Knee Surg Sports Traumatol Arthrosc

(2003) 11 : 339–343 S P O RT S M E D I C I N E
DOI 10.1007/s00167-003-0402-7

Lars Öhberg Sclerosing therapy

Håkan Alfredson
in chronic Achilles tendon insertional
pain-results of a pilot study

Received: 8 December 2002

Accepted: 20 April 2003
Published online: 12 August 2003
© Springer-Verlag 2003
L. Öhberg (✉)
Department of Radiation Sciences,
Diagnostic Radiology, Umeå University,
90185 Umeå, Sweden
Tel.: +46-90-7851625,
Fax: +46-90-7852511,
e-mail: lars.ohberg.us@vll.se
H. Alfredson
Department of Surgical
and Perioperative Science,
Sports Medicine,
Department of Muskuloskeletal Research,
National Institute for Working LIfe,
Umeå University, 90187 Umeå, Sweden
Abstract The origin of Achilles
tendon insertional pain has not been
clarified. Treatment is considered
difficult, though tendon, bone, and
bursae, alone or in combination, may
all be the source of pain. Recently,
neovascularisation in the area with
tendon changes was shown to corre-
late with pain in patients with chronic
mid-portion Achilles tendinosis. In a
pilot study, sclerosing the neovessels
outside the tendon cured the pain in
the majority of patients. In this pilot-
study, ultrasonography and colour
Doppler was used for the investiga-
tion of eleven patients (nine men and
two women, mean age 44 years) with
a long duration (mean 29 months) of
chronic Achilles tendon insertional
pain. All patients had distal tendon
changes and a local neovascularisa-
tion inside and outside the distal ten-
don on the injured/painful side, but
not on the noninjured/pain-free side.
In nine patients there was also a
thickened retrocalcaneal bursae, and
in four patients also bone pathology
(calcification, spur, loose fragment)
in the insertion. The sclerosing agent
polidocanol was injected against the
local neovessels found in all patients.
At follow-up (mean eight months),
sclerosing of the area with neoves-
sels had cured the pain in eight out
of eleven patients, and in seven of
the eight patients there was no neo-
vascularisation. Pain during tendon-
loading activity, recorded on a VAS-
scale, decreased from 82 mm before
treatment to 14 mm after treatment in
the successfully treated patients. In
conclusion, treatment only focusing
on sclerosing the area with neovessels
showed promising short-term clinical
results in this small pilot study. The
findings support further studies,
preferably in a randomised manner.
Keywords Achilles tendon ·
Insertional pain · Sclerosing therapy

pain [3]. A prominent posterior angle of the calcaneal bone,

Introduction
Painful conditions in the Achilles tendon insertion, often
called insertion tendinopathy, are well-known to be diffi-
cult to treat [4]. In the acute phase conservative treatment
is recommended, but with persisting (chronic) pain sur-
gery is often required [3, 4, 5]. The origin of insertional
pain has not been clarified, but is considered to be multi-
factorial [4]. The calcaneal bone, the superficial and retro-
calcaneal bursae and the Achilles tendon, separately or in
combination, have all been suggested to be the origin of
Haglund’s deformity, has been suggested to cause a painful
“impingement” on the tendon with ankle dorsiflexion [5, 6].
In the mid-portion of the Achilles tendon, recent stud-
ies have indicated that the occurrence of a neovascularisa-
tion in association with the area with tendon changes (tendi-
nosis) was associated with pain symptoms from the ten-
don. First, an investigation using ultra-sonography and colour
Doppler demonstrated the occurrence of a neovascularisa-
tion in 28 consecutive tendons with chronic painful tendi-
nosis, but not in any of 20 consecutive normal and pain-
free tendons [8]. Secondly, in a recent study, infiltrating
340

