Copyright 0 Munksgaard 1996

ACTA NEUROLOGICA
SCANDINAVICA
ISSN OOO1-6314

Acta Neurol Scand 1996: 94: 264-268

Printed in UK - all rights reserved

Epidemiological and clinical characteristics

of ALS in Belgrade, Yugoslavia
AlEaz S, Jarebinski M, Pekmezovic T, Stevic-Marinkovic Z, PavloviC S,
Apostolski S. Epidemiological and clinical characteristics of ALS in
Belgrade, Yugoslavia.
Acta Neurol Scand 1996: 94: 264-268. 0 Munksgaard 1996.

We present the results of the first epidemiological study of ALS in Belgrade.

The distribution of 58 newly discovered cases in a 7-year survey period
(1985-1991) showed that the average annual age-adjusted incidence rate was
0.42 per 100,000 population (95 YOconfidence interval, 0.18483). The rate
for males was 1.5 times higher than the rate for females. The greatest agespecific average incidence rate was observed in patients between 60 and 64
(3.66 per 100,000population; 95% confidence interval, 2.17-5.78). The actual
age-adjusted prevalence rate on December 31, 1991 was 1.07 per 100,000
(95% confidence interval, 0.71-1.71). The mean age at onset of the disease
was 56.21t9.8and it ranged from 24 to 74. We studied the natural course of
the disease through the mean duration and cumulative probability of
survival. The mean duration of the disease was 27.7f18.2 months. The
cumulative probability of survival was 27% for the whole population in a
5-year interval. Elderly patients and those with bulbar signs at onset had a
poorer prognosis. Patients under 49 at onset and those with the spinal form
of the disease survived longer.

The frequency of ALS is relatively uniform

throughout the world with the exception of
endemic regions. Recent epidemiological studies
have shown that the average annual incidence rates
range from 0.59 (1) to 2.4 (2) per 100,000 inhabitants. Lower average rates were reported mainly
for less developed countries, probably due to inaccurate diagnosis (3). All morbidity and mortality
studies showed male preponderance, indicating
that the sex is a risk factor for ALS. An increasing
age-specific risk with increasing age and well
defined peak of incidence between 55 and 75 years
has been demonstrated. Age-specific incidence
rates in Rochester appeared to rise continuously
with age, but without the earlier described peak
(4). Mortality studies in Sweden ( 5 ) and Italy (6)
confirmed the results of American survey, but some
differences were observed. Age-specific rates were
maximal in the 70-79 age group in Sweden and in
the 65-69 age group in Italy. Nevertheless, cohort
analysis in both studies proved that the age-specific
rates tend to increase progressively without decline
in the elderly. These results support the hypothesis
that ALS is pathogenetically related to ageing.
264

S. AICaz', M. Jarebinski',
T. PekmezoviC',
2. SteviC-MarinkoviC3, S. PavloviC3,
S. Apostolski3
Institutes of 'Addictions - Neurocenter, Belgrade,
2Epidemiology, 3Neurology. School of Medicine,
University of Belgrade, Yugoslavia

Key words: amyotrophic lateral sclerosis;

epidemiology; clinical characteristics

S. Apostolski. Institute of Neurology, School of

Medicine, University of Belgrade, 11000 Belgrade,
Dr SubotiCa 6. Yugoslavia
Accepted for publication May 29. 1996

Material and methods

The district of Belgrade is divided into 16 units and

extends over an area of 3,222 km2. According to
the 1991 census, the area has a population of
1,602,226 inhabitants. ALS cases were collected by
analysing hospital registers in the Institute of
Neurology which is the referential neurological
center, and in all other departments of neurology
which belong to additional two Clinical centers and
one Military Hospital in Belgrade. Due to the
severity of the disease and diagnostic procedures
required, hospitalisation was necessary for a
correct diagnosis of ALS in our conditions. Only
clearly diagnosed and fully documented cases were
included. The presented rates are the minimum
rates, since some cases may not have been hospitalized and therefore not registered. The diagnostic
criteria included progressive upper and lower
motor neuron signs in bulbar and two spinal
regions, or upper and lower motor neuron signs
in three spinal regions, absence of sensitive
and bladder function disturbances and typical
EMG findings. Radiological examinations (X-rays,

