Non-alcoholic steatohepatitis(NASH)is considered as the hepatic counterpart of the metabolic syndrome.
To investigate the determinants of NASH reversal in patients undergoing biliopancreatic diversion(BPD)in a 5-year follow-up study.
Prospective Study, Policlinico Universitario Agostino Gemelli.
37 patients underwent fine-needle liver biopsy during BPD. Nine of them had a negative liver biopsy for NASH and were excluded. Ultrasonography-guided percutaneous liver biopsy was obtained at 5 years after operation.
Biliopancreatic Diversion and liver biopsy.
The primary outcome of our study was histologic NASH reversal, at 5-years follow-up. To better characterize the clinical variables involved in the resolution of NASH, we also compared patients without histologic NASH resolution at 5 years, with those in whom NASH had disappeared.
At follow-up, NASH reversed in 56.5% of the patients. NAFLD Activity-Score(NAS)improved from 3.33±1.15 to 1.84±1.10(P<0.0001).Fibrosis reversed in 16% of the patients(P=0.022)and in 32% improved(95% CI, 0.05-0.54).No significant differences in BMI or clinical parameters changes explained the effect of surgery on NASH, apart from the measure insulin sensitivity post-surgery. HOMA-IR decreased from 3.31±1.72 at baseline to 1.73±1.08(P<0.0001)after BPD and Matsuda index improved from 2.66±1.79 to 4.73±3.05(P<0.0001).Lipid profile normalized(total-cholesterol from 4.75±1.18 to 3.32±0.77mmol/l, P<0.0001; LDL-cholesterol from 2.92±0.91 to 1.60±0.51mmol/l, P=0.0001; HDL-cholesterol from 0.97±0.33 to 1.10±0.35mmol/l, P=0.023; triglycerides from 2.52±1.6 to 1.47±0.67 mmol/l, P=0.003).Neural network analysis showed that end-study Matsuda index discriminated between responders and non-responders with high accuracy[receiver operating characteristic curve(ROC)area-under-the curve(AUC)0.98%].
Remission of NASH is driven by reversal of whole-body insulin resistance post-intervention.

© The Author(s) 2020. Published by Oxford University Press on behalf of the Endocrine Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Author