Sally G, a 38-year-old white woman from Atlanta, has type 2 diabetes mellitus (T2DM). She stands 64 inches tall and weighs 220 pounds. Her body mass index (BMI) is 38.2 kg/m2 and her glycated hemoglobin (A1C) level remains 8.8%—results that persist despite the use of metformin and a sulfonylurea. Sally’s BMI alone makes her a candidate for gastric bypass surgery. But how would the surgery affect her diabetes? It might be a cure, according to several groups of investigators.
Since 2012, studies of bariatric surgery’s effect on diabetes have appeared at a furious pace. These evaluations cover how much glycemic levels can be reduced with the use of bariatric surgery, the link between bariatric surgery and complete diabetes remission, and the feasibility of this approach to other subpopulations.
Results from bariatric surgery studies have received prominent billing at the most recent meeting of the American College of Cardiology (ACC), which convened in Washington, DC, in late March, and at the 74th Scientific Sessions of the American Diabetes Association (ADA), held in San Francisco in June.
A Flurry of Research
A study highlighted on the last day of the ACC meeting, and simultaneously published in the New England Journal of Medicine, involved Brigham and Women’s Hospital, Cleveland Clinic, and Harvard Medical School.1 The researchers found that gastric bypass surgery plus intensive medical therapy for diabetes, in patients with BMIs >40 kg/mg2, was significantly more effective in controlling glycemic levels after 3 years than intensive medical therapy alone.
group, only 5% and 18% of patients were able to attain an A1C level of ≤6.0% or ≤6.5%, respectively, after 3 years. In the Roux-en-Y gastric bypass combination group, however, 38% reached the ≤6.0% mark (P <.001) and 48% reached ≤6.5% (P <.003). In the sleeve gastrectomy combination group, 24% got to ≤6.0% A1C after 3 years (P <.01), and 47% achieved ≤6.5% (P <.003). The results of this study demonstrate not only how hard it is to control glycemic levels in obese patients using only lifestyle modification and intensive medical treatment, but also show the possible combined benefit of lost weight and better control through bariatric surgery.
But does this apply to Sally G? She is not currently receiving intensive medical therapy. Will bariatric surgery improve her poor glycemic control sufficiently to justify the risk associated with surgery? In the study cited above, the mean baseline A1C level was 9.3%. Both of the combined bariatric surgery/medical therapy groups registered a mean drop in A1C of 2.5%, compared with a 0.6% drop for medical therapy alone. Perhaps of greater note, other researchers have found that diabetes mellitus resolves in a proportion of obese patients undergoing gastric bypass.2-4
In a study from Cleveland Clinic and Loyola University School of Medicine, Chicago, investigators revealed a mean excess weight loss of 55% after an average follow-up of 6 months in more than 200 patients who had undergone various types of bariatric surgery. The patients’ mean A1C levels dropped from 7.5% at baseline to 6.5% after the follow-up period (P <.001), and they also registered a drop in fasting blood glucose levels of 41.1 mg/dL (P <.001) during that time. One-fourth of these patients experienced a long-term complete remission of T2DM during which they needed no hyperglycemic medications.5
Opening the Floodgates?
Investigators from Chile wanted to study the effects on T2DM of bariatric surgery in patients who were less obese, who had BMIs below 35 kg/m2. They found even greater likelihood of T2DM remission: after a 36-month follow-up, more than 50% of 100 patients had complete remission of their T2DM, 10% had partial remission, and 25% showed significant improvement.6
Researchers from Genoa, Italy,7 saw similar results in a smaller controlled study of 20 patients and 27 control individuals with diabetes. From a mean BMI of 32.9 kg/m2 and an A1C of 7.5% despite intensive medical therapy before Roux-en-Y gastric bypass, the study subjects’ mean BMI decreased to 25 kg/m2 and mean A1C level dropped to 7.0% over 36 months. Five of the 20 registered complete diabetes remission; 9 achieved control with medications, with A1C below 6.0%; and the remaining 6 had improved control with medications.
