Cannabis constituent synergy in a mouse neuropathic pain model

Abstract

Cannabis and its psychoactive constituent Δ9-tetrahydrocannabinol (THC) have efficacy against neuropathic pain however, this is hampered by their side-effects. It has been suggested that co-administration with another major constituent cannabidiol (CBD) might enhance the analgesic actions of THC and minimise its deleterious side-effects. We examined the basis for this phytocannabinoid interaction in a mouse chronic constriction injury (CCI) model of neuropathic pain. Acute systemic administration of THC dose-dependently reduced CCI-induced mechanical and cold allodynia, but also produced motor incoordination, catalepsy and sedation. CBD produced a lesser dose-dependent reduction in allodynia, but did not produce the cannabinoid side-effects. When co-administered in a fixed ratio, THC and CBD produced a biphasic dose-dependent reduction in allodynia. At low doses, the THC:CBD combination displayed a 200-fold increase in anti-allodynic potency, but had lower efficacy compared to that predicted for an additive drug interaction. By contrast, high THC:CBD doses had lower potency, but greater anti-allodynic efficacy compared to that predicted for an additive interaction. Only the high dose THC:CBD anti-allodynia was associated with cannabinoid side-effects and these were similar to those of THC alone. Unlike THC, the low dose THC:CBD anti-allodynia was not cannabinoid receptor mediated. These findings demonstrate that CBD synergistically enhances the pain relieving actions of THC in an animal neuropathic pain model, but has little impact on the THC-induced side-effects. This suggests that low dose THC:CBD combination treatment has potential in the treatment of neuropathic pain.

Full text

Research Paper
Cannabis constituent synergy in a mouse
neuropathic pain model
Sherelle L. Casey*, Nicholas Atwal, Christopher W. Vaughan
Abstract
Cannabis and its psychoactive constituent D9-tetrahydrocannabinol (THC) have efficacy against neuropathic pain, however, this is
hampered by their side effects. It has been suggested that co-administration with another major constituent cannabidiol (CBD) might
enhance the analgesic actions of THC and minimise its deleterious side effects. We examined the basis for this phytocannabinoid
interaction in a mouse chronic constriction injury (CCI) model of neuropathic pain. Acute systemic administration of THC dose-
dependently reduced CCI-induced mechanical and cold allodynia, but also produced motor incoordination, catalepsy, and
sedation. Cannabidiol produced a lesser dose-dependent reduction in allodynia, but did not produce the cannabinoid side effects.
When co-administered in a fixed ratio, THC and CBD produced a biphasic dose-dependent reduction in allodynia. At low doses, the
THC:CBD combination displayed a 200-fold increase in anti-allodynic potency, but had lower efficacy compared with that predicted
for an additive drug interaction. By contrast, high THC:CBD doses had lower potency, but greater anti-allodynic efficacy compared
with that predicted for an additive interaction. Only the high dose THC:CBD anti-allodynia was associated with cannabinoid side
effects and these were similar to those of THC alone. Unlike THC, the low dose THC:CBD anti-allodynia was not cannabinoid
receptor mediated. These findings demonstrate that CBD synergistically enhances the pain-relieving actions of THC in an animal
neuropathic pain model, but has little impact on the THC-induced side effects. This suggests that low dose THC:CBD combination
treatment has potential in the treatment of neuropathic pain.
Keywords: Cannabis, Neuropathic pain model, Phytocannabinoid, Synergy, Receptor
1. Introduction
Neuropathic pain is a debilitating condition resulting from damage
to the peripheral or central nervous system and can be caused by
physical trauma, such as accidents, surgery and stroke, or by
diseases such as diabetes, cancer, and immune disorders.
16
This
chronic pain state is difficult to manage, with many patients
experiencing pain that is refractory to currently available
pharmacotherapies and is often associated with disabling side
effects.
9
Given these problems, there is a need for novel
therapeutic options as either first-line medications or as adjuvants
to current therapies.
The plant Cannabis sativa has been used for thousands of
years to treat various medical conditions including pain,
neuralgia, cramps, nausea, diarrhoea, convulsions, and migraine.
Cannabis contains a multitude of phytocannabinoids including
the psychoactive constituent D9-tetrahydrocannabinol (THC) and
other constituents such as cannabidiol (CBD) which do not
produce THC-like psychotropic side effects.
21
To date, most
clinical trials on chronic pain have used either whole raw
cannabis, THC, or nabiximols, which are combinations of THC
and CBD. The most recent meta-analyses of these trials suggest
that cannabinoids have potential in the treatment of neuropathic
pain, although there are differing views on their clinical efficacy
and safety.
11,22,36
Given this uncertainty, it is crucial that we
understand the basis for the actions of these 2 phytocannabi-
noids alone, and in combination.
22
Numerous preclinical animal studies have shown that synthetic
cannabinoids have high pain-relieving efficacy in animal models
of neuropathic pain.
22
By contrast, relatively few studies have
examined the actions of phytocannabinoids in these neuropathic
pain models, which is surprising given their ubiquitous use in
clinical trials. These studies have shown that THC and CBD
reduce the allodynia associated with rodent neuropathic pain
models.
5–8,13,30,34,37
Recently, it has been demonstrated that
THC and CBD synergistically reduce the development of
allodynia in a chemotherapy-induced model of neuropathic
pain.
18
A limitation of these studies is that they do not provide
a systematic dose–response analysis of both the anti-allodynic
actions and side effects of phytocannabinoids. This is important
because it provides information on the efficacy and therapeutic
window (TW) of cannabinoids,
1,12,17,27
both of which are
important factors in their clinical use. Furthermore, although it
has been proposed that nabiximols might offer a superior,
synergistic alternative to individual cannabis constituents,
24
their
actions in neuropathic pain models are unknown. Thus, the
preclinical basis for the use of specific phytocannabinoids in
neuropathic pain states, either alone or in combination, remains
unclear. To address this, we used an isobolographic approach to
examine whether there is a synergistic interaction between THC
Sponsorships or competing interests that may be relevant to content are disclosed
at the end of this article.