Epidemiology of ALS
myelography, CAT scans and MRI) were carried
out when other conditions had to be ruled out. The
outcome in 54 cases was determined by directly
contacting patients and their families. Four patients
(2 men and 2 women) could not be reached for
follow-up and dropped out of the study after
diagnostic hospitalisation.
Clinical methods included a physical and neurological examination (48 patients examined by S.A.
and Z.S-M.) or a careful review of hospital records
with special attention to the complaints of the
patients and their neurological status (10 patients).
A11 rates were adjusted to a standard in order to
make them comparable with rates in other populations. We used the world population according to
Segi (7) and the method of direct standardization.
The calculation of the confidence interval (CI) was
based on a Poissons frequency distribution for rare
events (8). Cumulative risks were derived from
age-specific rates using the density method (9).
The differences between two arithmetic means
were tested by Students t-test. Pearsons X2 test
was used to assess differences between frequencies.
The probability of survival was obtained by the life
table method (10). Differences in survival between
subpopulations were evaluated by long-rank test
(11).
The source of population data were official published and unpublished documents of the Republic
Institute for Statistics of Serbia.
Results

In the 1985-1991 period, 58 patients with ALS were

discovered in the Belgrade district. In five patients
the diagnosis was confirmed during a later hospitalisation, not once was it denied. Thirty-two
patients died before December 31, 1991. Four
patients dropped out of the study after the diagnosis was established, since no contact with them was
possible. Our series comprised 35 men and 23
women. The youngest patient (woman) was 24 at
onset and the two oldest patients (women) were
74. The mean age at onset of symptoms was
56.2-t-9.8(55.9f7.6 in the male and 56.6k12.5 in the
female group). Even though women were somewhat older than men at onset of the disease, this
difference was not statistically significant (t=0.805;
p>0.05).
?he clinical presentation at the onset of the
disease was bulbar in nine (15.5%) and spinal in
49 (84.5Y0)patients. The latter group comprised of
42 patients with classic ALS and 7 patients with the
pseudo-polyneuritic (peroneal) form of ALS. The
mean age at onset was 56.of9.5 in the group with
spinal ALS. The patients with bulbar ALS were
older at onset (57.3f9.Q but this difference was

not statistically significant (t=0.49; p>0.05). Only

one case of familial ALS was registered. This
patient noticed the first symptoms at the age of 62
and died 23 months later. He was affected with the
classic form of the disease. The patients father and
uncle died from the same disease at the age 51 and
76, respectively.
Clinical characteristics

Special attention was paid to the initial symptoms

during data collection. The time of onset was not
obtained for one patient since part of her medical
documentation was lost. Due to the slowly progressive course of the disease, the patients were often
unsure as to the time and localisation of their initial
symptoms. Some patients with clear signs of bulbar
palsy, including hypotrophy of the tongue muscles,
claimed that their speech and swallowing disturbances were of abrupt onset, often after stress
situations. Others complained of disturbances
which can in no way be related to motor neuron
disease. In all cases where, in our opinion, the
symptom in question could not be caused by
pathological processes in ALS, the first following
symptom was taken as the initial symptom (Table
1).The most prominent initial symptom was weaknesS No sensitive symptoms were verified objectively at neurological examination. Cramps and
pain were the cause of diagnostic difficulties in
some patients; rheumatic disease, gonarthrosis and
lumbago had to be ruled out in these cases. In 30%
cases the weakness at onset was focal, i.e. it was
limited to a single limb.
incidence and prevalence

Using the distribution of new cases of the disease

beginning in a given year, we calculated the so
called crude rates. The average annual incidence
rate during the period 1985 through 1991 was 0.68
per 100,OOO population (95 % CI, 0.22-1.58) for men,
0.41 per 100,000 population (95% CI, 0.09-1.20) for
Table 1. Initial symptoms in ALS population in Belgrade
Symptoms

No.

Weakness
Amyotrophy
Fasciculations
Stiffness
Cramps and pain
Dysaesthesiae
Dysarthria and dysphagia
Loss of weight
No data
Total

28
1

(49.1)

5
3
6

il .El
(8.8)
15.21
(10.5)

4
9

(7.0)
(15.8)

(1.8)

(1.8)
(1 00)

58

265

AlEaz et al.
Table 2. Age-adjusted incidence rates of ALS in Belgrade in comparison with the results in other studies

Risk

Study

Rate

Male
Minnesota. 1925-842
Mexico. 1962-693
Ferrara. Italy. 1964-824
Florence, Italy, 1967-7615
Middle Finland, 197641
Messina. Italy. 1976-8516
Harris county, Texas 1985-88
Belgrade, Yugoslavia. 1985-91