Getting patients to a glycemic level below 7.0% (without factoring in an effect from continuing medical therapy) would be enticing for endocrinologists and bariatric surgeons, as well as patients. With that in mind, will gastric bypass surgery one day be considered a primary treatment for diabetes mellitus? Health plans have well-established requirements in place for the reimbursement of gastric bypass surgery. Payers must decide whether to make bariatric surgery available for T2DM patients with a lower BMI. Michael Fine, MD, medical director of Health Net of California told Evidence-Based Diabetes Management, “Our plan covers bariatric surgery for patients with BMI >35 kg/m2 and a diagnosis of diabetes. It is unlikely we would lower that BMI [threshold] until there is long-term evidence that patients with BMIs between 30 and 35 do better with bariatric surgery than with medical management of their diabetes and weight.”
Glycemic Regulatory Changes After Bariatric Surgery
Allison Goldfine, MD, and Mary Elizabeth Patti, MD, from Joslin Diabetes Center, Boston, wrote in a commentary in Diabetes Care, “Understanding the physiologic processes and sequence of change may help inform optimal selection of candidates most likely to benefit from different surgical interventions for diabetes and weight management and [from] less invasive, safer alternative therapeutic approaches to manage obesity and T2DM.”8
This seems to be the puzzling question: how does bariatric surgery influence the physiologic processes, resulting in improved glycemic control or even remission? The most obvious answer would be that the insulin resistance associated with excess adipose tissue dissolves away with the decrease in weight. This could spare the beta cells. However—although the precise mechanism of action in bariatric surgery’s effect on glycemic control is not yet known—investigators suspect the answer may be more complicated than that.
A study published in 2014 proposes that the effect may arise from altered hepatic insulin sensitivity, which then spreads to the peripheral tissues. The combination with increased postprandial insulin secretion may be responsible for the effect.9 Researchers in that study also found that changes in other factors are involved as well, such as improved insulin secretion and a spike in postprandial glucagon-like peptide 1 (GLP-1) production.10 Yet, many other mechanisms have been proposed, as listed below and illustrated in Figure 18:
• Exclusion of inhibitory factors from the proximal intestine
• Morphological changes of the Roux limb with increased cellular size and mass, resulting in reprogramming of intestinal glucose metabolism
• Increased energy expenditure
• Changes in branched-chain amino acids
• Bile-acid composition
• Gut microflora
• Unfolded protein response in adipose tissue11
Using a pig model, Lindquist and colleagues performed Roux-en-Y or sham operations, fed the pigs low-calorie diets, and examined the changes in pancreatic tissue. They found that beta-cell mass had actually grown significantly in those undergoing Roux-en-Y, along with islet number and the number of extra-islet beta cells. These investigators also observed greater pancreatic expression of insulin and glucagon as well as a greater number of cells with the GLP-1 receptors in pigs undergoing gastric bypass surgery.12
It seems that many different pathways may be responsible for the change in glycemic control. It also seems clear that whereas some changes, like increased hepatic insulin sensitivity, happen rather soon, the others likely occur over greater timespans, as the body adapts to its new anatomy.
And these positive changes may not be limited to patients with T2DM. Patients with type 1 diabetes mellitus (T1DM) seemed to experience a positive effect in a small study published earlier this year. Harvard investigators found that severely obese women with T1DM who underwent gastric bypass also had 2 positive short-term changes after a mean of 7.7 weeks postsurgery: Their A1C levels had dropped from 8.0% to 7.1%, and they experienced an 11% reduction in BMI.13 Although both factors may help improve glycemic management, each of these patients will continue to need insulin supplementation.
Although improvement or remission of T2DM is a potential positive side effect of bariatric surgical procedures, the procedures are not without serious risk and negative side effects. Perioperative mortality and morbidity are serious concerns, for providers, patients, and payers. As discussed in a previous issue of Evidence-Based Diabetes Management, Roux-en-Y surgery is associated with a risk of substance abuse, including alcohol abuse, after surgery.14 In addition, the longer-term effects of bariatric surgery on T2DM are not yet completely known.
