Miriam E Tucker
May 08, 2017
Findings in support of the DiaRem score were presented May 6 here at the American Association of Clinical Endocrinologists (AACE) 2017 Annual Scientific & Clinical Congress by Thiyagarajan Thangavelu MD, of Hennepin County Medical Center, Minneapolis, Minnesota.
The DiaRem score, developed by Christopher D Still, DO, director of Geisinger Obesity Institute, Danville, Pennsylvania, and colleagues, uses just four preoperative clinical variables — insulin use, age, HbA1c, and diabetes medication type — to generate a score of 0 to 22, where the lower the score means the greater the chance of remission 1 year after Roux-en-Y gastric-bypass surgery.
The Geisinger group have subsequently published 8-year validation data, but the few external validation studies have produced mixed results and haven't included surgery types other than gastric bypass, Dr Thangavelu noted.
"Bariatric surgery can have perioperative and long-term complications and significant cost, so it's important to see which candidate patients will actually benefit. Among the various models, the DiaRem score is very easy to use in daily clinical practice. We've demonstrated that it is predictive of which patients will have diabetes remission," he said during his presentation here.
Indeed, Gregory Dodell, MD, assistant clinical professor of medicine, endocrinology, diabetes, and bone disease at the Icahn School of Medicine at Mount Sinai, New York told Medscape Medical News that the DiaRem score could help facilitate discussion with patients, who often ask whether they'll be able to come off their diabetes and cholesterol medications following bariatric surgery.
"We've kind of intuitively known that if a patient has had diabetes for a long time, is on insulin, or their A1c is high, it's going to be a little less likely they'll get off those medications. But if you have a score you can put in front of the patient, and you can say 'these are your odds,' it lets them make a more educated decision about whether they want to go through a life-changing operation."
Score Predicts Remission Similarly to Original Cohort
The current study was a retrospective review of electronic health records of adults with preoperative type 2 diabetes who underwent bariatric surgery at Hennepin County Medical Center from 2006 to 2015. Of the 95 patients (68 women and 27 men) for whom 1-year data were available, 72 (76%) had undergone Roux-en-Y gastric bypass, while 23 (24%) had other bariatric procedures, including adjustable gastric band and sleeve gastrectomy.
At baseline, their body mass indexes ranged from 36 to 50 kg/m2, and they had an average HbA1c of 7.48%. Overall, the population is fairly comparable to the one used by the Geisinger group to develop the score, Dr Thangavelu noted.
Complete and partial type 2 diabetes remission were defined as HbA1c of less than 6.0% and 6.0% to 6.5% respectively, with no glucose-lowering medications.
The patients with higher DiaRem scores were less likely to achieve diabetes remission. When divided into five groups based on score (0–2, 3–7, 8–12, 13–17, and 18–22), remission rates were 100%, 83%, 64.2%, 59.2%, and 16.6%, respectively. The score was significantly predictive of both partial and complete remission (P < .001 for both).
"Across all groups, the remission rate is similar to the original cohort," Dr Thangavelu said.
While HDL cholesterol and triglycerides showed some improvement at 1 year, LDL cholesterol, total cholesterol, and blood pressure did not change significantly, and the DiaRem score wasn't predictive of those.
Dr Dodell noted that because those parameters were fairly well-controlled to begin with — mean LDL was 108 mg/dL, systolic blood pressure 132 mm Hg, and diastolic 80 mm Hg — "you're not going to see much improvement," adding, "I'd be curious to see things like testosterone or sleep apnea to see whether the score applies to those."
During the discussion period, Dr Thangavelu said that his team plans to further analyze their data by prior medications the patients were taking and also by demographic factors such as race. And he said they are examining 5-year data in their population.
Overall, Dr Dodell commented, "This was a validation study. You can use the score and see how it works in your own practice. It's useful, just as we use FRAX for osteoporosis or any other algorithm. It's a tool. It's probably not perfect, but it opens up a conversation to have with patients."
Dr Thangavelu and Dr Dodell have no relevant financial relationships.