Long-Term GLP-1RA Use Does Not Lower Delirium Risk in T2D

Delirium risk initially decreases but eventually increases among patients with type 2 diabetes using glucagon-like peptide-1 receptor agonists (GLP-1RAs). These findings, from a retrospective cohort study, were published in Diabetes, Obesity and Metabolism.

Delirium often precedes the onset of dementia, for which patients with type 2 diabetes are at increased risk.

To assess the association between GLP-1RA use and delirium, investigators from the People’s Hospital of Zhengzhou University in China sourced data from the TriNetX database which houses data from 142 health care organizations. Adults (N=817,263) with type 2 diabetes who initiated GLP-1RAs (n=63,468) or metformin (n=753,795) between 2005 and 2025 were evaluated for a diagnosis of delirium from 181 days after initiating treatment.

To balance for cohort differences, a 1:1 propensity score matching approach was used, with final sample sizes of 63,096 per treatment group.

The pre-matched GLP-1RA and metformin cohorts comprised 56.76% and 47.45% women and 58.34% and 54.11% White individuals, respectively, and had mean (SD) ages of 58.75 (13.12) and 60.45 (13.29) years, BMI of 36.50 (8.13) and 32.52 (7.85) kg/m², and glycated hemoglobin (HbA_1c) of 7.60% (2.06%) and 7.72% (2.12%), respectively.

Among the matched cohort, 4.24% of GLP-1RA recipients and 7.10% of metformin recipients had a diagnosis of delirium. After adjusting for covariates, the researchers observed no between-group difference in the likelihood of delirium (adjusted hazard ratio [aHR], 1.01; 95% CI, 0.97-1.07) but cumulative risk for delirium was lower with GLP-1RA use (risk ratio [RR], 0.60; 95% CI, 0.57-0.63).

In the time-dependent analysis, GLP-1RAs were associated with lower risk for delirium within the first 5 years of use (aHR, 0.89; 95% CI, 0.86-0.92; P <.0001), higher risk for delirium at 5 to 10 years of use (aHR, 1.15; 95% CI, 1.04-1.26; P =.0046), and no significant risk difference at more than 10 years of use (aHR, 0.81; 95% CI, 0.58-1.12; P =.1978) relative to metformin.

In the subgroup analysis, GLP-1RA use, compared with metformin use, was associated with lower risk for delirium among:

• Individuals aged 40 to 79 years;
• Black or White individuals;
• Those with HbA_1c of 7.49% at most; and,
• Those with concomitant use of insulins, sulfonylureas, a-glucosidase inhibitors, statins, antiepileptics, antipsychotics, and antidepressants.

However, GLP-1RA vs metformin use was associated with higher risk for delirium among individuals who were Asian, Native Hawaiian, or Pacific Islander.

The all-cause mortality rates were 6.28% and 9.95% among the GLP-1RA and metformin cohorts, respectively, indicating an increased likelihood (aHR, 1.16; 95% CI, 1.12-1.21) but lower cumulative risk (RR, 0.63; 95% CI, 0.61-0.66) for all-cause mortality with GLP-1RA use.

This study was limited by absence of data about frailty or cognitive scores.

The study authors concluded, “Contrary to the assumption that newer therapies are always superior, our findings challenge the perceived advantages of GLP-1RAs. While GLP-1 RA use was associated with a lower risk of delirium in the early treatment period, this advantage disappeared with long-term use and metformin remained associated with superior survival.”

This article originally appeared on Endocrinology Advisor