Verhaegh P, et al. – This meta–analysis was conducted on non–invasive tests for differentiating non–alcoholic steatohepatitis (NASH) from simple steatosis (SS), focusing on blood markers. Although blood markers as part of scoring systems in single studies, showed promising results as non–invasive diagnostics for NASH, none of the pooled results revealed good sensitivity and specificity (≥80%). Replication studies and more standardized study designs were urgently needed. No marker or scoring system could be recommended for use in clinical practice to differentiate NASH from SS at present.

• Using defined keywords, the authors performed a systematic search in PubMed, Medline, and Embase (1990-2016), limited to full-text papers in English and human adults, resulted in 2608 hits.
• A total of 122 eligible articles using liver biopsy as gold standard were identified after screening by two independent reviewers.
• They determined pooled sensitivity (sens_p) and specificity (spec_p) using metafor in R if at least 2 studies were available.

• The authors identified 122 studies, in which 219 different blood markers, either as single marker (n=107) or as part of scoring systems (n=112), and 22 other diagnostic tests were studied.
• In this study, markers identified related to several pathophysiological mechanisms.
• Alanine aminotransferase (sens_p 63.5%, spec_p 74.4%) within routine biochemical tests, adiponectin (sens_p 72.0%, spec_p 75.7%) within inflammatory markers, CK18-M30 (sens_p 68.4%, spec_p 74.2%) within markers of cell death/proliferation and HOMA-IR (sens_p 69.0%, spec_p 72.7%) within the metabolic markers were the most extensively studied.
• As per the outcomes, two scoring systems could also be pooled, NASH test (NASH vs borderline NASH+SS sens_p 22.9%, spec_p 95.3%) and the GlycoNASH test (sens_p 67.1%, spec_p 63.8%).