Pain Management Research Institute, Kolling Institute of Medical Research,
Northern Clinical School, Royal North Shore Hospital, University of Sydney, Sydney,
New South Wales, Australia
*Corresponding author. Address: Pain Management Research Institute, Kolling
Institute of Medical Research, Northern Clinical School, Royal North Shore Hospital,
University of Sydney, St Leonards, Sydney, NSW 2065, Australia. Tel.: 61(2)9926
4963. E-mail address: scas5381@uni.sydney.edu.au (S. L. Casey).
PAIN 158 (2017) 2452–2460
©2017 International Association for the Study of Pain
http://dx.doi.org/10.1097/j.pain.0000000000001051
2452 S.L. Casey et al.·158 (2017) 2452–2460 PAIN
®
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

and CBD in a mouse neuropathic pain model and whether this is
associated with an improvement in their analgesic efficacy
and TW.
2. Methods
All experiments in this study were performed on 8- to 9-week-old
male C57BL/6 mice, and are reported in compliance with the
Animals In Research Reporting In Vitro Experiment (ARRIVE)
guidelines and those of the “NH&MRC Code of Practice for the
Care and Use of Animals in Research in Australia”, and were
approved by the Royal North Shore Hospital Animal Ethics
Committee. Mice were obtained from the Kolling Institute Animal
Facility, and were initially housed in groups of 4 and then
individually after surgery. Individually ventilated cages were
maintained at 22˚C to 23˚C and humidity 65% to 75%, with
a 12:12 hours light:dark cycle. Animals had ad libitum access to
food and water throughout all stages of the study. Cages were
enriched with a mouse house igloo, tissues for nesting, and either
a straw or paddle pop stick for play.
2.1. Neuropathic pain model
A commonly used nerve injury model of neuropathic pain, the
chronic constriction injury (CCI) model of neuropathic pain, was
used for this study.
2,17
For the CCI surgery, anaesthesia was
induced with 4% isoflurane and then maintained with 2%
isoflurane in saturatedoxygen (1 mL·min
21
). To avoid hypothermia,
the mice were positioned on a heat mat for the duration of the
procedure. The fur over the mid-thigh region was clipped to avoid
contamination of the wound. The left common sciatic nerve was
exposed at the mid-thigh level and 2 size 6-0 chromic gut
ligatures placed around the nerve 2 to 3 mm apart. The ligatures
were tightened until a twitch of the foot was observed, while
ensuring that the blood supply to the nerve was not compro-
mised. The muscle incision was closed with 6-0 silk and the skin
incision closed with either a simple continuous suture pattern
using 6-0 silk or with tissue glue. The animals were recovered and
monitored before being returned to their cages. Animals were
then monitored daily until the day of the experiment.
2.2. Behavioural testing
To assess nerve injury–induced mechanical and cold allodynia,
animals were placed in elevated Perspex cages with a mesh wire
floor and were left to acclimatise for 30 minutes before any
testing. Mechanical allodynia was tested by applying a series of
von Frey filaments (0.2-6.84 g; North Coast Medical, San Jose,
CA) to the plantar surface of the operated hind paw. The measure
of the paw withdrawal threshold (PWT) was calculated using the
simplified up-down protocol.
3
A positive response was noted as
a rapid withdrawal of the paw from the hair. Cold allodynia was
tested by applying 20 mL of acetone to the plantar surface of the
left hind paw to induce evaporative cooling. The number of pain-
like responses (licking or chewing the paw, shaking the paw, or
looking at the paw) was counted.
Common cannabinoid side effects including motor impair-
ment, catalepsy, and sedation were also measured. Motor
impairment was tested using the rotarod. For this assay, mice
were placed on a bar which gradually increased its speed of
rotation from 4 to 30 rpm over a 300-second timeframe. The time
at which each mouse fell off the bar, or just held on to the bar for 2
consecutive rotations, was recorded. Catalepsy was tested using
the bar test. For this assay, the forepaws of the mouse were
placed on a raised bar (2 cm high) and the time taken to either
remove the forepaws from the bar or jump up on to the bar with
the hind paws was recorded, with a cut-off time of 120 seconds.
Spontaneous locomotor activity was tested using the dark open
field test. For this assay, mice were placed in a 25 cm 325 cm
open-topped Perspex box and their activity was recorded for 2.5
minutes. A 4 34 square grid was super-imposed over the
recording and the number of times the hindquarters of the animal
crossed a line on the grid was recorded. All tests were performed
under low-level red light (,3 lux).
2.3. Experimental protocol
After initial habituation to their holding cages and equipment used
for testing, animals underwent baseline testing for the allodynia and
side effect measures with the exception of the open field test. They
then underwent CCI surgery 2 days after baseline testing. Drug
testing was conducted at 8 to 10 days after surgery. Each animal
underwent only 1 drug testing experiment, and was euthanised by
carbon dioxide asphyxiation after testing. The experimenter was
blinded to the drug or drug combination being tested.
For the time course experiments, only mechanical PWT and
rotarod latency were measured (because of short inter-test
intervals). Measurements were taken immediately before drug
administration and then at 0.5, 1, 2, 4, and 6 hours after drug
administration. For the dose response and antagonist experi-
ments, all allodynia and side effect assays (with the exception of
the open field) were tested immediately before drug injection and
at 1 and 2 hours after injection. These postinjection time points
were selected to coincide with the time of peak drug effect as
determined by the time course study (see Results, Fig. 1). To
maintain novelty, the open field test was performed only once at
1.5 hours after injection. Acetone, rotarod, and bar tests were
performed twice at each specified time point. Von Frey and open
field tests were performed only once at each specified time point.
2.4. Drugs and administration
The phytocannabinoids THC and CBD were obtained from
THCPharm (Frankfurt, Germany). The CB1 receptor antagonist
AM281 and the CB2 receptor antagonist AM630 were
obtained from Cayman Chemicals (Ann Arbor, MI). Stock
solutions of all drugs used were prepared in dimethyl sulfoxide.
These were administered as a subcutaneous injection (volume
of 0.01-0.015 mL·g
21
in a vehicle consisting of 15% dimethyl
sulfoxide, 5% Tween80, and saline) in the loose skin between
the shoulder blades. Drug injection solutions were made up
immediately before administration.
