Infrequent occurrence of amphotericin B lipid complex-associated nephrotoxicity in various clinical settings at a university hospital: a retrospective study.

BACKGROUND: Lipid-based formulations of amphotericin B (AMB) have been shown to significantly lessen the occurrence of nephrotoxicity associated with the conventional form of AMB. A MEDLINE search of literature published from 1983 to 2002, using the search terms amphotericin B and nephrotoxicity, identified only 1 large, randomized, prospective trial that has tried to compare the nephrotoxicity rates among lipid-based AMB formulations. Using the nephrotoxicity surrogate marker of doubling of serum creatinine (SCr) level, the investigators reported a high rate of AMB lipid complex (ABLC)-associated nephrotoxicity (42.3%). However, enrollment in that study was limited to only febrile neutropenic patients.

OBJECTIVE: This retrospective study estimated the rate of ABLC-associated nephrotoxicity in various clinical settings at a university hospital and compared that rate with previously reported rates of nephrotoxicity.

METHODS: Data from adult neutropenic and nonneutropenic patients receiving ABLC were collected and the degree of nephrotoxicity was determined using 2 definitions: (1) doubling of baseline SCr level using the peak value within the first 7 days, and (2) end-of-therapy doubling of baseline SCr level using the end-of-therapy value.

RESULTS: Data from 33 patients (20 men, 13 women; mean age, 48.6 years) were collected. Using these definitions of ABLC-associated nephrotoxicity, only 2 cases (6.1%) were observed. This rate was significantly below the 42.3% rate reported in the only large published study (95% CI, 1.7-19.6; P < 0.001). The median change in SCr level was 0.1 mg/dL (range, -1.1 to 4.3 mg/dL). Rates of change were higher in patients who died during hospitalization, but the difference was not significant. Use of concomitant nephrotoxic agents did not account for significant changes in SCr level.

CONCLUSIONS: Data from this study suggest that ABLC infrequently causes clinically significant nephrotoxicity. Therefore, when formulary decisions are made in the selection of a drug for use in various clinical settings, earlier data derived from a single study in febrile neutropenic patients that suggested a significantly higher rate of nephrotoxicity should be interpreted cautiously. Larger trials with more diverse patient populations are needed to better characterize institutional rates of ABLC-associated nephrotoxicity and to aid formulary decision makers.