DAPA-HF & DEFINE-HF: Dapagliflozin in heart failure with reduced ejection fraction
Bottom line:
Among patients with symptomatic heart failure with reduced ejection fraction (HFrEF), the SGTL2 inhibitor dapagliflozin reduced the risk of a composite of CV death, HF hospitalization or urgent visit for HF requiring IV diuretics vs placebo (NNT 21), death (NNT 44) & HF hospitalizations (NNT 28) at 1.5 years in patients with OR WITHOUT diabetes.
Dapagliflozin also improves quality of life beyond the clinically important difference as early as 12 weeks (NNT 10-14), without any differences in adverse effects (e.g. hypovolemia, kidney injury, severe hypoglycemia).
Patients (n=4744)
From Feb 2017-Aug 2018, 8134 screened -> 4744 randomized
Included
HF with ejection fraction of 40% or less (HF with reduced ejection fraction [HFrEF])
NYHA functional class 2-4
NT-proBNP
>900 pg/mL if AF/atrial flutter
>400 pg/mL if HF hospitalization within 1 year
>600 pg/mL if neither of the above
+/- type 2 diabetes
Stable doses (at least 4 weeks) of standard HF medications (ACEI/ARB/ARNI + beta-blocker as tolerated, + MRA)
Key exclusion criteria: Type 1 diabetes, symptomatic hypotension or SBP <95, eGFR <30
Baseline (average/proportions)
66 y/o
Female 23%
White 70%, Asian 24%, Black 5%
North America 14%
HF characteristics
Etiology: Ischemic (56%), non-ischemic (36%), unknown (8%)
Prior HF hospitalization 47%
NYHA 2 (68%), 3 (32%), 4 (<1%)
LVEF 31%
NT-proBNP ~1400
PMHx: Diabetes 42%, AF 38%
Clinical variables: SBP 122, eGFR 66
Therapies
Diuretic 93%, digitalis 19%
ACEI 56%, ARB 27%, ARNI 11%
Beta-blocker 96%
MRA 71%
ICD 26%, CRT 7%
Antihyperglycemics (% of those with T2DM): Metformin (51%), sulfonylurea (22%), DPP4i (16%), GLP1RA (1%), insulin (27%)
Intervention & Control
Intervention: Dapagliflozin 10 mg daily
Dose reduced to 5 mg/d or temporary discontinuation if acute, unexpected decline in eGFR, volume depletion, or hypotension (or to avoid these)
Control: Matching placebo
Standardized monitoring: Follow-up at 14 days & 60 days (focus on HF/volume assessment, adverse events, & evaluation of renal function & potassium), then q4 months
Outcomes
CV outcomes @ median 18 months
Primary outcome (CV death, HF hospitalization, or urgent visit for HF resulting in IV therapy): Dapagliflozin 16.3% vs placebo 21.2%
Hazard ratio (HR) 0.74 (95% confidence interval 0.65-0.85); Absolute difference 4.9%, NNT 21
CV death: 9.6% vs 11.5%; HR 0.82 (0.69-0.98)
HF hospitalization: 9.7% vs 13.4%; HR 0.70 (0.59-0.83)
Consistent across subgroups (HR 0.75 for type 2 diabetes, HR 0.73 for those without diabetes)
All-cause mortality: Dapagliflozin 11.6% vs placebo 13.9%; HR 0.83 (0.71-0.97)
Quality of life (QoL) @ month 8
Measured using Kansas City Cardiomyopathy Questionnaire [KCCQ] total symptom score, range from 0 [worst] to 100 [best], Minimal clinically-important difference is a 5-point improvement/worsening)
Mean change: +6.1 vs +3.3 (/100); “win ratio” 1.18 (1.11-1.26)
Improvement 5+ points: Dapagliflozin 58.3% vs placebo 50.9%, odds ratio (OR) 1.15 (1.08-1.23)
Deterioration 5+ points: Dapagliflozin 25.3% vs placebo 32.9%, OR 0.84 (0.78-0.90)
Safety @ median 18 months (none statistically significant vs placebo)
Discontinuation due to adverse events: 4.7% vs 4.9%
Volume depletion: 7.5% vs 6.8% (p=0.4); serious in 1.2% vs 1.7%
Worsening renal function (sustained eGFR reduction >50%, ESRD [eGFR <15 >28 days], or death from renal disease): 1.2% vs 1.6%
Amputations: 0.5% in both groups
Fournier’s gangrene: 0 vs <0.1%
Bone fractures: 2.1% in both groups
Major hypoglycemia: 0.2% in both groups
DKA: 0.1% vs 0
Lab changes (difference vs placebo)
Weight -0.9 kg, SBP -1.3 mm Hg
NT-proBNP -303 pg/mL
HbA1c -0.24%
SCr +1.8 umol/L
Internal validity: Low risk of allocation, performance, detection & attrition bias
Computer-generated randomization, stratified by presence of type 2 diabetes
Allocation concealed by interactive voice/web-response system
Participants, clinicians unaware of treatment assignment (blinded)
Blinded adjudication of outcomes
Loss-to-follow-up <0.8%
Analysis of the intention-to-treat (ITT) population
Other trial of dapagliflozin in HFrEF: DEFINE-HF (PMID: 31524498)
Participants (n=263)
Included: HF with LVEF 40% or less, NYHA 2-3, elevated natriuretic peptide (NT-proBNP 400+ pg/mL or BNP 100+ pg/mL), +/- T2DM
Excluded: Type 1 diabetes, HF hospitalization within last 30 days, eGFR <30
Baseline characteristics:
61 y/o, male 73%, white 55%/black 40%
ischemic etiology 53%, prior HF hospitalization ~80%, HF duration 7 years
NYHA 2 (70%), 3 (30%), KCCQ overall summary score 67/100
LVEF 26%
T2DM 62%, AF 40%
SBP 123, HR 72, eGFR 65-70, HbA1c ~7%
Meds: ACEI/ARB 60%/ARNI 30%, BB 95-100%, MRA 60%, ICD 60%, CRT 20-30%
Dapagliflozin 10 mg/d (intervention) vs placebo (control)
Outcomes @ 12 weeks:
QoL measured using KCCQ overall summary score
Mean 3-point improvement with dapagliflozin vs placebo
Improvement of 5+ points: Dapagliflozin 43% vs placebo 33% (NNT 10)
Difference in KCCQ sub-domains of clinical summary score, total symptom score, physical limitation score & QoL score, but not social limitation score
No difference between dapagliflozin vs placebo in:
Change in NT-proBNP
Any adverse effects
Internal validity: Low risk of bias.