HbA1c Change and Diabetic Retinopathy During GLP-1 Receptor Agonist Cardiovascular Outcome Trials: A Meta-analysis and Meta-regression

Abstract

Background: Long-term glycemic control reduces retinopathy risk, but transient worsening can occur with glucose control intensification. Glucagon-like peptide 1 receptor agonists (GLP-1RA) lower glucose, but the long-term impact on retinopathy is unknown. GLP-1RA cardiovascular outcome trials (CVOTs) provide long-term follow-up, allowing examination of retinopathy outcomes.

Purpose: To examine the associations between retinopathy, HbA_1c, systolic blood pressure (SBP), and weight in GLP-1RA CVOTs.

Data sources: Systematic review identified six placebo-controlled GLP-1RA CVOTs reporting prespecified retinopathy outcomes.

Study selection: Published trial reports were used as the primary data sources.

Data extraction: HbA_1c, SBP, and weight data throughout follow-up by treatment group were extracted.

Data synthesis: Random-effects model meta-analysis showed no association between GLP-1RA treatment and retinopathy (odds ratio [OR] 1.10; 95% CI 0.93, 1.30), with high heterogeneity between studies (I ² = 52.2%; Q statistic P = 0.063). Univariate meta-regression showed an association between retinopathy and average HbA_1c reduction during the overall follow-up (slope = 0.77, P = 0.007), but no relationship for SBP or weight. Sensitivity analyses for HbA_1c showed a relationship at 3 months (P = 0.006) and 1 year (P = 0.002). A 0.1% (1.09 mmol/mol) increase in HbA_1c reduction was associated with 6%, 14%, or 8% increased Ln(OR) for retinopathy at the 3-month, 1-year, and overall follow-up, respectively.

Limitations: CVOTs were not powered to assess retinopathy outcomes and differed in retinopathy-related criteria and methodology. The median follow-up of 3.4 years is short compared with the onset of retinopathy.

Conclusions: HbA_1c reduction was significantly associated with increased retinopathy risk in meta-regression for GLP-1RA CVOTs. The magnitude of HbA_1c reduction was correlated with retinopathy risk in people with diabetes and additional cardiovascular risk factors, but the long-term impact of improved glycemic control on retinopathy was unmeasured in these studies. Retinopathy status should be assessed when intensifying glucose-lowering therapy.

Figure 1

Meta-analysis of retinopathy outcomes in GLP-1RA CVOTs. Data were assessed using the random-effects model.

Figure 2

The association of HbA_1c (A), SBP (B), and body weight reduction (C) vs. retinopathy Ln(OR) at the overall follow-up period. Data are meta-regression estimations and the 95% CI (represented by the dotted lines). The average of the differences of HbA_1c (A), SBP (B), and body weight (C) between the two treatment groups (GLP-1RA or placebo) weighted by follow-up (years) are presented. The area of each circle is proportional to the study’s variance.