Susceptibility testing of bacteria recovered from patients with peritonitis complicating continuous ambulatory peritoneal dialysis

Abstract

Antagonism of antibiotic activity by peritoneal dialysate has been postulated to be a cause of failure of treatment of peritonitis complicating continuous ambulatory peritoneal dialysis. We evaluated by a case-control study whether unexpected treatment failure could be attributed to such antagonism. Bacteria isolated from 34 patient episodes of peritonitis treated with the same regimen of ciprofloxacin monotherapy were studied. Ciprofloxacin was significantly less active in dialysate than in Iso-Sensitest broth (IB). The median MIC in IB was 0.5 microgram/ml, increasing to 2.0 micrograms/ml for both fresh dialysate (FD) (P = 0.003) and pooled dialysis effluent (PDE) (P = 0.03); the median MBC in IB was 8.0 micrograms/ml, increasing to 128.0 micrograms/ml in FD (P = 0.0002) and 64.0 micrograms/ml in PDE (P = 0.02). However, no significant differences were found in the results for patients suffering unexpected treatment failure (relapse of peritonitis) compared with the results for patients whose infection resolved without sequel. In IB the median MICs for relapsers and nonrelapsers were 1.0 and 0.5 microgram/ml, respectively (P = 0.88); median MBCs were 32.0 and 4.0 micrograms/ml (P = 0.19). In FD median MICs for relapsers and nonrelapsers were 2.0 and 1.0 micrograms/ml (P = 0.06); median MBCs were 128.0 micrograms/ml for both groups (P = 0.84). In PDE the median MICs were 2.0 micrograms/ml for both groups (P = 0.78); median MBCs were 256.0 and 64.0 micrograms/ml (P = 0.17). We therefore found no evidence to suggest that antagonism of antibiotic activity by dialysate is a cause of treatment failure or that conventional methods for laboratory susceptibility testing in peritonitis complicating continuous ambulatory peritoneal dialysis should be abandoned in favor of testing in media containing dialysate.