Background: Gastrointestinal anastomotic complications represent serious events; methods to evaluate anastomotic integrity seem to be suboptimal. Since endoscopic intraoperative anastomotic testing allows direct visualization of anastomosis, complication rates may be theoretically reduced by the use of this technique.
Methods: A prospective study involving 118 consecutive oncologic patients undergoing endoscopically tested gastrointestinal stapled anastomoses was carried out. As controls, 148 historical patients without anastomotic testing were used for comparisons.
Results: In the study group, anastomotic testing revealed 16 defects: 11 (9.3%) air leaks and five (4.3%) bleeding anastomoses. All leaks were oversewn and secured. Bleeding anastomoses were managed under direct visualization, and one non-patent anastomosis was redone. Forty-one (15.4%) postoperative anastomotic complications were observed: eight (3%) bleeding anastomoses, seven (2.6%) stenoses, and 26 (9.8%) clinical leaks. No early dehiscence or bleeding occurred if anastomoses were intraoperatively checked, while these complications were significantly more frequent in non-checked anastomoses (6.1% and 5.4%, respectively). Conversely, late leak and stenosis rates were similar between the two groups.
Conclusion: Endoscopic anastomotic testing was a safe and reliable method to assess integrity of gastrointestinal anastomoses, to correct any defect under direct visualization, and to avoid early complications. However, this method seemed inadequate to predict late anastomotic complications.