Association between femoral cortical thickness index and hip fracture in older persons

Abstract

Purpose This study aimed to reevaluate a simple radiographic measurement, the cortical thickness index (CTI), using standard plain radiographs to investigate its association with the presence of hip fracture in older persons. Methods The study group (SG) consisted of 46 patients (35 females, 11 males) aged 65 years and older who presented to the emergency department with hip fractures. Two control groups were formed: One comprising 46 older persons aged 65 years and older (CG-G) and another of young adults aged 20-30 (CG-YA), both referred to the outpatient clinic and had available plain hip radiographs. CTI was measured for all participants by the first author. Additionally, two observers independently measured CTI to assess intraobserver and interobserver agreement. Results Intraobserver and interobserver agreement was almost perfect based on Cohen's kappa classification (kappa > 0.8). The mean CTI for the SG was 0.389 +/- 0.05, while the mean CTI for the CG-G was 0.571 +/- 0.048, and for the CG-YA was 0.585 +/- 0.048. A statistically significant difference was found between the SG and both CG-G and CG-YA groups (p < 0.001). No significant difference was observed between CG-G and CG-YA (p > 0.05). A CTI threshold of 0.485 demonstrated high discriminative ability for distinguishing older persons with hip fracture from those without fracture, yielding a sensitivity of 97.83% and a specificity of 95.65%. Conclusion CTI is significantly reduced in older persons with hip fracture. As a quick and accessible measurement on standard radiographs, CTI may assist clinicians in distinguishing individuals with hip fracture from those without fracture in outpatient settings.