2.5. Analysis and statistics
For the time course experiments, raw data were compared using
2-way repeated analysis of variance, with time and drug
treatment as within- and between-subjects factors, respectively
(Prism; Graphpad Software, La Jolla, CA). When time-drug
treatment interactions were significant, within-treatment group
post hoc comparisons were performed using Dunnett correction
for multiple comparisons. For the dose response and antagonist
experiments, all data (with the exception of open field) were
normalised as a percentage of the maximum possible effect. For
mechanical PWT, this was calculated as (post-drug 2pre-drug)/
(cut-off 2pre-drug), where the cut-off was 6.84 g. For acetone
and rotarod, this was calculated as (pre-drug 2post-drug)/
(drug), and for the bar test as (post-drug 2pre-drug)/(cut-off)
December 2017·Volume 158 ·Number 12 www.painjournalonline.com 2453
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

where the cut-off was 120 seconds. Raw data were used for the
open field as predrug values were not obtained. For the
cannabinoid receptor antagonist experiments, data were com-
pared using 1-way analysis of variance, with the Tukey correction
for post hoc comparisons. All data are presented as mean 6
SEM. Data were considered significantly different when P,0.05.
There were 6 animals per treatment group for all experiments, as
in our previous synergy study in a mouse neuropathic pain
model.
17
Monophasic dose response curves were constructed by fitting
sigmoidal curves with variable slope, of the form
Effect ¼EMax
Dosep
Dosep1EDp
50(1)
where each drug had a maximal effect of E
Max
, a half maximally
effective dose (ED
50
), and a Hill slope of p. Biphasic dose
response curves were constructed by fitting the sum of 2
independent sigmoidal curves as in Equation 1. Comparisons of
these parameters between dose–response fits were performed
using an extra sum-of-square F-test (Prism). The TW of each drug
treatment group was calculated as the average of the ratio of
ED
50
s for each side effect measure divided by that for mechanical
and cold anti-allodynia.
Isobolographic analysis was used to determine whether an
interaction between THC and CBD was present when they were
administered in combination, as follows.
29
In the present study,
the THC:CBD combination was tested at a 1:1 ratio of their
ED
50
s. The predicted additive effect of THC and CBD was
calculated from their individual dose response curve parameters,
using an approach which takes into account different maximal
effects and Hill slopes for the 2 drugs. In this approach, the
predicted additive effect of the 2 drugs A and B is:
Eða;bÞ¼
EBb1CB
k
1
=
p
p
b1CB
k
1
=
p
p
1Cp
B(2)
k¼EB
EA11Cq
A
aq21 (2)
at doses a and b for drugs A and B, respectively, with maximal
effects of E
A
and E
B
(where E
B
.E
A
), ED
50
sofC
A
and C
B
, and Hill
slopes of q and p. The experimental combination dose response
curve was then compared with the theoretical predicted additive
dose response curve for THC plus CBD obtained from Equation 2
using a modified ttest to compare data at individual doses.
29
The
nonlinear isobole which describes the relationship between the
drugs at doses a and b, for a specified effect level B
i
(eg, the 50%
effect level), was calculated using the equation given below:
b¼Bi2CB
hEB
EA11Cq
A
aq21i
1
=
p
(3)
3. Results
3.1. Time course of action of THC and cannabidiol
We first examined the time course of action of near maximal doses
of THC and CBD (30 mg·kg
21
) to establish their time of peak effect.
At 8 to 10 days after CCI surgery, the effect of acute administration
of THC, CBD and matched vehicle on both PWT and rotarod
latency differed over time (Figs. 1A and B, F(12,90) 538, 12.9, P,
0.0001, 0.001). THC produced a significant increase in mechanical
PWTat0.5to4hoursafterinjection(Fig. 1A). In these animals, THC
also produced a significant decrease in rotarod latency at 0.5 to 6
hours after injection (Fig. 1B). Cannabidiol produced a significant
increase in mechanical PWT at 2 hours after injection (Fig. 1A).
Cannabidiol did not have a significant effect on rotarod latency (Fig.
1B). The vehicle for THC and CBD did not have a significant effect
on mechanical PWT or rotarod latency (Figs. 1A and B). These data
indicate that THC offers good analgesia but with significant side
effects at this dose, although CBD is a less efficacious analgesic at
the same dose but produces no cannabinoid side effects. The time
of peak effect when averaged over THC and CBD for these assays
was 1 to 2 hours after injection. We therefore used this time point in
all subsequent experiments.
3.2. Dose–response profiles of D9-tetrahydrocannabinol
and cannabidiol
We next examined the individual dose–response profiles of THC
and CBD (0.01-56 mg·kg
21
) for a wider range of anti-allodynia
and side effect assays. THC produced a dose-dependent
reversal of the CCI-induced decrease in mechanical PWT and
the CCI-induced increase in acetone responses, with similar
ED
50
s and maximal effects (Figs. 2A and B,Table 1,P.0.05).
THC also produced a dose-dependent decrease in rotarod
latency and open field crossings, and a dose-dependent increase
in bar latency (Figs. 2C–E,Table 1). The ED
50
s of THC for rotarod
and bar latency were greater than that for both mechanical PWT
Figure 1. Time course of action of THC and CBD. Time plots of the effects of
THC (30 mg·kg
21
), CBD (30 mg·kg
21
), and matched vehicle on (A) mechanical
PWT and (B) rotarod latency (n 56 per treatment group). Animals received
a single subcutaneous injection at time 0 hours, 8 to 10 days after CCI surgery
(post-CCI); pre-CCI data are also displayed. *, **, ***, and **** denote P,0.05,
0.01, 0.001, and 0.0001, respectively, compared with 0 hours post-CCI for
each treatment group. CBD, cannabidiol; CCI, chronic constriction injury ; paw
withdrawal threshold; THC, D9-tetrahydrocannabinol.
2454 S.L. Casey et al.·158 (2017) 2452–2460 PAIN
®
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

and acetone responses (Table 1,P,0.01-0.0001). The ED
50
of
THC for open field crossings was not significantly different from
that for both mechanical PWT and acetone responses (Table 1,
P.0.05). The TW of THC, measured as the ratio of the ED
50
of
side effects to anti-allodynia, was 5.0 when averaged across all
measures of anti-allodynia and side effects (Table 1).
