October 25, 1993

Predicting the Course of Peritonitis in Patients Receiving Continuous Ambulatory Peritoneal Dialysis

Author Affiliations

From the Ambulatory Care Service and the Renal Section, Veterans Affairs Medical Center and the Department of Medicine, University of New Mexico School of Medicine, Albuquerque.

Arch Intern Med. 1993;153(20):2317-2321. doi:10.1001/archinte.1993.00410200031003

Background:  Peritonitis is a common problem for patients receiving continuous ambulatory peritoneal dialysis. Episodes that do not respond to antibiotics within 96 hours are associated with substantial morbidity and mortality. The purpose of this study was to develop a method for identifying these patients at the time of hospital admission.

Methods:  We reviewed all cases of peritonitis associated with continuous ambulatory peritoneal dialysis that occurred at the Albuquerque (NM) Veterans Affairs Medical Center during a 10-year period. Episodes of peritonitis were randomly assigned to a training set or a validation set. Persistent infections were those lasting more than 96 hours. For training cases, stepwise logistic regression was used to develop a predictive model for persistent infection using information available at the time of hospital admission. The model was then used to assign validation cases to "high-" and "low-risk" categories. The group difference in the proportion of persistent cases was tested by X2 analysis.

Results:  Sixty patients had 120 episodes of peritonitis during the study period. Of 63 episodes assigned to the training set, 26 (41.3%) lasted more than 96 hours (persistent cases) and 37 were cured in 96 hours or less (usual cases). Compared with usual cases, persistent episodes were characterized by a higher age at presentation and a greater decline from preinfection values for hemoglobin and serum potassium, serum urea nitrogen, creatinine, albumin, and calcium. Advanced age and marked declines in serum potassium and albumin levels were identified by logistic regression as independent risk factors for persistent infection. The model identified 28 of 57 validation cases as high risk. Compared with low-risk cases, these episodes were much more likely to be persistent (64.3% vs 24.1%; P=.002) and result in death (32.1% vs 3.4%; P=.005).

Conclusions:  Advanced age and marked declines in serum albumin and potassium levels are poor prognostic signs in peritonitis associated with continuous ambulatory peritoneal dialysis. Patients with these findings should be treated aggressively.(Arch Intern Med. 1993;153:2317-2321)