Knee Surg, Sports Traumatol, Arthrosc

(1999) 7 : 378–381 E X P E R I M E N TA L S T U D Y
© Springer-Verlag 1999

H. Alfredson In situ microdialysis in tendon tissue:

K. Thorsen
R. Lorentzon high levels of glutamate,
but not prostaglandin E2
in chronic Achilles tendon pain

Received: 30 March 1999

Accepted: 30 May 1999
H. Alfredson (Y) · K. Thorsen ·
R. Lorentzon
Sports Medicine Unit,
Department of Surgical
and Perioperative Science,
Umeå University,
S-90187 Umeå, Sweden
e-mail: hakan.alfredson@idrott.umu.se
Tel.: +46-90-7853951
Fax: +46-90-135692
K. Thorsen · R. Lorentzon
Department of Musculoskeletal Research,
National Institute for Working Life,
S-90187 Umeå, Sweden
Abstract This investigation was to
our knowledge the first to use the
microdialysis technique to study con-
centrations of substances in a human
tendon. In four patients (mean age
40.7 years) with a painful nodule in
the Achilles tendon (chronic Achilles
tendinosis) and in five controls (mean
age 37.2 years) with normal Achilles
tendons (confirmed by ultrasonogra-
phy) the local concentrations of glu-
tamate and prostaglandin E2 were
measured under resting conditions.
A standard microdialysis catheter
was inserted into the Achilles tendon
under local anesthesia. Sampling was
performed every 15 min over a 4-h
period. The results showed signifi-
cantly higher concentrations of gluta-
mate in tendons with tendinosis than
in normal tendons (196 ± 59 vs. 48 ±
27 µmol/l, P < 0.05), and there were
no significant changes in glutamate
concentration over the period of in-
vestigation. There were no signifi-
cant differences in the mean concen-
trations of prostaglandin E2 (83 ±
22 vs. 54 ± 24 pg/ml) between ten-
dons with tendinosis and normal ten-
dons. In conclusion, in situ micro-
dialysis appears a useful method to
study certain metabolic events in ten-
don tissue. The higher concentrations
of the excitatory neurotransmitter
glutamate in Achilles tendons with a
painful nodule may possibly be in-
volved in the pain mechanism in this
chronic condition. Furthermore, there
were no signs of inflammation in the
tendons with painful nodules, as in-
dicated by the normal prostaglandin
E2 levels.
Key words Achilles tendon ·
Chronic pain · Microdialysis ·
Glutamate · Prostaglandin E2

Achilles tendinosis, in addition to the lack of inflamma-

Introduction
The Achilles tendon is the strongest tendon in the human
body, with a high capacity to withstand tensional forces
[1]. However, chronic painful conditions in the tendon,
often expressed as a localized painful swelling of the ten-
don, are relatively common especially among runners [2].
These painful conditions are still referred to as inflamma-
tory conditions [3], despite the fact that histological ex-
aminations have shown an absence of inflammatory cells
[4, 5]. We prefer to use the terms chronic tendinosis for
a localized painful swelling of the tendon with a long
duration (more than 3 months). Histologically, chronic
tory cells, is characterized by high amounts of interfibril-
lar glycosaminoglycans and changes in the collagen fiber
structure and arrangement [4]. It is a condition of un-
known cause and pathogenesis [6], most often seen among
recreational male runners in the age group 35–45 years,
and it is considered to be associated with overuse [6].
However, very little is known about the cause of chronic
pain or the pain mechanism associated with chronic
Achilles tendinosis.
In situ microdialysis has been shown to be a useful
technique to study metabolism of substances in various
types of human tissue, but the method has, to our knowl-
edge, never been used in human tendon tissue. The micro-

