The independent effect of emergency general surgery on outcomes varies depending on case type: A NSQIP outcomes study

Timothy Feeney, Manuel Castillo-Angeles, John W. Scott, Stephanie L. Nitzschke, Ali Salim, Adil H. Haider, Joaquim M. Havens

Research output: Contribution to journalArticlepeer-review

34 Citations (Scopus)

Abstract

Background: Emergency general surgery (EGS) is an independent risk factor for morbidity and mortality, and seven procedures account for 80% of the National burden of operative EGS. We aimed to characterize the excess morbidity and mortality attributable to these procedures based on the level of procedural risk. Methods: Retrospective analysis of the ACS National Surgical Quality Improvement Project (ACS-NSQIP) database. (2005–2014). Seven EGS procedures were stratified as high risk and low risk. Primary outcomes were overall mortality, overall morbidity, major morbidity. Multivariable logistic regression was performed. Results: There were 619,174 patients identified. Comparing EGS to non-EGS in high-risk cases the OR for overall mortality was 1.39(1.33,1.45), overall morbidity 1.07 (0.98, 1.16), and major morbidity 1.15(1.03,1,27). In low-risk cases the OR for overall mortality was 1.03 (0.89, 1.19) overall morbidity 1.35 (1.23, 1.48), and major morbidity 2.18(1.90, 2.50). Conclusions: Using a Nationally representative clinical database we identified significant heterogeneity in the outcomes of EGS depending on procedural risk. Risk stratification and benchmarking strategies need to account for the inherent heterogeneity of EGS.

Original languageEnglish
Pages (from-to)856-862
Number of pages7
JournalAmerican Journal of Surgery
Volume216
Issue number5
DOIs
Publication statusPublished - Nov 2018
Externally publishedYes

Keywords

  • EGS
  • Emergency general surgery
  • NSQIP
  • Outcomes

Fingerprint

Dive into the research topics of 'The independent effect of emergency general surgery on outcomes varies depending on case type: A NSQIP outcomes study'. Together they form a unique fingerprint.

Cite this