2.91
0.31
0.62
0.68
2.04
0.90
1.27
0.56

1.83440
0.10-0.72
0.41-2.90
0.50-0.91
1.1 4-3.37
0.61-1 27
0.95169
0.1&1.30

0.0038
0.0001
0.0008
0.0008
0.0023
0.0010
0.0015
0.00075

1
1
1
1
1
1
1
1

263
8461
1250
1250
435
1000
667
1508

Female
Minnesota, 1925-84
Mexico. 1962-693
Ferrara, Italy, 1964-132~
Florence, Italy, 1967-7615
Middle Finland, 1976-812
Messina, Italy. 1976-8516
Harris county, Texas 1985-88
Belgrade, Yugoslavia. 1985-91

1 .89
0.46
0.34
0.47
2.29
0.22
1.03
0.36

1.18-2.85
0.30-0.82
0.20-0.55
0.33-0.64
1.42-3.51
0.10-0.43
0.75-13 8
0.190.87

0.0025
0.0002
0.0004
0.0005
0.0027
0.0002
0.0013
0.00045

1
1
1
1
1
1
1
1

400
5709
2500
2000
370
5000
769
2252

~~~

95% CI

Cumulative Incidence*

Rates per 100,000population. 1970 US total population as standard t 95% Confidence interval
using incidence density method (to age 74 years)

females, and 0.54 per 100,000 population (95% CI,

0.23-1.06) for both sexes. These rates were then
adjusted according to the Segis world population
and the method of direct standardization. The
average age-adjusted annual incidence rate was
0.52 (95% CI, 0.17-1.21) for men and 0.34 (95%
CI, 0.07-0.99) for women; for both sexes it was 0.42
(95% CI, 0.18-0.83) per 100,000 population. The
male/female ratio of age-adjusted rates was 1.5 : 1.
Age standardized to the US population of 1970,
the incidence rate was 0.56/100,000 for men, 0.36/
100,000 for women and 0.44/100,000 for the total
population. In comparison with other populations,
our incidence rates are close to Italian (Table 2).
The incidence rates in USA and Finland are three
to five times greater. Cumulative incidence served
to assess the individual risk of acquiring ALS in
Belgrade. In the total Belgrade population under
74 the risk for ALS was 1 : 1 869 (1 : 1508 for men
and 1 : 2 252 for women). It is obvious that the
maximal risk of being affected with ALS for all
three categories lies within the 60-64 age range
with a tendency to decline further on (Fig. 1).The
age-specific incidence rate in the 60-64 age group
was 4.85/100,000 (95% CI, 2.42-8.68) for males, and
3.02/100,000 (95% CI, 1.30-5.95) for women, while
it was 3.66/100,000 (95% CI, 2.17-5.78) for both
sexes.
On December 31, 1991, 22 patients were alive
(11 men and 11women). Age-adjusted point prevalence rates for 1991 were: 1.00 (95% CI, 0.50-1.79)
per 100,000 population for males, 1.14 per 100,000
population (95% CI, 0.57-2.04) for females, and
1.07 per 100,000 (95% CI, 0.71-1.70) for the total

266

t Computed using incidence density method (to age 80 years)

Computed

Fig. 1. Average annual age- and sex-specific incidence rates of

ALS per 100,OOO population.

population. In correlation with the annual incidence rates for 1991 which were 1.6 times greater
for males, this fact indicates a longer duration of
the disease in women.
In order to gain better insight into the natural
course of ALS in our population, we focused
special attention on 32 patients who died during
the survey period. The onset of the disease was
determined by the moment of appearance of the
first symptoms and not by the time the diagnosis
was established; this was done for greater precision.
The mean duration of ALS was 27.7f18.2 months
(approximately 2 years and 4 months). Women

Epidemiology of ALS
lived almost 16 months longer. The mean duration
of the disease was 38.6e5.6 months in the female
group and 22.8H1.0 in the male group. This difference was statistically significant (t=3.8; ~ ~ 0 . 0 1In) .
relation to the form of the disease, the mean
survival for patients with spinal ALS was approximately 2 months longer (28.W17.7) than for those
with bulbar ALS (26.0f19.8). This difference was
not statistically significant (t=0.27; ~>0.05).
Patients under 49 lived almost 10 months longer
than those over 49 (36.3k12.0 and 26.8f16.9,
respectively); the difference in survival between
these two subgroups was also statistically insignifican t (t=0.92; p>0.05).
