Principal Findings:

The primary outcome of cardiovascular death, hospitalization for heart failure, or urgent heart failure visit occurred in 16.3% of the dapagliflozin group compared with 21.2% of the placebo group (p < 0.001). The primary outcome was the same in prespecified subgroups, including according to diabetes status.

Secondary outcomes:

• Cardiovascular death: 9.6% with dapagliflozin vs. 11.5% with placebo
• Hospitalization for heart failure: 9.7% with dapagliflozin vs. 13.4% with placebo
• Worsening of renal function: 1.2% with dapagliflozin vs. 1.6% with placebo (p = 0.17)

Decline in GFR and association with adverse events:

• Mean change in eGFR between day 0 and 14: -4.2 mL∙min⁻¹∙1.73 m⁻² in the dapagliflozin group vs. -1.1 mL∙min−1∙1.73 m⁻² in the placebo group
• Cardiovascular death or worsening heart failure in patients in the dapagliflozin group experiencing a >10% decline in eGFR compared with ≤10% decline in eGFR: HR 0.73 (95% CI 0.59-0.91)
• Cardiovascular death or worsening heart failure in patients in the placebo group experiencing a >10% decline in eGFR compared with ≤10% decline in eGFR was 1.45 (95% CI, 1.19-1.78) (p for interaction < 0.001)

Results according to baseline risk:

Subjects were risk stratified according to the MAGGIC (Meta-Analysis Global Group in Chronic Heart Failure) score. Dapagliflozin/placebo hazard ratios for the primary outcome according to MAGGIC risk score quintile (p for interaction = 0.71):

• Quintile 1: 7.6% with dapagliflozin vs. 10.2% with placebo
• Quintile 2: 12.9% with dapagliflozin vs. 17.7% with placebo
• Quintile 3: 13.6% with dapagliflozin vs. 20.6% with placebo
• Quintile 4: 21.9% with dapagliflozin vs. 26.8% with placebo
• Quintile 5: 28.7% with dapagliflozin vs. 32.4% with placebo

Dapagliflozin and ventricular arrhythmias:

• Ventricular arrhythmia, resuscitated cardiac arrest, or sudden death: 5.9% with dapagliflozin vs. 7.4% with placebo (p = 0.037)
• Ventricular arrhythmia: 2.1% with dapagliflozin vs. 2.7% with placebo (p = 0.037)
• Resuscitated cardiac arrest: 0.2% with dapagliflozin vs. 0.1% with placebo (p = 0.037)
• Sudden death: 3.9% with dapagliflozin vs. 4.8% with placebo (p = 0.037)

Serial change of high-sensitivity cardiac troponin T (hsTnT) and effect of dapagliflozin:

• The primary outcome was worse with increasing quartiles of baseline hsTnT. The benefit with dapagliflozin was consistent across quartiles of baseline hsTnT (p for interaction = 0.55), although patients in the top quartile had the greatest absolute risk reduction (absolute risk difference, 7.5%; 95% CI, 1.0%-14.0%).

Results according to age:

Dapagliflozin/placebo hazard ratios (HRs) of the primary outcome for various age groups:

• <55 years: HR 0.87 (95% confidence interval [CI] 0.60-1.28)
• 55-64 years: HR 0.71 (95% CI 0.55-0.93)
• 65-74 years: HR 0.76 (95% CI 0.61-0.95)
• ≥75 years: HR 0.68 (95% CI 0.53-0.88) (p for interaction = 0.76)

Results according to diuretic use:

Dapagliflozin/placebo hazard ratios (HRs) of the primary outcome for various age groups:

• No diuretic use: HR 0.57 (95% CI 0.36-0.92)
• Furosemide equivalent dose <40 mg daily: HR 0.83 (95% CI 0.63-1.10)
• Furosemide equivalent dose 40 mg daily: HR 0.77 (95% CI 0.60-0.99)
• Furosemide equivalent dose >40 mg daily: HR 0.78 (95% CI 0.63-0.97) (p for interaction = 0.61)

Dapagliflozin and health status:

Dapagliflozin/placebo HRs of the primary outcome according to Kansas City Cardiomyopathy Questionnaire (KCCQ) health status (tertiles):

• Lowest tertile: HR 0.70 (95% CI 0.57-0.86)
• Middle tertile: HR 0.77 (95% CI 0.61-0.98)
• Highest tertile: HR 0.62 (95% CI 0.46-0.83) (p for heterogeneity = 0.52)

Dapagliflozin vs. placebo was associated with a 2.3-point increase in KCCQ overall summary score from baseline to 8 months (p < 0.0001).

