Relationship of peritoneal transport rate and dialysis adequacy with inflammation in peritoneal dialysis patients
Fein, P.A.; Fazil, I.; Rafiq, M.A.; Schloth, T.; Matza, B.; Chattopadhyay, J.; Avram, M.M.
Advances in Peritoneal Dialysis. Conference on Peritoneal Dialysis 22: 2-6
2006
ISSN/ISBN: 1197-8554 PMID: 16983929 Document Number: 598913
Inflammation, dialysis adequacy, and peritoneal transport rate (PTR) influence clinical outcomes in peritoneal dialysis (PD) patients. The present study examined the relationship of C-reactive protein (CRP), a marker of inflammation, to PTR and residual renal function (RRF) in PD patients. We recorded the baseline dialysate-to-plasma creatinine (D/P Cr) of 210 PD patients starting in 1986. In a subgroup of 42 patients, we serially measured high-sensitivity CRP levels and.dialysis adequacy, including weekly Kt/V urea and creatinine clearance (CCr), starting in May 2003. The patients were followed to January 2006. Mean age was 53 +/- 16 (standard deviation) years, and 70% of the patients were African American. Enrollment mean and median CRP levels were 13.53 +/- 20.8 (range: 0.2-95.8) and 7.15 mg/L respectively. Mean weekly residual CCr and Kt/V during follow-up were 7.11 +/- 15.47 L/1.73 m2 and 0.14 +/- 0.30 respectively. The mean enrollment D/P Cr was 0.649 +/- 0.12 (range: 0.429-0.954). Patients with CRP > 10 mg/L had significantly lower weekly residual CCr (0.59 L/1.73 m2 vs. 10.1 L/1.73 m2, p = 0.01), residual Kt/V (0.01 vs. 0.20, p = 0.01), total CCr (56 L/1.73 m2 vs. 62 L/1.73 m2, p= 0.047), and total Kt/V (2.09 vs. 2.49, p = 0.001) than did those with CRP < or = 10 mg/L. Levels of CRP correlated negatively with weekly residual CCr (r = -0.42, p = 0.006), residual Kt/V (r = -0.43, p = 0.006), and total Kt/V (r = -0.44, p = 0.004). Enrollment D/P Cr was inversely correlated with serum albumin (r = -0.24, p = 0.001) and directly correlated with peritoneal protein loss (r = 0.34, p = 0.028). Higher enrollment D/P Cr was associated with lower observed cumulative survival (Kaplan-Meier) in PD patients. However D/P Cr was not an independent predictor of long-term survival in PD patients. Using multivariate Cox regression analysis, and including D/P Cr and residual Kt/V in the model, enrollment CRP was an independent predictor of mortality (relative risk = 1.036, p = 0.018). We conclude that elevated CRP is associated with lower RRF As a predictor of mortality, CRP may be better than RRF and D/P Cr.