Cannabidiol produced a dose-dependent reduction in the CCI-
induced decrease in mechanical PWT and the CCI-induced
increase in acetone responses, with similar ED
50
s and maximal
effects (Figs. 2A and B,Table 1,P.0.05). Although their ED
50
s
were not significantly different (P.0.05), CBD had a lesser max-
imal effect than THC on mechanical PWT and acetone-induced
Figure 2. Dose–response curves for THC and CBD. Dose–response curves for the effect of THC and CBD on (A) mechanical (PWT, (B) acetone-induced
responses, (C) rotarod latency, (D) bar latency, and (E) open field crossings. Where appropriate, the sigmoidal parametric fit is shown (parameter estimates in
Table 1). CBD, cannabidiol; PWT, paw withdrawal threshold; THC, D9-tetrahydrocannabinol.
Table 1
Dose–response characteristics and TW of WIN55212, morphine, and their combination.
Anti-allodynia Side effects TW
Mechanical, PWT Acetone, responses Rotarod latency Bar latency Open field crossings
THC
ED
50
4.3 (0.6) 2.7 (0.5) 15 (2) 29 (0.2) 6.0 (1) 5.0 (1.4-10.7)
E
Max
94 (0.1) 74 (0.1) 75 (17) 100 (2) —
Hill slope 1.7 (0.1) 1.4 (0.4) 1.6 (0.3) 5.4 (0.4) 21.1 (0.2)
CBD
ED
50
3.9 (0.1) 3.6 (0.4) .56.2 .56.2 .56.2 .15.0
E
Max
65 (5) 49 (16) ND ND —
Hill slope 1.9 (0.7) 0.9 (0.7) ND ND ND
Combination THC:CBD—predicted additive
ED
50
4.8 3.2 ND ND ND ND
Combination THC:CBD—high potency
ED
50
0.021 (0.007) 0.016 (0.005) — — — 1360 (762-2000)
E
Max
57 (2) 35 (4) — — —
Hill slope 1.6 (0.4) 2.1 (1.0) — — —
Combination THC:CBD—low potency
ED
50,
THC comp. 21 (2) 20 (2) 26 (8), 12 (3) 32 (1), 15 (0.1) 16 (3), 7.7 (1.6) 1.2 (0.8-1.6)
E
Max
39 (5) 60 (12) 76 (57) 71 (1)
Hill slope 10 (4) 1.6 (0.6) 1.4 (1.3) 3.5 (0.1) 22.0 (0.7)
Sigmoidal curve fit parameters for THC and CBD administered individually, the predicted additive effect of combination THC:CBD, the high potency combination THC:CBD actions, and the low potency combination THC:CBD
actions. For the low potency THC:CBD actions, the ED
50
s for the THC dose component are also shown (THC comp.). The parameters include the ED
50
(mg·kg
21
), E
max
(%MPE), Hill slope, and the TW. The values of these
parameters are shown as mean (SEM), mean (range), or ND (not determined).
CBD, cannabidiol; ED
50
, median effective dose; PWT, paw withdrawal threshold; THC, D9-tetrahydrocannabinol; TW, therapeutic window.
December 2017·Volume 158 ·Number 12 www.painjournalonline.com 2455
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

responses (Figs. 2A and B,Table 1,P,0.05). Cannabidiol did
not have an effect on rotarod latency, bar latency, or the number
of open field crossings over the range of doses tested
(Figs. 2C–E). Given that CBD failed to produce any side effect,
its TW was estimated to be greater than 15 (Table 1). These
observations indicate that THC has a greater anti-allodynic
efficacy, but lesser TW compared with CBD.
3.3. Effect of a fixed-ratio combination of
D9-tetrahydrocannabinol and cannabidiol on allodynia
We next examined the effect of combined treatment with THC
and CBD on allodynia, using an isobolographic approach which
accounts for their differing maximal effects and Hill slopes.
17,29
THC and CBD were administered in a fixed 1:1 ratio based on
their individual ED
50
s. When averaged across both anti-allodynia
assays, the ED
50
ratio for THC:CBD was 1:1.13 by weight, which
is similar to their relative content in nabiximols. The predicted
additive effect of combined THC and CBD treatment on
mechanical PWT and acetone responses had monophasic dose
dependence, with ED
50
and maximum values similar to that for
THC alone (Figs. 3A and B,Table 1).
Co-administration of THC and CBD produced a dose-
dependent increase in mechanical PWT and decrease in acetone
responses (Figs. 3A and B). Unlike the predicted additive dose
response curves, the experimentally obtained THC:CBD combi-
nation dose response curves were biphasic, displaying high- and
low-potency anti-allodynic actions (Figs. 3A and B). Given this
biphasic profile, we assessed the level of synergy for both the
high- and low-potency sites of action.
The high potency THC:CBD induced reduction in mechanical
and cold allodynia had an ED
50
which was 200 to 230 times less
than that predicted for an additive interaction and accounted for
57% and 35% of the maximum effect of THC:CBD (Figs. 3A and
B,Table 1). Consequently, the effect of THC:CBD on mechanical
PWT and acetone responses was greater than its predicted
additive effect at doses below 1.3 to 4.0 mg·kg
21
(Figs. 3A and
B). For this high potency site, the combination THC:CBD dose
which reduced mechanical and cold allodynia by 50% (ED
50
) was
similar to the isobole predicted to produce a 1% reduction if the
interaction was additive (Figs. 4A and B).
The low potency THC:CBD induced reduction in mechanical
and cold allodynia had an ED
50
which was 4.4 to 6.3 times greater
than that predicted for an additive interaction, and accounted for
39% and 60% of the maximum effect of THC:CBD (Figs. 3A and
B,Table 1). The effect of THC:CBD on mechanical PWT and
acetone responses was greater than its predicted additive effect
at the highest doses tested (13-40 mg·kg
21
), but not at
intermediate doses (Figs. 3A and B). For this low potency site,
the combination THC:CBD dose which reduced mechanical and
cold allodynia by 50% (ED
50
) was similar to the isobole predicted
to produce a 55% to 65% reduction if the interaction was additive
(Figs. 4A and B). Together, these observations indicate that
combination THC:CBD treatment has 2 modes of action,
including a low dose synergistic reduction in allodynia and a high
dose sub-additive reduction in allodynia.
