Perforation in colorectal stenting: a meta-analysis and a search for risk factors.

BACKGROUND Recent studies suggest that there is a substantial risk of perforation after colorectal stent placement. OBJECTIVE To identify risk factors for perforation from colonic stenting. DESIGN A meta-analysis of 86 studies published between 2005 and 2011. SETTING Multicenter review. PATIENTS All patients who underwent colorectal stent placement. INTERVENTION Colorectal stent placement. MAIN OUTCOME MEASUREMENTS The occurrence of perforation with subgroup analyses for stent design, stricture etiology, stricture dilation, and concomitant chemotherapy, including the use of bevacizumab. RESULTS A total of 4086 patients underwent colorectal stent placement; perforation occurred in 207. Meta-analysis revealed an overall perforation rate of 7.4%. Of the 9 most frequently used stent types, the WallFlex, the Comvi, and the Niti-S D-type had a higher perforation rate (>10%). A lower perforation rate (<5%) was found for the Hanarostent and the Niti-S covered stent. Stenting benign strictures was associated with a significantly increased perforation rate of 18.4% compared with 7.5% for malignant strictures. Dilation did not increase the risk of perforation: 8.5% versus 8.5% without dilation. The subgroup of post-stent placement dilation had a significantly increased perforation risk of 20.4%. With a perforation rate of 12.5%, bevacizumab-based therapy was identified as a risk factor for perforation, whereas the risk for chemotherapy without bevacizumab was 7.0% and not increased compared with the group without concomitant therapies during stent therapy (9.0%). LIMITATIONS Heterogeneity; a considerable proportion of data is unavailable for subgroup analysis. CONCLUSIONS The perforation rate of colonic stenting is 7.4%. Stent design, benign etiology, and bevacizumab were identified as risk factors for perforation. Intraprocedural stricture dilation and concomitant chemotherapy were not associated with an increased risk of perforation.