ABSTRACT
BACKGROUND AND AIM: FAST score has a good performance for diagnosing the composite of NASH + NAS ≥ 4 + F ≥ 2. However, it has not been evaluated in Latin American individuals with nonalcoholic fatty liver disease (NAFLD). We aimed to analyze the performance of the FAST score in a Brazilian NAFLD population. METHODS: Cross-sectional study was held in ≥ 18 years NAFLD patients diagnosed by ultrasonography and submitted to liver biopsy (LB). Liver stiffness (LSM) and CAP measurements were performed with FibroScan®, using M (BMI < 32 kg/m2) or XL probes. Area under receiver operating characteristic (AUROC) curves were calculated as well as sensitivity (S), specificity (Spe), positive predictive value (VPP) and negative predictive value (NPV) for the previously established FAST score cut-offs. RESULTS: Among 287 patients included (75% female; mean age 55 ± 10 years), NASH + NAS ≥ 4 + F ≥ 2 was reported in 30% of LB. For the FAST cut-off of 0.35, the S and NPV to rule out NASH + NAS ≥ 4 + F ≥ 2 were 78.8% and 87.8%, respectively. Regarding the cut-off of 0.67, the Spe and PPV to rule-in NASH + NAS ≥ 4 + F ≥ 2 were 89.1%, 61.8%, respectively. The AUROC of FAST for all included patients was 0.78 (95% CI 0.72-0.84) and for those with ≥ 32 kg/m2 was 0.81 (95% CI 0.74-0.88). CONCLUSION: FAST score has a good performance in a Brazilian NAFLD population, even in patients with higher BMI when the XL probe is adopted. Therefore, FAST can be used as a noninvasive screening tool mainly for excluding the diagnosis of progressive NASH, reducing the number of unnecessary liver biopsies.
Subject(s)
Elasticity Imaging Techniques , Non-alcoholic Fatty Liver Disease , Humans , Female , Middle Aged , Aged , Male , Non-alcoholic Fatty Liver Disease/diagnostic imaging , Non-alcoholic Fatty Liver Disease/epidemiology , Liver Cirrhosis/diagnosis , Cross-Sectional Studies , Brazil/epidemiology , Biopsy , Liver/diagnostic imaging , Liver/pathologyABSTRACT
OBJECTIVES: Recently, controlled attenuation parameter (CAP) was incorporated for XL probe. However, its performance through M and XL probes has been scarcely evaluated in nonalcoholic fatty liver disease (NAFLD). The performance of probes regarding transient elastography by Fibroscan is still under debate. AIM: Compare the performance of CAP and transient elastography in NAFLD patients obtained through XL with M probes using histological analysis as gold standard. METHODS: NAFLD patients underwent liver biopsy and FibroScan/CAP with M and XL probes the same day. C-statistic evaluated CAP performance in the identification of moderate/severe (≥33%) and severe (≥66%) steatosis by both probes and transient elastography performance for identification of significant fibrosis (≥F2). RESULTS: Eighty-one patients (74% female; age 54.2 ± 9.9 years; BMI 32.8 ± 5.2/ BMI ≥ 25 92.6%; 96% metabolic syndrome; 60% diabetes mellitus) were included. Mean CAP with M and XL probes was 314 ± 39 and 325 ± 47 dB/m, respectively. The areas under receiver operating characteristic curves (AUROCs) of the M and XL probes for steatosis detection ≥33% were 0.75 (0.64-0.84) and 0.76 (0.65-0.84) (P = 0.95) and for steatosis ≥66% 0.83 (0.73-0.90) and 0.82 (0.71-0.89) (P = 0.73), respectively, with similar performances for both degrees of steatosis. Regarding transient elastography, AUROCs of M and XL probes for ≥F2 were 0.82 (0.71-0.93) and 0.80 (0.69-0.92) (P = 0.66). CONCLUSION: Performance of M and XL probes is similar for the diagnosis of moderate and severe steatosis and significant fibrosis even on a overweight population with NAFLD.