3.4. Side effects of the fixed-ratio combination of
D9-tetrahydrocannabinol and cannabidiol
A synergistic anti-allodynic interaction will only lead to an
increase in the TW of a drug combination if its side effects do not
display an equivalent level of synergy. We therefore examined
thesideeffectdose–responseprofileoftheTHC:CBD
combination. THC:CBD produced a dose-dependent increase
in rotarod and bar latency, and a decrease in open field
crossings (Figs. 5A–C). In contrast to its anti-allodynic actions,
the dose–response profiles of the THC:CBD for the rotarod, bar,
and open field tests were monophasic, and had similar ED
50
s
(Figs. 5A–C,Table 1,P.0.05). The TW of THC:CBD was 1360
when side effects were compared with the high potency anti-
allodynic action, but was only 1.2 when compared with the low
potency anti-allodynic action (Table 1).
The side effect assays could not be analysed using the
isobolographic approach due to the lack effect of CBD alone. To
determine whether there was a THC:CBD interaction for the side
effects, we compared the dose response profiles of the THC
component of THC:CBD with that obtained for THC when
administered alone. For the rotarod and open field latencies,
there was no difference between the ED
50
of the THC component
of THC:CBD compared with that of THC when administered
alone (Figs. 5A and C,Table 1,P.0.05). In contrast, the ED
50
of
the THC component of THC:CBD for bar latency was less than
that of THC when administered alone (Fig. 5B,Table 1,P,
0.0001). It might be noted, however, that the maximal effect of the
Figure 3. Dose–response curves for the effect of combined THC and CBD
treatment on allodynia. Dose–response curves showing the effect of
administration of THC and CBD in a fixed ratio (THC:CBD at 1:1.13 fixed
weight ratio) on (A) mechanical PWT and (B) acetone-induced responses. The
biphasic sigmoidal fit for the combination THC:CBD data is shown (ED
50
and
E
Max
estimates shown in Table 1). Also shown are the predicted additive
dose response curves for combination THC:CBD, and the sigmoidal fits for
THC and CBD alone (reproduced from Fig. 2). **, ***, and **** denote P,0.01,
0.001, and 0.0001, respectively, for THC:CBD experimental combination
vs predicted additive at equivalent doses. CBD, cannabidiol; ED
50
,
median effective dose; PWT, paw withdrawal threshold; THC, D9-
tetrahydrocannabinol.
2456 S.L. Casey et al.·158 (2017) 2452–2460 PAIN
®
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

THC component of THC:CBD on bar latency was less than that of
THC when administered alone (Fig. 5B,Table 1,P,0.0001).
Together, these observations indicate that the side effects of
THC:CBD are largely produced by THC, and these are not
substantially affected by CBD. Thus, the differences in the TW of
combination THC:CBD at the low and high potency sites are
largely due to altered anti-allodynic potency.
3.5. Role of cannabinoid receptors
We finally examined the role of cannabinoid receptors in the
actions of THC and CBD by co-administering them with the
selective CB1 and CB2 antagonists, AM281 and AM630
(3 mg·kg
21
). AM281 and AM630 alone had no effect on any of
the pain, or behavioural scores (Figs. 6A–E). AM281 abolished
the effect of a maximal dose of THC (30 mg·kg
21
) on mechanical
PWT, but only partially reduced its effect on acetone responses
(Figs. 6A and B). AM281 abolished the effect of THC on rotarod
latency, bar latency and open field crossing (Figs. 6C–E). AM630
did not alter the effect of THC on mechanical PWT or acetone
responses (Figs. 6A and B). Although AM630 did not alter the
effect of THC on rotarod latency and open field crossing, it
partially reduced its effect on bar latency (Figs. 6C–E).
AM281 did not alter the effect of a maximal dose of CBD
(30 mg·kg
21
) on mechanical PWT, but partially reduced its effect
on acetone responses (Fig. 6B). Similarly, AM630 did not alter the
effect of CBD on mechanical PWT, but partially reduced its effect
on acetone responses (Figs. 6A and B). Cannabidiol did not have
an effect on rotarod latency, bar latency, and open field crossing,
and this was not altered by AM281, or AM630 (Figs. 6C–E).
We also examined the effect of these antagonists on
combination THC:CBD at a low dose (0.4 mg·kg
21
) which was
maximal for the high potency anti-allodynia, and at a high dose
(40 mg·kg
21
) which was maximal for the low potency anti-
allodynia. AM281 and AM630 did not alter the effect of low dose
THC:CBD on mechanical PWT, or acetone responses (Figs. 6A,
B). Low dose THC:CBD did not have an effect on rotarod latency,
bar latency and open field crossing, and this was not altered by
AM281, or AM630 (Figs. 6C–E).
AM281 abolished effect of high dose THC:CBD on mechanical
PWT, and partially reduced its effect on acetone responses (Figs.
6A and B). In addition, AM281 abolished the effect of high dose
THC:CBD on rotarod latency, bar latency and open field
crossings (Figs. 6C–E). AM630 did not alter the effect of high
dose THC:CBD on mechanical PWT, or acetone responses
(Figs. 6A and B). AM630 did not alter the effect of high dose THC:
Figure 4. Isoboles for combination THC and CBD treatment at a range of effect levels. Isoboles for THC:CBD co-administration on (A) mechanical PWT and (B)
acetone-induced responses. The experimental data for the combination, at its ED
50
(50% effect level, back symbol), are shown as part of the continuum of fixed-
ratio effects (black dotted lines). Theoretical isoboles of additivity for effect levels of 5, 10, 25, and 50% are shown for comparison (Equation 3). CBD, cannabidiol;
ED
50
, median effective dose; PWT, paw withdrawal threshold; THC, D9-tetrahydrocannabinol.
Figure 5. Dose–response curves for the side effects of combination THC and CBD treatment. Dose–response curves showing the effects of a THC and CBD
combination treatment on (A) rotarod latency, (B) bar latency, and (C) open field crossings. The x-axis represents the total dose of THC and CBD when given in
combination (THC:CBD), the dose of THC when given in combination with CBD (THC in THC:CBD), and the dose of THC when given alone (THC). With the
exception of open field (for which raw data is shown), all data are displayed as %MPE. CBD, cannabidiol; MPE, maximum possible effect; PWT, paw withdrawal
threshold; THC, D9-tetrahydrocannabinol.
December 2017·Volume 158 ·Number 12 www.painjournalonline.com 2457
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

CBD on rotarod latency, bar latency, and open field crossings
(Figs. 6C–E).
