Corynebacterium peritonitis in Australian peritoneal dialysis patients: predictors, treatment and outcomes in 82 cases.

BACKGROUND

Infection due to Corynebacterium species has been reported with increasing frequency over recent decades. The impacts of enhanced laboratory detection together with widespread use of new peritoneal dialysis (PD) connection technology and antimicrobial prophylaxis strategies on Corynebacterium PD-associated peritonitis have not been well studied.

METHODS

We investigated the frequency, predictors, treatment and clinical outcomes of Corynebacterium peritonitis in all Australian adult patients involving 66 centres who were receiving PD between 1 October 2003 and 31 December 2006.

RESULTS

Eighty-two episodes of Corynebacterium peritonitis (2.3% of all peritonitis episodes) occurred in 65 (1.4%) PD patients. Ten (15%) patients experienced more than one episode of Corynebacterium peritonitis and additional organisms were isolated in 12 (15%) episodes of Corynebacterium peritonitis. The incidence of Corynebacterium peritonitis was significantly and independently predicted only by BMI: RR 2.72 (95% CI 1.38-5.36) for the highest tertile BMI compared with the lowest tertile. The overall cure rate with antibiotics alone was 67%, which was similar to that of peritonitis due to other organisms. Vancomycin was the most common antimicrobial agent administered in the initial empiric and subsequent antibiotic regimens, although outcomes were similar regardless of antimicrobial schedule. Corynebacterium peritonitis not infrequently resulted in relapse (18%), repeat peritonitis (15%), hospitalization (70%), catheter removal (21%), permanent haemodialysis transfer (15%) and death (2%). The individuals who had their catheters removed more than 1 week after the onset of Corynebacterium peritonitis had a significantly higher risk of permanent haemodialysis transfer than those who had their catheters removed within 1 week (90% versus 43%, P < 0.05).

CONCLUSIONS

Corynebacterium is an uncommon but significant cause of PD-associated peritonitis. Complete cure with antibiotics alone is possible in the majority of patients, and rates of adverse outcomes are comparable to those seen with peritonitis due to other organisms. Use of vancomycin rather than cephazolin as empiric therapy does not impact outcomes, and a 2-week course of antibiotic therapy appears sufficient. If catheter removal is required, outcomes are improved by removing the catheter within 1 week of peritonitis onset.