Bone mass in Guamanian patients with amyotrophic lateral sclerosis and parkinsonism-dementia.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 80:107-113 (1989) Bone Mass in Guamanian Patients With Amyotrophic Lateral Sclerosis and Parkinsonism-Dementia RALPH M. GARRUTO, CHRIS C. PLATO, RICHARD YANAGIHARA, KATHLEEN FOX. JAMES DUTT. D. CARLETON GAJDUSEK. AND JORDAN TOBIN Laboratory of Central Nervous System Studies, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20892 (R.M.G.,R.Y., D.C.G.);Applied Physiology Section, National Institute on Aging, Francis Scott Key Medical Center, Baltimore, Maryland 21224 (C.C.P., K.F., J.T.); Department of Computer Science, University of South Alabama, Mobile, Alabama 36688 (5.0.) KEY WORDS Bone loss, ALS, PD Guam, Longitudinal study ABSTRACT Bone mass, as assessed by measurements of total subperiosteal diameter and medullary width of the second metacarpal bone on hand-wrist radiographs, was evaluated for 31 Guamanian patients (15 males, 16 females) with amyotrophic lateral sclerosis (ALS), 67 patients (39 males, 28 females) with parkinsonism-dementia (PD), and 66 (34 males, 32 females) nonaffected Guamanian controls. Comparisons between the two disease groups and between each disease group and the nonaffected controls were made taking into account the sex, age, and disability status ofeach participant. At all ages, ALS patients of both sexes had significantly lower percent cortical areas (PCA) than did nonaffected controls. The ALS males also had significantly lower PCA than PD males, although no significant differences were observed between female ALS and PD patients. The PD patients of either sex had a lower PCA when compared to controls, but the differences were not statistically significant. The observed differences in PCA were due solely to increased medullary width, suggesting that the diminished cortical bone thickness resulted from greater bone resorption rather than differential bone growth. Longitudinal studies support the cross-sectional findings of accelerated bone loss among ALS patients. It is not possible to determine from the present data whether the observed differences in PCA of the second metacarpal of the ALS patients are due to atrophy of the first interosseous muscle, to a generalized resorption process inherently associated with the development and progression of ALS,or to factors not accounted for by the present analysis. We previously demonstrated a lower bone mass in neurologically normal children, adolescents, and adults from the island of Guam, compared to other ethnic groups in the continental United States (Plato et al., 1982,1984). In the present report, we evaluated the bone mass in Guamanian patients with amyotrophic lateral sclerosis (ALS)and parkinsonism-dementia (PD), two neurodegenerative disorders occurring in high incidence among Guamanians. Clinically, ALS on Guam is the same as that described by Charcot more than one hundred years ago. It is a disease of the motor neurons and the @ 1989 ALAN R. LISS. INC. corticospinal tract, characterized by muscle weakness, progressive muscle atrophy, paralysis and spasticity. Neuropathologically, Guamanian ALS is similar to classical ALS worldwide, except that it is often associated with neurofibrillary tangles in the brain and spinal cord. The second neurological disorder, PD, is found in equally high incidence on Guam and is often encountered together Received October 12,1987; accepted November 4,1988. Address reprint requests to Dr. Ralph M. Garmto, Building 36, Room 5B-21, National Institutes of Health, Bethesda, Maryland 20892. 108 R.M. GARRUTO ET AL. with ALS in the same family, the same sibship, and occasionally in the same individual. Parkinsonism-dementia is characterized clinically by bradykinesia (slowness of voluntary motor activity), muscular rigidity, and tremor. It is always accompanied by an early onset progressive dementia. Neuropathologically PD is associated with severe neuronal loss and neurofibrillary tangles but, unlike Alzheimer disease, senile plaques are not usually seen. The objective of the current study was to determine if a lower bone mass was primarily associated with the disease process or was a secondary consequence ofimmobility or other factors, such as age, sex, and menopause. During the past decade, systematic and intensive investigations on the etiology and pathogenesis of, and pathogenetic relationship between, ALS and PD of Guam have