC recurrence.
Methods: Retrospective analysis from a two-institution
transplant program, of 10 consecutive waitlisted HCC
patients receiving SBRT + TACE matched with 10 such
patients receiving TACE only.
Results: Median treatment follow-up: SBRT + TACE group
45 vs 43 mo TACE-only group; median LT follow-up: 34 and
38 mo, respectively. Mean wait-time for the SBRT-group was
330 d vs 150 d for the TACE-only group. At last follow-up
8/10 SBRT-group patients were alive vs 6/10 in the TACEonly group. One HCC recurrence occurred in the SBRTgroup vs two in the TACE-only group. All HCC recurrences
died. Pre-transplant median AFP was higher in the SBRTgroup. Tumor response by explant percent necrosis was
similar between groups. LOS, ICU days and median EBL
trended higher for the SBRT group (p = 0.60). There were no
waitlist drop-outs in either group.
Conclusions: Despite longer waitlist time, HCC patients
receiving SBRT + TACE had no waitlist drop-out, and lower
HCC recurrence post-LT. SBRT + TACE patients trended
toward more difficult operations, but no significant difference
in post-LT survival. Thus, addition of pre-LT SBRT to TACE
appears safe and effective as a bridge to LT. Future prospective randomized clinical trials are warranted.
SUNDAY, MARCH 15, 2015,

7:30AM8:30AM
VIDEO A LIVER
V-A.01 TIPS AND TRICKS FOR
GLISSONIAN APPROACH DURING
LAPAROSCOPIC RIGHT HEPATECTOMY
F. F. Makdissi, R. C. Surjan, M. A. Machado
Sirio Libanes Hospital, Sao Paulo, SAO PAULO
Background: Laparoscopic Glissonian approach is a technique that avoids dissection of the hilar plate. It is useful for
anatomical liver resection and can reduce the operative time.
However, it has not been frequently used because needs
expertise and knowledge of liver anatomy.
Aim: To present a video of a laparoscopic right hepatectomy
using the Glissonian approach. The detailed technique for
control of the right pedicle is highlighted in this video.
Patient and Methods: A 29-year-old man with anabolic
steroids abuse presented with an acute abdominal pain. CT
scan showed 11-cm mass in the right liver. Patient is referred
for surgical treatment. Operation begins with liver mobilization. Glissonian approach of the right pedicle is achieved
with two small incisions made in specific anatomic landmarks around hilar plate. Clamp is introduced through these
incisions to occlude right Glissonian pedicle. The vascular
clamp is then replaced by an endoscopic vascular stapling
HPB 2015, 17 (Suppl. 1), 181
71
device, the right liver ischemic delineation is confirmed, and

the stapler is fired. All of these steps are performed without
the Pringle maneuver and without hand assistance. Liver
transection and vascular control of the right hepatic vein are
accomplished with as usual. The specimen is extracted
through suprapubic incision.
Results: Operative time was 160 minutes with minimum
blood loss. Recovery was uneventful and patient was discharged on the third postoperative day. Final pathology
showed liver cell adenoma with no signs of malignancy and
free margins.
Conclusion: Laparoscopic Glissonian approach for right
hepatectomy is safe, fast and can be accomplished with
proper technique.
V-A.02 CAVO-ATRIAL THROMBECTOMY

COMBINED WITH RIGHT
HEPATECTOMY EXTENDED TO
SEGMENT IV FOR VASCULAR INVASION
FROM HEPATOCELLULAR CARCINOMA
THROUGH AN ABDOMINAL,
TRANSDIAPHRAGMATIC AND
INTRAPERICARDIAC APPROACH
E. Vicente1,2,3, Y. Quijano1,2,3, B. Ielpo1,2,3, H. Duran1,2,3,
I. Fabra1,2,3, E. Diaz1,2,3, S. Olivares1,2,3, R. Caruso1,2,3,
A. Prestera1,2,3, M. De Luca1,2,3, J. Maupoey1,2,3,
E. Vicente1,2,3
1
Sanchinarro University Hospital., Madrid, MADRID;
2
Clara Campal Oncological Center, Madrid, MADRID;
3
San Pablo University. CEU, Madrid, MADRID
Vascular invasion of supra-hepatic veins is a major complication of primary liver tumours. The tumoral thrombus, when
extended to the vena cava and right atrium, may produce
occlusion of the tricuspid valve or pulmonary embolism with
sudden cardiac death. The presence of macroscopic vascular
infiltration represents an advanced stage of the tumour
contraindicating liver transplantation, thus liver resection
with thrombectomy is the only therapeutic option in this
setting despite the concerns of postoperative liver failure and
the dismal results at distance A 62-year-old female without
chronic liver disease was referred to our Hospital diagnosed
of the tumour located in the right hemi-liver with infiltration
of the right hepatic veins and a tumour thrombus extension to
the retrohepatic and suprahepatic inferior vena cava and right
atrium. Routine blood tests revealed normal hematological
results and liver biochemistry function. A computed tomography (CT) scan, Nuclear magnetic resonance (MRI) demonstrated the tliver tumor and the thrombus extending into
retrohepatic, supraphrenic-intrapericardial IVC and right
atrium. Right portal vein embolization was performed before
the surgical resection The video shows the preoperative diagnostic procedures to evaluate the location of the tumor, the
vascular invasion and extent of thrombus as well as the surgical procedure. Through a bilateral subcostal laparotomy,
right hepatectomy extended to segment IV was performed.
The transabdominal and transpericardial approach to the
intrapericardial IVC and right atrium avoided median
sternotomy. Intravascular trombectomy with intraoperative
transesophageal echocardiogram was also performed. Two
years after the operation, the patient remains alive and well
without evidence of tumoral recurrence.
2015 The Authors
72
Abstracts
V-A.03 EXTENDED LEFT

HEPATECTOMY WITH COMPLEX
PORTAL VEIN RECONSTRUCTION AND
IN SITU COLD PERFUSION OF THE
LIVER FOR HILAR
CHOLANGIOCARCINOMA
G. Sapisochin, R. Smoot, J. Qu, A. Fung, J. B. Conneely,
P. Kelly, G. Tait, L. Hotoyan, P. D. Greig, I. D. McGilvray
Department Of Surgery. Toronto General Hospital.
University Of Toronto, Toronto, ONTARIO
Tumors of the biliary confluence can pose significant technical challenges when attempting to safely achieve negative
margins. Major vascular involvement precludes standard
extended resection, often reducing therapeutic options to
transplantation or non-resective modalities. We present a
case of a left-dominant Klatskin tumor in a young patient,
involving the middle hepatic vein and abutting the portal vein
bifurcation. Significantly, the portal inflow to the Right Posterior Section comprised of segment VI and accessory
segment VI/VII portal vein branches, which would require
separate reconstruction if resection was to be considered. We
performed and extended left hepatectomy with resection of
the portal bifurcation and reconstruction of both branches
of the portal vein. In order to protect the remnant liver during
the planned vascular reconstruction, in-situ cold perfusion of
the liver with preservation solution was utilised. The accompanying video is comprised of real-time operative video. This
complex case illustrates the facility of employing advanced
transplantation techniques to expand therapeutic options in
the oncology setting.
V-A.04 TRANSTHORACIC PORT

PLACEMENT INCREASES SAFETY OF
TOTAL LAPAROSCOPIC POSTERIOR
SECTIONECTOMY
L. Schwarz1, T. A. Aloia1, C. Eng2, G. J. Chang1,
J. Vauthey1, C. H. Conrad1
1
UT MD Anderson Cancer Center, Department Of Surgical
Oncology, Houston, TEXAS; 2UT MD Anderson Cancer
Center, Department Of GI Medical Oncology, Houston,
TEXAS
Background: An anatomic posterior sectionectomy is infrequently performed due to the challenges of controlling the
right posterior portal pedicle (RPPP) while preserving the
anterior pedicle (RAPP), difficulty of visualizing the drainage of the right hepatic vein into the IVC, and the potential
for significant blood loss during the caval and hepatovenous
dissection.
Patient: A 62-year-old woman with 3 liver metastases to
SVI and SVII from sigmoid colon cancer underwent 5 cycles
of neoadjuvant chemotherapy with FOLFOX and 4 of
bevacizumab with a type I response. She underwent a
Primary First robotic low anterior rectosigmoid resection
followed by a laparoscopic posterior sectionectomy.
Technique: The patient was placed in a Modified French
Position. A transthoracic trocar was placed for optimal
laparoscopic visualization and access of the superior
retrohepatic IVC and drainage of the right hepatic vein into
IVC. Intraoperative ultrasound was crucial to assess tumor
2015 The Authors
location, define transection plane and preserve flow to RAPP

before division of RPPP. The parenchymal transection
follows an oblique angle and exposes the right hepatic vein.
Conclusion: Transthoracic port placement augments the
safety of the dissection along the IVC inferiorly and the right
hepatic vein superiorly due to direct visualization. Also,
it provides a direct instrument-to-target axis without the
typical fulcrum of dissecting the postero/superior liver.
Laparoscopic ultrasound is critical to confirm preserved flow
to the RPPP and guide the parenchymal transection. Liver
volumetry should be obtained prior to surgery to determine
adequate future liver remnant if conversion to a right lobectomy becomes necessary.
V-A.05 TOTALLY LAPAROSCOPIC TWO

STAGE HEPATECTECTOMY WITH
PORTAL VEIN EMBOLIZATION
W. Kuo1,3, C. U. Corvera1,2
University Of California, San Francisco, San Francisco,
CA; 2VA Medical Center San Francisco, San Francisco,
CA; 3Chiayi Christian Hospital, Chia-yi City, TAIWAN
1
54 year-old woman with synchronous metastatic rectal

cancer to the liver. Imaging showed several hepatic tumors
distributed in the both lobes of the liver; R > L. A two stage
procedure with right PVE was planned. This video shows a
limited partial left hepatectomy that was followed on POD #1
with an immediate right PVE. Six weeks later, a right hepatectomy was completed. Stage 1. Using three 5 mm ports and
a 12 mm periumbilical port, a partial segment # 2 liver resection and wedge resection of segment #3 was done using the
bipolar cautery device and vascular stapler. Stage 2. Six
weeks after PVE, she was returned to the operating room for
R hepatectomy. The hepatic veins were exposed. Hilar dissection began by exposing the RHA lying posterior to the
common hepatic duct. It was clipped and divided. The main
PV was dissected until the bifurcation was identified. A
caudate hepatotomy was done to allow isolation and division
of the main RPV by a single firing of a stapler. The line of
demarcation was marked using electrocautery and the parenchymal transection was started. As the liver was opened, the
right bile duct was transected intrahepatically. Coursing
branches of the middle and right hepatic veins were divided
intrahepatically using a stapler. The parenchymal transection
plane was aligned along the anterior surface IVC until the
liver was completely divided. The RHV was identified and
divided and the right hemi-liver was mobilized from diaphragm. The specimen was extracted via a low transverse
incision.
HPB 2015, 17 (Suppl. 1), 181
Abstracts

