Suboptimal diagnostic accuracy of ultrasound and CT for compensated cirrhosis: Evidence from prospective cohort studies

Abstract

Introduction: Abdominal ultrasound (US) and CT are important tools for the initial evaluation of patients with liver disease. Our study aimed to determine the accuracy of these methods for diagnosing cirrhosis. Methods: In all, 377 participants from 4 prospective cohort studies evaluating patients with various liver diseases were included. All patients were included between 2017 and 2022 and had undergone a liver biopsy as well as US and/or CT. Using the histological assessment as the gold standard, we calculated diagnostic accuracy for US and CT. Liver biopsies were evaluated by expert histopathologists and diagnostic scans by experienced radiologists. Results: The mean age was 54 ± 14 years and 47% were female. Most patients had NAFLD (58.3%) or alcohol-associated liver disease (25.5%). The liver biopsy showed cirrhosis in 147 patients (39.0%). Eighty-three patients with cirrhosis had Child-Pugh A (56.4% of patients with cirrhosis) and 64 had Child-Pugh B/C (43.6%). Overall, the sensitivity for diagnosing cirrhosis by US was 0.71 (95% CI 0.62–0.79) and for CT 0.74 (95% CI 0.64–0.83). The specificity was high for US (0.94, 95% CI 0.90–0.97) and for CT (0.93, 95% CI 0.83–0.98). When evaluating patients with Child-Pugh A cirrhosis, sensitivity was only 0.62 (95% CI 0.49–0.74) for US and 0.60 (95% CI 0.43–0.75) for CT. For patients with Child-Pugh B/C, sensitivity was 0.83 (95% CI 0.70–0.92) for US and 0.87 (95% CI 0.74–0.95) for CT. When limiting our analysis to NAFLD (20% with cirrhosis), the sensitivity for US was 0.45 (95% CI 0.28–0.64) and specificity was 0.97 (95% CI 0.93–0.99). Conclusion: US and CT show moderate sensitivity and may potentially overlook compensated cirrhosis underlining the need for additional diagnostic testing.