Anatomic rupture location, leukocyte levels and diabetes mellitus as factors influencing in-hospital mortality following percutaneous repair of post-infarction ventricular septal rupture: a single centre study.

Background: Post-infarction ventricular septal rupture (PIVSR) carries high mortality despite therapeutic advances. This study evaluates outcomes and mortality predictors in 22 PIVSR patients treated at H. Adam Malik Hospital, Medan, where percutaneous closure has become preferred due to surgical limitations. Methods: This single-center, retrospective cohort study analyzed 22 consecutive patients with post-infarction ventricular rupture (January 2022-May 2025), stratified by closure eligibility (n = 11 per group). Comparative analyses used independent t-tests (normal data), Mann-Whitney-U tests (non-normal), and Fisher's exact tests (categorical variables). Survival analysis employed Kaplan-Meier curves with log-rank testing. Effect sizes (mean differences, risk differences, odds ratios) are reported with 95% confidence intervals. Results: The non-closure group had higher leukocyte counts [14.9 ± 5.9 vs. 11.0 ± 4.0 × 10⁹/L, mean difference -3.9 (95% CI, -7.6 to -0.1) × 109 /L; P = 0.045], greater diabetes prevalence [54.5% vs. 9.1%; risk difference -45% (-75 to -16); P = 0.032], and shorter pre-closure survival (Mean 6 ± 5 days vs. 9 ± 6 days; P < 0.001). Among closure patients, apical rupture predicted universal mortality [7/7 deaths vs. 0/4 with mid-ventricular ruptures; risk difference 75% (25-100); P = 0.024], while LVEF (P = 0.92) and complexity (P = 1.000) showed no association. Survival favored closure (log-rank P < 0.001). Conclusion: Percutaneous PIVSR closure improved survival, but outcomes depended on anatomic rupture location, with apical rupture exhibiting prevalent mortality in the closure group. Non-closure patients had shorter survival, higher leukocytes, and diabetes. While rupture location and systemic factors influenced results, further research is needed to explain these associations and optimize patient selection.