An increase in ALS incidence on the Kii Peninsula, 1960-2009: A possible link to change in drinking water source

We investigated changes in the incidence of amyotrophic lateral sclerosis (ALS) in the Koza/Kozagawa/Kushimoto area (K. area) in the Kii Peninsula, Japan in 1960–2009. Probable and definite ALS cases diagnosed using El Escorial criteria were collected during a five-decade period: period I-V, 1960–2009. Forty-three ALS patients matched the selection criteria in the overall K. area, including three patients on Oshima, a small island opposite the mainland K. area. The age- and gender-adjusted incidence of ALS in the overall K. area (standardized for the 2005 Japanese population) decreased from 5.47/100,000 (95% CI 1.86–9.08) in period I to 0.61/100,000 (95% CI-0.28–1.50) in period III, and then increased to 4.39/100,000 (95% CI 1.70–7.07) in periodV. On Oshima, the age- and gender-adjusted incidence of ALS was 9.45/100,000 (95% CI—7.39–26.29) in period V. The present research indicates an increase of ALS incidence in the K. area, especially on Oshima. A limitation of this study was the small population.


Introduction
Amyotrophic lateral sclerosis (ALS) is a devastating adult-onset degenerating disease of unknown etiology of the motor neuron systems. The Koza, Kozagawa and Kushimoto (K.) area in the Kii Peninsula of Japan was reported to have a higher incidence of ALS in the 1950s than other areas of the world (1 -5). Epidemiologic research showed that drinking water sourced from Kozagawa River in the K. area contained severely low levels of Ca and Mg, and Ca/Mg defi ciency was speculated to have a role in the development of ALS in these areas (5,6). On Oshima, a small island municipally included in the K. area, the source of drinking water was changed from regional water to the Kozagawa River in 1975. To clarify whether ALS epidemiology on Oshima changed after altering the water source, we investigated changes in ALS incidence on Oshima and in the K. area in 1960 -2009.

ORIGINAL ARTICLE
An increase in ALS incidence on the Kii Peninsula, 1960 -2009: A possible link to change in drinking water source period II, 1970period II, -1979period III, 1980period III, -1989period IV, 1990period IV, -1999period V, 2000-2009. Regional physicians including neurologists in hospitals and clinics and the staff of the public healthcare and welfare center in the K. area were requested by our team to report all patients with possible motor neuron disease every year (7). To ensure complete case identifi cation, Wakayama Prefecture ' s List of Patients with Intractable Disease certifi ed by the Ministry of Health and Welfare of Japan was used. The selection criteria were as follows: 1) patients who our neurologists examined and diagnosed using the El Escorial criteria (8); 2) patients with probable or defi nite ALS who had been living in the K. area, including Oshima, for at least one year before diagnosis. Patients who showed Parkinsonism or dementia during the disease course (ALS/Parkinsonism dementia complex: ALS/PDC) were also included.
The present research was approved by the ethics committee of Wakayama Medical University and Kansai University of Health Sciences.

Statistics
A direct method was used to standardize the annual incidence rates by age and gender, using populations in the 2005 census in Japan: 127,537,189. Unpaired t -test was performed and two-sided p Ͻ 0.05 was considered signifi cant.

Patient population and clinical characteristics
We enrolled 50 patients with defi nite or probable ALS in the K. area, including Oshima, in 1960Oshima, in -2009, and 43 patients matched the selection criteria (Table I)

Follow-up study on Kii ALS 349
(seven were excluded because they were living outside the K. area when diagnosed). Patient distribution was restricted to the mainland side of the K. area during periods I -IV. Three ALS patients were found on Oshima (two males and one female) in period V (Figure 1). The mean age at onset in period V was the highest among the periods (Table I). The male: female ratio was low in period V compared to periods I -IV. The frequency of cases with a positive familial history in a detailed interview was 14.0%. Cu/Zn superoxide dismutase (SOD1) genes were analyzed in three of six familial cases; none had a SOD1 gene mutation. Three ALS/PDC patients were found in the K. area in period IV and one patient on Oshima in period V.

Discussion
The age-and gender-adjusted ALS incidence in the K. area decreased from 5.47 in 1960 -1969 to 0.61 in 1980 -1989, and then increased to 4.39 in 2000 -2009. The declining trend of the male: female ratio in the past 10 years was comparable with other reports (9,10), and the recent increase of ALS incidence in females could be related to some type of environmental or cultural factor pertaining in females in this area and the increased confi rmation of older female patients (11). The reason for the decline in 1980 -1989 is not clear, but might be partially due to missing cases from the data set, emigration, and environmental, lifestyle and cultural changes. Recent reports of annual age-adjusted ALS incidences ranged from 0.42/100,000 (12) to 2.96/100,000 (13) in other areas in the world (14 -17). Taken together, the ageand gender-adjusted incidence in the K. area and Oshima in 2000 -2009 was higher than in other areas. On Oshima, no patient with ALS was found in previous research in 1946 -1965 (18). It is noteworthy that a high ALS incidence was fi rst found on Oshima in 2000 -2009 after the drinking water source was changed to the Kozagawa River in 1975. The present research indicated an increase of ALS incidence in the K. area, especially on Oshima. A limitation of this study was the small population. Continuous study over a longer period is needed in this area. Comparing the incidence by the periods, the age-and gender-adjusted incidence of ALS in the overall K. area for the 2005 Japanese population was 5.47/100,000 (95% CI 1.86 Ϫ 9.08), male: 8.29/100,000 (95% CI 1.84 Ϫ 14.73), female: 2.98/100,000 (95% CI Ϫ 0.69 Ϫ 6.66) in period I, markedly decreased to 0.61/100,000 (95% CI Ϫ 0.28 Ϫ 1.50), male: 1.36/100,000 (95% CI Ϫ 0.58 Ϫ 3.30), female: 0 in period III, but recently increased again to 4.39/100,000 (95% CI 1.70 Ϫ 7.07), male: 4.01/100,000 (95% CI 0.22 Ϫ 7.81), female: 4.71/100,000 (95% CI 0.93 Ϫ 8.49) in period V.
Muro disease (Chairman, Yasumasa Kokubo), the Ministry of Health, Labour and Welfare of Japan, and a grant-in-aid for Scientifi c Research of Japan.