Importance: Sodium-glucose cotransporter-2 inhibitors (SGLT2i) improve outcomes in patients with heart failure and a reduced ejection fraction (HFrEF). The association with cardiac remodeling has not been investigated. Objective: To investigate the outcome of the SGLT2i empagliflozin, compared with placebo, on cardiac remodeling in patients with HFrEF. Design, Setting, and Participants: This exploratory post hoc analysis included participants with stable HFrEF and ejection fractions of 40% or less, who were randomly enrolled in an investigator-initiated, multicenter, double-blind, placebo-controlled randomized clinical trial in Denmark. Enrollment commenced on June 29, 2017, and continued through September 10, 2019, with the last participant follow-up on December 20, 2019. Interventions: Randomization (1:1) to empagliflozin (10 mg once daily) or matching placebo in addition to recommended heart failure therapy for 12 weeks. Main Outcomes and Measures: Efficacy measures were changes from baseline to week 12 in left ventricular end-systolic and end-diastolic volume indexes, left atrial volume index, and left ventricular ejection fraction adjusted for age, sex, type 2 diabetes, and atrial fibrillation. Secondary efficacy measures included changes in left ventricular mass index, global longitudinal strain, and relative wall thickness. Results: A total of 190 patients were randomized (95 each receiving empagliflozin and placebo), with a mean (SD) age of 64 (11) years; 162 were men (85.3%), 97 (51.1%) had ischemic HFrEF, 24 (12.6%) had type 2 diabetes, and the mean (SD) latest recorded left ventricular ejection fraction was 29% (8%). Of the 190, 186 completed the study. Empagliflozin significantly reduced left ventricular end-systolic volume index (-4.3 [95% CI, -8.5 to -0.1] mL/m2; P =.04), left ventricular end-diastolic volume index (-5.5 [95% CI, -10.6 to -0.4] mL/m2; P =.03), and left atrial volume index (-2.5 [95% CI, -4.8 to -0.1] mL/m2; P =.04) compared with placebo at 12 weeks' follow-up, with no change in left ventricular ejection fraction (1.2% [95% CI, -1.2% to 3.6%]; P =.32). These findings were consistent across subgroups. Of secondary efficacy measures, left ventricular mass index was significantly reduced by empagliflozin (-9.0 [95% CI, -17.2 to -0.8] g/m2; P =.03). Conclusions and Relevance: In this small, randomized, short-term study, empagliflozin was associated with modest reductions in left ventricular and left atrial volumes with no association with ejection fraction. Effects beyond 12 weeks of SGLT2i use require further study. Trial Registration: ClinicalTrials.gov Identifier: NCT03198585.