Comment on: Multivariable prediction model for both 90-day mortality and long-term survival for individual patients with perihilar cholangiocarcinoma: does the predicted survival justify the surgical risk?

Dear Editor

We read with great interest the article by van Keulen et al.¹. The authors used a multicentre international database to develop preoperative risk models for 90-day mortality and overall survival (OS) to help identify patients with resectable perihilar cholangiocarcinoma (pCCA) who are unlikely to benefit from surgical resection. This is a great study; however, we still have the following comments.

First, we note that BMI is included in both the Cox and logistic regression models as a continuous variable, which is inappropriate. Low BMI usually indicates malnutrition or cachexia. Including BMI as a continuous variable in the regression model would ignore the effect of low BMI on 90-day mortality risk and OS. Second, when discussing the factors that affect long-term survival, more attention is given to the factors that affect tumour recurrence. Therefore, those who died 90 days after surgery should be excluded from the cohort when Cox regression analysis was performed because the cause of death in these patients was commonly not tumour recurrence but liver failure. Third, the authors did not specify the definition of preoperative jaundice, but it is a very important preoperative indicator. In clinical practice, higher preoperative bilirubin means poorer liver function and greater risk of surgery. We believe that preoperative bilirubin should be included as a continuous variable in a regression model to more accurately reflect liver function, especially when the majority of pCCA patients have jaundice. Fourth, the link for the model in the body of the manuscript is inconsistent with the link in the legend. This may cause great confusion to other readers.

Finally, we are very appreciative of van Keulen et al.¹ for the development of such a valuable prediction model to better guide surgical decision-making.

Author contributions

Zhi-Peng Liu (Conceptualization, Writing—original draft), Yi Gong (Conceptualization, Writing—original draft), Hai-Su Dai (Writing—original draft), Xian-Yu Yin (Project administration, Resources), and Zhi-Yu Chen (Conceptualization, Funding acquisition, Writing—review & editing).