Multiple system atrophy Clinical Autonomic Research 1, °,13 (1991)
Research Paper WE studied 60 patients with multiple system atrophy and autonomic failure and 60 control subjects matched for age, sex and race. Their psychosocial history, pedigree and occupation were obtained by personal interview. An inventory of autonomic and neurologic symptoms was obtained from 148 first-degree relatives of the patients and 80 controls by a self-administered questionnaire. Patients with multiple system atrophy had significantly more potential exposures to metal dusts and fumes, plastic monomers and additives, organic solvents, and pesticides than the control population. The potential exposures were determined in most subjects by their reported usual occupation. Clinical symptoms of multiple system atrophy were reported by a significantly larger group of patients' relatives than controls. These findings are possibly consistent with the hypothesis that multiple system atrophy develops as a result of a genetically determined selective vulnerability in the nervous system. Specific neuronal systems may become targets for environmental insults or toxins, and the disease state may occur when ageing neuronal systems can no longer sustain functional capacity. This preliminary study supports the need to further explore possible environmental, occupational, and familial contributions to the aetiology of multiple system atrophy. Key words: Multiple system atrophy, Shy-Drager syndrome, Environmental exposure, Occupational exposure, Familial disease
Environmental-
occupational risk factors and familial associations in multiple system
atrophy: a preliminary investigation Linda E. Nee 1'cA, M S W , Manuel R. Gomez 2, MS, James Dambrosia z, PhD, Sherri Bale, 4 PhD, Roswell Eldridge, M D and Ronald J. Polinsky, M D 4Family Studies Section, Environmental Epidemiology Branch, National Cancer Institute; =Mathematical Statistics Section, and 5Clinical Neurogenetics Studies Unit, Neuroepidemiology Branch, National Institute of Neurological Disorders and Stroke; 2Occupational Studies Section, Environmental Epidemiological Branch, National Cancer Institute; ~Clinical Neuropharmacology Section, Clinical Neuroscience Branch, National Institute of Neurological Disorders and Stroke; National Institutes of Health, Bethesda, MD 20892, USA. CACorresponding Author
Introduction Multiple system atrophy (MSA) with autonomic failure, also called Shy-Drager syndrome, is a progressive disorder of the central and autonomic nervous systems, ending in death 7-10 years after the onset of symptoms. The nosology of autonomic failure has been in flux for many years, but the term MSA is now used to refer to the overlapping degenerative disorders that include olivopontocerebellar atrophy and striatonigral degeneration. ~ MSA usually begins in the sixth decade of life and affects men more often than women. ~ Previous research efforts to determine the cause of MSA have yielded few clues. Reduced survival of cultured lymphoblasts from patients with MSA was observed after the cells were exposed to X-rays. 2 An association between HLA-Aw32 and MSA was found in a small group of patients, but this finding was not substantiated in a larger population. 3 Neuropathologic studies reveal neuronal loss and gliosis in numerous areas of the central nervous system. I These changes are nonspecific, however, and provide little information about the basic underlying abnormality in MSA. Similarly, abnormal function of neurotransmitters and neuropep© Rapid Communicationsof Oxford Ltd.
tides in MSA offers potential avenues for therapeutic intervention at the symptomatic level, but gives no indication of a specific aetiologic mechanism. 4 Despite the report of familial cases, s MSA has not been the focus of systematic epidemiologic investigation. We report a detailed analysis of environmental, occupational and familial factors in a large series of patients with MSA and a matched control population.
