Pain 105 (2003) 255–264

www.elsevier.com/locate/pain

Chronic inflammation and compression of the dorsal root contribute to

sciatica induced by the intervertebral disc herniation in rats
Shu-Xun Houa,*, Jia-Guang Tanga, Hui-Sheng Chenb, Jun Chenb
a
Institute of Orthopedics, 304 Hospital of PLA, 51 Fucheng Road, Beijing 100037, PR China
b
Pain Research Centre, Institute of Neuroscience, The Fourth Military Medical University, Xi’an 710032, PR China
Received 1 January 2003; received in revised form 5 May 2003; accepted 13 May 2003

Abstract
The pathophysiological mechanisms underlying sciatica and back pain are not well understood. In the present study, a sciatica model was
developed to investigate the contributions of inflammation and compression of the dorsal root (DR). The procedure used autologous disc to
apply direct pressure to the L5 DR (disc compression, DC group). For control, five additional groups were included: (1) mechanical
compression of L5 DR without disc (compression, CP group); (2) epidurally placed disc without mechanical compression (disc group);
(3) epidurally placed nucleus pulposus (NP) without mechanical compression (NP group); (4) epidurally placed annulus fibrosus (AF)
without mechanical compression (AF group) and (5) sham group. The paw withdrawal latency to heat stimulation, paw withdrawal threshold
to mechanical stimulation, body weight, and motor function were determined pre- and post-surgery. It was observed that all experimental
groups with the exception of the sham group showed a progressive and prolonged mechanical hyperalgesia with the DC group having the
strongest effect. Furthermore, the disc group showed a greater mechanical hyperalgesia with earlier onset in comparison with the CP group
and disc, AF, and NP groups developed thermal hyperalgesia in addition to mechanical hyperalgesia following surgery. Finally, rats in all
groups showed normal motor function and body weight increase. These data suggest that this model is suitable to investigate the mechanisms
of sciatica and inflammation as well as mechanical compression is involved in the pathogenesis of this condition. Moreover, AF and NP may
contribute similarly to the development of sciatica and back pain.
q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Keywords: Disc herniation; Compression; Inflammation; Annulus fibrosus; Nucleus pulposus; Sciatica; Hyperalgesia

