Intake of animal products and stroke mortality in the Hiroshima/Nagasaki Life Span Study.

BACKGROUND

To determine whether intake of animal products was associated with a reduced risk of stroke mortality in a large-scale population-based cohort in Japan.

METHODS

A self-administered questionnaire, including questions on dietary habits, was mailed to the members of the Life Span Study, a cohort of people exposed and non-exposed to atomic bomb radiation, who were alive as of 1 September 1979. Animal products included frequency intake of beef/pork, chicken, ham/sausage, milk, dairy products, eggs, fish, and broiled fish. Responses were obtained from 40 349 people (72%): 15 350 men (mean age 54 years) and 24 999 women (mean age 58 years). The subjects were followed for 16 years, and deaths were ascertained by linkage to the nationwide family registration system of Japan. The associations between diet and stroke mortality were examined using a Cox proportional hazard model.

RESULTS

During the follow-up period, 1462 stroke deaths occurred. Four animal products comprising eggs, dairy products, fish, and broiled fish were independently associated with a decreased risk of stroke mortality; while beef/pork, chicken, ham/ sausage, and milk consumption were not associated with stroke death. A composite measure of eggs, dairy products, fish, and broiled fish intake was calculated, and the highest tertile was significantly inversely associated with total stroke mortality (Hazards Ratio [HR] = 0.80, 95% CI: 0.68, 0.93) compared with the lowest tertile. The protective effect of animal product intake on total stroke death was largely confined to intracerebral haemorrhage death; the RH of intracerebral haemorrhage death for the highest tertile of consumption was 0.72 (95% CI: 0.53-0.98) compared with the lowest tertile; animal products intake was not related to cerebral infarction mortality (HR = 0.84; 95% CI: 0.67-1.06).

CONCLUSIONS

Intake of animal products such as eggs, dairy products, and fish may be protective against intracerebral haemorrhage, but is not related to cerebral infarction mortality.