Materials and Methods MSA patients: We studied 60 patients with MSA from 1978 to 1988. They learned about the MSA research programme of the Clinical Neuropharmacology Section, National Institute of Neurological Disorders and Stroke, from health consumer newsletters or their private physicians. The diagnosis of MSA was confirmed after admission to the study, according to previously established criteria. 4 Fifteen patients who had been referred were excluded from the study because they could not provide extensive medical, family or occupational data. Clinical Autonomic Research.vol 1.1991
9
L. Nee et al. Table 1. Demographic characteristics of study subjects Characteristics
Patients (n = 60)
Controls (n = 60)
Sex and Race Male Black White Female Black White
3 35
3 35
1 21
1 21
48-80 years (mean, 64 years) 40-75 years (mean, 61 years)
40-75 years (mean, 63 years) 39-80 years (mean, 62 years)
Religion Protestant Catholic Jewish Other
36 12 10 2
42 12 5 1
Ancestry At least half Eastern European At least half Western European Western and Eastern European African
27 22 7 4
19 26 11 4
Education Junior high-*some high school High school graduate Some college ~ college graduate Graduate school
8 21 20 11
3 10 25 22
Occupation Manual job White collar job
20 40
13 47
Age Male Female
Sixty healthy control subjects were recruited to match the patients in age, sex and race. They consisted primarily of the patients' spouses, friends and normal volunteers. None of the controls is a known blood relative of a patient with MSA or is related to another control subject. The patients and controls were interviewed to obtain their psychosocial and occupational histories and pedigrees. They all were from the US and Canada; their other demographic characteristics are shown in Table 1. Although controls had a significantly (p < 0.01) higher level of education, their occupational status (white collar vs. manual) did not differ from the patient group.
First-degree relatives: We studied the symptoms of MSA in 148 first-degree relatives (parents, offspring and siblings) of 33 patients; 80 healthy control subjects were selected randomly and primarily from the patients' spouses and friends. Although all patients were asked to participate, a subgroup of 33 patients volunteered to participate in the study of first-degree relatives because they had living family members thought to be interested enough to return a questionnaire. The questionnaire, consisting of 46 questions about autonomic and neurologic symptoms, was delivered to first-degree relatives by the patients. The relatives (148 out of 193 in the 33 10
Clinical Autonomic Research.vol 1 -1991
Table 2. Sex and age of first-degree relatives Sex
Male < age 50 _> age 50 Female < age 50 >_ age 50
Relatives ( n = 148) n (%)
Controls (n = 80) n (%)
years years
42 (28%) 35 (24%)
16 (20%) 12 (15%)
years years
37 (25%) 34 (23%)
28 (35%) 24 (30%)
families, 77% response rate) completed the questionnaires and returned them to us in the preaddressed envelopes provided. All control subjects completed the same questionnaire. Table 2 shows the age and sex of the relatives and controls.
Environmental-occupational risk factors: The subjects' usual occupations were examined for potential exposure to six broad categories of common workplace contaminants, including numerous agents reportedly toxic to the nervous system/'-14 The categories were organic solvents (OS), metal dusts and fumes (MDF), organic dusts (OD), inorganic dusts (ID), pesticides (PEST), and plastic monomers and additives (PMA). The small number
Multiple system atrophy of subjects precluded investigation of any single substance. All the usual occupations reported by the subjects were evaluated by an industrial hygienist, who did not know the subjects' case~control status. As has been done in other studies, each occupation was classified as potentially exposed or unexposed to each of the six groups of agents. 15 These assignments were based on the limited information available, which was often restricted to job title. The assessments for all exposures except pesticides were based on occupational information. Most of the assessments of pesticide exposure were based on responses to open-ended questions from the interviewer about previous exposure to toxic agents and reflected self-reported home use of pesticides (e.g. in gardening) or indirect farm exposures (e.g. spraying of nearby fields).
Informed consent: All subjects gave their written informed consent for the study. The protocol was approved by the Institute's clinical investigational review committee. Statistical analysis: Each risk factor was examined individually by computing the odds ratios and corresponding chi-squared test. Logistic regression was a step-down variable selection procedure was used to evaluate the joint effect of the industrial exposures, job type, and age at onset. All second order interactions between exposures were tested using a score statistic. 1~19 The joint effects of exposures and family history of neurologic disease on MSA were also explored by logistic regression. An explanatory variable for family history of neurologic disease was defined as '1' if neurologic disease was present in the family and as '0' otherwise.
