SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY, cilt.49, sa.11, ss.1343-1348, 2014 (SCI-Expanded)
Background: Noninvasive markers that purport to distinguish patients with non-alcoholic fatty liver disease (NAFLD) with fibrosis from those without must be evaluated rigorously for their classification accuracy. Herein, we seek to compare the diagnostic performances of three different noninvasive methods (FibroMeter (TM) NAFLD score, NAFLD Fibrosis score (NFSA), and Transient Elastrography [TE]) for the detection of liver fibrosis in NAFLD patients. Methods: A total of 88 patients with biopsy-proven NAFLD were included. The Kleiner system was used for grading fibrosis in liver biopsies. The FibroMeter (TM) NAFLD score was determined using a proprietary algorithm (regression score). The NFSA score was calculated based on age, hyperglycemia, body mass index, platelets, albumin and serum aminotransferase levels. TE was performed using the Fibroscan apparatus. Results: The sensitivities/specificities for the FibroMeter (TM) NAFLD score, NFSA, and TE for the diagnosis of significant fibrosis (F2 + F3 + F4 fibrosis) were 38.6%/86.4%, 52.3%/88.6%, and 75.0%/93.2%, respectively. The areas under the receiver operating characteristic curves of TE were significantly higher than those of both the FibroMeter (TM) NAFLD score and NFSA. No significant differences were found between the FibroMeter (TM) NAFLD score and NFSA for the detection of significant and severe fibrosis, although the diagnostic performance of the FibroMeter (TM) NAFLD score was higher than that of the NFSA score for cirrhosis. Conclusions: In summary, TE showed the best diagnostic performance for the noninvasive assessment of liver fibrosis in NAFLD patients. The diagnostic performances of the FibroMeter (TM) NAFLD score and NFSA did not differ significantly for the detection of both significant and severe fibrosis.