N. Balasundaram1, R. D. Burgon1, M. F. Assefa1, I. S. Chandra1, J. M. Bath1, T. R. Vogel1 1University Of Missouri, Vascular Surgery, Columbia, MO, USA
Introduction: The Emergency Surgery Score (ESS) has been previously validated as a reliable tool to predict postoperative outcomes in emergency general surgery. The purpose of this study was to assess the performance of the ESS for lower extremity endovascular procedures in non elective setting (neLEE).
Methods: The American College of Surgeons’ National Surgical Quality Improvement Program database was retrospectively analyzed for patients undergoing neLEE between 2015 and 2019. The performance of the ESS in predicting mortality in each procedure was assessed using receiver operating characteristic analyses.
Results: 4583 patients underwent neLEE with median age 68 (+ 12.3 SD), with 1802 females (39.3%). The 30-day mortality rate was 3.4%. The ESS correlated with 30-day mortality (area under the curve -AUC was 0.729). The predictive ability of the ESS decreased with increasing age, with the ESS performing best for patients between 60 and 69 years in age (AUC 0.735) and worst for patients above 80 years (AUC 0.650).
The score gradually and accurately predicted 30-day mortality; 0%, 1.6%, 4.8%, 31.3% and 100% of patients with ESS of 0, 4, 8, 12 and 16 died within 30 days after surgery, respectively. ESS was also utilized to calculate risk of death/discharge to hospice (AUC 0.777), discharge to rehab (AUC 0.638), Renal failure (AUC 0.667), postintervention ventilation requirement (0.680), stroke (0.656) and UTI (0.676). A cumulative frequency table of mortality with ESS was used to partition patients into quartiles of ESS < 4, ESS between 5 and 7, ESS between 8 and 9, and ESS > 10. A Cochran- Armitage test showed linear trend towards increased 30-day mortality among the quartiles with increasing ESS (p<0.001), with Quartile 4 (ESS>10) having 15 times odds of increased 30-day mortality compared to reference quartile 1 (ESS<4) (Table 1)
Conclusion: ESS performance accurately predicts mortality for neLEE procedures. Its use may be useful for preoperative risk stratification and has the potential to be utilized for national benchmarking after lower extremity endovascular procedures
