New Score Predicts Diabetes Remission Post Bariatric Surgery

Larry Hand

September 12, 2013

Researchers in Pennsylvania have developed a tool comprising 4 preoperative clinical variables that surgeons and patients can use to predict the likelihood of type 2 diabetes remission after Roux-en-Y gastric-bypass surgery.

Christopher D. Still, DO, director of Geisinger Obesity Institute, Danville, Pennsylvania, and colleagues developed their algorithm, known as the DiaRem score, on the basis of a retrospective cohort study of 690 patients who underwent gastric-bypass surgery. They verified the results in 2 additional cohorts; their findings were published online September 13 in the Lancet Diabetes and Endocrinology.

DiaRem scores range from 0 to 22, with low scores consistently predicting higher remission rates and high scores predicting lower remission rates.

"Bariatric surgery is a very effective tool not only for weight loss but also for curing or resolving comorbid medical problems," Dr. Still told Medscape Medical News in a telephone interview. "The surgery is the best we have for long-term success, but it's not without potential risks and costs.

"The system we developed, the DiaRem the physician and a patient a tool to say, 'Let's see if we're actually going to remit diabetes before we go into this drastic realignment of the intestine and the stomach.' Now we can predict that," senior author George Argyropoulos, PhD, also of Geisinger, explained to Medscape Medical News.

But the study should not be misinterpreted to suggest that only patients who will achieve diabetes remission should undergo this type of surgery, stress 2 experts in an accompanying comment.

"Often those whose diabetes is the most difficult to control are the ones who could benefit most from surgery in the short and long terms, even if they do not achieve diabetes remission," say Dimitri J. Pournaras, PhD, of Imperial College London, United Kingdom, and Carel W. le Roux, PhD, of University College Dublin, Ireland. "A focus on so-called hard end points — eg, cardiovascular risk, diabetes-related complications, and mortality — is the way forward," they add.

259 Variables Whittled Down to Just 4

The researchers conducted a retrospective cohort study of 690 patients who underwent gastric-bypass surgery at Geisinger between January 1, 2004, and February 15, 2011, and whose electronic medical records were available for analysis. Of the 690 patients, 436 (63%) achieved partial or complete diabetes remission. The patients all had high body mass indexes (BMIs) and were mainly white. The mean age of the patients was older than 50 years.

The researchers examined 259 clinical variables, including 51 comorbidities, 93 drugs, 78 laboratory measurements, 19 survey scores, and 18 other factors such as age, sex, and tobacco and alcohol use. Using multiple logistic regression models, they attempted to identify predictors of early remission (beginning within 2 months) and late remission (beginning after 2 months) of diabetes.

There were 4 preoperative clinical variables included in their final model: insulin use, age, HbA1c concentration, and type of diabetic drugs used. BMI was not a strong predictor of remission so was not included, the researchers write; the latter finding lends supports to the argument of those who say that BMI is not appropriate as a selection criterion for Roux-en-Y gastric bypass, they note.

Of patients whose DiaRem score was 0-2, 88% achieved early remission, as did 64% of those who scored 3-7, 23% of those who scored 8-12, 11% of those whose score was 13-17, and just 2% of patients who scored 18-22.

As in the primary cohort, the proportion of patients achieving remission in the replication cohorts was highest for the lowest scores and lowest for the highest scores.

The greatest weight given to the score was from insulin use before surgery; the researchers estimate that 42.4% of patients not using insulin achieved early remission, compared with just 4.4% of patients using insulin. After 5 years, 77.7% of patients not using insulin achieved complete remission, compared with 15.4% of patients using insulin.

Early and late remission also correlated with younger age, low HbA1c concentration, and high serum insulin concentration. And use of insulin-sensitizing drugs (but not metformin) and sulfonylureas was associated with a lower chance of remission. Loss of excess body weight after surgery was also associated with late remission.

The researchers note that the DiaRem score method does not apply to sleeve gastrectomy, but they plan to develop another scoring method for that in the future.

A Simple Decision-Making Tool for Docs and Patients

The chief biostatistician for the study, G. Craig Wood, also of Geisinger, told Medscape Medical News that this research is unique, representing a different thought process to the average work. "We're trying to get to really good science but also make it adaptable and useful in a clinical sense. We wanted to make sure…the scoring [is] not too complex, so the [surgeon and patient] can arrive at the score without getting a calculator."

Dr. Still agrees. "This is a decision-making tool for both surgeons and patients to decide whether or not [Roux-en-Y gastric-bypass surgery] has a high chance of curing one's diabetes," he stressed.

"A lot of people only focus on weight loss and don't think about the profound medical benefits that bariatric surgery can achieve. This goes out 5 years. That gives you a pretty good handle on what your chances are for resolving diabetes with gastric bypass."

This research was funded by Geisinger Health System and the National Institutes of Health. None of the authors report any relevant financial relationships.

Lancet Diab Endocrinol. Published online September 13, 2013. Full article



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.