Optimum Radiographic Assessment of the Knee

Overview

X-rays are the most frequently used imaging test when evaluating the knee for joint replacement. They are non-invasive, safe and cost effective. They allow assessment of: progression of disease, appropriateness for joint replacement, in particular unicompartmental knee replacement (UKR), as well as likely prognosis following replacement. Despite a multitude of standardised views there is a lack of consensus regarding the optimum views to evaluate joint space narrowing within each compartment (lateral, medial and patellofemoral). This study will evaluate the status of knee cartilage in 225 patients with varying degrees, and patterns, of knee osteoarthritis (OA) using standing extension anteroposterior, 15 degrees flexion posteroanterior, 45 degrees flexion posteroanterior and valgus and varus stress views as well as MRI. These results will be compared to the gold standard imaging technique of stress views as well as to direct measurements of retrieved tissue in those patients who undergo knee replacement surgery. The sensitivity and specificity of each of the imaging techniques at predicting suitability for UKR will be calculated, the optimum imaging views proposed, and ultimately the results of this study will be used to develop a decision aid, based on optimum views, to help clinicians decide between likelihood of a patient being a candidate for UKR based on pre-operative X-ray findings.

The optimum imaging protocol is one that is acceptable to patients, involves the fewest X-rays to obtain the most clinically relevant information and one that utilizes the least resources in terms of staff and equipment. Currently standard assessment involves: standing anteroposterior, lateral and skyline views. In addition in patients being considered for joint replacement valgus/varus stress X-rays are used to evaluate the lateral compartment (as well as medial collateral ligament) and medial compartment respectively to assess the status of the cartilage. In patients with loss of cartilage on one side, typically medial, but preserved cartilage on the other a UKR, as opposed to total knee replacement (TKR), may be indicated. Currently there is a lack of consensus amongst orthopaedic surgeons as to the best way to image the knee joint to establish degree and pattern of arthritis. For a long time it has been known that weight bearing views are a better method at establishing the true joint space compared to non-weight bearing views width due to the increased forces across the joint. In addition it is known that full extension views, despite being the most commonly used view, may also underestimate joint space narrowing as in full extension the femur and tibia articulate across an area of the joint that is not typically not involved during activity, and hence can have relatively well preserved cartilage, giving a false impression of the joints disease state. These findings have been adopted, and continue to be adopted, into routine clinical practice and there continues to be an increase in the proportion of surgeons performing standing and semi-flexed views. However the best method of evaluating the disease state in each compartment has yet to be defined. When deciding between UKR and TKR the detection of the degree and pattern of arthritis is of critical importance. UKR are known to perform poorly in partial thickness disease and require full thickness cartilage in the retained tibiofemoral compartment. To establish whether a patient meets the indications for UKR X-rays are used with stress views being the gold standard, as well as the standard assessment that the studies of long term outcomes on UKR are based. Gibson and Goodfellow, who were first to describe stress X-rays in the workup of a patient for UKR, reported that those patients with a joint space width of more than 5mm in the lateral compartment had intact lateral cartilage during surgery making them appropriate for UKR. More recently Waldstein et al. reported that patients with a lateral joint space width of more than 4mm may be appropriate for UKR however overall they noted poor correlation between joint space width measured on valgus stress views and intra-operative Outerbridge grade. In addition to the low quantity of evidence regarding the relationship between stress views and intra-operative status of the joint the feasibility of performing stress X-rays also limits their use. Stress X-rays are resource dependent, can be uncomfortable for patients and require an additional practitioner. As such they are often not performed with many clinicians adopting alternative X-ray views, MRI or direct observation via arthroscopy. It has been proposed that standing views with the knee in 15 degrees then 45 degrees flexion may load the medial and lateral compartments respectively and that these views may be an alternative to stress views without the requirement for an additional practitioner. However the outcomes based on these forms of assessment, and the relationship between the joint space width measured using these contemporary techniques has not been reported. An alternative would include a stress device that allows a stress X-ray to be performed without the requirement of the clinician. This study will evaluate the status of knee cartilage in 225 patients with varying degrees, and patterns, of knee OA using standing extension anteroposterior, 15 degrees flexion posteroanterior, 45 degrees flexion posteroanterior and valgus and varus stress views as well as MRI. These results will be compared to the gold standard imaging technique of stress views as well as to direct measurements of retrieved tissue in those patients who undergo knee replacement surgery. The sensitivity and specificity of each of the imaging techniques at predicting suitability for UKR will be calculated, the optimum imaging views proposed, and ultimately the results of this study will be used to develop a decision aid, based on optimum views, to help clinicians decide between likelihood of a patient being a candidate for UKR based on pre-operative X-ray and MRI findings.