revealed subtle defects in mineral metabolism and selective intra- neuronal deposition of calcium, aluminum, and silicon in brain and spinal cord of these patients (Garruto, 1988; Garruto et al., 1984; Garruto and Yase, 1986; Yanagihara et al., 1984). MATERIALS AND METHODS Bilateral hand-wrist radiographs were obtained from 31 Guamanian patients (15 males, 16 females) with ALS,67 patients (39 males, 28 females) with PD, and 66 neurologically normal Guamanian controls (34 males, 32 females). Diagnoses were clinically verified by neurologists at the National Institute of Neurological Disorders and Stroke (NINDS) Research Center on Guam, and registered patients were examined at three- to six-month intervals. Informed consent was obtained from all participants. Patients were rated for degree of disability (see criteria in Table 1) with 0 indicating no TABLE 1. Disability categories for patients with amyotrophic lateral sclerosis a n d parkinsonism-dementia of Guam and percent of patients in each category Disability score 0 Disability category None 1 Mild 2 Moderate 3 Marked Amyotrophic lateral sclerosis Criteria Percent No weakness or muscle atrophy Mild weakness and muscle atrophy Unable to work, yet able to care for self; fully ambulatory Moderate weakness and muscle atrophy Requiring help in dressing, bathing, eating, daily functions; needs assistance with walking Marked weakness and muscle atrophy Requiring total or nearly total nursing care; confined to bed or wheelchair 0 41 25 34 Parkinsonism-dementia Criteria Percent No bradykinesia', tremor, rigidity, dementia Mild bradykinesia, tremor, rigidity, dementia Unable to work, yet able to care for self; fully ambulatory Moderate bradykinesia, tremor, rigidity, dementia Requiring help in dressing, bathing, eating, daily functions; needs assistance with walking Marked bradykinesia, tremor, rigidity, dementia Requiring total or nearly total nursing care; confined to bed or wheelchair 2 43 28 27 'Bradykinesia refers to the slowness of voluntary motor activity. TABLE 2. Mean age, age range, a n d mean disability scores of Guamanian patients with amyotrophic lateral sclerosis a n d parkinsonism-dementia a n d nonaffected controls Disease category Amyotrophic lateral sclerosis Male Female Parkinsonism-dementia Male Female Control Male Female No. of individuals mean i SE (years) range (years) 15 16 53.13 k 2.72 52.19 2.49 + 35-71 33-66 39 28 60.74 f 1.39 60.07 f 1.54 43-81 48-76 34 32 53.76 5 1.57 52.25 1.83 38-82 33-71 * 109 BONE LOSS IN ALS AND PD OF GUAM disability, 1 indicating mild, 2 indicating moderate, and 3 indicating marked disability. All controls had disability scores of zero. Percentages of patients characterized by the various degrees of disability are also listed in Table 1;thus, it becomes evident that while the end result of disability in each score is comparable between ALS and PD, the components and the diagnostic criteria for each score are not the same in both diseases. Table 2 summarizes the mean age and age range for ALS and PD patients and controls. To evaluate differences in the rate of bone loss over time, we collected serial radiographs from 17 ALS patients (7 males, 10 females), 27 PD patients (17 males, 10 females), and 22 nonaffected controls (9 males, 13 females) between 1977 and 1981. Most participants had only two serial X-rays, although some had as many as five. Total subperiosteal diameter (T) and medullary width (M) of the second metacarpal bone were measured directly from the radiographs (Fig. l),utilizing a dial-reading caliper with 0.05 mm readout capability (Garn, 1970; Plato and Norris, 1980). Percent cortical area (PCA), derived from T and M, where PCA = [(T2 - M2)/T21 x 100, was selected, instead of cortical thickness, as the indicator of bone mass, because it accounts for differences in bone size. The data were analyzed utilizing standard methods by t-tests, analysis of variance, and linear regression analysis. Analysis of variance comparisons were tested using Duncan's multiple range test (Duncan, 1975) and Bonferroni's t-test (Miller, 1981). RESULTS The PD patients were approximately seven years older than ALS patients and controls (Table 2). Mean age differences be- 100 I I I I 1 I 90 - 80 - 70 0 I- 8 0 601 1 30 20 30 40 50 60 70 80 90 AGE Fig. 1. Scatter diagram and regression lines of percent cortical area of the left second metacarpal bone on age in male Guamanian patients with amyotrophic lateral