Loop Diuretic Absorption


James Blake, MD, 340 East 64th Street, New York, NY 10021


Objectives. The present study was undertaken to determine if direct blockade of angiotensin II receptors by losartan potassium as compared to ACE inhibition might result in greater tolerability in patients with congestive heart failure in whom ACE inhibition resulted in hyperkalemia and azotemia.

Background. Blockade of angiotensin II receptors by losartan potassium may produce similar benefits in congestive heart failure as ACE inhibition. However, some observations suggest losartan potassium may have different effects on renal function than ACE inhibition.

Methods. Five consecutive patients with severe congestive heart failure were identified in whom treatment with ACE inhibition was complicated by hyperkalemia (K>5.7) and azotemia. In three of these patients losartan potassium was substituted for ACE inhibition while losartan potassium was added to treatment with ACE inhibition in the remaining two. The mean of four serial values of potassium, blood urea nitrogen, and creatinine were compared before and after change in treatment using a two tailed t test.

Results. The addition or substitution of losartan potassium resulted in statistically insignificant reductions in blood urea nitrogen and creatinine and a significant reduction in potassium from 5.7±0.1 to 4.9±0.3, p<0.03. Two patients who had required kayexalate were withdrawn. In all patients, hyperkalemia resolved and did not reoccur.

Conclusions. There appear to be fundamental differences between the effects of losartan potassium and ACE inhibitors on potassium excretion in congestive heart failure patients with mild to moderate renal insufficiency. Losartan potassium may also be associated with less azotemia in such patients. These differences may have important clinical implications, particularly in the subset of patients in whom ACE inhibition is poorly tolerated as a result of hyperkalemia and azotemia.

Congestive heart failure (CHF) is an increasingly common illness with a high mortality. Recently, antagonism of activity of the renin angiotensin system with ACE inhibitors has been documented to reduce mortality and symptoms in patients with CHF.1–4 While such ACE inhibition therapy is generally well tolerated, a minority of patients develop renal insufficiency with associated hyperkalemia, forcing reduction in dosage or withdrawal of ACE inhibition. In clinical trials of ACE inhibitor therapy for CHF, hyperkalemia has been noted in 4%–6% of patients and is more common in patients with underlying renal insufficiency, diabetes mellitus, and advancing age.5,6 As the population ages, patients with CHF in whom ACE inhibition results in hyperkalemia are more frequently seen in clinical practice.

Losartan potassium represents the first oral angiotensin II antagonist approved for treatment of hypertension. Although preliminary, early studies on the use of this agent for CHF indicate favorable hemodynamic changes and short term functional improvements similar to the results seen with ACE inhibitors.7–11 A larger, recently completed trial that compared ACE inhibition with losartan in elderly patients with CHF found that losartan was better tolerated and was associated with a reduction in mortality.12 However, there are few data directly comparing the effects of this newer agent with those of ACE inhibitors on renal function and potassium regulation in patients with CHF. In light of common actions on the renin angiotensin system, such effects are anticipated to be comparable. This assumption may represent an oversimplification, as hyperkalemia (serum K>5.7) occurred in 0.4% of hypertensive patients receiving losartan as opposed to 1%–3% of similar patients treated with ACE inhibitors.5,6,13 Furthermore, whereas ACE inhibitors are associated with decreased potassium excretion, losartan has been accompanied by a significant kaliuresis in normal volunteers.14,15

The observation that losartan therapy may produce less hyperkalemia than ACE inhibition in hypertensive patients prompted us to initiate therapy with losartan in five elderly patients with severely impaired left ventricular (LV) function in whom hyperkalemia and azotemia complicated ACE inhibitor therapy and who did not benefit from alternate hydralazine therapy.


Five consecutive patients form the subject of this report (Table). All five patients were elderly (mean: 73±6 years) with ejection fractions <35% (mean: 27.6±4). All patients had documented CHF for which they were initially treated with full doses of an ACE inhibitor, either captopril (50 mg t.i.d) or enalapril (10 mg b.i.d). Three of the five patients were subsequently titrated to 50 mg of losartan while two were treated with full doses of ACE inhibitors and were titrated to 25 mg of losartan. Both doses of losartan have been associated with a comparable degree of improvement in symptoms, exercise tolerance, and hemodynamics.8–10

 Patient 1Patient 2Patient 3Patient 4Patient 5
AGE (years)6985687569
Ejection fraction (%)2128273230
Diabetes mellitusNoYesYesYesYes
Coronary diseaseBypassMIBypassMI/CathBypass
Peripheral vascular diseaseYesYesYesYesYes
F=female; M=male; MI=myocardial infarction; cath=coronary disease confirmed by cardiac catheterization.

