Introduction: Nearly 1/3 rd of patients undergoing CABG have concurrent left ventricular systolic dysfunction (LVEF<50%). But the perioperative changes in LVEF is not well understood in these patients. Hypothesis: We hypothesized that patients with preoperative LV systolic dysfunction would have an improvement in LVEF post-CABG. Methods: We studied 549 patients with LVEF<50% who underwent CABG as part of the Surgical Treatment for Ischemic Heart Failure (STICH) trial and had pre- and post-operative (at 4 months) LVEF assessment using the same cardiac imaging modality. All imaging tests were interpreted at a core lab. Patients were classified into 3 groups according to change in LVEF: improved (>5% increase), unchanged, and worsened (>5% decrease). Results: Of the 549 patients (mean age 61.4±9.6 years, 13% women), 240 (43.7%) had improved LVEF (from 25.9±8.6% to 39.5±10.5%; p<0.001), 220 (40.1%) had unchanged, and 89 (16.2%) had worsened LVEF (from 34.4±7.8% to 23.6±7.8%; p<0.001) following surgery (Figure). On multivariate logistic regression, improvement in LVEF was inversely associated with baseline LVEF (OR 0.74, p<0.0001) and prior myocardial infarction (OR 0.56, p=0.028), but had a positive association with age (OR 1.10, p=0.045) and concurrent surgical ventricle restoration (OR 1.93, p=0.0004). Most notably, a >5% decrease in LVEF post-CABG was associated with higher long-term mortality (log-rank p=0.019). Conclusions: Almost 50% of patients with LV systolic dysfunction had a significant improvement in LVEF after CABG. However, 16% of patients, who had worsened LVEF perioperatively had a higher long-term mortality risk and may benefit from further investigation.