вход по аккаунту


Incidence OF systemic sclerosis in allegheny county Pennsylvania. A twenty-year study of hospital-diagnosed cases 19631982

код для вставкиСкачать
Vol. 40, No. 3, March 1997, pp 441-445
0 1997, American College of Rheumatology
A Twenty-Year Study of Hospital-Diagnosed Cases, 1963-1982
Objective. To determine the incidence of hospitaldiagnosed systemic sclerosis (SSc) among residents of
Pittsburgh and Allegheny County, PA, from 1963
through 1982.
Methods. Medical records from all Allegheny
County hospitals were searched using International
Classification of Diseases codes for patients diagnosed
with SSc. Each case was reviewed and verified by a
physician, according to specified criteria. Age-adjusted
incidence rates were computed for race and sex, and
results were expressed as new cases per million population per year, with the 1970 Allegheny County population serving a s the standard.
Results. Four hundred forty-four cases of SSc
were identified during this 20-year survey, for a total
annual incidence of 13.9 per million population. Overall, the incidence rate doubled during 1973-1982 compared with the first time interval of the study, with the
greatest increase occurring in women. Among the
younger population (ages 15-24), black women had the
highest incidence of SSc (21.2 per million population).
Overall, the female-to-male incidence ratio was 3:1, and
was slightly higher (3.41) during the childbearing years
Supported by grants from the Commonwealth of Pennsylvania (Department of Health), Harrisburg; the RGK Foundation, Austin, Texas; and the Arthritis Foundation, Western Pennsylvania Chapter (Shoemaker Fund), Pittsburgh.
Virginia D. Steen, MD: Georgetown University, Washington,
DC; Chester V. Oddis, MD, Claudia G. Conte, MA, MPH, Janine
Janoski, PhD, Gayle Ziegler Casterline, RN, Thomas A. Medsger, Jr.,
MD: University of Pittsburgh School of Medicine, Pittsburgh,
Address reprint requests to Thomas A. Medsger, Jr., MD,
Division of Rheumatology and Clinical Immunology, Liliane Kaufman
Building, Suite 502, 3471 Fifth Avenue, Pittsburgh, PA 15213.
Submitted for publication February 28, 1996; accepted in
revised form September 18, 1996.
(ages 15-44). Men under age 35 and children infrequently developed SSc.
Conclusion. The incidence of SSc reported in this
study increased with time. Although this trend was most
likely due to improved detection and medical record
techniques, the incidence rate stability in men compared with women suggests that there may have been a
true increase in the incidence of SSc in this population,
particularly in women.
Systemic sclerosis (SSc; scleroderma) is a generalized, chronic, connective tissue disease that is characterized by vascular, skin, joint, and internal organ abnormalities. Epidemiologic studies have been hampered by
the rarity of the disease, the variability of its clinical
manifestations, and the tendency for features of scleroderma to overlap with those of other recognized connective tissue disorders.
Several studies have described the epidemiologic
features of SSc in the US (1-3). A hospital-based
epidemiologic survey of scleroderma in Memphis and
Shelby County, TN from 1947 to 1968 showed an
average annual incidence of 4.5 cases per million population (4). Michet et a1 (5) and Medsger and Masi (6)
both found an annual incidence of 10 cases per million in
Rochester, MN and in US male veterans, respectively. A
somewhat lower annual incidence has been reported in
other countries, including Poland (7), New Zealand,
with 2.3 cases per million (8), Russia, with 7 cases per
million, and England, with 3.7 cases per million (9).
The present study was designed to determine the
incidence of hospital-diagnosed SSc in Pittsburgh and
Allegheny County, PA from 1963 to 1982. Based on our
findings, we have defined patterns of age-, sex-, and
race-specific incidence and have identified temporal
trends during this 20-year time period.
Case detection and disease classification. The protocol
followed in this study was identical to that reported in a study
of patients with polymyositis/dermatomyositis in the same
community (10). The medical records from all 35 hospitals in
Allegheny County were searched for patients with a discharge
diagnosis of SSc (International Classification of Diseases codes
710.1 [seventh revision], 734.0 [eighth revision], and 701.0 and
710.1 [ninth revision]) during the period 1963-1982. All inpatient charts showing these diagnostic codes were then individually abstracted by a trained research assistant. The following
data were obtained: 1) demographic data on age at first
hospital diagnosis, sex, race, and location of residence; 2) date
of disease onset (first symptom), first diagnosis by any physician, and first Allegheny County hospital diagnosis; and 3)
clinical information confirming scleroderma, including extent
of skin thickening, presence of Raynaud’s phenomenon, digital
pitting scars, esophageal dysmotility, pulmonary fibrosis, renal
crisis, and pulmonary arterial hypertension.