sclerosis (ALS), parkinsonism-dementia (PD), and in nonaffected controls. tween PD patients and ALS patients by sex were significant (P < .Ol), as were differences between PD patients and controls (P < .01). Significant differences were not found in ages of male or female ALS patients and their respective controls. The percent of ALS and PD patients in each disability category is shown in Table 1.There was no significant difference between A L S and PD in the number of patients in each disability category. There was also no significant difference in the mean PCA value between patients with different disability scores, except in male ALS patients between those with disability scores of 1and 3. It seems, therefore, that the degree of overall disability has little effect on the PCA of the second metacarpal of male ALS patients and no effect at all on the PCA TABLE 3. Bilateral measurements o f the second metacarpal bone of Guamanian patients with amyotrophic lateral sclerosis (ALS) and parkinsonism-dementia (PO)and nonaffected controls Sex Male Female Disease category N ALS PD Control ALS PD Control 15 39 34 16 28 32 Left hands Subperiosteal Medullary diameter width Mean f SE Mean k SE 8.71 f 0.18 8.94 !r 0.1 1 8.76 i 0.11 7.98 f 0.16 7.81 0.11 7.74 0.09 * + 4.01 3.65 3.14 3.63 3.66 3.02 f 0.39 -I 0.14 i 0.17 f 0.27 f 0.16 0.17 + Right hands Subperiosteal Medullary diameter width Mean SE Mean f SE + 9.23 + 0.21 9.23 k 0.14 9.12 !r 0.11 8.21 f 0.17 8.10 k 0.12 8.07 0.10 + 4.29 f 0.44 3.96 f 0.16 3.40 0.18 3.70 f 0.32 4.01 k 0.18 3.16 f 0.19 + 110 R.M. GARRUT0 ET AL. of female ALS patients, or of PD patients of either sex. No significant differencesin total subperiosteal diameter and medullary width were found between disease categories (Table 3). Likewise, with the exception of male and female controls, no significant bilateral differences were found. However, male and female controls had a significantlylower medullary width (I' < .01)than either ALS or PD patients. Comparisons of bone mass between ALS patients, PD patients, and controls were performed using the PCA of the left (or nondominant) hand. Since analysis of variance of PCA in terms of age, sex, diagnosis, and disabilitystatus gave significant F values for TABLE 4. Analysis of variance: Percent cortical area of the second metacarpal bone of Guamanian patients with amyotrophic lateral sclerosis and parkinsonism-dementia and nonaffected controls Source d.f. Total Error Model Age Sex Diagnosis Disability 109 102 7 Sum squares Mean squares F value P 10,663 7,585 3,078 1,207 325 758 788 74 440 1,207 325 379 263 5.94 16.31 4.39 5.12 3.55 ,0001 ,0001 .039 .008 .017 1 1 2 3 I00 I b A Male I - Disease category N PCA Mean =k SE ALS PD Control ALS PD Control 15 39 34 16 28 32 76.05 k 2.21 84.09 3z 1.41 85.24 f 1.48 16.72 f 2.05 79.25 f 1.61 82.70 k 1.46 ~~~ Female ~~~ TABLE 6. Multivariate analysis of variance: Percent cortical area (PCA) of the second metacarpal bone of Guamanian patients with amyotrophic lateral sclerosis (ALS) and parkinsonism-dementia (PD) and nonaffected controls adjusted for age Sex Contrast Sum of squares F value P Male ALS vs. PD ALS vs. Control PD vs. Control ALS vs. PD ALS vs. Control PD vs. Control 647 883 21 59 380 156 9.05 12.34 0.30 0.90 5.78 2.36 .0035 ,0007 .5856 .3453 ,0188 ,1287 Female - _-A. - Ao OiSo - $3 80 - A 70 - 0. *O 0. 60 - 40 - Sex all four variables (Table 4), further comparisons were made by diagnosis, sex, and age ~ O U D Fimre . 1 shows the scatter diamam tion for the female comparisons. Table 5 presents the age-adjusted mean percent cortical area by sex and diagnosis. In both sexes, nonaffected controls have the highest mean PCA and the ALS patients have the lowest. The mean PCA of PD patients lies between those of the controls and ALS patients, but closer to the former. Statistical comparisons of these means, through multivariate analysis of variance, shown in Table 6, indicate that 1)the differences in PCA between ALS patients and controls are statistically significant in both males and females, 2) differences between PD patients and controls are not statistically significant in either sex, and KO - =A."