Table 5.
Adjusted risk of cardiovascular outcomes among patients initiating SGLT2i compared to non-gliflozin therapies, using alternative definitions for HF subtypes
| HFrEF | SGLT2i versus DPP4i | SGLT2i versus GLP-1RA |
|---|---|---|
| HF alternative definition 1a | ||
| HF hospitalizations | 0.68 (0.64, 0.72) | 0.84 (0.79, 0.91) |
| MI or stroke hospitalizations | 0.92 (0.83, 1.03) | 1.01 (0.87, 1.11) |
| HF alternative definition 2a | ||
| HF hospitalizations | 0.68 (0.64, 0.71) | 0.85 (0.80, 0.90) |
| MI or stroke hospitalizations | 0.91 (0.82, 1.01) | 1.00 (0.87, 1.14) |
| HF alternative definition 3a | ||
| HF hospitalizations | 0.71 (0.66, 0.77) | 0.86 (0.79, 0.94) |
| MI or stroke hospitalizations | 0.88 (0.75, 1.03) | 1.04 (0.86, 1.26) |
| Year ≤ 2016 | ||
| HF hospitalizations | 0.67 (0.59, 0.77) | 0.82 (0.71, 0.94) |
| MI or stroke hospitalizations | 0.70 (0.53, 0.94) | 1.00 (0.75, 1.34) |
| Year > 2016 | ||
| HF hospitalizations | 0.61 (0.55, 0.67) | 0.84 (0.76, 0.92) |
| MI or stroke hospitalizations | 0.93 (0.77, 1.11) | 0.93 (0.76, 1.22) |
| HFpEF | SGLT2i versus DPP4i | SGLT2i versus GLP-1RA |
|---|---|---|
| HF alternative definition 1a | ||
| HF hospitalizations | 0.63 (0.60, 0.67) | 0.85 (0.80, 0.91) |
| MI or stroke hospitalizations | 0.88 (0.79, 0.98) | 0.97 (0.85, 1.10) |
| HF alternative definition 2a | ||
| HF hospitalizations | 0.63 (0.60, 0.66) | 0.83 (0.79, 0.88) |
| MI or stroke hospitalizations | 0.90 (0.81, 0.99) | 0.96 (0.85, 1.08) |
| HF alternative definition 3a | ||
| HF hospitalizations | 0.66 (0.62, 0.72) | 0.88 (0.81, 0.95) |
| MI or stroke hospitalizations | 0.90 (0.78, 1.03) | 0.98 (0.83, 1.17) |
| Year ≤ 2016 | ||
| HF hospitalizations | 0.70 (0.63, 0.79) | 0.91 (0.83, 1.05) |
| MI or stroke hospitalizations | 0.89 (0.71, 1.12) | 1.05 (0.81, 1.35) |
| Year > 2016 | ||
| HF hospitalizations | 0.59 (0.54, 0.64) | 0.84 (0.76, 0.92) |
| MI or stroke hospitalizations | 0.89 (0.75, 1.05) | 0.93 (0.76, 1.13) |
SGLT2i: sodium–glucose cotransporter-2 inhibitors; DPP4i: dipeptidyl peptidase 4 inhibitors; GLP-1RA: Glucagon-like peptide-1 receptor agonists; CI: confidence intervals; IR: incidence rate; HR: hazard ratio
Hazard ratios adjusted for variables described in Table 1 using stabilized inverse probability of treatment weighting
aFor the primary analysis, the cohort was restricted to patients with codes corresponding to HFrEF or HFpEF in either the first or second position in the discharge diagnosis. We conducted sensitivity analysis where we reconstructed the cohort to patients with the relevant diagnoses at any inpatient discharge diagnosis (alternative definition 1), or any inpatient or outpatient diagnosis (alternative definition 2). Finally, in the third sensitivity analysis, we excluded patients with a recent HF hospitalization within 30 days of index (alternative definition 3). For all analysis, no patient could have diagnosis of HFrEF and HFpEF at the same time, and IPTW was recalculated for each cohort