There is increasing interest in the use of ultrasonography (US) in rheumatology (1). Ultrasonography is noninvasive, safe (uses no ionizing radiation), and can be used repeatedly in an outpatient setting which provides immediate access for patients. Availability of US varies widely between hospitals, with most of the referrals being for specific conditions such as rotator cuff tears. This usually requires a separate visit to the radiology department and then a return visit to the referring physician. There is accumulating evidence that US is more accurate than clinical examination in the detection of synovitis and tenosynovitis in small joints (2, 3), and its use in musculoskeletal conditions is becoming increasingly validated (4). There is, however, a paucity of data assessing its actual impact on patient management. This study evaluated the diagnostic and therapeutic impact of musculoskeletal US in rheumatology outpatient clinics.

Of 520 consecutive rheumatology outpatients seen, 100 were referred for US, and were enrolled in the study following provision of informed consent. All patients underwent a routine assessment, including a detailed history and clinical examination by experienced physicians. The indication for US, the site of interest, and site-specific diagnosis (SSD; e.g., synovitis, tenosynovitis), which was diagnosed clinically by the attending physician, were documented. The overall diagnosis (OD; e.g., rheumatoid arthritis, gout) and management plan were also recorded.

Patients had US performed during the same clinic visit (on the requested sites only) by a rheumatology research fellow experienced in US, using an on-site ATL HDI 3000 machine (Advanced Technology Laboratories, Bothel, Washington). A linear array 10-5 MHz “hockey stick” transducer was used to examine most joints and a curvilinear array 5-3 MHz transducer was used to examine the hip. The referring physician subsequently reviewed the US report for each patient in the same clinic, and any change in the diagnosis or management as a result of US was documented.

Of the 100 patients referred for US, 73 were female and the mean age was 50 years (range 17–87 years). Sixty-four patients were referred to confirm a diagnosis alone, while 36 were referred for diagnosis and local corticosteroid injection. Twenty of these 36 patients (56%) had reported a poor response to a previous “blind” corticosteroid injection. A total of 121 sites were examined by US (Table 1). US was requested to confirm the presence or absence of synovitis in 86 of the 121 sites (71%), enthesitis in 11 of 121 sites (9%), and tenosynovitis in 9 of 121 sites (7%).

Table 1. Sites examined by ultrasonography, with frequency of change in site-specific diagnosis (SSD)
Site of scanNo. (%) of sites examinedFrequency of SSD change, no. (%)
Overall121 (100)60/121 (50)
Metacarpophalangeal joints45 (37)24/45 (53)
Wrist17 (14)8/17 (47)
Proximal interphalangeal joints15 (12)5/15 (33)
Hindfoot12 (10)7/12 (58)
Knee10 (8)5/10 (50)
Forefoot7 (6)6/7 (86)
Elbow5 (4)1/5 (20)
Other10 (8)4/10 (40)

Following review of the US findings, the SSD was changed in 53 of 100 patients (53%) and 60 of 121 sites (50%). In order of frequency, the changes in clinical SSD were synovitis in 43 of 60 sites (72%), tenosynovitis in 7 of 60 sites (12%), and enthesitis in 5 of 60 sites (8%). The frequency of SSD change, by site, is listed in Table 1. The OD was changed in 5 of 100 patients (5%). There were a further 8 of 100 patients (8%) in whom US helped confirm a provisional OD. The management plan was altered in 53 of 100 patients (53%) after US, of which 39 were due to a change in SSD and 14 were a result of US confirming a provisional SSD.

The corticosteroid regimen was affected in 43 patients. Planned intraarticular corticosteroid injections were altered in 22 patients, and in 14 patients, a new injection was given after US. Parenteral corticosteroid therapy was affected in 7 patients. Only 14 (39%) of the 36 intended injections were given at the planned intraarticular site. Disease-modifying antirheumatic drug (DMARD) therapy was affected in 13 patients, of which 10 were due to the detection of extensive subclinical synovitis.

