DIAGNOSTIC ACCURACY OF TRIPHASIC CT VERSUS ULTRASONOGRAPHY IN DETECTING LIVER LESIONS IN DIABETIC PATIENTS: A PILOT CASE SERIES

Abstract

Background: Diabetes mellitus is increasingly associated with hepatic complications, including NAFLD, cirrhosis, and hepatocellular carcinoma (HCC). Liver lesions in diabetic patients may remain clinically silent, making imaging essential for early detection. Although abdominal ultrasound (USG) is commonly used as the initial imaging modality due to its availability and safety, its accuracy may decrease in patients with fatty liver and chronic liver disease. Triphasic computed tomography (CT) provides improved lesion detection and characterization through multiphasic contrast enhancement but involves higher cost, radiation exposure, and contrast-related risks. Objective: This study aimed to compare the diagnostic performance of abdominal ultrasound and triphasic CT for detection and characterization of liver lesions in diabetic patients and to evaluate the ability of ultrasound to identify lesions detected by CT. Methods: This prospective pilot diagnostic accuracy case series included 5 male patients with type 2 diabetes mellitus and suspected liver lesions. All participants underwent abdominal ultrasound and triphasic CT within two weeks. Ultrasound was considered the index test, while triphasic CT was used as the reference standard because histopathological confirmation was unavailable. Clinical parameters, including HbA1c, liver function tests, and imaging findings, were recorded. Ultrasound sensitivity and accuracy were calculated using CT findings as the reference. Results: The five patients had a mean age of 57.8 years (range: 36–68 years) and a mean HbA1c level of 8.9%. Triphasic CT detected and characterized all lesions, achieving 100% sensitivity. Ultrasound detected lesions in four out of five patients, resulting in a sensitivity and accuracy of 80%. The single false-negative case occurred in a 65-year-old patient with poor glycemic control (HbA1c 11.2%) and chronic liver disease, where ultrasound missed a large 9.3 cm HCC and reported only a small 13 mm echogenic lesion suggestive of hemangioma. Ultrasound also showed limitations in lesion characterization, including misclassification of malignant lesions as benign lesions. Conclusion: Triphasic CT demonstrated superior diagnostic performance compared with abdominal ultrasound for detecting and characterizing liver lesions in diabetic patients, particularly those with poor glycemic control and chronic liver disease. Ultrasound remains useful as an accessible screening tool; however, it may not be sufficient as the sole imaging modality in high-risk diabetic patients. Larger multicenter studies with histopathological confirmation are required to establish definitive imaging protocols.