Results according to diabetes status:

Dapagliflozin/placebo HRs of the primary outcome according to diabetes:

• Diabetes: HR 0.75 (95% CI 0.63-0.90)
• No diabetes: HR 0.73 (95% CI 0.60-0.88) (p for interaction = 0.80)

Dapagliflozin and baseline medication use:

• Angiotensin-converting enzyme (ACE) inhibitor/angiotensin-receptor blocker (ARB) target dose <50%: HR 0.78
• ACE inhibitor/ARB target dose ≥50%: HR 0.64 (p for interaction = 0.21)
• Beta-blocker target dose <50%: HR 0.71
• Beta-blocker target dose ≥50%: HR 0.44 (p for interaction = 0.76)
• Mineralocorticoid receptor antagonist (MRA) target dose <50%: HR 0.71
• MRA target dose ≥50%: HR 0.74 (p for interaction = 0.82)

Dapagliflozin and recurrent events (instead of time-to-first events):

• Dapagliflozin vs. placebo, time-to-first event (Cox model): HR 0.75 (p < 0.0001)
• Dapagliflozin vs. placebo, recurrent events (LWYY model): HR 0.75 (p = 0.0002)
• Dapagliflozin vs. placebo, total heart failure hospitalization (joint frailty model): HR 0.71 (p < 0.0001)
• Dapagliflozin vs. placebo, cardiovascular death (joint frailty model): HR 0.81 (p = 0.0282)

Dapagliflozin and outpatient worsening of heart failure:

• Outpatient intensification of heart failure therapy: 11.1% in the dapagliflozin group vs. 14.3% of the placebo group (p = 0.0003)

Results in patients treated with sacubitril/valsartan:

Efficacy of dapagliflozin vs. placebo for the primary outcome according to use of sacubitril/valsartan:

• Taking sacubitril/valsartan: HR 0.75 (95% CI 0.50-1.13)
• Not taking sacubitril/valsartan: HR 0.74 (95% CI 0.65-0.86) (p for interaction = 1.0)

Efficacy of dapagliflozin vs. placebo for the primary outcome according to use of an MRA:

• Taking an MRA: HR 0.74 (95% CI 0.63-0.87)
• Not taking an MRA: HR 0.74 (95% CI 0.57-0.95) (p for interaction = 0.97)

IGFBP-7 and outcomes in heart failure:

• Insulin-like growth factor–binding protein-7 (IGFBP-7) is a potential prognostic marker in heart failure.
• Adjusted hazard ratio for the primary endpoint for tertile 3 vs. tertile 1: 1.48 (95% CI 1.17-1.88). There was no modification of the benefit of dapagliflozin according to IGFBP-7 (p for interaction = 0.34).

Effect of dapagliflozin according to endothelin-1 level:

• Primary outcome for tertile 3 vs. tertile 1 of baseline endothelin-1 level: adjusted hazard ratio (aHR) 1.95 (95% CI 1.53-2.50). Primary outcome for tertile 2 vs. tertile 1 of baseline endothelin-1 level: aHR 1.36 (95% CI 1.06-1.75).
• Estimated glomerular filtration rate (eGFR) slope for tertile 3 of baseline endothelin-1 level: -3.19 (95% CI -3.66 to -2.72); eGFR slope for tertile 2 of baseline endothelin-1 level: -2.08 (95% CI -2.52 to -1.63); eGFR slope for tertile 1 of baseline endothelin-1 level: -2.35 (95% CI -2.79 to -1.91); p = 0.002.
• The efficacy of dapagliflozin in preventing the primary endpoint was consistent regardless of baseline endothelin-1 concentration (p for interaction = 0.47).

Deterioration in estimated glomerular filtration rate (eGFR) <25ml/min/1.73m² (pooled analysis of DELIVER and DAPA-HF): 3.2% experienced a deterioration in eGFR to <25 at least once during follow-up. The median time to deterioration in kidney function to eGFR <25ml/min/1.73m² was 121 days. The risk of the primary outcome was lower with dapagliflozin compared with placebo both among patients who did (HR 0.53, 95% CI 0.33-0.83) and did not (HR 0.78, 95% CI 0.72-0.86) experience deterioration in kidney function below eGFR 25ml/min/1.73m² (p for interaction = 0.17). Rates of drug discontinuation were similar.