*.A"$ 90 - 50- $ TABLE 5. Age-adjusted mean percent cortical area (PCA) of the second metacarpal bone of Guamanian patients with amyotrophic lateral sclerosis (ALS) and parkinsonism-dementia (PD) and nonaffected controls i-np0-i -* ALS ---A CONTROLS 08 - - Fig. 2. Scatter diagram and regression lines of percent cortical area of the left second metacarpal bone on age in female Guamanian patients with amyotrophic lateral sclerosis (ALS), parkinsonism-dementia (PD), and in nonaffected controls. the PEA of ALSYpatients is lower than those ofPD and controls at all ages of both sexes. Because of differences in the mean age of onset between ALS and PD patients, the 111 BONE LOSS IN ALS AND PD OF GUAM participants were separated into three age groups: younger than 50 years (<50), between 50 and 59 years (50-59), and 60 years or older (360). Table 7 shows the mean ages and mean PCA by sex, disease category, and age group. This age grouping was more critical in females, as 9 of 32 controls, 5 of 16ALS patients, and none of the 28 PD patients were below the average age of menopause (46.6 years) of Guamanian women (Plat0 et al., 1982). No statistically significant differences in the mean age were found between disease categories within age groups. Among females in the <50 age group, no significant differences in PCA were found between any of the disease categories. Both female PD patients in this age group were postmenopausal, while the ages of 5 of the 7 ALS patients and 7 of the 11 controls were below the average menopausal age. Among younger males, ALS patients had the lowest PCA. However, the only significant difference was betweenALS and PD patients (P < .01). In the 50-59 age group, male and female ALS patients had the lowest PCA, and controls of either sex had the highest. In males, the differences in PCA between ALS patients and controls and between ALS and PD patients were statistically significant (P < .01). In females, significant differences ( P < .01) were found only between ALS patients and controls. In the 260 age group, the mean PCA of the ALS patients was again the lowest, and those of the nonaffected controls the highest. However, the differences between disease categories were not significant. The mean age of onset for ALS was 46.9 years in males and 46.3 in females, while the mean age of onset for PD was 55.8 in males and 55.3 in females. A significantly negative correlation ( r = -0.52, P < .001) was found between the age of onset of PD in males and PCA. By contrast, no significant correlation was found between the age of onset and PCA in male and female patients with ALS, and in female PD patients. The duration of disease (from onset to radiographic examination) for ALS was 6.26 years in male patients and 5.9 years in female patients. The duration for PD was 4.7 and 4.2 in male and female patients, respectively. No significant correlations were found between duration of disease of either ALS or PD and PCA, except in males the youngest age group ( 4 0years) had a significantly negative correlation of -0.79 ( P < .05). Comparisons between disability and disease duration for male or female ALS patients were nonsignificant when age groups were studied separately. There were, however, significant correlations ( r = 0.56, P < .03 in ALS males; r = 0.69, P < .003 in ALS females, and r = 0.53, P < .006 in PD females) when age groups were pooled. Correlations between disability and duration of PD in males were not significant. The longitudinal analysis of PCA by disease category is summarized in Table 8. The average annual regression coefficient is the average of the individual regression coefficients of participants from their first to last visit. With the exception of the control females (who were also the youngest), all groups had a negative correlation with age. Only the coefficients for ALS male and PD female were significantly (P < .05) different from zero. Comparisons between pairs of coefficients resulted in statistically significant differences between ALS and PD males (P < .Ol), ALS and control males (P < .05), and PD and control females (P < .05). TABLE 7 . Mean percent cortical area (PCA) of the second metacarpal bone of Guamanian patients with amyotrophic lateral sclerosis (ALS) and parkinsonism-dementia (PD) and nonaffected controls Age group (years) <50 50-59 260 Disease category ALS PD Control ALS PD Control ALS PD Control N Male Age f SE (years) PCA f SE N 7 2 9 3 16 16 5 21 9 44.1 f 2.0 44.5 f 1.5 43.0 f 1.2 53.7 f 0.7 54.4 f 0.7 53.6 f 0.7 65.4 f 2.0 67.1 1.3 64.9 f 2.3 82.9 f 0.4 89.0 f 3.5 87.8 f 2.0 66.2 f 9.3 85.0 f 1.6 87.3 f 1.8 75.6 f 7.9 80.6 f 1.3 82.5 f 3.3 6 2 * 11 6 11 14 4 15 7 Female Age f SE (years) * 41.7 2.4 48.5 5 0.5 40.8 f 1.6 53.3 k 