the neovessels with a local anaesthetic temporarily took
away the pain, and destroying the same neovessels by in-
jecting a sclerosing agent cured (six months follow-up)
the pain in most patients [9].
The aim with this study was to try the hypothesis that
destroying (sclerosing) the vessels and most likely also
nerves accompanying the vessels, but not addressing any
treatment to the tendon, bursae, or bone in itself, would
affect chronic Achilles tendon insertional pain. For scle-
rosing, we chose polidocanol that is well-known and widely
used as a sclerosing agent. Polidocanol has been used as a
sclerosing agent in the treatment for varicose veins in the
legs and oesophagus, haemorrhoids and telangiectasias,
and also to treat gastroduodenal lesions, with very few
side-effects reported [1, 2, 7].
Material and methods
Material
The investigation was approved by the ethical Committee at the
medical Faculty of the University of Umeå.
Eleven patients, nine males and two females, participated in the
study. The mean age was 44 years (range 24–67), mean height was
176 cm (range 162–185), and mean weight 78 kg (range 63–105).
All patients had a long duration of symptoms (mean 29 months,
range 7–96).
All patients had pain during Achilles tendon-loading activity,
and seven patients had shoe-problems. Clinically, all patients had
a widened and thick distal Achilles tendon that was painful during
palpation, and in 4 patients there were a painful diffuse bony
prominence in the Achilles tendon insertion. There was slight pain
during palpation of the retrocalcaneal bursa in seven patients.
All patients had been treated with nonsteroidal anti-inflamma-
tory drugs (NSAIDs), and four patients had had altogether 12 cor-
ticosteroid injections. Nine patients had tried eccentric calf-muscle
training without any effect.
The majority of the patients were recreational athletes (4 jog-
gers, 1 floor-ball, 1 beach volleyball), whilst one patient was an elite
long distance runner, and one a second division soccer player.
Three patients had walking as their activity level.
Methods
Ultrasonography and colour Doppler
The examinations were performed with high resolution grey scale-
ultrasound with the aid of colour Doppler, Acuson Segoia. A linear
multifrequency (8–13 MHz) probe was used. Pathological changes
in the painful thickened Achilles tendon and the bursae, and the
occurrence of bone pathology (calcification, spur, fragment), were
registered. Colour Doppler was used to diagnose the neovasculari-
sation. The contralateral non-painful tendon was also examined.
All examinations were done by the same experienced radiologist
(L.Ö). The test-to-test reliability for the observer (L.Ö.) was eval-
uated by examining one tendon ten times during a short period of
time, with repositioning between investigations. The coefficient of
variation (CV) was estimated to be 1.1%.
Sclerosing therapy
Polidocanol (5 mg/ml) was used as a sclerosing agent. The active
substance is an aliphatic non-ionised nitrogen-free surface anaes-
thetic, with molecule weight approximately 600. Before the treat-
ment the skin was washed with a solution of chlorhexidine and al-
cohol. The skin was draped with a sterile paper-cover with a hole
only for the distal part of the Achilles tendon. The injection was
performed with a 0.7×50 mm needle connected to a 2 ml syringe.
Since polidocanol also is a local anaesthetic there was no need for
anaesthesia prior to the sclerosing therapy.
Evaluation
Neovascularisation in the Achilles tendon insertion was estimated
(0), (1+), (2++), (3+++), (4++++) according to the appearance of
vessels inside the tendon, in close relation to the bursae wall, and
close to bone (loose fragment, spur, calcification). When there
were no visible vessels the estimation was (0). When there were
one or two small vessels the estimation was (1+). When there were
several irregular vessels the estimation was (2+ to 4+).

