Enhanced epidermal dose caused by localized electron contamination

from lead cutouts used in kilovoltage radiotherapy

J. E. Lye,a D. J. Butler, and D. V. Webb
Australian Radiation and Protection and Nuclear Safety Agency, Yallambie, Melbourne 3085, Australia

Received 25 January 2010; revised 22 April 2010; accepted for publication 10 June 2010;
published 12 July 2010
Purpose: To investigate and quantify electron contamination from the lead cutouts used in kilovoltage x-ray radiotherapy.
Methods: The lead cutouts were modeled with the Monte Carlo EGSnrc user codes DOSXYZnrc
and DOSRZnrc for x-ray beams ranging from 50 to 300 kVp. The results from the model were
confirmed with Gafchromic film measurements. The model and measurements investigated the dose
distribution with and without gladwrap shielding under the lead, and dose distributions with
round, square, and serrated edge cutouts.
Results: Large dose enhancement near the edges of the lead was observed due to electron contamination. At the epidermal/dermal border, there is double the dose at the edge of the lead compared to
the central dose due to electron contamination for a 150 kVp beam and three times the dose for a
300 kVp beam. gladwrap shielding effectively removes the contaminant dose enhancement using
ten and four layers for 300 and 150 kVp beams, respectively.
Conclusions: The contaminant dose enhancement is undesirable as it could cause unnecessary
erythema and hyperpigmentation at the border of the treated and untreated skin and lead to a poorer
cosmetic outcome. The contamination is easily removed by gladwrap shielding placed under or
around the lead cutout. 2010 American Association of Physicists in Medicine.
DOI: 10.1118/1.3458722
Key words: kilovoltage radiotherapy, electron contamination, lead cutout, skin dose
I. INTRODUCTION
The problem of electron contamination from the metal cones
used to shape the x-ray beams in kilovoltage kV radiotherapy is an acknowledged complication in kV dosimetry.1
Protocols such as TRS-398 recommend using thin films to
remove the electron contaminant dose when using parallel
plate chambers to measure the x-ray output.2 The AAPM
TG-61 kV protocol discusses the issue in some detail, noting
that the contamination may have clinical, as well as dosimetric, ramifications, and should not be ignored during
treatment.3 Electron contamination from lead eye shields is
also a known problem and commercial eye shields now have
a protective low Z layer to soak up the contaminant dose.4
An issue that has been largely neglected to date is the question of electron contamination from the custom lead cutouts
that are placed directly on the patients skin to define the
treatment area, as shown in Fig. 1. One study by Lee and
Chang5 considered this from a dosimetric point of view,
measuring the central surface dose change with and without
a thin film electron filter on a parallel plate chamber. They
observed around a 10% change in surface dose for small
diameter cutouts. A recent paper by Thomas and Clark6 investigates using a thin coating painted onto the custom lead
cutouts to reduce the electron dose, again measuring the efficiency of this method with a thin window parallel plate
chamber.
The work presented in the current paper confronts the
problem from a new direction, using Monte Carlo MC
3935

Med. Phys. 37 8, August 2010

tools to evaluate the contaminant dose over the whole treatment field, focusing particularly on the dose deposited locally near the edge of the lead cutout. A very large localized
epidermal edge dose enhancement of up to 400% was discovered using MC modeling and Gafchromic film measurements. At a depth equivalent to the epidermal/dermal border,
the edge dose enhancement remained high, up to 300%, and
could lead to unnecessary erythema and hyperpigmentation
at the border of the treatment field.
II. METHODS
II.A. Monte Carlo modeling of the lead cutouts

The EGSnrc user codes DOSRZnrc and DOSXYZnrc

were used to model the dose from round and square lead
cutouts, respectively. The MC transport parameters chosen
were those recommended for kV calculations in Ref. 7. The
DOSRZnrc and DOSXYZnrc simulations used 1 109 primary histories and 1 108 primary histories, respectively,
resulting in relative voxel dose accuracy of the order of 1%.
No variance reduction was used in the simulations. The
source was a photon point source. The beam radius at the
surface of the lead was 1 mm larger than the cutout under
investigation. The source to surface distance was 30 cm. The
source energies were the measured spectra of the Australian
Radiation Protection and Nuclear Safety Agency ARPANSA kilovoltage radiotherapy calibration beams from a
constant potential, tungsten target, Seifert x-ray unit Seifert
X-Ray Corp., Fairview Village, PA. The spectra were mea-

