The natural history of type 2 diabetes (T2D) is mostly progressive and patients often die from diabetes-related complications. However, T2D can sometimes improve enough to discontinue medications, and this phenomenon remains poorly understood. Using health examination data from the Korean National Health Insurance Service (KNHIS) for a new-onset T2D cohort, we investigated the incidence of diabetes remission and associated clinical factors. Remission was defined as normal fasting blood glucose levels < 126 mg/dL, an alternative criterion for remission based on expert consensus; patients who reached normal fasting blood glucose levels measured at 2-year intervals after discontinuation of treatment were defined as achieving remission. New-onset T2D patients (N=114,874) from 2009 to 2012 were followed until 2017, and 2,429 patients (2.1%) experienced remission during the study period. No significant difference in age was reported between the remission and non-remission groups. Compared to the non-remission group, the remission group had lower baseline body mass index (25.4 kg/m2 vs. 25.7 kg/m2), lower baseline fasting glucose levels (157.7 mg/dL vs. 177.1mg/dL), and a lower prevalence of hypertension (47.7% vs. 52.2%) and hyperlipidemia (44.4% vs. 48.2%). When investigating treatment methods, the remission rate was lower in patients who used more than three oral hypoglycemic agents including sulfonylurea (5.43% vs. 14.4%), while the rate of insulin users was higher in the remission group (10.5% vs. 8.0%). When weight change within 2 years before and after remission was reviewed through health examination results, the group that succeeded in weight loss of 5% or more had a significantly higher rate of remission.

In summary, remission was observed at rates as low as 2.1% in patients in standard medical care settings. Weight loss and insulin therapy in early T2D patients are thought to lead a favorable prognosis in terms of remission.

Disclosure

K. Song: None. J. Kim: None. K. Han: None. H. Kwon: None.

Funding

National Research Foundation of Korea (2021R1A2C2013890)

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.