NICE has been considering the role of bariatric surgery for the treatment of diabetes among people with BMI under 35 kg/m2; specifically, in cases of early onset insulin resistance, and in cases that are severe and complex. The paper by Boza et al (summarised alongside) is, therefore, timely.
Physicians often react with wonder, after having pharmacotherapy withdrawn or non-authorised on dubious grounds, to find that a surgeon can wake up one morning having dreamt up a new procedure and perform it the next day. Thus, this is a retrospective analysis of operations performed since 2002 in contravention of most international guidelines.
Bariatric surgery is successful in alleviating obesity and diabetes, but not for the reasons the surgeons anticipated. The efficacy of the restrictive element of Roux-en-Y gastric bypass (RYGB) is clear: after RYGB, one channel of the “Y” contains food and the other contains gastric juices; only when the two merge in the common channel does digestion commence. This is thought to be a factor in the degree of weight loss. Theories around alterations in bile metabolism are also convincing (Kohli et al, 2013).
However, what the pioneers of RYGB could not anticipate was the regulation of gut peptides induced by mechanical interference to the gut itself. The influence of the incretin system emanating from the L cells of the intestine, which produce glucagon-like peptide-1 among other peptides, on diabetes control is now widely accepted.
The resolution of diabetes in obese individuals is an important and unique challenge. The paper by Boza et al provides some important answers (primarily that surgery can reverse type 2 diabetes in individuals with BMI <35 kg/m2) and poses some equally important questions. It seems likely, as confirmed by previous papers (Ricci et al, 2014; Zhang et al, 2014), that RYGB is a viable option in people with type 2 diabetes. In this study, excess weight loss percentage (EWLp) for individuals with BMI <35 kg/m2 was 93%, which does not suggest that they lost more weight than people with BMI >35 kg/m2 who have had surgery, just that EWLp is grossly misleading. If a person is an ounce overweight and they lose an ounce, EWLp is 100%, so the lower a person’s excess weight to begin with, the higher EWLp will be for any given weight loss.
Another anomaly in the methods of this analysis is the use of remission rate for diabetes, quoted as 53.2% for complete remission (defined as HbA1c <42.1 mmol/mol [<6%] and off medication) and 9.6% partial remission (defined as HbA1c <47.5 mmol/mol [<6.5%] and off medication). Most bariatric physicians now consider it a mistake to prematurely stop glucose-lowering drugs when HbA1c returns towards normal. Individuals should ideally remain on diabetes registers and drug treatments that are effective and do not cause hypoglycaemic episodes in order to reap the long-term benefits of ongoing screening and reduced chance of recurrence. It is interesting to note that participants in this analysis were diagnosed with type 2 diabetes at a median of only 4 years preoperatively, and that remission occurred exclusively in those not taking insulin prior to surgery.
The paper concludes: “These findings suggest that surgery should be offered earlier to diabetic patients and not when there is an impaired pancreatic function with little chance of recovery”. NICE take note.
To view the summaries of each paper, please download the PDF of this article.
