doi:10.1093/brain/awn169
Brain (2008), 131, 2387^2400
Residual spinothalamic tract pathways predict
development of central pain after spinal cord injury
Gunnar Wasner,1,2,3 Bonsan Bonne Lee,2,4 Stella Engel2,4 and Elspeth McLachlan1,2
1
Prince of Wales Medical Research Institute, 2University of New South Wales, Australia, 3Department of Neurology,
Division of Neurological Pain Research and Therapy, University-Clinic of Schleswig-Holstein, Campus Kiel, Germany and
4
Prince of Wales Hospital, Randwick, Sydney, NSW, Australia
Correspondence to: Priv.-Doz. Dr Med. Gunnar Wasner, Department of Neurology, Division of Neurological Pain Research
and Therapy, University Hospital of Schleswig-Holstein, Campus Kiel, Schittenhelmstrasse 10, 24105 Kiel, Germany
E-mail: [email protected]
Central neuropathic pain following lesions within the CNS, such as spinal cord injury, is one of the most excruciating types of chronic pain and one of the most difficult to treat. The role of spinothalamic pathways in this
type of pain is not clear. Previous studies suggested that spinothalamic tract lesions are necessary but not sufficient for development of central pain, since deficits of spinothalamic function were equally severe in spinal cord
injured people with and without pain.The aim of the present study was to examine spinothalamic tract function
by quantitative sensory testing before and after activation and sensitization of small diameter afferents by
applying menthol, histamine or capsaicin to the distal skin areas where spontaneous pain was localized.
Investigations were performed in matched groups each of 12 patients with and without central pain below the
level of a clinically complete spinal cord injury, and in 12 able-bodied controls.To test peripheral C fibre function,
axon reflex vasodilations induced by histamine and capsaicin applications were quantified. In eight patients with
pain, sensations of the same quality as one of their major individual pain sensations were rekindled by heat
stimuli in combination with topical capsaicin (n = 7) or by cold stimuli (n = 1). No sensations were evoked in
pain-free patients (P50.01). Capsaicin-induced axon reflex vasodilations were significantly larger in pain patients
with heat- and capsaicin-evoked sensations in comparison to pain patients without capsaicin-provoked sensations. These results suggest that intact thermosensitive nociceptive afferents within lesioned spinothalamic
tract pathways distinguish people with central pain from those without.The ability to mimic chronic pain sensations by activation of thermosensory nociceptive neurons implies that ongoing activity in these residual spinothalamic pathways plays a crucial role in maintaining central pain. We propose that processes associated
with degeneration of neighbouring axons within the tract, such as inflammation, may trigger spontaneous activity in residual intact neurons that act as a ‘central pain generator’ after spinal cord injury.
Keywords: spinal cord injury; central neuropathic pain; spinothalamic tract; sensitization; nociceptive afferents
Abbreviations: NRS = numeric rating scales; SCI = spinal cord injury; ZPP = zone of partial preservation
Received April 25, 2008. Revised June 19, 2008. Accepted July 2, 2008. Advance Access publication July 31, 2008
Introduction
Central pain that develops after lesions within the CNS,
such as stroke, spinal cord injury (SCI) and multiple sclerosis, is one of the most excruciating and therapeutically
refractory types of neuropathic pain (Finnerup et al., 2005;
Siddall and Middleton, 2006). One reason for the lack of
success in treating these types of pain is that the mechanisms underlying central pain are not fully understood.
A particularly puzzling question has been the role of the
spinothalamic tract in central pain syndromes (Finnerup
and Jensen, 2004). Multiple studies in central pain patients
performed over the last two decades have demonstrated
deficits of temperature and pain sensation within the painful area indicating that lesions of the spinothalamic projections are essential for development of pain (Beric et al., 1988;
Boivie et al., 1989; Parrent et al., 1992; Vestergaard et al.,
1995; Eide et al., 1996; Bowsher et al., 1998; Defrin et al., 2001;
Finnerup et al., 2003; Finnerup and Jensen, 2004; Österberg
et al., 2005; Ducreux et al., 2006; Finnerup et al., 2007a).
However, as differences in altered sensitivity to thermal
and nociceptive stimuli were not found in patients with
pain compared to patients without pain after central lesions,
ß The Author (2008). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: [email protected]
2388
Brain (2008), 131, 2387^2400
it has been suggested that such lesions alone are not sufficient to generate central pain (Defrin et al., 2001; Finnerup
et al., 2003; Finnerup and Jensen, 2004; Ducreux et al.,
2006; Siddall and Middleton, 2006; Finnerup et al., 2007a).
Thus, despite clear evidence for their relevance, the exact
impact of lesions of the spinothalamic tract remains unclear.
About 30–40% of spinal cord injured people suffer from
neuropathic pain due to lesions of nervous structures after
SCI (Störmer et al., 1997; Siddall et al., 1999, 2003). In
the taxonomy of the International Association for the Study
of Pain (IASP), at-level pain, i.e. neuropathic pain related to
the level of the injury, is distinguished from below-level pain
localized distal from the site of injury (Siddall et al., 2000,
2002, 2003; Vierck et al., 2000). While at-level pain may be
driven by both peripheral and central components, belowlevel pain is considered to be a central pain syndrome directly
caused by the lesion to the spinal cord (Siddall et al., 2000;
Finnerup and Jensen, 2004; Wasner et al., 2007). In terms
of the underlying pathophysiology, it has been proposed
that, after SCI, abnormal activity may arise as a result of
a ‘spinal pain generator’ formed by neuronal hyperexcitability at the lesion site due to release of the excitatory amino acid
glutamate, up-regulation of sodium channels, activation of
glial cells during inflammatory processes triggered by the
lesion and loss of intraspinal and descending inhibitory pathways (Vierck et al., 2000; Finnerup and Jensen,
2004; Yezierski, 2005; Waxman and Hains, 2006). This is
in accord with a recent patient study indicating that neuronal hyperexcitability in the spinal cord is a key element in
central neuropathic pain, whereas loss of spinothalamic
function does not appear to be a pain predictor (Finnerup
et al., 2007b). As preserved neuronal function is essential
for development of hyperexcitability, the question arises as
to whether residual intact spinothalamic tract neurons can
be identified in subjects with central pain following SCI.
In the present study, spinothalamic tract function was
tested quantitatively in people with and without below-level
pain following clinically complete SCI by quantitative
sensory testing before and after activation and sensitization
of small diameter afferents by applying menthol, histamine
or capsaicin to the distal skin areas where spontaneous pain
was localized. The responses to these three chemicals can be
used to activate specific spinothalamic tract pathways. We
reasoned that stimulation after sensitization of small fibre
peripheral afferents including nociceptors and thermal
afferents in a below-level area with severe sensory deficits
might generate sensations that were missed by conventional
quantitative sensory testing (Wasner et al., 2007). This
would imply the existence of persisting connections
between the periphery and the central pathways.
Materials and Methods
Subjects
The function of the spinothalamic tract was tested in three groups
each of 12 age- and sex-matched subjects: (i) SCI people with
G. Wasner et al.
central neuropathic pain below the spinal lesion, (ii) SCI people
without any pain below the lesion and (iii) able-bodied pain-free
control subjects (Tables 1–3).
SCI subjects were included according to the following criteria.
To ensure a central lesion affecting the spinal cord above the
conus medullaris, we required (i) that the traumatic SCI was not
lower than the 10th thoracic vertebra and (ii) at least one clinical
sign at or proximal to the neurological level of the test area
indicated a central lesion (e.g. brisk Achilles jerks when testing
the lower leg). Patients had to be classified as having a complete
spinal cord lesion according to the American Spinal Injury
Association’s standards for classification of SCI (ASIA, grade A,
see later). In cases in which patients had a zone of partial
preservation (ZPP, see later), this needed to be well localized
outside the test area.
Neuropathic pain was defined as chronic pain in an area of
sensory abnormalities developed after SCI. The definition of
below-level pain was that the painful area was at least two
dermatome segments below the lesion level to ensure that the pain
was localized below the neurological level (Finnerup et al., 2003)
(Table 1). The pain needed not to have any primary relation to
spasm or any other movement.
SCI pain-free control subjects should not have had any
sensation and particularly not pain (including any spasm- or
movement-related pain) below the neurological level since the
time of the injury (Table 2).
Able-bodied control subjects had to be free of any chronic or
acute pain, any medication and any neurological disorder.
Subjects with a mental disease that potentially interfered with
their ability to be tested psychophysically, those with evidence of
any other disease affecting the central or the peripheral nervous
system, those with evidence of allergies to menthol, histamine or
capsaicin, those younger than 18 years old and those who were
pregnant were excluded from the study.
