Weight Fluctuations May Be Deadly in Type 2 Diabetes

Miriam E. Tucker

January 30, 2019

Body weight fluctuations may be independently predictive of worse outcomes in patients with type 2 diabetes, new research suggests.

The findings, from a novel post-hoc analysis of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, were published in the February issue of the American Journal of Cardiology by Phyllis Yeboah, MD, of Wake Forest Baptist Health, Winston-Salem, North Carolina, and colleagues. 

The glycemic control group of the landmark ACCORD trial of 10,251 participants with type 2 diabetes was stopped early because of an unexpected increase in mortality in the group randomized to intensive glycemic control. 

The current study is believed to be the first to examine ACCORD data for the relationship between change in body weight and outcomes.

"This study confirms what we all know, that obesity isn't good and weight gain should be discouraged in people with diabetes. But there is also a need to minimize weight fluctuations...The more you go up and down, the worse you do," senior investigator Joseph Yeboah, MD, a cardiologist at Wake Forest, told Medscape Medical News.

The implications, Yeboah said, are to be careful that advice given to patients to diet doesn't turn into a "yo-yo" situation.

"It's good [for them] to lose weight, but it's not good for a patient with diabetes to have their weight fluctuate up and down. Maintaining a BMI of 28 kg/m2 is better than going to 25 kg/m2, then 35 kg/m2, then back again."

Does Pushing to Lower HbA1c Lead to Weight Gain, Poorer Outcomes?

At baseline, 8.9% of ACCORD participants were normal weight, 29.1% were overweight, and 62% were obese.

Mean baseline BMI, mean change in body weight, and body weight variability during ACCORD were 32.2 kg/m2, 0.1 kg, and 3.4 kg, respectively.

After a mean 3.7 years of follow-up, 10.2% of patients had experienced one of the primary outcome events (nonfatal MI or nonfatal stroke or CV death), 4.3% had congestive heart failure, and 60.7% reported a microvascular event.

Overall, 7.0% of patients died from any cause.

Deaths occurred in 5% of those in the intensive treatment group versus 4% in the standard glycemic control group.

But notably, the current analysis reveals that all individuals who died during the trial were in the highest quartile for weight gain (mean, 8.8 kg) during the 3.7 years of follow-up before the trial was halted.

A previous ACCORD analysis had found that weight gain was greater in the intensive than the standard glycemic control group and that patients taking insulin or thiazolidinediones were most likely to gain weight (Diabetes Care. 2013;36:2162-2168). However, not all of the weight gain was in the intensive treatment group, Yeboah noted.

"I think we can say there's an association, that the intensive control arm was associated with weight gain, and weight gain is associated with worse outcomes."

"This suggests a possibility that weight gain could have contributed to poor outcomes in ACCORD," he said, adding, "we need more data on whether if you push hard to lower HbA1c you're likely to gain weight and whether that predicts outcomes." 

Weight Fluctuations Also Associated With Worse Outcomes

Next, the researchers explain, "Our study used the publicly available data of the ACCORD trial to show that while obesity is a risk factor for complications in persons with diabetes mellitus, the fluctuation of body weight (body weight variability) which often accompanies weight loss prescription may even be associated with a higher risk for macrovascular and microvascular complications."

Specifically, after full adjustment for confounders, including BMI, body weight variability (per unit standard deviation) was associated with increased risks of a primary outcome event (hazard ratio [HR], 1.25), heart failure (HR, 1.59), death (HR, 1.74), and microvascular events (HR, 1.18) (all P < .0001).

And participants who had normal baseline weight but were in the highest quartile of body weight variability had nearly three times the risk for a primary outcome event compared with those who were obese at baseline but remained in the lowest quartile for body weight variability (HR after adjustments, 2.91; P = .007).

Yeboah cited previous findings showing that when people lose weight and then regain it, they end up with more fat and less lean mass than they had before, even at the same weight. "That's one explanation for why weight fluctuations are harmful," he said.

Bottom-line, Yeboah said, the approach must be individualized. "Our recommendations shouldn't be a one-size-fits-all 'lose weight.' If the patient is motivated to lose and maintain, good. But if they lose and gain, let them know there are risks associated with that."

The authors have reported no relevant financial relationships.

Am J Cardiol. 2019;123:576-581. Abstract

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.