AIMS/HYPOTHESIS: Lack of insulin and infection/inflammation are the two most common causes of diabetic ketoacidosis (DKA). We used insulin withdrawal followed by insulin administration as a clinical model to define effects on substrate metabolism and to test whether increased levels of counter-regulatory hormones and cytokines and altered adipose tissue signalling participate in the early phases of DKA.METHODS: Nine individuals with type 1 diabetes, without complications, were randomly studied twice, in a crossover design, for 5 h followed by 2.5 h high-dose insulin clamp: (1) insulin-controlled euglycaemia (control) and (2) after 14 h of insulin withdrawal in a university hospital setting.RESULTS: Insulin withdrawal increased levels of glucose (6.1 ± 0.5 vs 18.6 ± 0.5 mmol/l), NEFA, 3-OHB (127 ± 18 vs 1837 ± 298 μmol/l), glucagon, cortisol and growth hormone and decreased HCO3- and pH, without affecting catecholamine or cytokine levels. Whole-body energy expenditure, endogenous glucose production (1.55 ± 0.13 vs 2.70 ± 0.31 mg kg-1 min-1), glucose turnover, non-oxidative glucose disposal, lipid oxidation, palmitate flux (73 [range 39-104] vs 239 [151-474] μmol/min), protein oxidation and phenylalanine flux all increased, whereas glucose oxidation decreased. In adipose tissue, Ser473 phosphorylation of Akt and mRNA levels of G0S2 decreased, whereas CGI-58 (also known as ABHD5) mRNA increased. Protein levels of adipose triglyceride lipase (ATGL) and hormone-sensitive lipase phosphorylations were unaltered. Insulin therapy decreased plasma glucose concentrations dramatically after insulin withdrawal, without any detectable effect on net forearm glucose uptake.CONCLUSIONS/INTERPRETATION: Release of counter-regulatory hormones and overall increased catabolism, including lipolysis, are prominent features of preacidotic ketosis induced by insulin withdrawal, and dampening of Akt insulin signalling and transcriptional modulation of ATGL activity are involved. The lack of any increase in net forearm glucose uptake during insulin therapy after insulin withdrawal indicates muscle insulin resistance.TRIAL REGISTRATION: ClinicalTrials.gov NCT02077348 FUNDING: This study was supported by Aarhus University and the KETO Study Group/Danish Agency for Science Technology and Innovation.