Systematic review of MRI, ultrasound, and scintigraphy as outcome measures for structural pathology in interventional therapeutic studies of knee arthritis: focus on responsiveness.

OBJECTIVE: validated imaging outcome tools to assess response to therapies in a single joint are required. Our aim was to review the published literature to ascertain the responsiveness of novel imaging techniques as outcome measures in interventional therapeutic studies of knee arthritis.
METHODS: an Ovid Medline search was performed for original articles in English that used imaging techniques to assess response at the knee joint to therapy in osteoarthritis, rheumatoid arthritis, and psoriatic arthritis. Changes in response to therapy were assessed with regard to both internal and external responsiveness.
RESULTS: in the studies that presented appropriate statistical data to allow responsiveness to be assessed, MRI was generally found to be internally responsive to pathologies imaged, and externally responsive, referenced against both other imaging modalities and biochemical biomarkers of arthritis. Ultrasonography was found to demonstrate internal responsiveness with regard to synovial thickness, effusion size, and popliteal cyst size. External responsiveness was demonstrated against several referenced health status measures. Scintigraphy was found to be externally responsive in the majority of studies, with internal responsiveness demonstrated in 1 study.
CONCLUSION: while the imaging techniques appear to be responsive from the data we present, further inspection reveals that interpreting the responsiveness of imaging techniques was difficult, largely because of a lack of standardization of image acquisition, definitions of pathology, and scoring systems. Refined pathological definitions and scoring systems are required to enable the development of valid and responsive tools for interventional clinical trials.