Four of the patients had angiographically documented coronary artery disease; in the remaining patient, such involvement was suspected on the basis of a history of angina and segmental LV wall motion abnormalities documented by echocardiography. All five patients manifested azotemia with serum creatinine levels of 1.6–3.3 mg/dL. Of particular interest, the blood urea nitrogen (BUN) was more than 20 times the creatinine level in all five patients, suggesting an element of prerenal azotemia. While receiving ACE inhibitor treatment, hyperkalemia with peak levels of 5.9–6.4 mEq/dL was present in all five patients; in two, it was sufficiently severe to warrant regular therapy with kayexalate.

Four of the five patients had insulin dependent adult onset diabetes; all five had peripheral vascular disease. In three of the patients this disease was documented by peripheral angiography, which also demonstrated the absence of renal artery stenosis.

Statistical analysis was performed to evaluate the significance of changes in potassium, BUN, and creatinine before and after the initiation of therapy with losartan. The mean of four consecutive measurements for each patient before and after the change in therapy were compared using a paired, two tailed, Student t test with p<0.05 considered to be statistically significant.


Patient 1 was clinically stable on ACE inhibition. As noted in each patient, he was intolerant of even small doses of diuretics (i.e., furosemide 20 mg), which resulted in worsened azotemia and exacerbation of hyperkalemia (Fig. 1). Even in the absence of diuretic therapy, hyperkalemia was of a sufficient degree that he was maintained on 30 g of kayexalate per week.

Figure 1.

In patient 1, ACE inhibitor therapy was associated with hyperkalemia necessitating use of kayexalate; after institution of losartan serum potassium levels normalized with a decrease in azotemia.

When this patient developed a drug rash, ACE inhibition was discontinued. He was intolerant of hydralazine and, therefore, was started on therapy with losartan potassium 50 mg. This change in therapy was associated with a substantial reduction in BUN with a smaller decrease in serum creatinine (Fig. 1). Along with the improvement in the apparent degree of prerenal azotemia, his hyperkalemia completely abated, permitting cessation of kayexalate.

Patient 2 was older with diffuse vascular disease and azotemia. She was free of symptoms of CHF while receiving an ACE inhibitor but, like Patient 1, also required kayexalate twice weekly to continue to receive this agent. When an intercurrent illness exacerbated her azotemia and hyperkalemia resulting in hospitalization, the patient's ACE inhibitor was discontinued. She was intolerant of hydralazine and was started on losartan 25 mg with close inhospital monitoring. Once again there was a substantial reduction in BUN and the degree of prerenal azotemia, and her hyperkalemia abated (Fig. 2) permitting withdrawal of kayexalate. This improvement allowed the addition of furosemide 40 mg once daily which produced a mild increment in BUN, although not to the extent previously noted with ACE inhibition, and caused no increase in serum potassium. Unfortunately, overt CHF subsequently recurred and was treated with reinstitution of therapy with captopril while continuing losartan. This change controlled her CHF, but unlike the situation prior to initiation of losartan, less aggravation of her azotemia, was caused and no elevation of her potassium level.

Patients 3 and 4 are similar to the other patients in that, despite the presence of edema, both were intolerant of diuretics, which precipitated worsening azotemia and hyperkalemia. However, with the initiation of losartan therapy, BUN and the degree of hyperkalemia diminished, which permitted the indicated increases in the diuretic dose, resulting in clinical improvement but offset the initial improvement in BUN. Patient 3 is of interest because, like Patient 2, he had very severe CHF and continues treatment with a combination of full ACE inhibition plus losartan, deriving a benefit similar to the other patients in whom therapy with losartan was substituted.

Patient 5 presented with significant hyperkalemia on ACE inhibition and remained hyperkalemic even after such therapy was withdrawn. Because of persistent symptoms, losartan 50 mg was cautiously introduced with clinical improvement and resolution of hyperkalemia.

For this group of five patients, the slight reductions in BUN (46.4±5.3–45.2±7.2) and creatinine (2.1±0.3–2.0±0.1) following the onset of therapy with losartan were not statistically significant. However, the reduction in serum potassium level with losartan treatment (5.7±0.1–4.9±0.3, p<0.03) was statistically significant. Following the change in therapy, hyperkalemia did not reoccur in any of these patients.


