Sodium-glucose cotransporter 2 inhibitors and cardiovascular events among patients with type 2 diabetes and low-to-normal body mass index: a nationwide cohort study

Abstract

Background: Patients with low-to-normal body mass index (BMI; < 25.0 kg/m²) were underrepresented in major randomized controlled trials on sodium-glucose cotransporter 2 (SGLT2) inhibitors for type 2 diabetes. The present study aims to investigate the effectiveness of SGLT2 inhibitors for cardiovascular outcomes among patients with type 2 diabetes and low-to-normal BMI, using finer stratification than previous trials.

Methods: This cohort study with a target trial emulation framework was conducted using insurance claims and health screening records of more than 30 million working-age citizens in Japan acquired from April 1, 2015 to March 31, 2022. 139,783 new users of SGLT2 inhibitors matched to 139,783 users of dipeptidyl protease (DPP) 4 inhibitors with stratification by BMI category (< 20.0, 20.0-22.4, 22.5-24.9, 25.0-29.9, 30.0-34.9, and 35.0 ≤ kg/m²). The primary outcome was a composite of all-cause death, myocardial infarction, stroke, or heart failure. Secondary outcomes were the components of the primary outcome. Cox proportional hazard models were used to compare SGLT2 inhibitors with DPP4 inhibitors in the whole population and subgroups defined by the BMI category.

Results: Among participants, 17.3% (n = 48,377) were female and 31.0% (n = 86,536) had low-to-normal BMI (< 20.0 kg/m², 1.9% [n = 5,350]; 20.0-22.4 kg/m², 8.5% [n = 23,818]; and 22.5-24.9 kg/m², 20.5% [n = 57,368]). Over a median follow-up of 24 months, the primary outcome occurred in 2.9% (n = 8,165) of participants. SGLT2 inhibitors were associated with a decreased incidence of the primary outcome in the whole population (HR [95%CI] = 0.92 [0.89 to 0.96]), but not in patients with low-to-normal BMI (< 20.0 kg/m², HR [95%CI] = 1.08 [0.80 to 1.46]; 20.0-22.4 kg/m², HR [95%CI] = 1.04 [0.90 to 1.20]; and 22.5-24.9 kg/m², HR [95%CI] = 0.92 [0.84 to 1.01]).

Conclusions: The protective effect of SGLT2 inhibitors on cardiovascular events among patients with type 2 diabetes appeared to decrease with lower BMI and was not significant among patients with low-to-normal BMI (< 25.0 kg/m2). These findings suggest the importance of considering BMI when initiating SGLT2 inhibitors.

Fig. 1

Kaplan–Meier curve for the primary outcome among the entire population. Number of events: 3,956/139,783 in SGLT2 inhibitors and 4,209/139,783 in DPP inhibitors (log-rank p = 0.003). DPP4 dipeptidyl peptidase 4, SGLT2 sodium-glucose cotransporter 2

Fig. 2

The effect of SGLT2 inhibitors on the primary outcome by six BMI categories. Hazard ratios were adjusted for age, sex, history of cardiovascular disease diagnosis, BMI, systolic blood pressure, estimated glomerular filtration rate, prescription histories of insulin, statin, and other glucose-lowering agents within the last six months before the matched month, and the duration between the month of entering the base cohort and the month of being propensity-score matched. BMI body mass index, CI confidence interval, DPP4 dipeptidyl peptidase 4, SGLT2 sodium-glucose cotransporter 2

Fig. 3

Secondary analysis: The effect of SGLT2 inhibitors on death, MI, stroke, and HF by BMI of < 22.5 vs. ≥ 22.5 kg/m². Hazard ratios were adjusted for age, sex, history of cardiovascular disease diagnosis, BMI, systolic blood pressure, estimated glomerular filtration rate, prescription histories of insulin, statin, and other glucose-lowering agents within the last six months before the matched month, and the duration between the month of entering the base cohort and the month of being propensity-score matched. The values of p-for-interaction were based on models that additionally included interaction terms between exposure (SGLT2 inhibitors) and a binary variable of BMI < or ≥ 22.5 kg/m.². BMI body mass index, CI confidence interval, DPP4 dipeptidyl peptidase 4, HF heart failure, MI myocardial infarction, SGLT2 sodium-glucose cotransporter 2

Fig. 4

Subgroup analysis: The effect of SGLT2 inhibitors on the primary outcome among subgroups by age, sex, and prevalent use of DPP4 inhibitors. Hazard ratios were adjusted for age, sex, history of cardiovascular disease diagnosis, BMI, systolic blood pressure, estimated glomerular filtration rate, prescription histories of insulin, statin, and other glucose-lowering agents within the last six months before the matched month, and the duration between the month of entering the base cohort and the month of being propensity-score matched. The values of p-for-interaction were based on models that additionally included interaction terms between exposure (SGLT2 inhibitors) and a binary variable of BMI < or ≥ 22.5 kg/m.². BMI, bodymass index, CI confidence interval, DPP4 dipeptidyl peptidase 4, HF heart failure, MI myocardial infarction, SGLT2 sodium-glucose cotransporter 2