Impact of Excessive Weight Gain on Cardiovascular Outcomes in Type 1 Diabetes: Results From the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study

Abstract

Objective: Intensive treatment (INT) of type 1 diabetes reduces the incidence of cardiovascular disease (CVD) events compared with conventional treatment (CONV), but it also results in more weight gain. Our objective was to examine whether excessive weight gain from INT of type 1 diabetes is independently associated with subsequent CVD events.

Research design and methods: Quartiles (Q) of weight gain in 1,213 participants aged 18 years and older at enrollment in the Diabetes Control and Complications Trial (DCCT) were determined within randomized treatment groups (INT vs. CONV) using change in BMI from baseline to the closeout DCCT visits. Effects of this weight gain on CVD risk factors and outcomes during an additional 20 years of observational follow-up were then determined.

Results: The Q4 INT group experienced greater proportional weight gain (median change in BMI, 6.08 kg/m²), increases in CVD risk factors, and need for medications for hypertension and lipids compared with the Q1-3 INT and comparable CONV groups. Over a mean of 26 years of follow-up, the numbers of major and total CVD events were not statistically different in Q4 compared with Q1-3 of either the INT or CONV group. By year 14, however, the incident CVD event curve became significantly higher in the Q4 INT group than in the Q1-3 INT groups (P = 0.024) and was similar to that for the CONV group.

Conclusions: For the first 13 years after DCCT, INT for type 1 diabetes reduced macrovascular events compared with CONV, even when excessive weight gain occurred. After this, total CVD events significantly increased in the Q4 INT group, becoming equivalent to those in the CONV group. Longer follow-up is needed to determine whether this trend continues and results in more major CVD events.

Trial registration: ClinicalTrials.gov NCT00360815 NCT00360893.

Figure 1

BMI of participants at the DCCT baseline, DCCT closeout, and EDIC years 5, 10, and 15 by quartile of weight gain between the DCCT baseline and closeout (excessive = fourth quartile of weight gain [red lines]; minimal = first through third quartiles combined [blue lines]) in both the INT and CONV groups. Data are mean ± SE.

Figure 2

Cumulative incidence during the EDIC study of the first occurrence of any CVD event with INT (P = 0.23, log-rank test) (A) or CONV (P = 0.81, log-rank test) (B) for type 1 diabetes and of nonfatal MI, stroke, or death from CVD (MACE) with INT (P = 0.93, log-rank test) (C) or CONV (P = 0.52, log-rank test) (D) for type 1 diabetes therapy. Excessive = fourth quartile of weight gain (red lines); minimal = first through third quartiles of weight gain combined (blue lines).

Figure 3

Cumulative incidence during the EDIC study of the first occurrence of any CVD (P = 0.21, log-rank test) (A) and nonfatal MI, stroke, or death from CVD (MACE) (B), comparing the group with excessive weight gain group (fourth quartile of weight gain, n = 152 [red solid lines]) with the group with minimal weight gain (first through third quartiles combined, n = 457 [blue solid lines]) in the INT group and with all participants in the CONV group (n = 609 [green dotted lines]). Using a Cox proportional hazards model, the unadjusted difference in hazards after year 14 of EDIC follow-up between the excessive and minimal weight gain groups receiving INT was significant (P = 0.024).