Results Environmental-occupational risk factors: Patients with MSA had significantly more potential exposures to metal dusts and fumes, plastic monomers and additives, organic solvents, and pesticides than controls (Table 3). Neither age at onset of MSA nor type of job was related to exposure to the various substances under consideration. The distribution by type of MSA (i.e. cerebellar, parkinsonian, or mixed) 4 were similar for both early and late onset of the disorder (_< 50 or > 50 years). Patients with early or late onset of MSA were equally likely to have had a manual job. Affected men had many more identified exposures than affected women. Industrial exposures retained in the logistic regression model, after step-down variable selection and tests for second order interactions, were MDF, PEST and PMA. Organic solvent
Table 3. Environmental-occupational risk factors Exposure
Organic solvents Metal dusts and fumes Organic dusts Inorganic dusts Pesticides Plastic monomers and additives
Patients (n = 60)
Controls (n = 60)
Odds ratio
24 12 11 7 10
13 1 4 1 2
2.41 * 14.75* 3.14 7.79 5.80*
13
3
5.25*
* Denotes that the associated 95% confidence interval for the odds ratio does not contain 1.
which was significant individually did not contribute additional discriminatory information in the presence of other exposures. The resulting regression coeft:icients and their standardized values are given in Table 4.
Joint effects of exposures and familial neurologic disease: Logistic regression including both industrial exposures and family history of neurological disease provides an estimate of increased relative risk for family history of 4.22 over the relative risks of the three exposures alone in the initial logistic regression. The resulting coefficients and their standardized values for the model containing both exposures and family history of neurological disease are given in Table 4.
Familial association: Twenty-three per cent of the patients had a first-degree relative with a neurologic disorder compared with 10% 0f the controls. Sixty-four per cent of the positive case families had a parent with a neurologic disorder compared with 16% of the control families. Men outnumbered women in the M S A population by 1.7:1 (63% men and 37% women), but women represented 52% of the patients with an affected first-degree relative. Two of the women with onset of MSA during their 30s and 40s had fathers with amyotrophic lateral sclerosis beginning in their 60s and 70s, two had a mother or brother with Alzheimer's disease,
Table 4. Logistic regression model coefficients Exposures
INTERCEPT M PD PEST PMA Family History
/Exposures and family history -0.5073 ( - 2 . 3 0 2 ) * 2.329 (2.134) 2.024 (2.499) 1.346 (1.866) --
-0.8052 ( - 3 . 1 3 9 ) 2.379 (2.159) 2.329 (2.723) 1.624 (2.218) 1.440 (2.562)
* Quantities in parentheses are the standardized value of the regression coefficients, i.e. the estimated regression coefficient divided by its standard deviation. Their asymptotic distributions are normal with mean 0 and variance 1. Clinical Autonomic Research. vol 1 -1991
11
L. Nee et al.
two had a mother or father with parkinsonism, one had a brother die of postpolio syndrome, and one had a mother and three brothers with neurofibromatosis (NF2)--she also had paralytic polio in childhood. T w o of the men with MSA had fathers and brothers with parkinsonism or tremors, one had a mother with head tremor, one had a fraternal twin with postpolio motor neuron disease, one had a brother with a history of chorea at eight years of age, and one had a brother with epilepsy. Among the controls, two women had a brother or sister with multiple sclerosis and one had a mother with senile head and hand tremor; three men had brothers or a sister with polio. We were not aware of any family in which there was more than one member with MSA. Data were inadequate to determine if there was an aggregation of exposures in the families. The first-degree relatives of MSA patients had a significantly higher incidence of loss of consciousness, staggering or unsteadiness, ringing in the ears, speech changes, trouble swallowing, and decline in mental capabilities than controls (Table 5). Because the controls were mostly patients' spouses and friends, there was the possibility that their familiarity with MSA symptoms could bias their responses. This potential bias was explored by giving the questionnaire to relatives of patients with pure autonomic failure, using the same method by which relatives of MSA patients received their questionnaires. Patients with pure autonomic failure share many common symptoms with MSA patients. Responses of the first-degree relatives of patients with pure autonomic failure (n = 55) did not differ from those of the controls used in our MSA comparisons. Furthermore, the spouse-caregivers (n = 14), who had the greatest awareness of the patients' symptoms, generally had fewer positive responses than other controls.
