Fig. 4

Decision curve analysis for different models with different clinical outcomes. A In-ICU mortality; B In-hospital mortality; C 90-days mortality; D 1-year mortality. Model 1 was adjusted for admission information and severity score, including admission type, age, gender, aortic rupture, SOFA, SAPSII and GCS; Model 2 was adjusted for complication and operation, including sepsis, chronic pulmonary diseases, peripheral vascular diseases, hypertension, renal failure, coagulopathy, fluid and electrolyte disorders, extracorporeal circulation, bypass surgery, ventilation on first day and urine output on first day; Model 3 was adjusted for laboratory indicators, including bicarbonate, creatinine, blood urea nitrogen, hematocrit, hemoglobin, PTT, PT, INR, white blood cell count and platelet count; Model 4 was adjusted for full variables, including admission information, severity score, complication, operation and laboratory indicators above. Compared with other models and crude AG, model 4 showed advantages in net benefit when applied to predict in-ICU mortality, in-hospital mortality, 90-day mortality and 1-year mortality of aortic aneurysm patients that had received open surgery. Abbreviation: AG, anion gap; PTT, partial thromboplastin time; PT, prothrombin time; INR, international normalized ratio; GCS, Glasgow Coma Scale; SOFA, sequential organ failure assessment; SAPSII, simplified acute physiology score II; ICU, intensive care unit