Impact of Moderate Alcohol Discontinuation on Insulin Action and Secretion in Latinos With and Without Hepatitis C

Background Insulin resistance ( IR ) is associated with hepatitis C virus ( HCV ), and Latinos are both at risk of IR and are disproportionately affected by HCV . Moderate alcohol consumption improves insulin sensitivity and may modify HCV ‐associated IR . We investigated the impact of moderate alcohol discontinuation on insulin sensitivity and secretion in Latinos using direct measurements. Methods Twenty‐five nondiabetic, noncirrhotic Latino adults without ( n = 17) or with ( n = 8) HCV underwent 3‐day metabolic assessment before and after prescription of 6 weeks of moderate alcohol discontinuation. Peripheral IR was measured via steady‐state plasma glucose ( SSPG ) and hepatic IR using endogenous glucose production during a 2‐step 240‐minute insulin suppression test. Insulin secretion was measured using graded glucose infusion test. Results Baseline mean age was 46 ± 11 years, 63% male, 29% had HCV , and mean body mass index was 27 ± 4 kg/m 2 . Compared to non‐ HCV , HCV patients had a higher median SSPG (132 vs. 98.8 mg/dl, p = 1.0), hepatic IR (13.5 vs. 11.3, p = 0.24), and insulin secretion rate ( ISR ‐ AUC , 1,290 vs. 1,250 pmol/min, p = 0.98). After confirmed alcohol discontinuation, hepatic IR was the only parameter that changed significantly (increased, mean change 2.6 ± 4.8, p = 0.02). Higher baseline alanine aminotransferase ( ALT ) was also associated with a greater change in hepatic IR (average 4.0 points/ ALT doubling, p = 0.004), and HCV was associated with a lesser change (average −7.3 points, p = 0.002), independent of ALT . Conclusions Short‐term moderate alcohol discontinuation adversely impacted hepatic IR in Latinos which was influenced by level of ALT at baseline independent of etiology. Although reduction in ALT through weight loss and HCV eradication remains a priority in improving IR , the observed nonharmful effect of moderate alcohol use represents a potentially confounding variable that warrants further study.