Neuroepidemiology 5: 29-38 (1986) Amyotropliic Lateral Sclerosis and Occupational Heavy Metal Exposure: A Case-Control Study Louise S. Gresham-a, Craig A. Molgaard-a, Amanda L. Golbeck -b, Richard Smith-c a-Division of Epidemiology and Biostatistics, Graduate School of Public f Icahn, San Diego State University; b-department of Mathematical Sciences, San Diego State University, and c-center for Neurologic Study. San Diego, Calif., USA Key Words. Amyotropilic lateral sclerosis - Occupation - licavy metals Neurotoxicity - Lead - Mercury - Etiology Abstract. A retrospective case-control study of occupational heavy metal exposure was conducted using 66 amyotropliic lateral sclerosis (ALS) patients and 66 -age- and sex- matched controls. Cases were ascertained primarily through a neurology support and research clinic. The self-administered questionnaire probed potential exposure to nine heavy metals: aluminum, lead, lead alkyl, magnesium, manganese, mercury, mercurv alkyl, nickel and selenium. Using McNemar's test and a Mantel Haenszel extended alial),- sis, no association was found between heavy metal exposure and the pathogenesis of ALS in this patient population. Demographic factors, fracture history, ii-nmunizatiotis, travel and other variables were similar in ALS patients and controls. Introduction Amyotrophic lateral sclerosis (ALS) is considered a major enigma ill neurology. Its symptomatology is associated with the selective dat-nage of the upper and lower motor neurons, leading to death within an average of 2-5 years. ALS appears to be distributed rather uniformly around the world, with mortality rates averaging 1.5 per I 00,000 population, incidence imaging mortality at 1-2 per I 00,000, and prevalence at 5-7 per I 00,000. In the course of searching for etiological factors, the role of heavy metals, lead and mercury in particular, has surfaced for discussion. Other fhctors of proposed etiologic significance include aging [ I ], viral infection [2, 3], disease of nerve growth factor [4], deficiency of DNA repair enzymes [5], impaired immune function [61, and fractures and skeletal abnormalities [7, 81. Concern that ALS may be related to a history of increased heavy metal exposure is not unexpected, given that lead and mercury are endemic envi- ronmental neurotoxins. Wilson in 1907 and Brown in 1954 respectively reported chronic lead poisoning and chronic mercurialism in farmers as causing ALS-like symptomology [9]. There is evidence to support [$, 10- 12] and refute [7, 13-15] a hypothesis of heavy metal exposure as a risk factor in ALS. Roelofs-Iverson et al. [16] recently reported a case-control study conducted at the Mayo Clinic where occupational exposure to heavy metals was significantly-higher among cases than controls. The metals were not studied individually and no information was obtained on the duration of exposure. The present study is unique and complementary in that it probes exposures to specific metals and the duration of exposure. Methods and Materials This study was carried out from January 1985 to May 1985. Sampling was done from a case listing developed by the Center for Neurologic Study, a patient support and research facility in San Diego, Calif In addition, letters were' sent to 60 area neurologists detailing the goals of the study, and seeking to increase case identification. Identification of Cases (1) Objective diagnostic criteria: clinical picture consisting of (a) progressive muscular atrophy and weakness, (b) variable signs of fasiculation, (c) pyramidal tract involvement, and (d) no sensory involvement. (2) Eligibility criteria: incident cases diagnosed during 1975-1985. Definition of Controls Friends or neighbors of ALS patients, not a former co-worker, matched I to I case on age (+ 5 years) and sex, and free of known neurologic disease. Data Collection Data collection was by means of a mailed self-administered questionnaire. Sixty ALS cases and 15 surrogates for ALS patients responded; 66 were used after screening for completeness and data quality. There were 33 men, mean age at diagnosis 55.2; and 33 women, mean age at diagnosis 57.7 (fig. 1). There were no twins. Each ALS case was given an additional questionnaire to be completed by a neighbor/friend of similar age. Subjects completing the questionnaire responded to demographic items on race, sex, marital status, and military status. Occupational history of jobs with potential exposure to individual heavy metals (Al, Pb, Pb alkyl, Hg, Hg alkyl, Mg, Mn, Ni, Se), as delimited by the National Institute for Occupational Health and Safety (NIOSH), was examined [17-201. This included 293 distinct occupations where exposure to specific heavy metals was at least a possibility according to NIOSH. Other environmental and medical items included were travel history (Mariana Islands and southern and western Pacific), history of polio, thyroid disorders, fractures, and immunization history, with specific reference to swine flu vacci- nation. Analysis Retaining matching, McNemar's formula was used to test the hypothesis that the odds ratio is unity for potential