All experiments were performed in the supine position in a
room held at a temperature of 22–23 C and a relative humidity
of 50–60%.
The study was approved by the local human research ethics
committees of the University of New South Wales and the South
Eastern Sydney and Illawarra Area Health Service. All participants
gave informed written consent to participate in the study.
Medical history and pain characteristics
Medical histories were obtained (Tables 1–3). Patients localized
their neurological level and pain on a body chart in two
dimensions (anterior and posterior) (Finnerup et al., 2003).
SCI patients with below-level pain were asked to characterize the
quality of their below-level pain by a maximum of three descriptors chosen from a modified short-form McGill Pain
Questionnaire or by using their own words when the pain
quality was not represented by any of the given descriptors. The
Questionnaire contained the following items: throbbing, shooting,
stabbing, sharp, cramping, gnawing, burning, hot, cold, aching,
heavy, tender, dull, touch-evoked, tingling, wrenching (Wasner
et al., 2002). Furthermore, the patients rated their average pain
intensity and the average intensity of any temperature component accompanying the pain on numeric rating scales (pain
rating: NRS 0–10, 0 = no pain, 10 = maximum conceivable
pain; temperature rating: NRS 10 to +10, 10 = maximum
Central pain in spinal cord injury
Brain (2008), 131, 2387^2400
2389
conceivable sensation of cold, 0 = neutral temperature sensation,
+10 = maximum conceivable sensation of heat).
Clinical examination
A complete neurological and physical examination was performed.
Spinal lesions were classified according to the American Spinal
Injury Association’s (ASIA) standards (Maynard et al., 1997).
In addition to the required testing for touch and pinprick to
estimate sensory function according to the ASIA standard,
vibration was tested with a graded tuning fork (64 Hz, 8/8 scale)
that was placed over a bony prominence close to the test area
in the lower extremity and left there until the subject could no
longer feel vibration. Vibration detection threshold was determined as a disappearance threshold (Whitton et al., 2005).
Spasticity was assessed by using the spasm frequency scale
[ranging from 0 (no spasm) to 4 (410 spasm per day)] (Snow
et al., 1990) and the modified Ashworth Scale [ranging from 0
(no increase in tone) to 4 (two people required to abduct the hips
to 45 )] (Bohannon and Smith, 1987).
Quantitative sensory testing
Quantitative mechano- and thermosensory testing was conducted
in all subjects. Initially, all sensory tests to be used were
demonstrated in an area of unaffected skin (e.g. the arm in SCI
people with thoracic lesions and healthy controls or the neck in
people with cervical lesions) to make the subjects familiar with the
applied stimuli. After these trials were complete, measurements
were first made in the face of all subjects. The face served as an
unaffected control area well above the neurological level primary
affected by the SCI. Next, these tests were performed before and
after chemical sensitization of small fibre afferents in the area of
below-level pain in the subjects with pain and topographically
corresponding areas in pain-free SCI people and healthy controls.
To distinguish any stimulus-evoked sensation in the area below
the neurological level from ongoing sensations or any imagined
sensation, the stimulus was applied when the subject had their
eyes closed and the site of stimulation identified afterwards.
In cases of severe sensory loss, failure to respond to cut-offlimits of the test apparatus resulted in assignment of the cut-off
limits as the threshold value.
Figure 1 gives an overview over the experimental protocol.
Quantitative mechano-sensory testing
Mechanical detection and pain thresholds to punctuate stimuli
were measured using a set of 12 von Frey hairs with a bending
force ranging from 0.25 to 512 mN (Optihair2-Set, Marstock
Nervtest, Germany). Using the method of limits, two determinations for detection and pain thresholds were made, each with a
series of ascending and descending stimulus intensities. Thresholds
were indicated verbally by the subjects. Stimulus-evoked pain was
rated on a numerical rating scale (NRS 0–10, see above). The final
thresholds and pain ratings at threshold were the geometric mean
of these two series (Rolke et al., 2006).
Mechanical dynamic allodynia was investigated by gently
moving a cotton swab over the skin three times at a frequency
of 0.2 Hz. The subjects rated the pain intensity by giving one
average NRS value (0–10) for these three stimuli.
When punctuate hyperalgesia was indicated by a decreased pain
threshold to punctuate stimuli or by the presence of mechanical
Fig. 1 Summary of experimental procedures.
dynamic allodynia, the corresponding areas of mechanically
evoked pain were determined by marking the border between
areas of non-painful and painful stimulation (Wasner et al., 2004).
Quantitative thermotesting
Quantitative thermotesting was performed with a thermotest
device (TSA 2001-II, Medoc, Israel) to investigate perception
thresholds for warmth, cold, heat pain and cold pain (Fruhstorfer
et al., 1976; Yarnitsky and Ochoa, 1991). According to the
standards of the German Research Network on Neuropathic Pain,
the method of limits was used by applying ramp stimuli at a
velocity of 1 C/s starting from 32 C applied to the skin via a
Peltier type thermode (3 3 cm2) (Rolke et al., 2006). The subjects
were asked to press a button when the respective thermal
sensation was perceived. In case of inability to press the button
due to SCI-related paresis, sensations were reported verbally to
the investigator who pressed the button immediately. Cut-off
temperatures were 0 and 50 C.
Thermally induced sensations
During investigations of thermal pain thresholds, the subjects were
instructed to rate the perceived pain and temperature sensations at
the thresholds on numeric rating scales (pain rating: NRS 0–10;
temperature rating: NRS 10 to +10, see above). The mean
ratings for three consecutive measurements each of temperature
and pain during the testing of thermal pain thresholds were
calculated (Wasner and Brock, 2008).
Furthermore, pain patients were asked to characterize the
quality of thermally induced pain by choosing descriptors from
the modified short-form McGill Pain Questionnaire or by using
their own words when the thermally induced pain quality was
not represented by any of the given descriptors (see above).
2390
Brain (2008), 131, 2387^2400
If thermal stimulation induced no pain within the cut-off limits,
but some other sensation, pain patients were asked to indicate
the temperature at which these sensations were induced and to
report their quality. To ensure their validity, the thermally
induced sensations needed to be reproduced in three consecutive
measurements.
Chemical activation and sensitization of small
fibre afferents
Small fibre afferents were chemically activated and sensitized by
L-menthol, histamine and capsaicin, in sequential applications. All
drugs were topically applied within the skin area of below-level
pain and corresponding sites in pain-free SCI patients and healthy
controls. In all pain patients, the area of below-level pain was large
enough to allow a distance of at least 15 cm between each site of
drug application. The time between each drug application was
about 30–45 min including 10–15 min time for the patient to rest
before the next session.
Menthol
Topical menthol activates and sensitises peripheral cold C fibre
nociceptors and non-noxious cold-specific A delta fibre afferents
via TRPM8, a menthol- and cold-sensitive ion channel of the
transient-receptor potential family (Wasner et al., 2004).
A 0.7 ml aliquot of a solution containing 400 mg/ml of
L-menthol (40% w/v) dissolved in 90% ethanol was placed on a
3 3 cm2 gauze pad, which was applied to the skin for 15 min. To
protect the ethanol from evaporation, the gauze pad was covered
with adhesive film and fixed by a 2.5 cm wide rubber band to
ensure adequate contact between the solution and the skin
(Wasner et al., 2004). Before and after topical menthol, mechanosensory testing and thermotesting (cold detection and cold pain
thresholds, as described earlier) were performed in the same area
as menthol application (Fig. 1).
Histamine
Cutaneous application of histamine induces an itch sensation due
to activation of histamine-sensitive C-fibre afferents synapsing
with a distinct neural pathway within the spinothalamic tract that
mediates itch-sensations (Schmelz et al., 1997; Andrew and Craig,
2001; Schmelz, 2001).
Histamine was ionophorized intracutaneously (histamine hydrochloride 1%, charge 4.8 mC: 80 mA for 60 s) via a silver disc electrode
of 0.5 cm diameter. Skin temperature, itch sensation and blood flow
were evaluated before and after topical histamine (Fig. 1).
G. Wasner et al.
pain thresholds, as described earlier) were performed in the
same area as capsaicin application, except for setting baseline
temperature during thermotesting at 35 C to equal skin temperature (Fig. 1).
Drug-induced sensations
During application of menthol and capsaicin, the subjects were
instructed to rate the perceived temperature and pain sensations on numeric rating scales every minute (pain rating: NRS
0–10; temperature rating: NRS 10 to +10; see earlier) (Wasner
et al., 2004).
