Two are better than one: noninvasive assessment of liver fibrosis in nonalcoholic fatty liver disease

Nonalcoholic fatty liver disease (NAFLD) is currently the most common cause of chronic liver disease in many regions, affecting 15–40 % of the general population [1, 2]. One of the key challenges in managing NAFLD is differentiating simple steatosis from advanced disease, i.e., those with nonalcoholic steatohepatitis (NASH), advanced fibrosis or even cirrhosis [3]. The challenge is further substantiated by the poor correlation between liver fibrosis and routine blood tests, e.g., the serum alanine aminotransferase (ALT) level [4]. Patients may have progressed to advanced fibrosis or cirrhosis without any abnormalities in these common blood parameters [5]. Liver biopsy is not always feasible or acceptable to patients, especially to those largely asymptomatic subjects; this has promoted the popularity of noninvasive assessments of liver fibrosis for NAFLD patients (Table 1) [6, 7]. Among the numerous noninvasive assessments, liver stiffness measurement (LSM) by transient elastography is one of the most popular tools [8]. This tool out-performed other serum parameters in detecting advanced fibrosis or cirrhosis [9]. Nonetheless, a false-positive diagnosis of advanced fibrosis is not uncommon, in addition to LSM results lying in the gray zone [9]. This would be an even more serious problem in obviously obese subjects, as the LSM is higher among those with a body mass index (BMI) above 30 kg/m in the same fibrosis stage [10]. On the other hand, the NAFLD fibrosis score (NFS) is the beststudied serum-based combined clinical algorithm; it was recommended in the latest American practice guidelines for NAFLD [11]. All these observations have raised the interest in combining these two well-studied noninvasive parameters. On the other hand, magnetic resonance elastography (MRE) is probably even more accurate than LSM and NFS [12]. Nonetheless, MRE is not as widely available as these two tools. Recently, Chan et al. [13] evaluated the diagnostic performance a novel two-step approach combining NFS and LSM. The beauty of this study was the robust methodology of including two independent groups of patients as the training and validation cohort, respectively. The detailed comparison of the performance of various approaches, be it either parameter alone or combining two parameters in different fashions, established this novel twostep approach as the best among the five to avoid liver biopsy. Similar approaches have been proposed in patients with chronic hepatitis B [14, 15]; this was the first time in NALFD patients. The unresolved questions remaining include the suboptimal performance of NFS in Asian subjects because of the different BMI cutoffs for obesity [16]. This was echoed by the relatively low sensitivity (40 %) of NFS to predict advanced fibrosis in Chan’s study. Investigators should pay special attention when applying any parameters derived from the west to Asian patients, such that different cutoffs or even different algorithms should be considered. It would be helpful if another serum parameter with better performance could be found. Cytokeratin-18 (CK18) fragments have been investigated extensively as novel biomarkers for G. L.-H. Wong H. L.-Y. Chan Institute of Digestive Disease, The Chinese University of Hong Kong, Shatin, Hong Kong