Meta-analysisindependent cross-checkfull text checked★ key study I-ODD-121 in-depth review
Trials do not prove that sulfonylureas cause more deaths
The Association between Sulfonylurea Use and All-Cause and Cardiovascular Mortality: A Meta-Analysis with Trial Sequential Analysis of Randomized Clinical Trials.
Overview
Varvaki Rados D et al.. PLoS Med 2016 · PMID 27071029 · DOI 10.1371/journal.pmed.1001992
All-cause mortality OR 1.12 (0.96-1.30); MI OR 0.92.
Does the study support the specific claim? No · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Assess sulfonylurea safety in RCTs.
- How was the study done?
Meta-analysis of 47 RCTs with TSA.
- How many people, animals or samples were studied?
37,650 patients.
- What was actually found?
No significant increase in death, MI or stroke.
- How large was the observed effect?
OR 1.12 (0.96-1.30).
- What do the results mean in everyday life?
Sulfonylureas have not been shown to cause heart deaths in trials, but low blood sugar remains their main risk.
- What are the study's limitations?
Trials not designed for safety.
- What does it say about Elawady's specific claim?
Contradicts 'cause' as an established fact.
- Do the results apply to Germany and Europe?
High.
- What can explicitly not be concluded from this study?
A small harm.
Scientific in-depth review
1. Full reference
- Authors
- Varvaki Rados D et al.
- Journal
- PLoS Med
- Year
- 2016
- Identifier
- PMID 27071029 DOI 10.1371/journal.pmed.1001992
- Verification
- verified – PubMed efetch 2026-10-26; Crossref works API 2026-10-26
- Retracted
- no
- Corrections
Correction (PLoS Med 2016;13:e1002091): article republished to correct indexing of author names; no change to data (correction text read).
- Extent checked
- full text
- Conflicts of interest and funding
CNPq (Brazil); funder had no role.
2. Research question
Are second/third-generation sulfonylureas associated with all-cause or cardiovascular mortality in RCTs?
3. Study design and population
- Study design
- Meta-analysis – Meta-analysis of RCTs (>=52 weeks) with trial sequential analysis
- Population
Adults with type 2 diabetes in RCTs
- Country
International
- Period
To 31 December 2014
- Sample size
47 RCTs, 37,650 patients, 890 deaths
- Inclusion and exclusion criteria
RCTs >=52 weeks of second/third-generation sulfonylureas reporting outcomes
4. Methods and endpoints
- Exposure / intervention
Sulfonylureas
- Comparator
Placebo/diet or active comparators (including metformin)
- Endpoints
All-cause and CV mortality; MI; stroke
- Statistical methods
Peto OR; trial sequential analysis
5. Actual results
- Results
All-cause mortality OR 1.12 (0.96-1.30); CV mortality OR 1.12 (0.87-1.42); MI OR 0.92 (0.76-1.12); stroke OR 1.16 (0.81-1.66). Versus active comparators all-cause OR 1.16 (0.98-1.38). As add-on to metformin all-cause OR 1.26 (0.94-1.68). Glipizide OR 1.68 (1.06-2.66) (exploratory). TSA excluded an absolute difference of 0.5%.
- Absolute risks
TSA excluded an absolute mortality difference of 0.5%.
- Authors' conclusion
Sulfonylureas are not associated with increased all-cause or CV mortality, MI or stroke.
6. Strengths
- RCT-only
- Trial sequential analysis
- Full text reviewed
7. Weaknesses
- Most trials not designed for safety outcomes
- Point estimates above 1 versus active comparators
8. Risk of bias
- Orienting tool
- AMSTAR 2
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Unclear randomisation/concealment in many trials.
9. Assessment of the statistical interpretation
CIs log-symmetric (stats/C-ODD-11.py). Point estimates of 1.12-1.26 with upper limits up to 1.68 do not exclude a modest relative increase; 'no association' means no proven harm.
10. Relevance to the specific claim
Randomised evidence does not confirm that sulfonylureas cause more deaths or MIs than comparators (including metformin).
11. Independent contrary findings
Observational cohorts (A-ODD-07); Shi 2023 NMA low-certainty mortality increase (I-ODD-107).
12. Transferability to Germany and Europe
High.
13. Overall methodological judgement
No proven CV harm; modest harm not excluded.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | partly replicated |
| Consistency with the overall evidence | mixed |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Differences between individual sulfonylureas.
15. Linked fact checks
- Sulfonylureas cause more myocardial infarctions, strokes and deaths than metformin (21% higher in 253,690 veterans). – Studies contradicting or limiting the claim