| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Developments |
1 From the Centre for Rheumatic Diseases, University Department of Medicine, Royal Infirmary, Queen Elizabeth Bldg, 10 Alexandra Parade, Glasgow G31 2ER, Scotland (W.R.F., P.V.B., R.D.S.); and the Department of Biological Sciences, University of Paisley, Scotland (C.G.E., J.C.L.). Received August 18, 2000; revision requested September 26; revision received November 13; accepted January 9, 2001. Supported by Integrated Clinical Arthritis Centre award S0590 from the Arthritis Research Campaign. Address correspondence to W.R.F. (e-mail: w.ferrell@bio.gla.ac.uk).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
Index terms: Arthritis, rheumatoid, 437.71 • Extremities, 437.71 • Fingers and toes, 437.71 • Hand, 437.71 • Lasers, 437.12984 • Lasers, Doppler study, 437.1299 • Ultrasound (US), power Doppler studies, 437.12984
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
Ultrasonography (US) can depict joint effusion, synovial tissue proliferation, and erosions in the metacarpophalangeal (MCP) joint (6,7), but the lower limit of its accuracy for depicting these abnormalities is not known. Other abnormalities associated with the MCP joint in RA (ie, pannus and effusion around tendons, tendon rupture, and rheumatoid nodules) can be detected with US (8,9). The power Doppler technique creates a color flow map through a sample volume on the basis of the total integrated power of the Doppler spectrum. Power signals are generated by the movement of blood cells within vessels, and power Doppler US relates to the volume of blood flowing within the imaging field (10).
Laser Doppler perfusion imaging is a recently developed technique for noninvasive assessment of blood flow through vascular beds, on the basis of the well-known Doppler shift principle, that yields a spatial map of tissue perfusion (11). The red (635-nm) wavelength used in conventional laser Doppler imagers limits penetration to the skin, but the optical properties of skin are such that longer wavelengths have greater tissue penetration power (12); thus, imaging of perfusion in deeper tissues is possible. Previous work (13) has demonstrated that elevated perfusion associated with the proximal interphalangeal joints is detectable in patients with RA with use of a near-infrared laser. However, the MCP joints of patients with RA are commonly affected early in the disease process. At present, there are no simple and noninvasive but objective measures of inflammatory activity associated with these joints.
The purpose of this cross-sectional study was to perform laser Doppler imaging and US of the hands of patients with known RA who were judged on clinical grounds to have pain and tenderness of the MCP joints and to establish whether elevated perfusion associated with MCP joints 2 and 3 was detectable.
|
Materials and Methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
A separate group of 13 healthy control subjects (10 women and three men; mean age, 41.2 years ± 12.8; age range, 27–63 years), who were not age matched to the patient group, were recruited consecutively. None of the control subjects had a history of any hand injury or disease; at the time of examination, their hands were clinically normal and asymptomatic. The MCP joints 2 and 3 of both hands were examined with laser Doppler imaging.
All study participants were asked to avoid physical activity before the examination, and none had applied any cream to the hand or recently undergone physical therapy. For the patient group, the joints were also examined with both gray-scale and Power Doppler US. The room temperature was monitored, as was the skin temperature over the dorsum of the fourth finger to ensure that the differences between groups were not the result of variations in environmental and, consequently, skin temperature. This study was approved by the institutional ethics committee, and informed consent was obtained from each participant.
Laser Doppler Imaging
A laser Doppler imager (Moor Instruments, Axminster, UK) was especially modified to incorporate a near-infrared (835-nm) laser, to increase tissue penetration, in addition to the standard red (635-nm) laser. The imager was positioned 60 cm above the surface of the hand for all participants. The laser beam (
1-mm diameter) was scanned in a raster fashion across the dorsum of the hands. From the backscattered light, a spatial image of tissue perfusion was generated that depended on the extent of the Doppler shift of this light (11). An array of as many as 256 x 256 measurement points was obtained, and the typical scanning time was about 3 minutes. A perfusion measurement was obtained for each point by calculating the product of erythrocyte velocity and concentration, to yield a flux value in arbitrary perfusion units. Red and near-infrared scans were obtained sequentially; the sequence was randomly varied. Subsequent image analysis was performed (W.R.F.) with the manufacturer’s dedicated software, which displayed a color-coded image of tissue perfusion on a monitor.
