Apolipoprotein(a) Phenotype and Lipoprotein(a) Level Predict Peritoneal Dialysis Patient Mortality

Abstract

To examine the associations between lipoprotein(a) [Lp(a)] level, apolipoprotein(a) [apo(a)] phenotype, and patient mortality in peritoneal dialysis (PD) patients. Observational prospective study of prevalent PD patients. Tertiary-care health sciences center. 54 prevalent PD patients were followed prospectively for 24 months. The exposures were Lp(a) level and apo(a) phenotype, designated by the apo(a) isoform size (number of kringle 4 repeats). Outcome was death from any cause. There were 24 deaths in 77.9 patient-years' follow-up. The independent predictors of death in the multivariate survival analysis were age [relative risk (RR) = 1.03, p = 0.23], diabetes (RR = 3.00, p = 0.03), diastolic blood pressure < or = 70 mmHg (RR = 2.94, p = 0.03), serum albumin (RR = 0.87, p < 0.01), and Lp(a) level (RR = 1.004, p < 0.01). There was strong inverse correlation of Lp(a) with apo(a) isoform size (r = -0.62, p < 0.01). With Lp(a) removed from the model, apo(a) isoform size was a significant predictor of death (RR = 0.91, p = 0.0497). Lipoprotein(a) level and apo(a) phenotype are associated with PD patient mortality. Measurement of Lp(a) level and apo(a) phenotype may be useful in clinical practice to identify patients at high risk for cardiovascular disease. Large prospective studies are needed to determine if a reversal of the increase in Lp(a) level associated with renal disease and dialysis is feasible and beneficial in reducing the risk of cardiovascular disease and mortality.