Home / Resources / Articles / Reversal of Prediabetes with Medication in High-Risk Populations

Reversal of Prediabetes with Medication in High-Risk Populations

Dec 1, 2018
Editor: Steve Freed, R.PH., CDE

Author: Arsalan Hashmi, PharmD. Candidate, LECOM College of Pharmacy

New study shows treating prediabetes risk for developing diabetes could reverse the disease before it starts.

According to the CDC, approximately 90% of people with prediabetes are unaware of the condition since their daily lives are relatively unaffected. This, combined with the fact that risk factors, including lack of exercise, unhealthy diet, and heavier weight, are not uncommon in American society, leads prediabetes to go undetected for years.


One-third of Americans living with prediabetes will develop diabetes. Prediabetes, which is diagnosed by an increased A1C and fasting blood sugar, has several different pathophysiologies that lead to type 2 diabetes. Previous studies have shown successful correction of pathophysiological factors in a controlled setting, with the addition of medications, lifestyle changes, or surgeries. This study applied a personalized medication regimen to the outpatient setting for patients at high risk for developing diabetes.

This study was a retrospective observational evaluation of clinical data on the management of healthy individuals at high risk for developing diabetes. There were 1,769 high-risk participants who were assessed by an internal medicine and endocrinology community practice. Management of these participants was as follows 1) identification of risk level for future diabetes, 2) oral glucose tolerance tests were performed by measuring plasma glucose concentrations, 3) severity of insulin resistance, beta-cell dysfunction and glycemia was determined, 4) stratification of risk based on how severe the pathophysiologic abnormalities were, and, finally, patients were given medications based on how severe the condition was. Patients were healthy at baseline. With the use of an oral glucose tolerance test, patients were classified into mild, moderate, or severe for glycemic response, insulin sensitivity, and response of beta cells.

Participants had an overnight fast followed by a 2-hour 75 g oral glucose tolerance test. No medications were taken before the test. Plasma glucose, insulin, and C-peptide were monitored in increments for up to 2 hours. Baseline C-reactive protein, plasma lipid, and HbA1C were recorded.

Glucagon-like peptide 1 (GLP-1) agonists are medications that stimulate insulin and inhibit glucagon secretion. They are useful in this case because the body is still able to  secrete and react to insulin. Patients with the highest risk of developing type 2 diabetes were given a combination of a GLP-1 agonist, pioglitazone 15mg/day, metformin 850mg/day, and lifestyle changes. Those at intermediate risk were not given the GLP-1 agonist. Matsuda index of insulin sensitivity was calculated and compared. Patients were considered severely insulin resistant if they ranked less than the lowest 5% on the index, and moderate 6th to 25th percentile. Insulin secretion was also calculated with severe being < 50% normal secretion, and moderate 50% to 70% of normal. Participants followed up every six months to test for type 2 diabetes, with oral glucose tests done every two years. Statistical analysis was done using Cox proportional hazard model, multi-variant logistic regression, ANOVA, and X2 tests.

Of the 1,796 people screened, only 422 completed all follow-ups for the study. There were 81 patients who were at high risk for T2D and received pioglitazone, metformin, GLP-1 agonists, and lifestyle advice. An additional 76 high-risk patients denied treatment and only received lifestyle advice. There were 141 patients who were classified as intermediate and given the same medications minus the GLP-1 agonist. In this intermediate group, an additional 124 patients denied therapy. Weight loss was only significant in patients receiving the GLP-1 agonist who lost on average 1.8 kg (p= 0.038). On follow-up, 21 participants who only took lifestyle recommendations had incident T2D, whereas only 7 participants developed T2D in the group receiving metformin and pioglitazone (HR 0.29 CI 0.11-0.78 p=0.0009). The group receiving a GLP-1 agonist had no T2D reported (HR 0.12 CI 0.02-0.94). Normal glucose tolerance was re-established in 39% of the participants receiving lifestyle therapy, 52% of participants receiving metformin and pioglitazone, and 77% of participants receiving the metformin/pioglitazone/GLP-1 agonist combination. There was a significant drop in fasting plasma glucose across all three groups. Insulin sensitivity increased 17% in those taking metformin/pioglitazone and increased 47% with the addition of GLP-1 agonists. Beta cell function also increased significantly in patients on metformin/pioglitazone +/- GLP-1:  77% and 162% respectively. In the numbers, we can see a connection between beta cells and T2D.

This study shows that even in individuals with seemingly normal glycemic control, beta cell dysfunction plays a big factor in the development of diabetes, and a simple method of testing this must be developed. The addition of metformin and pioglitazone can delay the onset of the disease substantially in patients with declining beta cells. Currently the only prevention guidelines on the American Diabetes Association’s’ website are the lifestyle changes that this study has shown have minimal effect. Before the FDA can approve these medications for at-risk patients, additional long-term studies must be done.

Practice Pearls:

  • Functional beta cells are the basis of prevention of diabetes. The use of metformin 850mg/day and pioglitazone 15mg/day increased their function by 77%.
  • 4.9% of people receiving metformin/pioglitazone developed diabetes, and none receiving the GLP-1 agonist developed diabetes.
  • The long-term effects of using these medications in prediabetes should be studied in longer trials.


Armato, John P, et al. “Successful Treatment of Prediabetes in Clinical Practice Using Physiological Assessment (STOP DIABETES).” The Lancet Diabetes & Endocrinology, vol. 6, no. 10, 2018, pp. 781–789., doi:10.1016/s2213-8587(18)30234-1.

“Diagnosing Diabetes and Learning About Prediabetes.” American Diabetes Association, 22 Sept. 2014, www.diabetes.org/diabetes-basics/diagnosis/.

CDC. “CDC Features.” Centers for Disease Control and Prevention, Centers for Disease Control and Prevention, 12 Jan. 2018, www.cdc.gov/features/diabetesprevention/index.html.

Arsalan Hashmi, PharmD. Candidate, LECOM College of Pharmacy