7:30AM8:30AM
VIDEO B PANCREAS/BILIARY
V-B.01 INTRAOPERATIVE AIR LEAK
TEST PREVENTS POST-HEPATECTOMY
BILE LEAK
R. W. Day, W. R. Burns, C. Conrad, J. Vauthey, T. A. Aloia
Department Of Surgical Oncology, The University Of Texas
MD Anderson Cancer Center, Houston, TEXAS
Background: After major liver resection, the incidence of
bile leak is reported at 410%. We have previously reported
that an intraoperative Air Leak Test (ALT) reduces the rate of
bile leak by five-fold with no additional morbidity or mortality.
Purpose: Demonstrate the steps of an intraoperative ALT
following major liver resection.
Results: A 29 year-old man presented with T3N1M1 rectal
cancer and synchronous liver metastases. There were two
right liver metastases, one at the right portal junction necessitating right hepatectomy. After neoadjuvant FOLFOX and
bevacizumab chemotherapy, chemoradiotherapy to his
primary rectal cancer, and abdominoperineal resection, he
presented for liver resection. His liver metastases had radiographically responded with CEA decrease from 8.5 to 2.3 ng/
mL. The functional liver remnant volume was calculated to
be 31%. The video briefly describes key components of the
right hepatectomy including extrahepatic vascular inflow
control, two-surgeon technique for hepatic parenchymal
transection, and intrahepatic stapled division of the right bile
duct. It then details the sequential steps of the ALT including
cannulation of the cystic duct, Doppler of intrahepatic vascular flow, manual occlusion of the bile duct, injection of air
under ultrasound visualization to confirm patency of the
biliary confluence, and repeated injection of air to identify
and oversew occult open bile ducts. At the completion of the
ALT, the patient had no air leaks and no drain was placed. He
experienced an uneventful recovery.
Conclusion: For patients undergoing major hepatectomy,
intraoperative ALT is a rapid, safe, and inexpensive means to
decrease the incidence of post-operative bile leak.
V-B.02 MINIMALLY INVASIVE PARTIAL

RIGHT HEPATECTOMY WITH
TRANSDIAPHRAGMATIC RIGHT
LOWER LOBE LUNG RESECTION FOR
COLORECTAL METASTASES
R. W. Day1, R. J. Mehran2, N. De Rosa1, C. Conrad1,
T. L. Moon3, J. Vauthey1, T. A. Aloia1
1
Department Of Surgical Oncology, The University Of
Texas MD Anderson Cancer Center, Houston, TEXAS;
2
Department Of Thoracic And Cardiovascular Surgery,
The University Of Texas MD Anderson, Houston, TEXAS;
3
Department Of Anesthesiology And Perioperative
Medicine, The University Of Texas MD Anderson,
Houston, TEXAS
73
both sites of metastases are resected. The main obstacle to

simultaneous resection is the additional morbidity of a thoracic incision. We have recently published an experience
with open hepatectomy and transdiaphragmatic lung resection, demonstrating the feasibility and safety of this
approach.
Purpose: In an effort to further improve outcomes we investigated the ability to perform the simultaneous procedure
using minimally invasive techniques.
Results: A 40 year-old woman with pT1N0 colon cancer
status post low anterior resection represented with an
elevated carcinoembryonic antigen (CEA = 4.5 ng/mL).
Imaging demonstrated three liver lesions, biopsy positive for
metastasis and one right lower lobe pulmonary lesion.
Preoperatively, she received three cycles of FOLFOX/
bevacizumab treatment and one cycle of FOLFOX alone
with radiological and biochemical response (CEA = 2.5 ng/
mL). In this video we demonstrate the key components of her
hand-assisted, laparoscopic simultaneous multifocal liver
and lung tumor resection including the use of a rubber band
retraction technique, resection of a disappeared 3 mm lesion,
and the transdiaphragmatic lung resection with EndoStitch
diaphragm closure. Final pathology confirmed margin negative resection of all 4 tumors. The patient was discharged
from the hospital on POD #4 with no medical or surgical
complications and a rapid return to normal function, as measured by a validated quality of life assessment tool.
Conclusion: In carefully selected patients with metastatic
colorectal cancer, minimally invasive transdiaphragmatic
approaches to lung resection simultaneous with liver resection are technically feasible and safe to perform.
V-B.03 LAPAROSCOPIC MAJOR

ARTERIAL RESECTION DURING
B. Franssen, M. Kendrick
With recent improvements in multimodality treatment for
borderline and locally advanced pancreatic cancer, major
vascular resection during pancreatic surgery has become
more frequent. Although portal vein and SMV resection in
both open and laparoscopic pancreaticoduodenectomy (PD)
has been previously described, the feasibility of major arterial resection in laparoscopic pancreatectomy has not been
established. While major arterial resection remains controversial, it is performed in our center in very select patients in
order to achieve an R0 resection during PD. The aim of this
video presentation is to demonstrate technical aspects and
plausibility of a common hepatic artery resection and reconstruction in combination with a portal vein resection in a
patient undergoing laparoscopic PD with a radiographically
stable ductal adenocarcinoma diagnosed two years prior to
surgery.
Background: Patients with metastatic colorectal disease to

the liver and lungs have improved long-term outcomes when
HPB 2015, 17 (Suppl. 1), 181
2015 The Authors

74
Abstracts
V-B.04 MODIFIED ROBOTIC APPLEBY

PROCEDURE WITH LEFT
NEPHRECTOMY AND
ADRENALECTOMY FOR
LOCALLY ADVANCED DISTAL
PANCREATIC CANCER
M. Radomski, A. Zureikat, H. Zeh, M. Hogg
University Of Pittsburgh, Pittsburgh, PA
Patients presenting with Stage III pancreatic adenocarcinoma
involving the distal pancreas and celiac axis have historically
been treated by nonsurgical means. However, almost a third
of these patients will die without metastatic disease suggesting that aggressive surgical resection involving visceral vascular structures may offer a survival advantage. Appleby first
described resection of the celiac axis for gastric cancer. We
have extended this technique to include stage III distal pancreatic adenocarcinoma involving the celiac axis without evidence of metastatic disease via a combined laparoscopic and
robotic approach. We describe a robotic distal pancreatectomy, splenectomy, left nephrectomy, and left adrenalectomy
in a 65 year old female patient with pancreatic adenocarcinoma involving the distal pancreas without evidence of
distant metastatic disease. She previously underwent six
cycles neoadjuvant chemotherapy with gemcitabine and
abraxane in which she had a good response with normalizing
CA19-9 values. Her postoperative course was complicated
by gastric ischemia which was treated with short course of
total parenteral nutrition, and a portal vein thrombus treated
with subcutaneous low molecular weight heparin. Her
pathology showed a 4.0 cm moderately differentiated adenocarcinoma with negative margins and twenty-six lymph
nodes negative for disease. She was discharged home on
postoperative day 25 and started on adjuvant chemotherapy
with gemcitabine and abraxane on postoperative day 76.
V-B.05 ROBOTIC
PANCREATICODUODENECTOMY WITH
PORTAL VEIN RESECTION AND PATCH
VENOPLASTY
M. Girgis, M. Hogg, H. Zeh, A. Zureikat
University Of Pittsburgh School Of Medicine, Pittsburgh,
PA
Introduction: Robotic pancreaticoduodenectomy is safe
and feasible for resectable pancreatic head adenocarcinoma.
Vascular resection during pancreaticoduodenectomy for
borderline resectable PDAC may be associated with slightly
increased morbidity but provides good oncologic outcomes.
We describe a robotic pancreaticoduodenectomy with
planned portal vein resection and patch venoplasty in a
patient with borderline resectable pancreas cancer.
Methods: This patients operation was recorded by the Da
Vinci Surgical System. The file footage was compiled and
edited. The patients consent was obtained.
Results: This is the case of a 57-year-old patient with
locally advanced pancreas cancer. He underwent 6 cycles of
neoadjuvant chemotherapy. Restaging CT scan showed
down-staging of the tumor to borderline resectable status.
The procedure was completed robotically in 452 minutes
with an estimated blood loss of 900 ml. A large 4 cm partial
2015 The Authors
portal vein resection was performed with reconstruction

using bovine pericardium in the form of a patch venoplasty.
Final pathology revealed a 2.9 cm tumor with 4/37 nodes
positive. He recovered uneventfully from the surgery and was
discharged from the hospital on postoperative day 6. Within
45 days of surgery, he had initiated adjuvant chemotherapy.
Conclusions: We describe a robotic portal vein resection
and reconstruction with patch venoplasty for borderline
resectable PDAC. We were able to complete the procedure
safely and in a timely manner with minimal morbidity.

8:30AM10:30AM
LONG ORAL J LIVER
PERIOPERATIVE/TECHNIQUES
LO-J.01 ELEVATED INR
INDEPENDENTLY PREDICTS MAJOR
COMPLICATIONS FOLLOWING MAJOR
HEPATECTOMY
J. N. Leal, T. P. Kingham, P. J. Allen, R. P. DeMatteo,
W. R. Jarnagin, M. I. DAngelica
Memorial Sloan Kettering Cancer Center, New York, NY
Background: The impact of major hepatectomy on
hemostasis is poorly characterized and accuracy of INR as an
indicator of coagulation status in the peri-operative setting
is unknown. This study aimed to characterize posthepatectomy hemostatic profiles and to evaluate subsequent
impact on clinical outcomes.
Methods: All patients undergoing major hepatectomy
between 20012012 were identified from a prospective database. Clinicopathologic and blood product transfusion data
were extracted. Post-operative hemostatic profiles were
analyzed with respect to plasma transfusion and major complications.
Results: 961 patients were identified. The mean number
of Couinaud segments resected was 4.3(+/0.85).Postoperatively peak INR was 1.50 +/ 0.26, platelet nadir
166.1 +/ 61.7 K/mcL, and hemoglobin nadir 9.5 +/ 1.34 g/
dL. INR 1.8 occurred in 152 (16%) patients. Major bleeding
complication rates in the INR 1.8 group were not different
from INR <1.8 group (0.7% vs. 0.8%, p = 0.90). Alternatively, major complication rates were significantly higher
(37% INR 1.8 vs. 17% INR <1.8, p < 0.0001). Among the
INR 1.8 group, plasma was transfused in 66% of cases and
was not associated with a significant change in major complication rates (40% in peak INR 1.8 with plasma vs. 30%
in peak INR 1.8 without plasma, p = 0.30). On multivariate
analysis, including standard variables, peak INR 1.8 was
independently associated with major complications (OR
2.48, 95% CI 1.643.74, p < 0.0001) while FFP transfusion
was not (OR 1.35 95% CI 0.902.05, p = 0.152).
Conclusions: Elevated INR following major hepatectomy
does not increase the risk of post operative hemorrhage but
rather portends increased overall morbidity that is not mitigated by FFP transfusion.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
LO-J.02 EARLY INCREASES IN

