Coronal T2 weighted fat-suppressed image demonstrating anterolateral bone bruising following rupture of the anterior cruciate ligament
Coronal T2 weighted fat-suppressed image demonstrating bony oedema secondary to trauma resulting in posterolateral corner injury
In current practice, proton density imaging is the mainstay of morphological assessment of articular cartilage. Accurate assessment of articular cartilage is achieved through adequate spatial resolution, contrast resolution and orthogonal image acquisition, which allow for a fuller appreciation of curved articular surfaces. Assessment for bone marrow oedema subjacent to areas of cartilage irregularity is important to help recognise areas of full thickness fissure. As interest increases in the use of MRI for OA, new sequences are being employed including morphological and compositional techniques which are covered in more detail below .
The meniscus is rich in proteoglycans; this makes the inherent signal characteristics of the meniscus low on all sequences. PD and T2 sequences make for excellent tissue contrast with the adjacent joint fluid and articular cartilage and the addition of fluid sensitive techniques such as fat suppression makes the appreciation of tears better. The meniscus is depicted in two planes, coronal and sagittal, and although the orientation of the collagen fibres cannot be demonstrated on MRI, the type of meniscal tear can be. The stability of meniscal tears can be assessed in order to ascertain the requirement for surgical intervention. The findings of these on MRI include tears greater in size than 1 cm and displacement of meniscal tissue.
A three-dimensional appreciation of the meniscus is key to understanding the tear configuration and how this appears on MRI. Challenging areas include visualisation of the meniscal root, which can be easily overlooked, and the assessment of previously repaired menisci. In the case of diagnostic difficulty, the use of an arthrographic study can aid diagnosis.
The assessment of ligament injury is one of the main indications for MRI examinations. The ligamentous anatomy is usually best assessed by fluid sensitive proton density or T2 images, in at least two planes. Sagittal images can be taken obliquely, to be oriented along the course of the ligament in question, to avoid averaging artefacts which may produce false-positive findings for ligamentous injury. High intrasubstance signal in T2 images suggests injury but in partial ruptures there may be some fibres which retain their normal signal (Fig. 2.3) .
Sagittal image demonstrating intrasubstance high signal following partial rupture of the ACL
2.3.6 Whole Joint Imaging (Semi-quantitative Scores)
Osteoarthritis is a disease of the whole joint. Whilst a broad understanding of the degree of joint involvement is possible using systems such as that of Kellgren and Lawrence, elements of the disease such as synovitis and degree of meniscal involvement are only visualised using MRI. Whilst the degree of cartilaginous damage is predicted using joint space width on plain radiographs, it is only possible to directly measure it using MRI. Likewise, the ability to predict disease progression is higher when there is co-localised non-cartilaginous pathology, such as bone marrow lesions and meniscal pathology .
A number of scoring systems have been devised to categorise the degree of damage to the joint on MRI . These systems are important in the research setting as they allow standardisation of MRI findings in epidemiological and interventional studies in OA. They may have clinical relevance as predictors of progression, but as MRI is not in widespread use for the monitoring of OA, their direct clinical relevance is limited. Such systems include the Whole-Organ MRI Score (WORMS), the Knee Osteoarthritis Scoring System (KOSS), the Boston-Leeds osteoarthritis Knee Score (BLOKS) and the MRI Osteoarthritis Knee Score (MOAKS) [23–26]. In all these systems, scores are attributed to a number of cartilage, synovial, subchondral, ligamentous and meniscal findings, subdivided by the area of the joint affected. All have excellent inter- and intra-rater reliability. Such scores have been used in randomised controlled trials and epidemiological studies of OA. They also appear to be clinically relevant, with the degree of bone marrow change and synovitis being related to the degree of pain in OA. They also appear to be of use in predicting progression of disease.
2.4 Compositional MRI
2.4.1 Imaging Protocols
Compositional MRI aims to assay the biochemical composition of tissues in order to detect early osteoarthritis prior to when it is evident on morphological MRI, at which point cartilage damage is likely irreversible. Potential applications include the identification of individuals who may benefit from early intervention, as well as an outcome measure to evaluate the efficacy of novel interventions within a short timeframe.
A plethora of MRI sequences are available that appear responsive to different cartilage properties . The best validated compositional sequence is delayed Gadolinium Enhanced MRI of Cartilage (dGEMRIC), which provides a measure of cartilage glycosaminoglycan content. Several studies show that dGEMRIC correlates well with the histological severity of osteoarthritis  and may predict future knee osteoarthritis , however, clinical applicability is limited by a long scanning protocol and requirement for potentially nephrotoxic contrast agent. There are a number of sequences with potentially greater clinical value that do not require contrast agent or specialist hardware, and that have acceptable scan times. These include T2 mapping, T1rho, T2* mapping and diffusions protocols . These sequences are responsive to a range of cartilage properties including glycosaminoglycan content, collagen fibre orientation and cartilage hydration. T2 mapping and T1rho are increasingly used in clinical studies, and like dGEMRIC, values correlate with histological degeneration  and baseline values are able to predict progressive osteoarthritis in the knee . T2 mapping is more sensitive than morphological MRI for the identification of early cartilage degeneration , and T2 mapping has been frequently employed to monitor the outcomes of cartilage repair procedures . Compositional MRI may therefore offer diagnostic and prognostic capabilities, as well as providing a measure of intervention efficacy.
2.4.3 Non-cartilaginous Structures
Compositional MRI has also been adopted for the evaluation of meniscus. There appears to be concomitant degeneration of the articular cartilage and meniscus in early osteoarthritis, supported by delayed Gadolinium Enhanced MRI of Meniscus (dGEMRIM) . Meniscal T1rho and T2 mapping values also correlate with the severity of knee osteoarthritis . Given the critical role played by the meniscus, addressing any associated meniscal pathology may be a critical step in delaying or preventing the development of osteoarthritis. Compositional MRI of non-cartilaginous structures is likely to represent an area for future development.
2.5 Computerised Tomography (CT)
CT has a limited but potentially valuable application for imaging early osteoarthritis. It offers excellent three-dimensional assessment of bone morphology, and bone shape is increasingly thought to have predictive value for the development and progression of osteoarthritis. CT arthrography can be employed to both identify cartilage defects and estimate cartilage glycosaminoglycan content . However, MRI has several advantages in that it permits the imaging of all joint structures alongside non-invasive compositional assessment. Combined single photon emission CT and conventional CT (SPECT/CT) may have a potential role in assessing loading of compartments in relation to leg alignment . SPECT/CT tracer uptake in the medial and lateral compartment correlates with varus or valgus alignment, respectively.