PHILADELPHIA – A nomogram that assigns a disease severity score to individuals with type 2 diabetes may provide a tool that helps surgeons, endocrinologists, and primary care physicians determine which weight-loss surgical procedure would be most effective, according to an analysis of 900 patients from Cleveland Clinic and University Hospital Clinic, Barcelona, reported at the annual meeting of the American Surgical Association.
“This is the largest reported cohort with long-term glycemic follow-up data that categorizes diabetes into three validated stages of severity to guide procedure selection,” said Ali Aminian, MD, of Cleveland Clinic. The study also highlighted the importance of surgery in early diabetes. The study involved a modeling cohort of 659 patients who had bariatric procedures at Cleveland Clinic from 2005 to 2011 and a separate data set of 241 patients from Barcelona to validate the findings. Roux-en-Y gastric bypass (RYGB) was performed in 78% of the Cleveland Clinic group and 49% of the Barcelona group, with the remainder having sleeve gastrectomy (SG).
Dr. Ali Aminian
RYGB and SG account for more than 95% of all bariatric procedures in people with type 2 diabetes, Dr. Aminian said, but outcomes of clinical trials have been variable, some reporting up to half of patients having long-term relapses. The Cleveland Clinic study involved all patients with type 2 diabetes who had RYGB or SG from 2005 to 2011 with 5 years or more of glycemic data, with a median follow-up of 7 years. The study used American Diabetes Association targets to define remission and glycemic control.
“Long-term response after bariatric surgery in patients with diabetes significantly differs according to diabetes severity,” Dr. Aminian said. “For example, the outcome of surgery in a patient who has diabetes for 2 years is significantly different than a patient who has diabetes for 15 years taking three medications, including insulin.”
The researchers generated the nomogram based on these four independent preoperative factors:
- Number of preoperative diabetes medications (P less than .0001).
- Insulin use (P = .002).
- Duration of diabetes (P less than .0001).
- Glycemic control (P = .002).
“These factors are readily available in clinical practice and are considered a proxy of functional pancreatic beta-cell reserve,” Dr. Aminian said. The nomogram scores the severity of each factor on a scale of 0 to 100. For example, one preoperative medication scores 12, but five scores 63; duration of diabetes of 1 year scores five points, but 16 years scores 60. The patient’s total points represent the Individualized Metabolic Surgery score.
The researchers assigned three categories: A score of 25 or less represents mild disease, a score of 25-95 represents moderate, and 95 to the maximum 180 is severe disease.
Based on the Cleveland Clinic and Barcelona cohorts, the researchers next developed recommendations for average risk patients in each category.
RYGB is “suggested” in mild disease based on remission rates of 92% in the Cleveland subgroup and 91% in the Barcelona subgroup vs. SG remission rates of 74% and 91%, respectively. On the other end of the spectrum, in patients with severe diabetes, both procedures were less effective in achieving long-term diabetes remission. The disparities were more pronounced for the moderate group: 60% and 70% for RYGB and 25% and 56% for SG in the Cleveland Clinic and Barcelona subgroups, respectively. Hence the nomogram highly “recommends” RYGB. An online calculator to determine Individualized Metabolic Surgery score is available at http://riskcalc.org/Metabolic_Surgery_Score/.
“Obviously, this is the first attempt toward individualized procedure selection and more work needs to be done,” Dr. Aminian said. “Our findings also highlight the importance of surgical intervention in early stages of diabetes in order to achieve sustainable remission.”
In his discussion, Matthew M. Hutter, MD, FACS, of Harvard Medical School, Boston, offered to “quibble” with Dr. Aminian’s conclusions. “I challenge you on your conclusion for the mild and severe categories,” Dr. Hutter said. “My rate-of-cure data makes me want to recommend bypass for any patient with diabetes – mild, moderate or severe.”
Dr. Aminian acknowledged that the number of SG cases in the severe subgroup – 51 – was not great, and long-term diabetes remission was comparable between the two procedures. The key distinguishing measure in the severe category was a net 8% difference between two procedures in glycemic control at last follow-up. “This is the difference that we must decide whether it’s clinically important or not – whether we’re willing to recommend a riskier procedure for an extra 8% achieving glycemic control,” Dr. Aminian said. “If someone thinks it’s worth the risk, then they may suggest gastric bypass. If someone thinks it’s not worth the risk, then they may suggest a sleeve gastrectomy. But, we should remember that patients in the severe group are very high-risk patients.”
Dr. Aminian reported no financial disclosures. Dr. Hutter disclosed receiving conference reimbursement from Olympus.
The complete manuscript of this study and its presentation at the American Surgical Association’s 137th Annual Meeting, April 2017, in Philadelphia, Pennsylvania, is to be published in Annals of Surgery pending editorial review.