Objective To investigate the predictive value of extracorporeal membrane oxygenation (ECMO) pre-computer multiple scoring systems in the mortality of patients with cardiogenic shock. Methods A retrospective analysis was performed on 100 patients with cardiogenic shock due to various reasons who were treated with veno-arterial ECMO (VA-ECMO) from July 2020 to July 2022. The patients were followed up for 30 days and divided into a survival group (35 cases) and a death group (65 cases) according to whether they survived 30 days after withdrawal. General clinical data, blood biochemistry data within 24 hours before ECMO, ventilator parameters, past medical history and other data were collected, and sequential organ failure score (SOFA) before VA-ECMO, acute physiology and chronic health evaluation Ⅱ (APACHEⅡ), survival after veno-arterial ECMO (SAVE) score and modified SAVE score were calculated. Blood biochemical indicators and clinical scores related to patient prognosis were screened using two-independent sample t test or Man-Whitney U test. The predictive efficacy of each score on short-term prognosis (30-day post-discharge mortality) was evaluated by receiver operating characteristic curve and area under curve (AUC). Results There were significant differences in APACHEⅡ score, SAVE score and modified SAVE score between two groups (P<0.05). The AUC and its 95%CI of APACHEⅡ score was 0.696 (95%CI 0.592 - 0.801), of SAVE score was 0.617 (95%CI 0.498 - 0.736), and of post SAVE score was 0.664 (95%CI 0.545 - 0.782), respectively. All AUCs were relatively low (<0.75). Conclusion SOFA, APACHEⅡ, SAVE score and modified SAVE score have limited clinical value in the prognosis assessment of ECMO patients, and do not show obvious advantages.