0094-2405/2010/378/3935/5/$30.00

2010 Am. Assoc. Phys. Med.

3935

3936

Lye, Butler, and Webb: Enhanced epidermal dose from lead cutouts

FIG. 1. Schematic of geometry used in DOSRZnrc and DOSXYZnrc for a

modeling the dose in water and b modeling the dose to film. The model is
run with and without the polyethylene layer for each case. For DOSRZnrc
0.5 mm radius, cylinders are used, and for DOSXYZnrc, 0.5 0.5 mm2
voxel widths are used.

3936

FIG. 2. DOSRZnrc results and the film measurements of irradiation with a

300 kV HVL 3.8 mm Cu beam, with and without ten layers of gladwrap
over the 25 mm diameter lead cutout.

III. RESULTS
sured with a Compton spectrometer at 1.6 m from the tube
and showed good agreement with spectra obtained from a
semiempirical model by Tucker.8 MC modeling of our kilovoltage beam qualities have been validated in a previous
work.9 Two different geometries were used to simulate dose
in a water phantom, and dose to film in air, as shown in Fig.
1. For DOSRZnrc, 0.5 mm radius cylinders were used, and
for DOSXYZnrc, 0.5 0.5 mm2 voxel widths were used.
The slabs of water shown in Fig. 1a were used for beams
from 150 to 300 kVp. Water slab thickness was reduced
from 0.05 to 0.02 mm for 100 kVp and to 0.002 mm for
50 kVp beams.

II.B. EBT film measurements

Gafchromic EBT film is well suited for superficial measurements as it does not require a light tight cardboard envelope that would absorb much of the contaminant electrons
before they could be measured. The film was placed flush
against the lead cutout at a distance of 30 cm from the x-ray
source. Cones are not used at ARPANSA, but a 20 mm lead
aperture 15 cm from the source constrains the beam diameter. The EBT has approximately 100 m of Mylar before a
50 m active layer and therefore measures the dose deposited between 100 and 150 m depth in Mylar. The EBT was
calibrated with the ARPANSA 300 kVp HVL 3.8 mm Cu
beam. Known doses were delivered to ten film samples,
which were then scanned after a period of 24 h, and a calibration curve of dose to scanner output was obtained. The
scanner was an Epson Perfection V700 photo scanner. An
optical density dependant scanner nonuniformity correction
was applied. Comparison of film profiles to profiles obtained
with ionization chamber measurements indicate a relative
film dose accuracy of the order of 1%.
Medical Physics, Vol. 37, No. 8, August 2010

III.A. Dose enhancement with and without gladwrap

The first measurements were taken with the highest energy 300 kVp beam incident on a 25 mm diameter 3 mm
thick round lead cutout. Figure 2a shows images of dose
deposited between 100 and 150 m depth in Mylar from the
film measurement and the MC model. The dose profiles, normalized to the central dose, are shown below the images. The
clear dose enhancement near the edges of the lead perturbs
the dose distribution that ideally should be uniform. There is
good agreement between the model and the film measurements, with both showing a ratio of the edge dose to the
central dose of approximately 1.4. The edge dose is slightly
lower in places on the film compared to the MC in Fig. 2a
where no gladwrap is used. This is presumably due to air
gaps between the film and the lead as discussed in Sec. III C.
Figure 2b shows the same measurement when ten layers
of gladwrap approximately 200 m of polyethylene is
wrapped around the lead. The polyethylene absorbs the electron contamination and the desired uniformity in the dose
distribution is seen.
III.B. Dose enhancement with depth in a water
phantom

The dose enhancement was modeled with depth in a water

phantom using DOSXYZnrc to simulate a 20 20 mm2
square lead cutout with 300 HVL 3.8 mm Cu and 180 kVp
HVL 0.99 mm Cu beams. Figure 3 shows the profiles for
five different average depths in water from 35 to 235 m
and images showing the dose at three depths. The profiles are
taken across the center of the square and normalized to the
central dose at a depth of 235 m. The edge dose enhancement reduces rapidly with depth in water and reduces much
more rapidly for the lower energy beam due to the reduced
penetration of the low energy electrons.

