Objective To analyze the value of ultrasonic spot tracking imaging (STI) combined with serum creatine kinase (CK)/creatine kinase isoenzyme (CK-MB) ratio in evaluating the early withdrawal of extracorporeal membrane oxygenerator (ECMO) in acute myocardial infarction (AMI).Methods A total of 82 patients with AMI who received ECMO treatment in Qingdao Municipal Hospital of Shandong Province from June 2022 to July 2023 were selected. Color Doppler ultrasonography was used to detect the global outer myocardium circumferential strain value (GCS-epi), middle myocardium circumferential strain value (GCS-mid) and inner myocardium circumferential strain value (GCS-endo) of the left ventricle. The levels of CK and CK-MB were detected by ELISA, and the ratio of CK/CK-MB was calculated. The patients were divided into failure group and successful group according to the withdrawal status. The factors influencing the early withdrawal failure of ECMO were analyzed, and the value of STI combined with serum CK/CK-MB ratio in predicting the early withdrawal result of ECMO was also analyzed.Results The incidence of multiple organ failure in the failure group was higher than that in the successful group (P < 0.05). The CK/CK-MB ratio and LVEF level in the failed group were lower than those in the successful group (P < 0.05), while the levels of GDS-ENDO, GDS-EPI and GDS-MID were higher than those in the successful group (P < 0.05). After multivariate Logistic regression analysis, the results showed that: multiple organ failure [O^R = 5.114 (95% CI: 1.748, 14.962) ], CK/CK-MB ratio [O^R = 3.618 (95% CI: 1.237, 10.586) ], LVEF level [O^R = 6.135 (95% CI: 2.097, 17.949) ], GDS-EPI level [O^R =7.501 (95% CI: 2.564, 21.945) ], GDS-MID level [O^R = 8.183 (95% CI: 2.797, 23.940) ] and GPS-ENDO level [O^R = 15.565 (95% CI: 1.841, 29.289) ] were risk factors for early ECMO withdrawal failure in AMI patients (P < 0.05). ROC curve analysis showed that the sensitivity of multiple organ failure, CK/CK-MB ratio, LVEF, GCS-epi, GCS-mid and GCS-endo in predicting early ECMO withdrawal failure was 92.31% (95% CI: 0.803, 1.000); The specificity was 94.64% (95% CI: 0.824, 1.000), AUC was 0.924. They combined predicted the early withdrawal failure of ECMO.Conclusion Multiple organ failure, CK/CK-MB ratio, LVEF, GCS-epi, GCS-mid, and GCS-endo have higher value in predicting early ECMO withdrawal failure in AMI patients.