All values were stored on a computer disk. On the light intensity (photo) image (Fig 1a, left), an area over the MCP joint was designated as the region of interest. A rheumatologist (R.D.S.) initially defined the region of interest on the basis of the surface anatomy. The region was an ovoid area (size range, 1.8–2.6 cm2) in which the median flux value was computed. The same region-of-interest placement technique was used for both patients and healthy subjects, and the same person performed the measurements. The laser Doppler imaging method was essentially similar to that previously described (13) for examination of the proximal interphalangeal joints.
|
|
|
The power Doppler zero level was established before the study. The power color gain was always adjusted to such a level that no power Doppler sign (red pixels inside the active green box) appeared in the active state of the probe with air contact or after gel was applied to the surface of the probe. With this setup, there was no power Doppler sign when healthy MCP joints were imaged. A low wall filter and low flow optimum were chosen from the software. The pulse repetition frequency varied between 500 and 1,000 Hz. During the study, scans were obtained when stable red pixels were observed with no pixels present under cortical bone. In this way, we attempted to exclude the main disadvantages of the power Doppler technique, namely motion sensitivity and common flash artifacts. Quantification of the hyperemia was not possible with power Doppler imaging; we could observe only the presence or absence of the pixels in the region of interest.
We used a dorsal approach in this study because it was the plane used at laser Doppler imaging, and we wanted to avoid the relatively large pulsatile digital arteries, which lie laterally along the joints. For gray-scale US, the region of interest was centered across the MCP joint line, and its size was strictly dependent on the 26-mm footprint of the transducer. The region of interest at power Doppler US was necessarily smaller and was based on pathologic features but always included the whole joint. All the US images were stored on magneto-optical disks for off-line analysis.The laser Doppler imaging and US measurements were performed separately, one immediately after the other, by different operators (C.G.E. and P.V.B., respectively); the order was randomly varied in successive patients. Laser Doppler imaging and US measurements were analyzed independently by different investigators (W.R.F. and P.V.B., respectively) so that they were blinded to results of the other tests until the comparison stage of the study. For each imaging modality, the same person sized and placed the regions of interest in each case.
Statistical Analysis
Before the main study of patients with RA was started, measurements were taken on two occasions for the first seven control subjects recruited, to obtain data for power calculations and between-day and within-day variability assessment. The other six control patients had not yet been recruited.
On the basis of the data obtained in the seven healthy subjects, power calculations indicated that six participants in each group would require 90% power to depict a 30% change in flux (
= 0.05). Data analysis was performed with software (MINITAB; Minitab, State College, Pa). Linear correlations were calculated with the Pearson product moment correlation coefficient for two sets of interval scale data (eg, flux, visual analog score for pain). The point biserial correlation coefficient was used for comparing interval scale (continuous) data (eg, flux) with nominal scale (dichotomous) data (eg, presence or absence of the power Doppler sign). The 
2 test was used for comparing two nominal scale data sets.
Perfusion values between groups were compared by calculating the mean value for all four MCP joints to yield a single value per participant. This calculation was necessary because RA characteristically affects multiple finger joints; thus, individual joints can not be considered independent of one another. Independent treatment of the MCP joints was appropriate for comparisons between laser Doppler imaging and the power Doppler sign, however, since the latter yields only dichotomous data that can not be summed. Comparisons were between two methods, which could give rise to contradictory results between individual joints. Two interval scale data sets were compared with a paired or unpaired Student t test, as appropriate. Interval scale data were expressed as the mean ± SD.
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
The threshold perfusion value for separating the two groups was 200 perfusion units, which lies clearly above the maximum value of 129 perfusion units obtained in healthy subjects in similar environmental conditions. Image analysis strongly confirmed this classification, with the high perfusion group showing on average at least a threefold higher perfusion in MCP joints 2 and 3 bilaterally compared with that in the low perfusion group (Table). This finding was supported by the mean value for the high perfusion group (395.5 perfusion units ± 118.9), which differed significantly (P = .002, unpaired t test) from the value in the low perfusion group (114.8 perfusion units ± 30.1).
|
Elevated MCP perfusion and hand dominance were significantly correlated (point biserial r = 0.9; P < .001; 52 joints examined). The high perfusion group had a significantly (P = .027) lower disease duration (6.1 years ± 5.5) than the low perfusion group (15.5 years ± 8.5). Perfusion values from the MCP joints of the 13 healthy subjects ranged from 57 to 129 perfusion units. The mean value (94.3 perfusion units ± 16.6, 13 subjects examined) differed significantly from that in the high perfusion group (P = .002) but not from that in the low perfusion group (P = .135). Differences between the high and low perfusion groups could not be explained on the basis of temperature variations, because there was no significant difference in room temperature (24.2°C ± 0.82 and 24.6°C ± 0.34, respectively; P = .22) or skin temperature (33.6°C ± 0.43 and 32.2°C ± 0.34, respectively; P = .48) between the groups.