Table 1 Sclerosing therapy in 11 patients with chronic Achilles VAS-score in millimetres (mm) represents the patients’ evaluation
tendon insertional pain. Age, gender and pathologically changed of the amount of pain during Achilles tendon-loading activity.
type of tissue. The occurrence of neovascularisation registered Number of injections, and patient satisfaction with treatment (Yes
with colour Doppler ultrasound (estimated as 0, 1+, 2+, 3+, 4+). or No)
Gender/age Pathology Neovascularisation VAS No Satis-
injection fied
Before After Before After
Female/46 Tendon, bone, bursa 3+++ 0 100 50 4 No
Male/53 Tendon, bone, bursa 4++++ 1+ 99 25 4 Yes
Male/46 Tendon, bone, bursa 4++++ 2++ 90 49 5 No
Female/54 Tendon, bone, bursa 2++ 0 74 8 4 Yes
Male/32 Tendon, bursa 2++ 0 99 11 1 Yes
Male/37 Tendon, bursa 3+++ 0 84 3 2 Yes
Male/23 Tendon, bursa 2++ 0 77 29 1 Yes
Male/68 Tendon, bursa 4++++ 0 64 4 3 Yes
Male/32 Tendon, bursa 2++ 2++ 70 74 1 No
Male/47a Tendon 3+++ 0 91 40 4 Yes
Male/43a Tendon 2++ 0 67 20 1 Yes
aPrevious surgical treatment (bursectomy + bone resection)

