Dear Editor,

We thank Dr Weng J et al. for their interest in our study [1] and are pleased to address two issues raised in their letter [2].

Concerning the first query on multivariate analysis, the risk factors with p values of less than 0.05 used from the univariate analysis may have had a large effect on outcome. Accordingly with the suggestion, we performed an additional multivariate analysis using variables with p value of less than 0.2 in the univariate analysis (male sex, previous abdominal surgery, low tumor, open surgery, low anterior resection (LAR), D3 node dissection, defunctioning ileostomy, prolonged operation time, and pT3≤). The results indicated that defunctioning ileostomy formation (odds ratio (OR) 5.38, 95% CI 1.37–24.17, p = 0.0152) and D3 node dissection (OR 10.55, 95% CI 1.72–208.29, p = 0.0076) were independent risk factors for early postoperative small bowel obstruction. The results were similar to those mentioned in our article, showing that in our study the effect of p value, though potentially large, did not significantly affect the findings.

The second issue raised by Weng J et al. regarded wide 95% CIs for LAR and D3 dissection indicating multicollinearity. We calculated correlation coefficients among all variables and all were lower than 0.7. Thus, we considered that LAR and D3 dissection, though having wide 95% CIs, do not need to be omitted in the analysis.

As mentioned by Weng J et al., we did emphasize that the limitation of our study is its exploratory nature and the small sample size. Further investigation is mandatory.