Diuretic-Related Side Effects: Development and Treatment

Domenic A. Sica, MD

In This Article

Hypokalemia and Hyperkalemia

A serum K+ value of ≤3.5 mmol/L, which is the most common definitional criterion for a diagnosis of hypokalemia, is a common finding in patients treated with loop and/or high-dose thiazide diuretics.[12] During the first several days of thiazide-diuretic therapy, plasma K+ falls an average of 0.6 mmol/L (in a dose-dependent manner) in subjects not taking K+ supplements, as compared with a 0.3 mmol/L drop in those taking furosemide.[13] However, it is unusual for serum K+ values to settle <3.0 mmol/L in diuretic-treated outpatients, apart from a high dietary Na+ intake and/or when a long-acting diuretic is being given (as is the case with chlorthalidone). Mechanisms that contribute to the onset of hypokalemia during diuretic use include: increased flow-dependent distal nephron K+ secretion (more commonly observed with a high Na+ intake), a fall in distal tubule luminal chloride (Cl-) metabolic alkalosis, and/or secondary hyperaldosteronism.[14,15]

The cardiac implications of diuretic-induced hypokalemia remain controversial. It would seem logical to infer that arrhythmia-related event rates are connected to the degree of hypokalemia, but this is in no way an unambiguous relationship (at least in an outpatient setting). This theme is confused by several factors including: the inconstant relationship between serum K+ concentrations and total body K+ deficits in the face of diuretic therapy; the fact that in most clinical trials evaluating arrhythmia risk (and/or sudden cardiac death [SCD]), serum K+ values have not been measured frequently enough or under sufficiently standardized conditions to allow for anything more than an educated guess as to the "average" K+ value at the time of an event; that the range of serum K+ values most commonly associated with increased ventricular ectopy is very small typically between 3.0-3.5 mmol/L and finally the issue of whether hypokalemia produced by transcellular shifts of K+ manufactures a similar risk as that generated by a reduced serum K+ on the basis of total body losses.

It has been observed that even mild degrees of diuretic-induced hypokalemia can be coupled with ventricular ectopy.[16,17] For example, the Multiple Risk Factor Intervention Trial[16] observed a significant inverse relationship between the serum K+ concentration and the frequency of premature ventricular contractions (PVCs); however, in this trial patients on chlorthalidone with the greatest decrease in serum K+ levels had the best outcomes. However, this relationship has not been detected in all studies, possibly because of the brief duration of many of these trials.[18,19] For example, in the Medical Research Council (MRC) study, 287/324 patients with mild hypertension underwent ambulatory electrocardiographic (ECG) monitoring. In the short-term (8 weeks), there was no increase in the frequency of PVCs; however, after 24 months of therapy, a significant difference, which correlated with serum K+ concentrations, emerged in the PVC rate (20% [diuretic treated] vs. 9% [placebo]).[19]

The hazards central to diuretic-related hypokalemia are most apparent in patients with left ventricular hypertrophy, congestive heart failure (CHF), and/or myocardial ischemia; particularly when they become acutely ill and have need of hospitalization.[20,21,22,23] As mentioned previously, outpatient forms of diuretic-related hypokalemia are seldom of a severe enough nature to demand urgent attention; however, these mildly lowered serum K+ values create a basis for more significant degrees of hypokalemia when transcellular shifts of K+ are interposed, as occurs during stressful circumstances marked by high endogenous epinephrine levels[24] ; therein lies one of the major at-risk scenarios of diuretic-related hypokalemia.

Despite a sometimes monotonous level of concern about CVR risk (rather than benefit) with diuretic therapy, in part, due to associated electrolyte abnormalities, several clinical trials, including the Systolic Hypertension in the Elderly Program, Swedish Trial in Old Patients with Hypertension, and MRC have shown that low-dose diuretic therapy reduces CVR event rates by 20%-25%.[25,26,27,28] Perhaps the use of lower doses of thiazides or their combination with a K+ -sparing diuretic explains these favorable results as compared with earlier trials, such as the Multiple Risk Factor Intervention Trial, in which higher doses of diuretics were employed (and PVCs were more frequent). In the Systolic Hypertension in the Elderly Program (SHEP),[28] 7.2% of those actively treated developed hypokalemia (serum K+ <3.5 mmol/L at year 1). Those subjects who developed hypokalemia did not secure the treatment benefits on CVR and coronary events as well as stroke recognized in similarly-treated, but normokalemic patients, with isolated systolic hypertension.

Two additional issues are considerations in the milieu of diuretic-related hypokalemia: first, the hemodynamic benefit of normalizing serum K+ [29] and second, the consequences of different doses, combinations of diuretics and/or K+ -sparing diuretics on SCD.[30,31] To the former, K+ supplementation (average increase in serum K+ of 0.56 mmol/L) in hypokalemic (serum K+ values <3.5 mmol/L), diuretic-treated patients have been followed by a 5.5 mm Hg average fall in mean arterial pressure.[29] As to the latter, the risk of SCD among patients receiving combined thiazide and K+ -sparing diuretic therapy has been shown to be lower than that found in patients treated with thiazides alone, with odds ratios for an event increasing significantly as the monotherapy dose of HCTZ increased from 25-100 mg/d.[30] Of note in these studies, the addition of K+ supplements to thiazide therapy had little effect on the risk of SCD, suggesting that other properties of K+ -sparing diuretics (such as decreasing urinary losses of magnesium [Mg2+]) may have been at play.[30]

K+ -sparing diuretics (such as triamterene and amiloride) and aldosterone-receptor antagonists (such as spironolactone and eplerenone) are often used for their ability to conserve K+ when it might otherwise be lost with thiazide and loop diuretic therapy. In certain instances, significant enough K+ retention occurs so as to result in hyperkalemia. Hyperkalemia with K+ -sparing diuretics is usually encountered in patients with an existing reduction in their glomerular filtration rate (when also given K+ supplements or salt substitutes), individuals who develop acute-on-chronic renal failure, those on an angiotensin-converting enzyme (ACE) inhibitor/angiotensin receptor blocker (ARB) and/or a nonsteroidal anti-inflammatory drug, or in other situations that predispose to hyperkalemia, such as metabolic acidosis, hyporeninemic hypoaldosteronism, or heparin therapy (including subcutaneous heparin regimens).[32]