In this small series of five consecutive patients, the addition or substitution of losartan potassium for ACE inhibition resulted in correction of hyperkalemia as well as lessening of the degree of prerenal azotemia. The change in mean serum potassium level with losartan was statistically significant, while the effects on BUN and creatinine were less pronounced and did not achieve statistical significance. However, the effect on these latter parameters may be underestimated as amelioration of hyperkalemia in these patients permitted increased diuretic doses that aggravated the degree of azotemia.

In addition to its statistical significance, the reduction in serum potassium following the addition or substitution of losartan had important clinical implications. In two of the patients, hyperkalemia was severe enough to warrant kayexalate, which was discontinued following the change in therapy to losartan without recurrence of hyperkalemia in either patient. The remaining three patients with less severe hyperkalemia also benefited from the change in therapy with amelioration of hyperkalemia and increased ability to tolerate needed diuresis.

Preliminary data indicate similar hemodynamic and symptomatic benefit with direct angiotensin blockade as with ACE inhibition.7–11 The recently completed Evaluation of Losartan In The Elderly (ELITE) study, comparing ACE inhibition with losartan in elderly patients with mild to moderate CHF demonstrated similar efficacy in reducing symptoms and hospital admissions for CHF, while demonstrating an apparent reduction in short term mortality in patients receiving losartan.12 It is generally believed that CHF results in activation of the renin angiotensin system and that elevation of angiotensin II activity potentiates progression of the heart failure syndrome by producing adverse hemodynamic effects, enhancing the release and activity of norepinephrine, and by the toxic effects angiotensin II activity has on the arteries and the myocardium. In this regard, direct blockade of angiotensin II might be expected to be as or more effective than ACE inhibition in patients with CHF. In fact, tissue based ACE activity and the intracellular enzyme chymase have been identified as potential modes of synthesis of angiotensin II by mechanisms unrelated to intravascular converting enzyme.16–18 ACE inhibition may be relatively ineffective in inhibiting such locally derived angiotensin II that might be more effectively antagonized by direct blockade of angiotensin receptors.16–19

On the other hand, ACE inhibition, not direct angiotensin II blockade, prevents the breakdown of bradykinin, which may be responsible for some of the clinical benefit of ACE inhibition.20,21 Animal models exist in which all or most of the benefit of ACE inhibition on adverse cardiovascular remodeling can be inhibited by direct antagonists of bradykinin, but not by antagonists of angiotensin II.21 In this regard, our observation that the CHF in our sickest two patients was not well managed on losartan alone is of interest. The combination of losartan and ACE inhibition led to excellent control of CHF in these patients while, at the same time, maintaining the improvement in azotemia and the clinically important reduction in potassium level as well as permitting needed diuresis.

The mechanism by which direct antagonism of angiotensin II reduced potassium in our patients is unclear. Kaliuresis has been observed in normal volunteers taking losartan while on a low sodium diet similar to that commonly prescribed for CHF patients.13–15 Although potassium excretion was not measured in our study, the data in volunteers and the trend toward a reduction in prerenal azotemia in our patients suggest that the losartan induced reduction in serum potassium may have been mediated by an improvement in renal excretion. Future studies will be needed to confirm enhanced kaliuresis in patients with CHF treated with losartan and to determine whether this effect is due to a change in systemic hemodynamics, an alteration of intrarenal hemodynamics with more complete inhibition of tissue derived angiotensin II,20,22 or due to an as yet unexplained direct renal effect of losartan similar to its known uricosuric effect.15

The current study is limited by its small size and observational nature. It is interesting, however, that the recently completed ELITE study made similar observations. The primary objective of the ELITE randomized controlled trial was to determine which class of therapy would be better tolerated from the standpoint of renal dysfunction. The investigators found no difference in this primary end point but noted a trend toward a greater frequency of potassium elevation in patients receiving ACE inhibitors (persistent increase >0.5 mmol/L, p = 0.069) and that six subjects receiving ACE inhibitors had to be withdrawn because of hyperkalemia compared to the withdrawal of only two receiving losartan. The observation that the prognostic benefit of losartan in comparison with ACE inhibition in this trial was primarily due to a lower incidence of sudden death, raises the possibility that the difference in potassium homeostasis between these agents might have an effect on cardiac electrical stability. While the small differences in potassium retention rarely required specific intervention in the less ill population studied in ELITE, which excluded subjects with renal insufficiency, these differences may be extremely important in patients, such as those represented in the present study who are more ill and are seen with increasing frequency in clinical practice. Our results suggest that, in such patients for whom standard therapy with an ACE inhibitor and a loop diuretic is complicated by hyperkalemia and prerenal azotemia, treatment with losartan potassium may offer well tolerated, potent therapy, directed against the activity of the renin angiotensin system.