1. Introduction
Intervertebral disc herniation is a major cause of sciatica
and low back pain seriously affecting the health and life
quality of patients. Chronic mechanical compression of
dorsal roots (DRs) or dorsal root ganglions (DRGs) has long
been regarded as the sole cause of the sciatica pain based on
the finding that there is a close relationship between
herniated disc and sciatica by Mixter and Barr (1934).
However, more recent clinical observations showed that it
might not be the case. For example, sciatica is not present in
some subjects with obvious disc herniation (Halperin et al.,
1982; Wiesel et al., 1984) and vise versa (Ohnmeiss et al.,
1997, 1999). In addition, the severity of radicular symptoms
* Corresponding author. Tel.: þ 86-10-68417282; fax: þ 86-10-
68429998.
E-mail address: shuxunhou@yahoo.com.cn (S.X. Hou).
0304-3959/03/$20.00 q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
doi:10.1016/S0304-3959(03)00222-7
is often not related to the size of the herniated disc (Hou and
Hu, 1992; Thelander et al., 1992). These suggest that
additional mechanisms may be responsible for sciatica
besides chronic mechanical compression. There are increas-
ing evidence from clinical and animal studies that the
inflammation of DRs and DRGs might be involved. Simple
placement of autologous or heterologous discs (AF and/or
nucleus pulposus (NP)) to DRs or DRGs without com-
pression can induce pain-related behaviors (hyperalgesia/
allodynia) (Kawakami et al., 1996, 1998) and structural and
functional changes in the affected DRs (Kayama et al.,
1996; Yabuki et al., 1998). Thus, it is plausible that both
mechanical compression and inflammation contribute to
sciatica development.
Animal models have been developed to investigate
peripheral neuropathic pain (Bennett and Xie, 1988; Seltzer
et al., 1990; Kim and Chung, 1992) and these models have
greatly increased our understanding of the underlying
256 S.-X. Hou et al. / Pain 105 (2003) 255–264
mechanisms. However, these models may not be suitable to
determine the mechanism of sciatica because of its unique
pathogenesis. Recently, some animal models have been
developed for this purpose (Kawakami et al., 1994, 1996,
1998, 2000; Hu and Xing, 1998; Olmarker and Myers, 1998;
Winkelstein et al., 2001, 2002). In these models, DRs or
DRGs were ligated, deformed, or compressed to mimic the
clinical stage of chronic mechanical compression. In
addition, annulus fibrosus (AF) and/or NP were placed on
or near DRs or DRGs without compression to induce
inflammation following disc herniation. In the present
study, we developed a new model to simulate the clinical
pathogenesis of sciatica by a direct compression of L5 DR
with autologous disc. We determined behavioral changes of
the rats to mechanical and heat stimulations. Furthermore,
we investigated the relative contributions of mechanical
compression and inflammation to the development of
hyperalgesia in this new model.
2. Methods
2.1. Animals and surgery
The experiments were performed on 51 Sprague –
Dawley albino male rats weighing 250 –350 g. The animals
were provided by Laboratory Animal Center of the Fourth
Military Medical University (FMMU) and the experimental
protocol was reviewed and approved by the FMMU Animal
Care and Use Committee. IASP’s guidelines for pain
research in animals were followed (Zimmermann, 1983).
The animals were housed in plastic boxes in groups of three
at 22 – 26 8C with food and water available ad libitum in the
colony room. A 12:12 h light – dark cycle with lights on at
08:00 hours was maintained and testing was done between
09:00 and 18:30 hours. The animals were acclimatized to
the laboratory and habituated to the test boxes for at least
30 min each day for 5 days before surgery.
Under sodium pentobarbital anesthesia (50 mg/kg,
intraperitoneal (i.p.)), a midline incision was made. All
surgical procedures were performed on the left side. During
the procedure, cautions were taken to prevent infection and
to minimize the injury. The paraspinous muscles were
dissected free from the left spinous processes to expose the
transverse processes. Left L4 – 5 laminectomy and L5
facetectomy were performed to expose and identify the L5
DR and DRG under a dissecting microscope. To exclude the
possibility that the immune response to allografted inter-
vertebral disc will play a role in the development of
hyperalgesia, autologous disc, NP, and AF were obtained
from the coccygeal intervertebral discs according to the
reported surgical protocol (Kawakami et al., 1998). The tail
was tightly ligated with a rubber band to prevent bleeding
and then was amputated distal to the ligation. The
amputated tail was denuded of skin and the disc including
NP (2 mg) and AF (2 mg) was obtained by curetting
coccygeal intervertebral discs. All surgical procedures were
performed under sterile condition.
The rats were randomly divided into six groups subjected
to various procedures detailed below. After surgery, all
wounds were washed with sterile saline and antibiotics, and
then sutured by layers. For the first 3 days after surgery,
antibiotics was injected i.p. twice a day.
2.1.1. Sham group ðn6Þ
To rule out the effect of surgery (laminectomy,
facetectomy, and tail amputation) on mechanical and heat
sensitivity of the hind paw, surgical exposure of the DRs
and DRGs and amputation of the tail without other
manipulations were performed as a control.
2.1.2. Disc compression group ðn11Þ
The autologous disc (4 mg) containing NP (2 mg) and
AF (2 mg) was placed proximal to the L5 DRG between the
DRG and the underlying bone to form direct compression of
DR. Using a glass needle with a smooth hook at the end, DR
was lifted towards midline and disc was carefully placed to
avoid damage to DR or DRG. Continued compression was
verified before wound closure and at the termination of the
experiments.
2.1.3. Compression (CP) group ðn8Þ
To determine the role of DR compression, the bone
fraction from a spinous process with similar size used in the
disc compression (DC) group was placed on the DR at the
same place as the DC group. The smooth facet of bone
fraction was placed against the DR.
2.1.4. Disc group ðn11Þ
To investigate the role of inflammation, autologous disc
(4 mg) containing NP (2 mg) and AF (2 mg) was placed on
the L5 DR, but without compression. The size of the disc is
the same as that used for the DC group.
2.1.5. Nucleus pulposus group ðn8Þ
The NP and AF groups were designed to investigate the
respective contributions of NP and AF to the pathogenesis
of sciatica. NP (2 mg), with the same size as that of NP in
disc group, was placed on the L5 DR, but without
mechanical compression.
2.1.6. Annulus fibrosus group ðn7Þ
AF (2 mg), with the size same as that of AF in the Disc
group, was placed on the L5 DR, but without mechanical
compression.
2.2. Behavioral tests
2.2.1. Examination of mechanical and thermal hyperalgesia
Paw withdrawal thermal latency (PWTL) and paw
withdrawal mechanical threshold (PWMT) of both hind
limbs were examined before and after surgery. Before
 S.-X. Hou et al. / Pain 105 (2003) 255–264
surgery, PWTL and PWMT were examined for 3 –4 days
until a steady value was obtained. For examination of
PWTL, the rat was placed on the surface of a 2 mm thick
glass plate covered with a Plexiglas chamber
(20 £ 20 £ 25 cm), the latency of paw withdrawal reflex
to heat stimuli was measured with a RTY-3 radiant heat
stimulator (Chen et al., 1999, Xi’an Fenglan Instrumental
Factory, PR China). The radiant heat source was a high
intensity projector halogen lamp bulb (100 W, 10.5 V)
positioned under the glass floor directly beneath the
targeting area on the dorsal surface of the hind paw. The
distance between the projector lamp bulb and the lower
surface of the glass floor was adjusted to produce a 5 mm
diameter light spot on the floor surface. Five stimuli were
repeated to the same site and the mean PWTL was obtained
from the latter three to four stimuli. The inter-stimulus
interval for each heat test was more than 10 min at the same
region and 5 min at the different region. To avoid excessive
tissue injury, a cut-off of 40 s was imposed. For examination
of PWMT, mechanical stimuli were applied by 11
individual monofilaments with bending forces at 58.5,
78.4, 98.0, 147.0, 196.0, 294.0, 343.0, 392.0, 441.0, 490.0,
588.0 mN. The rat was placed on a metal mesh floor covered
with the same Plexiglas chamber and von Frey filaments
were applied in an ascending intensity order from under-
neath the metal mesh floor to the testing site of both hind
paws. A single von Frey filament was applied 10 times
(once every several seconds) to each testing site. A bending
force sufficient to evoke 50% of the paw withdrawal
occurrence was expressed as PWMT (Chen et al., 1999).
The changes in PWTL and PWMT were calculated as the
percentage of difference between PWTL or PWMT pre- and
post-surgery: PWTLpost/PWTLpre £ 100 and PWMTpost/
PWMTpre £ 100, respectively. Less than 100% indicated
hyperalgesia and more than 100% hypoalgesia.
2.2.2. Evaluation of motor function and body weight
The rats were placed in the big Plexiglas chamber and
walking behaviors and posture of resting position were
monitored. Motor function was evaluated as follows:
1 ¼ normal gait without paw deformities; 2 ¼ normal gait
with a marked paw deformity; 3 ¼ slight gait disturbance in
which a drop paw was present and 4 ¼ severe gait
disturbance with motor paresis of the ipsilateral hind paw
(Kawakami et al., 1994). The body weight was measured
before and after surgery as an indication of general health.
2.3. Statistical analysis
All data were expressed by means ^ SEM. The statisti-
cal difference within a group was analyzed by Friedman test
followed by Wilcoxon signed-ranks matched-pairs test. The
statistical difference between groups was analyzed by two
way ANOVA followed by Scheffe test. The difference at a
given testing time point postoperatively between groups
257
was further compared by Mann –Whitney U-test. P , 0:05
was considered to be statistically significant.
3. Results
3.1. Mechanical hyperalgesia
With the exception of the Sham group, all experimental
groups showed significant decrease in PWMT for bilateral
paws over time (Friedman test, Fig. 1, Table 1). Compared
with the baseline value, the rats in DC group demonstrated a
significant and prolonged decrease in PWMT of the
ipsilateral paw, indicative of mechanical hyperalgesia,
from 1 day to 6 weeks after surgery (Fig. 1a). The decrease
was progressive, reaching peak at 3 weeks after surgery and
was not resolved at the end of 6 weeks observation period.
The other groups also exhibited a similar time course of
decrease in PWMT (Fig. 1b – e).
In order to distinguish the differential roles of mechanical
and inflammatory factors in the development of mechanical
hyperalgesia, further comparison was performed as to the
degree of decrease in PWMT of the ipsilateral paw between
DC, CP, and disc groups by two way ANOVA followed by
Scheffe test. The results showed that DC group demon-
strated a greater mechanical hyperalgesia in the ipsilateral
paw compared with CP ðP , 0:001Þ and disc groups
ðP ¼ 0:011Þ, while disc group displayed a greater hyper-
algesia than CP group ðP ¼ 0:002Þ. In order to determine
the differential roles of AF and NP in the development of
mechanical hyperalgesia identified in disc group, further
comparison was performed as to the degree of decrease in
PWMT of the ipsilateral paw between disc, NP, and AF
groups by two way ANOVA followed by Scheffe test. The
results showed that disc group demonstrated a greater
mechanical hyperalgesia in the ipsilateral paw compared
with NP ðP ¼ 0:019Þ and AF groups ðP ¼ 0:028Þ, but there
was no significant difference between NP and AF groups
ðP ¼ 0:999Þ.
Interestingly, contralateral paw to the treated side in
all the groups, but not sham group, also displayed a
prolonged, progressive decrease in PWMT with similar
pattern to that of ipsilateral paw. In comparison with
baseline value, Wilcoxon analysis revealed the following
difference: DC group, 1 week ðP ¼ 0:025Þ, 2 –6 weeks post-
surgery ðP ¼ 0:003 – 0:005Þ; CP group, 4 – 6 weeks
ðP ¼ 0:012 – 0:018Þ; disc group, 5 days –4 weeks ðP ¼
0:005 – 0:018Þ post-surgery; NP group, 3 days ðP ¼ 0:035Þ,
3 weeks ðP ¼ 0:017Þ, and 4 weeks ðP ¼ 0:028Þ post-
surgery; AF group, 2 – 6 weeks ðP ¼ 0:028 – 0:043Þ post-
surgery. No significant difference was demonstrated
between groups ðP ¼ 0:575, two way ANOVA), neither at
all the testing time points post-surgery between groups
(Mann– Whitney U-test).
258 S.-X. Hou et al. / Pain 105 (2003) 255–264

