On‐line hemodiafiltration reduces serum gentamicin levels much more efficiently than conventional hemodialysis

Abstract

Although gentamicin pharmacokinetics during conventional hemodialysis (C-HD) have been investigated,1, 2 there is no information regarding gentamicin removal during on-line hemodiafiltration (OL-HDF). Here we present the results of a study that was undertaken to clarify this issue, after receiving the approval of the hospital scientific and ethic committees and the consent of patients. In 8 stable chronic hemodialysis patients undergoing OL-HDF (group A), and 6 other patients undergoing C-HD (group B), a single dose of 2 mg/kg gentamicin was administered intravenously over a period of 10 min. None of the patients had received gentamicin during the previous month, none had significant residual renal function, and all had well-functioning vascular accesses. Patient demographics are shown in Table I. Blood samples for gentamicin serum level measurements were drawn before its administration (time 0 = T0), 1 1/2 hr later, immediately before the start of the dialysis session (time 1 = T1), and 3 min after the end of the session (time 2 = T2). Serum was stored at −4°C until measurement. Gentamicin serum levels were measured on an AxSYM-FPIA analyzer (Abbott, Abbott Park, Ill). The dialysis session length was 240 min in all patients. The dialyzer blood flow rate was maintained between 330 and 350 mL/min, while the dialysate flow rate was kept at 700 mL/min in all patients. ON-HDF was performed using high-flux polyester-polymer-alloy (PEPA) dialyzers (FDX18GW, surface area 1.8 m2, Nikkiso, Tokyo, Japan) with a total of 18 L replacement fluid administered post-filter, while C-HD was performed using low-flux polysulfone dialyzers (Fresenius-F8, surface area 1.8 m2, Fresenius Medical Care, Bad Homburg, Germany). BUN and creatinine levels were also measured using conventional biochemical methods at times T1 and T2. Table II shows the results of the study. None of the parameters measured differed significantly between groups A and B, with the exception of gentamicin serum levels at the end of the study sessions. These end-dialysis gentamicin levels were significantly lower in group A patients compared with group B patients (1.2 ± 0.3 ng/ml vs. 2.3 ± 0.5 ng/mL, respectively; p < .001). No correlation was found between BUN or creatinine serum levels changes and the corresponding gentamicin changes during the study sessions. Although our investigation is not a detailed pharmacokinetic study, it clearly indicates that gentamicin is removed much more efficiently during ON-HDF compared with C-HD. This is an expected outcome since ON-HDF is performed with high-flux dialyzers, and the method used (diffusion plus convection) is more efficient than C-HD (diffusion only) in removing middle-sized molecular substances such as aminoglycosides (MW approximately 500 Da).3 It has been reported that high-flux polysulfone dialyzers permit better removal of gentamicin.4 However, it is not clear from our study whether the dialyzer, the method, or both are responsible for this difference. Although blood level monitoring is the gold standard method to adjust gentamicin dose in patients with end-stage renal disease undergoing dialysis, the difference between ON-HDF and C-HD should be kept in mind, especially when gentamicin is administered before dialysis, as has been proposed recently.5 More studies are needed to clarify the cause of lower gentamicin serum levels at the end of ON-HDF compared with C-HD, and to determine practical guidelines for gentamicin administration in patients undergoing OL-HDF.