Heart failure & cardiomyopathiesSecondary analysis
Effects of Dapagliflozin on Heart Failure Hospitalizations According to Severity of Inpatient Course: Insights from DELIVER and DAPA-HF
- Secondary analysis of the DELIVER and DAPA-HF randomised trials
Authors' conclusion
Treatment with dapagliflozin consistently reduced HF hospitalizations regardless of severity of inpatient course or LOS.
Study context
- Population
- Patients with heart failure (HF) across the spectrum of EF (DELIVER and DAPA-HF)1,209 + 799 HF hospitalizations
- Comparison
- Dapagliflozin vs placebo
- Primary endpoint
- Total adjudicated HF hospitalizations by complexity (uncomplicated vs complicated: ICU stay, IV vasoactive therapies, ventilation, mechanical fluid removal or mechanical circulatory support) and by LOS
Primary result
As reported in the abstract
Total HF hospitalizations by severity of inpatient course (subgroups)
DapagliflozinRelative risk (95% CI)
uncomplicated, DELIVER
0.67(0.55–0.82)
uncomplicated, DAPA-HF
0.69(0.54–0.87)
complicated, DELIVER
0.82(0.63–1.06)
complicated, DAPA-HF
0.75(0.58–0.97)
Further findings
HF hospitalizations by length of stay (LOS)
| Comparison | RR (95% CI) | |
|---|---|---|
| Dapagliflozin · LOS <5 days, DELIVER | 0.76 (0.58–0.99) | |
| Dapagliflozin · LOS <5 days, DAPA-HF | 0.58 (0.42–0.80) | |
| Dapagliflozin · LOS ≥5 days, DELIVER | 0.71 (0.58–0.86) | |
| Dapagliflozin · LOS ≥5 days, DAPA-HF | 0.77 (0.62–0.94) | |
| 0.512Relative risk (log scale) |
Proportion of HF hospitalizations complicated
- DELIVER: uncomplicated 854 (71%), complicated 355 (29%)
- DAPA-HF: uncomplicated 453 (57%), complicated 346 (43%)
In-hospital mortality, complicated vs uncomplicated first HF hospitalization
- DELIVER: 16.7% vs. 2.3%
- P < 0.001
- DAPA-HF: 15.1% vs. 3.8%
- P < 0.001
Original publication
Effects of Dapagliflozin on Heart Failure Hospitalizations According to Severity of Inpatient Course: Insights from DELIVER and DAPA-HF. Eur J Heart Fail. Published 8 June 2023. doi:10.1002/ejhf.2912
Summarised from the published abstract; every number is checked against it