The Pharmaceutical Journal
Vol 269 No 7211 p206
17 August 2002

This article
Reprint
Photocopy

News summary

Related websites
Neurology (archneur.ama-assn.org)

Low potassium levels may increase stroke risk for elderly taking diuretics

Low serum potassium levels in patients taking diuretics, and low potassium intake in those not taking diuretics, is associated with an increased incidence of stroke in the elderly, say American researchers.

Dr Deborah Green, of the Neuroscience Institute at the Queen's Medical Centre in Honolulu, Hawaii, and colleagues conducted an observational study involving 5,600 men and women over 65 years of age. The participants, who had not had a stroke when they started the study, were followed for four to eight years and the number and type of strokes that occurred were recorded.

The researchers found that serum potassium levels less than or equal to 4.0mEq/L were associated with an increased relative risk for stroke, but only for people receiving diuretics. Among users of hydrochlorothiazide diuretics, those with the lowest level of potassium in their blood were 3.1 times more likely to have a stroke than those with the highest level (95 per cent confidence interval 1.7–5.8, P<0.0005). For loop diuretics and potassium-sparing diuretics the relative risks were 2.4 (1.2–4.7, P<0.001) and 2.0 (1.1–3.7, P<0.05), respectively.

However, the researchers stress that the results do not imply that diuretics create an excessive risk of stroke. "It is important to note that we are neither implying, nor does our data support, any excess risk for stroke associated with the use of diuretics. The question can be raised, however, as to whether diuretics would be even more effective if potassium levels were maintained at normal levels," they say. Furthermore, they found that stroke risk was associated with lower dietary intake of potassium only in people not taking diuretics. In this group, people with the lowest amount of potassium in their diet were 1.76 times (1.21–2.57, P<0.025) more likely to have a stroke than those with the highest amount of potassium in their diet.

The researchers also examined the small number of diuretic users who also had atrial fibrillation. They found that those with atrial fibrillation and low serum potassium were nearly 10 times more likely to develop a stroke than diuretic users with regular heart rhythms and higher blood potassium levels. They say that, although the numbers were small, this finding is intriguing enough to warrant further study (Neurology 2002;59:314).

In an accompanying editorial, Dr Steven Levine, of the Mount Sinai School of Medicine in New York, and Dr Bruce Coull, of the University of Arizona Health Science Centre, comment that the study suggests diuretics may modulate the risk of stroke associated with potassium levels or intake. "Even a slightly increased risk of low potassium added to established stroke risk factors, such as hypertension, diabetes mellitus, atrial fibrillation and cigarette smoking, could have a large effect, given the extensive use of diuretic drugs and the presumed ease of dietary manipulation of potassium," they say (ibid, p302).