Abdominal catastrophe revisited: the risk and outcome of enteric peritoneal contamination

Peritonitis from a visceral source is associated with striking morbidity and mortality in patients treated with peritoneal dialysis (PD). Surgical intervention for both diagnosis and repair is definitive. However, because the antecedents of enteric injury leading to peritonitis are unpredictable, no preventive strategy has been proposed or adopted. The goal of this study was to examine risk factors influencing the occurrence and outcome of anatomically documented peritonitis of enteric origin.

Retrospective chart and database review.

Peritoneal dialysis unit in tertiary-care referral hospital.

330 patients treated with PD for end-stage renal disease between 1988 and 2000.

Prevalence of peritonitis of anatomically documented enteric origin over two consecutive time periods within the study interval: period 1, from 1 January 1988 through 30 June 1996; period 2, from 1 July 1996 through 30 June 2000.

At least 1 episode of peritonitis occurred in 202 of 330 patients during the entire study period of 12.5 years (600.74 patient-years of care). There were 543 episodes of peritonitis. Anatomically documented visceral Injury caused bacterial peritonitis in 41 patients with a total of 63 discrete episodes, an incidence rate of 0.1048 per patient-year. Peritonitis-free survival was compared between the two periods using Kaplan-Meier analysis. The curve representing risk distribution for anatomically documented visceral peritonitis remained constant over the two periods, in contrast to improvements found in all other types of peritonitis, taken as a group (p= 0.044). Logistic regression modeling showed that the only risk factor associated with development of anatomically documented visceral peritonitis was older age. There was no influence of race, sex, time on PD, and underlying disease etiology. 31 deaths were attributed to peritonitis during the study period. The mortality rate from enteric peritonitis due to visceral injury was 46.3% (19/41 cases), compared to 7.5% for all other peritonitis taken as a group (12/161 cases, p < 0.0001).

The experience at University Hospitals of Cleveland suggests that abdominal catastrophe occurs in approximately 10% of all patients treated with PD, and is associated with high mortality, which has not changed over time. Therefore, peritonitis due to spontaneous visceral injury presents a great diagnostic and therapeutic challenge. It is important to develop a research strategy to understand this devastating complication.