Recurrent event analyses in patients receiving maintenance dialysis and the effect of higher- versus lower-dose chronic intravenous iron therapy: A report focusing on heart failure events from the PIVOTAL trial.

Abstract

AIMS In the PIVOTAL trial, a proactive high-dose regimen of intravenous iron sucrose, compared to a lower-dose reactive regimen, reduced the risk of first and recurrent events of the primary endpoint in haemodialysis patients. We present the various approaches of recurrent event analyses for the primary endpoint and for the composite of cardiovascular (CV) death or heart failure hospitalization and their non-fatal components. METHODS AND RESULTS Patients were randomized to a proactive maintenance dose of iron sucrose (average 264 mg/month) or a reactive treatment regimen (average 145 mg/month). We compared the results of time-to-first event analyses with recurrent event analysis using the negative binomial model and methods proposed by Wei-Lin-Weissfeld, Andersen and Gill, Lin, Wei, Yang and Ying [LWYY], Mao and Lin, and Rondeau and colleagues. The 2141 haemodialysis patients were followed for a median of 2.1 years and experienced 936 primary recurrent events, which is 42% higher than the number of 658 first events. Proactive regimen patients had 429 primary events (19.4/100 patient-years) compared with 507 events in the reactive regimen patients (24.6/100 patient-years) (rate ratio 0.77, 95% confidence interval [CI] 0.66-0.92, p = 0.0027, LWYY). Recurrent events were also reduced in the proactive regimen for the composite of CV mortality and heart failure hospitalizations (rate ratio 0.73, 95% CI 0.56-0.93, p = 0.013). Recurrent event analyses based on other approaches were very similar to those given above based on the method of LWYY. CONCLUSION A higher dose of chronic intravenous iron therapy compared to a lower dose substantially reduced the total burden of important recurrent events of death and CV disease in patients with end-stage kidney disease receiving maintenance haemodialysis therapy.