patients, information derived from both pedigree and questionnaire data, suggests a genetic predisposition for the disorder. Further, MSA patients were more likely than controls to have potentially experienced occupational exposure to metal dusts and fumes, plastic monomers and additives, organic solvents, and pesticides. These findings indicate the need for further study to determine if the development of MSA depends on a genetically determined neuronal vulnerability, signified by the symptoms in the first-degree relatives, which creates a target for environmental--occupational toxic insult, resulting in disease when the ageing neurologic system, with generalized neuronal fall-out, 2° can no longer maintain a functional threshold. We know of no other study implicating both familial and environmental-occupational factors in MSA. The male preponderance in MSA could be consistent with an occupational bias. Although females were outnumbered in the MSA population by 1.7:1, they were underrepresented in the environmental-occupational exposure group and overrepresented in the group with a positive family history. This suggests a possible polygenic mechanism of disease with greater disease liability in females. Toxic effects on the nervous system of many workplace contaminants are well documented. 6~14 The overlapping of exposures (Table 6), the attention and knowledge required to document exposures, and the limited number of subjects precluded more specific identification of the exposures. Exposure to environmental toxins could contribute to the decreased D N A repair found by
Discussion
Total no.
The higher incidence of neurologic symptoms and disease among first-degree relatives of MSA Table 5. Self-reported neurologic symptoms in first-degree relatives Symptom
Trouble swallowing Mental deterioration Tinnitus > age 50 years Loss of consciousness Females < 5 0 years Speech changes Staggering or unsteadiness 12
Table 6. Coexisting exposures Cases
OS MDF OD ID PEST PMA
Relatives (n = 148)
Controls (n = 80)
p
15 14
1 1
Research Paper WE studied 60 patients with multiple system atrophy and autonomic failure and 60 control subjects matched for age, sex and race. Their psychosocial history, pedigree and occupation were obtained by personal interview. An inventory of autonomic and neurologic symptoms was obtained from 148 first-degree relatives of the patients and 80 controls by a self-administered questionnaire. Patients with multiple system atrophy had significantly more potential exposures to metal dusts and fumes, plastic monomers and additives, organic solvents, and pesticides than the control population. The potential exposures were determined in most subjects by their reported usual occupation. Clinical symptoms of multiple system atrophy were reported by a significantly larger group of patients' relatives than controls. These findings are possibly consistent with the hypothesis that multiple system atrophy develops as a result of a genetically determined selective vulnerability in the nervous system. Specific neuronal systems may become targets for environmental insults or toxins, and the disease state may occur when ageing neuronal systems can no longer sustain functional capacity. This preliminary study supports the need to further explore possible environmental, occupational, and familial contributions to the aetiology of multiple system atrophy. Key words: Multiple system atrophy, Shy-Drager syndrome, Environmental exposure, Occupational exposure, Familial disease
Environmental-
occupational risk factors and familial associations in multiple system
atrophy: a preliminary investigation Linda E. Nee 1'cA, M S W , Manuel R. Gomez 2, MS, James Dambrosia z, PhD, Sherri Bale, 4 PhD, Roswell Eldridge, M D and Ronald J. Polinsky, M D 4Family Studies Section, Environmental Epidemiology Branch, National Cancer Institute; =Mathematical Statistics Section, and 5Clinical Neurogenetics Studies Unit, Neuroepidemiology Branch, National Institute of Neurological Disorders and Stroke; 2Occupational Studies Section, Environmental Epidemiological Branch, National Cancer Institute; ~Clinical Neuropharmacology Section, Clinical Neuroscience Branch, National Institute of Neurological Disorders and Stroke; National Institutes of Health, Bethesda, MD 20892, USA. CACorresponding Author
Introduction Multiple system atrophy (MSA) with autonomic failure, also called Shy-Drager syndrome, is a progressive disorder of the central and autonomic nervous systems, ending in death 7-10 years after the onset of symptoms. The nosology of autonomic failure has been in flux for many years, but the term MSA is now used to refer to the overlapping degenerative disorders that include olivopontocerebellar atrophy and striatonigral degeneration. ~ MSA usually begins in the sixth decade of life and affects men more often than women. ~ Previous research efforts to determine the cause of MSA have yielded few clues. Reduced survival of cultured lymphoblasts from patients with MSA was observed after the cells were exposed to X-rays. 2 An association between HLA-Aw32 and MSA was found in a small group of patients, but this finding was not substantiated in a larger population. 3 Neuropathologic studies reveal neuronal loss and gliosis in numerous areas of the central nervous system. I These changes are nonspecific, however, and provide little information about the basic underlying abnormality in MSA. Similarly, abnormal function of neurotransmitters and neuropep© Rapid Communicationsof Oxford Ltd.
