Reducing or Discontinuing Insulin or Sulfonylurea When Initiating a Glucagon-like Peptide-1 Agonist
Background: Hypoglycemia and weight gain are well-known adverse effects of insulin and sulfonylureas in patients with type 2 diabetes. Glucagon-like peptide-1 (GLP-1) agonist monotherapy has a low incidence of hypoglycemia, but when combined with insulin or sulfonylureas, patients have higher rates of hypoglycemia.
Methods: The primary study outcome was to determine the percentage of patients with dose reductions or discontinuation of insulin or a sulfonylurea at baseline, 3, 6, and 12 months. Secondary outcomes included changes in hemoglobin A1c and body weight measured. This was a single-center, quality assurance, quality improvement, retrospective chart review of patients prescribed a GLP-1 agonist while on insulin and/or a sulfonylurea between January 1, 2019, and September 30, 2022, at a pharmacist-led patient aligned care team (PACT) clinic at the Wilkes-Barre Veterans Affairs Medical Center.
Results: This retrospective chart review included 136 patients. Throughout 12 months of following the pharmacist-led PACT clinic, 55 patients (57.3%) had ≥ 1 insulin dose reduction, 16 patients (29.6%) had ≥ 1 dose reduction of a sulfonylurea, 14 patients (14.6%) discontinued insulin, and 21 patients (38.9%) discontinued their sulfonylurea. Conclusions: Patients taking insulin and/or sulfonylurea in combination with a GLP-1 agonist may require a dose reduction or discontinuation of the diabetic agents. Patients on both insulin and sulfonylurea may need closer monitoring due to the higher incidences of discontinuations compared with patients on only 1 of these agents. Dose reductions or discontinuations of these diabetic agents can promote positive patient outcomes, such as preventing hypoglycemia, minimizing weight gain, increasing weight loss, and reducing hemoglobin A 1c.
Results
This retrospective review included 136 patients; 96 patients taking insulin and 54 taking a sulfonylurea when they started a GLP-1 agonist. Fourteen patients were on both. Criteria for use, which are clinical criteria to determine if a patient is eligible for the use of a given medication, are used within the VA. The inclusion criteria for a patient initiating a GLP-1 agonist is that the patient must have atherosclerotic cardiovascular disease or chronic kidney disease with the patient receiving metformin (unless unable to use metformin) and empagliflozin (unless unable to use empagliflozin).
The baseline mean age and weight for the patient population in this retrospective chart review was 70.7 years and 238.2 lb, respectively. Ninety-six patients (70.6%) were started on semaglutide, 27 (19.9%) on dulaglutide, 12 (8.8%) on liraglutide, and 1 (0.7%) on exenatide. The mean HbA1c when patients initiated a GLP-1 agonist was 8.6%. When starting a GLP-1 agonist, 34 patients (25.0%) had an HbA1c < 8%, 89 (65.4%) had an HbA1c between 8% to 10%, and 13 (9.6%) had an HbA1c > 10% (Table).
For the primary results, 25 patients (26.0%) had a dose reduction of insulin when they started a GLP-1 agonist, and 55 patients (57.3%) had at least 1 insulin dose reduction within the year follow-up. Seven patients (13.0%) had a dose reduction of a sulfonylurea when they started a GLP-1 agonist, and 16 patients (29.6%) had at least 1 dose reduction of a sulfonylurea within the year follow-up. Six patients (6.3%) discontinued insulin use when they initially started a GLP-1 agonist, and 14 patients (14.6%) discontinued insulin use within the year follow-up. Eleven patients (20.4%) discontinued sulfonylurea use when they initially started a GLP-1 agonist, and 21 patients (38.9%) discontinued sulfonylurea use within the year follow-up (Figure).
Fourteen patients were on both insulin and a sulfonylurea. Two patients (14.3%) had a dose reduction of insulin when they started a GLP-1 agonist, and 5 (35.7%) had ≥ 1 insulin dose reduction within the year follow-up. Three patients (21.4%) had a dose reduction of a sulfonylurea when they started a GLP-1 agonist, and 6 (42.9%) had ≥ 1 dose reduction of a sulfonylurea within the year follow-up. Seven patients (50.0%) discontinued sulfonylurea and 3 (21.4%) discontinued insulin at any time throughout the year. The majority of the discontinuations were at the initial start of GLP-1 agonist therapy.
The mean HbA1c for patients on GLP-1 agonist was 8.6% at baseline, 8.0% at 0 to 3 months, 7.6% at 3 to 6 months, and 7.5% at 12 months. Patients experienced a mean HbA1c reduction of 1.1%. The mean weight when a GLP-1 agonist was started was 238.2 lb, 236.0 lb at 0 to 3 months, 223.8 lb at 3 to 6 months, and 224.3 lb after 12 months. Study participants lost a mean weight of 13.9 lb while on a GLP-1 agonist.
Discussion
While this study did not examine why there were dose reductions or discontinuations, we can hypothesize that insulin or sulfonylureas were reduced or discontinued due to a myriad of reasons, such as prophylactic dosing per guidelines, patients having a hypoglycemic event, or pharmacists anticipating potential low blood glucose trends. Also, there could have been numerous reasons GLP-1 agonists were started in patients on insulin or a sulfonylurea, such as HbA1c not being within goal range, cardiovascular benefits (reduce risk of stroke, heart attack, and death), weight loss, and renal protection, such as preventing albuminuria.13,14
This retrospective chart review found a large proportion of patients had a dose reduction of insulin (57.3%) or sulfonylurea (29.6%). The percentage of patients with a dose reduction was potentially underestimated as patients were not counted if they discontinued insulin or sulfonylurea. Concomitant use of GLP-1 agonists with insulin or a sulfonylurea may increase the risk of hypoglycemia and reducing the dose of insulin or a sulfonylurea may be necessary.9-11 The dose reductions in this study show that pharmacists within pharmacy-led PACT clinics monitor for or attempt to prevent hypoglycemia, which aligns with the prescribing information of GLP-1 agonists. While increasing the insulin or sulfonylurea dose is an option for patients, this approach can increase the risk of hypoglycemia, especially in an older population, like this one with a mean age > 70 years. The large proportions of patients with dose reductions or insulin and sulfonylurea discontinuations suggest that pharmacists may need to take a more cautious approach when initiating a GLP-1 agonist to prevent adverse health outcomes related to low blood sugar for older adults, such as falls and fractures.
Insulin was discontinued in 20.4% of patients and sulfonylurea was discontinued in 38.9% of patients within 12 months after starting a GLP-1 agonist. When a patient was on both insulin and a sulfonylurea, the percentage of patients who discontinued insulin (21.4%) or a sulfonylurea (50.0%) was higher compared with patients just on insulin (14.6%) or a sulfonylurea (38.9%) alone. Patients on both insulin and a sulfonylurea may need closer monitoring due to a higher incidence of discontinuations when these diabetes agents are administered in combination.
