To the editor: Brunner and Frick (1) drew attention to the development of hypokalemia and metabolic alkalosis with high-dose sodium benzylpenicillin therapy; and electrolyte disturbances have been recorded occasionally in patients receiving carbenicillin (2). During the last 2 years 90 patients were treated with carbenicillin (30 g daily) in our hospital; 25 received carbenicillin alone; 29 were treated with carbenicillin and gentamicin and 36 with carbenicillin and cephalothin. In 16 (18%) patients abnormally low levels of serum potassium were observed during therapy with carbenicillin. In 8 of these 16 patients hypokalemia was reported only once during antibiotic therapy; moreover, in