Dapagliflozin and Timing of Prior Heart Failure Hospitalization: A Patient-Level Meta-Analysis of DAPA-HF and DELIVER

Background Patients recently hospitalized for heart failure (HF) are at a higher risk of adverse clinical outcomes, but they may experience a greater absolute and relative benefit from effective therapies than individuals who are considered more “stable.” Objectives The authors examined the effects of dapagliflozin according to the timing of prior HF hospitalization in a patient-level pooled analysis of DAPA-HF (Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure) and DELIVER (Dapagliflozin Evaluation to Improve the Lives of Patients with Preserved Ejection Fraction Heart Failure). Methods A total of 11,007 patients were randomized in DAPA-HF and DELIVER. The primary outcome was the composite of worsening HF or cardiovascular death. Results In total, 12.4% were hospitalized for HF within 3 months of randomization, 14.2% between 3 and 12 months, and 16.8% more than 1 year before randomization, whereas 56.5% had not been hospitalized. The risk of the primary endpoint was inversely associated with time from prior HF hospitalization, and patients with a recent HF hospitalization had the highest risk. Compared with placebo, dapagliflozin reduced the risk of the primary outcome across HF hospitalization category (0-3 months, HR: 0.66 [95% CI: 0.55-0.81]; 3-12 months, HR: 0.73 [95% CI: 0.59-0.90]; >1 year, HR: 0.91 [95% CI: 0.74-1.12]; and no prior hospitalization, HR: 0.83 [95% CI: 0.73-0.94]; Pinteraction = 0.09). The number of patients needed to treat with dapagliflozin to prevent 1 event over the median follow-up of 22 months was 13, 20, 23, and 28, respectively. The beneficial effect was consistent across the range of LVEF regardless of HF hospitalization category. Conclusions The relative benefits of dapagliflozin were consistent across the range of LVEF regardless of the timing of the most recent HF hospitalization with a greater absolute benefit in patients with recent hospitalization.