Fig. 1. Showing the time course of changes in paw withdrawal mechanical threshold (PWMT) of both sides in DC (a), CP (b), Disc (c), NP (d), AF (e), and
sham (f) groups, respectively. The statistical difference between pre-surgery (21 day) and post-surgery in all groups was indicated by asterisk (*). The
statistical difference at a given time point between DC and CP groups was indicated by triangle (K (a)), whereas the statistical difference at a given time point
between disc and CP groups by square (A (c)). *P , 0:05; **P , 0:01. Vertical bars: ^SEM.
 S.-X. Hou et al. / Pain 105 (2003) 255–264 259

Table 1
The time course of changes in paw withdrawal mechanical threshold (mN) of bilateral paws in all groups

DC group Disc group CP group NP group AF group Sham group

Baseline
Ipsilateral 38.75 ^ 1.25 40.00 ^ 1.34 39.38 ^ 1.99 36.88 ^ 3.40 36.43 ^ 2.61 33.83 ^ 5.04
Contralateral 38.75 ^ 1.25 40.63 ^ 1.13 39.38 ^ 1.75 36.88 ^ 3.26 36.43 ^ 2.61 33.83 ^ 5.04
1 day
Ipsilateral 34.13 ^ 2.88 31.25 ^ 4.90 41.88 ^ 1.32 35.06 ^ 4.47 33.64 ^ 5.02 32.83 ^ 5.57
Contralateral 41.25 ^ 0.82 38.75 ^ 1.83 40.63 ^ 1.75 36.88 ^ 3.40 39.29 ^ 2.02 33.67 ^ 5.84
2 day
Ipsilateral 22.98 ^ 4.65 23.28 ^ 5.67 36.63 ^ 2.49 31.49 ^ 5.04 30.79 ^ 4.49 34.17 ^ 4.73
Contralateral 41.88 ^ 1.32 37.50 ^ 2.11 40.00 ^ 0.94 36.25 ^ 3.63 32.14 ^ 3.25 34.67 ^ 3.13
3 day
Ipsilateral 22.56 ^ 5.14 25.83 ^ 4.90 31.00 ^ 4.36 25.06 ^ 4.45 27.93 ^ 3.99 35.83 ^ 2.39
Contralateral 39.38 ^ 1.13 36.25 ^ 3.37 32.88 ^ 4.08 26.50 ^ 4.46 31.43 ^ 3.22 34.50 ^ 3.84