POSTOPERATIVE SERUM
PHOSPHOROUS AND CREATININE
LEVELS ARE ASSOCIATED WITH
MORTALITY FOLLOWING
MAJOR HEPATECTOMY
G. Herbert1, W. Jarnagin1, R. DeMatteo1, P. Allen1,
M. DAngelica1, M. Gonen1, K. Prussing2, T. P. Kingham1
1
Memorial Sloan Kettering Cancer Center, New York, NY;
2
University Of Minnesota Medical School, Duluth, MN
Introduction: Post-hepatectomy liver failure (PHLF) is
defined by the International Study Group of Liver Surgery as
a bilirubin >2.9 mg/dL and INR >1.7 on or after postoperative day 5 (the 50/50 criteria), and greatly increases the risk
of death after hepatectomy. Other than pre-existing liver
disease and small future liver remnant, few patient factors or
early post-operative indicators identify patients at elevated
risk of PHLF.
Methods: We reviewed demographics, comorbidities,
operative procedures, and post-operative laboratory trends of
patients undergoing major hepatectomy (>= 3 Couinaud segments) for malignancy from 19982013 at our institution.
These factors were compared between patients dying within
90 days of surgery, survivors meeting the 50/50 criteria, and
all remaining survivors.
Results: 1536 patients underwent major hepatectomy
during the time period. The majority were performed for
metastatic colorectal cancer, with an average of 4.24 resected
segments. 49 (3.2%) died within 90 days of surgery. 47
(3.1%) patients met ISGLS criteria for PHLF, of whom 15
died within 90 days of surgery. Operative blood loss was
twice as high in patients dying within 90 days as compared to
survivors. In spite of greater perioperative resuscitation
(achieving equivalent urine output) when compared to survivors, non-survivors had significantly higher creatinine and
phosphorous levels on POD 1. Predicted residual liver
volume did not differ between patients dying within 90 days
and survivors who met the 50/50 criteria (32.1% vs 31.1%).
Conclusions: An increase in both creatinine and phosphorous on POD 1 is an early indicator of patients at risk for
PHLF.
HPB 2015, 17 (Suppl. 1), 181
75
LO-J.03 EFFECTS OF BLOOD

TRANSFUSION ON PERIOPERATIVE
AND LONG-TERM OUTCOMES AFTER
MAJOR HEPATECTOMY FOR
METASTATIC COLORECTAL CANCER:
A MULTI-INSTITUTIONAL STUDY OF
456 PATIENTS
L. M. Postlewait1, M. H. Squires1, D. A. Kooby1,
S. M. Weber2, C. R. Scoggins3, K. Cardona1, C. S. Cho2,
R. C. Martin3, E. Winslow2, S. K. Maithel1
1
Division Of Surgical Oncology, Emory University, Atlanta,
GA; 2Division Of Surgical Oncology, University Of
Wisconsin, Madison, WI; 3Division Of Surgical Oncology,
Background: Data on the prognostic implications of blood
transfusion at the time of major hepatectomy (3 segments)
for colorectal cancer metastases are conflicting. Our aim was
to assess the association of perioperative transfusion with
postoperative complications and disease-specific survival
(DSS).
Methods: Patients who underwent major hepatectomy for
metastatic colorectal cancer from 20002010 at three US
academic institutions were included. 30-day mortalities and
patients who died of unknown cause were excluded from
survival analyses. Transfusion was analyzed based on timing
and volume of transfusion.
Results: Of 456 patients, 140 (30.7%) received blood transfusions. Perioperative transfusion was associated with
extended hepatectomy (40.8%vs27.9%; p = 0.020),
increased tumor size (5.7 vs 4.2 cm; p < 0.001), and
increased estimated blood loss (917 vs 390 mL; p < 0.001).
There were no differences between patients with or without
transfusion regarding age, ASA class, margin, number of
lesions, cirrhosis, or lymphovascular invasion. On multivariate analysis, perioperative transfusion was associated with
major complications (HR3.14; 95%CI: 1.835.39; p < 0.001)
and 90-day readmission (HR2.14; 95%CI: 1.183.89;
p = 0.012). 388 patients were included in survival analyses;
median follow-up was 38.8 mos. Perioperative transfusion
was not associated with DSS; however, patients who received
blood postoperatively had decreased DSS (37.4 vs
42.7 mos;p = 0.044). Increasing volume of transfusion at any
time was associated with shortened DSS (Perioperative: 0
units: 42.1 vs 12 units: 39.6 vs 3 units: 37.4 mo, p = 0.046;
2015 The Authors
76
Abstracts
Postoperative: 0 units: 42.7 vs 12 units: 38.6 vs 3 units:

27.2 mo, p = 0.031; Figure). Transfusion, regardless of
timing and volume, was not independently associated with
worsened DSS on multivariate analyses.
Conclusion: Perioperative blood transfusion with major
hepatectomy for colorectal cancer metastases is independently associated with increased complications and readmission but not reduced disease-specific survival. Judicious use
of blood transfusion in the perioperative period is warranted.
Final pathologic diagnoses were SLC (n = 73), BCA

(n = 15), BCAC (n = 3) and other primary liver cancers
(n = 3). On multivariate analysis, preoperative imaging features such as solitary lesion, septations and solid component
were associated with a final diagnosis of BCT (p < 0.05;
Table). However, one-third of SLC presented with septations
or solitary lesions. Five patients (23.8%) were initially misdiagnosed as SLC and 24 patients (32.9%) were misdiagnosed as BCT of which 15 patients unnecessarily underwent
resection. Fenestration and frozen section (FS) was performed in 47 patients (diagnostic accuracy = 100%),
impacting surgical management in 12 patients (12.7%) by
avoiding (n = 8) or mandating (n = 4) liver resection. No
patients with BCT (n = 8), submitted to fenestration and FS
followed by resection, developed recurrence.
Conclusion: Worrisome imaging features are associated
with a higher risk of BCT. However, one-third of patients
with SLC present with such imaging characteristics. In the
absence of a strong suspicion of malignancy, fenestration and
FS should be considered prior to complete resection.
LO-J.04 THE ACCURACY OF

PREOPERATIVE IMAGING AND
INTRAOPERATIVE FROZEN
SECTION IN THE MANAGEMENT
OF HEPATIC CYSTS
A. Doussot1, B. Groot Koerkamp1, P. J. Allen1,
R. P. De Matteo1, J. Shia2, T. P. Kingham1, W. R. Jarnagin1,
S. R. Gerst3, M. I. DAngelica1
1
Department Of Surgery Memorial Sloan Kettering Cancer
Center, New York, NY; 2Department Of Pathology
Memorial Sloan Kettering Cancer Center, New York, NY;
3
Department Of Radiology Memorial Sloan Kettering
Cancer Center, New York, NY
Introduction: Biliary cystic tumors (BCT) (biliary
cystadenoma (BCA) and cystadenocarcinoma (BCAC))
warrant complete resection. Simple liver cysts (SLC),
however, require only fenestration if symptomatic. Distinguishing between BCT and SLC with preoperative imaging
is not well studied.
Methods: All patients undergoing surgery for a preoperative
diagnosis of SLC or BCT were included. Perioperative clinical and imaging features from radiology reports were
analyzed for their ability to predict the final diagnosis.
Results: Ninety four patients underwent fenestration
(n = 53), enucleation (n = 6) or liver resection (n = 35) for
suspected SLC (n = 54), BCA (n = 32) and BCAC (n = 8).
2015 The Authors
LO-J.05 THE IMPACT OF PORTAL

PEDICLE CLAMPING ON SURVIVAL
FROM COLORECTAL LIVER
METASTASES IN THE CONTEMPORARY
ERA OF LIVER RESECTION: A
MATCHED COHORT STUDY
M. E. Tsang1, P. J. Karanicolas1,2, R. Habashi1, E. Cheng1,
S. Hanna1,2, N. G. Coburn1,2, C. H. Law1,2, J. Hallet2
1
Division Of General Surgery, University Of Toronto,
Toronto, ON; 2Sunnybrook Health Sciences Centre Odette
Cancer Centre, Toronto, ON
Introduction: Portal pedicle clamping (PPC) reduces bleeding during hepatectomy, but may impact micro-metastases
HPB 2015, 17 (Suppl. 1), 181
Abstracts
growth through ischemia-reperfusion injury. We sought to
examine the association between PPC and long-term survival
following hepatectomy for colorectal liver metastases
(CRLM).
Methods: We conducted a matched cohort study using our
prospective hepatectomy database to identify all patients
undergoing hepatectomy for CRLM from 2003 to 2013.
Cohorts were selected based on use of PPC, with 1 : 1 matching for age (5-year increments), time period (20032007 vs.
20072013), and Clinical Risk Score (0 to 5 scale). Primary
outcome was overall survival (OS). Conditional logistic and
Cox regression analyses determined odds ratios (OR) and
hazard ratios (HR).
Results: Of 481 hepatectomies for CRLM, 187 (39%)
patients underwent PPC. 110 pairs of patients were matched
in the cohorts, and the remainder excluded. Peri-operative
chemotherapy (p = 0.183), major hepatectomy (>= 3 segments) (p = 0.345) or resection status (R0 vs. R12)
(p = 0.132) did not differ. 30-day major morbidity (OR 0.73;
p = 0.332) and mortality (OR 2.4; p = 0.100) were not significantly associated with PPC. Median follow-up was 35
(range: 0130) months. When adjusting for extent of resection, blood loss, and operative time, no significant difference
was observed in OS (HR 1.76; p = 0.129) for PPC, with
5-year OS of 59.2% (95%CI: 54.059.2%) for PPC and
62.3% (95%CI: 61.867.3%) without PPC (log-rank
p = 0.415). Excluding 90-day deaths did not substantially
alter the results (log-rank p = 0.930).
Conclusions: PPC was not associated with a significant difference in OS in patients undergoing hepatectomy for
CRLM. It does not appear to adversely affect oncologic
outcomes.
HPB 2015, 17 (Suppl. 1), 181
77
LO-J.06 HEPATO-PANCREATECTOMY:
HOW MORBID? RESULTS FROM THE
NATIONAL SURGICAL QUALITY
IMPROVEMENT PROGRAM
T. B. Tran, M. M. Dua, D. A. Spain, B. C. Visser,
J. A. Norton, G. A. Poultsides
Stanford University School Of Medicine, Stanford, CA
Background: Simultaneous resection of both the liver and
pancreas carries significant complexity. The objective of this
study is to investigate perioperative outcomes following synchronous hepatectomy and pancreatectomy (SHP).
Surgical Quality Improvement Program (ACS-NSQIP)
database was queried to identify patients who underwent
SHP. Resections were defined as follows: <hemihepatectomy, hemihepatectomy (hemihepatectomy and
trisectionectomy), PD (pancreaticoduodenectomy), and
distal (distal pancreatectomy and enucleation).
Results: From 2005 to 2012, 377 patients underwent SHP,
representing 1% of 38,568 patients who underwent hepatectomy and/or pancreatectomy. Median age was 60 years. Indications included pancreatic malignancy (60%), hepatobiliary
malignancy (10%), benign disease (6%), gastric cancer (4%),
retroperitoneal neoplasm (3%), colon cancer (2%), and other
(15%). Patients were stratified based on the extent of combined resection: <hemihepatectomy + distal (n = 170),
hemihepatectomy + distal (n = 37), <hemihepatectomy +
PD (n = 151), and hemihepatectomy + PD (n = 19).
Perioperative morbidity and mortality gradually increased as
resections became more extensive (Figure). Although the
first 3 groups had a reasonable morbidity and mortality
profile, in the latter group of hemihepatectomy + PD the
rate of any complication was 84%, septic shock 26%,
reintubation 16%, hemodialysis 10%, 30-day mortality 10%,
and in-hospital mortality 33%.
Conclusions: Liver resections less extensive than hemihepatectomy can be performed concurrently with any pancreatectomy resulting in an acceptable safety profile. Synchronous hemihepatectomy (or trisectionectomy) and
pancreaticoduodenectomy remains a highly morbid combination and should be reserved for patients who have undergone extremely cautious selection.
2015 The Authors

