The advent of high-efficiency hemodialyzers has afforded improved efficiency of urea clearance; however, increased clearance of other substances, particularly antibiotics, also may occur, necessitating changes in clinical practice. Accordingly, we compared the efficiency of gentamicin removal using two different hemodialyzers, a conventional saponified cellulose ester (CD 135) and a high-efficiency cuprammonium rayon dialyzer (TAF 175L), in eight hospitalized patients undergoing antibiotic therapy for suspected or proven gram-negative infection. The rate of dialysis, estimated as the ratio of dialyzer urea clearance (K) to urea distribution volume (V) (K/V urea), and the total elimination rate constant (k) of gentamicin were measured during 17 hemodialysis treatments. The K/V urea for the two dialyzers, TAF 175L and CD 135, was 0.390 +/- 0.024 hr-1 and 0.413 +/- 0.129 hr-1 (P = NS), respectively. The TAF 175L hemodialyzer was almost twice as efficient in removing gentamicin as the CD 135: TAF 175, k = 0.263 +/- 0.024 hr-1; CD 135, k = 0.132 +/- 0.027 hr-1 (P < 0.001). Moreover, the rate of dialysis (K/V urea) was correlated with k of gentamicin for the TAF 175L dialyzer (r2 = 0.50, P < 0.02) but not for the CD 135 dialyzer. We conclude that dialyzer characteristics and the rate of dialysis (K/V urea) should be taken into consideration when determining the dosage of gentamicin in patients on hemodialysis.