Rheumatology 2000;39:1214–1217
Novel use of laser Doppler imaging for
investigating epicondylitis
W. R. Ferrell, P. V. Balint and R. D. Sturrock
Centre for Rheumatic Diseases, Department of Medicine, Royal Infirmary,
10 Alexandra Parade, Glasgow G31 2ER, UK
Abstract
Objective. This investigation evaluated a novel form of tissue perfusion measurement, laser
Doppler imaging (LDI ), in a case of lateral epicondylitis to establish if it might have
applications in assessing soft tissue lesions. LDI was used in conjunction with
ultrasonography to provide information about tissue oedema as well as the power Doppler
signal as an alternative method of assessing blood flow.
Methods. A laser Doppler imager with a near-infrared (NIR) laser source was used to
improve tissue penetration and yield measurements of perfusion (flux) from structures under
the skin. Skin temperature over the lateral epicondylar region was also measured.
Ultrasonography was used in both grey-scale and power Doppler modes. LDI, temperature
measurements and ultrasonographic data were obtained before treatment and serially after
local injection of methylprednisolone.
Results. Before treatment there was increased perfusion and skin temperature and the
presence of a power Doppler sign associated with the right lateral epicondyle as well as
oedema at the extensor origin. None of these was present at the asymptomatic contralateral
epicondylar region. Twenty-four hours after methylprednisolone administration, both
perfusion and skin temperature had increased, and they declined over the subsequent 48 h.
Although skin temperature had declined to normal (referenced to the contralateral epicondyle)
by the third day after injection, it took until the eleventh day after injection for perfusion to
normalize.
Conclusions. LDI using an NIR laser source appears to be an effective non-invasive method
for the examination of inflammatory responses in soft tissue, with greater sensitivity than
thermally based methods. In addition, LDI was found to correlate with power Doppler
ultrasonography.
K : Laser Doppler imaging, Ultrasonography, Epicondylitis.
Objective investigation of soft tissue injury or inflammation is difficult, as there are few techniques which
adequately permit the non-invasive assessment of tissue
perfusion. Infrared thermography has been used to
investigate soft tissue lesions, including epicondylitis [1],
and although it is useful it requires careful control of
environmental temperature and acclimatization of the
subject before measurement. Laser Doppler flowmetry
is a non-invasive technique based on the well-known
Doppler shift principle for detecting changes in skin
blood flow (perfusion) at a single point [2]. Laser
Doppler imaging (LDI ) extends this technique by scanning a laser beam across the tissue to generate a spatial
map of perfusion [3] with hundreds or thousands of
measurement points per scan, depending on the area
scanned. This technique has proved to be of value in
assessing burns [4], and when a near-infrared (NIR)
laser is used hyperaemia in inflamed finger joints is
Submitted 24 January 2000; revised version accepted 9 June 2000.
Correspondence to: W. R. Ferrell.
detectable in patients with rheumatoid arthritis [5]. In
this preliminary study we describe the novel use of laser
Doppler imaging (LDI ) in conjunction with ultrasonography to investigate a case of lateral epicondylitis.
Materials and methods
Following overuse, one of the authors ( WRF ) developed
lateral epicondylitis of the right elbow, characterized by
localized pain and tenderness over the epicondyle and
increased pain on resisted wrist extension. These
symptoms had remained stable over the 8 weeks before
examination. The left elbow was asymptomatic.
A laser Doppler imager (Moor Instruments,
Axminster, UK ) scanned a laser beam across the elbow,
producing a two-dimensional perfusion map of
256 × 256 measurement points. The instrument incorporated an NIR laser (830 nm), although on each occasion a scan was also taken using a red (635 nm) HeNe
laser. The optical properties of skin are such that
wavelengths longer than the visible spectrum penetrate
1214
© 2000 British Society for Rheumatology
Doppler imaging in epicondylitis
1215
F. 1. Colour panels showing progressive LDI near infrared scans from the affected right and unaffected left elbows on various days before (0) and after injection of
methylprednisolone. Scans are colour-coded: dark blue represents the lowest degree of perfusion, and increasing perfusion values are represented by green, yellow, red and
white. Power Doppler scans are shown at the appropriate time points along with photographic images of the elbows and diagrams of the bony structure. CE, common
extensor origin; LEC, left edge of the capitulum; RH, radial head; LEH, left edge of humerus. The position of the ultrasonic transducer over the lateral epicondyle of the
right arm is also shown.
