Predicting Response to Intra-articular Steroid Therapy in Symptomatic Knee Osteoarthritis

  • Nasimah Maricar

    Student thesis: Phd


    Background: Intra-articular steroid injections (IASI) are widely used in the management of knee osteoarthritis (OA); however, the response to this intervention is variable.Aim: The broad aim of this thesis was to determine disease, treatment and patient-related factors associated with both short-term and longer-term response to IASI in patients with symptomatic knee OA. Methods: A series of systematic reviews were undertaken to summarise and synthesise the literature relating to the project including i) predictors of treatment response, ii) accuracy of IASI and the impact of site of injection and use of image guidance on accuracy and treatment response, and iii) reliability and validity of clinical assessment of knee effusion. Predictors of response were determined from data collected as part of a large open-label clinical trial of IASI in knee OA. Over 200 men and women with painful knee OA had baseline assessment including questionnaires and clinical examination and, in a subsample, contrast-enhanced magnetic resonance imaging scans (MRI), followed by an intra-articular injection of 80 mg Depo-Medrone. A proportion of participants had their injections performed under ultrasound-guidance. Participants were reassessed within 2 weeks to determine response using the OARSI-OMERACT criteria. Responders at this time-point were followed by regular telephone calls until their pain [assessed using the Knee Injury and Osteoarthritis Outcome Score (KOOS)] relapsed within 20% of baseline or till 6 months when they were deemed as longer-term responders. Linear and logistic regression were used to determine whether baseline parameters predicted 'response' both in the short and longer-term.Results: The results of the systematic literature reviews confirmed a paucity of data on factors linked with IASI treatment response in knee OA; variation in accuracy of injection based on site and use of imaging; and also evidence of poor to moderate reliability and poor sensitivity though higher specificity in the assessment of knee effusion. 199 participants (52.3% female), mean age 62.8 years (SD 10.3), were included in the IASI trial. Based on OARSI-OMERACT criteria, 146 (73%) were responders at 2 weeks, while 40 (20%) were longer-term responders. Factors linked with short-term responder status included the presence of medial tibiofemoral and patellofemoral joint tenderness, nocturnal sleep disturbances from knee pain more than three times in the last 2 weeks, and patients' positive perception to IASI. Predictors of longer-term responder status included being in employment and pain-relief lasting greater than 3 months from a previous IASI to the knee. Factors linked with short-term non-responder status included previous ligament injury, a positive click on patella tap test and withdrawal of synovial fluid. Predictors of longer-term non-responder status were depression, greater number of pain sites, the presence of chronic widespread pain, and more severe structural disease as assessed using MRI and radiographs. Other patient, disease and treatment-related factors such as age, duration of disease, the numbers of previous IASI, and the use of guided injection had no influence on the response to the therapy.Conclusion: In individuals with symptomatic knee OA, disease, treatment and person-related factors impact on outcome following IASI. More severe pain and the presence of tenderness at the knee are linked with a better response in the short-term. Those with more severe structural disease on imaging, the presence of chronic widespread pain and higher number of pain sites, and depression appear to limit longer-term successful outcome of the therapy. The results of the study should help improve targeting of therapy in patients with symptomatic knee OA.
    Date of Award31 Dec 2016
    Original languageEnglish
    Awarding Institution
    • The University of Manchester
    SupervisorTerence O'Neill (Supervisor) & Michael Callaghan (Supervisor)


    • injection
    • steroid
    • predictors of response
    • osteoarthritis
    • knee
    • intra-articular

    Cite this