Dear Editor
We would like to thank Liu et al.1 for their comment on our article2.
First, we agree that that the use of BMI as a continuous variable may underestimate the negative effect of low BMI on outcomes. Yet, continuous BMI was the better fit in the Cox and logistic regression models, compared with BMI categories, albeit slightly (Cox regression: Akaike's information criterion (AIC) continuous = 11 169.31 versus AIC categories = 11 172.84; logistic regression: AIC continuous = 1048.41 versus AIC categories = 1049.649).
Second, Liu et al.1 propose excluding the patients who died within 90 days after surgery from the model for long-term overall survival (OS). We would agree with excluding patients with postoperative mortality if the aim of the study had been to develop a model to inform patients and physicians ‘after’ surgery. Such models, however, have been published already3. Our aim in van Keulen et al.2 was to predict long-term OS for shared decision-making ‘before’ surgery. Excluding patients with 90-day mortality from this model would have resulted in overestimating long-term OS.
Third, Liu et al.1 recommend using bilirubin rather than jaundice at presentation. We agree that bilirubin at presentation may have been a better prognostic factor. We used jaundice at presentation, because bilirubin in our multicentre data set was not consistently reported at presentation; frequently, it was available only after biliary drainage or immediately before surgery. This inconsistency will be resolved in future updates of the multicentre data set.
Finally, we thank Liu et al.1 for noticing a discrepancy between the main text and Figure 1 legend regarding the weblink for the online calculator. We apologize for any confusion this has caused. The appropriate weblink is: https://dhoppener.shinyapps.io/risk_vs_harm_app/.
Contributor Information
Stefan Buettner, Department of Surgery, Erasmus MC Cancer Institute, Rotterdam, The Netherlands.
Anne-Marleen van Keulen, Department of Surgery, Erasmus MC Cancer Institute, Rotterdam, The Netherlands.
Bas Groot Koerkamp, Department of Surgery, Erasmus MC Cancer Institute, Rotterdam, The Netherlands.
Pim B Olthof, Department of Surgery, Erasmus MC Cancer Institute, Rotterdam, The Netherlands; Department of Surgery, Amsterdam University Medical Centres, University of Amsterdam, Amsterdam, The Netherlands.
References
- 1. Liu Z-P, Gong Y, Dai H-S, Yin X-Y, Chen Z-Y. 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? Br J Surg 2023;110:873 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. van Keulen A-M, Buettner S, Erdmann JI, Pratschke J, Ratti F, Jarnagin WRet al. 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? Br J Surg 2023;110:599–605 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Groot Koerkamp B, Wiggers JK, Gonen M, Doussot A, Allen PJ, Besselink MGHet al. Survival after resection of perihilar cholangiocarcinoma—development and external validation of a prognostic nomogram. Ann Oncol 2015;26:1930–1935 [DOI] [PMC free article] [PubMed] [Google Scholar]