78
Abstracts
LO-J.07 PORTAL VEIN EMBOLISATION

AND APPLICATION OF
HAEMATOPOIETIC STEM CELLS IN
PRIMARILY NON-RESECTABLE
COLORECTAL LIVER METASTASES
FOR INSUFFICIENT FUTURE LIVER
REMNANT VOLUME
V. Treska1,2
University Hospital, Pilsen, PILSEN; 2School Of
Medicine, Pilsen, PILSEN
Background: Insufficient future liver remnant volume

(FLRV) is the cause of the low resectability of colorectal
liver metastases (CLMs).Portal vein embolisation (PVE)
with the application of autologous haematopoietic stem cells
(HSCs) is option for increasing resectability of CLMs. The
aim of our study was to compare the effect of PVE only and
PVE with application of HSCs in patients with primarily
non-resectable CLMs due to insufficient FLRV.
Methods: In prospective, randomised study PVE with the
application of HSCs was used in 14 patients (GI). The control
group (GII) consisted of 14 patients in whom only PVE was
performed. We evaluated FLRV growth, CLM volume
growth, median survival and progression-free survival (PFS).
Results: In all GI patients sufficient FLRV growth occurred
within three weeks. In the first two weeks, FLRV increased in
most of the patients (p < 0.006). In 13 (92.9%) of the GII
patients, optimum FLVR growth was observed within three
weeks following PVE (p < 0.002). More rapid FLVR growth
was observed in the GI patients (p < 0.01). CLM volume was
significantly increased in the GII (p < 0.0005) and also GI
(p < 0.006) at the time of liver resection. There was no significant difference in the growth of the CLM volume between
the groups (p < 0.17). The median survival of the GI and GII
patients was 7.4 and 6.8 months and the two-year PFS was 27
and 22% (n.s.), respectively.
Conclusion: We can conclude that PVE with HSC application is a promising method for effectively stimulating FLVR
growth in primarily non-resectable CLMs.
LO-J.08 ENHANCED ULTRASOUND

WITH NAVIGATION LEADS TO
IMPROVED LIVER LESION
IDENTIFICATION AND
NEEDLE PLACEMENT
R. C. Martin
Background: The aim of this study was to evaluate whether
3-dimensional (3D) enhanced ultrasound could increase the
accuracy and efficiacy for liver tumor identification and
needle placement.
Methods: A prospective study of 30 surgeons of various
training level where evaluated lesion identification success
and accuracy of needle placement. All surgeons were evaluated for time(seconds) to identify the liver lesions and placement of needles after review of 3 phase CT scan of the liver
and using standard B-mode ultrasound and then enhanced 3D
ultrasound (E-3DUS).
Results: Participants included 10 HPB surgeons, 5 Surgical
Fellows, 10 PGY-4&5 Surgical Residents and 5 PGY-3 resi 2015 The Authors
dents. Liver lesions were correctly identified in 73% of the

cases using B-mode ultrasound alone and 100% in E-3DUS.
The Mean time to identification is B-mode was 51.9 seconds
(SD +/ 37.1), which was significantly longer than with
E-3DUS (time 17.9 seconds, SD +/ 10.7)(p = 0.002). There
was significant improvement in time to lesion identification
using E-3DUS across all training levels (Figure,
*-P < 0.002). There was also a significant reduction in time
for accurate needle placement across all training levels (mean
reduction of 60%, within enhanced accuracy (p = 0.001)
Conclusions: E-3DUS significantly enhances lesion identification regardless of size and enhances needle accuracy for
all surgeons. This adjunctive system should be considered for
both training as well as for all complex liver tumor ablations.
MO-A.02 BRANCHED DNA ENHANCED

ALBUMIN RNA IN SITU
HYBRIDIZATION PROVIDES A
DEFINITIVE DIAGNOSIS FOR
INTRAHEPATIC
CHOLANGIOCARCINOMA
C. R. Ferrone, D. T. Ting, M. Shahid, I. T. Konstantinidis,
L. Goyal, N. Bardeesey, D. Borger, K. D. Lillemoe,
A. X. Zhu, M. N. Rivera, V. Deshpande
Massachusetts General Hospital, Boston, MA USA
Purpose: Intrahepatic cholangiocarcinoma (ICC) is often a
diagnosis of exclusion after patients have undergone numerous studies and procedures. Currently there is no
histopathologic or immunohistochemical marker for ICC.
Albumin expression is restricted to the liver. Since liver
parenchymal cells are derived from a common embryonic
progenitor, the aim of our study was to determine if albumin
is a potential biomarker for ICC utilizing a novel and highly
sensitive RNA in situ hybridization (ISH) platform.
Methods: Modified branched DNA probes were developed
for albumin RNA ISH. A total of 467 patient samples of
primary and metastatic lesions were evaluated.
Results: Of the 467 samples, 43 were ICCs, 42 were hepatocellular carcinomas and 332 were non-hepatic carcinomas
including tumors arising from the perihilar region, bile duct,
pancreas, stomach, esophagus, colon, lung, breast, ovary,
endometrium, kidney, and urinary bladder. Albumin RNA
ISH was highly sensitive for cancers of liver origin, staining
positive in 42 of 43 ICCs (99%) and 42 HCCs (100%).
Perihilar and distal bile duct carcinomas as well as carcinomas arising at other sites were negative for albumin. Notably,
6 of 27 (22%) intrahepatic tumors previously diagnosed
as carcinomas of undetermined origin were positive for
albumin.
Conclusions: Albumin RNA ISH is a sensitive and highly
specific diagnostic tool to distinguish ICC from metastatic
adenocarcinoma to the liver or carcinoma of unknown origin.
The specificity of albumin RNA ish could replace the extensive diagnostic work up patients undergo to confirm the
diagnosis.
HPB 2015, 17 (Suppl. 1), 181
Abstracts
MO-A.03 HIGH HLA CLASS I ANTIGEN

EXPRESSION IN COMBINATION WITH
LOW PD-L1 EXPRESSION AS A
FAVOURABLE PROGNOSTIC
BIOMARKER IN INTRAHEPATIC
CHOLANGIOCARCINOMA
1
F. Sabbatino , V. Villani , J. H. Yearley , L. Cai ,

V. Deshpande3, I. T. Konstantinidis1, S. P. Nota4, Y. Wang1,
A. X. Zhu5, L. Goyal5, D. T. Ting5, N. M. El-Bardeesy5,
T. S. Hong6, K. K. Tanabe7, C. Moon2, S. Ferrone1,4,
K. D. Lillemoe1, C. R. Ferrone1
1
Department Of Surgery, Massachusetts General Hospital,
Boston, MA; 2Merck Research Laboratories, Palo Alto,
CA; 3Department Of Pathology, Massachusetts General
Hospital, Boston, MA; 4Department Of Orthopedic
Surgery, Massachusetts General Hospital, Boston, MA;
5
Cancer Center, Massachusetts General Hospital, Boston,
MA; 6Department Of Radiation Oncology, Massachusetts
General Hospital, Boston, MA; 7Department Of Surgical
Oncology, Massachusetts General Hospital, Boston, MA
Introduction: For immune checkpoint molecule-specific
monoclonal-antibodies (mAbs) to be effective patients must
be able to mount an immune response to their tumor. This
requires a functional HLA class I antigen-processing machinery in the targeted tumor cells. We evaluated i) lymphocyte
infiltrate as a measure of a patients immune response to their
intrahepatic cholangiocarcinoma (ICC) ii) HLA class I
antigen expression iii) expression of the checkpoint molecules programmed cell death 1 (PD-1) and its primary
ligand programmed cell death ligand 1 (PD-L1).
Methods: Clinicopathologic data for 30 patients undergoing resection for ICC was collected. Tumors were
immunohistochemically stained with CD8-, CD4-, HLA-,
PD-1- and PD-L1-specific mAbs.
Results: Median age was 64 years, 47% had stage II disease
and median follow-up was 28.1 months. All tumors had
lymphocyte infiltrates. CD8s in the fibrous septa (FS)
between tumor lobules (TL) was higher than within TL
(mean 104 vs 12, P < 0.0001), but not for CD4s. CD8s and
CD4s in the FS and TL correlated with HLA expression
(P < 0.05). CD8s were decreased, while CD4s were
increased if HLA was down-regulated. PD-1 expression was
present in 83% of cases and highest on T cells in the FS.
CD8s/CD4s, HLA, and PD-1/PD-L1 expression in the tumor
microenvironment did not correlate with survival. In contrast
high HLA expression and low PD-L1 expression was associated with a longer overall survival (P = 0.04).
Conclusions: Increased PD-L1 expression and HLA class I
antigen defects provide ICC with an escape mechanism from
immune recognition. This provides rationale for implementing antibodies to checkpoint molecules for ICC.
79
MO-A.04 PIVOTAL ROLE OF INTRAPLATELET VEGF-A DURING