However, a truly long-term study of patients who had undergone gastric bypass (using various methods), the Swedish Obese Subjects (SOS) trial, may serve as a guide to the diabetes outcomes of patients decades after bypass surgery.4 This prospective, matched-cohort study recruited and followed thousands of patients from 25 surgical departments and 480 primary healthcare centers in Sweden. In a study published in June 2014, the investigators followed the outcomes of those with T2DM: 260 control patients who did not undergo gastric bypass and 343 patients who did. The patients entered the study between 1987 and 2001 and were followed through 2013. The median follow-up time for glycemic status in patients in the control and surgery groups was 10 years. The researchers also assessed patients’ T2DM complication outcomes, and the median follow-up time for this evaluation was roughly 18 years in both groups.4
The T2DM remission rates have been considerably higher in the gastric bypass group in both the 2-year and 15-year analyses (Figure 2). The improved glycemic control seemed to translate over the long term to fewer T2DM complications: The cumulative incidence of microvascular complications was 41.8 events per 1000 person-years for controls and 20.6 per 1000 person-years for those in the surgery group, yielding a hazard ratio (HR) of 0.44 (P <.001). For macrovascular complications, controls experienced 44.2 events per 1000 person-years compared with 31.7 per 1000 person-years for the surgical group (HR, 0.68; P = .001).4
It is not yet known what happens to glycemic levels in patients who undergo gastric bypass but fail to maintain their initial weight loss. Although this is a concern, the SOS trial and other population-based studies consider this in their overall evaluations.4,15 Overall, control of patients’ T2DM is better. More research is needed into the effect on T2DM in those whose weight loss is not maintained, as this may help to better define prior authorization criteria for gastric bypass surgery and to clarify the mechanism of glycemic regulation after surgery.
Predicting Who Will Experience Remission of Diabetes
As the studies discussed above confirm, not every patient undergoing bariatric surgery will be able to discard their T2DM medications. Currently, no mechanisms exist to predict who will or won’t benefit most from such surgery, in terms of T2DM improvement. Still, colleagues from the Geisinger Health System conducted a retrospective analysis to try to determine the factors that favor a T2DM remission. They evaluated the medical records of 630 Geisinger members with T2DM who underwent Roux-en-Y surgery between 2004 and 2011, for whom complete medical records were available.16
From this analysis, they developed a scoring system, dubbed DiaRem, which rates an individual patient from 0 to 22. The lowest scores seem to presage a clinical remission of T2DM within 5 years of the operation. The scores were determined based on 4 principal preoperative variables that they identified: insulin use, age, A1C level, and type of antidiabetic drugs used. Insulin use before surgery seemed to have the greatest (negative) effect on likelihood of remission.16 Additional validation testing will need to be conducted before DiaRem can be incorporated into practice.
New research presented at the June 2014 ADA meeting supported several components of the DiaRem formula. Stanford University researchers concurred that age, insulin use, and duration of insulin use were among several factors that seemed to differentiate those with partial versus complete remissions. Furthermore, some postoperative factors that seemed to be correlated with complete remission were greater weight loss, lower postop BMI, and lower triglyceride levels.17
Austrian researchers investigated a differentiation by gender, and determined that men may be more likely to experience full remission than women, although many more women undergo bariatric surgery than men. In this study’s sample, 37.7% of men initially had T2DM, compared with 15.8% of women. Two years after bariatric surgery, only 1.6% of men had diabetes compared with 2.6% of women. The researchers also observed improved insulin resistance, low-density lipoprotein cholesterol, and triglyceride levels in men compared with women (P<.03). In this study, however, each subject had a starting BMI above 45 kg/m2, so we cannot assume the results could be predictive for those with lower levels of preoperative obesity.18
The Value Proposition
The possibility of long-term benefit, including reduction or remission of long-term T2DM complications, is tantalizing. From a financial perspective, the chance to reduce the costs of treating T2DM, and of managing chronic kidney disease and heart disease common in patients with T2DM, may convince payers to cover bariatric surgery as an effective primary treatment for individuals like Sally G. But payers will likely move cautiously, especially in the short term, and still require initial treatment with intensive medical therapy. As Fine noted, “Assuming the long-term data demonstrates better outcomes on average with surgery than with medical management, we would [still] require a significant trial of participation in a program of supervised diet, exercise, and medication—including a GLP-1—without reaching an A1C goal <7%.” He added that his organization would also seek “psychological profiles [of patients with T2DM] that indicated likely adherence to the post surgery diet restrictions.”
Still, in the future, screening tools may help differentiate those patients who may experience the greatest glycemic benefit from this expensive and risky procedure. This would increase the value of bariatric surgery further to some patients with T2DM, perhaps even to some who are not morbidly obese. Answers to questions about the cost effectiveness of bariatric surgery for people with T2DM may well shift in coming years.References
1. Schauer PR, Bhatt DL, Kirwan JP, et al. Bariatric surgery versus intensive medical therapy for diabetes—3-year outcomes. N Engl J Med. 2014;370(21):2002-2013.
