Enterocutaneous fistula and ventral hernia after absorbable mesh prosthesis closure for trauma: The plain truth

Hypothesis: The severity of abdominal injury is the determining factor for the development of enterocutaneous fistula and ventral hernia after absorbable mesh prosthesis closure (AMPC) for trauma. Methods: We conducted a retrospective analysis of case series that included 140 consecutive trauma patients with AMPC surviving more than 48 hours from October 1,1989, to March 31, 2000, at a Level I trauma center. The days until abdominal wall reconstruction was used as a measure of exposure of the viscera to the mesh. The abdominal trauma index (ATI) was used as the measure of injury severity. Statistical analysis included t test comparisons, logistic regression analysis, and life-table analysis for hernia development. results: Enterocutaneous fistula occurred in 10 patients (7.1%). The ATI (mean, 32.5 ± 23.1) was the only variable independently associated with fistula formation (p = 0.01). The risk of fistula increased by 4% for each 1 unit increase in ATI (95% confidence interval [CI], 1-7%). One hundred seventeen patients (84%) survived to completion of abdominal wall reconstruction over a mean of 18.9 ± 22.5 days and 3.6 ± 1.9 operations. The number of days until abdominal wall reconstruction was the only variable independently associated with ventral hernia development (p < 0.001). The likelihood of fascial closure decreased by 26% (95% CI, 16-44%) per day and the risk of ventral hernia increased by 16% (95% CI, 9-23%) per day. The hernia development rate at 4 years (per life table) was 67% for the total, 13% for patients with delayed fascial closure, and 80% for patients requiring other closure techniques. Conclusion: Although the severity of abdominal injury is the most important factor for fistula formation, the most important factor for ventral hernia development is the duration of AMPC. Daily interventions, such as mesh tightening, may be necessary to limit ventral hernia in these high-risk patients.