3937

Lye, Butler, and Webb: Enhanced epidermal dose from lead cutouts

3937

FIG. 4. The modeled dose enhancement at the edges of the lead cutout for
round, square, and serrated edge cutouts, all for 300 kV HVL 0.38 mm Cu,
from the MC model. There is no polyethylene shielding.

FIG. 3. Reduction of the contaminant edge dose with depth in water with no
gladwrap from the MC model. A 300 and a 180 kV beam are considered.
A 2 2 cm2 square lead cutout is investigated. The profiles show a slice
across the center of the profile. Note that the contaminant edge dose is
higher in the corners of the square.

The images in Fig. 3 show that the edge enhancement is

much higher in the corners of the square as opposed to the
sides of the square. This suggests that the shape of the lead
cutout is important in determining the edge dose enhancement. A variety of different shape and size cutouts were
modeled, including a serrated edge cutout or pie crust cutout. Serrated edge cutouts are sometimes used to improve the
cosmetic outcome by softening the residual scar border at the
radiotherapy treatment site.10 An interesting question arises
as to whether the improved cosmetic result is due to the
serrations smearing the edge dose over a larger area over
many fractions, or whether the serrated shape physically reduces the local electron contamination. To address this point,
the maximum and minimum edge dose enhancement in the
serrated cutout is compared to the round and square cutouts.
The image inset in Fig. 4 shows a film image of the dose
from such a serrated edge cutout using the 300 kVp beam.
The maximum edge dose enhancement occurs in the triangular sections surrounded by the lead, while the minimum occurs in between the triangular sections. The serrated cutout
was modeled by a 2 cm square with serrations spaced every
two millimeters on one edge.
Figure 4 shows the change in edge dose enhancement
with depth in water for 1 and 2 cm diameter round cutouts,
the side and corner dose enhancement for a 2 cm square
cutout, and the maximum and minimum edge dose enhancement for the serrated lead cutout. For each series, the edge
Medical Physics, Vol. 37, No. 8, August 2010

dose is shown relative to the central dose at the maximum

depth of 235 m. The minimum edge dose enhancement in
the serrated cutout was the same as the side dose enhancement in the square cutout. The maximum edge dose enhancement from the serrated cutout was much higher than the
square cutout side dose enhancement. This suggests that the
serrated shape does not physically reduce the electron contaminant dose enhancement, but actually increases the local
dose at the edge of the treatment field in a single fraction.
The improved cosmetic outcome with a serrated edge may be
due to averaging out the edge dose over a larger area over
many fractions and the reduction of the hard line at the edge
of the treatment field. The serrated edge cutout was also
modeled with 200 m polyethylene under the lead. The serrated field shape is unchanged, but the ratio of edge dose:central dose is reduced to less than 1.1 for all depths and
edge positions.
A larger range of energies was modeled to detail the behavior of the electron contaminant dose enhancement with
depth in water. A series of 2 cm square lead cutouts with an
incident beam energy of 300 HVL 3.8 mm Cu, 250 HVL
2.5 mm Cu, 180 HVL 0.99 mm Cu, 150 HVL 0.60 mm
Cu, 100 HVL 6.53 mm Al, and 50 kVp HVL 0.79 mm
Al were modeled. An average of the side and corner dose
relative to the central dose at 235 m was taken as a broad
indication of edge dose enhancement for all cutout shapes.
Exponential fits to the edge dose enhancement versus depth
in water is shown in Fig. 5 for all energies excluding 50 kV;
50 kV was omitted for clarity as the dose enhancement did
not penetrate past the first 10 m, equivalent to the dead
outer layer of the epidermis.
The exponential fits to the edge dose enhancement have
been overlayed on a schematic of the skin. The schematic
shows the stratum corneum the dead layer, the basal layer
of the epidermis at the epidermal/dermal border, and the upper dermis. The thickness of the epidermis is critical to un-

3938

Lye, Butler, and Webb: Enhanced epidermal dose from lead cutouts

FIG. 5. The edge dose enhancement with depth in water for energies ranging
from 300 to 100 kVp from the DOSXYZnrc model with no polyethylene
shielding. The edge dose enhancement is for a 2 2 cm2 square lead cutout,
taking an average of the side and corner doses relative to the central dose at
235 m depth. An exponential fit of the MC data points is shown in the
figure. The figure is overlaid on a schematic of the structure of facial skin.