heavy metal exposure, number of fractures, metabolic deficiencies and history of immunization. Ninety-five percent approximate confidence intervals (Cls) were constructed [2 1 J. For heavy metal analysis, only jobs performed by a subject for more than 6 months were included. The Mantel Haenszel (MH) extended (dose-response) analysis [221, ignoring rpatching, was calculated to test for a significant progressive increase or decrease in the odds ratio with increased duration of occupational exposure. Results Demographics There were no significant differences between the groups on demo- graphic variables of ethnicity (X2 = 1.33, p 0.5 1), history of military ser- vice (X2 = 1.02, p = 0.60), marital status (X2 2.59, p = 0.63), or education (X2 = 4.62, p = 0.20). Table I summarizes the demographic profile of patients and controls. -------------------------------------------------------------------------------------------------------- Table 1. Summary of demographics: ALS cases and controls -------------------------------------------------------------------------------------------------------- ALS cases % Controls % -------------------------------------------------------------------------------------------------------- Mean age, years 59 (11.70) 61.5 (11.40) Age range, years 26-84 31-83 Ethnicity Caucasian 64 97 64 97 Hispanic 1 1.5 2 3 Black 1 1.5 0 - Twins 0 - 0 - Military experience 23 35 22 33 Marital status Married 47 71 49 74 Divorced 8 12.1 5 8 Widowed 6 9.1 9 14 Single 5 7.8 3 3 Education Postgraduate 13 20 1 7 26 College 25 38 20 30 High school 21 32 27 41 Elementary 7 10 2 3 -------------------------------------------------------------------------------------------------------- Antecedent Events Occupational Heavy Metal Exposure. The types of occupations re- ported demonstrated that the assumption of constant occupational expo- sure ratios between males and females was not met. Among women, only 2 cases reported potential occupational heavy metal exposure (table 11). Therefore, in calculation of the matched odds ratio for heavy metal expo- sures, only males were used. No association was found between cases and controls on the variable of heavy metal occupational exposure. The matched analysis is presented in table 111. Since there were so few exposures to the other metals, only lead, mercury and 'all metals' were analyzed. Ninety-f-ive percent Cls are given. To examine any possible dose-response phenomenon, an age-adjusted MH extended analysis was performed, with matching suspended, on months of lead and mercury exposures. It should be noted that some cells contained zero and did not contribute to the denominator in the odds ratios calculated from the MH dose-response test. Zero months exposure was Table 11. Occupational heavy metal exposures: ALS cases and controls, male and female, by type of metal exposure -------------------------------------------------------------------------------------------------------- Type of metal exposure ALS cases % ('ontrols "/o -------------------------------------------------------------------------------------------------------- Males Al 3 9 5 1 5 Mg - 2 6 Mn 1 3 Ni 3 9 Se 1 3 1 3 Pb (+ alkyl) 14 42 1 2 36 Hg (+ alkyl) 7 2 1 6 1 8 -------------------------------------------------------------------------------------------------------- Females Al - 1 3 Mg - 1 3 Mn - - Ni - 3 9 Se - 1 3 Pb 2 6 - Hg - 6 1 8 A single respondent can have more than one type of metal exposure. -------------------------------------------------------------------------------------------------------- -------------------------------------------------------------------------------------------------------- Table III. Matched analyses, male ALS cases and controls: occupational heavy metal expo- sures, Pb and Hg -------------------------------------------------------------------------------------------------------- Variable Exposure OR cl -------------------------------------------------------------------------------------------------------- All metal occupations > 1 6 1 3 6 8 2.20 0.69, 7.47 Pb occupations > 1 5 1 2 6 10 2.00 0.63, 7.00 Hg occupations >,- 1 2 6 4 2 1 1.50 0.31, 8.26 ++ = Both case and control exposed; case exposed, control not exposed; case not exposed, control exposed; both case and control not exposed. OR = Odds ratio; ('I confidence interval. -------------------------------------------------------------------------------------------------------- -------------------------------------------------------------------------------------------------------- Table IV. Number of persons with fractures by site: ALS cases, controls -------------------------------------------------------------------------------------------------------- Site ALS cases Controls --------------------------- ---------------------------- male female total male female total -------------------------------------------------------------------------------------------------------- Head 0 0 0 3 1 4 Neck 0 0 0 0 1 1 -Spine 1 2 3 1 3 4 Extremities 14 2 1 35 1 3 20 33 -------------------------------------------------------------------------------------------------------- Total 38 (58%) 42 (64%) -------------------------------------------------------------------------------------------------------- compared to 1-60 months (level 1), * MHPB = 8.397, * MHHG = undefined; and zero months exposure compared to 61+ months exposure (level 11), * MHPB = 0.974, * MHHG = 1. 