Background In severe peritoneal dialysis (PD)-related peritonitis, patients’ response to antibiotic can be poor. We postulated that adjunctive lavage may improve the outcome in severe cases by enhancing the removal of bacteria and inflammatory cells from the peritoneum. Methods Severe PD peritonitis was defined as poor clinical response to empirical cefazolin/ceftazidime and a PD effluent (PDE) leukocyte count > 1,090/mm3 on day 3. Enrolled patients were randomized into either the lavage group ( n = 20) or control group ( n = 20). In the lavage group, continuous lavage by an automated PD machine from day 3 to 5 or 6 was performed, whereas the usual PD schedule was maintained in the control group. The primary outcome was treatment success. Post hoc analysis was also performed to compare the outcome between subgroups with different severity. Results Baseline parameters were similar in the lavage and control groups, including PDE leukocyte count on day 3 (4,871/mm3 vs 4,143/mm3, p = 0.46). Treatment success rates were high in both groups (75% vs 70%, p = 0.72). C-reactive protein (CRP) on day 3 was found to be the only predictor of treatment failure and was used to stratify all patients into tertiles of severity. Whilst a significant decline in treatment success was evident across the tertiles of increasing CRP in the control group (100% vs 85.7% vs 28.6%, p = 0.005), treatment success was relatively maintained in the lavage group (85.7% vs 71.4% vs 66.7%, p = 0.43). Conclusions Adjunctive lavage did not improve the overall outcome, although it may be beneficial for the more severe peritonitis patients who have high CRP.