The clinical results were evaluated with the use of a visual ana-
logue scale (VAS) for pain, where the patient recorded the amount
of pain during their activity (walking, jogging, running etc.) on a
100 mm-long pain scale. The amount of pain was recorded from
0–100 mm, where no pain is recorded as 0 and severe pain as 100.
Patient satisfaction with treatment was also recorded. Results were
evaluated before and after each treatment.
Results
Before treatment, ultrasonography showed a widening of
the distal Achilles tendon in all patients. In the widened
area there were structural tendon changes demonstrated as
irregular fibre structure and hypo-echoic areas. In all pa-
tients, colour Doppler showed small, irregular vessels, in-
side and outside the ventral part of the structurally changed
distal tendon. The vessels were entering the tendon from
the fat pad anterior to the Achilles tendon. In seven cases,
there were vessels in close relation to the wall of the retro-
Fig. 1 The sclerosing procedure. Note the position of the probe
covered with a sterile rubber cover and the position of the injection
needle on the medial side of the Achilles tendon
Fig. 2 Neovascularisation in
chronic painful Achilles tendon
insertional pain. Longitudinal
ultrasound scan. Thickened
and irregular distal tendon.
Colour Doppler is presented in
gray scale and the neovessels
are the coloured structures
341
calcaneal bursa. In three cases, there were a bony spur and
calcifications, and in 1 case there was a loose bone frag-
ment, in the tendon insertion. At the first investigation, also
the contralateral (normal) tendon was investigated, and
found to be thin with a regular fibre structure and without
any hypo-echoic areas and vessels inside. Furthermore,
there were no bony spurs, calcifications, or bone frag-
ments, a normal bursae, and no neovascularisation in close
relation to these structures.
The mean VAS-score (pain during activity) before treat-
ment was 84.5 mm (range 64–100), and the neovasculari-
sation was estimated to be 2+ in five tendons, and 3+ or
4+ in 6 tendons (Table 1).
Polidocanol was injected against the vessels entering the
Achilles tendon from the ventral side of the tendon, and
against vessels in close relation to the bursae wall and bone
pathology (spurs, calcifications, fragment). The injection
was performed dynamically, with the aid of linear high-
resolution ultrasound, in order to inject polidocanol into
or close to the vessels. The ultrasound probe was held on
the dorsal side of the Achilles tendon, parallel with the fi-
bres. The injection was always done from the medial side
of the tendon to minimise the risk of contact with the sural
nerve (Fig. 1). It was necessary to use colour Doppler to
identify these small vessels (Fig. 2).
It was possible to place the tip of the needle into or
close to the vessels. When the tip of the needle was in the
correct position a small amount of polidocanol (2–4 ml)
was injected under continuous monitoring with real time
ultrasound. It was possible to observe the immediate ef-
fect of the injection (Fig. 3). If the position of the needle
was correct (inside or very close to the vessels), the circu-
lation in the target vessels stopped momentarily. The in-
jections against the vessels continued until the circulation
had stopped in all the vessels. All patients were pain-free
immediately after the injection (local anaesthetic effect),
but the symptoms returned after a few hours.
The patients were allowed free activity such as walk-
ing, bicycling, cross-country skiing and jogging the day
after treatment. They were allowed to go back to strenu-
ous tendon loading activities one week after injection.
342
Fig. 3 The same tendon as in
Fig. 2 immediately after injec-
tion of polidocanol. There is
no remaining circulation in the
vessels inside the Achilles ten-
don
The patients were controlled clinically and by ultrasound
after three to six weeks. If the patient still had Achilles
tendon insertional pain, and if there was a remaining neo-
vascularisation, the treatment was repeated in the same
manner. In the successful cases, one to four (mean 2) treat-
ments (injections) were given, until the patients had no
pain from the Achilles tendon insertion, and there were no
recognisable neovessels. At follow-up (mean eight months),
the mean VAS-score after treatment (last injection) was
14 mm in the successfully treated patients (range 3–40),
and 58 mm in the three poor cases (range 49–74). Of the
eight successfully treated cases,the neovascularisation
was reduced to 0 in seven cases, and 1+ in one case. There
was no obvious change in thickness or structure of the
Achilles tendon during the short-term follow-up.
Eight of the eleven patients were satisfied with the re-
sults of treatment, and three were not satisfied. These three
patients were those who had a high VAS-score. One pa-
tient (tendon and bursa pathology) had remaining neovas-
cularisation (2+) in close relation to the tendon and retro-
calcaneal bursa, and one patient (tendon pathology) had
remaining neovascularisation (2+) in close relation to the
tendon, after treatment. In the third patient (tendon, bur-
sae and bone pathology), there was no remaining neovas-
cularisation but a loose bone fragment.
There were shoe problems in seven patients before treat-
ment. After treatment, the three patients with a poor result
had remaining shoe problems, whilst the other patients had
no remaining shoe problems.
There were no side effects of the treatment.
Discussion
In this pilot study on patients with chronic Achilles ten-
don insertional pain we demonstrated promising short-
term clinical results, with reduced pain level and possibil-
ity to take part in previous (before injury) Achilles tendon-
loading activities, by addressing the treatment (sclerosing
therapy) to the area with neovascularisation in close rela-
tion to pathology in the tendon, retrocalcaneal bursa, and
bone. No treatment was given to the tendon, bursa, and
bone, in itself.
There are many theories around the origin of Achilles
tendon insertional pain, such as the bursae, the tendon,
and the calcaneus (impingement, spurs, calcifications, and
loose fragments). It is well known that the retrocalcaneal
bursae might be focus for a chronic inflammation [3, 4],
and consequently could be a source for nociceptive pain.
Also, an impingement between a spur or prominence of
the upper posterior calcaneus, and a thickened or normal
tendon, might mediate nociceptive pain from the bone
and/or tendon [6]. Furthermore, a partially or totally de-
tached bone fragment, and maybe also calcifications, may
cause pain. These different issues alone, or together, might
all be responsible for the painful condition. Therefore, in
clinical practice it is often difficult to judge where to ad-
dress the treatment. Importantly, the scientific evidence
for these theories is missing, and there might be other sources
for the pain.
In recent studies, of patients with chronic pain from mid-
portion Achilles tendinosis, we have demonstrated that
the pain was associated with a neovascularisation outside
(on the ventral side) and inside the affected part of the ten-
don [8], and that the pain was cured in most patients if the
neovessels were destroyed by injections of a sclerosing
agent [9]. The neovessels did not re-appear, at least not
during the six months follow-up. The conclusions from
those studies were that the neovessels and accompanying
nerves most likely were associated with the pain-symp-
toms during tendon-loading activities.
Based on the conclusions from our previous studies,
we decided to try the hypothesis that a neovascularisation
in the Achilles tendon insertion might be associated with
pain in patients with chronic Achilles tendon insertional
pain. All patients selected were severe cases. They had had
a long duration of pain-symptoms, had tried conservative
treatment (NSAID, rest, strength training) without effect
 343

