Hall, Michelle C.
An ultrasonographic study of knee joints: features of inflammation and their relationship to radiographic osteoarthritis and pain.
PhD thesis, University of Nottingham.
Background. Knee osteoarthritis (OA) can result in considerable pain and disability for some people. Inflammation within the joint may be partly responsible for the pain associated with OA and a link between inflammation and disease progression has been suggested. Ultrasound (US) imaging has been successfully employed in the evaluation of knee joint effusion, synovial hypertrophy and power doppler signal (PDS) which are said to represent joint inflammation. The associations between US features of inflammation, knee pain and radiographic OA have yet to be firmly established.
Objectives. The objectives of this thesis were to compare the frequency of US features of inflammation in 4 groups from a community sample,  those with normal knees (controls)  knee pain - without radiographic OA (KP)  radiographic OA (without pain) (ROA) and  symptomatic OA (SOA). Associations between US features, knee pain, radiographic change and clinical signs of inflammation could then be explored. Secondary objectives were to determine if US features change in tandem with fluctuations in knee pain (1) over time and (2) with improved pain following a therapeutic intervention in people with SOA.
Methods. In a cross-sectional multiple group comparison study, 243 participants were divided into 4 groups based on the presence of absence of knee pain and ROA. All underwent an US examination for effusion, synovial hypertrophy, peri-articular cysts and PDS. The presence or absence of features, absolute measures (millimetres) and grade of PDS (0-3) was recorded for both knees. Radiographs and clinical evaluation of knee pain, biomechanical stiffness and function were also undertaken.
Follow-up examination of control and SOA groups was undertaken at 3 months. Participants with SOA were then invited to take part in a randomised placebo-controlled study of intra-articular (IA) cortico-steroid and a saline placebo.
Results. The frequency of US features in the control group (effusions (29%) synovial hypertrophy (8%), popliteal cysts (12%) and PD signal (2%)) was not significantly different from those in the KP group. US features were more common in ROA and higher again in SOA (effusion 81% and 92% respectively, synovial hypertrophy 41% and 82%, popliteal cysts 22% and 39%). PDS was not significantly different between ROA (6.3%) and SOA (16%).
Synovial hypertrophy was the only US feature independently associated with knee pain after adjusting for ROA (aOR 6.6; 95% CI 2.85, 15.11). All grey-scale features were strongly associated with ROA and remained so after adjusting for pain (effusion aOR 13.39, 95%CI 6.14, 29.02; synovial hypertrophy aOR 14.39, 95%CI 6.28, 32.94; popliteal cysts aOR 2.82, 95%CI 0.76, 10.43). PDS was not association with either knee pain or radiographic OA.
Change in pain severity was not found to correlate with and change in US measures among the participants followed up at 3 months or following improved pain among participants in the intervention study.
Conclusion. These findings show that US features suggestive of inflammation are higher in participants with SOA but was only significant for synovial hypertrophy. Synovial hypertrophy was confirmed as an independent risk factor for knee pain but was not found to be responsive to temporal changes in pain or improved pain following an IA cortico-steroid or placebo injection. Further studies to understand the contribution of US features of inflammation to pain in knee OA are warranted.
Thesis (University of Nottingham only)
||W Medicine and related subjects (NLM Classification) > WE Muscoskeletal system
||UK Campuses > Faculty of Medicine and Health Sciences > School of Medicine
UK Campuses > Faculty of Medicine and Health Sciences > School of Clinical Sciences
||09 Dec 2013 13:29
||14 Sep 2016 23:05
Actions (Archive Staff Only)