dialysis technique allows continuous measurements of in
vivo concentrations of substances with molecular size be-
low the cutoff limit of the dialysis membrane. We exam-
ined prostaglandin E2 (PGE2) because it is well known to
be involved in inflammatory reactions [7] and the excita-
tory neurotransmitter glutamate that recently has been
shown to be involved in central and peripheral pain regis-
tration [8]. PGs and glutamate are well suited for exami-
nation by the microdialysis technique. Here we describe,
to our knowledge, the first experiment using the micro-
dialysis technique to detect and study the local concentra-
tions of PGE2 and glutamate in patients with Achilles
tendinosis and in healthy tendons.
Methods
The study included four male patients (mean age 40.7 years, range
34–53) with a long duration of pain symptoms (> 6 months) from
Achilles tendinosis located at the 2–6 cm level in the tendon (from
the calcaneal insertion), and five controls (mean age 37.2 years,
range 27–42) recruited after advertisement at the local hospital.
The study protocol was approved by the Ethics Committee of the
University of Umeå Medical College, and the experiment was con-
ducted according to the principles of the Declaration of Helsinki.
All patients and controls were otherwise healthy and not on any
medication. In the patients, the occurrence of tendinosis and the
exact location of the tendinotic changes in the tendon was con-
firmed by ultrasonography. In the controls, ultrasonographic ex-
amination confirmed that the tendon had a normal structure.
The microdialysis was performed as a surgical procedure under
strict sterile conditions. With the patient in a supine position, the
skin over the Achilles tendon and heel was disinfected, and a local
anesthetic (2–4 ml prilocaine hydrochloride, 10 mg/ml, Södertälje,
Sweden) was injected into the skin, subcutaneous tissue, and
paratenon, close to the medial border of the lower 6 cm of the
Achilles tendon. A small skin incision was made 5–6 cm above the
Achilles tendon insertion into the calcaneus, and the paratenon was
opened. The microdialysis catheter (with a diameter of 1.4 mm and
length of the membranous covered active part of 30 mm) was in-
troduced into the tendon under visual control, placed longitudi-
nally and parallel to the tendon fibers into the central part of the of
the tendon, and fixed to the skin. In the tendons with tendinosis the
catheter was placed longitudinally and into the central part of the
palpable nodule.
The microdialysis system consists of a battery-driven infusion
pump (CMA 106; CMA/Microdialysis, Stockholm Sweden) with a
fixed infusion rate at 0.3 µl/min. The dialysis catheter (CMA 60;
CMA/Microdialysis), perfused with Ringer’s acetate, has a pore
dimension of 20 kD, and at the given perfusion rate 0.3 µl/min al-
most full recovery of PGE2 is achieved in the dialysate [9]. Also,
at the given perfusions rate and the given membranous dimension
almost full recovery of glutamate can be expected. The void vol-
ume from the probe to the sample collector is 3 µl, equivalent to
a 10-min fraction. After flushing the system, sampling was per-
formed every 15 min, and every dialysis sample (4.5 µl) was di-
luted to 25 µl and immediately frozen to –70 °C. Samples were
taken during a 4-h period with the patient resting in a supine posi-
tion. Samples from two consecutive 15-min periods were pooled.
After derivatization with OPA 40 mmol/l, glutamate was deter-
mined by HPLC (precolumn Nucleosil C18. 5 µm, 5 × 4 mm,
Knauer; column Nucleosil C18. 5 µm, 60 × 4 mm, Knauer). The
mobile phase A contained 0.1 M sodium acetate, pH 6.95 and the
mobile phase B methanol:tetrahydrofuran (97.5 : 2.5), and the flow
rate was 1.0 ml/min. Fluorescence detecting used a CMA/280 Flu-
379
Glutamate in tendon tissue
350
Tendinosis
300
Normal tendons
250
Glutamate micromol/l
200
150
100
50
0
1 2 3 4 5 6 7 8
Sample number
Fig. 1 The concentrations (µmol/l) of glutamate (mean ± SD) in
tendons with tendinosis and in normal tendons, during the 4-h
sampling period
orescence Detector (CMA Microdialysis) at excitation 330–365
nm and emission 440–530 nm.
PE2 was analyzed using a commercially available PGE2 radio-
immunoassay kit (DuPont, Boston, Mass., USA). Samples or stan-
dards, together with 125I-labeled PGE2 tracer, were incubated with
rabbit anti-PGE2 antibodies in a 0.0255 M phosphate buffer, pH 6.8,
overnight at 4°C. The samples were then precipitated by adding
polyethylene glycol, centrifuged, and decanted, and the radioactiv-
ity in the pellet was determined by using a γ-counter. The sensitiv-
ity of the assay is 10 pg/ml. The antibody used shows a 3.7%
cross-reactivity to PGE1 but one of less than 1% to dehydro-keto-
PGE2, PGA2, 6-keto-PGF1α, PGF2, and thromboxane B2.
Statistical analysis used SPSS software (Chicago, Ill., USA).
An independent-samples test (Mann-Whitney) compared at
each sampling the mean concentration of the substance in the
Achilles tendons with tendinosis with the mean concentration of
the substance in the normal tendons. Analysis of variance was car-
ried out to compare the mean concentrations of the substances in
the Achilles tendons with tendinosis and the mean concentrations
of the substances in the normal tendons at each sampling. P < 0.05
was considered significant.
Results
In the Achilles tendons with tendinosis (n = 4), the con-
centration of glutamate (196 ± 59 µmol/l) was significantly
higher than the concentration of glutamate (48 ± 27 µmol/l)
in the normal tendons (n = 5). There were no significant
differences in the mean concentrations of glutamate over
time (4 h). The results of the measurements of glutamate