This study suggests that US has a diagnostic and therapeutic impact in the majority of referred patients who attend rheumatology clinics. When these findings are applied in the context of all patients attending clinics during the study period, US has an impact on diagnosis and management in at least 10% (53 of 520) of all cases. Consistent with previous reports, this study demonstrates a poor correlation between US and clinical examination in the detection of synovitis; the changes to DMARD therapy were mainly a result of detection of subclinical synovitis by US. This would suggest patients with clinically stable disease are often undertreated, and may help explain the continued bone damage reported in this group of patients (5).

Response to corticosteroid injections is known to vary considerably, and there is evidence that accurately placed injections result in improved patient outcome (6, 7). Interestingly, as a consequence of US, less than half the referred patients received an injection at the preplanned site. The impact of US on corticosteroid injections therefore reflects the limitations of clinical assessment in accurately localizing pathologic sites and may explain, in part, the variation in response to conventionally placed injections. This study also documents the referral pattern when rheumatologists have direct access to US, with most patients referred for assessment of small-joint synovitis and guided injections.

After initial capital expenditure, the only running costs for US are the service contract and operators' time, in this case, taking 10–15 minutes per patient assessment. In the United Kingdom, charges for an individual small-joint US (which may include the cost of injection) vary from $50 in the public sector to $200 (in United States dollars) in the private sector. The US examinations took place during the outpatient visit, where direct access has the advantage of immediate alteration in the management plan, thereby avoiding additional hospital visits and associated costs.

There are limitations to this study. There was an initial selection bias, since the patients assessed were all referred for US in the first instance, suggesting a degree of uncertainty in the diagnosis or management at the outset. Although a change in the clinical diagnosis or management plan is an important first step, it does not necessarily follow that outcome will be improved. Further work is required to determine whether these changes translate to an improvement in patient outcome. Any benefit will have to be offset with the additional cost of US equipment and training, but should also take into account the direct and indirect cost benefit of a “one-stop” patient service.

In this preliminary, nonrandomized observational study, US has a major impact on the diagnosis and management of musculoskeletal conditions. US is safe, relatively inexpensive, and can be performed effectively by rheumatologists, raising important questions for the future training and practice of rheumatology. A randomized study assessing change in diagnosis and management after US, with specific clinical outcome measures, is warranted.


  1. Top of page
  • 1
    Grassi W, Cervini C. Ultrasonography in rheumatology: an evolving technique. Ann Rheum Dis 1998; 57: 26871.
  • 2
    Backhaus M, Kamradt T, Sandrock D, Loreck D, Fritz J, Wolf KJ, et al. Arthritis of the finger joints: a comprehensive approach comparing conventional radiography, scintigraphy, ultrasound, and contrast enhanced magnetic resonance imaging. Arthritis Rheum 1999; 42: 123245.
  • 3
    Conaghan PG, Wakefield RJ, O'Connor P, Gibbon WW, Brown C, McGonagle D, et al. MCPJ assessment in early RA: a comparison between Xray, MRI, high-resolution ultrasound and clinical examination [abstract]. Arthritis Rheum 1998; 41 suppl 9: S246.
  • 4
    Wakefield RJ, Gibbon WW, Conaghan PG, O'Connor P, McGonagle D, Pease C, et al. The value of sonography in the detection of cortical bone erosions: a comparative study with conventional radiography. Arthritis Rheum 2000; 43: 276270.
  • 5
    Mulherin D, Fitzgerald O, Bresnihan B. Clinical improvement and radiological deterioration in rheumatoid arthritis: evidence that the pathogenesis of synovial inflammation and articular damage may differ. Br J Rheumatol 1996; 35: 12638.
  • 6
    Eustace JA, Brophy DP, Gibney RP, Bresnihan B, Fitzgerald O. Comparison of the accuracy of steroid placement with clinical outcome in patients with shoulder symptoms. Ann Rheum Dis 1997; 56: 5963.
  • 7
    Jones A, Regan M, Ledingham J, Pattrick M, Manhire A, Doherty M. Importance of placement of intra-articular steroid injections. BMJ 1993; 307: 132930.