4. Discussion
This study has demonstrated that co-administration of the
phytocannabinoids THC and CBD synergistically attenuates
allodynia in a mouse model of neuropathic pain. Interestingly,
this nabiximol combination had 2 modes of action. At low doses,
the THC:CBD combination had partial anti-allodynic efficacy and
was devoid of THC-like side effects. At high doses, the THC:CBD
combination had maximal anti-allodynic efficacy, but a reduced
TW compared with THC alone. These findings have major
implications for the clinical use of the phytocannabinoids THC
and CBD, individually and in combination.
The psychoactive cannabis constituent THC produced dose-
dependent reductions in the mechanical and cold allodynia
induced by the CCI nerve-injury model of neuropathic pain. This
high efficacy anti-allodynia was consistent with that previously
reported for single-dose and full-dose–response studies.
7,8,13,37
In these animals, THC also produced impairment of motor
performance on the rotarod, catalepsy in the bar test, and
sedation in the dark open field test. These side effects were
consistent with those previously reported in na¨
ıve, unoperated
animals.
14,19,20,28,31,39
Interestingly, THC reduced allodynia with
a potency 5 times greater than that at which it produced side
effects. This is important because it demonstrates that THC has
a greater TW (ED
50
side effects/ED
50
anti-allodynia) than that of
synthetic high-efficacy cannabinoid receptor agonists which
have no window between their anti-allodynic actions and side
effects.
12,17,27
The difference between the TW of THC and
synthetic cannabinoid receptor agonists may be due to a number
of factors, such as the lower efficacy of THC compared with
WIN55212, or to differences in their mechanisms of action. These
observations indicate that THC is a more viable therapeutic option
for the treatment of neuropathic pain than the high-efficacy,
synthetic pan-cannabinoid receptor agonists which have been
the subject of most animal studies.
The cannabis constituent CBD also produced dose-
dependent reductions in nerve injury–induced mechanical and
cold allodynia. This was qualitatively similar to some, but not all,
previous studies using rodent models of neuropathic
pain.
6,13,30,34
Although the anti-allodynic potency of CBD was
similar to that of THC, it displayed lower efficacy. Unlike THC,
CBD did not produce catalepsy, impair motor performance, or
produce sedation over a wide range of doses, as observed
previously in unoperated animals.
14,20,28,31
This indicates that
CBD has potential in chronic pain states, although this may be
more as an adjuvant because of its moderate efficacy compared
with THC. It might be also noted that the anti-allodynic efficacy of
Figure 6. Role of cannabinoid receptors in the actions of THC and CBD. Bar charts depicting the effect of the cannabinoids THC (30 mg·kg
21
), CBD (30 mg·kg
21
),
and THC plus CBD (THC:CBD, 0.4 and 40 mg·kg
21
), or vehicle on (A) mechanical PWT (B) acetone responses, (C) rotarod latency, (D) bar latency, and (E) open
field crossings. Cannabinoids/vehicle were administered either alone (Control) or with the cannabinoid CB1 or CB2 receptor antagonists AM281 and AM630 (3
mg·kg
21
). With the exception of open field (for which raw data is shown), all data are displayed as %MPE. *, **, and **** denote P,0.05, 0.01, and 0.0001,
respectively, for each cannabinoid (THC, CBD, and THC:CBD) plus antagonist vs the cannabinoid without antagonist; #, ###, and #### denote P,0.05, 0.01,
0.0001 for each cannabinoid (alone or with antagonist) vs vehicle. CBD, cannabidiol; MPE, maximum possible effect; PWT, paw withdrawal threshold; THC, D9-
tetrahydrocannabinol.
2458 S.L. Casey et al.·158 (2017) 2452–2460 PAIN
®
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

CBD in neuropathic pain models is improved with repeated
administration.
6
The major finding of this study was that combination THC:CBD
treatment, given in a dose ratio similar to that used in clinical
trials,
36
synergistically reduced mechanical and cold allodynia in
a nerve injury–induced neuropathic pain model. This is similar to
a recent chemotherapy-induced neuropathic pain model.
18
It
differs, however, from previous studies on acute nociceptive
pain,
4,10,26,31,35
suggesting that THC:CBD synergy may be
restricted to chronic pain states. Isobolographic analysis dem-
onstrated that this synergistic interaction had a complex, biphasic
nature. First, there was a high potency THC:CBD mode of action.
Thus, at low doses the THC:CBD combination displayed
a 200-fold increase in anti-allodynic potency compared with that
predicted for a simple additive interaction. This synergistic
increase in potency was far greater that that previously observed
between cannabinoid and opioid agonists.
17,37
Although the
reduction in allodynia at low doses of the THC:CBD combination
only displayed partial efficacy, it was not associated with
cannabinoid side effects. Second, there was a low potency
THC:CBD mode of action. Thus, at higher doses the THC:CBD
combination reduced allodynia with higher efficacy, but reduced
potency compared with that predicted for an additive interaction.
In addition, the high dose THC:CBD anti-allodynia was associ-
ated with motor disruption, catalepsy, and sedation, similar to
that observed for THC when administered alone, which differs
from the THC:CBD potentiation of hypo-locomotion observed in
a recent study.
4
These observations indicate that low and high
dose THC:CBD combination treatment have different benefits
and risks. At low doses, combination THC:CBD has a greatly
improved TW at the expense of low anti-allodynic efficacy,
whereas high doses have higher anti-allodynic efficacy at the
expense of a reduced TW when compared with THC alone. It
might be noted that cannabis produces a number of other
problematic side effects, such as dry mouth, nausea, euphoria,
and anxiety,
36
which would have to be examined in future studies.
The low and high dose modes of action of the THC:CBD
combination differed in the involvement of cannabinoid receptors.
The high dose THC:CBD induced anti-allodynia and side effects
displayed a cannabinoid receptor profile which was similar to that
of THC alone, with both being mediated by cannabinoid CB1, and
to a lesser extent CB2 receptors. Although the pharmacological
profile of THC has not been examined previously in a neuropathic
pain model, our observations are consistent with the profile of
THC in acute pain and side effect assays.
22
By contrast,
cannabinoid receptors made a lesser contribution to the
CBD-induced anti-allodynia, which is consistent with previous
neuropathic pain studies.
