Abstract Background Red cell distribution width (RDW) is a routinely measured haematological biomarker. It is uncertain whether RDW remains valuable when added to other prognostic variables in a contemporary population with heart failure and reduced ejection fraction (HFrEF). Purpose To assess the relationship between RDW and clinical characteristics and outcomes in patients with HFrEF. To quantify the effect of omecamtiv mecarbil according to baseline RDW in GALACTIC-HF. Methods We analysed data from GALACTIC-HF, a phase 3, randomised, placebo-controlled trial evaluating the efficacy and safety of omecamtiv mecarbil in patients with HFrEF. We used Cox proportional hazard models to examine the relationship between RDW (quartiles) and clinical outcomes. The incremental value of RDW added to a validated prognostic model (PREDICT-HF, for the composite of HF hospitalisation or cardiovascular CV death), built using standard clinical and laboratory variables, including NT-proBNP, was evaluated using Harrell’s C statistic, integrated discrimination index (IDI) and net reclassification index (NRI). Results Of 8232 randomised patients, 7864 (95.5%) had baseline RDW data (median 15.9%, 25th-75th percentile 14.6-17.5%). Patients with higher RDW were more frequently Black and had a greater prevalence of comorbidities and worse HF status. They were also more likely to have an ischaemic aetiology and elevated NT-proBNP and troponin I. Compared to patients with lower RDW (quartile 1 Q1), the hazard ratios (95% CI) for the composite of HF hospitalisation or CV death were 1.54 (1.37-1.75), 1.99 (1.76-2.25), and 2.71 (2.41-3.06) for Q2 to Q4, respectively (Figure). When added to the PREDICT-HF risk model, RDW at baseline improved Harrell’s C statistic from 0.72 (95%CI 0.70-0.73) to 0.73 (0.71-0.74) at 1 year (P 0.001). IDI and NRI increased by 0.008 (0.004-0.012) and 0.119 (0.082-0.154), respectively (both P0.001). This was also true for additional models for CV death and all-cause death data not shown. Omecamtiv mecarbil consistently reduced the composite of HF hospitalisation or CV death across RDW quartiles. Conclusions RDW, a routinely available and inexpensive biomarker, independently predicts the risk of HF hospitalisation or CV death in HFrEF. In GALACTIC-HF, the benefit of omecamtiv mecarbil was consistent across the range of RDW examined. Figure: Red cell distribution width (RDW) in heart failure. The upper panel illustrates the distribution of RDW, and the lower panel shows the association between RDW and the composite of HF hospitalisation and CV death. CV, cardiovascular; HF, heart failure; HR, hazard ratio.
Yang et al. (Sat,) studied this question.