5 day
Ipsilateral 20.69 ^ 4.84 18.58 ^ 4.85 30.63 ^ 3.59 28.13 ^ 4.38 28.14 ^ 5.47 33.00 ^ 3.00
Contralateral 33.81 ^ 4.35 30.00 ^ 4.01 32.88 ^ 3.36 31.25 ^ 4.15 30.29 ^ 5.35 34.67 ^ 2.85
1 week
Ipsilateral 14.75 ^ 4.72 20.84 ^ 5.54 28.63 ^ 2.43 23.44 ^ 4.61 26.71 ^ 5.78 30.83 ^ 4.17
Contralateral 28.13 ^ 4.62 30.83 ^ 2.83 32.88 ^ 3.49 29.13 ^ 3.11 36.43 ^ 4.04 30.83 ^ 3.96
2 week
Ipsilateral 12.13 ^ 4.56 17.06 ^ 5.52 28.25 ^ 3.83 24.44 ^ 3.73 28.57 ^ 3.37 30.83 ^ 3.52
Contralateral 28.19 ^ 5.39 28.50 ^ 4.02 34.50 ^ 2.63 31.88 ^ 2.66 30.43 ^ 3.44 33.33 ^ 4.22

3 week
Ipsilateral 7.03 ^ 1.61 16.25 ^ 5.18 20.00 ^ 5.00 16.63 ^ 4.50 17.07 ^ 4.97 29.17 ^ 5.54
Contralateral 14.88 ^ 3.41 20.00 ^ 3.35 28.75 ^ 4.41 19.69 ^ 4.44 24.14 ^ 5.33 26.17 ^ 4.97
4 week
Ipsilateral 15.81 ^ 5.24 17.06 ^ 4.43 20.38 ^ 3.86 25.63 ^ 4.72 20.86 ^ 3.42 30.00 ^ 5.63
Contralateral 22.25 ^ 4.62 26.25 ^ 4.90 23.50 ^ 3.32 23.25 ^ 3.36 23.79 ^ 4.09 29.50 ^ 6.65
6 week
Ipsilateral 18.06 ^ 4.47 23.94 ^ 3.42 19.38 ^ 2.90 23.38 ^ 5.36 22.14 ^ 3.43 29.67 ^ 4.74
Contralateral 28.91 ^ 2.96 35.75 ^ 3.09 23.25 ^ 2.78 25.38 ^ 4.12 27.14 ^ 3.69 28.83 ^ 3.61
No. of animals 11 11 8 8 7 6

3.2. Heat hyperalgesia
In the present study, the sensitivity of bilateral hind paws
to heat stimuli was also investigated in all the treatment
groups. Unlike PWMT data, PWTL of bilateral paw did not
exhibit a significant decrease over time within a group
(Friedman test, Fig. 2, Table 2). Compared with the baseline
value, the statistically significant decrease in PWTL of the
ipsilateral paw was only identified at the time point of
1 week ðP ¼ 0:013Þ post-surgery and 2 days for the disc
group; 4 weeks ðP ¼ 0:018Þ and 6 weeks ðP ¼ 0:043Þ post-
surgery in the ipsilateral paw for AF group, and 5 days ðP ¼
0:03Þ in the ipsilateral paw for NP group (Fig. 2c – e). The
statistically significant decrease of the contralateral paw was
not identified at all the testing time points post-surgery in all
groups, but with a statistically significant increase at the
time point of 2 days ðP ¼ 0:022Þ in disc group. DC, CP, and
sham groups did not show any significant changes in PWTL
of both sides, compared with the baseline value (Fig. 2a,b
and f). Two way ANOVA analysis did not reveal the
significant difference between groups. No significant
difference was also identified at all the testing time points
when compared between groups (Mann – Whitney U-test).
3.3. Motor function and body weight
All rats in all groups displayed normal weight increase
and motor function over the observed time period post-
surgery (data not shown).
4. Discussion
4.1. New sciatica model
The present results showed that the direct compression of
DR by the autologous disc could produce a progressive,
long-term mechanical hyperalgesia that is greater than
260 S.-X. Hou et al. / Pain 105 (2003) 255–264

Fig. 2. Showing the time course of changes in paw withdrawal thermal latency (PWTL) of both sides in DC (a), CP (b), disc (c), NP (d), AF (e), and sham (f)
groups, respectively. The statistical difference between pre-surgery (21 day) and post-surgery in all groups was indicated by asterisk. *P , 0:05. Vertical bars:
^SEM.

compression alone or disc placement alone. Thus, this the treatment produced a prolonged and greater mechanical
model appears to be suitable to investigate the pathogenesis hyperalgesia. Third, all the rats tolerated the surgery and did
of sciatica. First, the direct compression of DR by disc is not exhibit obvious motor disfunction and general distress,
relevant to the pathogenesis of clinical sciatica. Second, and finally, the model has little variability since all 11 rats
 S.-X. Hou et al. / Pain 105 (2003) 255–264 261

Table 2
The time course of changes in paw withdrawal thermal latency(s) of bilateral paws in all groups

