Peritonitis influences mortality in peritoneal dialysis patients

J Am Soc Nephrol. 1996 Oct;7(10):2176-82. doi: 10.1681/ASN.V7102176.

Abstract

Mortality remains high in peritoneal dialysis (PD) patients. Known risk factors for mortality include age, diabetes, race, initial albumin level, and cardiovascular disease. Peritonitis is reported to cause death in 1 to 6% of PD patients but has not been well studied as a risk factor for mortality. This study examined 516 adults with a total of 896 yr on PD at one center to determine if peritonitis influenced mortality. Time at risk began on Day 1 of training and ended at death, transplant, or 60 days after transfer to hemodialysis or intermittent peritoneal dialysis. The overall mortality rate was 17.4/100 patient yr. Survival was lower for whites, men, diabetic patients, and older patients. Independent risk factors for mortality (by Cox proportional hazards) were race, diabetes, increased age, and increased peritonitis rate. Use of the Y-set was not associated with decreased mortality. Peritonitis was a risk factor only in whites, nondiabetic patients, and those patients over the age of 60. For every 0.5/yr increase in the peritonitis rate, the risk of death increased 10% in whites, 11% in those patients who were over the age of 60, and 4% for nondiabetic patients. Mortality rates did not decrease over time (1979 to 1995), although peritonitis rates fell significantly (P < 0.001). Rates of Gram-negative and fungal peritonitis showed no trend over time. Peritonitis contributed to 25 of 158 (15.8%) of deaths. Gram-negative/fungal peritonitis accounted for 14 deaths (9.5% of all Gram-negative/fungal episodes) whereas Staphylococcus epidermidis accounted for only 1 death (0.5% of all S. epidermidis episodes) (P < 0.001). Cardiovascular disease was more common in those patients whose deaths were unrelated to peritonitis (P < 0.01), whereas an infectious cause was more common in those patients whose deaths were peritonitis-related (P < 0.001). In this study, peritonitis was a risk factor for death in whites, nondiabetic patients, and older patients. However, the Y-set did not improve survival, perhaps because it does not decrease Gram-negative/fungal peritonitis. To have an impact on survival, efforts are needed to reduce the peritonitis that results from these more serious pathogens.

MeSH terms

  • Adult
  • Aged
  • Cause of Death
  • Female
  • Forecasting
  • Humans
  • Incidence
  • Male
  • Middle Aged
  • Mortality
  • Peritoneal Dialysis* / mortality
  • Peritoneal Dialysis, Continuous Ambulatory
  • Peritonitis / epidemiology
  • Peritonitis / microbiology
  • Peritonitis / mortality*
  • Proportional Hazards Models
  • Survival Analysis