A single-center retrospective study used 5 machine-learning classifiers to predict re-suturing versus non-operative management in 129 postpartum perineal wound dehiscence cases (2020–2023) with a 20% hold-out test set. Re-sutured wounds were more severe (deeper/longer/more disrupted) and had higher rectal mucosal involvement (p=0.031) and higher subspecialist involvement (50% vs 0%, p<0.001). The best-performing model was SVM with AUC 0.86 (95% CI 0.81–0.91) and accuracy 0.86; key predictors included wound depth, wound length, pain score, and across models clinician seniority and wound depth were consistently important. The study demonstrates that current re-suturing decisions are predictable from severity features and provider factors, supporting potential decision-support standardization.