High lipoprotein(a) levels and small apolipoprotein(a) size prospectively predict cardiovascular events in dialysis patients

Lipoprotein(a) [Lp(a)] levels are increased in dialysis patients, suggesting that they may play a role in the elevated atherosclerotic cardiovascular disease (ASCVD) risk in this population. Few prospective studies of Lp(a) level, apolipoprotein(a) [apo(a)] size, and ASCVD have been performed in the dialysis population. An inception cohort of 833 incident dialysis patients were followed prospectively. Baseline Lp(a) was measured by apo(a) size-independent ELISA and apo(a) size by Western blot after SDS-agarose gel electrophoresis. A combined prospective nonfatal and fatal ASCVD end point included myocardial infarction, coronary revascularization, cerebrovascular accident, carotid endarterectomy, peripheral revascularization, gangrene, or limb amputation. Survival analyses were performed with adjustment for baseline demographics, comorbid conditions, ASCVD risk factors, albumin, lipids, and C-reactive protein. Median follow-up was 27.4 mo, with 297 ASCVD events, 130 non-ASCVD deaths, and seven losses to follow-up over 1649 person-years. In multivariate Cox regression models, both high Lp(a) concentration (>= 53 nmol/L) and low molecular weight (LMW) apo(a) isoforms (<= 22 Kringle-IV repeats) predicted ASCVD events (relative hazard [RH] = 1.38, P = 0.02; RH = 1.58, P < 0.0005, respectively). In models that included both Lp(a) concentration and apo(a) size, only apo(a) size remained associated with ASCVD. Among those with both LMW apo(a) and Lp(a) level > 123 nmol/L, the relative hazard (RH) of ASCVD was 1.73 (P < 0.0005), compared with high molecular weight apo(a) and Lp(a) level < 123 nmol/L. No interactions by age, race, gender, diabetes, or ASCVD were present. Both LMW apo(a) size and high Lp(a) level predict ASCVD risk in dialysis patients, but the association of ASCVD with LMW isoforms is stronger than the association with high Lp(a) concentration.