A case was included in the study if the diagnosis of
scleroderma was accepted after physician (VDS) review of the
abstracted data and if the patient was a resident of Allegheny
County at the time of hospital diagnosis. The diagnosis of SSc
was accepted if the patient satisfied the American College of
Rheumatology (ACR) criteria for classification of definite
scleroderma (11) or if the patient had Raynaud’s phenomenon,
sclerodactyly, telangiectasias, or calcinosis and 1 organ involvement characteristic of SSc, including esophageal hypomotility,
small bowel hypomotility, pulmonary arterial hypertension, or
scleroderma renal crisis. In the circumstance of overlap with
other connective tissue diseases, a case was accepted only if
scleroderma was the primary diagnosis.
Population denominators. The setting for this survey
was Allegheny County, PA, including the city of Pittsburgh.
The total county population was relatively stable, ranging from
1.6 million (1960) to 1.3 million (1980) residents over the study
period. The population was racially heterogeneous, with 9%
nonwhite at the time of the 1970 US Census. The 20-year study
was divided into four 5-year time intervals: 1963-1967, 19681972, 1973-1 977, and 1978-1982. Denominator population
data for Allegheny County were derived from the US Census:
the 1970 and 1980 US Censuses were used for the second and
fourth time intervals, respectively; the mean of the 1960 and
1970 US Censuses was used for the first time interval; and the
mean of the 1970 and 1980 US Censuses was used for the third
time interval.
Calculation of incidence rates by age, race, and sex.
Direct age-adjusted incidence rates of hospital-diagnosed SSc
were computed for each race, sex, and disease classification
subgroup, and for each defined time interval. Incidence rates
were expressed as new cases per million population per year.
The 1970 Allegheny County total population was used as the
standard. Age-specific rates for each race and sex subgroup
were determined for the 20-year duration of the study, using
Allegheny County cases and population data. The 1960 US
Census data, used to calculate incidence rates in the first time
interval, did not include a breakdown of the non-white group
into blacks and other races. Therefore, we used the entire
nonwhite denominator for calculation of black female and
black male incidence rates, which resulted in slight underesti-
mation of incidence in the black population for the 1963-1968
interval only.
Statistical analysis. Differences between observed and
expected frequencies were tested using the chi-square method,
with Yates’ correction or Fisher’s exact test where appropriate.
Differences between incidence rates in these large populations
were tested by application of the binomial model to crude
person-time data (12).
Classification of cases. We identified 444 patients with SSc who were residents of Allegheny County
and who had a verified hospital diagnosis of SSc during
1963-1982. Ninety-one percent of these patients fulfilled
the ACR criteria for SSc (ll),by either major (n = 307)
or minor (n = 98) criteria. The 39 patients who did not
fulfill the ACR criteria all had Raynaud’s phenomenon,
sclerodactyly, and either telangiectasias or calcinosis. In
addition, 29 patients had esophageal dysmotility (2 of
whom also had small intestinal dysmotility), 4 had
pulmonary hypertension, 3 had renal crisis, and 3 had
pulmonary interstitial fibrosis.
Distribution of cases by age, race, and sex. The
race and sex distributions of all cases of SSc, according
to the time of initial diagnosis in an Allegheny County
hospital, are shown in Table 1. The number of cases
diagnosed dramatically increased with time, particularly
in women. Sixty-five percent of all patients were identified during the second period of the study.
Age-adjusted, age-specific incidence rates (Table
1). The overall age-adjusted annual incidence was 13.9
cases per million population for the 20-year study period. The incidence of 18.7 per million for the study
interval 1973-1982 (9.5% confidence interval [9.5% CI]
16.5-20.9) was 2-fold greater than the 9.6 per million for
the prior interval 1963-1972 (95% CI 8.1-11.1; P <
0.001). Incidence rates increased in both women and
men during the 20-year study period (Table 2). The
incidence in men increased from 5.5 (9.5% CI 3.8-7.1) to
8.8 cases per million (9.5% CI 6.6-11.0; P = 0.02). In
comparison, women showed a more impressive increase
in incidence, from 13.3 (9.5% CI 10.9-1.5.6) to 27.6 cases
per million (9.5% CI 23.9-31.3; P < 0.001).
