PAINÒ 145 (2009) 304–311

www.elsevier.com/locate/pain

Ketamine produces effective and long-term pain relief in patients

with Complex Regional Pain Syndrome Type 1
Marnix J. Sigtermans a,1, Jacobus J. van Hilten b,1, Martin C.R. Bauer a, M. Sesmu Arbous c, Johan Marinus b,
Elise Y. Sarton a, Albert Dahan a,*
a
Department of Anesthesiology, Leiden University Medical Center, P.O. Box 9600, 2300 RC Leiden, The Netherlands
b
Department of Neurology, Leiden University Medical Center, P.O. Box 9600, 2300 RC Leiden, The Netherlands
c
Department of Intensive Care, Leiden University Medical Center, P.O. Box 9600, 2300 RC Leiden, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: Complex Regional Pain Syndrome Type 1 (CRPS-1) responds poorly to standard pain treatment. We
Received 1 April 2009 evaluated if the N-methyl-D-aspartate receptor antagonist S(+)-ketamine improves pain in CRPS-1
Received in revised form 20 May 2009 patients. Sixty CRPS-1 patients (48 females) with severe pain participated in a double-blind randomized
Accepted 18 June 2009
placebo-controlled parallel-group trial. Patients were given a 4.2-day intravenous infusion of low-dose
ketamine (n = 30) or placebo (n = 30) using an individualized stepwise tailoring of dosage based on effect
(pain relief) and side effects (nausea/vomiting/psychomimetic effects). The primary outcome of the study
Keywords:
was the pain score (numerical rating score: 0–10) during the 12-week study period. The median (range)
CRPS
Ketamine
disease duration of the patients was 7.4 (0.1–31.9) years. At the end of infusion, the ketamine dose was
NMDA receptor 22.2 ± 2.0 mg/h/70 kg. Pain scores over the 12-week study period in patients receiving ketamine were
Disease modulation significantly lower than those in patients receiving placebo (P < 0.001). The lowest pain score was at
the end of week 1: ketamine 2.68 ± 0.51, placebo 5.45 ± 0.48. In week 12, significance in pain relief
between groups was lost (P = 0.07). Treatment did not cause functional improvement. Patients receiving
ketamine more often experienced mild to moderate psychomimetic side effects during drug infusion
(76% versus 18%, P < 0.001). In conclusion, in a population of mostly chronic CRPS-1 patients with severe
pain at baseline, a multiple day ketamine infusion resulted in significant pain relief without functional
improvement. Treatment with ketamine was safe with psychomimetic side effects that were acceptable
to most patients.
Ó 2009 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

1. Introduction ror approach. There is no evidence that commonly used treatments

with opioids, antidepressants, antiepileptics, and sympathetic
Complex Regional Pain Syndrome Type 1 (CRPS-1) is a chronic blockade are effective in CRPS [35]. Dimethylsulfoxide crème, N-
pain syndrome typically affecting an extremity after a local trauma acetylcysteine, and physiotherapy reduce symptoms of CRPS, but
or surgical intervention [41]. The initial phase of the syndrome is their effect is often limited [26,30,35]. Spinal cord stimulation is
characterized by pain, edema, changes in skin temperature and used as a last resort and is effective in the management of chronic
color, and hyperhydrosis [41]. Although the recovery rate of CRPS pain in CRPS-1, although its efficacy tends to decline over the years
is unknown, a substantial number of patients develop chronic dis- [12].
ease with severe pain, disability, and loss of quality of life [7]. In Recent studies implicate the N-methyl-D-aspartic acid receptor
the Netherlands the incidence of CRPS-1 is 26 per 100,000 person (NMDAR) in the etiology and perseverance of chronic pain. In
years, with predominance in women [6]. At present, the patho- chronic pain states the NMDAR is activated and upregulated in
physiology of CRPS-1 remains largely unknown. In contrast to neu- the spinal cord (central sensitization) [36,44]. This results in en-
ropathic chronic pain syndromes there is no proof of a clinically hanced signal transmission in the pain circuitry from the spinal
evident nerve lesion as a causative factor in CRPS-1 [1]. cord to the cortex leading to spontaneous pain, allodynia (pain per-
As in the treatment of most chronic pain syndromes, the com- ception from a non-noxious stimulus) and hyperalgesia (increased
mon strategy in managing CRPS-1 is characterized by a trial and er- pain sensitivity). Considering the involvement of the NMDAR in
chronic pain, antagonists of the NMDAR may play an important
* Corresponding author. Tel.: +31 71 526 2301; fax: +31 71 526 4824.
role in chronic pain treatment [34]. One such NMDAR antagonist
E-mail address: [email protected] (A. Dahan). is the intravenous anesthetic agent ketamine, which is currently
1
These authors contributed equally to this work. the only potent NMDAR antagonist clinically available. A major

