A common treatment for wounds that penetrate the colon (part of the large intestine) is to attach the colon, from at or above the injury, to a bag outside the body via the abdominal wall (this is called a colostomy or fecal diversion). This diverts feces from the injury, to prevent infection and death. With improved critical care techniques and antibiotic therapy many trauma centers now manage their patients with direct repair of the colon to close the injury. The potential advantages are avoidance of complications of having an opening of the colon in the abdomen wall (stoma) to a bag, the need for another procedure for stoma closure, and the psychological and financial burden of stoma care.
The review authors searched the medical literature and found six controlled studies in which patients were randomized to primary repair or fecal diversion. Results were reported for a total of 705 patients. The two groups sustained significant injuries with the primary repair patients at least as ill as the diverted patients. The studies were reported from 1979 to 2002 and involved increasingly 'high risk' patients. Five were conducted in the United States and one in South Africa. Primary closure was at least as safe as fecal diversion. The number of deaths was similar in both the primary repair (1.94%) and the diverted groups (1.74%). Total complications, total infectious complications, abdominal infections and wound complications all favored primary repair. The studies did not adequately report colostomy closure for trauma-related colostomies, which can itself result in complications and significant illness.
Meta-analysis of currently published randomized controlled trials favors primary repair over fecal diversion for penetrating colon injuries.
Primary repair of penetrating colon injuries is an appealing management option. However, uncertainty about its safety persists.
The objective of this review was to compare morbidity and mortality rates after primary repair to the rates after fecal diversion, in the management of penetrating colon injuries, using a meta-analysis of randomized controlled trials.
We searched the Cochrane Injuries Group Specialised Register, CENTRAL (The Cochrane Library 2008, Issue 3), MEDLINE 1950 to Sept 2008, EMBASE 1980 to Sept 2008, ISI Web of Science: Science Citation Index Expanded (SCI-EXPANDED) 1970 to Sept 2008, Conference Proceedings Citation Index-Science (CPCI-S) 1990 to Sept 2008, and PubMed (searched 26 Sept 2008). The most recent searches were carried out in September 2008.
Studies were included if they were randomized controlled trials comparing the outcomes of primary repair versus fecal diversion in the management of penetrating colon injuries.
Two authors independently extracted the data. Outcomes evaluated from each trial included mortality, total complications, infectious complications, intra-abdominal infections, wound complications and penetrating abdominal trauma index (PATI). We calculated Peto odds ratios (ORs) for combined effect with a 95% confidence interval (95% CI) for each outcome. Heterogeneity was assessed for each outcome, using a chi-squared test.
Six trials involving 705 patients were included. Mortality was not significantly different between groups, which was low in both the primary repair (1.94%) and the diverted groups (1.74%). The Peto OR for mortality was 1.22 (95% CI 0.40 to 3.74). However, the primary repair group experienced a significantly lower rate of complications (Peto OR 0.54; 95% CI 0.39 to 0.76), total infectious complications (Peto OR 0.44; 95% CI 0.17 to 1.1), abdominal infections including dehiscence (Peto OR 0.67; 95% CI 0.35 to 1.3), abdominal infections excluding dehiscence (Peto OR 0.69; 95% CI 0.34 to 1.39), wound complications including dehiscence (Peto OR 0.73; 95% CI 0.38 to 1.39), and wound complications excluding dehiscence (Peto OR 0.67; 95% CI 0.32 to 1.39). Statistical significance favoring primary repair over fecal diversion was achieved for all outcomes related to abdominal infections and wound complications when one study was excluded for both clinical and statistical heterogeneity in the sensitivity analysis.