tides in MSA offers potential avenues for therapeutic intervention at the symptomatic level, but gives no indication of a specific aetiologic mechanism. 4 Despite the report of familial cases, s MSA has not been the focus of systematic epidemiologic investigation. We report a detailed analysis of environmental, occupational and familial factors in a large series of patients with MSA and a matched control population.
Materials and Methods MSA patients: We studied 60 patients with MSA from 1978 to 1988. They learned about the MSA research programme of the Clinical Neuropharmacology Section, National Institute of Neurological Disorders and Stroke, from health consumer newsletters or their private physicians. The diagnosis of MSA was confirmed after admission to the study, according to previously established criteria. 4 Fifteen patients who had been referred were excluded from the study because they could not provide extensive medical, family or occupational data. Clinical Autonomic Research.vol 1.1991
9
L. Nee et al. Table 1. Demographic characteristics of study subjects Characteristics
Patients (n = 60)
Controls (n = 60)
Sex and Race Male Black White Female Black White
3 35
3 35
1 21
1 21
48-80 years (mean, 64 years) 40-75 years (mean, 61 years)
40-75 years (mean, 63 years) 39-80 years (mean, 62 years)
Religion Protestant Catholic Jewish Other
36 12 10 2
42 12 5 1
Ancestry At least half Eastern European At least half Western European Western and Eastern European African
27 22 7 4
19 26 11 4
Education Junior high-*some high school High school graduate Some college ~ college graduate Graduate school
8 21 20 11
3 10 25 22
Occupation Manual job White collar job
20 40
13 47
Age Male Female
Sixty healthy control subjects were recruited to match the patients in age, sex and race. They consisted primarily of the patients' spouses, friends and normal volunteers. None of the controls is a known blood relative of a patient with MSA or is related to another control subject. The patients and controls were interviewed to obtain their psychosocial and occupational histories and pedigrees. They all were from the US and Canada; their other demographic characteristics are shown in Table 1. Although controls had a significantly (p < 0.01) higher level of education, their occupational status (white collar vs. manual) did not differ from the patient group.
First-degree relatives: We studied the symptoms of MSA in 148 first-degree relatives (parents, offspring and siblings) of 33 patients; 80 healthy control subjects were selected randomly and primarily from the patients' spouses and friends. Although all patients were asked to participate, a subgroup of 33 patients volunteered to participate in the study of first-degree relatives because they had living family members thought to be interested enough to return a questionnaire. The questionnaire, consisting of 46 questions about autonomic and neurologic symptoms, was delivered to first-degree relatives by the patients. The relatives (148 out of 193 in the 33 10
Clinical Autonomic Research.vol 1 -1991
Table 2. Sex and age of first-degree relatives Sex
Male < age 50 _> age 50 Female < age 50 >_ age 50
Relatives ( n = 148) n (%)
Controls (n = 80) n (%)
years years
42 (28%) 35 (24%)
16 (20%) 12 (15%)
years years
37 (25%) 34 (23%)
28 (35%) 24 (30%)
families, 77% response rate) completed the questionnaires and returned them to us in the preaddressed envelopes provided. All control subjects completed the same questionnaire. Table 2 shows the age and sex of the relatives and controls.
