A Multi-center Clinic Trial Of Assessment Of Fatty Liver Disease By FibroScan And CK-18in China

Objective:1. To evaluate the performance of a novel non-invasive controlled attenuationparameter (CAP) to quantitative assess hepatic steatosis and differentiateconfounding factors.2. To appraise the interference of hepatic steatosis and subcutaneous-liver capsuledistance (SCD) for liver stiffness measurements (LSM). To assess and verifyLSM noninvasive and quantitative evaluation of fibrosis of non-alcoholic fattyliver disease (NAFLD) and chronic hepatitis B (CHB) patients.3. To investigate the capability of FibroScan (CAP+LSM) combined cytokeratin-18(CK-18) on non-invasive and quantitative assessing "hepatic steatosis,non-alcoholic steatohepatitis (NASH) and fibrosis" spectrum of NAFLD.Methods:1. This was a multi-center prospective cohort study. Consecutive patients (aged≥18years) who had undergone percutaneous liver biopsy and proven NAFLD andCHB were recruited from five Chinese liver centers.2. FibroScan-502equipped with M probe was used to capture both CAP and LSMsimultaneously. SCD was determined by TM model. Demographic,anthropometry and biochemical parameters were measured.3. Detcetion of cytokeratin-18(CK-18) with M30/M65by enzyme-linkedimmunosorbent assay (ELISA).4. The NAFLD activity score (NAS) and fibrosis stage were used for NAFLD andthe METAVIR classification was used for CHB. Steatosis was categorised as S0(<5%)，S1(5~33%), S2(34~66%), S3(>66%).5. Continuous variables and patient characteristics were expressed as either median(interquartile range, IQR) or n (%), as appropriate. The Chi-squared test orFisher’s exact test were used to compare categorical data. The Mann-Whitney Utest or Kruskal-Wallis H test were used to compare continuous variables. Multivariate analyses were used by multiple linear regression or binary Logisticanalysis. The receiver operating characteristic (ROC) curves were plotted, and theareas under the curves (AUROC) were calculated to determine the diagnosticvalue. The accuracy of the measurement values at the optimal thresholds wasdefined by maximizing the sum of sensitivity and specificity (maximum Youdenindex).Results:1. A total of381patients were recruited, including101NAFLD patients (aged18to57years,73.3%were male) and280CHB patients (aged18to66years,69.3%were male，84cases with fatty liver). No significant differences were found in theCAP values between the NAFLD group and the CHB group in each steatosisgrade.2. CAP was positively independently associated with liver steatosis and SCD, andnegativele independently with total bilirubin (TBIL). When SCD<25mm, TheCAP (dB/m) thresholds values and AUROC (95%confidence interval, CI) ofidentification all patients with steatosis more than>5%,>33%and>66%were265.5and0.883(0.847-0.920),283.5and0.914(0.877-0.951),293.5and0.878(0.827-0.929), respectively. When SCD≥25mm, the CAP threshold wassignificantly increased.3. The median (IQR) of LSM (kPa) was no significant difference in the degree ofsteatosis or SCD after adjusted fibrosis.The optimal cut-off values of LSM (kPa)identification of fibrosis F1, F2, F3and F4for NAFLD were5.45,7.75,8.65and14.3, while8.15,9.15,10.6and10.85for CHB patients.4. For quantitative measurement of "steatosis, NASH and fibrosis" spectrum,FibroScan combined CK-18were used in65NAFLD patients. For theidentification of steatosis>33%, AUROC (95%CI) of CAP was0.848(0.752-0.945), high than M30and M65. For steatosis>66%, AUROC (95%CI)of M65was0.772(0.647-0.896), higher than CAP and M30. AUROC (95%CI)of LSM in differentiating≥F1,≥F2,≥F3and≥F4were0.765(0.595-0.935),0.795(0.577-0.912),0.706(0.524-0.899) and0.952(0.891-1.000), all higher thanM30and M65. Binary Logistic regression analysis showed that M65and CAPwere independently prediction factors for NASH. A two-step approach combinedM65and CAP was recommended to discriminate NASH with PPV85.7%, NPV100%and diagnostic conforming rate92.0%.Conclusions:1. CAP appears to be a promising tool for the non-invasive detection andquantification of hepatic steatosis, but is limited by SCD.2. CAP values were not affected by etiology, such as NAFLD and CHB.3. LSM values were not significantly interfered by hepatic steatosis and SCD.4. FibroScan (CAP+LSM) combined CK-18(M65) had a good perfermance of"steatosis, NASH, fibrosis" spectrum of NAFLD.