Conversely, the black male population showed no
increase in incidence rates in the second interval of the
study; incidence declined from 12.2 (9.5% CI 6.6-18.5) to
8.8 cases per million (9.5% CI 6.2-17.3; P > 0.1, not
signficant [NS]). The white male population showed only
a slight (NS) increase, from 5.0 (95% CI 3.3-6.6) to 8.8
(9.5% CI 6511.2) cases per million population. The
incidence of SSc was highest in women in the second
Table 1. Number of cases and average annual age-adjusted incidence of systemic sclerosis by race, sex, and time interval in Allegheny County, PA,
Period of
diagnosis Number
* CI
All patients
(95% CI)
(95% CI)
(95% CI)
(95% CI)
(95% CI)
5.3 (2.9-7.7)
4.6 (2.4-6.8)
9.4 (6.1-12.7)
8.3 (5.2-11.4)
6.8 (5.4-8.2)
14.4 (10.5-18.3)
11.7 (8.4-15.0)
27.7 (22.4-33.0)
27.2 (21.9-32.5)
19.9 (17.7-22.0)
20.9 (4.6-37.0)
11.4 (1.0-21.8)
6.4 (0-15.4)
9.6 (4.5-14.7)
11.5 (0-23.7)
20.8 (6.7-34.9)
29.8 (12.9-46.0)
27.3 (11.0-43.6)
22.6 (15.0-30.2)
9.7 (7.5-11.8)
9.5 (7.3-11.6)
19.2 (16.1-22.3)
18.2 (15.0-21.3)
confidence interval.
study interval. The incidence in white women increased
significantly, from 13.0 (95% CI 10.5-15.6) to 27.5 cases
per million (95% CI 23.6-31.3) annually during the
second time interval ( P < 0.0001). Incidence rates also
increased in black women, but the difference did not
reach statistical significance. The increase in incidence
was variable across the age ranges. There was no particular trend for any one age group.
Age-specific incidence rates revealed a progressive increase corresponding with age, with the highest
rates in black women between the ages of 45 and 54 and
in a slightly older (ages 55-64) group of white women
(Figure 1). Cases were infrequent in children below age
15. In the 15-24-year age group, there were also very few
patients, except for black women, in whom the incidence
rate was very high. Black women in this age group
showed an incidence of 21.2 (95% CI 4.0-38.3) cases per
million population. In contrast, 15-24-year-old white
women had an annual incidence of only 1.9 (95% CI
0.1-3.6) cases per million population (P < 0.0001). The
incidence in young white men was also low, at 1.8 per
million, and there were no cases among black men.
Thus, it is no surprise that this 15-24-year age group had
a very dramatic incidence ratio of 11.4:l between blacks
and whites (P < 0.0001) (Figure 1).
In the 35-44-year age group, there was an increased incidence among black men (25.8 per million
population, 95% CI 6.9-44.6) compared with white men
(10.1 per million population, 95% CI 6.7-13.6) (P <
0.05). The race incidence ratio in this age group was
3.1:1, compared with <2.0:1 for the other age groups
(P < 0.05). Compared with the other groups, only white
men did not have a marked decrease in incidence after
age 65.
In white patients, the peak female-to-male incidence ratio was more than 4:1, observed in the 35-44year age group. The overall female-to-male ratio during
the childbearing years (ages 15-44) was 3.4:1, and
during the postmenopausal years (age 45+), was 2.4:l.
Table 2. Comparison of group incidence rates of systemic sclerosis
during the study period*
Group incidence rate (95% CI)
Black men
White men
Black women
White women
5.46 (3.8-7.1)
13.32 (10.9-15.6)
12.2 (6.6-18.5)
5.0 (3.3-6.6)
16.3 (7.1-25.4)
13.0 (10.5-15.6)
8.8 (6.6-11.0)
27.6 (23.9-31.3)
8.8 (6.2-17.3)
8.8 (6.5-11.2)
28.5 (16.8-40.2)
27.5 (23.6-31.3)
* CI
confidence interval; NS
not significant.
Figure 1. Age-specific incidence of systemic sclerosis in Pittsburgh
and Allegheny County, PA, 1963-1982.
