Trimethoprim-sulfamethoxazole has been used since 1968 and has been associated with major adverse reactions, including skin lesions, thrombocytopenia and leukopenia, and gastrointestinal dysfunction. In the past decade, hyperkalemia has been increasingly recognized as a side effect of this drug combination.

Allapan and colleagues [1] described 80 patients treated with standard-dose therapy (trimethoprim, <320 mg/d; sulfamethoxazole, <1600 mg/d) for various infections other than Pneumocystis carinii pneumonia. They found that hyperkalemia developed and serum creatinine levels increased 4 to 5 days after the start of therapy in many patients. No other electrolyte abnormalities were noted. Trimethoprim-sulfamethoxazole was believed to inhibit distal tubule sodium reabsorption and potassium secretion.

Velazquez and associates [2] and Greenberg and coworkers [3] observed hyperkalemia, hyponatremia, and increased creatinine levels in patients with human immunodeficiency virus infection (HIV) who received high-dose therapy (trimethoprim, 20 mg/kg of body weight per day; sulfamethoxazole, 100 mg/kg per day). These side effects peaked on day 5 [2] or between days 7 and 10 [3]. All patients had normal plasma renin activity and aldosterone levels.

We recently treated a 49-year-old HIV-positive patient (CD4 count, 34 cells/mm³) for P. carinii pneumonia using trimethoprim-sulfamethoxazole (20 mg/kg per day). On admission, his potassium level was 3.8 mmol/L, his sodium level was 136 mmol/L, his blood urea nitrogen level was 8 mg/dL, and his creatinine level was 0.9 mg/dL. On day 12, his potassium level was 6.0 mmol/L, his sodium level was 131 mmol/L, his blood urea nitrogen level was 22 mg/dL, and his creatinine level was 1.0 mg/dL. The patient showed no evidence of acidosis or dehydration. His aldosterone level was 2 µg/dL, and his renin was 1.4 µg/mL per hour after intravenous therapy with furosemide (40 mg). His baseline cortisol level was 7.8 mg/dL. One hour after cosyntropin (0.25 mg) was given intravenously, the cortisol level increased to 10.3 mg/dL. Despite therapy with corticosteroids, electrolyte abnormalities persisted. On day 14, trimethoprim-sulfamethoxazole therapy was discontinued, and pentamidine was started. Two days later, the patient's potassium level was 4.0 mmol/L, and his sodium level was 136 mmol/L.

Seney and colleagues [4] described several electrolyte disorders in HIV-positive patients that were caused by adrenal or pituitary dysfunction, hyporeninemic hypoaldosteronism, or renal sodium wasting. The adrenal and pituitary defects may be subtle and may explain the high incidence (20% to 50%) of electrolyte abnormalities seen in HIV-positive patients [5]. Trimethoprim-sulfamethoxazole may unmask or exacerbate preexisting defects. Patients with HIV infection may also be susceptible to hyperkalemia during trimethoprim-sulfamethoxazole therapy because of an unrecognized defect in the renal or adrenal axis [2]. We believe that if hyperkalemia occurs, an endocrinologic evaluation should be done to predict future overt adrenal insufficiency.