2. Kaska L, Kobiela J, Proczko M, Stefaniak T, SledziÅ„ski Z. Does the length of the biliary limb influence medium-term laboratory remission of type 2 diabetes mellitus after Roux-en-Y gastric bypass in morbidly obese patients? Wideochir Inne Tech Malo Inwazyjne. 2014;9(1):31-39.
3. Behbehani F, Ammori BJ, New JP, Summers LK, Soran H, Syed AA. Metabolic outcomes 2 years following gastric bypass surgery in people with type 2 diabetes: an observational cohort study [published online March 27, 2014]. QJM. http://www.ncbi.nlm.nih.gov/pubmed/24652654.
4. Sjöström L, Peltonen M, Jacobson P, et al. Association of bariatric surgery with long-term remission of type 2 diabetes and with microvascular and macrovascular complications. JAMA. 2014;311(22):2297-2304.
5. Brethauer SA, Aminian A, Romero-Talamás H, et al. Can diabetes be surgically cured? long-term metabolic effects of bariatric surgery in obese patients with type 2 diabetes mellitus. Ann Surg. 2013;258(4):628-636.
6. Boza C, Valderas P, Daroch DA, et al. Metabolic surgery: Roux-en-Y gastric bypass and variables associated with diabetes remission in patients with BMI <35 [published online April 10, 2014]. Obes Surg. http://www.ncbi.nlm.nih.gov/pubmed/24719226.
7. Scopinaro N, Adami GF, Papadia FS, et al. Effects of gastric bypass on type 2 diabetes in patients with BMI 30 to 35. Obes Surg. 2014;24(7):1036-1043.
8. Goldfine AB, Patti ME. Diabetes improvement following Roux-en-Y gastric bypass: understanding dynamic changes in insulin secretion and action. Diabetes. 2014;63:1454-1456.
9. Bojsen-Møller KN, Dirksen C, Jørgensen NB, et al. Early enhancements of hepatic and later peripheral insulin sensitivity combined with increased postprandial insulin secretion contribute to improved glycemic control after Roux-en-Y gastric bypass. Diabetes. 2014;63:1725-1737.
10. Madsbad S, Dirksen C, Holst JJ. Mechanisms of changes in glucose metabolism and bodyweight after bariatric surgery. Lancet Diabetes Endocrinol. 2014;2(2):152-164.
11. Voyias PD, Adaikalakoteswari A, Kumasaiyai W, et al. Bariatric surgery induced weight loss and systemic glucose regulates intracellular unfolded protein response within human adipose tissue from type 2 diabetes mellitus patients. Diabetes. 2014;63(suppl 1):Abstract 2006-P.
12. Lindqvist A, Spégel P, Ekelund M, et al. Gastric bypass improves beta-cell function and increases beta-cell mass in a porcine model. Diabetes. 2014;63:1665-1671.
13. Middelbeek RJ, James-Todd T, Patti ME, Brown FM. Short-term insulin requirements following gastric bypass surgery in severely obese women with type 1 diabetes [published online March 26, 2014]. Obes Surg. http://www.ncbi.nlm.nih.gov/pubmed/24668543.
14. Mehr SR. The link between a common type of bariatric surgery and increased alcohol use. Am J Manag Care. 2014;20(SP4):SP116-SP118.
15. Hedberg J, Sundström J, Sundbom M. Duodenal switch versus Roux-en-Y gastric bypass for morbid obesity: systematic review and meta-analysis of weight results, diabetes resolution and early complications in single-centre comparisons. Obes Rev. 2014;15(7):555-563.
16. Still CD, Wood GC, Benotti P, Petrick AT, et al. Preoperative prediction of type 2 diabetes remission after Roux-en-Y gastric bypass surgery: a retrospective cohort study. Lancet Diabetes Endocrinol. 2014;2(1):38-45.
17. Morton J, Rivas H, Rosas U, Garg T, Hemphill K. Complete and partial diabetes remission after bariatric surgery. Diabetes. 2014;63(suppl1):Abstract 2070-P.
18. Brix JM, Kopp HP, Krzizek EC, Schernthaner GH, Chernthaner G. Diabetes remission is higher in men compared to women after bariatric surgery. Diabetes. 2014;63(suppl1): Abstract 2083-P.