ANGIOGENIC PHASE OF LIVER
REGENERATION AFTER PARTIAL
HEPATECTOMY IN HUMANS
B. Aryal1, T. Shimizu2, J. Kadono1, A. Furoi3,
M. Yamakuchi2, M. Inoue1, T. Komokata4, T. Hashiguchi2,
Y. Imoto1
1
Cardiovascular And Gastroenterological
Surgery,Kagoshima University, Kagoshima, KYUSHU
Japan; 2Laboratory And Vascular Medicine, Kagoshima
University, Kagoshima, KYUSHU Japan; 3Kirishima
Medical Center, Kirishima, KYUSHU Japan; 4Kagoshima
Medical Center, Kagoshima, KYUSHU Japan
The ability of remnant liver to regenerate plays a decisive
role in morbidity and mortality after liver resection. Molecular mechanism of liver regeneration (LR) has still remained
as a major concern in modulating the strategies in order to
improve the outcome after resection. Two phases of LR have
been proposed lately; inductive phase marked by rapid proliferation of hepatocytes followed by a delayed angiogenic
phase with proliferation of non-parenchymal cells. Several
studies have already reported the beneficial role of platelet in
LR. We introduce platelet as a central player in the
angiogenic phase of LR after partial hepatectomy in human.
Materials and Methods: Twenty patients with diagnosed
hepatocellular carcinoma; eligible for hepatectomy were
enrolled in the study. Serum, plasma and intra-platelet
VEGF-A (vascular endothelial growth factor-A) along with
the major mitogens were monitored both before and after one
month of operation.
Result: The major serum VEGF-A pool is attributed to the
platelet, with plasma containing scarcer amount or no
VEGF-A. The change in serum VEGF-A following liver
resection strongly correlated with the altered platelet count.
Serum VEGF-A was found to be significantly elevated after
a month of partial hepatectomy; intra-platelet VEGF (IPVEGF) showed the similar trend whereas no significant difference was observed in the level of major mitogens
including hepatocyte growth factor. The soluble VEGF
receptors as important anti-angiogenic markers were down
regulated during the on-going angiogenic phase of LR.
Conclusion: IP-VEGF in angiogenic phase might clue to
anovel therapeutic target to accelerate LR after hepatectomy.
MO-A.05 PROGNOSTIC FACTORS OF

RECURRED HEPATOCELLULAR
CARCINOMA AFTER PRIMARY
LIVER RESECTION
J. Jeong, K. Suh, K. Lee, N. Yi
Seoul National University College Of Medicine, Seoul,
SEOUL
Background: Resection is the treatment of choice for hepatocellular carcinoma (HCC). However, recurrence is still a
major problem after primary resection and prognostic factors
after recurrence is not well known. So, we investigated prognostic factors of recurred HCC after primary resection to
determine who had an aggressive tumor biology.
HPB 2015, 17 (Suppl. 1), 181
2015 The Authors

80
Abstracts
Methods: We reviewed the patients who underwent surgical

resection for HCC between 2005 and 2011. Of a total 941
patients who underwent resection, 589 (62.5%) patients had
recurrences. Among them, 141 patients with extrahepatic
recurrences or positive margins were excluded. Finally, we
analyzed 448 patients who had intrahepatic recurrences after
curative resections.
Results: The median duration between primary resection
and recurrence was 9.0 (0.290.0) months. At recurrence,
353 (78.8%) patients had HCCs within Milans criteria.
After recurrence, 5-year disease-free survival rate was
12.8% and 5-year survival rate was 66.1%. >5 cm of
primary tumor size, <1 year of duration to recurrence, >200
of AFP level at recurrence, and beyond Milans criteria at
recurrence were revealed as independent poor prognostic
factors associated with disease-free survival. Whereas, only
Milans criteria and CTP class at recurrence were identified
as independent prognostic factors associated with overall
survival.
Conclusion: For recurred HCC after primary resection,
factors at recurrence including primary tumor size are significantly associated with disease-free survival. Milans criteria and CTP class at recurrence are associated with overall
survival. Therefore, it should be considered when deciding
the second-line treatment option for recurred HCC after
primary resection.
MO-B.01 GOING GREEN BEYOND THE

BORDERS: EXPANDING THE IRE KILL
ZONE WHILE CONSERVING
ELECTRICITY
S. C. Agle, Y. Li, C. A. Doughtie, C. R. Scoggins,
R. C. Martin
University Of Louisville, Louisville, KY USA
Introduction: Irreversible electroporation (IRE) is an ablation technique in which short, high-voltage pulses are applied
to tissue to irreversibly permeabilize the cell membrane resulting in apoptosis. The objective of this study was to determine
if a secondary zone of reversible electroporation can be used to
enhance tumor cell death.
Methods: Five nude mice with orthotopic human pancreatic
adenocarcinoma (PANC1) were in each group: gemcitabine
(15 mg/kg) alone, IRE alone, gemcitabine plus IRE and a
control group. IRE was set up in a 2-probe array around the
tumor to produce 2,000 volts/cm at a rate of 90 pulses per
minute for a total of 90 pulses. Pancreas tissue slides were
measured for hypoxia via the hypoxia-inducible factor (HIF1a) and apoptosis via TUNEL staining. The IRE tissue was
obtained from the lateral probe margin adjacent to the ablation zone and the gemcitabine only tissue was taken from the
tumor edge.
Results: The rate of hypoxia (Th area %) in the tissue
treated with IRE alone or combined with gemcitabine were
similar (45.2 8.6 and 68.0 15.3) and over 3 times lower
(p < 0.1) than the groups receiving only gemcitabine or no
treatment (218.8 29.2 and 351.8 28.4). IRE plus
gemcitabine group had an apoptotic index of 22.6% 7.7
which was more than 4 times higher than the IRE alone
group (4.6 2.7, p < 0.01).
2015 The Authors
Conclusion: In addition to tumor cell death within the irreversible ablation zone the use of IRE could potentially reduce
local recurrence by allowing for increased drug administration to the cells on the periphery of the ablation zone.
MO-B.02 THE POST-SIR-SPHERES

SURGERY STUDY (P4S): ANALYSIS OF
OUTCOMES FOLLOWING HEPATIC
RESECTION OR TRANSPLANTATION IN
101 PATIENTS PREVIOUSLY TREATED
WITH SELECTIVE INTERNAL
RADIATION THERAPY (SIRT)
F. Pardo1, M. Schoen2, L. Rheun-Chuan3, D. M. Manas4,
D. R. Jeyarajah5, G. Katsanos 6, G. A. Maleux7, B. Sangro8
1
Hepatobiliary, Pancreatic And Transplant Surgery,
Pamplona, NAVARRA; 2Surgery, Kalrsruhe,
BADEN-WRTTEMBERG; 3Interventional Radiology,
Taipei, BEITOU DISTRICT,; 4Hepatobiliary And
Transplant Surgery, Newcastle, TYNE AND WEAR;
5
Surgical Oncology, Dallas, TX; 6Hepatobiliary And
Transplant Surgery, Brussels, BRUXELLES; 7Radiology,
Leuven, FLEMISH BRABANT; 8Liver Unit, Pamplona,
NAVARRA
SIRT (or radioembolisation) is primarily used as palliative
treatment for inoperable primary or metastatic liver tumours,
and as bridge-to-liver transplantation in hepatocellular carcinoma (HCC). There have been reports of down-sizing to
surgical resection/transplantation but there are no robust
studies of safety outcomes. P4S is an international,
multicentre, retrospective study to assess outcomes associated with liver resection or transplantation following SIRT
using yttrium-90 resin microspheres (SIR-Spheres; Sirtex).
Primary endpoints were peri-operative and 90-day postoperative morbidity and mortality. Analysis used standard
statistical methods. Data were captured on SIRT, surgery
(between 08/199805/2014) and follow-up on 101 patients
with either primary liver cancer (HCC: 48.5%; cholangiocarcinoma: 6.9%) or secondary hepatic metastases from
HPB 2015, 17 (Suppl. 1), 181
Abstracts
colorectal (30.7%), neuroendocrine (6.9%) and other cancers
(6.9%). Resection was performed in 72 patients (71.3%;
40.6% major, 30.7% minor) and liver transplantation in 29
(28.7%). Mean interval between first SIRT and first hepatic
surgery was 8.7 months. Clavien-Dindo grade 3+ peri-/postoperative complications were: liver failure: 7 (6.9%); woundspecific: 4 (4.0%); cardiovascular: 0 (0%); pulmonary: 8
(7.9%); renal-specific: 2 (2.0%); other: 15 (14.9%). Cumulative 90-day all-cause mortality from first hepatic surgery
was 4 (4.0%). These 4 cases were all trisectionectomies
(colorectal: 3; cholangiocarcinoma: 1) and typically had 1
prior chemotherapy line, pre-surgical co-morbidities and
suffered post-hepatectomy multi-organ failure including
liver failure. Future liver remnant was targeted with SIRT in
1 of the 4 cases. The safety profile of post-SIRT resection and
transplantation appears similar to that previously reported for
hepatic surgery. No deaths appear to be directly related to
SIRT.
MO-B.03 #PANCSM INITIAL

EXPERIENCE WITH CREATION OF A
TWITTER-BASED MONTHLY ONLINE
PANCREATIC CANCER CHAT
COMMUNITY
N. J. Gusani1, L. Coker3, M. H. Katz5, D. Reidy-Lagunes4,
P. Bloomston2
1
Program For Liver, Pancreas, & Foregut Tumors; Penn
State Hershey Cancer Institute, Hershey, PENNSYLVANIA
USA; 2Surgical Oncology; Ohio State University,
Columbus, OH USA; 3Marketing & Communications; The
Ohio State University Comprehensive Cancer Center,
Columbus, OH USA; 4Division Of Gastrointestinal
Oncology; Memorial Sloan-Kettering Cancer Center, New
York, NY USA; 5Department Of Surgical Oncology; The
University Of Texas MD Anderson Cancer, Houston, TX
USA
81
services like Twitter often used for real-time communication

in pre-arranged chats around a topic or disease. The
authors created a monthly Twitter-based 1-hour chat
(#PancSM) to help increase collaboration and interaction
among patients, family/caregivers, advocates, health care
providers, researchers, and charitable organizations. We
report our initial experience.
Methods: #PancSM chats are held on the first Thursday of
each month, 910pm ET. A moderator introduces predetermined topics and discussion points, facilitating interaction with targeted questions. The chat progresses with freeform discussion among participants and is united and made
searchable by use of the hashtag #PancSM in all tweets.
Metrics for each chat session and participants were derived
using online tools (Simply Measured, Followerwonk).
Results: The first three monthly #PancSM chats were lively,
with 125262 tweets per session and 921 tweets per minute.
The chats had a maximal reach of 32,608, with a maximal
total exposure of 288,027 impressions. Tweeters participating in #PancSM chats initially included mostly physicians,
but by the third chat, advocacy groups and patient caregivers
were taking an important role.
Conclusions: A Twitter-based monthly Pancreatic Cancer
chat has the potential to reach large numbers of pancreatic
cancer community members and to discuss issues important
to many stakeholders. This forum is an innovative education
medium to help improve understanding and patient outcomes
in pancreatic cancer.
Background: Pancreatic Cancer (PancCa) is a highly lethal

disease about which many myths and much nihilism exists.
Social Media (SM) is a novel medium for communication
and interaction across time and space, with microblogging
HPB 2015, 17 (Suppl. 1), 181
2015 The Authors

