GLP-1 receptor agonists and pioglitazone showed comparable benefits for liver and cardiovascular outcomes in type 2 diabetes, but GLP-1 drugs reduced heart failure risk by 35%.
35% lower heart failure riskGLP-1 receptor agonist users vs pioglitazone users in type 2 diabetes (HR 0.65, p < 0.001)
What the researchers found
In 8,922 propensity-matched patients (4,461 per group), GLP-1 receptor agonists showed comparable risks versus pioglitazone for major adverse liver outcomes (HR 0.94, 95% CI 0.66–1.34) and MACE (HR 0.99, 95% CI 0.80–1.22). However, GLP-1 receptor agonists significantly reduced heart failure risk (HR 0.65, 95% CI 0.51–0.83) — a 35% reduction. Results were consistent across intention-to-treat and per-protocol analyses and robust across subgroup analyses.
Why it matters
Both GLP-1 receptor agonists and pioglitazone are recommended for patients with type 2 diabetes and cardiovascular or liver disease risk. Without head-to-head trials, clinicians have lacked direct comparison data. This study fills that gap and provides actionable guidance — especially the finding that GLP-1 drugs have a meaningful edge in preventing heart failure.
How the study worked
This was a target trial emulation using an active comparator, new user design. Type 2 diabetes patients newly prescribed GLP-1 receptor agonists or pioglitazone between 2008 and 2022 were identified from a territory-wide Hong Kong health database. Propensity score matching created balanced groups, and Cox proportional hazards models estimated outcomes via both intention-to-treat and per-protocol analyses.
What this study cannot tell us
This is an observational study using electronic health records, not a randomized trial, so unmeasured confounders may exist. The study was conducted in a Hong Kong Chinese population, which may limit generalizability. Specific GLP-1 drugs were not analyzed separately, and the per-protocol heart failure result did not reach significance, suggesting the ITT finding should be interpreted cautiously.
How to read the evidence
This is a large, well-designed observational study using target trial emulation methodology — considered the gold standard for causal inference from observational data. However, it remains non-randomized, and the per-protocol analysis showed a narrower heart failure benefit.
When this study was published
Published in 2025 using data from 2008–2022, this study reflects the full era of GLP-1 receptor agonist use and provides timely comparison data as these drugs are increasingly prescribed.
The bigger picture
As diabetes treatment shifts toward drugs with proven cardiovascular and metabolic benefits beyond glucose control, head-to-head comparisons become essential. This study positions GLP-1 receptor agonists as at least equivalent to pioglitazone for cardio-hepatic protection, with a clear advantage for heart failure — the leading cause of hospitalization in diabetes patients.
Questions still open
- Would a randomized controlled trial confirm GLP-1 drugs' superiority over pioglitazone for heart failure prevention?
- Do specific GLP-1 receptor agonists (e.g., semaglutide vs liraglutide) differ in their heart failure benefits compared to pioglitazone?
- Should GLP-1 drugs be preferred over pioglitazone as first-line add-on therapy in T2D patients with both liver and heart failure risk?
Common questions
Why compare GLP-1 drugs to pioglitazone specifically?
Should everyone with type 2 diabetes switch from pioglitazone to a GLP-1 drug?
Read the original research
Benefits of glucagon-like peptide-1 receptor agonists versus pioglitazone for cardio-hepatic outcomes: a territory-wide target trial emulation.
Cardiovascular diabetology, 24(1), 416
Citation
Li, Lanlan; Lui, David Tak-Wai; Fong, Carol Ho-Yi; Chow, Wing-Sun; Au, Ivan Chi-Ho; Xiong, Xi; Lang, Brian Hung-Hin; Wong, Carlos King-Ho; Lee, Chi-Ho. (2025). Benefits of glucagon-like peptide-1 receptor agonists versus pioglitazone for cardio-hepatic outcomes: a territory-wide target trial emulation.. Cardiovascular diabetology, 24(1), 416. https://doi.org/10.1186/s12933-025-02973-5