are shown in Fig. 1.
There was no significant difference between the con-
centration of PGE2 in the tendons with tendinosis (83 ±
22 pg/ml) and normal tendons (54 ± 24 pg/ml. There was
no significant difference in the mean concentration of
PGE2 over time (4 h). PGE2 values are presented in Fig. 2.
380
180 PGE2 in tendon tissue
Tendinosis
Normal tendons
160
140
120
PGE2 pg/ml
100
80
60
40
20
0 opinion that there is involvement of an inflammatory re-
1 2 3 4 5 6 7 8
Sample number
Fig. 2 The concentrations (pg/ml) of PGE2 (mean ± SD) in ten-
dons with tendinosis and in normal tendons, during the 4-h sam-
pling period
Discussion
Our experiment shows that it is possible to use the micro-
dialysis technique to detect and study concentrations of
low molecular substances in the human Achilles tendon.
We found a significantly higher (approx. fourfold) con-
centration of glutamate in Achilles tendons with tendi-
nosis than in normal tendons, but there were no signs of
inflammation in the tendons with tendinosis (as indicated
by normal PGE2 levels).
The microdialysis technique has previously been dem-
onstrated at our clinic to be useful for studying bone me-
tabolism in humans by determining the production of
PGE2 in the proximal tibia metaphysis [10]. Several diffi-
culties must be anticipated in using the microdialysis
technique. It is an invasive procedure, and caution must
be taken to prevent infection. Therefore a necessary re-
quirement is that the operating room be equipped with fa-
cilities ensuring sterile conditions when introducing the
microdialysis catheter into the tendon. Furthermore, the
catheter is soft, and the membrane surrounding the
catheter is very sensitive and must be handled carefully to
avoid damage. To ensure that the catheter is not being
damaged, and to make sure that it is inserted and placed
into the tendon, the catheter must be introduced under vi-
sual control and through a minor incision in the skin and
paratenon. To prevent clotting of the catheter, the system
must be flushed without delay after insertion of the catheter
into the tendon. The handling of samples is also of utmost
importance, and the samples must be frozen immediately
to –70°C because of the rapid metabolism of PGE2.
Our group of patients with chronic Achilles tendinosis
had tendinotic changes localized at the 2–6 cm level (from
the bone insertion) in the tendon. To ensure that the
catheter was placed in the area with tendinotic changes we
inserted the 30-mm-long microdialysis catheter longitudi-
nally into the palpable nodule.
We used a local anesthetic without adrenaline to infil-
trate the skin and paratenon locally at the insertion site.
This is unlikely to have any effect on the concentrations
of glutamate and PGE2. There is a possibility that the tis-
sue trauma caused by the insertion of the catheter could
affect especially the concentrations of the neurotransmit-
ter glutamate, but examining the glutamate concentrations
in the various samples over the 4-h sampling period re-
vealed no signs indicating that the insertion procedure had
any significant impact on glutamate concentrations.
In patients with chronic painful conditions in the
Achilles tendon, it has until recently been the general
action. PGs are well known to play a central role in in-
flammatory reactions [7], and treatment often focuses on
medication with PG antagonists. Even the terminology of
the conditions, with the suffix “-itis” (e.g., tendinitis and
tendonitis), emphasize an inflammatory component. How-
ever, although biopsies show an absence of inflammatory
cells, the role of inflammation in this condition is being
questioned. The results from our investigation (normal
PGE2 levels) further emphasizes that no inflammatory com-
ponent is involved in the chronic stage of this condition.
The background and mechanisms of pain in chronic
Achilles tendinosis are unknown. In this investigation we
found that tendons with tendinosis had approximately
four times higher mean concentrations of the neurotrans-
mitter glutamate than normal tendons. In recent years the
importance of glutamate as a mediator of pain has been
emphasized [11]. It is known that the ionotrophic gluta-
mate receptors N-metyl-D-aspartate, α-amino-3-hydroxy-
5-metyl-4-isoxazolepropionic acid, and kainate are pre-
sent in unmyelinated and myelinated sensory axons [12],
and that peripherally administered glutamate antagonists
diminish the response to formalin-induced nociception
[8]. The results of our investigation clearly show that
painful chronic Achilles tendinosis is associated with high
concentrations of the excitatory neurotransmitter gluta-
mate. It must be considered whether there is a local hy-
perproduction of glutamate by some cells in the tendinosis
area or an increased number of glutamate sensory nerve
terminals. Immunohistochemical localization and quan-
tification of N-metyl-D-aspartate receptors in normal
Achilles tendons and in tissue from painful nodules are
currently going on. It might be speculated that high con-
centrations of glutamate reflects a general phenomenon
occurring in chronic painful tendon conditions, i.e., many
so-called overuse injuries.
In conclusion, the microdialysis technique can be used
to detect and study concentrations of low molecular sub-
stances in tendon tissue. The painful condition chronic
Achilles tendinosis is associated with occurrence of the
excitatory neurotransmitter glutamate, which was found
in significantly higher concentrations in Achilles tendons
with tendinosis than in normal tendons. Furthermore,
there were no signs of inflammation in tendons with
tendinosis, as indicated by normal PGE2 levels.