5,6,33
Importantly, cannabinoid recep-
tors also made little contribution to the low dose (high potency)
THC:CBD induced allodynia. Although the receptors mediating
the low dose THC:CBD induced anti-allodynia were not de-
termined, it could be due to an interaction at multiple targets and
sites given the substantial synergy between these phytocanna-
binoids. These targets could include transient receptor potential
and voltage-gated Ca
21
channels, plus glycine, peroxisome
proliferator–activated receptor, 5-hydroxytryptamine receptors,
and orphan receptors, all of which have a role in neuropathic pain
and are known cannabinoids targets.
5,6,15,23,25,32,33,38
In summary, this study has demonstrated that CBD synergis-
tically enhances the anti-allodynic actions of THC in a neuropathic
pain model. Combination THC:CBD treatment had a complex,
biphasic response mode of action which proffers 2 divergent
medication regimes. High dose THC:CBD combination treatment
offers marginally higher pain-relieving efficacy than THC alone,
but at the expense of a reduced TW. Alternatively, low dose THC:
CBD treatment had the advantage of a greatly improved TW at
the expense of lower pain-relieving efficacy when compared with
THC. Although low dose THC:CBD combination is unlikely to be
used as a stand-alone first-line mediation, these findings highlight
the potential of nabiximols in the treatment of neuropathic pain.
Conflict of interest statement
The authors have no conflicts of interest to declare.
Acknowledgments
Support provided by the Lambert Initiative (University of Sydney).
S. L. Casey was in receipt of an Australian Pain Society and
Australian Pain Relief Association Seqirus scholarship.
S. L. Casey and C. W. Vaughan conceived and designed the
study. S. L. Casey and N. Atwal conducted the study. S. L. Casey,
N. Atwal, and C. W. Vaughan analysed the data. S. L. Casey and
C. W. Vaughan wrote the manuscript. S. L. Casey, N. Atwal, and
C. W. Vaughan checked and approved the manuscript.
Article history:
Received 7 June 2017
Received in revised form 26 July 2017
Accepted 21 August 2017
Available online 1 September 2017
References
[1] Adamson Barnes NS, Mitchell VA, Kazantzis NP, Vaughan CW. Actions of
the dual FAAH/MAGL inhibitor JZL195 in a murine neuropathic pain
model. Br J Pharmacol 2016;173:77–87.
[2] 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.
[3] Bonin RP, Bories C, De Koninck Y. A simplified up-down method (SUDO)
for measuring mechanical nociception in rodents using von Frey
filaments. Mol Pain 2014;10:26.
[4] Britch SC, Wiley JL, Yu Z, Clowers BH, Craft RM. Cannabidiol-Delta9-
tetrahydrocannabinol interactions on acute pain and locomotor activity.
Drug Alcohol Depend 2017;175:187–97.
[5] Comelli F, Giagnoni G, Bettoni I, Colleoni M, Costa B. Antihyperalgesic
effect of a Cannabis sativa extract in a rat model of neuropathic pain:
mechanisms involved. Phytother Res 2008;22:1017–24.
[6] Costa B, Trovato AE, Comelli F, Giagnoni G, Colleoni M. The non-
psychoactive cannabis constituent cannabidiol is an orally effective
therapeutic agent in rat chronic inflammatory and neuropathic pain. Eur J
Pharmacol 2007;556:75–83.
[7] De Vry J, Kuhl E, Franken-Kunkel P, Eckel G. Pharmacological
characterization of the chronic constriction injury model of neuropathic
pain. Eur J Pharmacol 2004;491:137–48.
[8] Deng L, Guindon J, Cornett BL, Makriyannis A, Mackie K, Hohmann AG.
Chronic cannabinoid receptor 2 activation reverses paclitaxel neuropathy
without tolerance or cannabinoid receptor 1-dependent withdrawal. Biol
Psychiatry 2015;77:475–87.
[9] Dworkin RH, O’Connor AB, Audette J, Baron R, Gourlay GK, Haanpaa
ML, Kent JL, Krane EJ, Lebel AA, Levy RM, Mackey SC, Mayer J,
Miaskowski C, Raja SN, Rice AS, Schmader KE, Stacey B, Stanos S,
Treede RD, Turk DC, Walco GA, Wells CD. Recommendations for the
pharmacological management of neuropathic pain: an overview and
literature update. Mayo Clin Proc 2010;85:S3–14.
[10] Finn DP, Beckett SR, Roe CH, Madjd A, Fone KC, Kendall DA, Marsden CA,
Chapman V. Effects of coadministration of cannabinoids and morphine on
nociceptive behaviour, brain monoamines and HPA axis activity in a rat
model of persistent pain. Eur J Neurosci 2004;19:678–86.
[11] Finnerup NB, Attal N, Haroutounian S, McNicol E, Baron R, Dworkin RH,
Gilron I, Haanpaa M, Hansson P, Jensen TS, Kamerman PR, Lund K,
Moore A, Raja SN, Rice AS, Rowbotham M, Sena E, Siddall P, Smith BH,
Wallace M. Pharmacotherapy for neuropathic pain in adults: a systematic
review and meta-analysis. Lancet Neurol 2015;14:162–73.
December 2017·Volume 158 ·Number 12 www.painjournalonline.com 2459
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.

[12] Fox A, Kesingland A, Gentry C, McNair K, Patel S, Urban L, James I. The
role of central and peripheral cannabinoid1 receptors in the
antihyperalgesic activity of cannabinoids in a model of neuropathic
pain. PAIN 2001;92:91–100.
[13] Harris HM, Sufka KJ, Gul W, ElSohly MA. Effects of delta-9-
tetrahydrocannabinol and cannabidiol on cisplatin-induced neuropathy
in mice. Planta Med 2016;82:1169–72.
[14] Hayakawa K, Mishima K, Hazekawa M, Sano K, Irie K, Orito K, Egawa T,
Kitamura Y, Uchida N, Nishimura R, Egashira N, Iwasaki K, Fujiwara M.
Cannabidiol potentiates pharmacological effects of Delta(9)-
tetrahydrocannabinol via CB(1) receptor-dependent mechanism. Brain
Res 2008;1188:157–64.
[15] Hejazi N, Zhou C, Oz M, Sun H, Ye JH, Zhang L. Delta9-
tetrahydrocannabinol and endogenous cannabinoid anandamide
directly potentiate the function of glycine receptors. Mol Pharmacol
2006;69:991–7.
