Early weight loss, diabetes remission and long-term trajectory after diagnosis of type 2 diabetes: a retrospective study

Aims/hypothesis Weight loss can improve glycaemic management in individuals with type 2 diabetes, but its long-term effects on remission, cardiovascular risk factors and complications remain unclear. We investigated clinical outcomes following non-interventional >= 10% body weight loss in people with newly diagnosed type 2 diabetes in a routine care setting. Methods We retrospectively analysed two cohorts of people with newly diagnosed type 2 diabetes. After exclusions, cohort 1 included 1934 individuals followed for up to 25 years; cohort 2 comprised 13,277 individuals followed for up to 10 years. Participants were categorised into two groups based on whether or not they lost at least 10% body weight. In a sensitivity analysis, a group of participants with intermediate weight loss (5% to <10%) was also considered. Outcomes included HbA(1c), diabetes remission, cardiovascular parameters and chronic complications. Results Participants (58% male) had a mean age of 62 years and a mean diabetes duration of <2 years at inclusion; mean baseline HbA(1c) was 57-64 mmol/mol (7.4-8.0%) and mean BMI was similar to 30 kg/m(2). Weight loss >= 10% was obtained in 15.9% (n=308) of participants in cohort 1 and in 8.8% (n=1167) in cohort 2. In cohort 1, weight loss >= 10% was associated with a sustained reduction in HbA(1c) (mean difference 2.1 mmol/mol; 0.19%) and a higher remission rate than in the <10% weight loss group (20.2% vs 5.5%; HR 4.2). These findings were confirmed in cohort 2, with remission rates of 13.2% and 4.1% (HR 2.6) in the >= 10% and <10% weight loss groups, respectively. Weight loss >= 10% improved systolic BP and HDL-cholesterol and triglyceride levels. Participants with weight loss of 5% to <10% (28.2% in cohort 1 and 17.4% in cohort 2) had marginal improvements in HbA(1c), lipids and remission rates compared with participants with weight loss <5%, and such results were inferior to those achieved with weight loss >= 10%. In cohort 1, compared with weight loss <5% (reference), the HR for remission was 5.2 with weight loss >= 10% vs 1.7 with weight loss 5% to <10%. Weight loss >= 10% was not associated with a reduced incidence of complications. On the other hand, remission was independently associated with a significantly lower rate of new-onset microangiopathy (adjusted HR 0.84; 95% CI 0.73, 0.97; p=0.019). Conclusions/interpretation Early weight loss of >= 10% in type 2 diabetes was associated with sustained glycaemic improvements, increasing by three to four times the rates of diabetes remission. Remission, in turn, more than weight loss was associated with a reduced risk of complications.