The statistical probability of survival of our
patients was derived from life table analyses of
ALS. This method includes patients who were lost
during the study as well as those who were living
at the end of the survey period. The cumulative
probabilities of survival of a two-, five- and sevenyear period were 62%, 27%, and 27%, respectively.
These findings refer to the total population.
We also analysed survival in relation to sex, age
and clinical form of ALS. There was a difference
of survival according to sex. A five-year survival
was 22% for men and 36% for women; the same
was noted after a seven-year survey period. The
long-rank test detected no statistically significant
difference (X2=2.65;p>0.05).
Survival was compared between subgroups of
patients with different forms of the disease (spinal
and bulbar ALS). At the end of a four-year period
not one of the patients with bulbar ALS was living,
while 38.4% patients with spinal ALS were still
alive (Fig. 2). This difference was statistically highly
significant (long-rank test; X2=3.44;k=l; p<O.Ol).
There was a significant difference in survival
between patients under and over 49 (Fig. 3). A twoyear survival for patients under 49 was 88% and
57% for those over 49. At the end of a five-year
period these figures were 53% and 23%, respectively. More than 50% of the patients under 49 and
only 19% of those over 49 were alive at the end of
the survey period (X2=21.87;k=l; ~ ~ 0 . 0 1 ) .
Discussion

During the period from 1985 to 1991, 58 patients

were diagnosed as the newly cases of ALS in
Belgrade. The most frequent initial symptom in the
population of our patients was weakness (49%).
Lopez-Vega et al. (18) and Gubbay et al. (19) noted
weakness as the initial complaint in 63% and 86%
cases, respectively. Other symptoms included dysphagia and dysarthria; the frequency of these symptoms correlates with the frequency of bulbar ALS.
Cramps and pain often present at onset (10%).

Fig.2. Survival of ALS patients according to clinical forms.

<,
3..
20.10

..

07

Stiffness as a spasticity equivalent is rarely encountered and it was noted in only 5% of our patients,
what is very similar to the results (1%)obtained in
Israel (19).
The adjusted incidence rate of ALS in Belgrade
was 0.42 per 100,000 population in a 7-year period
(1985-1991). Low rates were registered in Mexico
(0.4/100,000) (13) and Poland (0.22/100,000) (20).
The latter is the lowest rate reported so far. With
the exception of Italy where the incidence rates
were similar to ours (0.44-0.78 per 100,000 population) (14-16, 21-23), we are not aware of any
descriptive epidemiological study conducted in
neighbouring countries. Studies in some Northern
European countries, as well as in Canada and USA
show incidence rates up to 5 times greater (2,22-24).
267

Altaz et al.
The percentage of patients under 49 is relatively
high in our population (19%). Not one incident
case was found in the group of patients over 75.
Conversely, incidence rates in patients over 75 are
high in other populations. The maximal rate for this
group was found in Rochester (4). This fact may
indicate some missed diagnoses of ALS in our
oldest population. At older age, symptoms of ALS
are often interpreted by others as part of the ageing
process. Furthermore, physicians who first see the
patient are more motivated to refer younger
patient to a specialist. The elderly are also less agile
and rarely visit neurological centres.
Advanced age and bulbar signs at onset indicated poorer prognosis in our population. The
patients under 49 had significantly higher survival
rates than the older patients. The more benign
course of ALS in younger patients was also confirmed by the other authors (18,22-25). The bulbar
form of ALS had a significantly lower survival rate
in comparison with the spinal form. These findings
are also consistent with findings in other populations (19, 26).