Intensity of itch sensation due to ionophoresis of histamine
was also quantified on a NRS (0–10) for 10 min by asking the
subjects every minute for rating. If any of the drugs induced a
sensation other than pain, temperature or itch, respectively, the
subjects were asked to report the quality of this sensation (Fig. 1).
Cutaneous vascular perfusion
Skin temperature was measured at sites of drug application below
the neurological level with an infra-red thermometer (Linear
Instruments, USA). During histamine and capsaicin application,
cutaneous vascular perfusion was measured non-invasively by
laser Doppler flowmetry, yielding relative blood flow changes in
arbitrary perfusion units (Moor Instruments, Axminster, Devon,
UK). The laser Doppler probe was placed 1 cm distal to the electrode
for histamine ionophoresis, and also 1 cm from the edge of the
capsaicin gauze pad, for continuous measurement of skin blood
flow outside the application area. Axon reflex vasodilation (relative
increase in blood flow) was calculated as follows: (maximal flow
after drug applicationbaseline flow/baseline flow 100) (Baron
and Saguer, 1993). At the end of each session, the visible area
of histamine- or capsaicin-evoked flare was drawn on sheets of
plastic film and measured using a digital planimeter (Fig. 1).
Data acquisition and analysis
Because most of the measured parameters were not normally
distributed (Rolke et al., 2006), data are presented as median
(50th percentile) and interquartile range (25–75th percentiles),
the U-test was applied for unpaired data analyses and the
Wilcoxon-test for paired data analyses. Normally distributed
data are presented as mean SE and a parametric procedure
was used (Student’s t-test, see Table 3). The Chi-square test was
used to test for differences in the proportions of the responses
to the applied tests between the patient groups.
Statistical significance was assigned at the P50.05 level by
two-tailed analysis.
Capsaicin
Capsaicin strongly activates and sensitizes C nociceptors via TRPV1,
a heat sensitive ion channel of the transient-receptor potential
family (LaMotte et al., 1992; Baron, 2000; Caterina et al., 2000).
A 300 ml aliquot of a solution containing 0.02 M capsaicin
(0.6%) dissolved in 45% ethanol was placed on a 3 3 cm2 gauze
pad, which was applied to the skin for 15 min. Because effects
of capsaicin are substantially temperature-dependent, local skin
temperature was clamped at 35 C with a feedback-controlled
heat lamp (Baron et al., 1999; Wasner et al., 2000, 2007). To
protect the ethanol from evaporation, the gauze pad was covered
with adhesive film. Before and after topical capsaicin, mechanosensory testing and thermotesting (warm detection and heat
Results
Subject characteristics
In all patients, testing for touch, pinprick, vibration sense
and motor function revealed the absence of sensory and
motor function in the area of interest below the level of
injury. The absence of sensory and motor function in
the lowest sacral segment indicated a complete spinal cord
lesion according to the American Spinal Injury Association’s
standards for classification of SCI (ASIA, grade A)
in all patients (Maynard et al., 1997) (Tables 1 and 2).
Central pain in spinal cord injury
Brain (2008), 131, 2387^2400
Three patients in each group had a ZPP referring to
dermatomes and myotomes caudal to the neurological
level that remained partially innervated. In all of these
cases, this zone was well localized outside the test area
(Tables 1 and 2).
Traumatic SCI was the underlying cause of the lesion in
all but one patient who had suffered a spinal infarction
(Tables 1 and 2). There were no differences in time since
injury, distribution of neurological level and grading of
spasticity between the two patient groups (Table 3).
Five patients with below-level pain suffered also from
neuropathic pain at the neurological level (Table 1).
2391
However, in all pain patients, below-level pain was the
dominating symptom and was well discriminated both in
quality and site of origin from their at-level pain. Three SCI
patients in the pain-free group had a history of occasional
at-level pain of moderate intensity that was not present
during the testing period (Table 2).
All pain patients had a history of taking oral medication
against neuropathic pain and six patients still received
treatment. However, their medication did not provide
adequate pain relief in any of the patients (Table 1).
As confirmed by MRI, two patients in each group
developed posttraumatic syringomyelia, which was not
Table 1. Clinical characteristics of spinal cord injured people with central pain below the level of injury
No Age/Sex Neurological ASIA
level
1
2
22/M
33/M
T4
T4
3
25/M
T1
4
5
52/M
46/M
T5
T5
6
34/M
C7
7
40/M
8
9
Time since
Mechanism of injury
injury (months)
A
7
A, ZPP 95
(T7)
A, ZPP 28
(T4)
A
348
A
322
Water skiing accident
Bike accident
Area of
Test area
below-level pain
At-level Pain
pain
medication
Syrinx
Upper legs, feet Feet
Legs
Left lower leg
^
Yes
^
^
^
C6 ^T5
Yes
^
^
^
Yes
^
^
C7^T2
^
^
Gabapentin,
amitriptyline
Gabapentin,
amitriptyline
Gabapentin
Gabapentin,
oxycodone,
amitriptyline
Pregabalin
^
^
42
T10
A, ZPP
(T4)
A
Motor vehicle accident Lower abdomen, Left lower
right knee
abdomen
Motor vehicle accident Legs
Left upper leg
Motor vehicle accident Legs
Left lower leg
and foot
Diving accident
Left lower leg
Left lower leg
44
Motor vehicle accident Legs
61/F
29/M
T10
C7
A
A
34
53
Fall from height
Legs
Motor vehicle accident Feet
10
11
61/M
57/M
T10
T7
A
A
48
384
12
29/M
T4
A
34
Motor vehicle accident Upper legs
Motor vehicle accident Feet and left
lower leg
Motor vehicle accident Feet
Right lower leg
and both feet
Right lower leg
Left lower leg
and foot
^
Yes
^
Right upper leg
Feet
Left foot
^
^
Yes
Pregabalin,
amitriptyline
^
^
^
^
^
^
Loss of sensory and motor function in the lowest sacral segment indicated a complete spinal cord lesion according to the ASIA, grade
A in all patients (Maynard et al., 1997). Three patients had a ZPP referring to dermatomes and myotomes caudal to the neurological level
(in brackets) that remains partially innervated.
Table 2 Clinical characteristics of spinal cord injured people without pain below the level of injury
No.
Age/Sex
Neurological level
ASIA
Time since injury (months)
Mechanism of injury
At-level pain
Syrinx
1
2
3
4
5
6
7
8
9
10
11
12
39/M
22/M
45/M
61/M
35/M
38/M
36/M
60/M
30/M
67/M
44/M
46/M
T5
T4
T4
T4
C6
T5
T9
C4
T7
T10
T7
T8
A
A
A
A
A
A, ZPP (T8)
A
A, ZPP (C7)
A, ZPP (L2)
A
A
A
72
12
39
450
20
18
226
502
50
93
327
215
Motor vehicle accident
Motor vehicle accident
Motor vehicle accident
Motor vehicle accident
Diving accident
Spinal infarction
Motor vehicle accident
Motor vehicle accident
Motor vehicle accident
Fall from height
Motor vehicle accident
Motor vehicle accident
Yes
^
Yes
^
Yes
^
^
^
^
^
^
^
^
^
T2^T3
^
^
^
T6 ^T9
^
^
^
^
^
All patients were classified ASIA grade A as described in the text. Three patients had a ZPP referring to dermatomes and myotomes caudal
to the neurological level (in brackets) that remains partially innervated in complete patients.
2392
Brain (2008), 131, 2387^2400
G. Wasner et al.
accompanied by any change in the symptoms below the
neurological level (Tables 1 and 2).
Sensation in control areas above
the neurological level
None of the subjects experienced any spontaneous ongoing
sensation in the facial area which served as the control area
above the neurological level of the lesion. Quantitative
sensory testing in the face revealed no significant differences
in perception of mechanical or thermal stimuli and pain
thresholds between SCI patients with and without pain and
healthy controls (Fig. 2).
Neuropathic pain below the neurological level
All SCI pain patients suffered from ongoing pain with a
median pain intensity scored as NRS 5 (interquartile range
NRS 3–6.8) (Table 4). Burning was the predominant
pain descriptor chosen by 9 out of 12 patients; other pain
characteristics included stinging, dull, aching, heavy, shooting
and sharp. Interestingly, only three patients described the
pain as hot-burning with a median temperature intensity
of NRS 5 (interquartile range NRS 4–8) whereas, in eight
patients, the pain was not accompanied by any temperature
component irrespective of whether the pain was characterized
as burning. One patient described pain as an ongoing cold
aching sharp sensation with an intensity of NRS 5 and a
temperature component of NRS –5 (Table 4, no. 8).
