Patients with NASH are characterized by metabolic inflexibility, which translates into improper nutrient use when switching from fasting to postprandial states. In this study we assessed ketone levels during an oral glucose tolerance test (OGTT) in patients with vs. without NASH.


Overweight or obese patients (n=41) underwent a 2-hour OGTT with 75-grams of glucose and liver 1H-MRS. A small subset of patients also underwent a liver biopsy. Plasma insulin, c-peptide, and free fatty acid (FFA) were measured by ELISA. Beta-hydroxybutyrate (BHB) concentration was measured by gas chromatography and mass spectrometry.


Patients (age: 55±10 years; 65% male; BMI: 34.0±5.6 kg/m2; A1c: 6.3±1.0%) were divided based on the severity of liver disease: No NAFLD (n=29), isolated steatosis (n=5), and NASH (n=7). They were well matched for important clinical parameters, such as age (p=0.21), gender (p=0.99), and prevalence of diabetes (p=0.23). In the fasting state, patients with NASH only showed a trend towards higher BHB levels compared to patients with isolated steatosis and those with no NAFLD (137±31 vs. 103±28 vs. 101±44 µM, p=0.11, respectively). However, differences were unmasked during the OGTT, with significantly higher levels in patients with NASH (p≤0.02 for all time points: 30, 60, 90 and 120 min). This translated into a significantly higher BHB exposure over time when expressed as area under the curve (AUC) during the OGTT (14.7±3.9 vs. 10.3±2.9 vs. 9.1±3.1 mM x min, p=0.001). More important, BHB AUC showed an excellent correlation with the degree of liver fibrosis by histology (r=0.71, p=0.007).


Patients with NASH showed increased ketogenesis over time. This is likely a compensatory mechanism to deal with FFA overload, typical in these patients. If confirmed in larger studies, it would suggest that patients with NASH require overactive mitochondria to compensate for nutritional overload. In turn, this may be responsible for the development of liver fibrosis.