Letter to the Editor: Contrast-enhanced CT and liver surface nodularity for the diagnosis of portosinusoidal vascular disorder: a case-control study

To the editor, We read with great interest the article entitled "Contrast-enhanced CT and liver surface nodularity (LSN) for the diagnosis of portosinusoidal vascular disorder (PSVD): a case-control study," by Valainathan et al.1 The authors do a great job in assessing the performance of liver surface nodularity and liver segment IV size for a PSVD diagnosis in patients with signs of portal hypertension, however, there are a few concerns that need to be addressed. First, the author claimed that normal or enlarged segment IV was independently associated with PSVD, however, this sign was judged by visual impression of the readers. Neither did they describe the criteria of this sign nor the method of measuring transversal size of segment IV. Second, lots of research proved that PSVD patients seldomly developed in HCC.2,3 As HCC could be recognized through CT scan, if HCC was not a study factor in this research, it should be regarded as a baseline factor and matched between cirrhosis and PSVD patients in a case-control study. As a result, liver cirrhosis patients with HCC were supposed to be excluded from control cohort. Third, the authors did not describe the selection process of variables entering multivariate binary logistic regression analysis. Lots of methods could be considered such as stepwise logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO) and so on. Fourth, as past research proved that liver stiffness measurement could be a useful method to distinguish PSVD from cirrhosis, liver stiffness measurement results were supposed to be listed in Tables 1 or 2.4 On the contrary, the authors just briefly mentioned it in the results and discussion. Last but not least, the authors only described the area under the receiver operating characteristics curve of the multivariable model, however, they did not show the calibration curve, which might help further in evaluating the accuracy and efficiency of the model in learning and validation cohort. AUTHOR CONTRIBUTIONS Zijin Liu and Mingjie Tan wrote the letter. Huiguo Ding revised the letter. CONFLICT OF INTEREST The authors have no conflicts to report.