Environmental-occupational risk factors: The subjects' usual occupations were examined for potential exposure to six broad categories of common workplace contaminants, including numerous agents reportedly toxic to the nervous system/'-14 The categories were organic solvents (OS), metal dusts and fumes (MDF), organic dusts (OD), inorganic dusts (ID), pesticides (PEST), and plastic monomers and additives (PMA). The small number
Multiple system atrophy of subjects precluded investigation of any single substance. All the usual occupations reported by the subjects were evaluated by an industrial hygienist, who did not know the subjects' case~control status. As has been done in other studies, each occupation was classified as potentially exposed or unexposed to each of the six groups of agents. 15 These assignments were based on the limited information available, which was often restricted to job title. The assessments for all exposures except pesticides were based on occupational information. Most of the assessments of pesticide exposure were based on responses to open-ended questions from the interviewer about previous exposure to toxic agents and reflected self-reported home use of pesticides (e.g. in gardening) or indirect farm exposures (e.g. spraying of nearby fields).
Informed consent: All subjects gave their written informed consent for the study. The protocol was approved by the Institute's clinical investigational review committee. Statistical analysis: Each risk factor was examined individually by computing the odds ratios and corresponding chi-squared test. Logistic regression was a step-down variable selection procedure was used to evaluate the joint effect of the industrial exposures, job type, and age at onset. All second order interactions between exposures were tested using a score statistic. 1~19 The joint effects of exposures and family history of neurologic disease on MSA were also explored by logistic regression. An explanatory variable for family history of neurologic disease was defined as '1' if neurologic disease was present in the family and as '0' otherwise.
Results Environmental-occupational risk factors: Patients with MSA had significantly more potential exposures to metal dusts and fumes, plastic monomers and additives, organic solvents, and pesticides than controls (Table 3). Neither age at onset of MSA nor type of job was related to exposure to the various substances under consideration. The distribution by type of MSA (i.e. cerebellar, parkinsonian, or mixed) 4 were similar for both early and late onset of the disorder (_< 50 or > 50 years). Patients with early or late onset of MSA were equally likely to have had a manual job. Affected men had many more identified exposures than affected women. Industrial exposures retained in the logistic regression model, after step-down variable selection and tests for second order interactions, were MDF, PEST and PMA. Organic solvent
Table 3. Environmental-occupational risk factors Exposure
Organic solvents Metal dusts and fumes Organic dusts Inorganic dusts Pesticides Plastic monomers and additives
Patients (n = 60)
Controls (n = 60)
Odds ratio
24 12 11 7 10
13 1 4 1 2
2.41 * 14.75* 3.14 7.79 5.80*
13
3
5.25*
* Denotes that the associated 95% confidence interval for the odds ratio does not contain 1.
which was significant individually did not contribute additional discriminatory information in the presence of other exposures. The resulting regression coeft:icients and their standardized values are given in Table 4.
Joint effects of exposures and familial neurologic disease: Logistic regression including both industrial exposures and family history of neurological disease provides an estimate of increased relative risk for family history of 4.22 over the relative risks of the three exposures alone in the initial logistic regression. The resulting coefficients and their standardized values for the model containing both exposures and family history of neurological disease are given in Table 4.
