Non-alcoholic steatohepatitis (NASH) in patients with polycystic ovarian syndrome (PCOS)

Background. Both non-alcoholic steatohepatitis (NASH) and polycystic ovary syndrome (PCOS) are associated with metabolic syndrome (MS) and insulin resistance (IR). Except for a few case reports, there are no systematic assessments of NASH in PCOS patients. Aim. To determine the prevalence of NASH and independent factors associated with NASH in a cohort of patients with documented PCOS. Methods. Patients with established diagnosis of PCOS and matched controls (matched for gender, age, and body mass index (BMI)) were included in the study. Causes of other liver diseases were systematically excluded by clinical and laboratory tests. Excessive alcohol use was defined as alcohol consumption of greater than 10 g/day. All liver biopsies were read by a single pathologist blinded to the clinical data. Histologic NASH was defined as steatosis with lobular inflammation and ballooning degeneration of hepatocytes with or without Mallory--Denk bodies or pericellular fibrosis. Univariate and multivariate analyses with logistic regression were performed to compare PCOS to matched controls. Results. Sixty-six patients were included in the study (34 PCOS and 32 matched controls). Of PCOS patients, 73% had a liver biopsy while 78% of the matched controls had a liver biopsy. In comparing PCOS patients to the matched controls, clinical (BMI, waist circumference, type 2 diabetes, MS, or its components, any alcohol consumption in the prior year, ethnic background, age, gender, etc.) and laboratory data (aminotransferases, ferritin, glucose, etc.) were not significantly different (p > 0.05). However, PCOS patients tended to have more histologic NASH on their liver biopsies (44.0% vs. 20.8%, p == 0.08). Independent predictors of histologic NASH in PCOS patients were elevated aspartate aminotransferase (AST), high triglycerides and small amounts of alcohol consumption (p == 0.019, 10-fold cross-validated AUC == 0.80, 95% CI == 0.56--0.94). Although about half of PCOS patients did not report any alcohol consumption, 50% did report rare alcohol use. In fact, PCOS patients with histologic NASH tended to report higher alcohol consumption per week than PCOS without NASH (3.80 +/-+/- 6.16 vs. 1.11 +/-+/- 1.87 g/week, p == 0.1). Nevertheless, these amounts of alcohol consumption were quite minimal. Conclusions. Despite similar clinical and laboratory profiles to the matched controls, PCOS patients seem to have more histologic NASH. Although alcohol consumption was rare for both PCOS and controls, even rare alcohol consumption in PCOS patients was independently associated with histologic NASH.