In black patients, the peak female-to-male ratio was
observed in the 15-24-year age group.
20 1
This study estimated the age-adjusted annual
incidence of hospital-diagnosed SSc in Pittsburgh and
Allegheny County, PA, for a 20-year period, 1963-1982.
The predominantly white population (91%) had an
overall annual incidence of 13.9 newly diagnosed cases
per million population, with a significant increase from
the first period of study (9.6 per million) to the second
(18.7 per million). The present study used the largest
demographically representative US population yet studied
to determine the incidence of hospital-diagnosed SSc.
A primary concern is whether or not a hospitalbased study is acceptable to determine the incidence of
SSc. Prior to the 1980s, during the period when this
study was done, patients with this complicated and often
progressive multisystem disease were almost universally
hospitalized at some point in the natural history of their
illness. Today, however, given the current restrictions on
hospitalization, hospital surveys alone will tend to underestimate disease occurrence.
There are other reasons for considering both
underdiagnosis and overdiagnosis in this type of hospital
survey. Regarding underdiagnosis, patients with limited
scleroderma skin changes (sclerodactyly only) or no skin
thickening (scleroderma sine scleroderma) often do not
have the diagnosis of scleroderma recognized, but are
hospitalized by specialists for their scleroderma-related
problems, such as reflux esophagitis, small bowel malabsorption, or pulmonary interstitial fibrosis. In our
University of Pittsburgh Medical Center-based series,
316 (97%) of 325 consecutive scleroderma patients with
limited cutaneous involvement who were evaluated during a similar time period, 1972-1982, had a hospital
diagnosis. Thus, we believe that very few cases were lost
due to incomplete ascertainment.
With regard to overascertainment, we recognize
that Berkson’s bias may occur in such hospital surveys
(12). However, we found only 2 patients who were
excluded because scleroderma was not the primary
diagnosis. An additional 18 patients were diagnosed with
scleroderma but did not fulfill the established criteria,
and thus were also excluded. Furthermore, in our simultaneous survey of patients with hospital-diagnosed polymyositis (lo), all cases of “overlap” between scleroderma and polymyositis were given either 1 or the other
diagnosis, and thus were represented in only 1 of the 2
198872 187577 1870-02
Figure 2. Overall incidence of systemic sclerosis in Shelby County,
TN (1947-1967) and Allegheny County, PA (1963-1982).
The incidence of SSc is higher in this study than
that reported in previous surveys (1-6). Even the most
recent population study of scleroderma in South Carolina found a low incidence rate of pure scleroderma,
although there was a much higher prevalence of
scleroderma-spectrum disorders (10). When we compared the incidence in Shelby County, TN with that in
Allegheny County, PA during the period 1963-1967, in
which the same physician (TAM) screened all cases, the
Allegheny County annual incidence was double that in
Shelby County (9.7 versus 4.5 new cases per million per
year) (Figure 2). This variation might be attributable to
differences in population genetics or demographic factors or to environmental exposures, but the most likely
explanation is increased physician awareness of this
disease in Pittsburgh because of the great interest and
local educational efforts of the late Dr. Gerald Rodnan.
Also demonstrated in Figure 2 is the marked
increase in incidence during 1973-1982 compared with
the period 1963-1972 in Allegheny County (P < 0.0001).
Although this difference may represent increased physician awareness of the disease, it could also indicate a
true increase in incidence over time. The increase in
incidence in all male patients combined was slight (from
5.5 to 8.8 cases per million), and in black men, incidence
actually decreased, from 12.2 to 8.8 cases per million per
year. Differences in ascertainment by race are recognized by the US Census Bureau (1% for whites versus
5% for blacks) in the 1970 and 1980 Censuses (Robinson
JG: personal communication). Nevertheless, this small
difference in the denominators did not affect the significance of our results regarding racial differences in
incidence of SSc.
The age-specific rates in the present study confirmed a previously reported peak annual incidence at
ages 45-64 that declined in the 65+ age group. Incidence rates under age 34 are quite low, with the
exception of black women, who had a 20-fold higher
annual incidence than white women in the 15-24-year
age group, and a 2-fold higher incidence in the 25-34year age group. Similar findings of increased incidence
among young black women have been observed in
studies of other connective tissue diseases, e.g., in systemic lupus erythematosus (14) and in polymyositis (10).
