Imaging of knee osteoarthritis: a review of multimodal diagnostic approach
Claudia Lucia Piccolo,Carlo Augusto Mallio,Federica Vaccarino,Rosario Francesco Grasso,Bruno Beomonte Zobel
DOI: https://doi.org/10.21037/qims-22-1392
2023-11-01
Abstract:Knee osteoarthritis (KOA) is a common chronic condition among the elderly population that significantly affects the quality of life. Imaging is crucial in the diagnosis, evaluation, and management of KOA. This manuscript reviews the various imaging modalities available until now, with a little focus on the recent developments with Artificial Intelligence. Currently, radiography is the first-line imaging modality recommended for the diagnosis of KOA, owing to its wide availability, affordability, and ability to provide a clear view of bony components of the knee. Although radiography is useful in assessing joint space narrowing (JSN), osteophytes and subchondral sclerosis, it has limited effectiveness in detecting early cartilage damage, soft tissue abnormalities and synovial inflammation. Ultrasound is a safe and affordable imaging technique that can provide information on cartilage thickness, synovial fluid, JSN and osteophytes, though its ability to evaluate deep structures such as subchondral bone is limited. Magnetic resonance imaging (MRI) represents the optimal imaging modality to assess soft tissue structures. New MRI techniques are able to detect early cartilage damage measuring the T1ρ and T2 relaxation time of knee cartilage. Delayed gadolinium-enhanced MRI of cartilage, by injecting a contrast agent to enhance the visibility of the cartilage on MRI scans, can provide information about its integrity. Despite these techniques can provide valuable information about the biochemical composition of knee cartilage and can help detect early signs of osteoarthritis (OA), they may not be widely available. Computed tomography (CT) has restricted utility in evaluating OA; nonetheless, weight-bearing CT imaging, using the joint space mapping technique, exhibits potential in quantifying knee joint space width and detecting structural joint ailments. PET-MRI is a hybrid imaging technique able to combine morphological information on bone and soft tissue alterations with the biochemical changes, but more research is needed to justify its high cost and time involved. The new tools of artificial intelligence, including machine learning models, can assist in detecting patterns and correlations in KOA that may be useful in the diagnosis, grading, predicting the need for arthroplasty, and improving surgical accuracy.