Objective A nomogram prediction model of cardiopulmonary arrest patients was constructed based on nutritional status, sequential organ failure score (SOFA) and Acute Physical and Chronic Health Status scoring System II (APACHE-II). Methods Selected 102 cardiac arrest patients successfully rescued with CPR from January 2022 to December 2024 as the study subjects, Tracking its 28-d prognostic outcome, The 58 deceased patients were included in the death group (n=58), Other 44 patients were included in the survival group (n=44), To the clinical data of patients, SOFA, APACHE-II score assessment and determination of blood amylase, PNI and CAR within 24 hours after successful resuscitation, Comparing the differences between the death group and the surviving group, A logistic regression model was used to analyze the influencing factors of death 28d after resuscitation in cardiopulmonary arrest patients, A nomogram prediction model of death 28d after cardiopulmonary arrest was constructed based on SOFA, APACHE-II score, PNI and CAR level. Results The SOFA, APACHE-II scores, blood amylase and CAR levels were higher than the surviving group, and the PNI index was lower than the surviving group (P <0.05); there was no difference in leukocyte count, platelet count and hemoglobin level among the groups (P> 0.05). By logistic regression analysis, SOFA, APACHE-II score and blood amylase, PNI and CAR levels all affected the death 28d after resuscitation. The nomogram prediction model of death at 28d after cardiopulmonary arrest was constructed based on SOFA, APACHE-II score and the ROC curve and the predicted AUC value of 28d after resuscitation was 0.910(95%CI:0.847-0.973), which had a high predictive value.Conclusion SOFA, APACHE-II score and blood amylase, PNI and CAR levels all had effects on death 28d after cardiopulmonary resuscitation, so the established nomogram prediction model had good predictive efficacy on prognosis after cardiopulmonary resuscitation.