None of the pain patients suffered from any kind of
evoked pain below the neurological level. None of the SCI
control patients or the healthy control subjects suffered
from any ongoing sensation or any evoked pain below the
neurological level.
Sensations induced by activation and
sensitization of small fibre afferents
below the neurological level
SCI people
Mechanical stimulation with von Frey hairs or by moving a
cotton swab within the area of below-level pain and within
the corresponding pain-free area in SCI control patients did
not induce any sensation in the people with SCI in accord
with the clinical diagnosis of complete SCI. Thermal stimuli
between 0 C and 50 C did not induce any sensation in
either patient group except for two SCI pain patients. In
patient no. 3, an average heating temperature of 48.7 C
of three consecutive stimuli led to a sensation of shooting
pain with an intensity of NRS 2 (Table 4). The same
kind of shooting pain (underlined in Table 4) with an
intensity of NRS 3 was one of the pain descriptors chosen
by this patient to characterize his neuropathic pain.
In patient no. 8, skin cooling to a temperature of 26.1 C
led to an intense cold sensation rated with NRS –6. As
described earlier, coldness (underlined in Table 4) was
one of the major pain characteristics in this patient.
Therefore, in both these patients, thermal stimulation
rekindled characteristics typical of their individual neuropathic pain sensations.
Chemical activation and sensitization of small fibre
afferents by topical application of menthol, histamine or
capsaicin did not induce any sensation in either of the two
SCI patient groups. However, thermal heat (but not
mechanical or cold stimulation) after sensitization with
capsaicin evoked sensations in six additional SCI pain
patients rekindling typical individual neuropathic pain
characteristics (Table 4). In four of these patients
(Table 4, nos. 1, 2, 4 and 5), heat stimuli of certain
intensities induced painful sensations sharing at least one
quality of their neuropathic pain sensations (underlined in
Table 4). In two patients (nos. 6 and 7), who described
their neuropathic pain as hot, thermal stimulation after
capsaicin rekindled this temperature component by inducing non-noxious warm sensations within the painful area.
In four pain patients (Table 4, nos. 9–12), as well as
in all pain-free control SCI people, neither thermal nor
mechanical stimuli evoked any sensation after sensitization
of small fibre afferents. The pain characteristics of these
four SCI pain patients (burning, hot-burning, sharp,
stinging, aching and tingling) did not differ from those of
the SCI people with thermally evoked pain sensations.
Table 3 Group data on clinical characteristics of the three study groups of each 12 subjects
Age
Time since injury (months)
Male/female
Neurological level cervical/thoracic
Spasm frequency scale
Modified Ashworth Scale
SCI with pain
SCI without pain
Controls
P-value
40.8 14.2
120 142
11/1
2/10
2.8 1.3
1 0.6
43.6 13.4
169 175
12/0
2/10
2.7 1.5
1 0.8
37.7 12.8
^
9/3
^
0
0
NS*#z
NS*
NS*#z
NS*
NS*, 50.001#z
NS*, 50.01#z
Spasticity was assessed using the spasm frequency scale [ranging from 0 (no spasm) to 4 (410 spasm per day); Snow et al., 1990] and the
modified Ashworth Scale [ranging from 0 (no increase in tone) to 4 (two people required to abduct the hips to 45 ); Bohannon and Smith,
1987].
SCI with pain versus SCI without pain.
#
SCI with pain versus controls.
zSCI without pain versus controls; data are presented as mean SEM.
Central pain in spinal cord injury
Brain (2008), 131, 2387^2400
2393
sensation present in only 3/12 subjects (median NRS 0;
interquartile range: NRS 0–0.4) and significant cold
allodynia (Fig. 3A). Punctate hyperalgesia was found in
only 6/12 subjects measuring 9.6 cm2 (median; interquartile
range 6.4–19.9). Dynamic mechanical allodynia was not
present in any subject after menthol.
Ionophoresis of histamine induced an ongoing itch
sensation in all subjects of moderate intensity (NRS 0–10:
median NRS 2; interquartile range NRS 1.1–4.8).
Topical application of capsaicin induced an intense
ongoing hot burning sensation [NRS 6 (median; interquartile range 4–8) for temperature and NRS 7.3 (median;
interquartile range 6–8.8) for pain] as well as punctuate and
heat hyperalgesia (Fig. 3B). Areas of punctuate hyperalgesia
in 10/12 subjects measured 27.3 cm2 (median; interquartile
range 21.7–37.8). Dynamic mechanical allodynia was
present in 10/12 subjects and measured 18.4 cm2 (median;
interquartile range 15.2–23.2).
In summary, results on quantitative sensory testing in
combination with drug-induced activation and sensitization
of small fibre afferents in intact healthy controls were in
accord with previously published data (Baron and Saguer,
1993; Wasner et al., 2000, 2004; Rolke et al., 2006).
Cutaneous vascular perfusion
Fig. 2 Quantitative sensory testing in the face. Group data for
thermal detection thresholds (A), cold pain (B) and heat pain
thresholds (C), as well as mechanical detection and pain
thresholds (D) in SCI patients with central pain (filled squares),
SCI patients without pain (filled diamonds) and able-bodied
controls (filled circles). Data include pain and temperature ratings
at threshold (NRS = numeric rating scale). Pain rating: NRS 0 ^10,
0 = no pain, 10 = maximum conceivable pain; temperature rating:
NRS 10 to +10, 10 = maximum conceivable sensation of cold,
0 = neutral temperature sensation, +10 = maximum conceivable
sensation of heat). Data are presented as median (50th percentile)
and interquartile range (25 ^75th percentiles). There are no
significant differences between the groups.
In summary, 8 out of 12 SCI people with neuropathic
below-level pain responded to thermal stimulation in the
painful area either before or after chemical sensitization of
small fibre nociceptive afferents, whereas there were no
responders in the control group of pain-free SCI people
(P50.001; 2 test).
Control subjects
In healthy controls, thresholds for mechanical and thermal
stimuli prior to sensitization of small fibre afferents were all
within normal limits as described in other studies on
quantitative sensory testing (Fig. 3) (Rolke et al., 2006).
Topical application of menthol induced an ongoing cold
sensation rated with an intensity of NRS –3 (median;
interquartile range: NRS –0.4 to –4), a moderate pain
Skin temperature and blood flow prior to histamine
ionophoresis was not significantly different between SCI
patients and control subjects (Fig. 4A and B). However,
both blood-flow increase and flare size induced by
histamine were significantly larger in healthy controls
compared to the two patient groups (Fig. 4C and D).
There were no differences between SCI patients with and
without below-level pain. Histamine-induced axon reflex
vasodilation did not differ between SCI pain patients in
whom neuropathic pain sensations were rekindled by heat
stimuli prior to or after capsaicin-sensitization (Table 4,
nos. 1–7) and pain patients who did not respond to heat
(Table 4, nos. 8–12) (Fig. 4C and D).
During capsaicin application, skin temperature was held
at 35 C. Baseline blood flow was not significantly different
between the three subject groups (Fig. 5A). There was a
trend towards larger increases in blood flow and flare size
after capsaicin in healthy controls as compared to the two
patient groups but this did not reach statistical significance
(Fig. 5B and C, P = 0.18). However, within the group of
patients with central pain, the increases in blood flow of
people who developed pain in response to heat stimuli
before or after capsaicin (Table 4, nos. 1–7) were
significantly larger than those in pain patients who did
not respond to heat (Table 4, nos. 8–12, P50.05) and
reached values similar to those in healthy controls (Fig. 5B).
A similar trend was present in the size of the capsaicininduced flare (Fig. 5C; P = 0.11). Blood flow and flare size
of SCI pain patients who did not respond to capsaicin
resembled those in pain-free patients (Fig. 5B and C).
