Meta-analysis of the risk factors for clinical anastomotic leakage after resection of rectal cancer in China

Objective: To explore the risk factors for clinical anastomotic leakage after resection of rectal cancer in China. Methods: By meta-analysis we made a comprehensive analysis of the risk factors for clinical anastomotic leakage after resection of rectal cancer based on 19 articles published in China between January 1999 and January 2009. Results: The anastomotic leakage rate was higher in the patients aged 60 years old and above than in those younger, with the combined odds ratio (OR) value being 0.50 (95% CI: 0.33 -0.76) (P<0.01). The incidence rate was higher in the male patients than in the female ones, with the combined OR value being 2.17 (95% CI: 1.38 -3.42) (P<0.01). The incidence rate in the patients with the distance of tumor from the lower margin to anal verge being 7 cm and shorter was higher than that with longer distance, with the combined OR value being 1.79 (95% CI: 1.37-2.35) (P<0.01). The incidence rate in the patients who had received radiotherapy preoperatively was higher than that in those who had not, with the combined OR value of 3.66 (95% CI: 2.19-6.09) (P<0.01). The incidence rate in the patients who had received stapler anastomosis was higher than that in the patients who had received manual anastomosis, with the combined OR value being 0.70 (95% CI: 0.47-1.05), but there was no significant difference between them (P>0.05). The incidence rate was higher in the patients with diabetes mellitus than in the healthy ones, with the combined OR value being 3.16 (95% CI: 2.27-4.39) (P<0.01). The incidence rate was lower in the patients with Dukes A and B stages than in those with Dukes C and D stages, with the combined OR value being 0.61 (95% CI: 0.45-0.83) (P<0.01). The incidence rate in the patients with high malignance degree in clinicopathological types was higher than that with low malignance degree, with the combined OR value being 2.17 (95% CI: 1.38-3.42) (P<0.01). The incidence rate was lower in the patients who had received preventive colostomy than in those who had not, with the combined OR value being 0.39 (95% CI: 0.14-1.05), but there was no significant difference between them (P>0.05). The incidence rate was higher in the patients who had got selective operation than in those who had got emergency operation, with the combined OR value being 0.27 (95% CI: 0.13-0.56). Conclusion: The risk factors of anastomotic leakage after resection of rectal cancer are as follows: 60 years old and above, male patients, diabetes mellitus, preoperative neoadjuvant radiotherapy, the distance of tumor from the lower margin to the anal verge being shorter than 7 cm, Dukes C and D stages, high malignance degree in clinicopathological types, and emergency operation.