Impact of Gender and Menopause on Histologic Severity of Inflammation and Injury in patients with Nonalcoholic Fatty Liver Disease

Abstract

BACKGROUND/AIM: How gender and menopause impact severity of NAFLD is not fully understood. As estrogens modulate cellular injury response and immune response, gender/menopause may impact the pattern and/or degree of inflammation/injury among patients with NAFLD. We investigated whether gender and menopause are associated with histologic severity of inflammation/injury in patients with NAFLD. METHODS: We analyzed 722 adult patients enrolled at two different clinical settings who had 1) histologic diagnosis of NAFLD and 2) no co-existing liver conditions or excess alcohol use: referral liver clinic (N=378) and bariatric surgery (N=344). Menopausal state was classified based on selfreported reproductive information, history of oophorectomy, and age relative to the average US age at menopause (50 years old). Liver histology was graded for steatosis (STE), lobular inflammation (LOB), hepatocyte ballooning (BAL) and portal inflammation (PORT), and staged for fibrosis (FIB) according to the NASH CRN scoring system. NASH was defined as any STE (≥ G1) accompanied by 1) BAL (≥ G1) or LOB (≥ G2) or 2) any FIB (≥ ST1) AND [any LOB (≥ G1) or BAL (≥ G1)]. Clinical characteristics, prevalence of NASH, and individual features of inflammation (i.e., LOB, BAL, and PORT) were compared among the gender/menopause categories, considering site difference and other potential confounders. RESULTS: Overall, the prevalence of NASH was significantly lower in pre-menopausal as opposed to post-menopausal women and men (68% vs. 80%, 79%, P<0.004). The association was eliminated (p=0.97) after adjusting for sites, black race, diabetes, hypertension, BMI and estrogen-containing agent use. Both gender and menopausal status were significantly associated with grades of BAL in a site-specific manner (interaction, p<0.05). After adjusting for the other confounders, men were more likely than women to have less severe BAL at the referral liver clinic (adjusted cumulative odds ratio [ACOR] and 95% confidence interval [CI] =0.4[0.3, 0.7], p<0.0003) while in the bariatric surgery population men were more likely than pre- menopausal women to have more severe BAL (ACOR and 95%CI=2.2[1.1, 4.2], p<0.03). The gender/menopause categories did not show significant association with LOB or PORT. CONCLUSION: Pre-menopausal women are protected from having NASH, which, may be explained by associated risk factors for NASH as opposed to gender and/or menopausal status. Severity of BAL was associated with gender/menopause in a site-specific manner, suggesting the presence of selection bias and/or differential impact of gender/menopause on inflammatory process depending on targeted populations.