2394
Brain (2008), 131, 2387^2400
G. Wasner et al.
Table 4 Characteristics of below level pain and thermally induced sensations in each spinal cord injured patient with
central pain
No
1
2
3
4
5
6
7
8
9
10
11
12
Characteristics of below level pain
Thermally induced sensations
Pain descriptors
Pain (NRS)
Temperature
(NRS)
Threshold
Evoked sensations
Pain (NRS)
Temperature
(NRS)
Burning, stinging
Burning, sharp, pins and needles
Dull, shooting, squeezing
Dull, heavy, tight
Burning, unpleasant tingling
Hot-burning, tingling
Hot-burning, aching
Cold, aching, sharp
Hot-burning, sharp, cramping
Burning, stinging, tingling
Burning, aching, tingling
Burning, dull
5
5
3
8
3
8
6
5
7
6
3
2
0
0
0
0
0
5
4
5
8
0
0
0
48.1 (Capsaicin)
39.0 (Capsaicin)
48.7
40.2 (Capsaicin)
49.5 (Capsaicin)
47.8 (Capsaicin)
40.5 (Capsaicin)
26.1
^
^
^
^
Stinging
Burning
Shooting
Heavy, tight
Unpleasant tingling
Warm
Warm
Cold
^
^
^
^
4.7
4
2
2
1
0
0
0
^
^
^
^
0
0
0
0
0
3
2
6
^
^
^
^
Patients characterized quality of their below-level pain and of thermally induced sensations by descriptors chosen from a modified shortform Mc Gill Pain Questionnaire or by using their own words in case the pain quality was not represented by any of the given descriptors
(Wasner et al., 2002). Average pain intensity and the average intensity of any temperature component accompanying the pain was rated on
numeric rating scales (pain rating: NRS 0 ^10, 0 = no pain; 10 = maximum pain; temperature rating: NRS 10 to +10; 10 = maximum cold;
0 = neutral temperature sensation; +10 = maximum heat). Thermal thresholds at which evoked sensations were perceived are given. In six
subjects heat stimulation evoked sensation only after capsaicin (in brackets), whereas in two subjects (no. 3 and 8) thermal stimulation
alone was effective. Note that thermally evoked sensations rekindled typical characteristics of the patients’ individual neuropathic pain
sensations (underlined descriptors). In four pain patients (no. 9^12) no sensation was induced.
Discussion
The present study has revealed persisting spinothalamic
tract function in the majority of patients with central pain
following clinically complete SCI. Thermal stimulation after
topical activation and sensitization of peripheral small fibre
(mainly nociceptive) afferents in the area to which pain was
referred evoked sensations in 8/12 cases. The thermally
evoked sensations mimicked the characteristics of each
patient’s individual pain implying that these residual
connections were involved in pain generation. No sensations were induced in matched pain-free patients given
the same protocol. We conclude that partially preserved
spinothalamic tract pathways can be the site of the ‘central
pain generator’ after spinal cord injuries.
Several previous studies using quantitative sensory testing
including thermal stimulation in patients with central pain
have demonstrated deficits in spinothalamic tract function
indicating that lesions within the ascending nociceptive
pathways are essential for the development of central pain
(Defrin et al., 2001; Finnerup et al., 2003, 2007; Finnerup
and Jensen, 2004; Ducreux et al., 2006; Siddall and
Middleton, 2006). However, as these studies failed to
show differences in disturbed spinothalamic tract function
between patients with and without pain, the exact
involvement of the spinothalamic tract in mediating central
pain remained unclear. Our approach has enabled residual
function of heat- and cold-sensitive small fibre afferents to
be demonstrated. Thermal stimulation prior to and after
sensitization of different classes of small fibre afferents by
menthol, capsaicin and histamine was the pivotal methodological approach that enabled us to identify the
differences between patients with and without central pain.
‘Discomplete’ sensory lesions of spinothalamic
tract in central pain patients
In our experiments, sensations were evoked in two-thirds of
the SCI patients with pain, but in none of the patients
without pain. Two pain patients responded to thermal
stimuli even before peripheral sensitization, which is in
accord with the previous data showing that sensations can
be evoked below the lesion in a minority of people
classified as having complete SCI (Finnerup and Jensen,
2004). However, in the other six responders, sensations
could be elicited by heat stimuli but required sensitization
of nociceptive afferents with capsaicin. In these cases, the
stimuli were extremely strong and not in the range
tolerated by the able-bodied control subjects.
It is unlikely that our findings are biased by the
psychophysical nature of the methods, because (i) the
patients identified the exact site of the stimulus without
seeing it being applied and (ii) thresholds were reproducible
during repetitive stimulation. In addition, the psychophysical approach cannot account for the differences between
patients with and without pain. Furthermore, quantitative
sensory testing in the face revealed no differences between
SCI patients with and without central pain and healthy
controls indicating not only that small fibre afferents in this
control area well above the lesion had normal function,
Central pain in spinal cord injury
Fig. 3 Quantitative sensory testing in non-painful skin in ablebodied controls before and after menthol and capsaicin. Group
data for mechanical and thermal thresholds and perception ratings
in able-bodied controls in distal skin areas equivalent to the pain
areas below the lesion of SCI people prior to (filled circles) and
after (open circles) applying menthol (A) and capsaicin (B). (A)
Menthol raised significantly the threshold temperature for cold
pain indicating cold hyperalgesia without effects on mechanical
pain threshold or pain and temperature perceptions. (B) Capsaicin
induced mechanical and heat hyperalgesia by significantly decreasing mechanical and heat pain thresholds. Mechanical hyperalgesia
was accompanied by an increased pain perception that was not
seen for heat hyperalgesia. Perception ratings (NRS) defined as in
Fig. 2. Data are presented as median and interquartile range;
P50.05, P50.01.
but also that SCI people are as competent in dealing with
this methodology as able-bodied controls.
Preserved sensations following peripheral stimulation
of thermosensory afferents can only be explained by a
‘discomplete’ lesion within the spinothalamic tract that
left residual connections between the skin site where central
pain was perceived and the brain. As such discomplete lesions
were not detected in the group of comparable SCI patients
without below-level pain, these partially preserved pathways
could be involved in the development of central pain.
The concept of discomplete SCI was first introduced
to describe clinically complete lesions accompanied by
neurophysiological evidence of residual brain influences
Brain (2008), 131, 2387^2400
2395
Fig. 4 Responses to histamine. Group data for histamine-induced
axon reflex vasodilation in the distal area below the level of
lesion of all three subject groups [SCI patients with central pain
(filled squares), SCI patients without pain (filled diamonds) and
able-bodied controls (filled circles)]. There were no significant
differences between skin temperatures (A) and baseline blood
flow measured by laser Doppler flowmetry (B) before histamineionophoresis. Histamine induced a significantly larger vasodilation
(C) and flare (D) in able-bodied controls than in the two patient
groups. There were no significant differences between pain
patients in which heat stimulation after capsaicin evoked sensations
(open squares) and pain patients without this response (grey filled
squares) (A^D). Data are presented as median and interquartile
range; P50.05, P50.01.
on spinal motor function below the lesion (Dimitrijevic,
1988; Sherwood et al., 1992). Later, the term
‘sensory discomplete’ was used to describe a minority of
people classified by clinical criteria as having complete
SCI but in whom sensations could be evoked from below
the lesion (Finnerup and Jensen, 2004). The present
study demonstrates that many subjects with pain arising
below the SCI can be considered to have similarly
discomplete lesions of their spinothalamic pathways.
Which spinothalamic tract neurons are
preserved in central pain patients?
Several mechano- and thermosensitive afferents can be
activated by quantitative sensory testing in conjunction
2396
Brain (2008), 131, 2387^2400
Fig. 5 Responses to capsaicin. Group data for capsaicin-induced
axon reflex vasodilation in the distal area below the level of lesion
of all three subject groups. Skin temperature was held at 35 C.
There were no significant differences between SCI patients with
central pain (filled squares), SCI patients without pain (filled
diamonds) and able-bodied controls (filled circles) in baseline
blood flow (A) or capsaicin-induced vasodilation (B) and flare (C).
However, capsaicin-induced vasodilation was significantly larger
in pain patients in which heat stimulation after capsaicin evoked
sensations (open squares) compared to pain patients without this
response (grey filled squares) (B). Data are presented as median
and interquartile range; P50.05.
with topical drug application as data from healthy controls
demonstrate. Menthol and capsaicin induced ongoing pain,
thermal and mechanical hyperalgesia due to sensitization of
cold and heat-specific nociceptive afferents (LaMotte et al.,
1992; Baron, 2000; Wasner et al., 2004; Namer et al., 2005).
Histamine induced a prolonged itch sensation due to
activation of histamine-sensitive afferents (Schmelz et al.,
1997; Andrew and Craig, 2001; Schmelz, 2001).
The fact that the majority of our spinothalamic
discomplete central pain patients responded to heat after
capsaicin, whereas none of the other potent thermal or
mechanical stimuli evoked in them any sensation, implies
that only a subset of spinothalamic tract pathways
was preserved. These intact fibres must have been capsaicinas well as heat-sensitive, because stimulation with capsaicin
alone was not potent enough to induce any sensation, as
was demonstrated in a previous study (Finnerup et al.,
2007a). Capsaicin acts on two subpopulations of C fibre
nociceptors via TRPV1, the transient-receptor potential
cation channel V1 (Szolcsanyi, 2004). Activation and sensitization of mechano-insensitive nociceptors encode longlasting capsaicin-induced burning pain, heat hyperalgesia
and flare responses, whereas polymodal nociceptors are
activated only briefly by capsaicin and do not appear to
be sensitized (Schmelz et al., 2000; Schmidt et al., 2000).