References
1. Komi PV, Fukashiro S, Järvinen M
(1992) Biomechanical loading of the
achilles tendon during normal locomo-
tion. Clin Sports Med 11:521–531
2. Kvist M (1994) Achilles tendon in-
juries in athletes. Sports Med 18:173–
201
3. Myerson MS, McGarvey W (1998)
Disorders of the insertion of the
Achilles tendon and Achilles tendinitis.
J Bone Joint Surg Am 80:1814–1824
4. Movin T (1998) Aspects of aetiology,
pathoanatomy and diagnostic methods
in chronic mid-portion achillodynia.
Thesis, Karolinska Institute, Stockholm
5. Åström M (1997) On the nature and
etiology of chronic Achilles tendinopa-
thy. Thesis, University of Lund, Lund
6. Józsa L, Kannus P (1997) Human ten-
dons. Human Kinetics, Leeds
7. Solomon LM, Juhlin L, Kirchenbaum
MB (1968) Prostaglandins on cuta-
neous vasculature. J Invest Dermatol
51:280–282
8. Davidson EM, Coggeshall RE, Carlton
SM (1997) Peripheral NMDA and non-
NMDA glutamate receptors contribute
to nociceptive behaviors in the rat for-
malin test. Neuroreport 8:941–946
9. Darimont C, Vassaux G, Gaillard D,
Ailhaud G, Négrel R (1994) In situ mi-
crodialysis of prostaglandins in adipose
tissue: stimulation of prostacyclin re-
lease by angiotensin II. Int J Obes Re-
lat Metab Disord 18:783–788
381
10. Thorsen K, Kristoffersson AO, Lerner
UH, Lorentzon RP (1996) In situ mi-
crodialysis in bone tissue. Stimulation
of prostaglandin E2 release by weight-
bearing mechanical loading. J Clin In-
vest 98:2446–2449
11. Dickenson AH, Chapman V, Green
GM (1997) The pharmacology of exci-
tatory and inhibitory amino acid-medi-
ated events in the transmission and
modulation of pain in the spinal cord.
A review. Gen Pharmacol 28:633–638
12. Coggeshall RE, Carlton SM (1998) Ul-
trastructural analysis of NMDA,
AMPA and kainate receptors on un-
myelinated and myelinated axons in
the perifery. J Comp Neurol 391:78–86