Haemochromatosis resulting from HFE C282Y homozygosity causes increased iron absorption and is classically associated with liver disease and arthropathy. Prior evidence has suggested higher rates of limb fractures in people with haemochromatosis, but the relationship with vertebral fractures has been poorly characterised. This study aimed to investigate whether vertebral compression fractures identified radiographically on lateral dual-energy X-ray absorptiometry (iDXA) are associated with HFE C282Y homozygosity in a large community cohort drawn from the UK Biobank.
Participants were selected from UK Biobank imaging reassessment (2014–2020). The analysis included 227 participants of European genetic ancestry who were HFE C282Y homozygotes and 234 matched controls who did not carry common HFE haemochromatosis variants. Matching criteria were age, sex, and body mass index (BMI). Lateral vertebral assessment scans were obtained using iDXA (GE–Lunar) as part of the imaging protocol.
All lateral vertebral assessment scans were reviewed blind to participants’ HFE genotype for radiological evidence of vertebral fracture. The outcome of interest was radiographically identified vertebral compression fracture on iDXA lateral imaging. The reviewers recorded vertebral fractures across the assessed population; no additional imaging modalities or clinical adjudication beyond the blinded radiological review are reported in the source.
Matched logistic regression models were used to assess the association between HFE C282Y homozygosity and radiographically identified vertebral fracture. Analyses were stratified by sex. A sensitivity analysis excluded participants with a recorded diagnosis of haemochromatosis to evaluate whether the association persisted when clinically diagnosed cases were removed.
Across the 461 participants included in the imaging review, 78 vertebral fractures were identified, representing 16.9% of the cohort.
Among males, 22 of 89 HFE C282Y homozygotes (24.7%) had vertebral fractures identified on iDXA compared with 9 of 90 matched male controls without HFE alleles (10.0%). In matched logistic regression, male C282Y homozygotes had increased odds of vertebral fracture: Odds Ratio (OR) 2.95, 95% Confidence Interval (CI) 1.28–6.85, p = 0.01.
Among females, vertebral fracture prevalence was identical between groups: 23 of 138 female C282Y homozygotes (16.7%) and 24 of 144 matched female controls without HFE alleles (16.7%). The corresponding OR was 0.99 (95% CI 0.53–1.87, p = 1.00), indicating no observed excess risk in women in this dataset.
When participants with a recorded diagnosis of haemochromatosis were excluded, the association in males remained. The OR for vertebral fracture among male C282Y homozygotes (excluding clinically diagnosed haemochromatosis) was 3.37 (95% CI 1.41–8.10, p = 0.007). This indicates that the increased odds observed in males were not solely driven by those with a recorded clinical diagnosis.
In this community-based imaging study using iDXA lateral vertebral assessment, male HFE C282Y homozygotes had a substantially higher likelihood of radiographically identified vertebral fracture compared with matched controls who lack common HFE haemochromatosis variants. No difference in vertebral fracture prevalence was observed among females in this sample.
The authors suggest these findings support consideration of targeted vertebral fracture assessment in C282Y homozygous men to enable prompt identification and management that could reduce risk of future fractures if intervention is appropriate. As the report is a preprint, the results have not been peer reviewed and should not yet guide clinical practice without further confirmation.
The study used UK Biobank data under application number 14631. The authors state relevant ethical approvals and participant consent were obtained. Data used in the analysis are available to bona fide researchers from UK Biobank subject to approval and access restrictions; they are not publicly available. Declared funders include the University of Exeter and a National Institute for Health and Care Research (NIHR) Advanced Fellowship (NIHR301844). The preprint declares no competing interests.
The manuscript is presented as a preprint and has not undergone peer review. The source reports the core design, imaging method, matched comparison, and main effect estimates; additional methodological details, subgroup analyses beyond sex stratification, and other potential confounder adjustments are not reported in the source abstract. Readers should refer to the full preprint for further methodological detail and await peer review and replication before altering clinical practice.