Aims/hypothesis Diabetic ketoacidosis (DKA) is often caused by concomitant systemic inflammation and lack of insulin. Here we used an experimental human model to test whether and how metabolic responses to insulin are impaired in the early phases of DKA with a specific focus on skeletal muscle metabolism. Methods Nine individuals with type 1 diabetes from a previously published cohort were investigated twice at Aarhus University Hospital using a 120 min infusion of insulin (3.0/1.5 mU kg) after an overnight fast under: (1) euglycaemic conditions (CTR) or (2) hyperglycaemic ketotic conditions (KET) induced by an i.v. bolus of lipopolysaccharide and 85% reduction in insulin dosage. The primary outcome was insulin resistance in skeletal muscle. Participants were randomly assigned to one of the two arms at the time of screening using www.randomizer.org. The study was not blinded. Results All nine volunteers completed the 2 days and are included in the analysis. Circulating concentrations of glucose and 3-hydroxybutyrate increased during KET (mean ± SEM 17.7 ± 0.6 mmol/l and 1.6 ± 0.2 mmol/l, respectively), then decreased after insulin treatment (6.6 ± 0.7 mmol/l and 0.1 ± 0.07 mmol/l, respectively). Prior to insulin infusion (KET vs CTR) isotopically determined endogenous glucose production rates were 17 ± 1.7 μmol kg (p = 0.77) and urea excretion rates were 16.9 ± 2.4 g/day vs 7.3 ± 1.7 g/day (p = 0.01). Insulin failed to stimulate forearm glucose uptake and glucose oxidation in KET compared with CTR (p < 0.05). Glycogen synthase phosphorylation was impaired in skeletal muscle. Conclusions/interpretation In KET, hyperglycaemia is primarily driven by increased endogenous glucose production. Insulin stimulation during early phases of DKA is associated with reduced glucose disposal in skeletal muscle, impaired glycogen synthase function and lower glucose oxidation. This underscores the presence of muscle insulin resistance in the pathogenesis of DKA.