Texas Clinical Research Institute Arlington, Texas
Introduction: Nonalcoholic fatty liver disease (NAFLD) is a common cause of liver disease. Newer modalities such as FibroScan and MR elastography can quantify the amount of steatosis and fibrosis on the liver. FibroScan has gained wide acceptance since it is non-invasive, less costly, and performed in physician’s offices. Our study aims to evaluate the accuracy of FibroScan compared to liver biopsy in a community clinic setting.
Methods: Charts of 90 NAFLD patients were reviewed, and data were abstracted on FibroScan liver stiffness measures (LSM) and liver biopsy results. Accuracy of the LSM was defined as concordant with the Metavir fibrosis on biopsy if the difference was less than 1 stage. Diagnostic performance of the LSM was evaluated using area under the curve of the receiver operating characteristic (AUROC) curves with the recommended NAFLD fibrosis cut-off points.
Results: Of the 90 NAFLD patients, 54 (60%) met the diagnostic criteria for NASH with at least 1 point in each of steatosis, inflammation, and ballooning. Concordant LSM was identified in 59 patients (66%) with disconcordant LSM in 31 patients (34%). Under staging happened in 14 subjects (16%) and over staging in 17 subjects (19%). The usefulness of the LSM values based upon AUROC and sensitivity/specificity is in the sufficient to good level (0.6 – 0.7 and 0.7 – 0.8, respectively) (See Figure 1). The positive predictive value (PPV) was highest in the ≥F2 group and the negative predictive value (NPV) was highest in the F4 group (See Table 1). Therefore, the LSM results are better at identifying those with ≥F2 fibrosis and those without F4 fibrosis.
Discussion: We conclude that FibroScan should not be used as the single method to evaluate the severity of NAFLD and that a multi-modality approach is needed. Potential limitations of our study include small sample size, high percentage of morbidly obese patients, high NAS scores, and elevated liver enzyme levels which are potential confounders. In clinical research studies on other liver diseases, obesity, acute liver inflammation, elevated transaminases, extrahepatic cholestasis, and increased central venous pressure have been found to be independent risk factors affecting FibroScan LSM measurements; these factors may also impact LSM measurement in patients with NAFLD. Further studies are needed to evaluate the role of factors impacting the FibroScan as a fibrosis measurement tool and to develop specific guidelines for use in NAFLD patients in a community clinic setting.