Within-day and between-day variabilities measured at the second MCP joint in seven of the 13 healthy subjects on two occasions were 3.1% ± 3.4 and 3.9% ± 4.1, respectively. The cross-sectional nature of the present investigation precluded such assessment for the patients.
Comparison of Power Doppler US and Laser Doppler Imaging
The power Doppler sign was present in some MCP joints (Fig 1c), but the correlation between the power Doppler sign and laser Doppler imaging perfusion was weak (point biserial r = 0.244) and nonsignificant (P = .1, 44 joints examined). Comparison of the power Doppler sign (present or absent) and laser Doppler imaging flux signal (high or low), both as nominal scale data, showed that there was agreement in 20 of 44 joints (ie, the power Doppler sign was present in a high perfusion joint or vice versa). The 2 test indicated that this result did not differ significantly (P = .72, 44 joints examined) from that which occurred by chance alone.
Correlation with the Visual Analog Score
Patients in both the high and low perfusion groups experienced pains associated with their finger joints, but there was a noticeable difference between the groups. Patients in the high perfusion group showed a significantly positive linear correlation (r = 0.55) between the MCP perfusion values and the pain score (P < .005, 24 joints examined), which strongly suggests this pain had inflammatory origin (Fig 2). Analyzed separately, the low perfusion group showed an inverse but nonsignificant correlation (r = -0.34, 28 joints examined) between laser Doppler imaging perfusion and the visual analog score for pain, which suggests that for this patient group, joint pains are unlikely to be of inflammatory origin but are rather of mechanical origin. In contrast, the power Doppler sign was not correlated with the visual analog score for pain (point biserial r = -0.001, 44 joints examined).
|
2 test). Elevated laser Doppler imaging perfusion values were significantly (P = .001) correlated with anechoic or hypoechoic regions (point biserial r = 0.76, 40 joints examined). In contrast, comparisons with the power Doppler sign revealed no close agreement. There was agreement in 18 of 40 joints (ie, anechoic region in a joint with a power Doppler sign or vice versa) (P = .32, 2 test). |
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
The lack of correspondence between the power Doppler sign and laser Doppler imaging may be related to differences in the parameters measured with the two techniques. Power Doppler imaging can depict the amount of blood flowing in the tissue rather than the velocity of blood flow (10), whereas laser Doppler flux is the product of red blood cell velocity and the concentration of these cells. Thus, these two techniques measure rather different aspects of blood flow, and laser Doppler imaging appears to be more sensitive. Gray-scale US can be used to detect synovitis associated with inflamed joints in patients with RA (7).
Areas of synovitis are associated with anechoic or hypoechoic signals, and there was a strong association between the occurrence of such areas in joints that also showed high perfusion measured with laser Doppler imaging. There was no significant correlation with the power Doppler sign, however, which again suggests that at present this method is less sensitive than laser Doppler imaging. The lack of correspondence between the power Doppler sign and synovitis with gray-scale US observed in the present study contrasts with previous findings (7). This may be because of the greater scope for variation between technicians or machines. With laser Doppler imaging, however, there is no contact with the patient, which reduces variation between operators, and laser Doppler imagers automatically compensate for variations in laser power. Within-day and between-day variabilities of laser Doppler imaging measurements at the MCP joints were similar to values we previously reported for the proximal interphalangeal joints (13).
The observation that the high and low perfusion groups showed a significant positive correlation between perfusion and the pain score suggests that, for these groups, the pain was likely to be related to the inflammatory process. In the low perfusion group, there was no correlation between perfusion and the pain score, which suggests that for this group the pain was not inflammatory in origin but may have been related to other factors such as bone and cartilage damage. This information may be useful for determining the most effective drug therapy, since antiinflammatory drugs would be most appropriate for the high perfusion group. In the low perfusion group, however, such drugs would not confer any benefit unless inflammation was present in joints elsewhere.
Joint pain is the most common symptom in patients with RA, but, as findings in this study indicate, it does not help discriminate between pain of inflammatory origin and pain of mechanical origin. Such distinction is important, since inflammatory pain is likely to arise from hypervascularized pannus formation, which is associated with the invasive and destructive phase of RA (16). Clinical assessment of inflammatory status is more difficult at the MCP joints compared with at the proximal interphalangeal joints, since diameter is difficult to measure at the MCP joints, and joint tenderness does not help discriminate between articular pains of differing origins.
