Orthogeriatric admission biomarker profiles across hip fragility fracture phenotypes: a real-world cohort study

Hip fragility fractures are traditionally managed as a single clinical entity, despite long-standing anatomical, epidemiological and prognostic differences between intra- and extracapsular fractures. Whether routinely collected admission biomarker profiles differ between these fracture phenotypes remains insufficiently explored. The prospective cohort comprised 743 consecutive patients aged ≥ 60 years with hip fragility fractures admitted during 2023–2024, of whom 644 had complete data and constituted the analytical sample (266 ICF and 378 ECF). No missing values were imputed. Eighteen routinely assessed haematological and biochemical admission biomarkers, used as proxies for biological domains related to Chronic Risk Factors for Bone Damage (CRiF-BD), were analysed. Fracture phenotype was established as intracapsular and extracapsular by standard clinical and radiographic criteria before biomarker analysis. Group comparisons were followed by correlation analysis, hierarchical clustering and multivariable logistic regression to assess associations between admission biomarker profiles and the radiographically established fracture phenotype. The robustness of multivariable statistical separation was assessed using 10-fold cross-validation and the area under the receiver operating characteristic curve (AUC). Compared with extracapsular fractures, intracapsular fractures showed higher haematocrit, haemoglobin, albumin and free thyroxine levels, and lower potassium levels (all p < 0.05). Multivariable analysis identified a combined admission biomarker profile associated with fracture phenotype. The multivariable model showed moderate statistical separation between radiographically established ICF and ECF phenotypes (AUC 0.72 [95% CI 0.67–0.76]). Correlation and hierarchical clustering analyses revealed an exploratory multivariable structure broadly corresponding to haematological, transport/coagulation, renal, endocrine and metabolic biomarker domains. Routinely available admission biomarker profiles differ between radiographically established intra- and extracapsular hip fragility fracture phenotypes. Interpreted within the CRiF-BD framework, these exploratory associations are consistent with biological heterogeneity between fracture phenotypes and support further investigation into whether ICF and ECF reflect distinct or partially overlapping pathophysiological processes. Because the biomarkers were measured after fracture occurrence, the present study cannot determine to what extent the observed differences reflect pre-existing biological characteristics or acute fracture-related changes. Future studies incorporating pre-fracture or longitudinal biomarker data, comparisons with non-fractured older populations and direct bone assessment are needed to clarify their temporal origin and pathophysiological significance.