DC group Disc group CP group NP group AF group Sham group

Baseline
Ipsilateral 16.58 ^ 1.17 17.13 ^ 1.24 17.26 ^ 1.25 19.64 ^ 1.50 17.90 ^ 1.67 16.18 ^ 1.37
Contralateral 14.59 ^ 1.06 15.04 ^ 0.99 17.66 ^ 1.66 19.69 ^ 1.33 18.01 ^ 2.25 16.17 ^ 1.04
1 day
Ipsilateral 16.81 ^ 1.66 14.16 ^ 1.12 15.01 ^ 1.07 18.64 ^ 1.17 13.03 ^ 0.96 16.92 ^ 1.74
Contralateral 15.19 ^ 1.09 14.49 ^ 1.25 16.30 ^ 1.19 18.85 ^ 1.21 15.44 ^ 1.37 17.72 ^ 1.65
2 day
Ipsilateral 15.16 ^ 0.68 15.54 ^ 1.11 17.76 ^ 1.46 20.59 ^ 1.72 14.99 ^ 1.79 15.53 ^ 1.65
Contralateral 15.71 ^ 1.05 17.28 ^ 0.74 16.65 ^ 0.86 21.64 ^ 1.79 17.44 ^ 0.73 17.25 ^ 1.07
3 day
Ipsilateral 15.78 ^ 1.06 14.88 ^ 1.46 19.30 ^ 2.03 19.03 ^ 1.23 18.73 ^ 0.88 16.82 ^ 1.85
Contralateral 16.31 ^ 1.47 14.06 ^ 1.17 18.71 ^ 1.77 21.64 ^ 1.81 18.64 ^ 1.59 17.62 ^ 1.69

5 day
Ipsilateral 13.33 ^ 0.93 13.19 ^ 1.34 16.28 ^ 1.85 15.09 ^ 0.86 18.46 ^ 2.13 16.13 ^ 2.33
Contralateral 15.14 ^ 1.26 12.61 ^ 1.20 17.75 ^ 2.59 17.19 ^ 1.68 16.97 ^ 2.00 16.65 ^ 1.75
1 week
Ipsilateral 13.35 ^ 0.98 13.61 ^ 1.03 15.65 ^ 1.04 18.43 ^ 1.29 18.99 ^ 1.91 15.18 ^ 2.15
Contralateral 14.36 ^ 1.07 12.90 ^ 1.10 15.44 ^ 1.48 16.21 ^ 0.68 19.43 ^ 1.78 15.82 ^ 1.35
2 week
Ipsilateral 13.71 ^ 0.92 14.83 ^ 1.42 17.19 ^ 2.26 18.68 ^ 1.73 14.76 ^ 1.66 18.02 ^ 2.75
Contralateral 14.26 ^ 0.70 14.30 ^ 1.41 15.08 ^ 2.27 18.16 ^ 1.58 13.46 ^ 1.12 17.05 ^ 1.91

3 week
Ipsilateral 13.96 ^ 1.16 14.55 ^ 1.25 17.25 ^ 2.02 18.38 ^ 1.59 15.84 ^ 1.65 14.33 ^ 1.10
Contralateral 13.64 ^ 1.02 14.36 ^ 1.10 17.91 ^ 2.16 19.18 ^ 2.00 15.01 ^ 1.65 17.57 ^ 2.34
4 week
Ipsilateral 13.54 ^ 1.13 13.34 ^ 0.83 15.94 ^ 1.83 16.28 ^ 1.22 13.47 ^ 0.86 17.63 ^ 1.12
Contralateral 14.99 ^ 1.46 13.08 ^ 0.78 15.13 ^ 1.87 15.25 ^ 0.71 14.57 ^ 1.66 18.93 ^ 1.62
6 week
Ipsilateral 16.23 ^ 1.42 15.65 ^ 1.79 15.40 ^ 2.24 18.13 ^ 2.00 12.30 ^ 0.77 14.80 ^ 1.64
Contralateral 15.23 ^ 1.71 15.18 ^ 0.70 16.83 ^ 2.39 18.50 ^ 2.21 12.73 ^ 0.89 16.90 ^ 2.68
No. of animals 11 11 8 8 7 6