Author index
Aal, A. K.: 7
Abbott, D. E.: 20, 41
Abbott, D.: 10
Abdelhady, A.: 69
Abrams, P.: 16
Adams, D. B.: 63
Adams, R. B.: 19
Agle, S. C.: 80
Agopian, V. G.: 51
Ahmad, S. A.: 41
Ahmad, S.: 20
Ahuja, N.: 4, 41, 54
Aksoy, E.: 31
Akyuz, M.: 31
Alanazi, R.: 16
Albers, C.: 67
Aldakkak, M.: 5
Ali, N.: 53, 64
Aljiffry, M.: 16
Alkhalili, E.: 21
Allard, R.: 24
Allen, P. J.: 38, 74, 76
Allen, P.: 75
Allison, J.: 67
Almeida, M. A.: 27
Aloia, T. A.: 20, 39, 72, 73, 78
Alseidi, A.: 48, 60, 62, 64
Al-Shazly, M. A.: 69
Alshenaifi, J.: 16
Alsina, A. E.: 67
Alvarez, F.: 9
Amini, N.: 26
Anantha Sathyanarayana, S.: 34, 35,
65
Anaya, D. A.: 6
Anderson, C. D.: 44, 70
Anderton, A. C.: 29
Andreou, A.: 53, 68
Annamalai, A.: 11
Annigeri, P.: 35
Aoki, T.: 35
Armstrong, P. A.: 58
Arnoletti, P.: 21
Aryal, B.: 79
Attwood, K.: 57
Aucejo, F.: 31
Aviles, C.: 37
Aycart, S.: 18
Ayloo, S.: 68
Baker, E. H.: 8, 23, 59
Baker, E.: 37
Balaa, F. K.: 51
Balart, L. A.: 70
Ball, C. G.: 47
Barber, K.: 67
Bardeesey, N.: 78
Barin, B.: 52
Bassi, C.: 1
Bauer, T. W.: 19
Baumert, T.: 3
Bdeawey, E.: 25
Beal, E.: 69
Beane, J. D.: 15
Begnami, M. F.: 18
Behman, R.: 54
Behrman, S. W.: 61
Bendix, S.: 34, 65

Bennett, S.: 51
Berber, E.: 31
Berger, D. L.: 30
Berger, Y.: 18
Bertens, K. A.: 24
Bhullar, J. S.: 47
Biehl, T.: 60, 62, 64
Bindroo, S.: 47
Binkley, C. E.: 37
Bjrnbeth, B. A.: 33
Bjrnson, B.: 33
Black, S.: 69
Blazer, D. G.: 64
Bliss, L. A.: 15, 49
Bloomston, M.: 36, 69
Bloomston, P.: 81
Bolinger, B. L.: 37
Bonachi, R. S.: 18
Boone, B.: 57
Booth, C. M.: 17, 29, 32
Borg, B.: 70
Borger, D.: 78
Boucher, L.: 16
Bressan, A. K.: 47
Brown, M.: 36
Brunt, L. M.: 39
Buell, J. F.: 70
Bullock, A.: 55
Burns, W. R.: 73
Busuttil, R. W.: 51
Cai, L.: 79
Callery, M. P.: 15, 49, 56, 59
Cameron, J. L.: 4, 54
Cameron, J.: 41
Campbell, P. M.: 36
Cardona, K.: 22, 26, 75
Caruso, R.: 71
Cassera, M. A.: 36, 40
Cassera, M.: 31
Castellanos, A.: 2
Castillo, A. E.: 70
Castillo-Angeles, M.: 55, 56, 59
Castleberry, A. W.: 21
Ceppa, E. P.: 12, 62, 66
Chabot, J. A.: 5
Chalikonda, S.: 56, 64
Chan, S. T.: 34
Chang, C. K.: 37
Chang, G. J.: 72
Chapman, W. C.: 1
Chaudhary, S.: 47
Cheaito, A.: 51
Cheek, S.: 20, 37
Cheng, E.: 76
Cho, C. S.: 22, 75
Choi, J. S.: 37
Choi, J.: 34
Chojniak, R.: 18
Christians, K. K.: 3, 5, 7, 10, 63, 65
Christians, K.: 2
Chun, Y.: 25
Chung, J.: 24
Chung, M. H.: 50
Cillo, U.: 32
Cioffi, J. L.: 12, 62, 66
2015 The Authors

HPB 2015 Americas Hepato-Pancreato-Biliary Association, HPB, 17 (Suppl. s1), 8286
Clanton, J.: 62, 64

Clark, C.: 51
Clary, B. M.: 64
Claudio, R.: 67
Cleary, S.: 30
Cloutier, A.: 24
Coburn, N. G.: 42, 58, 65, 76
Coburn, N.: 54
Cochran, A.: 23, 37
Coimbra, F. J.: 18, 27, 29
Coker, L.: 81
Collin, Y.: 19, 24, 31
Colquhoun, S.: 11
Conneely, J. B.: 72
Conrad, C. H.: 39, 72
Conrad, C.: 20, 73, 78
Coordes, A.: 68
Corvera, C. U.: 72
Cosgrove, D.: 32
Costa, A.: 30
Costantino, C.: 29
Cowzer, D.: 32
Cozakov, Y.: 47
Crawford, J. H.: 9
Crocenzi, T. S.: 17
Croome, K. P.: 13, 45, 55
DAngelica, M. I.: 28, 38, 74, 76
DAngelica, M.: 75
DHondt, M.: 33
Da Costa, W. L.: 18, 27, 29
Dagenais, M.: 19, 31
Dahaba, A.: 69
Daily, M. F.: 2, 52
Davenport, D.: 2, 52
Day, R. W.: 20, 73, 78
De Farias, I. C.: 18, 27, 29
De Geus, S. W.: 49
De Godoy, A. L.: 18, 27, 29
De La Fuente, S.: 21
De Luca, M.: 71
De Matteo, R. P.: 76
De Mello, C. A.: 27
De Rosa, N.: 73
De Souza, R. S.: 18, 27, 29
Decker, M.: 47
DeGeus, S.: 15
Delman, K. A.: 48
DeMatteo, R. P.: 38, 74
DeMatteo, R.: 75
Demirjian, A.: 55
Denault, A.: 24
Deshpande, V.: 27, 78, 79
Devriendt, D.: 33
DeWitt, J. M.: 12
Diana, M.: 3
Diaz, E.: 71
Difronzo, A.: 63
Dillhoff, M.: 36
Dinh, K.: 6
Diniz, A. L.: 18, 27, 29
Dip, M.: 9
Dissanaike, S.: 8
Dixon, E.: 47
Dixon, M. E.: 58
Dixon, M.: 65
Donchev, V.: 11
Author index
Dong, M.: 53
Dorsett-Martin, W.: 44
Doughtie, C. A.: 80
Doussot, A.: 28, 76
Downs-Canner, S.: 57
Doyle, M. B.: 1
Drosdeck, J.: 36
Drummond, J.: 23, 37
Du, L.: 43
Dua, M. M.: 14, 42, 44, 77
DuBay, D. A.: 7, 9
Dural, C.: 31
Duran, H.: 71
Dutson, E. P.: 51
Earl, T. M.: 44, 70
Earle, C.: 65
Easler, J. J.: 12
Eaton, A. A.: 38
Eckhoff, D. E.: 7, 9
Edwards, G. C.: 43
Edwards, J. P.: 47
Ehrenfeld, J.: 43
Ehrenwald, E.: 25
Ejaz, A.: 32, 41
El-Bardeesy, N. M.: 79
Eldert, R. E.: 66
Elgendi, A. M.: 25
ElHayek, K.: 64
Ellison, C. E.: 36
El-mansy, M.: 69
Elmi, M.: 58, 65
El-Sedfy, A.: 58, 65
Elshafey, M.: 25
Enestvedt, K.: 70
Eng, C.: 72
Epelboym, I.: 5
Erickson, B. A.: 3, 5, 65
Erinjeri, J. P.: 28
Ertel, A.: 10, 20, 41
Eskander, M. F.: 49
Eskander, M.: 15
Etra, J. W.: 26
Eubanks, S.: 21
Evans, D. B.: 3, 5, 63, 65
Ewan, L. C.: 50
Fabra, I.: 71
Fairfull Smith, R.: 43
Falk, G. A.: 56
Farnell, M. B.: 13, 55
Farsad, K.: 70
Fathi, A.: 10
Fauda, M.: 9
Fernandez Del Castillo, C.: 55
Ferrone, C. R.: 27, 30, 55, 78, 79
Ferrone, S.: 79
Fields, R. C.: 11, 14, 39
Fino, N.: 51
Fleming, J. B.: 39
Flores, K.: 6
Fong, Y.: 38
Fonseca, V. H.: 18
Fowler, K. J.: 1
Franco, E. S.: 67
Franssen, B.: 73
Freedman, S. D.: 15
Freitas, H. C.: 29
Friedman, M.: 11
Friel, C. M.: 19
Fuchshuber, P.: 37
Fulp, W. J.: 58
Fung, A.: 72
Fung, J.: 31
Furoi, A.: 79
Fuss, M.: 70
Gabriel, E. M.: 57
Gabriel, E.: 60
Gall, T. M.: 28
Gallinger, S.: 43, 48
Gamblin, T. C.: 2, 7, 10
Gamblin, T.: 46
Garcia, S.: 49
Garcia-Monaco, R. D.: 17
Gasslander, T.: 33
Gedaly, R.: 2, 52
Gennaro, K. H.: 7
George, B.: 3, 5, 65
Gerst, S. R.: 76
Girgis, M.: 74
Glorioso, J. M.: 45
Goel, M.: 61
Gondolesi, G.: 9
Gonen, M.: 75
Goyal, L.: 27, 78, 79
Gray, S. H.: 7, 9
Greenbaum, A.: 21
Greig, P. D.: 72
Groeschl, R. T.: 7
Grondin, S. C.: 47
Groot Koerkamp, B.: 76
Gu, C.: 43
Gulenchyn, K. Y.: 43
Gupta, R.: 25
Gusani, N. J.: 81
Habashi, R.: 76
Habersetzer, F.: 3
Halac, E.: 9
Hall, B. L.: 15, 39, 61, 62
Hallet, J.: 3, 42, 54, 76
Halpern, E.: 55
Hammill, C. W.: 36, 40
Hammill, C.: 31
Hanna, E. M.: 53
Hanna, S.: 54, 76
Hanseman, D. J.: 41
Hansen, P. D.: 36, 40
Hansen, P.: 31
Harrigan, A. M.: 19
Harris, J. W.: 12
Hasegawa, K.: 35
Hashiguchi, T.: 79
Hassanain, M.: 16
Hauch, A. T.: 70
Hawkins, W. G.: 11, 14, 39
Haywood, N. S.: 7
He, J.: 4, 26, 54
Hedrick, T. L.: 19
Hefty, M. T.: 50
Helmy, A. S.: 69
Helton, S.: 60, 62, 64
Hemingway, K. T.: 55
Herbert, G.: 75
Hernandez-Alejandro, R.: 24
Hill, M.: 25
Hines, O. J.: 51
Hiotis, S.: 18
Hirose, K.: 26, 41, 54
83
Hiyama, D. T.: 51
Hochwald, S. N.: 38
Hochwald, S.: 57, 60
Hodul, P.: 58
Hoehn, R. S.: 20, 41
Hoehn, R.: 10
Hoen, H. M.: 40
Hoen, H.: 31
Hoffe, S.: 58
Hogg, M.: 57, 74
Hong, J. C.: 48
Hong, T. S.: 79
Hooper, J. E.: 70
Horeya, H. E.: 54
Hosny, A.: 69
Hotoyan, L.: 72
House, M. G.: 12, 15, 62, 66
Hruban, R. H.: 4
Hu, T.: 24
Huang, R.: 19
Hunter, S.: 20, 37
Husien, M.: 43
Hwa, K. J.: 14, 44
Iannitti, D. A.: 8, 53, 59
Iannitti, D.: 23, 37
Idrees, K.: 43
Ielpo, B.: 71
Illig, K.: 58
Imoto, Y.: 79
Imventarza, O.: 9
Inampudi, S.: 25
Inoue, M.: 79
Isaksson, B.: 33
Jacobs, M. J.: 34, 35, 65
Jacobs, M.: 47
Jalink, D. W.: 43
Jaques, D. P.: 39
Jarnagin, W. R.: 28, 38, 74, 76
Jarnagin, W.: 75
Jenkins, H.: 50
Jeong, J.: 79
Jernigan, P. L.: 41
Jeyarajah, D. R.: 20, 37, 47, 80
Jhaveri, K.: 30
Jiao, L. R.: 28
Johnson, B. L.: 58
Johnson, J.: 50
Johnston, C. W.: 40
Johnston, F. M.: 2, 3, 5, 7, 10, 63
Johnston, W. C.: 36
Johnstone, M.: 67
Joyce, D.: 56
Julian, J. A.: 43
Jutric, Z.: 40
Kadono, J.: 79
Kagedan, D. J.: 58, 65
Kandil, E.: 70
Kaneko, J.: 35
Karanicolas, P. J.: 42, 76
Karanicolas, P.: 54
Kather, R.: 34
Katsanos, G.: 80
Katz, M. H.: 39, 81
Kaugh, J.: 17
Kazantsev, G. B.: 37
Keith, S. W.: 62
Kelly, L. R.: 17
2015 The Authors

