Glycemic Patterns Assessed by Continuous Glucose Monitoring in People with Type 2 Diabetes Treated with Insulin Glargine 100 U/mL in Primary Care: The GPDetect Study

Abstract

INTRODUCTION: Hypoglycemia is a common complication in people with type 2 diabetes mellitus (T2DM) treated with basal insulin, yet its true prevalence in real-world primary care remains underexplored. This study used blinded continuous glucose monitoring (CGM) to assess glycemic patterns in patients with T2DM on basal insulin-supported oral therapy (BOT) with insulin glargine 100 U/mL (Gla-100) in Spain. METHODS: GPDetect was a prospective, multicenter, observational study in adults with T2DM treated with Gla-100 for ? 6 months in primary care. Participants underwent 14 days of blinded CGM (FreeStyle Libre(®) Pro iQ(®)). The primary objective of the present study was to estimate the proportion of patients with T2DM treated with Gla-100 whose TBR (glucose < 70 mg/dL) exceeded the internationally recommended threshold of 4%. Secondary endpoints included time in range (TIR), time above range (TAR), episodes of clinically significant hypoglycemia (< 54 mg/dL), cognitive function (MMSE), and fear of hypoglycemia (EsHFS). RESULTS: Among 136 patients (mean age 69 years), 16.9% had TBR > 4%, with a higher prevalence at night (23.5%) than during the day (12.5%). Clinically significant hypoglycemia (< 54 mg/dL > 1%) occurred in 8.8% of patients. Hyperglycemia was common: 46.3% spent > 25% of the day between 181-250 mg/dL, and 28.7% spent time > 250 mg/dL. Only 55.9% achieved TIR ? 70%. Patients receiving SGLT2i had lower mean glucose and EsHFS scores. Cognitive impairment (MMSE < 23) was associated with longer hypoglycemia duration. CONCLUSIONS: Blinded CGM revealed a relevant burden of unrecognized hypoglycemia-particularly nocturnal-as well as daytime hyperglycemia in patients with T2DM treated with Gla-100 in primary care. These results support the clinical value of CGM in informing treatment decisions and improving individualized diabetes management.