Serum concentrations of classic liver enzymes are unsuitable for diagnosing non-alcoholic steatohepatitis

Introduction The prevalence of non-alcoholic fatty liver disease (NAFLD) in the German population is 20–30 %. This large number of potentially affected people rules out performing a liver biopsy to confirm a possible diagnosis of non-alcoholic steatohepatitis (NASH) or for the purpose of risk assessment in all cases. Non-invasive methods for diagnosis and to support screening strategies are therefore required. Elevated serum concentrations of classic liver enzymes are currently used as an indicator for NAFLD or NASH.

Goals In this study we aimed to examine whether an adjustment of the upper limit of normal for classic liver enzymes (AST, ALT and GGT) leads to an improvement of the diagnostic accuracy.

Methods Data from 363 morbidly obese patients (42.5 ± 10.3 years old; mean BMI: 52 ± 8.5 kg/m²) who underwent bariatric surgery, with histologically confirmed NAFL or NASH (NAS and SAF score available), were evaluated using statistical methods.

Results In 121 women (45 %) and 45 men (46 %) we found elevated values for at least one of the three parameters ALT, AST or GGT. The serum concentrations of ALT (p  < 0.0001), AST (p  < 0.0001) and GGT (p = 0.0023) differed significantly between patients with histologically confirmed NAFL and NASH, regardless of whether the classification was based on the NAS or the SAF score. All three serum parameters correlated significantly positively with the NAS and the SAF score, with correlation coefficients between 0.33 (correlation of ALT with NAS) and 0.40 (correlation of GGT with SAF score). In order to assess whether it is possible to separate NAFL and NASH using classic liver enzymes, ROC curves were generated for all three parameters. The area under the ROC curve achieved a maximum of 0.70 (for ALT applying the NAS), which demonstrates a poor to moderate separation. To achieve 95 % specificity, the upper limit of normal for ALT would have to be 47.5 U/l; for 95 % sensitivity, the upper limit of normal for ALT would have to be 17.5 U/l, which leads to a diagnostic gap of 62 % uncategorized patients.

Discussion ALT, AST and GGT are unsuitable for the non-invasive diagnosis of NAFL or NASH. Furthermore they are not applicable as a screening tool for NASH in obese patients. The suitability of classic liver values as an indicator for NAFLD should generally be questioned.

Publikationsverlauf

Artikel online veröffentlicht:
04. Januar 2021