S1 Raw data -

Objective
This retrospective study aimed to explore the association and clinical value of sequential organ failure assessment (SOFA) score on the predictors of adverse events in patients with unruptured abdominal aortic aneurysms (AAA).
Methods
A total of 322 patients from Medical Information Mart for Intensive Care IV database were enrolled. Logistic regression was conducted to explore the association between SOFA and primary outcome (need for surgery, NFS). Receiver operating characteristic (ROC) and nomogram analyses were used to assess its performance for predicting NFS. We also explored the association and clinical value of SOFA on secondary outcomes including hospital length of stay (LOS), ICU-LOS, and in-hospital mortality by linear and logistic regression analyses, generalized additive model, ROC, and decision curve analysis.
Results
Totally 291 patients underwent the surgery. High SOFA score significantly correlated with NFS both in crude and adjusted models (all P<0.05). SOFA had a relatively favorable prediction performance on NFS (AUC = 0.701, 95%CI: 0.596–0.802). After adjusting for related diseases, its prediction performance was increased. When SOFA was combined with lactate and gender, the model showed an AUC of 0.888 (95%CI: 0.759–1.000) and 0.3–0.9 prediction possibility. Further, the SOFA also showed significant relationship with hospital-LOS, ICU-LOS, and in-hospital mortality (all P<0.05), and exerted some value in the prediction of 7-day hospital-LOS (AUC = 0.637, 95%CI: 0.575–0.686) and in-hospital mortality (AUC = 0.637, 95%CI: 0.680–0.845).
Conclusions
SOFA score was related to the NFS and can be regarded as a useful indicator for predicting the NFS in patients with AAA.