106. No increased risk was shown with in- creased occupational exposure to lead or mercury. Swine Flu Vaccination and Other Ii?it?iunizations. There was no signif- icant difference between cases and controls on history of immunizations, including the 1976 swine flue vaccine. Forty-five percent of ALS males and 48% control males had received at least one immunization, as did 50% ALS females and 45% control females. ThyroidIMineral Disorders. The number of reported mineral and thy- roid disorders was similar in ALS patients and control subjects. Travel History. Eighteen percent of ALS males had traveled to both the Mariana Islands and Japan; 30% to Japan and 21 % to the Marianas. Control subjects reported travel to both places at a significantly lower rate of 9%, yet reported similar rates to individual countries (30% Japan, 18% Marianas). In female respondents, 9% of both groups had traveled to Japan, and 6% of both groups visited the Marianas. Military travel by the male subjects seems to account for the difference between male and female percentages. Poliomyelitis. No significant difference was found between cases and controls on history of paralytic poliomyelitis. Fractures. Comparing the number of individuals with at least one frac- ture, no significant difference was found between ALS and control subjects. Table IV shows fractures by site for males and females. Fourteen percent of the fractures in the control group and 4 % in ALS cases were reported as involving the head, neck or spine. Of those reporting fractures, the average number of years from the time of fracture to the time of diagnosis of ALS was 24.5 years in males (s.d. = 13.5) and 15.3 years in females (s.d. = 15.0). The average number of years from diagnosis to fracture was 3.8 years for males (s.d. = 2.6) and 2.5 years for females (s.d. = 1.0). No significant difference was found between cases and controls. Discussion The nervous system of humans is constantly being burdened by manufactured as well as naturally occurring neurotoxic chemicals. Of these, heavy metals remain complex and, in many aspects, undefined as regards biologic effect, toxicity and metabolism. From the data present- ed, it can be concluded that no relationship has been established between potential occupational exposure to heavy metals and the pathogenesis of ALS in this patient population. The occupations investigated are ones in which workers are potentially exposed to the heavy metals. The true haz- ard of exposure depends upon such factors as concentration, how it is handled, duration of exposure, and perhaps overridingly, susceptibility of the individual. Nonoccupational sources of heavy metal exposure were not investigated. No difference was found between cases and controls on the number of fractures. In cases, only 16% occurred within 5 years of diagnosis. The majority were more than 10 years before. If susceptibility to an injury is an indication of the early stages of disease, one would expect to see most frac- tures shortly before onset. Fractures after diagnosis occurred within an average of 3 years. The control group reported three times as many fractures in the head, neck and spine region than did ALS cases. The extremities accounted for the majority of the fractures sites in both groups. General trauma was not assessed. No significant difference was found for the remaining variables of immunization history, thyroid/mineral disorder, prior poliomyelitis and travel history, many of which have their derivation in the western Pacific surveillance of ALS [23]. In terms of family neurology, 1 1.6 % ofthe ALS cases had a family history of ALS. According to the literature, clinical stud- ies estimate a 5-10% rate of familial cases [241. Potential errors in the estimation of exposure, such as recall, instru- mentation, and prevarication, can bias the estimate of the relative risk. The study was planned to minimize susceptibilities which could threaten the research outcome and associated interpretation. Closed questions in a self- administered questionnaire afforded standardization and required only rec- ognition of occupations queried. The use of surrogate respondents allowed increased representativeness and sample size. Research by Her?nann [25] and Kolonel et al. [26] has demonstrated reliable and valid responses by surrogates to medical and personal inquiries. Selection bias can occur when choosing patients from a support group such as the Center for Neurologic Study. Studies based on voluntary group members indicate that such individuals are often more apt to be white, younger, of a higher socioeconomic status and have a longer duration of disease [27, 28]. However, controls in this study were selected from the friends/neighbors of these patients, which would tend to minimize