Japanese-American participants), there was a significant inverse relationship (p = 0.002) between potassium intake and mortality from thromboembolic stroke (Lee et al., 1988). No significant association was noted for nonfatal thromboembolic stroke or for fatal or nonfatal hemorrhagic strokes. Additionally, inverse relationships between potassium intake and stroke mortality were noted in several cohort studies (Sasaki et al., 1995; Xie et al., 1992; Yamori et al., 1994) but these findings were not adjusted for caloric intake and/or were based on an ecologic analysis (Xie et al., 1992). In a 7-year follow-up report of 5,754 men and 5,875 women who were participants in the Scottish Heart Health Study, an inverse relationship between potassium intake and subsequent death, both from all causes and from coronary heart disease, was found (Tunstall-Pedoe et al., 1997).

Similarly, over the course of 8 years of follow-up in 43,738 U.S. men in the Health Professionals Study, there was a significant inverse relationship between baseline potassium intake and stroke (p = 0.007 for trend across quintiles of potassium intake) after adjustment for established cardiovascular disease risk factors, including blood pressure and caloric intake (Ascherio et al., 1998). The multivariate RR of stroke for men in the highest versus lowest quintile of potassium intake was 0.62 (95 percent CI = 0.43−0.88). The association was similar for both ischemic (n = 210) and all (n = 328) strokes. Use of potassium supplements was also inversely associated with the risk of stroke.

In a 14-year study of 85,764 U.S. women who participated in the Nurses Health Study, there was an inverse relationship between potassium intake and ischemic stroke (RR = 0.72; 95 percent CI = 0.51 to 1.01 for comparison of upper and lower quintiles of potassium intake; 347 strokes occurred during this time period), but much of the association was lost following adjustment for calcium intake (Iso et al., 1999). Two analyses of NHANES I follow-up study have been reported. In a 17-year analysis of subsequent stroke mortality in approximately 10,000 men and women (during which there were 304 strokes), there was a significant inverse relationship between potassium intake and stroke mortality in hypertensive and African-American men, but not in other subgroups (Fang et al., 2000). In a 19-year follow-up of the same cohort, the relationships of potassium intake with fatal and nonfatal strokes (total n = 927) and coronary heart disease (n = 1847) events were assessed (Bazzano et al., 2001). Overall, stroke hazard was significantly different among quartiles of potassium intake (p = 0.03), but the relationship was nonlinear. Participants in the lowest quartile of potassium intake at baseline