Efficacy of intermittently scanned continuous glucose monitoring in patients with types 1 or 2 diabetes receiving insulin therapy: a systematic review and meta-analysis

Abstract

BackgroundMonitoring glucose levels is crucial for managing glycemic control. Methods include self-monitored blood glucose (SMBG), continuous glucose monitoring (CGM), and intermittently scanned continuous glucose monitoring (isCGM).ObjectiveTo assess the efficacy of isCGM versus SMBG in individuals with type 1 diabetes mellitus (T1DM) or type 2 diabetes mellitus (T2DM) on insulin therapy.MethodsWe conducted a systematic review including randomized controlled trials involving patients over 4 years old with T1DM or T2DM on multiple daily insulin regimens, comparing isCGM to SMBG. The outcomes analyzed were HbA1c (%), time below the target glucose range (TBR), patient satisfaction (DTSQ), device-related adverse events, time in range (TIR), and hypoglycemic events. Searches were performed in MEDLINE, EMBASE, and CENTRAL. Two independent reviewers screened studies, assessed the risk of bias, and extracted data. The meta-analyses employed a random-effects model, and the certainty of evidence was evaluated via the GRADE system.ResultsSeventeen studies with 1,860 participants were included. The isCGM demonstrated a moderate certainty of evidence for reducing HbA1c (Mean difference [MD]: -0.25%, 95% confidence interval [95% CI]: -0.39– -0.10%; I²: 82.6% 13 studies; 1,482 patients) and enhancing patient satisfaction (MD: 4.5, 95% CI: 2.18– 6.82; I²: 92.9%; 10 studies; 1,150 patients). Meta-regression revealed that intervention duration was a significant moderator of HbA1c reduction. isCGM also favored a reduction in TBR, with an MD of -0.15% (95% CI: -0.23– -0.07%; I²: 96.7% 8 studies; 1,094 patients; low certainty). Mild device-related adverse events were more common in the isCGM group (Relative risk: 2.69, 95% CI: 1.5– 4.81; I²: 0%; 7 studies; 991 participants; moderate certainty). The overall frequency of participants who discontinued isCGM due to cutaneous adverse events was 1% (95% CI: 0–6%; 7 studies; 533 participants). No clear effects were observed for TIR (MD: 0.02%, 95% CI: -0.05– 0.1%; I²: 79.6%; 11 studies; 1,318 patients; very low certainty) or hypoglycemic episodes.ConclusionsCompared with SMBG, isCGM reduces HbA1c, enhances patient satisfaction, and reduces TBR. However, it may increase the incidence of mild device-related adverse events. No definitive effects were observed on the TIR or hypoglycemia frequency.Prospero registrationCRD42024562805.