84
Kelly, P.: 72
Kelly, R. P.: 10
Kemmer, N.: 67
Kendrick, M. L.: 13, 55
Kendrick, M.: 73
Kent, T. S.: 15, 55, 56, 59
Khreiss, M.: 45
Khwaja, K.: 49
Kilbane, E. M.: 15, 61, 62
Kilbane, M.: 66
Killackey, M.: 70
Kim, I.: 11
Kim, K.: 17
Kim, Y.: 25
Kingham, P.: 28
Kingham, T. P.: 38, 74, 75, 76
Kirichenko, A.: 16
Kirks, R. C.: 53
Klausner, J. Q.: 14
Klein, A.: 11
Klein, R.: 4
Klompmaker, S.: 59
Kluger, M. D.: 5
Kohli, N. P.: 34
Kojouri, K.: 37
Kokudo, N.: 35
Kolbeck, K.: 70
Komokata, T.: 79
Konomos, M.: 48
Konstantinidis, I. T.: 27, 78, 79
Kooby, D. A.: 22, 26, 75
Krepline, A. N.: 5, 65
Krohmer, S.: 2
Krzywda, B. A.: 3
Krzywda, E. A.: 63
Kumar, R.: 24
Kunnimalaiyaan, M.: 46
Kunnimalaiyaan, S.: 46
Kuo, W.: 72
Kutlu, O. C.: 8, 49
Kuvshinoff, B. W.: 38
Kuvshinoff, B.: 57, 60
Labow, D.: 18
LaFemina, J.: 6
Lahiff, S. L.: 63
Lahiff, S. M.: 3
Laing, C. J.: 17
Lam, T.: 48
Lancaster, W. P.: 63
Lapointe, R.: 19, 24, 31
Lavu, H.: 62
Law, C. H.: 42, 43, 76
Law, C.: 54
Leal, J. N.: 74
Lee, J. E.: 39
Lee, K. K.: 57
Lee, K.: 79
Lee, S. J.: 62
Leiva Espinoza, J.: 9
Lemke, M.: 54
Lenarz, M.: 68
Lennon, A.: 4
Letourneau, R.: 19, 31
Levine, E.: 51
Levine, M. N.: 43
Lewis, C. E.: 51
Li, Q.: 58, 65
Li, Y.: 80
Lidsky, M. E.: 21
Author index
Lillemoe, K. D.: 27, 30, 55, 78, 79
Lin, B.: 60
Linehan, D. C.: 1
Litchman, T.: 28
Liu, J. Y.: 26
Liu, N.: 65
Luberice, K.: 4
Lundgren, L.: 33
Luo, T.: 25
Luque, C.: 9
Machado, M. A.: 71
Mackillop, W. J.: 17, 29, 32
MacLennan, P. A.: 9
Mahar, A. L.: 42
Mahendraraj, K.: 5
Maithel, S. K.: 22, 26, 48, 75
Makary, M. A.: 26, 41, 54
Makdissi, F. F.: 71
Makris, A.: 67
Malafa, M. P.: 58
Maleux, G. A.: 80
Malleo, G.: 1
Manas, D. M.: 80
Mancias, J.: 55
Mandeli, J.: 18
Mao, S. A.: 45
Maqsood, H.: 25
Marescaux, J.: 3
Marginean, C.: 51
Marques, M. C.: 18, 27, 29
Marshall, G.: 35
Martel, G.: 51
Martin, J. T.: 12
Martin, R. C.: 17, 22, 75, 78, 80
Martinie, J. B.: 8, 53, 59
Martinie, J.: 23, 37
Mason, M. C.: 6
Massarweh, N. N.: 6
Matlock, J. A.: 48
Mattera, J.: 9
Matus, D.: 9
Maupoey, J.: 71
Maurette, R.: 9
Maynard, E. C.: 12
Maynard, E.: 2, 52
McCormack, L.: 9
McDowell, D. M.: 3
McGilvray, I. D.: 72
McGrath, P. C.: 12
McIntyre, C. A.: 38
McMillan, M. T.: 1
McMurry, T. L.: 19
Medkhaly, A.: 16
Mehran, R. J.: 73
Meier, A. M.: 20
Merchant, N. B.: 43
Mercurio, N.: 39
Metrakos, P.: 16
Midura, E.: 41
Miller, C.: 31
Miller, J. R.: 43
Mimeault, R.: 51
Minter, R. M.: 22, 47, 48
Mir, H. R.: 43
Mise, Y.: 35
Misustin, S. M.: 63
Mitin, T.: 70
Mitra, A.: 61
Mittal, V. K.: 47
2015 The Authors