The female-to-male incidence ratio during the
childbearing years (ages 15-44) was 3.4:1, and during
the postmenopausal years, it was 2.4:l. This ratio in the
childbearing years was lower than that reported in
previous studies. In the Memphis survey, the female-tomale ratio was 15:l (l), and Barnett et a1 found a 5.1:l
ratio (15) between female and male incidence rates. The
explanation for an excess incidence in women is unknown, but hormonal factors are believed to play a role.
It is not clear why our female-to-male incidence ratio
was so much lower than that observed in other studies.
In summary, this community-based survey found
the overall incidence of hospital-diagnosed SSc to be
13.9 cases per million population annually. Although
scleroderma is an uncommon disease, its incidence
appears to be increasing. Increased incidence among
blacks and women, especially during the childbearing
years, suggests that genetic and hormonal factors are
important in the expression of this disease.
1. Masi AT: Classification of systemic sclerosis. In, Current Topics in
Rheumatology: Systemic Sclerosis (Scleroderma). Edited by CM
Black, AR Myers. New York, Gower Medical Publishing, 1985
2. Hochberg MC, Lopez-Acuna D, Gittelsohn AM: Mortality from
systemic sclerosis (scleroderma) in the United States, 1969 -1977.
In, Current Topics in Rheumatology: Systemic Sclerosis (Scleroderma). Edited by CM Black, AR Myers. New York, Gower
Medical Publishing, 1985
3. Lawrence RC, Hochberg DJ, McDuffie FC, Medsger TA Jr, Felts
WR, Schulman LE: Estimates of the prevalence of selected
arthritis and musculoskeletal disease in the United States. J Rheumatol 16:427-441, 1989
4. Medsger TA Jr, Masi AT: Epidemiology of systemic sclerosis
(scleroderma). Ann Intern Med 74:714-721, 1971
5. Michet CJ Jr, McKenna CH, Elveback LR, Kaslow RA, Kurland
LT: Epidemiology of systemic lupus erythernatosus and other
connective tissue diseases in Rochester, Minnesota, 1950 through
1979. Mayo Clin Proc 60:lOS-123, 1985
6. Medsger TA Jr, Masi AT: Epidemiology of systemic sclerosis
(scleroderma) among male United States veterans. J Chronic Dis
31~73-85, 1978
7. Bosmansky K, Zitnan D, Urbanek T, Svec U: Incidence of diffuse
disorders of the connective tissue with special reference to systemic lupus erythematosus in a selected district in the years
1961-1969. Fysiatr Rheumatol Vestn 49:267-272, 1971
8. Wigley RD, Borman B: Medical geography and the aetiology of
the rare connective tissue disease in New Zealand. SOCSci Med
14D:175-183, 1980
9. Silman A, Jannini S, Symmons D, Bacon P: An epidemiological
study of scleroderma in the West Midlands. Br J Rheumatol
27:286-290, 1988
10. Oddis CV, Conte CG, Steen VD, Medsger TA Jr: Incidence of
polymyositis-dermatomyositis:a 20-year study of hospital diagnosed cases in Allegheny County, PA 1963-1982. J Rheumatol
17:1329-1334, 1990
11. Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee: Preliminary criteria for the classification of systemic sclerosis
(scleroderma). Arthritis Rheum 23:581-590, 1980
12. Rothman KJ: Modern Epidemiology. Boston, Little, Brown, and
Co., 1986
13. Maricq HR, Weinrich MC, Keil JE, Smith EA, Harper FE,
Nussbaum AI, LeRoy EC, McGregor AR, Diat F, Rosa1 EJ:
Prevalence of scleroderma spectrum disorders in the general
population of South Carolina. Arthritis Rheum 32:998-1006, 1989
14. Masi AT, Kaslow RA: Sex effects in systemic lupus erythematosus:
a clue to pathogenesis. Arthritis Rheum 21:480-484, 1978
15. Barnett AJ, Miller MH, Littlejohn GO: A survival study of patients
with scleroderma diagnosed over 30 years (1953-1983): the value
of a simple cutaneous classification in the early stages of the
disease. J Rheumatol 15:276-283, 1988
Без категории
Размер файла
540 Кб
years, incidence, diagnosed, systemic, count, hospital, pennsylvania, stud, case, twenty, sclerosis, 19631982, allegheny
Пожаловаться на содержимое документа