A prospective look at the current state of open abdomens

Pedro G R Teixeira, Ali Salim, Kenji Inaba, Carlos Brown, Timothy Browder, Daniel Margulies, Demetrios Demetriades

Research output: Contribution to journalArticlepeer-review

78 Scopus citations


The present study examines the current management, closure rate, and complications of open abdomens in trauma patients admitted to an Academic Level I trauma center between May 2004 and April 2007. Variables examined include mechanism, injuries, use of antibiotics and paralytics, type of abdominal closure, days to closure, complications, ICU and hospital length of stay, and mortality. Stepwise logistic regression was performed to identify independent predictors of failed abdominal closure. Of 900 laparotomies, 93 (10%) were left open. Eighty-five (91%) patients survived for closure opportunity. Definitive fascial closure was achieved in 72 (85%) at 3.9 ± 3.7 days (range 1-21 days). Of the remaining 13 patients, seven were closed with biologic material, five by skin grafting, and one had skin-only closure. Entero-atmospheric fistulas occurred in 14 (15%) patients. Two independent risk factors associated with failed abdominal closure were the presence of deep surgical site infection [odds ratio (OR) 17.4; 95% confidence interval (CI) 2.6-115.8, P = 0.003] and intra-abdominal abscess (OR 7.4; 95% CI 1.1-51.0, P = 0.04). In conclusion, open abdomens are commonly necessary after trauma laparotomies. Definitive fascial closure can be achieved in 85 per cent of cases. In conjunction with biologics, closure can be achieved in 93 per cent of cases. Failure to primarily close the abdomen is associated with a significantly higher risk for entero-atmospheric fistula occurrence.

Original languageEnglish (US)
Pages (from-to)891-897
Number of pages7
JournalAmerican Surgeon
Issue number10
StatePublished - Oct 1 2008

ASJC Scopus subject areas

  • Surgery


Dive into the research topics of 'A prospective look at the current state of open abdomens'. Together they form a unique fingerprint.

Cite this