0304-3959/$36.00 Ó 2009 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.pain.2009.06.023
 M.J. Sigtermans et al. / PAINÒ 145 (2009) 304–311 305

issue with the use of ketamine is the development of psychomi- (iv) Exclusion of other conditions that would otherwise account
metic side effects such as hallucinations and drug high [34]. This for the degree of pain and dysfunction.
limits its use, especially when used at high dose. However, several
open-label case studies in which ketamine treatment was used in Exclusion criteria were: pain score of less than 5 of 10,
CRPS-1 patients showed a benefit in pain reduction with an accept- age < 18 years, pregnancy/lactation, increased intracranial pres-
able side effect profile [5,9,15,16,38]. Furthermore, few random- sure, a history of psychosis, a serious medical disease (e.g., cardio-
ized clinical trials of long-term ketamine administration in vascular, renal or liver disease) and use of strong opioid
cancer pain patients with and without neuropathic pain, demon- medication. Patients were allowed to continue the following pain
strated the efficacy and safety of ketamine infusion [2,22]. medications: paracetamol, non-steroidal anti-inflammatory drugs,
In this study we evaluate the short-term and long-term efficacy tramadol, amitryptilin, selective serotonin reuptake-inhibitors,
of prolonged ketamine administration on pain in patients with gabapentin and pregabalin. The pain medication was kept constant
CRPS-1. during the 12-week study period. Patients’ consent was obtained
according to the Declaration of Helsinki and the study was ap-
proved by The Medical Ethics Committee of the Leiden University
2. Methods Medical Center.

This study is registered in the Netherlands Trial Register

2.2. Treatment
(www.trialregister.nl) under No. NTR507 (ISRCTN 30472359).
Patients were randomly allocated to receive S(+)-ketamine
2.1. Patients (Ketanest S, Pfizer BV, Capelle aan de IJssel, The Netherlands) or
placebo (normal saline). A physician otherwise not involved in
Patients referred to our outpatient pain clinic between January the study performed randomization (using a computer-generated
2006 and January 2008, and who were diagnosed with CRPS-1, list) and prepared the syringes. Patients were admitted to a short
were eligible for inclusion in the study. The diagnosis of CRPS-1 stay ward for 5 days, and they received two iv lines on the morning
was based on the International Association for the Study of Pain of admission (day 1), one for drug infusion, and the other for blood
criteria [4,23]: sampling. The drug infusion rate started at 1.2 lg kg 1 min 1 (or
5 mg/h for a 70-kg patient) at 8 AM on day 1 and was titrated at
(i) The presence of an initiating noxious event, or a cause of regular intervals (max. thrice daily) to a maximum of
immobilization. 7.2 lg kg 1 min 1 (or 30 mg/h for a 70-kg patient). The infusion
(ii) Continuing pain, allodynia or hyperalgesia, with the pain rate was increased when pain relief was insufficient (based on re-
being disproportionate to any inciting event. ported visual analogue pain scores reported at 2 h (day)–8 h
(iii) Evidence at some time of edema, changes in skin blood flow (night) intervals) and side effects were acceptable to the patients.
and/or abnormal sudomotor activity in the region of pain. If side effects were unacceptable but pain relief insufficient, the