derstanding the problem posed by the shallow contaminant

dose enhancement. The thickness of the epidermis varies
greatly with body site, as well as varying between individuals. Whitton and Everall11 observed a forehead epidermis
thickness of 50 m versus a fingertip thickness of 369 m.
More recently, Gambichler et al.12 measured the average epidermal thickness on the forehead, pectoral, scapular, forearm, buttock, and calf to be approximately 70 m for the
younger group and 60 m for the older group. In this study,
the results of Gambichler et al.12 are used. The forehead
thickness of the older group was taken as appropriate for
kilovoltage radiotherapy, defining the epidermal thickness in
this study as 60 m.
From Fig. 5 we can extract the edge dose enhancement at
pertinent depths. Three depths were chosen as the most relevant for this study: The average epidermis dose, the
epidermal/dermal border, and the upper dermis. Figure 6
shows the edge dose enhancement as a function of x-ray tube

FIG. 6. The edge dose enhancement as a function of kV for three pertinent

depths. There is no polyethylene shielding in this model.
Medical Physics, Vol. 37, No. 8, August 2010

3938

voltage for these depths. The epidermal/dermal border was

defined as 60 m depth. The average viable epidermal depth
was defined as 35 m, excluding the first 10 m of dead
cells. The upper dermis was considered at a depth of
100 m. The epidermal/dermal border is radiobiologically
important. Radiation damage at this depth can cause
erythema due to inflammation from the breakdown of cells in
the actively dividing basal layer. Damage to the melanocytes
found in this layer can lead to melanin deposition in the
dermis and result in local hyperpigmentation.13
Edge dose enhancement to the average epidermis is large;
about four times the central dose at 235 m and roughly
constant from 150 to 300 kVp. This behavior can be qualitatively understood by considering that although more contaminant electrons will be produced from higher energy x
rays, the electrons will also penetrate further and will not
deposit all of their energy at the shallow depths around
35 m. At the epidermal/dermal border, we see about
double edge dose enhancement for 150 kVp increasing up to
three times edge dose enhancement for 300 kVp. For
100 kVp and below, there is no significant edge dose enhancement at the epidermal/dermal border depth. At the upper dermis, 250 and 300 kVp show about double edge dose
enhancement and below 150 kVp, there is no significant
edge dose enhancement.
For each of the kilovoltage energies, layers of gladwrap
20 m polyethylene were added to the model until the
dose enhancement was reduced to less than 10% in the
10 60 m epidermal layer. Table I shows the results from
this investigation.
III.C. Air gaps

The effect of an air gap between the lead cutout and the
skin has been investigated. Figure 7 shows the case of a
300 kVp beam incident on a 25 mm diameter lead cutout
modeled with the lead flush on the water surface, with 0.2

FIG. 7. The MC dose profile for a round 2.5 cm lead cutout at a depth
between 10 and 60 m with no shielding, 0.2 mm of polyethylene, and 0.2
mm of air gap.

3939

Lye, Butler, and Webb: Enhanced epidermal dose from lead cutouts

TABLE I. Layers of gladwrap required to reduce edge dose enhancement in

the 10 60 m epidermal layer to less than 10%.
Tube voltage
kV
300
250
180
150
100
50

HVL

Layers
of gladwrap

Edge dose: Central dose

between 10 and 60 m

3.8 mm Cu
2.5 mm Cu
0.99 mm Cu
0.60 mm Cu
6.53 mm Al
0.79 mm Al

10
9
5
4
2
0

1.10
1.07
1.06
1.09
1.03
1.00

mm shielding of polyethylene, or with 0.2 mm of air gap.

