Predicting 30-day postoperative mortality for emergent anterior abdominal wall hernia repairs using the American College of Surgeons National Surgical Quality Improvement Program database

Abstract

Purpose

Anterior abdominal wall hernias are among the most commonly encountered surgical disease. We sought to identify risk factors that are associated with 30-day postoperative mortality following emergent abdominal wall hernia repair using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database.

Methods

A retrospective analysis of data from the ACS NSQIP from 2005 to 2010 was performed. Patients were selected using Current Procedural Terminology (CPT) and International Classification of Disease 9 Clinical Modification (ICD9) codes for the repair of inguinal, femoral, umbilical, epigastric, ventral, or incisional hernias that were incarcerated, obstructed, strangulated, or gangrenous. Only emergent cases occurring within two days of admission and admitted as inpatients were included. Univariate and multivariable analysis was performed. A risk score was also created.

Results

There were 4298 cases of emergent anterior abdominal wall hernia surgery. The most common was inguinal (25.3 %), followed by incisional (23.8 %), umbilical (23.5 %), ventral (12.1 %), femoral (8.8 %), and epigastric (6.5 %) hernias. Multivariable analysis demonstrated six statistically significant predictors of short-term mortality, including history of congestive heart failure (CHF) [odds ratio (OR) 8.24, 95 % confidence interval (CI) 4.05–16.75), age (OR 5.52, 95 % CI 3.48–8.77), history of peripheral vascular disease (PVD) (OR 4.98, 95 % CI 2.08–11.92), presence of ascites (OR 3.16, 95 % CI 1.64–6.08), preoperative blood urea nitrogen (OR 1.35, 95 % CI 1.22–1.49), and preoperative white blood cell count (OR 1.22, 95 % CI 1.02–1.45). The C-statistic for the risk model was 0.858.

Conclusion

We present a large study on short-term mortality following emergent anterior abdominal wall hernia repairs based on the ACS NSQIP with a derived risk model that demonstrates excellent discriminative ability.

http://ift.tt/2cx9dAg