Fig. 1. Patients’ flowchart. Eighty-seven patients were considered for participation in the study; 16 were excluded because they did not fulfill the study criteria, while 11
others declined to participate, after which 60 patients underwent randomization. A few patients were lost for determination of secondary end-points but not for primary end-
points.
306 M.J. Sigtermans et al. / PAINÒ 145 (2009) 304–311

infusion rate was decreased one step for one interval and subse- pain) to 10 (unbearable pain) at baseline and weekly until week
quently increased again. In case of full pain relief, the infusion rate 12. Secondary outcomes were the course of parameters that
was not increased further and was kept constant until the end of were measured at baseline and weeks 1, 3, 6 and 12: (i) ability
the treatment period. The treatment ended on day 5 around noon. to use the affected limb in normal day-to-day activity assessed
Blood sampling was performed at regular intervals during drug using the Radboud Skills Questionnaire (RASQ) and the Walking
infusion and was continued up to 10 h after the infusion had Ability Questionnaire (WAQ) for upper and lower limbs, respec-
ended. After unblinding of the study, patients who had received tively [24,29]; (ii) active range of motion (AROM) assessed by
placebo were given the opportunity to receive an additional keta- goniometric measurements [25,28]; (iii) threshold for touch as-
mine treatment in an open fashion. sessed using Semmes–Weinstein monofilaments on the affected
and contralateral limbs [33]; (iv) skin temperature of the af-
2.3. Measurements fected and contralateral limbs measured by infrared thermome-
try [25,28]; (v) volumetric measurements of the affected and
Measurements were performed in the week prior to the contralateral limbs using the water displacement method
treatment week (baseline), during treatment (week 1) and in [25,28].
the 11-week follow-up period. The primary outcome of the During the treatment week liver functions (once daily) and
study was the course of spontaneous pain reported by the pa- blood pressure (thrice daily) were measured and the occurrence
tients during the 12-week study period. Pain scores were as- of nausea/vomiting and psychomimetic side effects (drug high
sessed by a numerical rating scale (NRS) ranging from 0 (no and hallucinations) was continuously monitored.

Table 1
Patients’ characteristics.
All patients Ketamine Placebo P-value
N (number of females) 60 (48) 30 (22) 30 (26) 0.33*
Age (year) 45.6 ± 12.4 43.7 ± 11.5 47.5 ± 13.1 0.23
Weight (kg) 79.1 ± 18.8 78.9 ± 17.8 79.3 ± 20.1 0.94
Height (cm) 171.9 ± 9.9 173.6 ± 9.9 170.1 ± 9.7 0.17
BMI (kg m 2) 26.7 ± 5.8 26.1 ± 5.4 27.3 ± 6.3 0.43
Disease duration (year) 7.8 ± 6.8 9.4 ± 8.0 6.1 ± 5.1 0.06
Number of extremities affected 0.29*
One 36 15 21
Two 16 11 5
Three 3 1 2
Four 5 3 2
First affected 0.61*
Lower extremity 22 11 11
Upper extremity 23 10 13
Both extremities 15 9 6
Pain at baseline (NRS) 7.0 ± 1.3 7.2 ± 1.2 6.9 ± 1.4 0.32
McGill [14]
PRI 27.0 ± 11.7 28.6 ± 10.4 25.1 ± 12.9 0.030
NWC 13.1 ± 4.5 13.1 ± 4.0 13.0 ± 5.1 0.93
RASQ (upper extremity) [24,29] 3.3 ± 0.9 3.4 ± 0.7 3.3 ± 1.2 0.95
Walking Ability Questionnaire [24,29]
Walking inside the house 5.9 ± 2.7 6.1 ± 2.5 5.6 ± 3.0 0.57
Walking outside 6.8 ± 2.6 7.1 ± 2.6 6.5 ± 2.7 0.58
Standing up 7.7 ± 3.0 7.1 ± 3.3 7.2 ± 2.7 0.91
TSK [42] 38.8 ± 10.9 38.0 ± 6.8 39.6 ± 14.2 0.63
Euroqol Health Valuation (VAS) [13] 54.1 ± 19.9 55.0 ± 18.8 53.1 ± 21.4 0.78
HADS [45]
Depression 3.8 ± 2.7 4.7 ± 3.1 2.8 ± 1.8 0.02
Anxiety 5.3 ± 2.9 6.1 ± 3.1 4.4 ± 2.4 0.04
PCI [40] 73.0 ± 11.7 74.1 ± 9.8 71.7 ± 13.5 0.50
Short-form 36 [3]
SF-36 PH 31.4 ± 19.2 30.0 ± 19.2 33.0 ± 19.6 0.60
SF-36 MHS 58.6 ± 19.0 51.8 ± 20.8 65.9 ± 13.7 0.009
CIRS co-morbidity [19]
Number affected organ systems 3.8 ± 3.0 4.1 ± 3.6 3.5 ± 2.3 0.47
Severity index 1.2 ± 0.5 1.2 ± 0.5 1.21 ± 0.5 0.90
Pain co-medication
Paracetamol 15 6 9 0.55*
NSAIDs 14 9 5 0.36*
Benzodiazepines 19 10 9 0.10*
Tramadol hydrochloride 15 6 9 0.55*
TCA 10 5 5 1.00*
SSRI 6 3 3 1.00*
Gabapentin/pregabalin 9 7 2 0.26*