As mechano-insensitive nociceptors respond not only
G. Wasner et al.
to capsaicin, but also to heat, their activation may also
account for the sensations induced by heat stimuli prior to
capsaicin application in one SCI pain patient. Therefore, we
suggest that a subset of spinothalamic tract neurons which
receive inputs from mechano-insensitive C fibre nociceptors
was partially preserved in these central pain patients.
This hypothesis is supported by data presented here on
drug-elicited axon reflexes. The axon reflex vasodilation is
due to release of vasoactive peptides (substance P and
mainly calcitonin gene-related peptide, CGRP) from the
endings of small fibre afferents (Szolcsanyi, 2004).
Capsaicin and histamine evoked reduced axon reflex
vasodilation in both patient groups compared to healthy
controls as previously demonstrated (Finnerup et al.,
2007a). It was suggested (Kuesgen et al., 2002) that
this may be due to increased vasoconstriction following
injury-related changes in sympathetic efferent activity
counteracting axon reflex vasodilation. However, as skin
temperature and baseline blood flow were not different
between the groups, no direct evidence for such a mechanism has been found in the present study. Notably,
however, capsaicin induced a significantly larger blood-flow
increase in the SCI pain patients who responded to heat
before or after capsaicin than in pain patients who did not
respond (Fig. 5B). This difference was specific for capsaicinsensitive axons because there was no difference between the
histamine-induced axon reflexes which involves different
axons. As it is suggested that dorsal root reflexes via spinal
pathways including the dorsal horn are involved in axon
reflex vasodilation (Lin et al., 1999; Weng and Dougherty,
2005), these data suggest that the pathways from mechanoinsensitive C nociceptors were more preserved in these
patients than in non-responders. CGRP-containing afferent
axons have been reported to sprout in the dorsal horn
of many segments above and below both transection
(Krenz and Weaver, 1998) and severe compression of rat
spinal cord (Weaver et al., 2001) and have been proposed
to play a role in development of central pain (Christensen
and Hulsebosch, 1997). Whether such sprouting axons are
CGRP-containing mechano-insensitive nociceptive afferents
is not clear.
Apart from the involvement of the dorsal horn in
axon reflexes, the lack of drug-induced vasodilation in the
non-responders and in the patients without central pain
implies that central lesions can affect the function or even
anatomical integrity of primary peripheral nociceptors that
are not directly damaged. One can speculate that uninjured
nociceptive afferents below the injury lose their expression
of vasodilator peptides, their excitability or even that they
degenerate. There might be retrograde transneuronal
degeneration after damage to the secondary spinothalamic
afferents, as occurs in the visual system where undamaged
retinal ganglion cells degenerate after striate cortical lesions
in primates (Cowey, 1974; Johnson and Cowey, 2000).
One of the eight SCI responders with central pain
reacted to cooling, but not to heating (Table 4, no. 8),
Central pain in spinal cord injury
indicating that the preserved spinothalamic pathways in
this patient were connected not with mechano-insensitive
C fibre nociceptors but with cold-specific afferents (Wasner
et al., 2004).
Are intact spinothalamic tract neurons
the site of the ‘spinal pain generator’?
In all eight central pain patients that responded to
peripheral thermal stimuli, the evoked pains had characteristics typical of their individual spontaneous pain sensations
(Table 4), i.e. their neuropathic pain was rekindled. This
indicates that central pain is at least partly driven by the
pathological activity of preserved spinothalamic afferents
and suggests that they act as ‘spinal pain generators’.
The idea that activity in uninjured fibres contributes to
neuropathic pain comes from animal studies demonstrating
ongoing discharge of uninjured C fibres in a peripheral
nerve trunk containing degenerating axons (Wu et al., 2001;
Djouhri et al., 2006). It is thought that inflammatory
mediators such as cytokines (e.g. tumour necrosis factoralpha) and neurotrophic factors (nerve growth factor)
released from non-neuronal cells around neighbouring
degenerating axons are the key players for eliciting
pathological activity in intact afferents (Wu et al., 2001;
Schafers et al., 2003; Djouhri et al., 2006; Scholz and Woolf,
2007). Similar inflammatory changes in the injured spinal
cord are not restricted to the injury site and the
surrounding region of secondary degeneration but extend
along the spinal cord rostral and caudal to the lesion, with
activation of microglia (Popovich et al., 1997; McKay et al.,
2007; Donnelly and Popovich, 2008) persisting for over a
year, as post-mortem studies in humans have demonstrated
(Schmitt et al., 2000). Microglial activation in both the
dorsal horn and thalamus after experimental SCI is
associated with pain-related behaviour as well as increased
hyperexcitability and upregulation of Nav1.3 channels in
nociceptive neurons (Hains and Waxman, 2007). Remote
activation of microglia with increases in pro-inflammatory
cytokines has been shown to predict the onset and severity
of below-level neuropathic pain after incomplete SCI in
rats (Detloff et al., 2008). However, it is not clear whether
such inflammatory processes are capable of maintaining
chronic pain for years, in particular as animals have rarely
been studied for longer than 1 month after injury. It is
possible that inflammatory processes initiate a cascade of
functional or structural changes which trigger chronic
ectopic activity in nociceptive pathways.
Where is the ‘pain generator’ located?
The present data indicate that central pain can be induced
by activity in intact spinothalamic afferents located in the
dorsal horn where inflammatory processes are prominent
(Hains and Waxman, 2006; McKay et al., 2007; Detloff et
al., 2008) and ascending spinothalamic tract neurons are
located (Saab et al., 2008). Degeneration of some
Brain (2008), 131, 2387^2400
2397
spinothalamic tract neurons with axons injured at the
level of the lesion (Fig. 6, encircled 1) could be responsible
for microglial activation around the surviving spinothalamic neurons triggering excitation. In addition, spinal
lesions will interrupt descending inhibitory pathways arising
from the brainstem (Fig. 6, encircled 2) that play a crucial
role in modulating the excitability of dorsal horn neurons
(Meller and Gebhart, 1993). Hyperexcitability of these
neurons located far below the level of the injury may then
be the basis of the ‘spinal pain generator’ (Fig. 6).
It is also possible that spontaneous activity in residual
spinothalamic pathways originates from ascending spinothalamic tract axons within the white matter particularly
close to the site of the lesion (Fig. 6, encircled 3).
Inflammation in the white matter is characterized by
persistent CD68+ macrophages involved in the prolonged
Fig. 6 Proposed role of residual spinothalamic tract neurons
in below-level pain following SCI. SCI (grey bar) lesioned the
majority of spinothalamic tract axons (dashed think line) leading
to retrograde reactions (possibly degeneration) of the corresponding dorsal horn neurons (shaded circle). Residual spinothalamic
afferents, which were not affected by the lesion (solid thin line),
developed ongoing activity contributing to central neuropathic
pain. Pathological activity may arise from intact dorsal horn
neurons (filled circle) due to inflammatory processes associated
with degeneration of neighbouring axon terminals and neurons
in the dorsal horn (1), as a result of disinhibition due to lesion of
descending pain inhibitory pathways (dotted line) following SCI (2),
from ascending spinothalamic tract axons within the damaged
white matter close to the lesion (3). Changes of excitability and
discharge patterns in thalamic neurons may be triggered by
ongoing activity in these intact spinothalamic afferents (4).
2398
Brain (2008), 131, 2387^2400
phagocytosis of myelin (Schmitt et al., 2000; McKay et al.,
2007).
Alternative central pain mechanisms
There is evidence that other mechanisms can also lead
to central pain (Finnerup and Jensen, 2004). In the present study, four of the pain patients did not respond
to peripheral stimulation. Clinical characteristics, pain
descriptors and pain intensities did not distinguish between responders and non-responders in our group of pain
patients (Table 4). It may have been that the chemical
stimulation was not intense enough to demonstrate any
residual spinothalamic connections. Alternatively, their pain
was generated by a different mechanism. Furthermore,
in the group of pain patients that responded to thermal
stimuli, not all of each individual’s neuropathic pain
sensations were rekindled (Table 4). Therefore, it is possible that other mechanisms in higher-order neurons are
also involved.
