Cohort studyindependent cross-checkabstract checked★ key study I-LEG-109 standard review
Antibiotics and haemolysis in G6PD deficiency
Quinolones and risk of haemolysis in patients with glucose-6-phosphate dehydrogenase deficiency: a nationwide retrospective active-comparator cohort study.
Overview
Lavon O et al.. J Antimicrob Chemother 2026 · PMID 42522747 · DOI 10.1093/jac/dkag265
In 465,083 antibiotic courses, coded haemolysis occurred after fewer than 1 in 1,000 courses, with no clear excess for quinolones.
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?
Test whether quinolones cause haemolysis in G6PD deficiency.
- How was the study done?
National database cohort vs penicillins.
- How many people, animals or samples were studied?
64,861 patients.
- What was actually found?
No clear increase.
- How large was the observed effect?
aOR 1.29 (0.92-1.77).
- What do the results mean in everyday life?
Ordinary antibiotics rarely cause haemolysis in G6PD deficiency.
- What are the study's limitations?
Coded data; abstract only.
- What does it say about Elawady's specific claim?
Contradicts the class-wide antibiotic claim.
- Do the results apply to Germany and Europe?
Applicable.
- What can explicitly not be concluded from this study?
It cannot exclude rare reactions in severe variants.
Scientific in-depth review
1. Full reference
- Authors
- Lavon O et al.
- Journal
- J Antimicrob Chemother
- Year
- 2026
- Identifier
- PMID 42522747 DOI 10.1093/jac/dkag265 · PMC13416391
- Verification
- verified – PubMed efetch 2026-10-26 + Crossref works API
- Retracted
- no
- Corrections
- none
- Extent checked
- abstract only
- Conflicts of interest and funding
Not stated in abstract.
2. Research question
Are oral quinolone dispensations associated with haemolysis compared with penicillins in G6PD-deficient patients?
3. Study design and population
- Study design
- Cohort study – Nationwide retrospective active-comparator cohort (Clalit Health Services database)
- Population
Patients with an ICD-coded diagnosis of G6PD deficiency receiving outpatient oral quinolone or penicillin
- Country
Israel
- Period
2010-2023
- Sample size
465,083 dispensations among 64,861 patients
- Inclusion and exclusion criteria
>=1 ICD code for G6PD deficiency; outpatient oral dispensation.
4. Methods and endpoints
- Exposure / intervention
Oral quinolone
- Comparator
Oral penicillin
- Endpoints
Haemolysis-related diagnosis code within 30 days; jaundice; laboratory markers
- Statistical methods
Logistic models adjusted for age, sex, Charlson index; within-patient subcohort
5. Actual results
- Results
Coded haemolysis after 52/65,906 quinolone (0.079%) vs 198/399,177 penicillin episodes (0.050%); aOR 1.29 (95% CI 0.92-1.77). Age <=18 y: 5/8,865 vs 91/160,423, aOR 1.02 (0.36-2.26). Within-patient: 0.065% vs 0.061%, aOR 0.95 (0.65-1.37).
- Absolute risks
Below 0.1% of courses in both groups (recalculated crude risk difference 0.029 percentage points).
- Authors' conclusion
No detectable major increase in coded haemolysis with quinolones; small risks or risks in severe deficiency/young children not excluded.
6. Strengths
- Very large sample
- Active comparator
- Within-patient analysis
7. Weaknesses
- Coded outcomes
- Variant severity unknown
- Abstract only
8. Risk of bias
- Orienting tool
- ROBINS-I
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Outcome misclassification; confounding by indication partly addressed.
9. Assessment of the statistical interpretation
CI approximately log-symmetric; crude NNH would exceed 3,000 courses if causal (stats/C-LEG-05_lavon2026_quinolones.py).
10. Relevance to the specific claim
Shows that common antibiotics (penicillins, quinolones) are rarely followed by haemolysis in G6PD deficiency; contradicts the class-wide antibiotic claim of C-LEG-05.
11. Independent contrary findings
Some hospital leaflets list quinolones as drugs to avoid (I-LEG-110).
12. Transferability to Germany and Europe
Applicable to Mediterranean G6PD variants.
13. Overall methodological judgement
Reassuring real-world data on antibiotics; supports not withholding needed antibiotics.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | not replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | moderate |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Risk in severe variants and infants.
15. Linked fact checks
- In G6PD deficiency, haemolysis is triggered by merely smelling fava beans and by eating lentils, as well as by aspirin and antibiotics. – Studies contradicting or limiting the claim