A prospective, randomized clinical study was undertaken to determine the effect of standard-dose trimethoprim-sulfamethoxazole combination treatment on serum potassium concentrations in outpatients treated in an ambulatory clinic. Ninety-seven patients were treated with oral antibiotics for a variety of infections. Fifty-one patients treated with trimethoprim-sulfamethoxazole (trimethoprim, 320 mg/day; sulfamethoxazole, 1,600 mg/day) constituted the treatment group, while 46 patients treated with other antibiotics served as controls. Serum potassium, sodium, and chloride concentrations, serum carbon dioxide content, blood urea nitrogen level, serum creatinine level, and serum glucose concentration were measured. The baseline serum potassium concentration in the treatment group was 4.30 +/- (SD) 0.36 mmol/l, and it increased significantly (p < 0.001) to 4.66 +/- 0.45 mmol/l on day 5 of therapy. Subgroup analysis of mean serum potassium concentration on day 5 of therapy failed to detect clinically relevant hyperkalemia. In patients with a serum creatinine level equal to or greater than 1.1 mg/dl (K+, 4.83 +/- 0.48 mmol/l), a nonsignificant difference (p = 0.3) in the potassium concentration was noted on day 5 as compared with patients with a serum creatinine level <1.1 mg/dl (K+, 4.63 +/- 0.44 mmol/l). Although diabetics had a higher serum potassium concentration (K+, 4.91 +/- 0.44 mmol/l) than nondiabetics (K+, 4.61 +/- 0.44 mmol/l), the difference was not statistically significant (p = 0.055). Patients aged >/=50 years (K+, 4.82 +/- 0.59 mmol/l) had a significantly different (p = 0.046) serum potassium concentration on day 5 than patients aged <50 years (K+, 4.55 +/- 0.28 mmol/l). In contrast, the baseline serum potassium concentration in the control group was 4.37 +/- 0.45 mmol/l, and it decreased (p = 0.1) to 4.22 +/- 0.4 mmol/l on 5 days of drug therapy. Trimethoprim-sulfamethoxazole therapy, when used to treat a variety of infections, leads to an increase in serum potassium concentration in most patients. After 5 days of therapy with this drug, the treatment group developed a statistically significant rise in the serum potassium concentration as compared with the control group. However, severe hyperkalemia (K+ >/=5.5 mmol/l) occurred in only 3 patients (6%) treated with trimethoprim-sulfamethoxazole. In addition, none of the subgroups of treated patients developed clinically important hyperkalemia. This suggests that outpatients, in contrast to acquired immunodeficiency syndrome patients and hospitalized patients with mild renal insufficiency, develop severe or life-threatening hyperkalemia less commonly when treated with this antimicrobial regimen. However, outpatients having risk factors which may predispose to the development of hyperkalemia should be carefully monitored when treated with trimethoprim-sulfamethoxazole.