AIM: Liver biopsy is still required for the diagnosis of hepatocellular ballooning and inflammation, which are important histological feature of nonalcoholic steatohepatitis (NASH). We conducted this multicenter, cross-sectional study to identify novel blood markers for the diagnosis of hepatocellular ballooning. METHODS: We enrolled 176 patients, and 132 patients proven by liver biopsy as having NAFLD were classified as non-ballooning (ballooning grade 0) (n = 83) or ballooning (ballooning grade 1 and 2) (n = 49) by a central pathology review. We performed gas chromatography-mass spectrometry, hydrophilic interaction liquid chromatography tandem mass spectrometry, and lipidomics with plasma. RESULTS: As correlates of hepatocellular ballooning, among the clinical parameters, serum type IV collagen 7S correlated the most significantly with the ballooning grade (correlation coefficient [CC] = 0.463; P < 0.001). Among the metabolic-/lipidomic-markers, phosphatidylcholine (PC)(aa-44:8) correlated the most significantly with the ballooning grade (CC = 0.394; P < 0.001). The area under the receiver operating characteristic curve of type IV collagen 7S, choline, and lysophosphatidylethanolamine (LPE) (e-18:2), was 0.846 (95% confidence interval: 0.772-0.919). CONCLUSIONS: Plasma levels of PC were positively correlated, and those of lysophosphatidylcholine and LPE were negatively correlated with hepatocellular ballooning in NAFLD patients. These noninvasive metabolic-/lipidomic-based plasma tests might be useful to distinguish between cases of NAFLD with and without hepatocellular ballooning. This article is protected by copyright. All rights reserved.
AIM: Liver biopsy is still required for the diagnosis of hepatocellular ballooning and inflammation, which are important histological feature of nonalcoholic steatohepatitis (NASH). We conducted this multicenter, cross-sectional study to identify novel blood markers for the diagnosis of hepatocellular ballooning. METHODS: We enrolled 176 patients, and 132 patients proven by liver biopsy as having NAFLD were classified as non-ballooning (ballooning grade 0) (n = 83) or ballooning (ballooning grade 1 and 2) (n = 49) by a central pathology review. We performed gas chromatography-mass spectrometry, hydrophilic interaction liquid chromatography tandem mass spectrometry, and lipidomics with plasma. RESULTS: As correlates of hepatocellular ballooning, among the clinical parameters, serum type IV collagen 7S correlated the most significantly with the ballooning grade (correlation coefficient [CC] = 0.463; P < 0.001). Among the metabolic-/lipidomic-markers, phosphatidylcholine (PC)(aa-44:8) correlated the most significantly with the ballooning grade (CC = 0.394; P < 0.001). The area under the receiver operating characteristic curve of type IV collagen 7S, choline, and lysophosphatidylethanolamine (LPE) (e-18:2), was 0.846 (95% confidence interval: 0.772-0.919). CONCLUSIONS: Plasma levels of PC were positively correlated, and those of lysophosphatidylcholine and LPE were negatively correlated with hepatocellular ballooning in NAFLD patients. These noninvasive metabolic-/lipidomic-based plasma tests might be useful to distinguish between cases of NAFLD with and without hepatocellular ballooning. This article is protected by copyright. All rights reserved.