“Weight loss surgery can dramatically reduce the odds of developing type 2 diabetes,” BBC News reports.
This group had undergone weight loss surgery, so researchers compared them with a comparison group matched for age, sex and BMI, who did not have surgery. They looked at the development of type 2 diabetes in both groups.
Using the maximum follow-up period in the study (seven years), they found that the “surgery group” had an 80% reduced risk of developing diabetes compared with the “no surgery” group.
These findings are mainly applicable to those with a very high BMI (over 40). Results at lower BMIs (30 to 35) were still positive, but did not have statistical significance.
It's important to stress that weight loss surgery is no magic bullet and is associated with both short- and long-term risks and complications, such as unsightly excess skin.
Regardless, the results are consistent with current English guidelines, which recommend offering weight loss surgery to people with a BMI of 40 or more if a number of additional conditions are fulfilled. People with a BMI of 35 to 40 can also be offered weight loss surgery if they have other medical conditions that are compounded by obesity.
Read more about who is eligible for weight loss surgery on the NHS.
Where did the story come from?
The study was carried out by researchers from London-based University and Hospital Departments, and was funded by the UK National Institute for Health Research.
Both the BBC and the Daily Express reported the study accurately.
What kind of research was this?
This was a (matched) cohort study in a large group of obese individuals, assessing the effect of weight loss surgery (also called bariatric surgery) on the risk of developing type 2 diabetes.
Cohort studies have the ability to give an indication of cause and effect, but not direct proof. Common limitations of such a study design include high dropout rates, and the possibility of confounding – that there are other differences between the people with the different exposures that are influencing the outcomes.
That said, due to the size of the reduction in relative risk in the surgery group, it would be surprising if surgery did not have at least some influence on the study's outcomes.
What did the research involve?
The research team recruited two groups of closely matched obese adults: one group underwent weight loss surgery and one group didn't. They then analysed whether the surgery influenced if they went on to develop type 2 diabetes over the following seven years.
The study recruited adults (age 20 to 100 years) identified from a UK-wide database of family practices, who were obese (BMI ≥30 kg/m2) and did not have diabetes.
They enrolled 2,167 patients who had undergone weight loss surgery between Jan 1 2002 and April 30 2014 and matched them according to BMI, age, sex, index year and a blood glucose measure for diabetes (HbA1c) with 2,167 controls who had not had surgery. Weight loss surgical procedures included:
- laparoscopic gastric banding (n=1053)
- gastric bypass (795)
- sleeve gastrectomy (317)
In two people, procedures were undefined.
The main outcome the team were interested in was development of clinical diagnosis of diabetes, which was extracted from electronic health records.
What were the basic results?
The group reported that they found a reduction in diabetes risk in both men and women due to surgery, across age groups, and after different types of surgical procedures.
The average BMI for both groups was 43 – well above the minimum threshold level for obesity (30). People who had bariatric surgery were more likely to have high blood pressure or cholesterol, and to be treated with medications for these conditions.
Maximum follow up was seven years after surgery; however, most were followed up for less. The average (median) follow up was 2.8 years (interquartile range: 1.3 to 4.5 years).
By the end of the maximum seven-year follow-up period, 4.3% (95% confidence interval (CI) 2.9 to 6.5) of the weight loss surgery group had developed diabetes, compared with 16.2% (13.3 to 19.6) in the matched control group. This analysis took into account the time between surgery and diabetes, so gives different figures from the above.
This meant that the number of newly diagnosed diabetes cases (incidence) was significantly lower in the weight loss group compared with the controls, giving a hazard ratio of 0.20 (95% CI 0.13 to 0.3). This analysis was adjusted for confounders, including comorbid cardiovascular disease and depression, smoking, high blood pressure, and cholesterol and their associated treatments. This means that the surgery reduced the relative risk of developing diabetes by 80% compared to not having surgery.
How did the researchers interpret the results?
Their interpretation was that, “bariatric surgery [weight loss surgery] is associated with reduced incidence of clinical diabetes in obese participants without diabetes at baseline for up to seven years after the procedure".
This research suggests that weight loss surgery may reduce the risk of developing diabetes in people who are morbidly obese (with an average BMI of 43) compared with no surgery. The beneficial effect appeared to increase over time and at the maximum follow-up period assessed in the study (seven years), the relative risk of developing diabetes had reduced by 80%.
There was variation in the risk reduction depending on age, BMI and the type of procedure, but all were beneficial.
The study had many strengths, but also some key limitations.
The obese participants were sampled from a database that indicated whether or not they had surgery. The comparison group was only matched for age, sex and BMI, so it is likely that there is some other differences between these people which influenced their selection for surgery. For example, it could have been for reasons such as personal choice, inadequate trial of non-surgical measures, or being unsuitable for anaesthesia and surgery.
Despite the results being adjusted for various medical confounders that could have an influence, these other unknown and unmeasured differences may have meant the groups had different diabetes risk to begin with.
This could make it harder to be certain how much of the difference in diabetes risk is specifically down to the effect of surgery, and how much is due to other influences.
It is also important to recognise that the results do not apply to all people who are categorised as obese. The average BMI of recruits was high overall, at 43, meaning the results may be less applicable to people with BMIs at the lower end of the obesity scale. Further evidence of this came from a sub-analysis by BMI category. They found significant risk reductions in BMI groups 35 to 39.9, and 40 and above. At BMI levels 30 to 34.9, there was still a 60% or so reduction in risk reported, but this failed to meet statistical significance, meaning it may be a chance finding.
However, in any case, most people with BMIs below 35 are not currently eligible for bariatric surgery on the NHS, in line with UK guidance.
A further factor to bear in mind is that the control group were not offered any intervention at all, such as an intensive weight loss programme. Hence, the results tell us how much surgery is better than doing nothing, rather than if it is better than specific non-surgical alternatives, such as the NHS Choices diet and exercise plan.
The results are consistent with current English guidelines, which recommend offering weight loss surgery to people with a BMI of 40 or more if a number of additional conditions are fulfilled. People with a of BMI 35 to 40 can also be offered weight loss surgery if they have other medical conditions. For full details, see Weight Loss Surgery – who can use it?
As with any surgery, weight loss surgery has risks. The balance of risks and potential benefits would need to be discussed between doctor and patient on a case-by-case basis. Information from studies like this may inform the conversation.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 3 November 2014
Daily Express, 3 November 2014
Links to the science
The Lancet – Diabetes & Endocrinology. Published online November 3 2014