References ;
1. GIAGHEDDU
M, PUGGIONI
G, MASALA
C et al. Epidemiological study of amyotrophic lateral sclerosis frequency in
Sardinia, Italy. Acta Neurol Scand 1983: 68: 394-404.
2. MURROS
K, FOGELHOM
R. Amyotrophic lateral sclerosis in
Middle Finland: an epidemiological study. Acta Neurol
Scand 1983: 67: 41-47.
3. KURTZKE
JF. Epidemiology of amyotrophic lateral sclerosis.
In: Rowland LP, ed. Human motor neuron disease. New
York: Raven Press, 1982: 281-300.
4. JUERGENS
SM, KURLAND
LT, OKAZAKI
H, MULDER
DW.
ALS in Rochester, Minnesota, 1925-1977. Neurology 1980:
30: 463-470.
5. GUNNARSON
LG, LINDBERG
G, SODERFELT
B, AXELSON0.
The mortality of motor neuron disease in Sweden, Arch
Neurol 1990 47: 42-46.
6. CHIO A, MAGNANI
C, SCHIFFERD. Amyotrophic lateral
sclerosis mortality in Italy, 1958 to 1987: A cross-sectional
and cohort study. Neurology 1993: 43.
7. DOLLR. Comparison between registries: age standardized
rates. In: Waterhouse J et al, eds Cancer Incidence in Five
Continents. Lyon: Iarc Scientific Publications, 1976: 111:
454-459.
8. HAENSZEL
W, LOWLAND
DB, SIRKENMG. Lung cancer
mortality as related to residence and smoking histories.
J Natl Cancer Inst 1962: 28: 1000-1001.
9. KLEINBAUM
DG, KUPPER
LL, MORGENSTERN
H. Epidemio-

logic research. Belrnont: Lifetime Learning Publications,

1982.
10. Cox DR. Regression models and life tables. J R Stat SOC
B 1972: 34.
11. MANTELN. Evaluation of survival data and two new
rank order statistics arising in its consideration. Cancer
Chernoter Rep 1966: 50: 163-170.
12. YOSHIDAS, MULDER DW, KURLANDLT, CHU CP,
OKAZAKIH. Follow-up study on amyotrophic lateral
sclerosis in Rochester, Minessota, 1925 through 1984.
Neuroepidemiology 1986: 5: 61-70.
L, SANESTEBAN
E, ALTERM. Mexican resist13. OLIVARES
ance to amyotrophic lateral sclerosis. Arch Neurol 1972:
27: 397-402.
E, CARRERAS
M, TOLAR, PINNAL, PAOLINO
E,
14. GRANIERI
DAGOSTINO
G. Motor neuron disease in the province of
Ferrare, Italy, 1964-1982. Neurology 1988: 38: 1604-1608.
I, ANTUONO
P, AMMADUCCI
L. Study of epidemio15. BRACCO
logical and etiological factors of amyotrophic lateral
sclerosis in the province of Florence, Italy. Acta Neurol
Scand 1979: 6 0 112-124.
P, MALARA
CE, MARABELLO
L et a]. Amyo16. DE DOMENCIO
trophic lateral sclerosis: an epidemiological study in the
province of Messina, Italy, 1976-1985. Neuroepidemiology
1988: 7: 152-158.
17. AEGERS
JF, APPELS, PERKINS
P, LEE J. Incidence and
prevalence of amyotrophic lateral sclerosis in Harris
County, Texas, 1985-1988. Arch Neurol 1991: 48: 589-593.
18. LOPEZ-VEGA
JM, CALLEJAJ, COMBARROS
0, POLOJM,
BERCIANO
J. Motor neuron disease in Cantabria. Acta
Neurol Scand 1988: 77: 1-5.
19. GUBBAY
SS, KAHANA
E, ZILBER
N, COOPERG, PINTOV
S,
LEIBOWITZ
Y. Amyotrophic lateral sclerosis. A study of its
presentation and prognosis. J Neurol 1985: 232: 295-300.
20. CENDROWSKI
W, WENDERM, OWSIANOWSKI
M. Analyse
epidemilogique de la Grande-Pologne. Acta Neurol Scan
1970 46: 609-617.
21. SALEMI G, FERRO B, ARCARA A, CASSATAM,
CASTIGLIONE
MG, SAWTTIERIG. Amyotrophic lateral
sclerosis in Palermo, Italy: An epidemiologic study. Ital J
Neurol Sci 1989: 10: 505-509.
22. SCARPA
M, COLOMBO
A, PANZETIIP, SORGATO
l? Epidemiology of amyotrophic lateral sclerosis in the province of
Modena, Italy. Influence of environmental exposure to
lead. Acta Neurol Scand 1988: 77: 456-460.
23. RosATi G, PIA L, GRANIERI
E et al. Studies on epidemiological, clinical and etiological aspects of ALS disease
in Sardinia, Southern Italy. Acta Neurol Scand 1977: 55:
231-244.
24. MURRAY
LP et al. Amyotrophic
TJ, CAMERON
J, HEFFERNAN
lateral sclerosis in Nova Scotia. Adv Exp Med Biol 1987:
209: 345-349.
25. BOZEKC, EISENA. Acquired motor neuron disease of the
young. Neurology 1985: 35 (Suppl 1): 242.
26. TYSNESOB, VOLLSETSE, AARLIJA. Epidemiology of
amyotrophic lateral sclerosis in Hordaland county, western
Norway. Acta Neurol Scand 1991: 83.