1216
W. R. Ferrell et al.
more deeply [6 ], and previous work indicates that
hyperaemia associated with inflamed [5] or injured [7]
finger joints is readily detectable using the NIR laser.
Measurements were performed in a room relatively free
of draughts and an ambient temperature of 22–25°C,
but no subject acclimatization was employed. Both the
affected right elbow and the asymptomatic left elbow
were scanned periodically before and after treatment,
the left side being an internal control. We had demonstrated previously that within- and between-day variability for LDI measurements of proximal interphalangeal
joints was between 3 and 4% [5]. Skin surface temperature over both epicondyles was also measured on the
same occasions as scans were taken, using thermocouples
connected to a monitor (Harvard Inc, MA, USA) with
a resolution of 0.1°C.
An ATL ( WA, USA) HDI 3000 ultrasound machine
with a compact linear 10–5 MHz, 26 mm probe was
used to obtain 2D grey-scale and power Doppler images
and to detect altered structure of the origin of the
common extensors and blood flow in a coronal plane
(Fig. 1). A previously described technique was used to
avoid artefacts [8]. Colour noise was set before examination, it was checked that there was no signal under the
cortical bone, and the contralateral side was also examined. Pulse repetition frequency setup was at 500 Hz
and low-flow optimum was selected from the software.
Using grey-scale ultrasonography, we had examined the
thickness of the iliofemoral ligament previously and
found intra- and inter-observer error to be 6.1 and
10.6% respectively [9]. To avoid errors [10] we examined
the contralateral asymptomatic side with the same
settings.
Ultrasound scans were taken before injection and 11
days after the injection of 10 mg methylprednisolone at
the site of maximal tenderness, close to the extensor
origin.
Results and discussion
Skin temperature over the affected epicondyle was initially elevated (by 1.1°C ) compared with the unaffected
side and was associated with the LDI perfusion value
being more than three times higher over the epicondylar
region ( Figs 1 and 2A and 2B). Power Doppler ultrasonography localized the increased blood flow volume
to deeper structures, with only weak signals in the
tendon (Fig. 1). Grey-scale ultrasonographic examination revealed a hypoechogenic region associated with
the common extensor origin on the affected side, suggestive of oedema.
Methylprednisolone administration resulted in an
inflammatory response 24 h later. Concurrently there
was an increase in epicondylar temperature ( Fig. 2A)
and perfusion (Figs 1 and 2B), which had declined by
48 h after injection ( Figs 1 and 2B). Epicondylar temperature also fell and became equivalent to that on
the unaffected side by 72 h after injection (Fig. 2A).
However, perfusion remained elevated compared with
the unaffected side and took a further 7 days to approxi-
F. 2. Temperature (A) and perfusion (flux) (B) values measured over the inflamed (right) and control ( left) epicondyles
as a function of time before and after steroid injection.
Although skin temperature normalized by the third day after
injection, a further 7 days was required for perfusion to
normalize. (C ) Skin temperature and perfusion were poorly
correlated below a skin temperature of 33°C. Data points for
both the affected right (filled circles) and unaffected left (open
circles) elbows are plotted, but R2 refers to the combined data
sets. No significance was obtained when the individual data
sets were assessed statistically.
Doppler imaging in epicondylitis
mate the value on the unaffected side (Fig. 2B), even
though the symptoms had resolved completely by the
third day after injection. On the eleventh day, when the
LDI signal approximated the unaffected side, the power
Doppler sign was virtually absent (Fig. 1). At this time
the grey-scale ultrasonogram showed reduced hypoechogenic features. The correlation between skin temperature and LDI perfusion proved to be non-significant
below a skin temperature of 33°C ( Fig. 2C ), although
there was some correspondence at higher skin temperatures in the 48 h after the administration of
methylprednisolone.
This investigation is novel in two respects: LDI has
never previously been used to examine a case of epicondylitis and, although ultrasonography has been used to
investigate normal elbow joint structure [11, 12] and
lateral epicondylitis [13, 14], there has never previously
been an investigation combining the two techniques.
The results from this longitudinal case study indicate
that LDI is sufficiently sensitive to detect hyperaemia
associated with epicondylitis. Although both techniques
are based on the Doppler shift principle, there are
differences in the signals obtained with these two techniques. The LDI perfusion signal is the product of the
velocity and the concentration of blood cells within a
volume of tissue. However, the power Doppler technique
creates a colour flow map through a sample volume
based on the total integrated power of the Doppler
spectrum, but all velocity information is sacrificed [8].