The principal limitation of the laser Doppler imaging method is that measurement depth is difficult to establish and is likely to vary depending on the skin properties of the individual patient. However, penetration depth increases with increasing wavelength (12); thus, the use of lasers that operate further into the infrared wavelength should improve detection of elevated perfusion associated with inflammation. The near-infrared (835-nm) laser can penetrate skin to a depth of approximately 1,300 µm before the incident optical energy density diminishes to one-third of its original value (12). Although skin hinders light penetration, only a small fraction of backscattered photons is necessary to yield the flux signal. Furthermore, the intensity of the backscattered light is normalized, and corrections are made for variations in laser power. Consequently, measurements can be obtained from vascular beds deeper than 1,300 µm, although inflammatory lesions may not be detectable in deeper, more proximal joints such as the shoulder.
Another limitation of laser Doppler imaging is that an external standard of reference is not available for comparison. In animal experiments, however, we found that laser Doppler imaging correlates well with absolute measurement of blood flow with radiolabeled microspheres (17). Although this can not be tested in humans, it is likely that tissue perfusion measured with laser Doppler imaging provides an accurate indication of changes in underlying blood flow.
These findings suggest that laser Doppler imaging has the potential to provide an objective assessment of inflammatory hyperemia in finger joints with RA and possibly in other soft tissues. This could prove to be a useful research tool for investigating efficacy of new treatments and providing further insight into the mechanisms of the disease process. It is possible that future development of this noninvasive and intrinsically safe technique could allow it to be used routinely as an initial assessment of inflammatory status.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, W.R.F.; study concepts, J.C.L., W.R.F., R.D.S., P.V.B.; study design, J.C.L., C.G.E., P.V.B.; literature research, J.C.L., W.R.F., P.V.B.; clinical studies, R.D.S., P.V.B.; data acquisition, P.V.B., C.G.E., J.C.L.; data analysis/interpretation, C.G.E., W.R.F., P.V.B.; statistical analysis, W.R.F.; manuscript preparation, P.V.B., W.R.F., C.G.E.; manuscript definition of intellectual content, W.R.F., R.D.S., P.V.B., J.C.L.; manuscript editing, revision/review, and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods Results DiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. V Balint, P. Mandl, and D. Kane "All that glistens is not gold" Separating artefacts from true Doppler signals in rheumatological ultrasound Ann Rheum Dis, February 1, 2008; 67(2): 141 - 142. [Full Text] [PDF]
|
|
|
|
|
|
D Kane, J C Lockhart, P V Balint, C Mann, W R Ferrell, and I B McInnes Protective effect of sensory denervation in inflammatory arthritis (evidence of regulatory neuroimmune pathways in the arthritic joint) Ann Rheum Dis, February 1, 2005; 64(2): 325 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. Murray, A. L. Herrick, and T. A. King Laser Doppler imaging: a developing technique for application in the rheumatic diseases Rheumatology, October 1, 2004; 43(10): 1210 - 1218. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Beckers, C. Ribbens, B. Andre, S. Marcelis, O. Kaye, L. Mathy, M.-J. Kaiser, R. Hustinx, J. Foidart, and M. G. Malaise Assessment of Disease Activity in Rheumatoid Arthritis with 18F-FDG PET J. Nucl. Med., June 1, 2004; 45(6): 956 - 964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Ribbens, B. Andre, S. Marcelis, O. Kaye, L. Mathy, V. Bonnet, C. Beckers, and M. G. Malaise Rheumatoid Hand Joint Synovitis: Gray-Scale and Power Doppler US Quantifications Following Anti-Tumor Necrosis Factor-{alpha} Treatment: Pilot Study Radiology, November 1, 2003; 229(2): 562 - 569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. G. Egan, J. C. Lockhart, W. R. Ferrell, S. M. Day, and J. S. McLean Pathophysiological basis of acute inflammatory hyperaemia in the rat knee: roles of cyclo-oxygenase-1 and -2 J. Physiol., March 1, 2002; 539(2): 579 - 587. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
= high perfusion group, 
= low perfusion group) compared with flux determined with laser Doppler imaging. Note that all but one of the MCP perfusion values in the low perfusion group lie below 200 perfusion units (PU), and all but one in the high perfusion group lie above that value.