displayed a prolonged decrease in PWMT of both side after
the direct compression of DR.
In the present study, DC and CP treatments did not
produce thermal hyperalgesia, while disc, AF, and NP
treatments showed a significant decrease in PWTL at
several time points post-surgery. Consistent with the current
results, previous studies have reported different behavioral
results induced by mechanical compression of DR or DRG,
for example, the coexistence of mechanical and thermal
hyperalgesia (Song et al., 1999) or dissociation between the
two (Tabo et al., 1999). In addition, results about behavioral
hyperalgesia induced by epidural disc without compression
were also conflicting (Olmarker and Myers, 1998;
Kawakami et al., 2000). The conflicting behavioral results
may be due to differences in models, experimental
manipulations, measuring methods, and animal species
and strains. Nonetheless, it is puzzling in the present study
that DC treatment, in contrast to disc, AF, and NP treat-
ments, do not produce thermal hyperalgesia, since disc in
DC group also contains AF and NP. One possible
explanation is that DC treatment produces mechanical
compression of DR, while Disc, AF, and NP treatments do
not.
Almost all treatments (DC, CP, disc, AF, and NP)
produced contralateral mechanical hyperalgesia, in addition
to ipsilateral side. In agreement with previous studies (Chen
et al., 1999; Zhang et al., 2000), contralateral hyperalgesia is
more delayed in onset and less severe. Previous studies have
reported the existence of contralateral hyperalgesia in
neuropathic pain models (Seltzer et al., 1990; Kim and
Chung, 1992; Zhang et al., 2000), cutaneous injury pain
models (Coderre and Melzack, 1985; Kissin et al., 1998;
Chen et al., 2000, 2001), and experimental arthritis models
(Donaldson, 1999). As to the mechanisms underlying the
contralateral hyperalgesia, a neural communication path-
way between bilateral dorsal horns has been proposed
(Koltzenburg et al., 1999). However, a non-neural pathway
or supraspinal modulation is also possible (Urban and
262 S.-X. Hou et al. / Pain 105 (2003) 255–264
Gebhart, 1999; Samad et al., 2001; Chen et al., 2003).
Clinically, mirror pain is also observed in patients with
radiculopathy who experience bilateral leg or arm pain. In
addition, contralateral effect is also observed in other
radiculopathy models (Kawakami et al., 1994; Hunt et al.,
2001). Nevertheless, the nature of contralateral hyperalgesia
in this model deserves further studies.
The present study showed that rats in the disc, AF, and
NP groups displayed a long-term mechanical hyperalgesia.
The results are in agreement with other studies showing a
long-term mechanical hyperalgesia (2 – 3 weeks post-
surgery) after the epidural application of NP without
compression (Kawakami et al., 1998, 2000). Surprisingly,
the present results are not in agreement with a recent study
reporting that epidural application of NP produced a short
lasting mechanical hyperalgesia, lasting only for 1 day
(Obata et al., 2002). The discrepancy is not clear. One
possible reason may be due to the difference of the methods
determining mechanical hyperalgesia. Decrease in PWMT
was used to indicate mechanical hyperalgesia in the present
and other studies (Kawakami et al., 1998, 2000), whereas
response frequency of paw withdrawal to mechanical
stimuli with a given intensity was used in the study by
Obata et al. (2002). Furthermore, the present results are also
not consistent with those from the experimental neuritis
models showing hyperalgesia lasting for only several days
(Eliav et al., 1999; Sorkin and Doom, 2000; Chacur et al.,
2001), although mechanical hyperalgesia in the present
study is also mediated by the inflammatory mechanism. As
to the discrepancy, several explanations can be given. First,
this discrepancy may be due to the different mechanism of
the induction of nerve inflammation in the two conditions.
When epidurally placed, disc, NP, or AF will release many
bioactive substances, for example, phospholipase A2,
cytokines, nitric oxide, etc. all of which will contribute to
the development of nerve root inflammation. In addition,
autoimmune response induced by NP previously secluded
from the immune system will play a role in the development
of nerve inflammation. Second, epidural application of disc,
NP, or AF will produce the inflammation of dura mater,
which may result in spreading of bioactive substances
released from Disc, NP, or AF to the spinal cord due to the
possible opening of blood – brain barrier caused by the
inflammation of dura mater. Third, secondary effect induced
by the necrosis of Disc, NP, or AF may contribute to the
long-term mechanical hyperalgesia.
4.2. Differential role of chronic compression and
inflammation in sciatica
Since Mixter and Barr (1934) reported the relationship
between herniated disc and sciatica, the mechanical
compression has been regarded as the sole cause of the
herniated disc-induced sciatica. Consistent with previous
animal studies and clinical observations (Weber, 1983; Hu
and Xing, 1998; Winkelstein et al., 2001, 2002), the present
results also suggest that mechanical compression plays a
role in the pathogenesis of sciatica, because the mechanical
compression without disc produced a prolonged mechanical
hyperalgesia. However, some studies have reported that
isolated mechanical compression could not produce signifi-
cant pain-related behaviors (Kawakami et al., 1994;
Olmarker and Myers, 1998). One possible reason might
result from the difference between the methods of
mechanical compression.
Based on the previous studies, several possible mechan-
isms are responsible for mechanical hyperalgesia identified
in CP group. First, it may result from electrophysiological
changes of peripheral and central nervous systems induced
by mechanical compression of nerve root, for example, a
reduction in conduction velocity of peripheral nerve
(Cornefjord et al., 1997), a maintained repetitive firing of
spinal wide-dynamic-range (WDR) neurons (Hanai et al.,
1996), and increased spontaneous discharge of myelinated
fibres (Hu and Xing, 1998). Second, nerve fibre damage
induced by mechanical compression (Yoshizawa et al.,
1995; Cornefjord et al., 1997) may play a role. Third,