Familial association: Twenty-three per cent of the patients had a first-degree relative with a neurologic disorder compared with 10% 0f the controls. Sixty-four per cent of the positive case families had a parent with a neurologic disorder compared with 16% of the control families. Men outnumbered women in the M S A population by 1.7:1 (63% men and 37% women), but women represented 52% of the patients with an affected first-degree relative. Two of the women with onset of MSA during their 30s and 40s had fathers with amyotrophic lateral sclerosis beginning in their 60s and 70s, two had a mother or brother with Alzheimer's disease,
Table 4. Logistic regression model coefficients Exposures
INTERCEPT M PD PEST PMA Family History
/Exposures and family history -0.5073 ( - 2 . 3 0 2 ) * 2.329 (2.134) 2.024 (2.499) 1.346 (1.866) --
-0.8052 ( - 3 . 1 3 9 ) 2.379 (2.159) 2.329 (2.723) 1.624 (2.218) 1.440 (2.562)
* Quantities in parentheses are the standardized value of the regression coefficients, i.e. the estimated regression coefficient divided by its standard deviation. Their asymptotic distributions are normal with mean 0 and variance 1. Clinical Autonomic Research. vol 1 -1991
11
L. Nee et al.
two had a mother or father with parkinsonism, one had a brother die of postpolio syndrome, and one had a mother and three brothers with neurofibromatosis (NF2)--she also had paralytic polio in childhood. T w o of the men with MSA had fathers and brothers with parkinsonism or tremors, one had a mother with head tremor, one had a fraternal twin with postpolio motor neuron disease, one had a brother with a history of chorea at eight years of age, and one had a brother with epilepsy. Among the controls, two women had a brother or sister with multiple sclerosis and one had a mother with senile head and hand tremor; three men had brothers or a sister with polio. We were not aware of any family in which there was more than one member with MSA. Data were inadequate to determine if there was an aggregation of exposures in the families. The first-degree relatives of MSA patients had a significantly higher incidence of loss of consciousness, staggering or unsteadiness, ringing in the ears, speech changes, trouble swallowing, and decline in mental capabilities than controls (Table 5). Because the controls were mostly patients' spouses and friends, there was the possibility that their familiarity with MSA symptoms could bias their responses. This potential bias was explored by giving the questionnaire to relatives of patients with pure autonomic failure, using the same method by which relatives of MSA patients received their questionnaires. Patients with pure autonomic failure share many common symptoms with MSA patients. Responses of the first-degree relatives of patients with pure autonomic failure (n = 55) did not differ from those of the controls used in our MSA comparisons. Furthermore, the spouse-caregivers (n = 14), who had the greatest awareness of the patients' symptoms, generally had fewer positive responses than other controls.
patients, information derived from both pedigree and questionnaire data, suggests a genetic predisposition for the disorder. Further, MSA patients were more likely than controls to have potentially experienced occupational exposure to metal dusts and fumes, plastic monomers and additives, organic solvents, and pesticides. These findings indicate the need for further study to determine if the development of MSA depends on a genetically determined neuronal vulnerability, signified by the symptoms in the first-degree relatives, which creates a target for environmental--occupational toxic insult, resulting in disease when the ageing neurologic system, with generalized neuronal fall-out, 2° can no longer maintain a functional threshold. We know of no other study implicating both familial and environmental-occupational factors in MSA. The male preponderance in MSA could be consistent with an occupational bias. Although females were outnumbered in the MSA population by 1.7:1, they were underrepresented in the environmental-occupational exposure group and overrepresented in the group with a positive family history. This suggests a possible polygenic mechanism of disease with greater disease liability in females. Toxic effects on the nervous system of many workplace contaminants are well documented. 6~14 The overlapping of exposures (Table 6), the attention and knowledge required to document exposures, and the limited number of subjects precluded more specific identification of the exposures. Exposure to environmental toxins could contribute to the decreased D N A repair found by
Discussion
Total no.
The higher incidence of neurologic symptoms and disease among first-degree relatives of MSA Table 5. Self-reported neurologic symptoms in first-degree relatives Symptom
Trouble swallowing Mental deterioration Tinnitus > age 50 years Loss of consciousness Females < 5 0 years Speech changes Staggering or unsteadiness 12
Table 6. Coexisting exposures Cases
OS MDF OD ID PEST PMA
Relatives (n = 148)
Controls (n = 80)
p
15 14
1 1
