Table 3.
Risk of cardiovascular hospitalizations among patients initiating SGLT2i compared to non-gliflozin therapies, by heart failure subtype
| HFrEF | Unadjusted | IPTW adjusted | IPTW adjusted HR (95% CI) | |
|---|---|---|---|---|
| No. events (IR)a | RD (95% CI) | |||
| SGLT2i | DPP4i | |||
| (n = 2503) | (n = 11,379) | |||
| Follow up, months: median (IQR) | 6.5 (5.1, 12.9) | 7.1 (4.0, 14.3) | ||
| Heart failure hospitalizations | 704 (52.2) | 5567 (80.9) | − 20.9 (− 16.1, − 25.7) | 0.67 (0.63, 0.72) |
| MI or Stroke hospitalizations | 144 (8.8) | 1240 (12.7) | − 1.6 (− 3.7, − 0.0) | 0.86 (0.75, 0.99) |
| SGLT2i | GLP-1RA | |||
|---|---|---|---|---|
| (n = 3214) | (n = 3737) | |||
| Follow up, median (IQR) | 6.6 (4.3, 14.7) | 7.3 (4.1, 15.1) | ||
| Heart failure hospitalizations | 893 (49.4) | 1316 (60.6) | − 5.5 (− 10.0, − 1.0) | 0.86 (0.79, 0.93) |
| MI or Stroke hospitalizations | 180 (8.4) | 282 (10.2) | 0.6 (− 1.4, 2.5) | 1.02 (0.85, 1.22) |
| HFpEF | Unadjusted | IPTW adjusted | IPTW adjusted HR (95% CI) | |
|---|---|---|---|---|
| No. events (IR)a | RD (95% CI) | |||
| SGLT2i | DPP4i | |||
| (n = 2846) | (n = 14,647) | |||
| Follow up, median (IQR) | 6.1 (4.0, 11.2) | 6.6 (4.0, 12.5) | ||
| Heart failure hospitalizations | 804 (51.2) | 7132 (80.5) | − 23.0 (− 14.5, − 31.4) | 0.65 (0.61, 0.69) |
| MI or Stroke hospitalizations | 167 (8.8) | 1624 (12.7) | − 1.0 (− 3.0, 1.1) | 0.90 (0.79, 1.02) |
| SGLT2i | GLP-1RA | |||
|---|---|---|---|---|
| (n = 3578) | (n = 5475) | |||
| Follow up, median (IQR) | 6.5 (4.1, 12.8) | 7.1 (4.1, 13.6) | ||
| Heart failure hospitalizations | 1059 (53.2) | 2004 (60.2) | − 4.7 (− 8.6, − 0.7) | 0.89 (0.83, 0.96) |
| MI or Stroke hospitalizations | 213 (8.8) | 443 (10.3) | − 0.4 (− 2.1, 1.3) | 0.97 (0.83, 1.14) |
HFrEF: Heart failure with reduced ejection fraction; HFpEF: Heart failure with preserved ejection fraction; 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; IQR: Interquartile range; RD: Rate difference
aRepresent the unadjusted number of events and incidence rates per 100 person-years of follow up
bHazard ratios were adjusted for variables described in Tables 1 and 2 using stabilized inverse probability of treatment weighting