Breakdown of intestinal repair after laparotomy for trauma: incidence, risk factors, and strategies for prevention.

BACKGROUND

Breakdown of intestinal repair and enteric leakage after trauma laparotomy can have dire consequences. Factors contributing to these failures when stratified according to location of intestinal injury and method of repair were examined.

METHODS

We retrospectively reviewed all intestinal injuries occurring in a recent 2-year time span in adult patients surviving for more than 48 hours at a Level I trauma center. Data included Injury Severity Score, Abdominal Trauma Index score, site (stomach, duodenum, small and large intestine), and type of repair (enterorrhaphy vs. resection and anastomosis). Physiologic parameters within 48 hours of repair were assessed. Nonparametric analysis was used with significance assessed at the 95% confidence interval.

RESULTS

Two hundred twenty-two intestinal repairs in 171 patients were evaluated. All repairs but one were performed at the initial surgery. Eleven (5%) of these failed in 11 patients (6.4%)--four duodenum, four small bowel, and three colon--and were not recognized for an average of 15 days. Breakdown of repair occurred in patients with higher Injury Severity Scores and Abdominal Trauma Index scores (30 vs. 21 and 29 vs. 14, respectively; p < 0.001) and higher intraoperative blood and fluid administration (8.8 vs. 2.2 U and 11.5 vs. 5.1 L, respectively; p < 0.05). This was associated with longer intensive care unit and hospital stays (15.1 vs. 1.9 and 68.4 vs. 10.4 days, respectively; p < 0.001). All small bowel leaks occurred after resection and anastomosis versus enterorrhaphy (p < 0.05). All anastomotic breakdowns (four small bowel, one colon) occurred in the setting of massive blood and fluid administration versus those that did not leak (12.5 vs. 1.7 U and 12.7 vs. 5.8 L, respectively; p < 0.05). Four of 12 duodenal enterorrhaphies failed. All were associated with pancreatic injury versus none without (p < 0.05). The abdominal compartment syndrome occurred in three patients. In each case, breakdown of a small bowel anastomosis occurred.

CONCLUSIONS

(1) Stomach repair and small bowel and large-bowel enterorrhaphy may be safely accomplished in any setting. (2) Associated pancreatic injury is a risk factor for disruption of duodenorrhaphy. (3) In patients with massive blood and fluid administration, delay of bowel anastomoses should be considered. (4) Disruption of small bowel anastomoses is associated with abdominal compartment syndrome.