Bariatric Treatment Seems to Stymie Type 2 Diabetes

/in , /by

Obese patients who underwent bariatric surgery had an 83% lower incidence of type 2 diabetes compared with a control group that did not have weight-loss surgery, data from a large Swedish study showed.

Almost four times as many patients in the control group developed type 2 diabetes during 15 years of follow-up. Patients who had impaired fasting glucose (IFG) prior to surgery appeared to benefit most from bariatric surgery’s preventive effect on type 2 diabetes, reported Lars Sjöström, MD, PhD, of the University of Gothenburg, and colleagues in the Aug. 23 issue of the New England Journal of Medicine.

“Among patients with impaired fasting glucose, bariatric surgery reduced the risk by 87%, and type 2 diabetes did not develop in approximately 10 of 13 obese patients who underwent bariatric surgery. This risk reduction is at least twice as large as that observed with lifestyle interventions in moderately obese, prediabetic persons,” they said.

Obese individuals in a prediabetic state have a high risk of type 2 diabetes, which can be reduced by 40% to 45% with lifestyle changes or medication. The benefits persist, at least in part, for 3 to 15 years, the group explained.

However, most trials of lifestyle interventions and weight-loss drugs have involved moderately obese patients. For the severely obese, only bariatric surgery has been shown to achieve large, sustained weight losses, the authors continued.

Two studies reported earlier this year showed substantial rates of diabetes remission and improved metabolic control in obese diabetic patients after bariatric surgery.

A substantial volume of evidence has documented bariatric surgery’s effect on diabetes remission, but few studies have demonstrated the effect of surgery on preventing diabetes. The paucity of data reflect, in part, the lack of long-term follow-up and control groups for most of the studies.

The prospective, nonrandomized, controlled Swedish Obese Subjects (SOS) trial began in 1987, and investigators had enrolled 4,047 obese patients by Jan. 31, 2001. Investigators also enrolled 5,335 patients with characteristics similar to those of the intervention group into a control group that received usual care.

An initial report from the trial, focusing on the effect of surgery on cardiovascular risk factors, showed a reduction in the risk of type 2 diabetes in the intervention group (N Engl J Med 2004; 351: 2683-2693).

The follow-up analysis examined the long-term effects of bariatric surgery on the risk of developing diabetes, a prespecified secondary endpoint of the study (N Engl J Med 2007; 357: 741-752).

The diabetes analysis included 1,658 patients who underwent bariatric surgery (primarily vertical banded gastroplasty) and 1,771 study participants from the control group. Beginning 6 months after enrollment, participants in both groups had regularly scheduled follow-up examinations out to 15 years.

Laboratory assessments, including blood glucose levels, were performed after 2, 10, and 15 years. Investigators defined type 2 diabetes as documentation of a fasting blood glucose level >e;110 mg/dL, a fasting plasma glucose level >e;126 mg/dL, or initiation of diabetes medication.

Patients in the intervention arm had a maximal mean weight loss of 68 pounds after 1 year, and weight loss averaged 44 pounds at 10 and 15 years. In the control group, weight gain or loss never exceeded 6 to 7 pounds.

After a median follow-up of 10 years, 392 patients in the control group had developed type 2 diabetes compared with 110 in the intervention arm, corresponding to incidences of 28.4 and 6.8 cases per 1,000 person-years, respectively.

The difference translated into an unadjusted hazard ratio of 0.22 for the intervention group and an HR of 0.17 by multivariate analysis (P<0.001 for both values). Among patients with IFG at baseline, bariatric surgery was associated with an HR of 0.13, or an 87% reduction in the risk of type 2 diabetes.

Analysis by type of bariatric procedure showed significant risk reductions irrespective of the surgery, resulting in HRs of 0.12 for gastric bypass, 0.20 for gastric banding, and 0.25 for vertical banded gastroplasty (P<0.001 for each type of surgery).

The multivariate analysis identified six predictors of diabetes incidence:

  • Surgery versus usual care: HR 0.17 (95% CI 0.13 to 0.21, P<0.001)
  • Older age: HR 1.23 (95% CI 1.12 to 1.36, P<0.001)
  • Baseline blood glucose: HR 2.18 (95% CI 1.99 to 2.39, P<0.001)
  • Waist-to-hip ratio: HR 1.18 (95% CI 1.06 to 1.32, P=0.003)
  • Urinary albumin excretion: HR 1.24 (95% CI 1.10 to 1.39, P<0.001)
  • High density lipoprotein (HDL) cholesterol: HR 0.87 (95% CI 0.78 to 0.98, P=0.02)

 

The postoperative mortality was 0.2%. In addition, 2.8% of patients who underwent bariatric surgery required re-operation within 3 months because of complications.

Factors that did not significantly influence diabetes risk included body mass index, insulin level, and leisure-time physical activity.

The study had some limitations: The diagnosis of type 2 diabetes was based on fasting glucose levels and patient self-reports of diabetes medication use.

“The long-term findings from the Swedish Obese Subjects (SOS) study are both provocative and exciting — especially the findings that suggest that bariatric surgery may prevent the conversion of abnormalities in glucose metabolism to frank diabetes,” wrote Danny Jacobs, MD, MPH, of Duke University, in an accompanying commentary.

“However, it remains impractical and unjustified to contemplate the performance of bariatric surgery in the millions of eligible obese adults. And to be certain, the authors do not suggest such an approach,” he said.

Jacobs emphasized that more research is needed to determine which obese patients are most likely to benefit from bariatric surgery. He pointed out that surgery did not prevent diabetes in all patients, underscoring the multifactorial nature of obesity.