A source of pain signals suggested by animal as well as
human research is hyperexcitability leading to ectopic
activity of thalamic neurons following SCI (Lenz et al.,
1994; Hains and Waxman, 2007). Changes in sodium
channel expression related to microglial activation following
deafferentation by degeneration of the terminals of ascending spinothalamic tract neurons are proposed to be the
underlying cause of these thalamic changes (Gao et al.,
2004; Hains et al., 2005; Hains and Waxman, 2007; Zhao
et al., 2007). The results of the present study suggest that
spontaneous activity arising from residual intact spinothalamic afferents, rather than from degeneration of their
terminals in the thalamus, may be the trigger for secondary
changes of excitability and discharge patterns in thalamic
neurons (Fig. 6, encircled 4). This idea parallels the notion
of hyperexcitability in secondary nociceptive dorsal horn
neurons generated following spontaneous activity in
primary afferents in peripheral neuropathic pain
(Campbell and Meyer, 2006).
Conclusions and implications
The present study provides evidence for the first time
that intact spinothalamic tract afferents projecting through
a damaged region of spinal cord can contribute to central
pain following SCI. It is suggested that spontaneous
activity is generated in these intact neurons by excitatory
substances released from microglia activated during the
chronic inflammation triggered by retrograde degeneration
of lesioned neighbouring neurons. Therefore, we suggest
that both degeneration and preservation are important for the development of central pain, because degeneration is essential for inducing the inflammatory processes
that generate neuronal hyperexcitability and activity in
preserved neurons.
G. Wasner et al.
While other mechanisms can also be responsible for
central pain, residual intact pathways may also contribute
to central pain syndromes other than those that follow SCI.
As well as other strategies, targeted treatment options such
as anti-inflammatory drugs may be successful to abolish the
generation of spontaneous activity. Data from animal
experiments have already demonstrated that this might be
a promising approach (Hains and Waxman, 2006).
Acknowledgements
We thank Danielle Burton for her excellent assistance and
recruiting of the patients. We thank also Associate Professor
Phil Siddall and Dr Paul Wrigley for their cooperation and
support in patient recruitment. The work was supported by
the Alexander von Humboldt-Foundation, the Spinal
Injuries Research Centre (Prince of Wales Medical
Research Institute), the National Health and Medical
Research Council of Australia, the International
Association for the Study of Pain (IASP collaborative
research grant) and the German Ministry of Research and
Education within the German Research Network on
Neuropathic Pain (BMBF, 01EM 05/04).
References
Andrew D, Craig AD. Spinothalamic lamina I neurons selectively sensitive
to histamine: a central neural pathway for itch. Nat Neurosci 2001; 4:
72–7.
Baron R. Capsaicin and nociception: from basic mechanisms to novel
drugs. Lancet 2000; 356: 785–7.
Baron R, Saguer M. Postherpetic neuralgia. Are C-nociceptors involved
in signalling and maintenance of tactile allodynia? Brain 1993; 116:
1477–96.
Baron R, Wasner G, Borgstedt R, Hastedt E, Schulte H, Binder A, et al.
Effect of sympathetic activity on capsaicin-evoked pain, hyperalgesia,
and vasodilatation. Neurology 1999; 52: 923–32.
Beric A, Dimitrijevic MR, Lindblom U. Central dysesthesia syndrome in
spinal cord injury patients. Pain 1988; 34: 109–16.
Bohannon RW, Smith MB. Interrater reliability of a modified Ashworth
scale of muscle spasticity. Phys Ther 1987; 67: 206–7.
Boivie J, Leijon G, Johansson I. Central post-stroke pain–a study of the
mechanisms through analyses of the sensory abnormalities. Pain 1989;
37: 173–85.
Bowsher D, Leijon G, Thuomas KA. Central poststroke pain: correlation
of MRI with clinical pain characteristics and sensory abnormalities.
Neurology 1998; 51: 1352–8.
Campbell JN, Meyer RA. Mechanisms of neuropathic pain. Neuron 2006;
52: 77–92.
Caterina MJ, Leffler A, Malmberg AB, Martin WJ, Trafton J,
Petersen-Zeitz KR, et al. Impaired nociception and pain sensation
in mice lacking the capsaicin receptor. Science 2000; 288: 306–13.
Christensen MD, Hulsebosch CE. Spinal cord injury and anti-NGF
treatment results in changes in CGRP density and distribution in the
dorsal horn in the rat. Exp Neurol 1997; 147: 463–75.
Cowey A. Atrophy of retinal ganglion cells after removal of striate cortex
in a rhesus monkey. Perception 1974; 3: 257–60.
Defrin R, Ohry A, Blumen N, Urca G. Characterization of chronic pain
and somatosensory function in spinal cord injury subjects. Pain 2001;
89: 253–63.
Detloff MR, Fisher LC, McGaughy V, Longbrake EE, Popovich PG,
Basso DM. Remote activation of microglia and pro-inflammatory
Central pain in spinal cord injury
cytokines predict the onset and severity of below-level neuropathic pain
after spinal cord injury in rats. Exp Neurol 2008 [Epub ahead of print].
Dimitrijevic MR. Residual motor functions in spinal cord injury. Adv
Neurol 1988; 47: 138–55.
Djouhri L, Koutsikou S, Fang X, McMullan S, Lawson SN. Spontaneous
pain, both neuropathic and inflammatory, is related to frequency of
spontaneous firing in intact C-fiber nociceptors. J Neurosci 2006; 26:
1281–92.
Donnelly DJ, Popovich PG. Inflammation and its role in neuroprotection,
axonal regeneration and functional recovery after spinal cord injury.
Exp Neurol 2008; 209: 378–88.
Ducreux D, Attal N, Parker F, Bouhassira D. Mechanisms of central
neuropathic pain: a combined psychophysical and fMRI study in
syringomyelia. Brain 2006; 129: 963–76.
Eide PK, Jorum E, Stenehjem AE. Somatosensory findings in patients with
spinal cord injury and central dysaesthesia pain. J Neurol Neurosurg
Psychiatry 1996; 60: 411–5.
Finnerup NB, Jensen TS. Spinal cord injury pain–mechanisms and
treatment. Eur J Neurol 2004; 11: 73–82.
Finnerup NB, Johannesen IL, Fuglsang-Frederiksen A, Bach FW,
Jensen TS. Sensory function in spinal cord injury patients with and
without central pain. Brain 2003; 126: 57–70.
Finnerup NB, Otto M, McQuay HJ, Jensen TS, Sindrup SH. Algorithm for
neuropathic pain treatment: an evidence based proposal. Pain 2005; 118:
289–305.
Finnerup NB, Pedersen LH, Terkelsen AJ, Johannesen IL, Jensen TS.
Reaction to topical capsaicin in spinal cord injury patients with and
without central pain. Exp Neurol 2007a; 205: 190–200.
Finnerup NB, Sorensen L, Biering-Sorensen F, Johannesen IL, Jensen TS.
Segmental hypersensitivity and spinothalamic function in spinal cord
injury pain. Exp Neurol 2007b; 207: 139–49.
Fruhstorfer H, Lindblom U, Schmidt WC. Method for quantitative
estimation of thermal thresholds in patients. J Neurol Neurosurg
Psychiatry 1976; 39: 1071–5.
Gao YJ, Ren WH, Zhang YQ, Zhao ZQ. Contributions of the anterior
cingulate cortex and amygdala to pain- and fear-conditioned place
avoidance in rats. Pain 2004; 110: 343–53.
Hains BC, Saab CY, Waxman SG. Changes in electrophysiological
properties and sodium channel Nav1.3 expression in thalamic neurons
after spinal cord injury. Brain 2005; 128: 2359–71.
Hains BC, Waxman SG. Activated microglia contribute to the maintenance of chronic pain after spinal cord injury. J Neurosci 2006; 26:
4308–17.
Hains BC, Waxman SG. Sodium channel expression and the molecular
pathophysiology of pain after SCI. Prog Brain Res 2007; 161: 195–203.
Johnson H, Cowey A. Transneuronal retrograde degeneration of retinal
ganglion cells following restricted lesions of striate cortex in the monkey.
Exp Brain Res 2000; 132: 269–75.
Krenz NR, Weaver LC. Sprouting of primary afferent fibers after spinal
cord transection in the rat. Neuroscience 1998; 85: 443–58.
Kuesgen B, Frankel HL, Anand P. Decreased cutaneous sensory axon-reflex
vasodilatation below the lesion in patients with complete spinal cord
injury. Somatosens Mot Res 2002; 19: 149–52.
LaMotte RH, Lundberg LE, Torebjörk HE. Pain, hyperalgesia and activity
in nociceptive C units in humans after intradermal injection of
capsaicin. J Physiol (Lond) 1992; 448: 749–64.
Lenz FA, Kwan HC, Martin R, Tasker R, Richardson RT, Dostrovsky JO.
