It is well recognised that one of the major limitations of musculoskeletal ultrasound (US) is its operator dependency.1 This issue is currently being addressed by the OMERACT and the EULAR Working Party in Imaging, who are working towards the production of guidelines for the acquisition and interpretation of US images particularly with respect to inflammatory arthritis.2

There are many variables that might influence an US image. These include factors pertaining to the machine (type of machine and transducer and control settings) and those that are not machine related (ambient temperature, transducer pressure3 and patient positioning). Doppler US appears particularly sensitive to all these factors as has recently been highlighted by Torp-Pedersen and Terslev4 and Teh.5 It has previously been reported that there are differences in Doppler signals within muscles4 5 and tendons4^–6 when viewed in either a relaxed or contracted state. However, it has not been reported whether a change of joint position affects the Doppler signal in synovitis.

We would like to illustrate the potential importance of joint positioning for the detection of Doppler signal within joints. To our knowledge there are no previous reports of this in the literature.

A 54-year-old man with active rheumatoid arthritis (RA) underwent an US assessment revealing grey-scale synovitis with power Doppler signal in several finger joints. In the left first metacarpophalangeal joint (fig 1A), grade 2 grey-scale synovitis (using a 0–3 semiquantitative score) was noted with no Doppler signal (fig 1B) by one observer (VL). However, an immediate subsequent scan by another observer (RJW) (fig 1C) revealed marked grade 2 Doppler in addition to the same grey-scale findings (fig 1D). After careful consideration, we determined that a difference in thumb position accounted for the discrepancy. The first observer had scanned the thumb in an extended position in order to improve acoustic access around the joint, while the second observer scanned the thumb in a neutral position. The sonography findings were reproduced by a third sonographer (AZ) who was blinded to the others’ findings.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Effect of extension of thumb joint on Doppler signal. When the first metacarpophalangeal joint was examined in an extended position (A), grade 2 grey scale was found but no power Doppler (B). However, when the thumb was examined in a relaxed or neutral position (C) grade 2 Doppler signal was noted (D). All images were taken using a Philips HDI 5000 machine employing a 15–8 MHz hockey-stick transducer.

On another occasion, a 69-year-old man with early RA was scanned using the same machine. The radial aspect of the second metacarpophalangeal joint (fig 2A) had marked grey-scale synovitis and positive Doppler signal (fig 2B) when scanned in a neutral position. However, when the scan was repeated with the finger in a flexed position (fig 2C) the Doppler signal completely disappeared (fig 2D). This observation was similarly reproduced by all three sonographers.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Effect of flexion on blood flow on the radial aspect of the index finger. When the second metacarpophalangeal joint was examined in a relaxed or neutral position (A), grade 2 grey scale and grade 2 power Doppler was noted (B). However, when the joint was examined in a flexed position (C) the Doppler signal disappeared (D). A small decrease in Doppler gain is shown in (D) compared with (B) but it is unlikely that this accounted for the marked decrease in observed Doppler signal.

The Doppler signal in RA has been shown to correlate with disease activity7 and may have a predictive value in identifying patients at risk of developing structural damage.8 Our two cases illustrate the importance of positioning technique with respect to the determination of the Doppler signal within a joint. It is presumed that overextension or flexion of the joint tightens the joint capsule and collateral ligaments resulting in compression of the vasculature. Additionally, compression may result from raised intracapsular pressure following certain joints positions especially if there is a synovial or fluid collection.9 In conclusion, we highlight a potentially important factor that should be considered when scanning joints for a Doppler signal. A larger study is warranted, however, before formal recommendations can be made.