1.6 53.5 f 0.8 54.4 0.8 63.2 3~ 1.6 66.5 f 1.3 66.0 f 1.5 + PCA & SE 85.7 f 4.3 81.0 f 4.7 87.3 f 2.6 73.2 4.2 79.7 f 3.0 85.8 f 1.4 73.5 3.5 74.8 2.0 74.9 f 2.7 + ** 112 R.M. GARRUTO ET AL. T A B L E 8. Longitudinal analysis of the progressive loss o f percent cortical area in Guamanian patients with amyotrophic lateral sclerosis ( A L S ) and parkinsonism-dementia (PD)and nonaffected controls Sex Male Female Disease category No. of individuals Average annual' regression + SE Mean age (years) 1st visit Age range (years) 1st VlSlt Mean2 interval (years) ALS PD Control ALS 7 17 9 10 10 13 -5.15 f 2.02 -0.62 0.39 -0.49 -+ 0.42 -2.49 1.72 -2.79 1.17 0.22 zk 0.52 50.1 61.3 54.2 53.1 59.2 44.2 35-68 53-82 26-82 39-67 50-75 24-60 1.7 2.9 2.6 2.5 2.4 2.3 PD Control + *+ 'Average of the individual annual regression ciiefficients of percent cortical area on age. "Mean number of years between first and last visit. DISCUSSION Our data indicate that nonaffected neurologically normal Guamanian males have a higher PCA than either ALS or PD patients of the same age and sex, and ALS patients have the lowest, although these differences are not always statistically significant. ALS patients of either sex have a significantly lower PCA than do controls of the same age and sex. Male ALS patients also had significantly lower PCA in the second metacarpal than did male PD patients. Differences between PD and nonaffected controls were nonsignificant in both males and females. In evaluating these results, one is confronted with the inherent differences in bone mass related to sex, age, menopausal status, age of onset of the disease, and degree and duration of immobility and disability. The contributions of sex and age were controlled (with one exception) by analyzing the male and female data separately by age groups. The only exception was in the 4 0 year-old group, where both female PD patients in this age group were postmenopausal. Their mean age was also significantly higher than that of female ALS patients, or controls in the same age group. This inter-group age difference could account for the lower PCA in the PD women. While the older age and the postmenopausal status of the <50-year PD females may be responsible for their lower PCA, these factors cannot account for the lower PCA values of the ALS male, or the female ALS patients in the remaining age groups. Age of onset of either ALS or PD did not show consistent correlations with PCA. It is not clear from our data why the later onset of PD in males resulted in less cortical bone. Immobilization is known to enhance bone loss. However, disability could not account for the differences in PCA between ALS and PD patients, since their disability scores were not significantly different. One could argue, however, that differences in age of onset and/or the duration of disability prior to radiographic examination may be important factors in determining the amount of bone loss. Even though our data showed associations between duration of disease and disability, we found no consistent correlation between the duration of either ALS or PD with PCA. It is possible that patients with a longer duration have a less severe form of the disease so that, even though the duration from onset to death is longer, the overall extent of immobilization may be no greater than in patients with a shorter duration. Indeed, the only significant correlation between duration and PCA was found in the youngest (<50 years) ALS males. Our results indicate that, while ALS and PD patients have similar disability scores, the former consistently have lower PCA values. Two questions emerge: first, is the measure of disability in the two diseases comparable, and second, how does the measure of disability relate to the bone being measured? Although the end result for the degree of disability in Table 1 is the same for ALS and PD, the criteria used in deriving the degree of disability for each disease are not. In ALS, the criteria revolve around the issue of weakness and muscle atrophy, whereas in PD the major criteria are slowness of voluntary movement and rigidity. An earlier study (Plato and Norris, 19801,indicated that bone remodelling responds to the amount of physical stress exerted on the bone by the muscle. Thus, progressive muscle atrophy in ALS may have a more profound effect on bone mass than on overall immobility. Furthermore, ALS involves the atrophy of one of the BONE LOSS IN ALS AND PD