Characteristics of somatotopic organization and spontaneous neuronal
activity in the region of the thalamic principal sensory nucleus in
patients with spinal cord transection. J Neurophysiol 1994; 72: 1570–87.
Lin Q, Wu J, Willis WD. Dorsal root reflexes and cutaneous neurogenic inflammation after intradermal injection of capsaicin in rats.
J Neurophysiol 1999; 82: 2602–11.
Maynard FM Jr, Bracken MB, Creasey G, Ditunno JF Jr, Donovan WH,
Ducker TB, et al. International standards for neurological and functional
classification of spinal cord injury. American Spinal Injury Association.
Spinal Cord 1997; 35: 266–74.
Brain (2008), 131, 2387^2400
2399
McKay SM, Brooks DJ, Hu P, McLachlan EM. Distinct types of microglial
activation in white and grey matter of rat lumbosacral cord after
mid-thoracic spinal transection. J Neuropathol Exp Neurol 2007; 66:
698–710.
Meller ST, Gebhart GF. Nitric oxide (NO) and nociceptive processing in
the spinal cord. Pain 1993; 52: 127–36.
Namer B, Seifert F, Handwerker HO, Maihofner C. TRPA1 and TRPM8
activation in humans: effects of cinnamaldehyde and menthol.
Neuroreport 2005; 16: 955–9.
Österberg A, Boivie J, Thuomas KA. Central pain in multiple sclerosis–
prevalence and clinical characteristics. Eur J Pain 2005; 9: 531–42.
Parrent AG, Lozano AM, Dostrovsky JO, Tasker RR. Central pain in the
absence of functional sensory thalamus. Stereotact Funct Neurosurg
1992; 59: 9–14.
Popovich PG, Wei P, Stokes BT. Cellular inflammatory response after
spinal cord injury in Sprague-Dawley and Lewis rats. J Comp Neurol
1997; 377: 443–64.
Rolke R, Baron R, Maier C, Tolle TR, Treede RD, Beyer A, et al.
Quantitative sensory testing in the German Research Network on
Neuropathic Pain (DFNS): standardized protocol and reference values.
Pain 2006; 123: 231–43.
Saab CY, Waxman SG, Hains BC. Alarm or curse? The pain of
neuroinflammation. Brain Res Rev 2008; 58: 226–35.
Schafers M, Lee DH, Brors D, Yaksh TL, Sorkin LS. Increased sensitivity of
injured and adjacent uninjured rat primary sensory neurons to
exogenous tumor necrosis factor-alpha after spinal nerve ligation.
J Neurosci 2003; 23: 3028–38.
Schmelz M. A neural pathway for itch. Nat Neurosci 2001; 4: 9–10.
Schmelz M, Michael K, Weidner C, Schmidt R, Torebjörk HE,
Handwerker HO. Which nerve fibers mediate the axon reflex flare in
human skin? Neuroreport 2000; 11: 645–8.
Schmelz M, Schmidt R, Bickel A, Handwerker HO, Torebjörk HE. Specific
C-receptors for itch in human skin. J Neurosci 1997; 17: 8003–8.
Schmidt R, Schmelz M, Torebjork HE, Handwerker HO. Mechanoinsensitive nociceptors encode pain evoked by tonic pressure to human
skin. Neuroscience 2000; 98: 793–800.
Schmitt AB, Buss A, Breuer S, Brook GA, Pech K, Martin D, et al. Major
histocompatibility complex class II expression by activated microglia
caudal to lesions of descending tracts in the human spinal cord is not
associated with a T cell response. Acta Neuropathol 2000; 100: 528–36.
Scholz J, Woolf CJ. The neuropathic pain triad: neurons, immune cells and
glia. Nat Neurosci 2007; 10: 1361–8.
Sherwood AM, Dimitrijevic MR, McKay WB. Evidence of subclinical brain
influence in clinically complete spinal cord injury: discomplete SCI.
J Neurol Sci 1992; 110: 90–8.
Siddall PJ, McClelland JM, Rutkowski SB, Cousins MJ. A longitudinal
study of the prevalence and characteristics of pain in the first 5 years
following spinal cord injury. Pain 2003; 103: 249–57.
Siddall PJ, Middleton JW. A proposed algorithm for the management of
pain following spinal cord injury. Spinal Cord 2006; 44: 67–77.
Siddall PJ, Taylor DA, McClelland JM, Rutkowski SB, Cousins MJ. Pain
report and the relationship of pain to physical factors in the first 6
months following spinal cord injury. Pain 1999; 81: 187–97.
Siddall PJ, Yezierski RP, Loeser JD. Pain following spinal cord injury:
clinical features, prevalence and taxonomy. IASP Newsletter 2000; 3:
3–7.
Siddall PJ, Yezierski RP, Loeser JD. Taxonomy and epidemiology of
spinal cord injury pain. In: Yezierski RP, Burchiel K, editors. Progress
in pain research and management, vol. 23. spinal cord injury pain:
assessment, mechanisms, management. Seattle: IASP Press; 2002.
p. 9–24.
Snow BJ, Tsui JK, Bhatt MH, Varelas M, Hashimoto SA, Calne DB.
Treatment of spasticity with botulinum toxin: a double-blind study. Ann
Neurol 1990; 28: 512–5.
Störmer S, Gerner HJ, Gruninger W, Metzmacher K, Follinger S,
Wienke C, et al. Chronic pain/dysaesthesiae in spinal cord injury
patients: results of a multicentre study. Spinal Cord 1997; 35: 446–55.
2400
Brain (2008), 131, 2387^2400
Szolcsanyi J. Forty years in capsaicin research for sensory pharmacology
and physiology. Neuropeptides 2004; 38: 377–84.
Vestergaard K, Nielsen J, Andersen G, Ingeman-Nielsen M, ArendtNielsen L, Jensen TS. Sensory abnormalities in consecutive, unselected
patients with central post-stroke pain. Pain 1995; 61: 177–86.
Vierck CJJ, Siddall P, Yezierski RP. Pain following spinal cord injury:
animal models and mechanistic studies. Pain 2000; 89: 1–5.
Wasner G, Binder A, Kopper F, Baron R. No effect of sympathetic
sudomotor activity on capsaicin-evoked ongoing pain and hyperalgesia.
Pain 2000; 84: 331–8.
Wasner G, Brechot A, Schattschneider J, Allardt A, Binder A, Jensen TS,
et al. Effect of sympathetic muscle vasoconstrictor activity on capsaicininduced muscle pain. Muscle Nerve 2002; 26: 113–21.
Wasner G, Brock JA. Determinants of thermal pain thresholds in normal
subjects. Clin Neurophysiol 2008 (in press).
Wasner G, Naleschinski D, Baron R. A role for peripheral afferents in the
pathophysiology and treatment of at-level neuropathic pain in spinal
cord injury? A case report. Pain 2007; 131: 219–25.
Wasner G, Schattschneider J, Binder A, Baron R. Topical menthol–a
human model for cold pain by activation and sensitization of C
nociceptors. Brain 2004; 127: 1159–71.
Waxman SG, Hains BC. Fire and phantoms after spinal cord injury:
Na+ channels and central pain. Trends Neurosci 2006; 29: 207–15.
G. Wasner et al.
Weaver LC, Verghese P, Bruce JC, Fehlings MG, Krenz NR, Marsh DR.
Autonomic dysreflexia and primary afferent sprouting after clipcompression injury of the rat spinal cord. J Neurotrauma 2001; 18:
1107–19.
Weng HR, Dougherty PM. Response properties of dorsal root reflexes in
cutaneous C fibers before and after intradermal capsaicin injection in
rats. Neuroscience 2005; 132: 823–31.
Whitton TL, Johnson RW, Lovell AT. Use of the Rydel-Seiffer graduated
tuning fork in the assessment of vibration threshold in postherpetic
neuralgia patients and healthy controls. Eur J Pain 2005; 9: 167–71.
Wu G, Ringkamp M, Hartke TV, Murinson BB, Campbell JN, Griffin JW,
et al. Early onset of spontaneous activity in uninjured C-fiber
nociceptors after injury to neighboring nerve fibers. J Neurosci 2001;
21: RC140.
Yarnitsky D, Ochoa JL. Warm and cold specific somatosensory systems.
Psychophysical thresholds, reaction times and peripheral conduction
velocities. Brain 1991; 114: 1819–26.
Yezierski RP. Spinal cord injury: a model of central neuropathic pain.
Neurosignals 2005; 14: 182–93.
Zhao P, Waxman SG, Hains BC. Modulation of thalamic nociceptive processing after spinal cord injury through remote activation of
thalamic microglia by cysteine cysteine chemokine ligand 21. J Neurosci
2007; 27: 8893–902.