C-reactive protein and mortality in hemodialysis patients: the Dialysis Outcomes and Practice Patterns Study (DOPPS)

Abstract

Background: We examined associations of C-reactive protein (CRP) levels with mortality in Japanese hemodialysis patients and trends in prevalence of CRP measurement at hemodialysis facilities internationally. To assess whether measurement of CRP may influence outcomes, we examined associations of facility prevalence of CRP measurement with mortality.

Methods: CRP measurements were from a cross-section of patients in the international Dialysis Outcomes and Practice Patterns Study (n = 610 facilities, 16,355 patients). Cox proportional hazards models assessed associations of mortality with CRP in Japan, and with a facility's frequency of measuring CRP internationally, (except in the USA and Canada).

Results: From 2002-2004, CRP was measured in 0-19% of patients in each country, except Japan (55%). From 2005-2007, CRP was measured in ≥ 50% of country patients except in Canada (15%) and the USA (2%). After multivariable adjustment, the hazard ratio (HR) of death was 1.6- to 2.4-fold higher (p < 0.05) for various categories of CRP levels >3 mg/l (vs. <1.0 mg/l). Cardiovascular mortality risk was lower in facilities measuring CRP for ≥ 50% of patients (HR = 0.72, p = 0.01) in multivariable-adjusted analyses.

Conclusions: CRP is informative regarding mortality risk beyond that provided by other inflammatory and nutritional markers, with significantly higher risk seen at CRP >3 mg/l. Greater use of CRP may lead to improved patient care as suggested by the association of greater CRP measurement with lower cardiovascular mortality.