Reply to the Letter to the Editor: Derivation and Internal Validation of a Clinical Prediction Tool to Predict Nonalcoholic Fatty Liver Disease in Patients With Crohn’s Disease
Year: 2020
Type: letter
Abstract: To the Editors, We greatly appreciate the comments and interest of Drs. Xu, Chen. and Yang regarding our article in which we developed a clinical prediction Ttool for nonalcoholic fatty liver disease (NAFLD) in Crohn’s disease (CPN-CD). In addressing their comments, we would like to highlight that the CPN-CD was designed to guide the clinical decision to screen CD patients for NAFLD during routine care and not specifically as a hypothesis-generating tool for the pathophysiology of NAFLD in CD. Therefore, the model derivation was performed without regard to the causal relationship of the predictors.1 For example, alanine aminotransferase is highly predictive for NAFLD and is appropriate for our anticipated use of the CPN-CD, whereas including alanine aminotransferase would not be appropriate in a model designed to study pathophysiology. We also wish to highlight that the proposed predictors (biologics, steroids, intestinal surgery, and subjective disease activity) were considered as candidate variables for the CPN-CD. They were not included in the final model as they did not improve the operating characteristics of the CPN-CD to justify increasing its complexity. Objective disease activity (endoscopic or radiographic) was not considered a candidate variable because the value of that predictor would be known only after the CPN-CD would have been used to triage screening for NAFLD in CD patients. The candidate variable of an age-sex interaction to model postmenopausal status was added in the model because it improved the performance characteristics of the CPN-CD compared with excluding them altogether or considering them separately. Additionally, variables not significant in a univariate analysis may still be included in a final multivariable model if biologically plausible and consistent with current statistical methodology.2 The role of a patient’s biologic sex in the pathophysiology of hepatic steatosis is by no means straightforward. Highlighting this, the Hepatic Steatosis Index considers female sex as a risk factor,3 whereas the Framingham Steatosis Index actually treats male sex as a risk factor.4 Although we agree that testosterone signaling is likely important, estrogen signaling and sex hormone–induced changes in insulin resistance5 seem to be similarly relevant. Finally, we would like to thank Drs. Xu, Chen, and Yang for catching our error in the body of the manuscript where the number of patients characterized as high-risk was mistakenly transcribed as 27% rather than 34%. Furthermore, we wholeheartedly agree that further elucidation of the pathophysiology of NAFLD in CD is warranted, and we are actively pursuing that avenue of research and the opportunity to improve upon the CPN-CD as new knowledge develops. Supported by: SM is supported by an Institutional National Research Service Award (T32-DK007130-45). NOD is supported by grants DK-119437, HL-38180, and DK-112378. PD is supported by a Junior Faculty Development Award from the American College of Gastroenterology. MAC is supported by DK109384, a Crohn’s and Colitis Foundation Daniel H Present Senior Research Award (Ref. 370763) and philanthropic support from the Givin’ It All for Guts Foundation (https://givinitallforguts.org) and the Lawrence C. Pakula MD IBD Research Innovation and Education Fund.
Source: Inflammatory Bowel Diseases
DOI: 10.1093/ibd/izaa043
Institution: Washington University in St. Louis
Language: English
Topic: Inflammatory Bowel Disease
Subfield: Genetics
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W3010201390