gms | German Medical Science

Introduction: A special problem in visceral surgery is recurrent anastomotic leak, which frequently requires multiple reoperations and sometimes leads to enteral fistula. Neither molecular genetic nor clinical risk factors for this kind of chronic wound healing disorder are known to date. Aim of this study was to identify the risk factors for recurrent anastomotic leak/intestinal perforation.

Material and methods: During a period of 24 Months 39 patients with reoperation due to anastomotic leak were registered in a prospective database. These Patients were devided into two groups. All patients with one reoperation were allocated to group I (n=18), patients with more than one reoperation to group II (n=21). Maximum number of revision in one patient was 16. Data analysis was conducted with SPSS. A statistical significance was tested using the Krusal-Wallis and Chi-Quadrat Tests. Adjusted analysis using multivariate logistic regression was performed to examine the independent effects of each factor. Statistical significance was considered at P value <0,05.

Results: The overall median age of patients was 62 years (range: 23-83) (Group I - 61 years (range 30-82) und Group II 63 years (range: 23-83)).

Anastomotic leaks were located in the whole gastrointestinal tract. (1x esophagus, 2x stomach, 1x duodenum, 11x small bowel, 20x colon und 4x rectum). Leak localization was similar in both groups. (Group I - 1xStomach, 4xsmall bowel, 11x colon, 2x rectum und Group II - 1x esophagus, 1x stomach, 1x duodenum, 7x small bowel, 9x colon und 2x rectum).

Univariate analysis revealed blood transfusion (p=0,001) and alcohol abuse (p=0,025) as significant risk-factors for recurrent anastomotic leak. Furthermore, patients in group II received a higher number of plasma products, yet statistical significance could not be achieved (p=0,09). Blood transfusion and alcohol abuse proved to be independent risk factors in multivariate analysis.

Conclusion: Postoperativ blood transfusion and alcohol abuse lead to impaired anastomotic healing and increases risk for recurrent anastomotic leak/ intestinal perforation. The molecular biological background is unclear and remains focus of our further Investigation on the prospectively collected tissue samples.