In addition to its blood pressure-reducing effects, increased potassium intake may have independent vascular protective properties. This possibility has been evaluated in experimental studies conducted in rodents over the last four decades. In a series of animal models, including both stroke-prone spontaneously hypertensive (SHRSP) and Dahl salt-sensitive rats, the addition of either potassium chloride or potassium citrate markedly reduced the mortality from stroke, a reduction that was unrelated to any measured attenuation of hypertension (Tobian, 1986; Tobian et al., 1984). In a more recent study with SHRSP rats in which aortic blood pressure was measured by continuous radiotelemetry, dietary potassium supplemented as either potassium bicarbonate or potassium citrate attenuated hypertension and prevented stroke (Tanaka et al., 1997). However, supplemental potassium chloride exacerbated hypertension, increased risk of stroke (Tanaka et al., 1997), and amplified renal microangiopathy (Tanaka et al., 2001), in comparison with potassium bicarbonate or citrate.
Hence, at least in this animal model, the anion accompanying potassium had a major qualitative effect on outcomes such that potassium citrate or bicarbonate was beneficial, while potassium chloride appeared to be harmful. Still, the discordant results between this study and the cited study of Tobian and coworkers (1984) are difficult to reconcile and therefore preclude firm conclusions.
An inverse relationship between dietary potassium intake at baseline and subsequent stroke-associated morbidity and mortality has also been noted in several, but not all, cross-sectional and cohort studies (Table 5-6). In a 12-year follow-up of 859 men and women enrolled in the Rancho Bernardo Study and who were 50 to 79 years of age at baseline, a significant (p = 0.01) inverse relationship between potassium intake and subsequent risk of stroke-related mortality was noted (Khaw and Barrett-Connor, 1987). Each standard deviation increase in potassium intake (0.4 g [10 mmol]/day) at baseline was associated with a 40 percent reduction in risk of stroke-related mortality relative risk (RR) = 0.6; 95 percent confidence interval (CI) = 0.44 to 0.81 after adjustment for age, gender, systolic blood pressure, caloric intake, and other potential confounders. Limitations of the study included the restricted characteristics of the cohort and the fact that the findings were based on only 24 stroke deaths. No significant relationship with coronary heart disease was detected.
Over a 16-year follow-up in the Honolulu Heart Study (n = 7,591