such bias. The results of this study represent a detailed investigation of chronic and acute exposure to heavy metal via occupational experiences in the eti- ology of amyotrophic lateral sclerosis. Because of the great disparity among men and women in terms of occupational heavy metal exposures, it is not likely that occupational exposure would be a primary etiologic factor in the development of this disease. More realistically, the probabilistic nature of pathogenesis depends upon several factors and interactions. When examin- ing males only for 293 occupations in which metal exposure is at least possible, evidence still does not exist to support occupational heavy metal exposure as a risk factor. The contribution of this study is toward refuta- tion of this etiological theory, and refocusing of research energies in the future. References I Sobue, G.; Sahashi, K.; Takahashi, A.; Matsuoka, Y.; Muroga. T.; Sobue, I.: Degenerating compartment and functioning compartment of motor neurons in ALS: possible process of motor neuron loss. Neurology, Minneap. 33: 654-657 (1983). 2 Antel, J.; Arnason, B.; Fuller, T.: Histocompatibility typing in amyotrophic lateral sclerosis. Archs Neurol., Chicago 33: 423-425 (1976). 3 Kohne, D.; Gibbs, C.; White, S.; Tracy, S.; Meinke, W.; Smith, R.: Virus detection by ALS and Occupational Heavy Metal Exposure 37 nuclei acid hybridization: examination of normal and ALS tissues for the presence of polioviruses. J. gen. Virol. 56: 223-233 (198 1). 4 Rowland,' L.: Looking for the cause of amyotrophic lateral sclerosis. New Engi. J. Med. 311: 979-981 (1984). 5 Bradley, W.; Krasin, F.: A new hypothesis of the etiology of amyotrophic lateral sclerosis. Archs Neurol., Chicago 39.- 677-680 (1982). 6 Antel, J.; Noronha, A.; Oger, J.; Amason, B.: Human motor neuron diseases (Raven Press, New York 1982). 7 Kondo, K.; Tsubaki, T.: Case-control studies of motor neuron disease association with mechanical injuries. Archs Neurol., Chicago 38: 220-226 (198 1). 8 Pierce-Ruhland, P.; Patten, B.: Repeat study of antecedent events in motor neuron disease. Ann. clin. Res. 13: 102-107 (198 1). 9 Kurtzke, J.: Motor neuron(e) disease. Br. med. J. 284: 141-142 (1982). 10 Mulder, D.: The diagnosis and treatment of ALS (Houghton Mifflin, Boston 1982). 1 1 Currier, R.; Haerer, A.: Amyotrophic lateral sclerosis and metallic toxins. Archs envir. Hlth 17., 712-719 (1968). 12 Felmus, M.; Rasool, C.; Bradley, W.: Calcium contents of RBC's from patients with amyotrophic lateral sclerosis. Archs Neurol., Chicago 39.- 454 (1982). 13 Conradi, S.; Ronnevi, L.; Vesterberg, O.: Abnormal tissue distribution of lead in amyotrophic lateral sclerosis. J. neurol. Sci. 29.- 259-265 (1976). 14 Pierce-Ruhland, R.; Patten, B.: Muscle metal in motor neuron disease. Ann. Neurol. 8: 193-195 (1980). 15 Campbell, A.; Williams, E.; Barritrop, D.: Motor neuron disease and exposure to lead. J.- Neurol. Neurosurg. Psychiat. 33: 977-985 (1970). 16 Roclofs-lverson, R.; Mulder, D.; Eiveback, L.; Kurland, L.; Molgaard, C.: ALS and heavy metals: a pilot case-control study. Neurology, Minneap. 34: 393-395 (1984). 17 National Institute for Occupational Health and Safety: Criteria for a recommended standard: occupational expsure to inorganic mercury (US Department of Health, Education, and Welfar, Washington 1973). 18 National Institute for Occupational Health and Safety: Occupational diseases: a guide to their recognition (US Department of Health, Education, and Welfare, Washington 1977). 19 National Institute for Occupational Health and Safety: Criteria for a recommended standard: occupational exposure to ino rganic nickel (US Department of Health, Edu- cation, and Welfare, Washington 1978). 20 National Institute for Occupational Health and Safety: Criteria for a recommended standard: occupational exposure to inorganic lead (US Department of Health, Edu- cation, and Welfare, Washington 1978). 21 Johnson, N.; Kotz, S.: Distribution in statistics (Houghton Mifflin, Boston 1969). 22 Garruto, R.; Gajdusek, D.: Risk Factors for Senility (Oxford University Press, Oxford 1984). 23 Mantel, N.: Chi-square tests with one degree of freedom: extensions of tile Mantel- Haenszel procedure. J. Am. stat. Ass. 58: 690-700 (1963). 24 Rosen, A.: Amyotrophic lateral sclerosis: clinical features and prognosis. Archs Neu- rol., Chicago 35: 638-642 (1978) Gresham/Molgaard/Goibeck/Smith 38 25 Hermann, N.: Retrospective information from questionnaires. 1. Comparability of primary respondents and their next-of-kin. Am. J. Epidem. 121: 939-947 (1985). 26 Kolonel, L.; Hirohata, T.; Nomura, A.: Adequacy of survey data collected from sub- stitute respondents. Am. J. Epidem. 106: 476-484 (1977). 27 Juergens, S.; Mulder, D.: Clinical neuroepidemiology (Pitman Medical, Tunbridge Wells 1980). 28 Palo, J.; Jokelainen, M.: Geographic and social distribution of patients with amyo- trophic lateral sclerosis. Archs Neurol., Chicago 34: 724 (1977). Craig A. Molgaard, PhD, MPH, Division of Epidemiology and Biostatistics, Graduate School of Public Health, San Diego State University, San Diego, CA 92181 (USA)