on the pain, and were on the waiting list for surgical treat-
ment. Because pain during Achilles tendon-loading activ-
ity was the main complaint from the patients, we decided
to have that as the primary factor when evaluating the re-
sults of treatment.
Sclerosing the neovessels outside the pathologically
changed Achilles tendons, close to the bursae walls and
bone (spurs, calcifications, fragment) markedly improved
the symptoms of eight out of eleven tendons, and allowed
these patients to return to pain-free tendon loading activ-
ity. Absence of neovessels after treatment correlated well
with reduced pain from the Achilles tendon insertion in
seven out of eight patients. In two patients with tendon,
bone and bursae pathology, there was remaining pain in
the tendon insertion after treatment, and these two pa-
tients had remaining neovessels. In one patient with re-
maining pain after treatment, there were no remaining neo-
vessels, but a loose bone fragment. Of course, it is un-
likely that this therapy would help patients with consider-
able bone pathology, such as spurs or loose bone frag-
ments, but studies on larger groups of patients are needed
to be able to draw any conclusions on what type of distal
pathology this therapy is best suited.
We do not know the exact background to the good re-
sults on tendon pain with this type of treatment. During
the study we observed that the patients became pain-free
and could do pain-free heel raises on one leg immediately
after the treatment. However, the pain returned within six
hours in all patients. This is most likely caused by the lo-
cal anaesthetic effect of polidocanol. Thereafter, the pain
gradually disappeared, in three patients after one injection
and in six patients after two to four injections. Very prob-
ably there is an effect of polidocanol not only on the ves-
sels but also on nerves accompanying the vessels, either
directly by destruction or indirectly by ischaemia. Polido-
canol has a selective effect in the vascular intima causing
thrombosis of the vessel. The agent also has an effect if
the injection is performed extravascular, which is impor-
tant when very small vessels are sclerosed.
It might be speculated that this type of treatment will
destroy parts of the normal circulation in the tendon, and
possibly could cause ischaemia and the risk of tendon
necrosis? Therefore, it’s important to note that the injec-
tions of polidocanol are given very locally, outside the ten-
don, and in the lowest concentration (5 mg/ml) available.
Compared to the tissue trauma during surgical treatment
(excision of tendinosis tissue or multiple longitudinal in-
cisions), the trauma caused by local sclerosing therapy is
minimal. Thus far, after altogether around 150 injections,
we have had no indications that this treatment is harmful
for the tendon.
In conclusion, in this small pilot- study ultrasound and
colour Doppler-guided sclerosing of neovessels in patients
with chronic painful Achilles tendon insertional pain sig-
nificantly reduced the pain level, and allowed the majority
of patients to go back to tendon loading activities. Further
studies on larger groups of patients, preferably in a ran-
domized pattern, are needed to further evaluate the effects
of this type of treatment.

References
1. Conrad P, Malouf GM, Stacey MC
(1995) The Australian polidocanol
(aethoxysklerol) study. Results at 2 years.
Dermatol Surg 21:334–336
2. Guex JJ (1993) Indications for the scle-
rosing agent polidocanol. J Dermatol
Surg Oncol 19:959–961
3. Järvinen M, Józsa L, Kannus P, Järvinen
TLN, Kvist M, Leadbetter W (1997)
Histopathological findings in chronic
tendon disorders. Scand J Med Sci
Sports 7:86–95
4. Kvist M (1991) Achilles tendon injuries
in athletes. Ann Chir Gynaecol 80:188–
201
5. Kvist M (1994) Achilles tendon injuries
in athletes. Sports Med 18:173–201
6. Vega MR, Cavolo DJ, Green RM, Co-
hen RS (1984) Haglund’s deformity.
J Am Podiatr Med Assoc 74:129–135
7. Winter H, Drager E, Sterry W (2000)
Sclerotherapy for treatment of hemangi-
omas. Dermatol Surg 26:105–108
8. Öhberg L, Lorentzon R, Alfredson H
(2001) Neovascularisation in Achilles
tendons with painful tendinosis but not
in normal tendons: an ultrasonographic
investigation. Knee Surg Sports Trau-
matol Arthrosc 9:233–238
9. Öhberg L, Alfredson H (2002) Ultra-
sound guided sclerosis of neovessels in
painful chronic Achilles tendinosis: pi-
lot study of a new treatment. Br J Sports
Med 36:173–177
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.