Association between ultrasound quality and test performance for HCC surveillance in patients with cirrhosis: a retrospective cohort study. Issue 6 (15th February 2022)

• Record Type: Journal Article
• Authors: Chong, Nicolas; Schoenberger, Haley; Yekkaluri, Sruthi; Fetzer, David T.; Rich, Nicole E.; Yokoo, Takeshi; Gopal, Purva; Manwaring, Carrie; Quirk, Lisa; Singal, Amit G.
• Is Part Of: Alimentary pharmacology & therapeutics. Volume 55:Issue 6(2022)
• Page Start: 683
• Page End: 690
• Publication Date: 2022-02-15
• Subjects: Digestive organs -- Diseases -- Treatment -- Periodicals; Digestive organs -- Effect of drugs on -- Periodicals; Gastrointestinal system -- Diseases -- Treatment -- Periodicals; Gastrointestinal system -- Effect of drugs on -- Periodicals; 615.73
• DOI: 10.1111/apt.16779
• Languages: English
• ISSNs: 0269-2813

Abstract

Summary: Background: Ultrasound visualisation is limited in approximately 20% of patients with cirrhosis undergoing hepatocellular carcinoma (HCC) surveillance; however, it is unknown if impaired visualisation directly impacts test performance. We aimed to evaluate the association between ultrasound visualisation and surveillance test performance. Methods: We performed a retrospective cohort study among patients with cirrhosis, with or without HCC, who underwent ultrasound‐based surveillance at two large health systems between July 2016 and July 2019. Ultrasound visualisation assessment was recorded by interpreting radiologists using the ultrasound LI‐RADS Visualisation score. We performed logistic regression analyses to evaluate the association between ultrasound visualisation and diagnostic test performance. We assessed sensitivity for HCC detection among ultrasounds performed in the year prior to HCC diagnoses and specificity using ultrasounds in those without HCC. Results: Among 186 patients with HCC, severely limited visualisation (Vis Score C) on ultrasound prior to HCC diagnosis was associated with increased odds of false‐negative results, that is lower sensitivity (OR 7.94, 95% CI 1.23‐51.16) in multivariable analysis. Ultrasound sensitivity with visualisation scores A or B exceeded 75%, compared to only 27.3% with visualisation score C. Among 2052 cirrhosis patients without HCC, moderate visualisation limitations (Vis score B) were associated with increased odds of false‐positive results (OR 1.60, 1.13‐2.27), although specificity exceeded 95% across all visualisation scores. Conclusions: Impaired ultrasound visualisation is associated with worse surveillance test performance. Alternative blood‐based biomarkers and imaging strategies are needed for patients at risk for ultrasound‐based surveillance failure. Abstract : Severely limited ultrasound visualization is associated with significantly lower sensitivity for early detection of HCC in patient with cirrhosis, although specificity and risk of false positive observations appears to be acceptable across visualization scores. Created in BioRender.com.