In 1921, diabetes was a terminal condition. In 1922, thanks to Frederick Banting and his team’s work on insulin, it became treatable. A cure for diabetes is still a pipe dream, but bariatric surgery is arguably getting close to a cure for type 2, specifically.
The STAMPEDE (Surgical Treatment and Medications Potentially Eradicate Diabetes Efficiently) study (Schauer et al, 2012; summarised on the facing page) presents new evidence for the management of diabetes in obese individuals by medical versus surgical means. The word “cure” is still forbidden, as the duration of resolution by any means is unknown and referring to a “cure” implies that long-term follow-up and risk management is no longer required, which is not the case. Nevertheless, a primary outcome target of HbA1c ≤42 mmol/mol (6.0%) as aspired to here, represents normoglycaemia. STAMPEDE’s results are remarkable and should change practice and be considered by those needing to balance cost and clinical effectiveness in austere times.
Walter Pories, pioneer of modern bariatric surgery research, once called an early paper on gastric bypass (GB) “Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus” (Pories et al, 1995). This was not a very scientific title, and slightly smug as the diabetes-resolving effect was a chance finding and the putative hormonal mechanism of action is still unknown. It is true, however, that the ability of surgery to modify the enteroinsular axis in some way is remarkable, and demonstrated here to an impressive degree. An average post-GB HbA1c of 46 mmol/mol (6.4%) in previously uncontrolled diabetes with baseline HbA1c of 77 mmol/mol (9.2%), is the best long-term treatment that can be offered; 49 mmol/mol (6.6%) with sleeve gastrectomy (SG) is not far behind.
SG describes the removal of the greater curvature of the stomach to restrict food intake, possibly lowering ghrelin secretion to reduce hunger. Originally designed to be the first part of a two-part procedure in patients too overweight for immediate GB, it was observed as safe and successful in its own right. A benefit of SG is maintenance of the stomach architecture – the cardiac and pyloric sphincters remain in control of gastric function. Opinions on the reason for the enhanced and rapid effects of GB on diabetes vary. The “hindgut theory” suggests an increase in the release of incretin hormones including GLP-1 from the L-cells of the intestine as nutrients are delivered more rapidly to the lower small intestine. The “foregut theory” describes the elimination of a putative “anti-incretin” hormone which, if it exists, promotes fat deposition and thus insulin resistance. Pories’ study hints at a world in which a patient with an inguinal hernia will be referred by the GP for a curative procedure, and a newly diagnosed person with diabetes will be referred for his or her definitive operation.
Evidence supporting Pories’ paper is increasing. The SOS (Swedish Obese Subjects) trial (Sjöström et al, 2004), looking at patients who underwent basic bariatric procedures >20 years ago, has shown that rates of hypertriglyceridaemia, diabetes and hyperuricaemia were lower in the surgically treated group than in controls after up to 10 years. A study by Adams et al (2007) showed that long-term all-cause mortality with surgery was reduced by 40% compared with controls, and that cause-specific mortality also decreased for coronary artery disease, diabetes and cancer. In 2009 a review by Picot et al concluded that bariatric surgery is more effective for weight loss than non-surgical options, is better at improving and resolving comorbidities and is cost-effective. The National Bariatric Surgery Registry (NBSR) analysed data from 8710 operations over a year and showed the effects of obesity surgery in treating a whole range of life-threatening diseases, including
type 2 diabetes (NBSR Data Committee, 2010).
Other notable papers provide the context in which STAMPEDE needs to be judged. Adjustable
gastric bands have been shown by Dixon and O’Brien (2002) to induce resolution of diabetes in 73% of patients, but only in drug-naïve, newly diagnosed individuals. There is no doubt that weight loss following gastric band insertion can lead to significant improvement in diabetes, but this seems to be dependant on weight loss. The Buchwald et al (2004) meta-analysis confirmed GB as the most effective of the commonly performed operations, demonstrating resolution of diabetes in 83.7% of patients, but the paper predated routine SG.
Recently, Himpens et al (2010) gave an appraisal of the sleeve, describing excellent initial weight loss, but questionable long-term maintenance. This put the effectiveness of SG under scrutiny. Did the operation provide long-term benefit, and did GB not offer far superior resolution of diabetes? STAMPEDE was published in this context.
The STAMPEDE results are impressive because of the effects of SG on diabetes. Patients who underwent SG had a non-significantly lesser degree of weight loss than those who underwent GB bypass, with good normalisation of glucose. Furthermore, 72% of the SG group were able to stop glucose-lowering therapy, in contrast to the bypass group who were all able to stop all medications. Although STAMPEDE seems to confirm bypass as the gold standard, it suggests an elevation of status of SG. Similarly impressive was the weight loss achieved by medical means in people with poorly controlled diabetes, of whom 12% achieved an HbA1c≤42 mmol/mol (6%), possibly through the frequent use of newer agents such as GLP-mimetics. Another observed outcome was the enhanced reduction of medication for cardiovascular (CV) risk. However, this is a complex area; studies like SOS demonstrate that, although a person’s global CV risk is significantly reduced by bariatric surgery, he or she is still at higher risk than the general population, so drugs should not necessarily be stopped on normalisation of a particular risk factor. The long-term treatment of a permanently post-obese individual needs careful study.
In summary, the SG is a relatively safe procedure, without long-term results to prove its effectiveness, although the GB is still judged to be superior in terms of weight loss, weight maintenance and resolution of comorbidities. STAMPEDE adds evidence relating to how the SG, an operation in its relative infancy, fits within the firmament of bariatric surgery, and paints a more optimistic picture than other studies.
To view the summaries of each paper, please download the PDF.
