acidogenic effect of the intended ketosis due to the lack of carbohydrate. There are no published studies of the long-term metabolic effects of this kind of diet in any group of individuals.

Substantial numbers of individuals receive treatment with diuretic therapies for medical conditions, primarily high blood pressure, but also congestive heart failure and chronic kidney disease. Thiazide diuretics (e.g., hydrochlorothiazide and chlorthalidone) and loop diuretics (e.g., furosemide) increase urinary potassium excretion, which in some instances leads to overt hypokalemia—that is, a serum potassium concentration of 3.5 mmol/L or less. Accordingly, many individuals on diuretic therapy are given a potassium supplement. In a recently completed trial (Furberg et al., 2002), approximately 8 percent of individuals assigned to low-dose chlorthalidone (12.5 to 25 mg/day) required a potassium supplement. Because of diuretic-induced urinary potassium losses, it is plausible that individuals on diuretic therapy should have an AI greater than 4.7 g (120 mmol)/day. However, available evidence is insufficient to confirm the need for a higher AI in such individuals.

Several relatively common clinical conditions can predispose individuals to hyperkalemia, even at levels of potassium intake that are below the AI. The most common of these conditions are chronic kidney disease, heart failure, and type 1 diabetes, each of which can impair renal excretion of potassium. Angiotensin converting enzyme (ACE) inhibitor drug therapy, which is a recommended therapy for each of these conditions, increases the risk of hyperkalemia (Schoolwerth et al., 2001).

The risk for hyperkalemia during ACE inhibitor therapy increases as kidney function declines. The conclusion from a case series of 33 hypertensive patients in which serum potassium levels were measured before and after ACE inhibitor therapy was that serum potassium levels rarely rose to greater than 5.0 mmol/dL unless the estimated glomerular filtration rate (GFR) was less than 40 mL/minute (Textor et al., 1982). In this study, patients consumed between 2.7 and 3.1 g (70 and 80 mmol)/day of potassium. Whether a higher dietary intake of potassium would precipitate hyperkalemia is uncertain. In a case-control study of 1,818 medical outpatients on ACE inhibitor therapy, severe hyperkalemia (defined as serum potassium