Reliability of methods in assessing proximal femur bone quality for hip prosthesis selection

Assessment of proximal femoral bone quality is essential for implant selection in total hip arthroplasty (THA). The Dorr classification, Canal-to-Calcar Ratio (CCR), and Cortical Thickness Index (CTI) are commonly used radiographic tools, but their reproducibility and associations with densitometric parameters remain unclear. This study aimed to evaluate the reproducibility of these three methods and their associations with DXA- and 3D-DXA-derived bone mineral density (BMD), to identify the most reliable tool for preoperative bone quality assessment in THA. A prospective cohort of 203 patients with hip osteoarthritis was assessed using standard radiographs, DXA-derived BMD, and 3D-DXA-derived cortical and trabecular bone parameters. Three independent observers assessed radiographs in two sessions separated by one year. Intra- and inter-observer reliability was tested with kappa, ICC, and Bland-Altman analysis. Correlations with BMD were assessed by ANOVA and Spearman correlation analysis. Subgroup analyses were performed according to sex and BMI. The Dorr classification showed only intra- and interobserver fair-to-moderate reproducibility, while CCR and CTI demonstrated intra- and interobserver moderate-to-excellent reliability. CTI was positively associated with total femoral BMD, whereas CCR showed a negative association; the association was stronger for CTI. Total femoral BMD and cortical sBMD differed across Dorr types. No other associations were found between morphometric indices and 3D-DXA-derived cortical or trabecular BMD. CTI showed the highest intra- and inter-observer reproducibility and the strongest association with DXA-derived BMD. Within this dataset, CTI appeared to be the most reproducible radiographic surrogate among the evaluated indices for preoperative bone quality assessment in THA.