[16] Jensen TS, Baron R, Haanpaa M, Kalso E, Loeser JD, Rice AS, Treede
RD. A new definition of neuropathic pain. PAIN 2011;152:2204–5.
[17] Kazantzis NP, Casey SL, Seow PW, Mitchell VA, Vaughan CW. Opioid
and cannabinoid synergy in a mouse neuropathic pain model. Br J
Pharmacol 2016;173:2521–31.
[18] King KM, Myers AM, Soroka-Monzo AJ, TumaRF, Tallarida RJ, WalkerEA,
Ward SJ. Single and combined effects of delta9 -tetrahydrocannabinol
and cannabidiol in a mouse model of chemotherapy-induced neuropathic
pain. Br J Pharmacol 2017;174:2832–41.
[19] Ledent C, Valverde O, Cossu G, Petitet F, Aubert JF, Beslot F, Bohme
GA, Imperato A, Pedrazzini T, Roques BP, Vassart G, Fratta W,
Parmentier M. Unresponsiveness to cannabinoids and reduced
addictive effects of opiates in CB1 receptor knockout mice. Science
1999;283:401–4.
[20] Long LE, Chesworth R, Huang XF, McGregor IS, Arnold JC, Karl T. A
behavioural comparison of acute and chronic Delta9-
tetrahydrocannabinol and cannabidiol in C57BL/6JArc mice. Int J
Neuropsychopharmacol 2010;13:861–76.
[21] Mechoulam R, Hanus LO, Pertwee R, Howlett AC. Early
phytocannabinoid chemistry to endocannabinoids and beyond. Nat
Rev Neurosci 2014;15:757–64.
[22] Rahn EJ, Hohmann AG. Cannabinoids as pharmacotherapies for
neuropathic pain: from the bench to the bedside. Neurotherapeutics
2009;6:713–37.
[23] Ross HR, Napier I, Connor M. Inhibition of recombinant human T-type
calcium channels by Delta9-tetrahydrocannabinol and cannabidiol. J Biol
Chem 2008;283:16124–34.
[24] Russo E, Guy GW. A tale of two cannabinoids: the therapeutic rationale
for combining tetrahydrocannabinol and cannabidiol. Med Hypotheses
2006;66:234–46.
[25] Ryberg E, Larsson N, Sjogren S, Hjorth S, Hermansson NO, Leonova J,
Elebring T, Nilsson K, Drmota T, Greasley PJ. The orphan receptor
GPR55 is a novel cannabinoid receptor. Br J Pharmacol 2007;152:
1092–101.
[26] Sanders J, Jackson DM, Starmer GA. Interactions among the
cannabinoids in the antagonism of the abdominal constriction response
in the mouse. Psychopharmacology (Berl) 1979;61:281–5.
[27] Scott DA, Wright CE, Angus JA. Evidence that CB-1 and CB-2
cannabinoid receptors mediate antinociception in neuropathic pain in
the rat. PAIN 2004;109:124–31.
[28] Taffe MA, Creehan KM, Vandewater SA. Cannabidiol fails to reverse
hypothermia or locomotor suppression induced by Delta(9)-
tetrahydrocannabinol in Sprague-Dawley rats. Br J Pharmacol 2015;
172:1783–91.
[29] Tallarida RJ. An overview of drug combination analysis with
isobolograms. J Pharmacol Exp Ther 2006;319:1–7.
[30] Toth CC, Jedrzejewski NM, Ellis CL, Frey WH II. Cannabinoid-mediated
modulation of neuropathic pain and microglial accumulation in a model of
murine type I diabetic peripheral neuropathic pain. Mol Pain 2010;6:16.
[31] Varvel SA, Wiley JL, Yang R, Bridgen DT, Long K, Lichtman AH, Martin
BR. Interactions between THC and cannabidiol in mouse models of
cannabinoid activity. Psychopharmacology (Berl) 2006;186:226–34.
[32] Wallace VC, Segerdahl AR, Lambert DM, Vandevoorde S, Blackbeard J,
Pheby T, Hasnie F, Rice AS. The effect of the palmitoylethanolamide
analogue, palmitoylallylamide (L-29) on pain behaviour in rodent models
of neuropathy. Br J Pharmacol 2007;151:1117–28.
[33] Ward SJ, McAllister SD, Kawamura R, Murase R, Neelakantan H, Walker
EA. Cannabidiol inhibits paclitaxel-induced neuropathic pain through 5-
HT(1A) receptors without diminishing nervous system function or
chemotherapy efficacy. Br J Pharmacol 2014;171:636–45.
[34] Ward SJ, Ramirez MD, Neelakantan H, Walker EA. Cannabidiol prevents
the development of cold and mechanical allodynia in paclitaxel-treated
female C57Bl6 mice. Anesth Analg 2011;113:947–50.
[35] Welburn PJ, Starmer GA, Chesher GB, Jackson DM. Effect of
cannabinoids on the abdominal constriction response in mice: within
cannabinoid interactions. Psychopharmacologia 1976;46:83–5.
[36] Whiting PF, Wolff RF, Deshpande S, Di Nisio M, Duffy S, Hernandez AV,
Keurentjes JC, Lang S, Misso K, Ryder S, Schmidlkofer S, Westwood M,
Kleijnen J. Cannabinoids for medical use: a systematic review and meta-
analysis. JAMA 2015;313:2456–73.
[37] Williams J, Haller VL, Stevens DL, Welch SP. Decreased basal
endogenous opioid levels in diabetic rodents: effects on morphine and
delta-9-tetrahydrocannabinoid-induced antinociception. Eur J
Pharmacol 2008;584:78–86.
[38] Xiong W, Cheng K, Cui T, Godlewski G, Rice KC, Xu Y, Zhang L.
Cannabinoid potentiation of glycine receptors contributes to cannabis-
induced analgesia. Nat Chem Biol 2011;7:296–303.
[39] Zimmer A, Zimmer AM, Hohmann AG, Herkenham M, Bonner TI.
Increased mortality, hypoactivity, and hypoalgesia in cannabinoid CB1
receptor knockout mice. Proc Natl Acad Sci U S A 1999;96:5780–5.
2460 S.L. Casey et al.·158 (2017) 2452–2460 PAIN
®
Copyright Ó2017 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.