Objective: To explore the clinical application value of vasoactive-inotropic score (VIS) in predicting veno-arterial extracorporeal membrane oxygenation (VA-ECMO) assisted weaning in patients with cardiogenic shock. Methods: This study was a retrospective single-center cohort study, including patients who received VA-ECMO treatment for cardiogenic shock. A total of 495 patients received ECMO treatment, and 150 patients met the criteria and were included in the analysis. The demographic characteristics, underlying diseases, etiology, clinical conditions, APACHE II score, lactic acid level, and VIS score of the patients were collected, and the differences between the two groups were compared. Independent predictors were analyzed by multivariate Logistic regression, and the ROC curve and subgroup analysis were further drawn to evaluate the predictive efficacy of VIS. Results: A total of 150 patients with cardiogenic shock (CS) receiving VA-ECMO were included, of whom 84 (56.0%) were successfully weaned and 66 (44.0%) failed. The proportion of acute myocardial infarction was higher in the failure group, while myocarditis was more common in the success group. The rates of concomitant CRRT and APACHE II scores were higher in the failure group (P 1, P < 0.05). ROC curve analysis demonstrated that the 24-hour VIS reduction rate had the highest predictive value for weaning outcome (AUC = 0.780). A combined prediction model incorporating the 24-hour VIS reduction rate, APACHE II score, VIS, and lactate indicators further improved predictive performance (AUC = 0.820). Subgroup analyses indicated that this prediction model maintained consistent predictive value across different etiologies, CPR statuses, and CRRT conditions. Conclusion: VIS, especially the VIS value at 24 hours after ECMO initiation, is an important index to predict the outcome of VA-ECMO weaning in patients with cardiogenic shock, and can be used as a powerful tool for clinical evaluation of the possibility of successful weaning.