CE-452774-3 SODIUM-GLUCOSE CO-TRANSPORTER 2 INHIBITOR USE IN TYPE II DIABETES MELLITUS IS ASSOCIATED WITH A LOWER RATE OF ATRIAL ARRHYTHMIAS IN A REAL-WORLD POPULATION

Abstract

Sodium glucose co-transporter 2 inhibitors (SGLT2i) have been reported, in post hoc analyses of clinical trial populations, to be associated with lower rates of atrial and ventricular arrhythmias. The real-world effect on cardiac arrhythmias is unknown. We sought to determine the effects of SGLT2 inhibitors on cardiac arrhythmias in a real-world population. We conducted a retrospective cohort study to evaluate the effects of SGLT2i on cardiac arrhythmias in South Australia. Using an existing database, 882 patients with type 2 diabetes mellitus (T2DM) on oral diabetic therapy (33.6% females, median age 62.3, IQR 53.9, 68.1), who received SGLT2 inhibitors were identified through their episodes of care at South Australian public hospitals from 2011-2019, a period when SGLT2i use was predominantly limited to the treatment of T2DM. Patients were temporally matched with 3282 controls who did not receive SGLT2i. Baseline characteristics were adjusted using inverse probability treatment weighting. The quality of propensity matching was confirmed using standard difference of the means. The incidence of outcomes was determined from hospital admission ICD-10 data. Cox regression analysis was performed. SGLT2i were associated with a significantly lower rate of atrial arrhythmias (HR 0.17, 95% CI 0.07-0.40, p<0.001) at 2-years follow up. When the analysis was repeated with added standard covariates, the independent relationship between SGLT2i use and atrial arrhythmias remained. The relationship between SGLT2i use and ventricular arrhythmias (HR 0.25, 95% CI 0.06-1.02, p=0.054), or cardiac arrest (HR 0.82, 95% CI 0.19-3.43, p=0.789) did not reach statistical significance, see table 1. When tachyarrhythmias as a group were evaluated, there was a reduction associated with SGLT2i use (HR 0.32, 95% CI 0.13-0.78, p=0.011). Interestingly, all-cause mortality was higher in the SGLT2i use group (HR 2.03, 95% CI 1.56-2.56, p<0.001), despite propensity matching for known patient factors. Although this is a limitation of this non-randomised and retrospective dataset, it does suggest that the ‘sicker’ cohort were indeed receiving SGLT2i and despite this, the cohort on SGLT2i therapy had less atrial arrhythmias observed, adding to the strength of the finding. In this real-world patient cohort, SGLT2i use was associated with a lower incidence of atrial arrhythmias. Prospective randomised trials are underway to determine the effects of SGLT2i as specific AF pharmacotherapy.Tabled 1Table 1. Hazard ratio associated with SGLT2i use, compared to propensity matched group not treated with SGLT2i.Hazard Ratio with SGTL2i use (95% CI)p-valueAtrial arrhythmias0.17 (0.07-0.40)p<0.001Ventricular arrhythmias0.25 (0.06-1.02)p=0.054Cardiac arrest0.82 (0.19-3.43)p=0.789Tachyarrhythmias0.32 (0.13-0.78)p=0.011 Open table in a new tab