neuroimmune response and neuroinflammation of central
nervous system may be involved in the mechanical
hyperalgesia (Hashizume et al., 2000a,b; Rutkowski et al.,
2002). In addition, nerve root ligation was reported to
induce an increase in the number of substance P- and
calcitonin gene-related peptide-like cells and vasoactive
intestinal polypeptide concentration in lumbar DRG
(Kawakami et al., 1994), which may contribute to
mechanical hyperalgesia in CP group.
Recent clinical and animal studies suggest that the
inflammatory responses induced by herniated disc also play
an important role in pathogenesis of sciatica. For instance,
the sciatica is not present in some subjects with herniated
disc and vice versa (Halperin et al., 1982; Wiesel et al.,
1984; Ohnmeiss et al., 1997, 1999). In animal studies,
epidural disc can induce the inflammatory responses, which
may result in morphological and functional changes in the
affected nerve (Olmarker et al., 1997; Kayama et al., 1996)
and electrophysiological changes in the DRG cells and
spinal WDR cells (Takebayashi et al., 2001; Anzai et al.,
2002). More importantly, epidural disc without compression
could produce obvious pain-related behaviors such as
hyperalgesia (Kawakami et al., 1996, 1998). The present
study also supports the above opinion: (1) epidural disc, AF,
or NP treatments without compression induced a prolonged
mechanical hyperalgesia and (2) DC of DR produced a
greater hyperalgesia than the sole compression.
Thus, the present data support the opinion that both
mechanical and inflammatory factors contribute to the
pathogenesis of sciatica. However, the combined contri-
butions of mechanical compression and inflammation are
not the simple summation of two insults, but the mutual
interaction between them (Olmarker and Myers, 1998).
Furthermore, the present data suggest that the mechanical
compression and inflammation play a differential role in
 S.-X. Hou et al. / Pain 105 (2003) 255–264
the different time course of the development of hyperalgesia
induced by DC. The comparison between DC, CP, and disc
groups showed that DC treatment produced a greater
mechanical hyperalgesia in the ipsilateral paw, compared
with CP ðP , 0:001Þ and disc groups ðP ¼ 0:011Þ, while
epidural disc without compression produced an earlier and
greater hyperalgesia than the sole compression (P ¼ 0:002,
Fig. 1b and c). This may implicate that the inflammatory
process plays a more important role in the onset stage
(1 – 2 days post-surgery) of hyperalgesia induced by DC,
whereas inflammatory and mechanical factors contribute to
the hyperalgesia in the subsequent stage.
Although the present data showed the role of disc-
induced inflammation in the development of hyperalgesia,
the role of AF vs NP in the inflammation-induced sciatica is
unclear (Olmarker and Myers, 1998; Kawakami et al., 1996,
1998). The present study also investigated the differential
contributions of AF vs NP to the inflammation. The results
showed that both AF and NP treatments led to decrease in
PWMT, suggesting that AF as well as NP is involved in the
development of inflammation and sciatica.
In conclusion, the present study shows that the direct
compression of DR by disc is a clinically relevant model to
investigate the mechanisms of sciatica. Furthermore, data
from the different treatment groups suggest that mechanical
compression and inflammation are involved in the develop-
ment of sciatica.
Acknowledgements
The authors are very grateful to Dr H.L. Pan from the
Pennsylvania State University College of Medicine, USA
and Dr X.J. Xu from Karolinska Institutet, Sweden for their
critical and helpful comments and correcting the English
usage of the manuscript.
References
Anzai H, Hamba M, Onda A, Konno S, Kikuchi S. Epidural application of
nucleus pulposus enhances nociresponses of rat dorsal horn neurons.
Spine 2002;27:50– 5.
Bennett GJ, Xie YK. A peripheral mononeuropathy in rat that produces
disorders of pain sensation like those seen in man. Pain 1988;33:
87–107.
Chacur M, Milligan ED, Gazda LS, Armstrong C, Wang H, Tracey KJ,
Maier SF, Watkins LR. A new model of sciatic inflammatory neuritis
(SIN): induction of unilateral and bilateral mechanical allodynia
following acute unilateral peri-sciatic immune activation in rats. Pain
2001;94:231 –44.
Chen HS, Chen J, Chen J, Guo WG, Zhen MH. Establishment of bee
venom-induced contralateral heat hyperalgesia in the rat is temporally
dependent upon central summation of ongoing primary afferent input
from the injured site. Neurosci Lett 2001;298:57–60.
Chen HS, Chen J, Sun YY. Contralateral heat hyperalgesia induced by
unilaterally intraplantar bee venom injection is produced by central
changes: a behavioral study in the conscious rat. Neurosci Lett 2000;
284:45–8.
263
Chen HS, Li MM, Shi J, Chen J. Supraspinal contribution to development
of both tonic nociception and referred mirror hyperalgesia: a
comparative study between formalin test and bee venom test in the
rat. Anesthesiology 2003;98:1231– 6.
Chen J, Luo C, Li HL, Chen HS. Primary hyperalgesia to mechanical and
heat stimuli following subcutaneous bee venom injection into the
plantar surface of hind paw in the conscious rat: a comparative study
with the formalin test. Pain 1999;83:67–76.
Coderre TJ, Melzack R. Increased pain sensitivity following heat injury
involves a central mechanism. Behav Brain Res 1985;15:259–62.
Cornefjord M, Sato K, Olmarker K, Rydevik B, Nordborg C. A model for
chronic nerve root compression studies: presentation of a porcine model
for controlled, slow-onset compression with analyses of anatomic
aspects, compression onset rate, and morphologic and neurophysiologic
effects. Spine 1997;22:946–57.
Donaldson LF. Unilateral arthritis: contralateral effects. Trends Neurosci
1999;22:495–6.
Eliav E, Herzberg U, Ruda MA, Bennett GJ. Neuropathic pain from an
experimental neuritis of the rat sciatic nerve. Pain 1999;83:169–82.
Hanai F, Matsui N, Hongo N. Changes in responses of wide dynamic range
neurons in the spinal dorsal horn after dorsal root or dorsal root ganglion
compression. Spine 1996;21:1408 –14.
Halperin N, Agasi M, Hendel D. Painless root compression following disc
extrusion. Arch Othrop Trauma Surg 1982;10:63–6.