Further work is required to establish the reproducibility
and repeatability of grey-scale and power Doppler ultrasonography at the lateral epicondyle of the humerus in
asymptomatic and symptomatic individuals.
An important issue is whether LDI penetrates sufficiently to detect hyperaemia of structures under the
skin. There is indirect evidence to support the concept
that the laser penetrates to deeper structures, as subcutaneous veins were clearly imaged with the 830 nm laser
but not with the less penetrating red source (not shown).
In animal experiments, in which variables can be controlled more tightly, it has been demonstrated that
changes in synovial perfusion are detectable through
intact skin [15].
Our grey-scale findings of a hypoechoic region at the
insertion of the extensor carpi radialis brevis confirm
previous observations [13, 14]. On power Doppler examination, a stronger signal was observed at the periosteum
than in the tendon itself, which correlates with the
tendon being poorly vascularized, whereas periosteum
has a rich vascular network.
Over the basal physiological (∏ 33°C ) range of perfusion and skin temperatures encountered, there was no
significant correlation between these two variables. This
may indicate that thermally based techniques involving
the measurement of surface temperature are less sensitive
than LDI. Although there was correspondence between
LDI and temperature initially, as the inflammation
resolved and the surface temperature dropped below
33°C this association diminished (Fig. 2A and B). Thus
1217
LDI can detect hyperaemic areas associated with injury
or inflammation with greater sensitivity than thermally
based methods. LDI has the potential to be an effective
diagnostic tool for the non-invasive assessment of soft
tissue injuries such as epicondylitis, and may prove to
be particularly useful for monitoring progress after
symptoms have resolved. This may have predictive value
in determining when exercise can be resumed and could
provide objective parameters for evaluating the effectiveness of interventions in clinical research.
Acknowledgement
The authors gratefully acknowledge the support of the
Arthritis Research Campaign (ICAC award S0590).
References
1. Thomas D, Siahamis G, Marion M, Boyle C.
Computerised infrared thermography and isotopic bone
scanning in tennis elbow. Ann Rheum Dis 1992;51:103–7.
2. Nilsson GE, Tenland T, Oberg PA. A new instrument for
continuous measurement of tissue blood flow by light
beating spectroscopy. IEEE Trans Biomed Eng
1980;27:597–604.
3. Wårdell K, Jakobsson A, Nilsson GE. Laser Doppler
perfusion imaging by dynamic light scattering. IEEE Trans
Biomed Eng 1993;40:309–16.
4. Niaza ZBM, Essex TJH, Papini R, Scott D, McLean NR,
Black MJM. New laser Doppler scanner, a valuable
adjunct in burn depth assessment. Burns 1993;19:485–9.
5. Ferrell WR, Sturrock RD, Mallik AK, Abbot NC,
Lockhart JC, Edmondson WD. Laser Doppler perfusion
imaging of proximal inter-phalangeal joints in patients
with rheumatoid arthritis. Clin Exp Rheumatol 1997;14:
649–52.
6. Anderson RR, Parrish P. The optics of human skin.
J Invest Dermatol 1981;77:13–9.
7. McDougall JJ, Ferrell WR, Bray RC, Wadey VMR, Frank
CB. Repetitive activity alters perfusion of proximal
interphalangeal joints of the human hand. Clin J Sports
Med 1998;8:106–10.
8. Rubin JM, Bude RO, Carson PL, Bree RL, Adler RS.
Power Doppler US: a potentially useful alternative to
mean frequency-based color Doppler US. Radiology 1994;
190:853–6.
9. Balint P, Sturrock RD. Intra and inter-observer variability
in ultrasound measurements of adult human hips: pilot
study [abstract 66 ]. Ann Rheum Dis 1999;58:27.
10. Cardinal E, Lafortune M, Burns P. Power Doppler US in
synovitis: reality or artifact? Radiology 1996;200:868–7.
11. Koski JM. Ultrasonography of the elbow joint. Rheumatol
Int 1990;10:91–4.
12. Barr LL, Babcock DS. Sonography of the normal elbow.
Am J Roentgenol 1991;157:793–8.
13. Maffulli N, Regine R, Carillo F, Capasso G, Minelli S.
Tennis elbow: an ultrasonographic study in tennis players.
Br J Sports Med 1990;24:151–5.
14. Ferrara MA, Marcelis S. Ultrasound of the elbow. J Belg
Radiol 1997;80:122–3.
15. Ferrell WR, Lockhart JC, Karimian SM. Tachykinin
regulation of basal synovial blood flow. Br J Pharmacol
1997;121:29–34.