Incidence and Risk Factors for Long-Term Mesh Explantation Due to Infection in More than 100,000 Hernia Operation Patients

Abstract

BACKGROUND: Infectious complications after hernia operation are potentially disastrous, often requiring long-term antibiotic administration, debridement, and mesh explantation. Our objective was to describe the long-term incidence and risk factors for synthetic mesh explantation due to infection after hernia operation in a large cohort. STUDY DESIGN: Retrospective database study using Veterans Affairs Surgical Quality Improvement Program and chart review of veterans undergoing abdominal or groin hernia repair with synthetic mesh implantation during 2008-2015. The main outcome was mesh explantation due to infection within 5 years. RESULTS: The study population consisted of 103,869 hernia operations, of which 74.3% were inguinal, 10.7% umbilical, and 15.0% ventral. Explantation incidence was highest among ventral (1.5%). Median explantation interval overall was 208 days. In multivariable logistic regression, all obesity levels from pre-obesity to obesity class III were associated with higher explantation risk. American Society of Anesthesiology physical status classification of 3 to 5 was associated with odds ratio (OR) of 1.7 (95% CI, 1.28 to 2.26), as was longer operative duration (OR 1.83; 95% CI, 1.51 to 2.21), and contaminated or dirty surgical wound classification (OR 2.27; 95% CI, 1.11 to 4.64). Umbilical repair (OR 6.11; 95% CI, 4.14 to 9.02) and ventral repair (OR 14.35; 95% CI, 10.39 to 19.82) were associated with higher risk compared with inguinal. Open repair was associated with a higher risk compared with laparoscopic (OR 3.57; 95% CI, 2.52 to 5.05). Deep incisional surgical site infection within 30 days of operation was more likely to result in long-term mesh explantation (29.2%) than either superficial (6.4%) or organ space infection (22.4%). CONCLUSIONS: Mesh explantation for infection is most common after ventral hernia repair. Risk factor optimization is crucial to minimize such an end point.