Abstract Introduction While the beneficial impact of enhancing left ventricular ejection fraction (LVEF) is well-documented in patients with heart failure with reduced LVEF, its relevance in cardiac sarcoidosis (CS) remains unclear. Purpose This study aimed to identify predictive factors for LVEF improvement and explore its association with prognosis in patients with CS. Methods A post-hoc analysis of the ILLUMINATE-CS registry, a retrospective observational study across multiple centers, was conducted. CS diagnosis adhered to either the 2016 Japanese Circulation Society guidelines or the 2014 Heart Rhythm Society expert consensus statement. Patients with baseline LVEF ≥50% were excluded in the current analysis. LVEF improvement was quantified as the variance between baseline and follow-up echocardiography. Patients were categorized into tertiles based on LVEF improvement: 1st tertile (≤0%, least improved), 2nd tertile (0–9%), and 3rd tertile (>9%, most improved) groups. The primary outcome was all-cause mortality after follow-up echocardiography. Results Among 188 enrolled patients (mean age 62.1±10.7 years; 42.6% male), 65, 60, and 63 were allocated to the 1st, 2nd, and 3rd tertile groups, respectively. Multivariate linear regression analysis revealed that narrower QRS duration, lower baseline LVEF, and absence of beta-blocker prescription at baseline were independently associated with greater LVEF improvement. Over a median follow-up of 407 days, 26 all-cause deaths occurred. Kaplan–Meier curves exhibited significant difference among the tertile groups (log-rank, P=0.002). Cox proportional hazard analysis, adjusted for possible prognostic factors including LVEF, demonstrated significantly higher mortality in the 1st and 2nd tertile groups compared to the 3rd tertile group (vs. 1st tertile: hazard ratio [HR] 22.14, 95% confidence interval [CI] 2.73–179.50, P=0.004; vs. 2nd tertile: HR 8.17, 95% CI 1.02–65.71, P=0.048). Additionally, treating LVEF improvement as a continuous variable revealed an independent association with increased mortality (LVEF improvement per 5%: HR 0.69, 95% CI 0.54–0.87, P=0.002). Conclusions Improvement LVEF in patients with CS is significantly associated with improved outcomes, highlighting its clinical importance in managing CS.Kaplan–Meier curves for all-cause death