The role of radiographic knee alignment in total knee replacement outcomes and opportunities for artificial intelligence-driven assessment

Prevalent knee osteoarthritis (OA) imposes substantial burden on health systems with no cure available. Its ultimate treatment is total knee replacement (TKR). Complications from surgery and recovery are difficult to predict, and numerous factors may affect them. Radiographic knee alignment is a key factor affecting TKR success, affecting outcomes such as postoperative pain or function. Recently, artificial intelligence (AI) has been introduced to automatically analyse knee radiographs, for example, to assess knee alignment. These tools reduce measurement variability, support high-throughput analysis of large imaging datasets, and help identify alignment features associated with postoperative outcomes. A focused review is therefore needed to clarify how knee alignment biomarkers are measured, how they relate to TKR outcomes, and how AI-based approaches may support their assessment and clinical application in surgical planning and patient risk stratification. In this review, we first examine the current scoring protocols for evaluating TKR outcomes and potential knee alignment biomarkers associated with these outcomes. We then discuss existing AI-based approaches for generating knee alignment biomarkers from knee radiographs, and explore future directions for AI-driven knee alignment assessment and TKR outcome prediction. Knee alignment biomarkers represent an important but underexplored domain for improving prediction of TKR outcomes. AI-driven radiographic assessment offers a promising means to generate consistent, scalable, and clinically actionable measures of alignment, though challenges remain in accuracy, validation, multi-view integration, and clinical deployment. Improved models, rigorous external validation, and systems that link alignment measurements to outcome prediction are needed to enhance reliability and clinical applicability.