Diagnostic Accuracy of Plain Computed Tomography to Detect the Morphological Variations of Liver: An Update

Abstract

Introduction: The prevalence of accessory fissures (AFs) in the liver ranges from 6% to 56%, as reported by cadaveric studies, which is much higher than the prevalence of 25% on computed tomography (CT) scans. Despite reporting many morphological variations in the liver by various cadaveric studies worldwide, the imaging studies are very few. Radiological imaging of patients undergoing liver surgery is a routine preoperative investigation. Despite there are several occasions where liver variations occur as a surprise during surgery which reflects on the diagnostic capacity of CT. Materials and Methods: Sixty cadaveric livers removed and stored in 10% Formalin were studied for all morphological abnormalities. The same livers were analyzed separately by anatomists by gross examination and radiologists by plain CT imaging. The radiologists were blinded from the gross examination findings. The data obtained by both methods were compared and the diagnostic accuracy of CT was estimated. Results: The common surface morphological variations detected were AFs, accessory lobes (ALs), pons hepatis (PH), and multilobed caudate and quadrate lobes (QLs). Out of the total 89 AFs identified by gross examination, only 73 could be detected by CT scan, thus the sensitivity of CT scan in detecting AFs is 82%. Similarly, the sensitivity of CT scan in detecting ALs, PH, and multilobed caudate, and QL was 64% (low), 81%, and 19% (very low), respectively. Thus, plain CT has variable sensitivity depending on the morphological variation. Conclusions: This study has highlighted some of the diagnostic inaccuracies that may arise during plain CT examination of the liver in a person posted for liver surgery or a person with abdominal trauma. Contrast-enhanced CT scans can circumvent many of these problems. Lack of awareness of these issues may affect the normal course of treatment and prognosis in such patients.