Daniel Korya, MD

Seth A, Mossavar-Rahmani Y, Kamensky V, Silver B, Lakshminarayan K, Prentice R, et al. Potassium Intake and Risk of Stroke in Women With Hypertension and Nonhypertension in the Women’s Health Initiative. Stroke. 2014

The old saying of “an apple a day sends the doctor away” is said to have started in the 1860’s in the region of Pembrokeshire in Wales. The original phrase was actually “eat an apple on going to bed, and you’ll keep the doctor from earning his bread.” The phrase eventually evolved from its origins to “an apple a day, no doctor to pay”; and finally, the familiar “an apple a day sends the doctor away.” 

Apples are low in calories, low in sodium and high in vitamin C. A meta-analysis published in Stroke (2013) showed an inverse relationship between vitamin C intake and stroke risk.  Recently, potassium has been investigated for its ability to ward off stroke. Everyone seems to associate bananas with potassium since they are believed to have high levels of potassium (the average banana has 118 g of potassium, other foods with high potassium are mentioned below). So, perhaps another saying should be “a banana a day sends the doctor away”.  Or, should it?
The study published by Arjun Seth and his colleagues presented statistical analysis of a subgroup of the Women’s Health Initiative Observational Study (WHI-OS) cohort. The population was women aged 50-79 from 40 states who were prospectively followed for a mean of 11.1 years. Outliers based on caloric intake were removed and the data of quantity of daily potassium intake was derived from the food frequency questionnaires (FFQs). There were 90,137 patients remaining who were evaluated for the development of ischemic stroke during the follow-up period. The study controlled for age, ethnicity, BMI, history of smoking, alcohol intake, hypertension, aspirin use, hormone therapy, diabetes, history of MI, dyslipidemia and activity level.
After identifying this sub-cohort, the authors described their statistical methods and analysis. In summary, the group was broken down into quartiles based on mean daily potassium intake and an analysis of variance or a chi squared was performed to compare the quartiles with the covariates. Then, hazard ratios were estimated from the Cox proportional hazards model comparing the highest quartile with the lowest quartile of potassium intake. 
With regard to the analysis of hypertensive versus non-hypertensive patients, the researchers used a history of hypertension as opposed to hypertension during the follow-up period.  The argument for this distinction was the belief that potassium intake may be related to blood pressure and the possibility of hypertension being a mediator between potassium intake and stroke.  Essentially, the idea is that potassium may be mitigating the development of stroke through its blood pressure lowering effects. 
The results of the study are interesting. It was found that blacks, current smokers, and non-drinkers had lower dietary potassium intake.  Patients who were more active actually had more potassium intake. The highest quartile of potassium intake had a 27% lower incidence of stroke than the lowest quartile. This effect was 30% in a subgroup of patients with normal BMIs. There was a lower rate of small vessel strokes in non-hypertensive patients as compared with hypertensives.
To conclude, the authors remind us that this is the largest U.S. cohort of post-menopausal women and they reiterate the results. But, more importantly, a summary of the information this study provided is as follows: the amount of daily potassium post-menopausal women consumed was calculated based on the foods they reportedly ate. Foods that are high in potassium are salmon, green leafy vegetables, avocados, beans, baked potatoes, yogurt and bananas. Patients who ate those foods also tended to be non-smokers, non-blacks, who were drinkers with higher activity levels. Also, these patients were less likely to have small vessel ischemic strokes if they were not hypertensive.