Determinants of the increase in ketone concentration during <scp>SGLT2</scp> inhibition in <scp>NGT</scp> , <scp>IFG</scp> and <scp>T2DM</scp> patients

<jats:sec> <jats:title>Aim</jats:title> <jats:p> To examine metabolic factors that influence ketone production after sodium‐glucose cotransport inhibitor ( <jats:styled-content style="fixed-case">SGLT2</jats:styled-content> ) administration. </jats:p> </jats:sec> <jats:sec> <jats:title>Research design and methods</jats:title> <jats:p> Fasting plasma glucose ( <jats:styled-content style="fixed-case">FPG</jats:styled-content> ), insulin, glucagon, free fatty acid and ketone concentrations were measured in 15 type 2 diabetes mellitus ( <jats:styled-content style="fixed-case">T2DM</jats:styled-content> ) and 16 non‐diabetic subjects before and at day 1 and day 14 after treatment with empagliflozin. </jats:p> </jats:sec> <jats:sec> <jats:title>Results</jats:title> <jats:p> Empagliflozin caused a 38 mg/d <jats:styled-content style="fixed-case">L</jats:styled-content> reduction in <jats:styled-content style="fixed-case">FPG</jats:styled-content> concentration in <jats:styled-content style="fixed-case">T2DM</jats:styled-content> patients. However, it caused only a small but significant (7 mg/d <jats:styled-content style="fixed-case">L</jats:styled-content> ) reduction in the <jats:styled-content style="fixed-case">FPG</jats:styled-content> concentration in impaired fasting glucose ( <jats:styled-content style="fixed-case">IFG</jats:styled-content> ) subjects and did not affect <jats:styled-content style="fixed-case">FPG</jats:styled-content> concentration in normal glucose tolerant ( <jats:styled-content style="fixed-case">NGT</jats:styled-content> ) subjects. Empagliflozin caused a significant increase in mean plasma glucagon, free fatty acid ( <jats:styled-content style="fixed-case">FFA</jats:styled-content> ) and ketone concentrations in <jats:styled-content style="fixed-case">T2DM</jats:styled-content> subjects. However, empagliflozin did not cause a significant change in mean plasma insulin, glucagon or ketone concentrations in non‐diabetic subjects. An index that integrates change in plasma glucose, insulin and <jats:styled-content style="fixed-case">FFA</jats:styled-content> concentration at day 1 strongly correlates with plasma ketone concentration at day 1 (r = 0.85, <jats:italic> <jats:styled-content style="fixed-case">P</jats:styled-content> </jats:italic> < .001) and day 14 (r = 0.63, r = 0.01) and predicts, with 86% sensitivity and 83% specificity, subjects at the top tertile for plasma ketone concentration after empagliflozin treatment. </jats:p> </jats:sec> <jats:sec> <jats:title>Conclusion</jats:title> <jats:p> Results of the present study demonstrate that <jats:styled-content style="fixed-case">SGLT2</jats:styled-content> inhibition exerts different metabolic effects in non‐diabetic individuals as compared to diabetic patients. </jats:p> </jats:sec>