Mittmann, N.: 58, 65

Miura, J. T.: 7, 10
Miura, J.: 2
Miyata, A.: 35
Mobley, C.: 51
Moehlen, M.: 70
Molinari, M.: 10, 68
Molla, N.: 16
Moon, C.: 79
Moon, T. L.: 73
Morgan, K. A.: 63
Morris-Stiff, G.: 53, 56, 64
Mortenson, M. M.: 37
Moser, A. J.: 15, 55, 56, 59
Moulton, C.: 43, 48
Muscarella, P.: 36
Mutter, D.: 3
Nagendran, M.: 28
Nagorney, D. M.: 13, 45, 55
Nakeeb, A.: 12, 62, 66
Nanji, S.: 17, 29, 32
Nardin, C.: 28
Naugler, W. E.: 29
Naugler, W.: 70
Negussie, E.: 35
Newell, P. H.: 36, 40
Newell, P.: 31
Ng, S.: 49
Nguyen, P.: 58
Nguyen, T. K.: 56
Nievas, F.: 9
Nir, I.: 21
Nissen, N.: 11, 52
Norton, J. A.: 14, 42, 77
Nota, S. P.: 79
Nurkin, S. J.: 38
Nurkin, S.: 57, 60
Nussbaum, D. P.: 64
Nyberg, S. L.: 45
Nydam, T. L.: 66
ORourke, C.: 53
Oliva, J.: 16
Olivares, S.: 71
Orloff, S. L.: 29, 67
Orloff, S.: 70
Osayi, S. N.: 36
Oshima, K.: 65
Osman, H.: 20, 37
Ostapoff, K. T.: 38
Owen, J. W.: 1
Pai, E.: 61
Pandharipande, P. V.: 55
Pappas, T. N.: 21
Paquette, I. M.: 41
Paquin-Gobeil, M.: 51
Paramesh, A.: 70
Parda, D.: 16
Pardo, F.: 80
Parikh, A. A.: 43
Parikh, J. A.: 34, 35
Parikh, J.: 65
Park, S.: 24
Parker, R.: 25
Parmar, A.: 15, 61, 62
Pasko, J. L.: 29, 67
Patkar, S.: 61
Pawlik, T. M.: 25, 26, 32, 48, 54
Author index
Pawlik, T.: 41
Payette, F.: 24
Pena, L.: 52
Peng, P. D.: 37
Perez, A.: 21
Perez, C. A.: 37
Pessaux, P.: 3
Petersen, B. D.: 67
Philips, P.: 17
Picozzi, V.: 60
Pimiento, J. M.: 58
Pitt, H. A.: 15, 61, 62
Pitt, S. C.: 15
Plasse, M.: 19, 31
Pomianowska, E.: 33
Postlewait, L. M.: 22, 26, 48, 75
Postow, M. A.: 28
Potkonjak, M.: 2
Pottel, H.: 33
Poultsides, G. A.: 14, 42, 77
Pratschke, J.: 53, 68
Prestera, A.: 71
Prussing, K.: 75
Qu, J.: 72
Quan, D.: 43
Que, F. G.: 13, 55
Quijano, Y.: 71
Quinonez, E.: 9
Quintini, C.: 31
Radomski, M.: 74
Raju, R.: 58, 65
Ramirez, R. M.: 37
Randhawa, S.: 35
Rashid, O. M.: 58
Redden, D. T.: 7
Reddy, S.: 25, 53
Redman, R. A.: 17
Reid-Lombardo, K.: 13, 55
Reidy-Lagunes, D.: 81
Rezaee, N.: 4, 54
Rheun-Chuan, L.: 80
Riall, R. S.: 61
Riall, T. S.: 15
Ribeiro, H. S.: 18, 27, 29
Rilling, W. R.: 17
Ritch, P. S.: 3, 5, 65
Rivera, M. N.: 78
Roach, L.: 38
Roch, A. M.: 12
Rocha, F. G.: 60
Rocha, F.: 62, 64
Rogers, R.: 52
Romero, P.: 9
Rong, Z.: 31
Rosas, E. E.: 37
Rose, J. B.: 60, 62, 64
Rosemurgy, A. S.: 4
Rosok, B. I.: 33
Ross, S. B.: 4
Ross, S. W.: 8, 53
Rouleau, E.: 6
Rowsell, C.: 58
Roy, A.: 19, 31
Rubenfeld, I. S.: 34
Ruo, L.: 43
Russell, M. C.: 26
Ryan, C.: 4
Saad, N. E.: 1
Sabbatino, F.: 30, 79
Saddekni, S.: 7
Sadot, E.: 38
Sadowitz, B.: 4
Sakamoto, Y.: 35
Salami, A.: 6
Salman, B.: 4
Sandstrm, P.: 33
Sanford, D. E.: 11, 14
Sangro, B.: 80
Sapisochin, G.: 72
Sarmiento, J. M.: 26
Sarpel, U.: 18
Sasadeusz, K.: 31
Sauer, P. F.: 9
Scally, C.: 22
Schenning, R. C.: 67
Schlieman, M. A.: 37
Schmidt, C. M.: 62, 66
Schmidt, C.: 12, 36, 69
Schoen, M.: 80
Schreeder, M. T.: 17
Schrope, B.: 5
Schwarz, L.: 39, 72
Scoggins, C. R.: 17, 22, 75, 80
Seawright, A.: 70
Seehofer, D.: 53, 68
Sela, N.: 24
Serrano, P. E.: 43
Seshadri, R. M.: 59
Seshadri, R.: 8, 23, 37, 53
Shah, M.: 2, 52
Shah, S. A.: 20, 41
Shah, S.: 10
Shahid, M.: 78
Sharma, V. R.: 17
Sheckley, M.: 11
Sheikh, M. R.: 20, 37
Shen, P.: 51
Shia, J.: 76
Shim, J.: 67
Shimizu, T.: 79
Shin, E.: 58, 65
Shridhar, R.: 58
Shrikhande, S. V.: 61
Shubert, C. R.: 13
Shyr, Y.: 43
Sielaff, T.: 25
Simoneau, E.: 16
Singh, N.: 44
Smith, A.: 70
Smith, T. J.: 70
Smoot, R. L.: 13
Smoot, R.: 30, 72
Soares, F. A.: 27
Sodergren, M. H.: 28
Soler, L.: 3
Sosa, J. A.: 64
Spain, D. A.: 77
Sparrelid, E.: 33
Spitzer, A. L.: 37
Spolverato, G.: 25, 26, 32, 41
Springett, G.: 58
Sprys, M. H.: 1
Squires, M. H.: 22, 26, 75
St. Martin, L.: 48
Stafford, A. T.: 64
Staley, C. A.: 26
Steve, J.: 57
85
Stilwell, K.: 50
Stock, P.: 52
Storino, A.: 55, 56, 59
Strasberg, S. M.: 17, 39
Stukenborg, G. J.: 19
Subar, D. A.: 50
Sugawara, Y.: 35
Suh, K.: 79
Sukharmwala, P.: 4
Sultenfuss, M. A.: 6
Surjan, R. C.: 71
Surraco, P.: 9
Sussman, J. J.: 20
Swan, R. Z.: 8, 53, 59
Swan, R.: 23, 37
Szramowski, M.: 16
Tabrizian, P.: 18
Tait, G.: 72
Takaki, H.: 28
Talbot, P.: 68
Tanabe, K. K.: 27, 30, 79
Tatum, C. M.: 17
Tee, M. C.: 13
Tee, M.: 55
TEH, S. H.: 37
Templin, M. A.: 59
Templin, M.: 8
Thai, N.: 16
Thiesing, J. T.: 31
Thirunavukarasu, P.: 38, 57, 60
Thomas, C.: 70
Thoolen, S. J.: 59
Tilak, J.: 47
Tillou, A.: 51
Ting, D. T.: 27, 78, 79
Tohme, S.: 45
Tolat, P.: 65
Tom, K.: 16
Tomalty, R. D.: 17
Torabi, R.: 30
Tran, T. B.: 14, 42, 77
Tremblay St-Germain, A.: 30
Treska, V.: 78
Truty, M. J.: 13, 55
Tsai, S.: 2, 3, 5, 7, 10, 63, 65
Tsang, M. E.: 76
Tsang, M.: 42
Tseng, J. F.: 15, 49, 56, 59
Tsung, A.: 13, 45
Turaga, K. K.: 2, 7, 10
Turcios, L.: 2
Turcotte, S.: 19, 31
Tzeng, C. D.: 12
Tzeng, C.: 2, 52
Uemoto, S.: 69
Uppal, R.: 31
Usatoff, V.: 34
Valsangkar, N.: 66
Van Der Vliet, W. J.: 59
Vandenbroucke-Menu, F.: 19, 24, 31
Vansteenkiste, F.: 33
Vargas, C. R.: 56
Vargas, C.: 55
Varley, P.: 13, 45
Varshney, N.: 47
Vauthey, J.: 20, 39, 72, 73, 78
Velduis, P.: 21
2015 The Authors

86
Vicente, E.: 71
Vilchez, V.: 2, 52
Villani, V.: 30, 79
Visser, B. C.: 14, 42, 44, 77
Vitale, A.: 32
Voidonikolas, G.: 11
Vollmer, C. M.: 1
Wachsman, A.: 11
Wachtel, M.: 8
Walsh, R. M.: 53, 64
Walsh, R.: 56
Wang, Y.: 79
Warner, S. G.: 22, 48
Warshaw, A. L.: 55
Wassef, W.: 6
Wasserman, J.: 51
Waters, J. A.: 62
Watkins, A. A.: 55, 56, 59
Weaver, J.: 51
Weber, S. M.: 22, 75
Weber, S.: 36
Wei, X.: 17, 29, 32
Weiss, M. J.: 4, 26, 41, 54
Author index
Welch, S.: 24
Welsh, M. T.: 12
Wey, J.: 64
White, J. A.: 7, 9
White, R. R.: 64
White, T.: 48
Williams, M. V.: 49
Wima, K.: 10, 20, 41
Winer, J. H.: 26
Winslow, E.: 22, 75
Wolf, R. F.: 36, 40
Wolf, R.: 31
Wolfgang, C. L.: 4, 26, 41, 54
Wong, L. L.: 29
Woo, S.: 63
Wood, L. D.: 4
Worhunsky, D. J.: 42
Worth, P.: 70
Wren, S. M.: 44
Wright, G.: 50
Wu, Y.: 31
Yamakuchi, M.: 79
Yang, C. J.: 15
2015 The Authors

Yazici, P.: 31
Yearley, J. H.: 79
Yi, N.: 79
Yoshihara, E.: 33
Yoshy, C. S.: 24
Youngwirth, L. M.: 64
Yousafzai, O. K.: 49
Yu, L.: 36
Zaki, A. M.: 54
Zarrinpar, A.: 51
ZarZaur, B. L.: 61
Zaydfudim, V. M.: 19
Zaytseva, Y.: 2
Zeh, H. J.: 57
Zeh, H.: 74
Zenoni, S.: 21
Zhang, J.: 44
Zhu, A. X.: 27, 78, 79
Zhu, X.: 21
Zilbert, N.: 48
Zureikat, A.: 57, 74
Zyromski, N. J.: 12, 62, 66

15 Abstracts AHPBA

Cargado por

Información del documento

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

15 Abstracts AHPBA

Cargado por

Copyright:

Formatos disponibles

SUPPLEMENT 1

THE OFFICIAL JOURNAL OF THE

International Hepato-Pancreato-Biliary Association

Abstracts of the Fifteenth Annual Americas Hepato-Pancreato-Biliary Congress

HPB Editors and Editorial Board

Honorary Regional Editors

P.-A. Clavien, Switzerland

concerning your journal subscription please go to www.wileycustomerhelp.com/ask

Abstracting and Indexing Services. The Journal is indexed by EMBASE/

Excerpta Media, Index Medicus/MEDLINE, SIIC database, Science Citation Index

Volume 17, Supplement 1, February 2015

Abstracts of the Fifteenth Annual Americas Hepato-Pancreato-Biliary Congress

Volume 17, Supplement 1, February 2015

HPB 2015, 17 (Suppl. 1), 181

PP.02 COLORECTAL LIVER

surgical pathology or 1 year follow-up imaging, 37 of 82

those with MVI and CD44 or CD133 is 30% and 27.7%;

PP.03 CANCER STEM CELL MARKER

FRIDAY, MARCH 13, 2015,

HPB 2015, 17 (Suppl. 1), 181

Results: We present a 52 year-old gentleman with a 3 cm

FRIDAY, MARCH 13, 2015,

CA19-9 or HbA1c was detected in 2 (5%) and 4 (9%)

LO-A.02 HAS SURVIVAL IMPROVED

2015 The Authors

LO-A.03 HIGH-GRADE INTRADUCTAL

LO-A.05 PRETREATMENT SERUM CA

Conclusion: This study illustrates criteria for the highest

FRIDAY, MARCH 13, 2015,

LO-A.06 ONE YEAR EXPERIENCE

Introduction: Transarterial chemoembolization (TACE) is

HPB 2015, 17 (Suppl. 1), 181

LO-B.04 SURGICAL RESECTION VERSUS

LO-B.03 THE MODIFIED RESPONSE

J. T. Miura, R. T. Groeschl, F. M. Johnston, S. Tsai,

2015 The Authors

LO-B.05 ROLE OF SURGICAL

LO-B.06 CLINICAL OUTCOMES OF

R. Seshadri, E. H. Baker, S. W. Ross, M. Templin,

Introduction: In advanced stages, hepatocellular carcinoma

2015 The Authors

O. C. Kutlu1, M. Wachtel2, S. Dissanaike1

HPB 2015, 17 (Suppl. 1), 181

FRIDAY, MARCH 13, 2015,

HPB 2015, 17 (Suppl. 1), 181

2015 The Authors

LO-C.03 A GLIMPSE INTO THE FUTURE:

LO-C.04 SOCIETAL REINTEGRATION

primary aim of this study was to assess the rate of social

LO-C.05 DEFINING THE INFLUENCE OF

HPB 2015, 17 (Suppl. 1), 181

FRIDAY, MARCH 13, 2015,

2015 The Authors

LO-D.03 INCREASED MORBIDITY AND

LO-D.02 IMPACT OF MARGIN STATUS

Introduction: Select patients with periampullary cancers

HPB 2015, 17 (Suppl. 1), 181

LO-D.05 FACTORS INFLUENCING

HPB 2015, 17 (Suppl. 1), 181

2015 The Authors

LO-D.07 PANCREATECTOMY WITH

LO-D.06 THE IMPACT OF RECENT

Introduction: A variety of techniques have been described

adenocarcinomas (63 and 60%) or neuroendocrine tumors