Left ventricular mass predicts cardiac reverse remodeling in patients treated with empagliflozin: an exploratory sub-analysis of the empa-heart cardiolink-6 randomized controlled trial

BackgroundLeft ventricular (LV) hypertrophy is associated with an elevated risk for cardiovascular (CV) disease events and all-cause mortality. The CV benefits of sodium-glucose transport protein 2 inhibitors (SGLT2i) have been attributed, in part, to cardiac reverse remodeling. The EMPA-HEART CardioLink-6 study reported that SGLT2 inhibition with empagliflozin was associated with a significant reduction in LV mass indexed to body surface area (LVMi) after 6 months. In this exploratory sub-analysis of the same trial, we evaluated how baseline LVMi may influence cardiac reverse remodeling with empagliflozin.Methods and ResultsA total of 97 patients with type 2 diabetes and coronary artery disease were randomized to empagliflozin (10 mg/d) or matching placebo for 6 months. To determine if the impact of empagliflozin on the 6-month change in LVMi was dependent on baseline LVMi, the EMPA-HEART CardioLink-6 cohort was stratified into two groups – baseline LVMi ≥60 g/m2 and < 60 g/m2. Between-group comparisons were conducted using a linear regression model adjusted for baseline differences in LVMi (ANCOVA) that included an interaction term between baseline LVMi sub-group and treatment. The effect of empagliflozin on 6-month change in LVMi was significantly different between patients with a baseline LVMi ≥60 g/m2 and those whose LVMi was < 60 g/m2 (Pinteraction=0.0064). The adjusted difference between those randomized to empagliflozin and those assigned placebo was -0.46 g/m2 (95% CI: -3.44 g/m2, 2.52 g/m2, P=0.76) and -7.26 g/m2 (95% CI: -11.40 g/m2, -3.12 g/m2, P=0.0011) in the LVMi < 60 g/m2 and LVMi ≥60 g/m2 subgroups, respectively. These associations persisted following multivariate adjustment for baseline characteristics with adjusted differences of 0.59 g/m2 (95% CI: -3.01 g/m2, 4.19 g/m2, P=0.74) in the LVMi < 60 g/m2 group and -7.03 g/m2 (95% CI: -11.06 g/m2, -2.99 g/m2, P=0.001) in the LVMi ≥60 g/m2 group (Pinteraction=0.0054). No significant associations were found between baseline LVMi and 6-month change in LV end systolic volume-indexed (Pinteraction=0.086), LV end diastolic volume-indexed (Pinteraction=0.34), or LV ejection fraction (Pinteraction=0.15).ConclusionPatients with larger LVMi at baseline experienced substantially greater cardiac reverse remodeling benefits with empagliflozin. The effect baseline LV mass has on the benefits derivable from SGLT2i therapies warrants further investigation. BackgroundLeft ventricular (LV) hypertrophy is associated with an elevated risk for cardiovascular (CV) disease events and all-cause mortality. The CV benefits of sodium-glucose transport protein 2 inhibitors (SGLT2i) have been attributed, in part, to cardiac reverse remodeling. The EMPA-HEART CardioLink-6 study reported that SGLT2 inhibition with empagliflozin was associated with a significant reduction in LV mass indexed to body surface area (LVMi) after 6 months. In this exploratory sub-analysis of the same trial, we evaluated how baseline LVMi may influence cardiac reverse remodeling with empagliflozin. Left ventricular (LV) hypertrophy is associated with an elevated risk for cardiovascular (CV) disease events and all-cause mortality. The CV benefits of sodium-glucose transport protein 2 inhibitors (SGLT2i) have been attributed, in part, to cardiac reverse remodeling. The EMPA-HEART CardioLink-6 study reported that SGLT2 inhibition with empagliflozin was associated with a significant reduction in LV mass indexed to body surface area (LVMi) after 6 months. In this exploratory sub-analysis of