Hashizume H, DeLeo JA, Colburn RW, Weinstein JN. Spinal glial
activation and cytokine expression after lumbar root injury in the rat.
Spine 2000a;25:1206 –17.
Hashizume H, Rutkowski MD, Weinstein JN, DeLeo JA. Central
administration of methotrexate reduces mechanical allodynia in an
animal model of radiculopathy/sciatica. Pain 2000b;87:159–69.
Hou SX, Hu RW. Relation of severity of low back and leg pain to type
of prolapsed intervertebral disk. J Neurol Orthop Med Surg 1992;13:
26– 9.
Hu SJ, Xing JL. An experimental model for chronic compression of dorsal
root ganglion produced by intervertebral foramen stenosis in the rat.
Pain 1998;77:15 –23.
Hunt JL, Winkelstein BA, Rutkowski MD, Weinstein JN, DeLeo JA.
Repeated injury to the lumbar nerve roots produces enhanced
mechanical allodynia and persistent spinal neuroinflammation. Spine
2001;26:2073–9.
Kawakami M, Tamaki T, Hayashi N, Hashizume H, Matsumoto T,
Minamide A, Kihira T. Mechanical compression of the lumbar nerve
root alters pain-related behaviors induced by the nucleus pulposus in the
rat. J Orthop Res 2000;18:257–64.
Kawakami M, Tamaki T, Hayashi N, Hashizume H, Nishi H. Possible
mechanism of painful radiculopathy in lumbar disc herniation. Clin J
Orthop 1998;351:241–51.
Kawakami M, Tamaki T, Weinstein JN, Hashizume H, Nishi H, Meller ST.
Pathomechanisms of pain-related behavior produced by allografts of
intervertebral disc in the rat. Spine 1996;21:2101–7.
Kawakami M, Weinstein JN, Spratt KF, Chatani K, Traub RJ, Meller ST,
Gebhart GF. Experimental lumbar radiculopathy. Immunohisto-
chemical and quantitative demonstrations of pain induced by lumbar
nerve root irritation of the rat. Spine 1994;19:1780– 94.
Kayama S, Konno S, Olmarker K, Yabuki S, Kikuchi S. Incision of the
annulus fibrosus induces nerve root morphologic, vascular, and
functional changes. An experimental study. Spine 1996;21:2539–43.
Kim SH, Chung JM. An experimental model for peripheral neuropathy
produced by segmental spinal nerve ligation in the rat. Pain 1992;50:
355– 63.
Kissin I, Lee SS, Bardley EL. Effect of prolonged nerve block on
inflammatory hyperalgesia in rats: prevention of late hyperalgesia.
Anesthesiology 1998;88:224–32.
Koltzenburg M, Wall PD, McMahon SB. Does the right side know what the
left is doing. Trends Neurosci 1999;22:122–7.
Mixter WJ, Barr JS. Rupture of the intervertebral disc with involvement of
the spinal canal. N Engl J Med 1934;211:210 –5.
264 S.-X. Hou et al. / Pain 105 (2003) 255–264
Obata K, Tsujino H, Yamanaka H, Yi D, Fukuoka T, Hashimoto N,
Yonenobu K, Yoshikawa H, Noguchi K. Expression of neurotrophic
factors in the dorsal root ganglion in a rat model of lumbar disc
herniation. Pain 2002;99:121–32.
Ohnmeiss DD, Vanharanta H, Ekholm J. Degree of disc disruption and
lower extremity pain. Spine 1997;22:1600–5.
Ohnmeiss DD, Vanharanta H, Ekholm J. Relation between pain location
and disc pathology: a study of pain drawings and CT/discography. Clin
J Pain 1999;15:210–7.
Olmarker K, Brisby H, Yabuki S, Nordborg C, Rydevik B. The effects of
normal, frozen, and hyaluronidase-digested nucleus pulposus on nerve
root structure and function. Spine 1997;22:471 –5.
Olmarker K, Myers RR. Pathogenesis of sciatic pain: role of herniated
nucleus pulposus and deformation of spinal nerve root and dorsal root
ganglion. Pain 1998;78:99–105.
Rutkowski MD, Winkelstein BA, Hickey WF, Pahl JL, DeLeo JA. Lumbar
nerve root injury induces central nervous system neuroimmune
activation and neuroinflammation in the rat: relationship to painful
radiculopathy. Spine 2002;27:1604–13.
Samad TA, Moore KA, Sapirstein A, Billet S, Allchorne A, Poole S,
Bonventrek JV, Woolf CJ. Interleukin-1 mediated induction of COX-2
in the central nervous system contributes to inflammatory pain
hypersensitivity. Nature 2001;410:471–5.
Seltzer Z, Dubner R, Shir Y. A novel behavioral model of neuropathic pain
behavior in rats. Pain 1990;43:205–18.
Song XJ, Hu SJ, Greenquist KW, Zhang JM, Lamotte RH. Mechanical and
thermal hyperalgesia and ectopic neuronal discharge after chronic
compression of dorsal root ganglia. J Neurophysiol 1999;82:3347– 58.
Sorkin LS, Doom CM. Epineural application of TNF elicits an acute
mechanical hyperalgesia in the awake rat. J Peripher Nerv Syst 2000;5:
96–100.
Tabo E, Jinks SL, Eisele JHJ, Carstens E. Behavioral manifestation of
neuropathic pain and mechanical allodynia, and changes in spinal
dorsal horn neurons, following L4–L6 dorsal root constriction in rats.
Pain 1999;80:503–20.
Takebayashi T, Cavanaugh JM, Ozaktay AC, Kallakuri S, Chen CY. Effect
of nucleus pulposus on the neural activity of dorsal root ganglion. Spine
2001;26:940–5.
Thelander U, Fagerlund M, Friberg S, Larsson S. Straight leg raising test
versus radiologic size, shape, and position of lumbar disc hernias. Spine
1992;17:395–9.
Urban MO, Gebhart GF. Supraspinal contributions to hyperalgesia. Proc
Natl Acad Sci USA 1999;96:7687–92.
Weber H. Lumbar disc herniation. A controlled, prospective study with
ten years of observation. Spine 1983;8:131–40.
Wiesel SW, Tsourmas N, Feffer HL, Citrin CM, Patronas N. A study of
computer-assisted tomography. I. The incidence of positive CAT scans
in an asymptomatic group of patients. Spine 1984;9:549– 51.
Winkelstein BA, Rutkowski MD, Weinstein JN, Deleo JA. Quantification
of neural tissue injury in a rat radiculopathy mode: comparison of local
deformation, behavioral outcomes, and spinal cytokine mRNA for two
surgeons. J Neurosci Methods 2001;111:49–57.
Winkelstein BA, Weinstein JN, Deleo JA. The role of mechanical
deformation in lumbar radiculopathy. An in vivo model. Spine 2002;
27:27–33.
Yabuki S, Kikuchi S, Olmarker K, Myers RR. Acute effects of nucleus
pulposus on blood flow and endoneurial fluid pressure in rat dorsal root
ganglia. Spine 1998;23:2517–23.
Yoshizawa H, Kobayashi S, Morita T. Chronic nerve root compression:
pathophysiologic mechanism of nerve root dysfunction. Spine 1995;20:
397 –407.
Zhang JM, Li HQ, Brull SJ. Perfusion of the mechanically compressed
lumbar ganglion with lidocaine reduces mechanical hyperalgesia and
allodynia in the rat. J Neurophysiol 2000;84:798–805.
Zimmermann M. Ethical guidelines for investigations of experimental pain
in conscious animals. Pain 1983;16:109 –10.