Authors: Sandoval G. Luis*, Tolaque A. Yolanda, Gatica Manuel, Ranero Jorge.
Background: Hepatocellular-carcinoma (HCC) diagnosis can be made by imaging, if imaging is not definitely we need to individualized and include additional imaging or biopsy. However is important to determine if there is cirrhotic or non-cirrhotic liver for prognosis and treatment. In this study we aimed to compare the findings obtained for liver masses between three phase computed tomography (CT) with the anatomopathological results and characterized them epidemiologically.
Methods: Using an observational, retrospective and analytic study, all the liver masses seen by functional three phase CT were included and compared with the pathology result during 2015 – 2017 at Instituto Guatemalteco de Seguridad Social (IGSS). Categorical variables were presented in frequency and percentages and analyzed by Chi squared of homogeneity. Normality was tested with Kolmogorov-Smirnov. Numerical data was evaluated with t-student of independent samples. At relational level a bivariate study was made, then elevated to multivariate level. To measure sensitivity and specificity we constructed receiver operating characteristics (ROC) curves and calculate area under the curve (AUC). Negative predictive value (NPV) and positive predictive value (PPV) were calculated.
Results: Of 76 liver masses evaluated by three phase CT, 53% were confirmed by biopsy; 84% of HCC diagnosis were non-cirrhotic livers. In ROC curves, AUC for three phase CT for all samples was 0.646, then we stratified in cirrhotic and non-cirrhotic, AUC for cirrhotic was 0.65, PPV of 0.54 and NPV of 0.69 for HCC diagnosis. AUC for non-cirrhotic patients was 0.665, PPV of 0.67 and NPV of 0.62.
Conclusion: Our findings were the opposite of global epidemiology, most of the liver masses diagnosed as HCC are in non-cirrhotic livers. Three phase CT can diagnose HCC with good sensitivity and specificity and it has better PPV in non-cirrhotic patients, nevertheless biopsy should be done because they are low risk patients.View/Download pdf