the same trial, we evaluated how baseline LVMi may influence cardiac reverse remodeling with empagliflozin. Methods and ResultsA total of 97 patients with type 2 diabetes and coronary artery disease were randomized to empagliflozin (10 mg/d) or matching placebo for 6 months. To determine if the impact of empagliflozin on the 6-month change in LVMi was dependent on baseline LVMi, the EMPA-HEART CardioLink-6 cohort was stratified into two groups – baseline LVMi ≥60 g/m2 and < 60 g/m2. Between-group comparisons were conducted using a linear regression model adjusted for baseline differences in LVMi (ANCOVA) that included an interaction term between baseline LVMi sub-group and treatment. The effect of empagliflozin on 6-month change in LVMi was significantly different between patients with a baseline LVMi ≥60 g/m2 and those whose LVMi was < 60 g/m2 (Pinteraction=0.0064). The adjusted difference between those randomized to empagliflozin and those assigned placebo was -0.46 g/m2 (95% CI: -3.44 g/m2, 2.52 g/m2, P=0.76) and -7.26 g/m2 (95% CI: -11.40 g/m2, -3.12 g/m2, P=0.0011) in the LVMi < 60 g/m2 and LVMi ≥60 g/m2 subgroups, respectively. These associations persisted following multivariate adjustment for baseline characteristics with adjusted differences of 0.59 g/m2 (95% CI: -3.01 g/m2, 4.19 g/m2, P=0.74) in the LVMi < 60 g/m2 group and -7.03 g/m2 (95% CI: -11.06 g/m2, -2.99 g/m2, P=0.001) in the LVMi ≥60 g/m2 group (Pinteraction=0.0054). No significant associations were found between baseline LVMi and 6-month change in LV end systolic volume-indexed (Pinteraction=0.086), LV end diastolic volume-indexed (Pinteraction=0.34), or LV ejection fraction (Pinteraction=0.15). A total of 97 patients with type 2 diabetes and coronary artery disease were randomized to empagliflozin (10 mg/d) or matching placebo for 6 months. To determine if the impact of empagliflozin on the 6-month change in LVMi was dependent on baseline LVMi, the EMPA-HEART CardioLink-6 cohort was stratified into two groups – baseline LVMi ≥60 g/m2 and < 60 g/m2. Between-group comparisons were conducted using a linear regression model adjusted for baseline differences in LVMi (ANCOVA) that included an interaction term between baseline LVMi sub-group and treatment. The effect of empagliflozin on 6-month change in LVMi was significantly different between patients with a baseline LVMi ≥60 g/m2 and those whose LVMi was < 60 g/m2 (Pinteraction=0.0064). The adjusted difference between those randomized to empagliflozin and those assigned placebo was -0.46 g/m2 (95% CI: -3.44 g/m2, 2.52 g/m2, P=0.76) and -7.26 g/m2 (95% CI: -11.40 g/m2, -3.12 g/m2, P=0.0011) in the LVMi < 60 g/m2 and LVMi ≥60 g/m2 subgroups, respectively. These associations persisted following multivariate adjustment for baseline characteristics with adjusted differences of 0.59 g/m2 (95% CI: -3.01 g/m2, 4.19 g/m2, P=0.74) in the LVMi < 60 g/m2 group and -7.03 g/m2 (95% CI: -11.06 g/m2, -2.99 g/m2, P=0.001) in the LVMi ≥60 g/m2 group (Pinteraction=0.0054). No significant associations were found between baseline LVMi and 6-month change in LV end systolic volume-indexed (Pinteraction=0.086), LV end diastolic volume-indexed (Pinteraction=0.34), or LV ejection fraction (Pinteraction=0.15). ConclusionPatients with larger LVMi at baseline experienced substantially greater cardiac reverse remodeling benefits with empagliflozin. The effect baseline LV mass has on the benefits derivable from SGLT2i therapies warrants further investigation. Patients with larger LVMi at baseline experienced substantially greater cardiac reverse remodeling benefits with empagliflozin. The effect baseline LV mass has on the benefits derivable from SGLT2i therapies warrants further investigation.