Objective: We investigated the relationship between insulin withdrawal (IW), endogenous insulin recovery, and concurrent SGLT2 inhibitor (SGLT2i) therapy.

Patients and Methods: Of the 78 type 2 diabetic patients admitted for glycemic control and subjected to meal tolerance tests (MTT) on the next day after admission and immediately before discharge using the same diabetes diet while off treatment, 39 patients who had received intensive insulin therapy after admission were subjected to retrospective analysis. Of these, 26 patients in whom insulin withdrawal had been achieved during admission were divided into those who had received concurrent SGLT2i therapy (n=19) and those who had not (n=7). Insulin secretion in terms of ΔCPR/ΔPG 0-30 min and AUC-CPR 0-2 h during the MTT was compared at admission and discharge.

Results: 1) While there was no significant difference between those with IW and those without IW in their age, BMI, HbA1c, C-peptide index and urinary CPR excretion at admission, significantly more patients with IW had received SGLT2i than those without IW (P < 0.001). Again, ΔCPR/ΔPG 0-30 min and AUC-CPR 0-2 h were significantly (p<0.001) increased in those with IW but not in those without IW. 2) In those with IW, while ΔCPR/ΔPG 0-30 min and AUC-CPR 0-2 h were significantly (p<0.05) increased in both of those with SGLT2i and those without, the time to IW was shown to be significantly shorter (p<0.01), the maximum daily insulin dose lower, and the duration of hospital stay shorter, in those with SGLT2i therapy than in those without.

Conclusions: Study results confirmed that endogenous insulin secretion was significantly restored in those with IW and that the time to IW was significantly shorter in those with concurrent SGLT2i therapy than in those without, suggesting a role for SGLT2i in achieving IW.


M. Matsui: None. H. Takahashi: None. S. Sawano: None. Y. Mori: None. K. Utsunomiya: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.