The profile is of the dose deposited between 10 and 60 m
water depth, equivalent to the viable epidermis. The 0.2 mm
of polyethylene cuts out the edge dose enhancement almost
completely, while the air gap only reduces the contaminant
dose slightly.
IV. CONCLUSIONS
The lead cutouts used in kilovoltage radiotherapy cause a
large electron contaminant dose enhancement at the edge of
the treatment field at shallow depths. Such a dose enhancement is undesirable as it could lead to unnecessary erythema
and hyperpigmentation at the border of the treated and untreated skin and a poorer cosmetic outcome. The edge dose
enhancement at a particular depth in water is highly energy
dependent. A pertinent depth to evaluate the likelihood of
clinical effect is the epidermal/dermal border, defined in this
study as 60 m depth in water. At this depth, there is significant edge dose enhancement for beams above 100 kVp.
A 150 kVp beam has approximately double edge dose enhancement increasing to three times edge dose enhancement
for 300 kVp. Plastic film such as gladwrap shielding
wrapped around the lead cutout, or placed under the lead
cutout, is a simple way to remove the electron contamination. Alternatively, an appropriate thickness of varnish could
be painted onto the lead.6 Note that if plastic film covers the
opening in the lead cutout for the treatment, the plastic film
should also be used in dosimetry.

Medical Physics, Vol. 37, No. 8, August 2010

3939

ACKNOWLEDGMENTS
The authors thank Marianne Rinks from the Northern
Sydney Cancer Centre for the provision of a serrated lead
cutout.
a

Electronic mail: jessica.lye@arpansa.gov.au

S. C. Klevenhagen, D. DSouza, and I. Bonnefoux, Complications in
low energy x-ray dosimetry caused by electron contamination, Phys.
Med. Biol. 36, 11111116 1991.
2
P. Andreo, D. T., Burns, K. Hohlfeld, M. S. Huq, T. Kanai, F. Laitano, V.
Smyth, and S. Vynckier, Absorbed dose determination in external beam
radiotherapy: An international code of practice for dosimetry based on
standards of absorbed dose to water, IAEA Technical Report Series No.
398, ed. V.11b, 2004.
3
C.-M. Ma, C. W. Coffey, L. A. DeWerd, C. Liu, R. Nath, S. M. Seltzer,
and J. P. Seuntjens, AAPM protocol for 40300 kV x-ray beam dosimetry in radiotherapy and radiobiology, Med. Phys. 28, 868893 2001.
4
A. S. Shiu, S. S. Tung, R. J. Gastorf, K. R. Hogstrom, W. H. Morrison,
and L. J. Peters, Dosimetric evaluation of lead and tungsten eye shields
in electron beam treatment, Int. J. Radiat. Oncol., Biol., Phys. 35, 599
604 1996.
5
C. H. M. Lee and K. K. D. Chan, Electron contamination from the lead
cutout used in kilovoltage radiotherapy, Phys. Med. Biol. 45, 18
2000.
6
D. W. Thomas and J. A. Clark, Reducing the skin dose from secondary
electrons in kilovoltage radiotherapy: A pliable coating for custom lead
cut-outs, J. Radiotherapy in Practice 8, 233237 2009.
7
B. Walters, I. Kawrakow, and D. W. O. Rogers, DOSXYZnrc users
manual, NRCC Report No. PIRS-794, rev. B, 2007.
8
D. M. Tucker, G. T. Barnes, and D. P. Charkraborty, Semiempirical
model for generating tungsten target x-ray spectra, Med. Phys. 18, 211
218 1991.
9
J. E. Lye, D. J. Butler, and D. W. Webb, Monte Carlo correction factors
for the ARPANSA kilovoltage free-air chambers and the effect of moving
the limiting aperture, Metrologia 47, 1120 2010.
10
B. Hernndez-Machin, L. Borrego, M. Gil-Garcia, and B. H. Hernandez,
Office-based radiation therapy for cutaneous carcinoma: Evaluation of
710 treatments, Int. J. Dermatol. 46, 453459 2007.
11
J. T. Whitton and J. D. Everall, The thickness of the epidermis, Br. J.
Dermatol. 89, 467476 1973.
12
T. Gambichler, R. Matip, G. Moussa, P. Altmeyer, and K. Hoffman, In
vivo data of epidermal thickness evaluated by optical coherence tomography: Effects of age, gender, skin type, and anatomic site, J. Dermatol.
Sci. 44, 145152 2006.
13
C. K. Bomford and I. H. Kunkler, Walter and Millers Textbook of Radiotherapy: Radiation Physics Therapy and Oncology, 6th ed. Churchill
Livingstone, Edinburgh, NY, 2003.
1