Diagnostic accuracy of Mirels score, CT-based Structural Rigidity Analysis, and Finite Element Analysis for predicting impending femoral pathologic fractures in metastatic bone disease: a systematic review.

Background: Patients with metastatic bone disease affecting the femur are at substantial risk of pathologic fracture, resulting in significant morbidity and frequently necessitating urgent surgical intervention. Accurate identification of lesions at imminent risk of fracture enables timely prophylactic fixation and improved patient outcomes. Although the Mirels score remains widely used, its limited specificity has prompted interest in CT-derived biomechanical approaches, including CT-based Structural Rigidity Analysis (CTRA) and Finite Element Analysis (FEA). Whole-femur CT is routinely obtained in some specialist orthopaedic oncology pathways, including our own, but this practice is not universal; many centres rely initially on clinical assessment and plain radiographs, with CT reserved for selected cases. We systematically reviewed the literature to compare evidence for Mirels, CTRA, and CT-based FEA in predicting pathologic femoral fractures. Methods: A systematic search of MEDLINE and Embase was performed for studies published between 2005 and 2025. Eligible studies evaluated Mirels scoring, CTRA, or CT-based FEA in adult patients with femoral metastatic disease and reported subsequent pathologic femoral fracture outcomes. Two reviewers independently screened studies and extracted data. Potential cohort overlap was assessed by comparing authorship, recruiting institutions, enrolment periods, eligibility criteria, and descriptions of prior datasets. Management after fracture-risk assessment was examined to identify intervention-related bias. Due to heterogeneity in study design, imaging methods, and outcome reporting, a qualitative narrative synthesis was undertaken. Results: Eight studies met the inclusion criteria, comprising prospective and retrospective clinical cohorts, an implementation study, and one illustrative comparative case series. In studies reporting formal diagnostic estimates, Mirels sensitivity ranged from 66.7% to 88%, while specificity ranged from 38% to 47.9%. In the principal prospective comparison, CTRA achieved 100% sensitivity and 60.6% specificity. Across formal FEA analyses, sensitivity ranged from 80% to 100% and specificity from 67% to 86%, although modelling methods, thresholds, and comparators varied. The evidence arose from small, partly overlapping cohorts, and prophylactic stabilization of lesions considered high risk limited observation of untreated outcomes and particularly the positive predictive value. Heterogeneity precluded meta-analysis. Conclusion: Mirels remains a useful, accessible, and sensitive first-line screening tool, particularly where plain radiography is the principal imaging modality. CTRA and CT-based FEA may offer additional specificity and objective mechanical information in selected patients where suitable calibrated CT imaging and technical expertise are available, but current evidence does not support replacing clinical assessment and radiography. Whole-femur CT is routine in some specialist pathways, including our own, but not across all centres. Further progress will require prospective multicentre validation, standardized methodologies, health-economic assessment, and integration within multidisciplinary team pathways.