OF GUAM muscles attached to the bone (second metacarpal) being measured. Hence, the similarity in the overall immobility between A L S and PD notwithstanding, the second metacarpal bone of ALS patients should be losing more bone than that of PD patients. In other words, the effect of disability and immobility on bone remodelling may be preferential and not always generalized. An analysis of the longitudinal data suggests that male ALS patients lose bone more rapidly than either male PD patients or nonaffected controls, and that female PD patients similarly have a higher rate of bone loss than do female controls. Our longitudinal data do not permit a full ascertainment of the sources of variation in bone loss. Because of the small number of participants in each category, it was impossible to adjust adequately for age or duration of disability. The size of the data set notwithstanding, the results of the longitudinal analysis offer additional support to the conclusions of the cross-sectional analysis. Our present results from hand-wrist radiographs of Guamanian patients with ALS and PD support the notion that once the intervening factors of age, sex, menopause status, disease duration, and immobility and disability are controlled, there is still a significant difference in bone PCA between ALS patients and Guamanian controls of either sex and between male ALS and male PD patients. This point is supported by a recent report by Glasberg et al. (1987a) who studied bone mineral content through single photon absorptiometry in the radius of United States ALS patients. They also found increased incidence of osteopenia in the patient group, which could not be accounted for by disability. Whether the PCA differences are the result of weakness and atrophy of the second interosseous muscle in ALS or are due to a generalized bone demineralization associated with calcium deficiency, toxic metals, and metabolic dysfunction in these disorders on Guam (Garruto et al., 1985; Garruto and Yase, 1986; Glasberg et al., 1987a; Glasberg et al., 1987b1,or a combina- 113 tion of both factors, could not be ascertained from the present data. LITERATURE CITED Duncan DB (1975) t-Tests and intervals for comparisons suggested by the data. Biometrics 32:339-359. Garn SM (1970) The Earlier Gain and Later Loss of Cortical Bone. Springfield, I L Charles C. Thomas. Garruto RM (1989) Amyotrophic lateral sclerosis and parkinsonism dementia of Guam: Clinical, epidemiological and genetic patterns. Am. J . Hum. Biol. (in press). Garruto RM, Fukatsu R, Yanagihara R, Gajdusek DC, Hook G, and Fiori CE (1984) Imaging of calcium and aluminum in neurofibrillarv tangle-bearing neurons in parkinsonism-dementia df GuLm. Proc. Gatl. Acad. Sci. U S A . 82:1875-1879. Garmto RM. Yanaeihara R. and Gaidusek DC (1985) Disappearance oT high incident aGyotrophic lateral sclerosis and Parkinson-dementia of Guam. Neurology 35:193-198. Garruto RM, and Yase Y (1986) Neurodegenerative disorders of the Western Pacific: The search for mechanisms of pathogenesis. Trends Neurosci. 9:36%374. Glasberg M, Gross B, Kleerekoper M (1987a) Bone mineral content and osteopenia in amyotrophic lateral sclerosis. Book of Abstracts, International Conference of Amyotrophic Lateral Sclerosis, October 29-31, 1987, Kyoto, Japan, p. 117. Glasberg M, Kleerekoper M, and Glasberg J (1987b) Parathyroid metabolism and calcium homeostasis in ALS. Book of Abstracts, International Conference of Amyotrophic Lateral Sclerosis, October 29-31, 1987, Kyoto, Japan, p. 77. Miller RG, Jr (1981) Simultaneous Statistical Inference. New York: Springer-Verlag. Plato CC, Garruto RM, Yanagihara RT, Chen KM, Wood JL, Gajdusek DC, and Norris AH (1982) Cortical bone loss and measurements of the second metacarpal bone: I. Comparisons between adult Guamanian Chamorros and American Caucasians. Am. J. Phys. Anthropol. 59:461465. Plato CC, Greulich WW, Garruto RM,and Yanagihara R (1984) Cortical bone loss and measurements of the second metacarpal bone: 11. Hypodense bone in postwar Guamanian children. Am. J. Phys. Anthropol. 6357-63. Plato CC, and Norris AH (1980) Bone measurements of the second metacarpal and grip strength. Hum. Biol. 52:131-149. Yanaghara R, Garruto RM, Gajdusek DC, Tomita A, Uchikawa T, Konagaya Y, Chen KM, Sobue I, Plato CC, and Gibbs CJ (1984) Calcium and vitamin D metabolism in Guamanian Chamorros with amyotrophic lateral sclerosis and parkinsonism-dementia. Ann. Neurol. 25:42-48.