BSL, baseline; BMI, body mass index; PRI, McGill pain rating index; NWC, McGill total number of words chosen; RASQ, Radboud Skills Questionnaire; TSK, Tampa Scale of
Kinesiophobia; HADS, Hospital Anxiety and Depression Scale; PCI, Pain Coping Inventory; SF-36 PH, short-form 36 physical health sumscore; SF-36 MHS, short-form 36
mental health status sumscore; CIRS, Cumulative Illness Rating Scale (co-morbidity evaluation); NSAIDs, non-steroidal anti-inflammatory drugs; TCAs, tricyclic antide-
pressants; SSRI, selective serotonin reuptake-inhibitor.
*
Group differences are tested with a Student’s t-test or v2-test; values are mean ± SD. Significant differences are printed in bold.
 M.J. Sigtermans et al. / PAINÒ 145 (2009) 304–311 307

Fig. 2. Mean infusion rate and plasma concentrations of ketamine and its active metabolite norketamine during and 10 h after the ketamine infusion. Ketamine and
norketamine concentrations rapidly declined upon the termination of ketamine infusion. Values are means ± SEM.

2.4. Sample size and data analysis and SF-36, Table 1). However, these differences were either well be-
low the cut-off point for psychopathology (HADS; cut-off = 8) or
This study was powered to detect a two-point reduction in NRS small (SF-36). Furthermore, no effect of the difference in these base-
pain for the ketamine group versus placebo group. Based on a line parameters was observed on the outcome in the univariate
power of 0.80, a of 0.05 (two-sided) and a standard deviation of regression analysis. The medical history did not reveal a significant
2.5, we calculated that 25 patients were needed per group. Assum- difference in prevalence of psychiatric diseases (CIRS, Table 1) A few
ing that some patients would be lost in follow-up, we planned to patients were lost for determination of secondary end-points but
enroll 30 patients per group. Statistical analysis included all pa- not for primary end-points (Fig. 1).
tients, according to the intention-to-treat principle.
A linear mixed model with an autoregressive correlation struc- 3.2. Treatment procedure
ture was used to determine the effect of ketamine versus placebo
during the 12-week study period on the course of spontaneous Drug infusion was carried out as planned in 58/60 patients. Due to
pain (primary end-point) and other (secondary) end-points. To a severe feeling of high, drug infusion was terminated in two patients
determine the effect of treatment on spontaneous pain at the on day 3 and day 4, respectively. These patients were monitored
end of the study a univariate linear regression was used to eval- throughout the complete follow-up period and were included in the
uate which baseline characteristics influenced pain. Next, in a data analysis. The mean ketamine infusion rate was 22.2 ± 2.0 mg/h
multivariable linear regression model, treatment, all variables (normalized to a 70-kg patient) at the end of the treatment phase
with a P-value <0.10 in the univariate analysis, and variables (Fig. 2). A steady state in the plasma concentration for ketamine and
determined clinically relevant (disease duration, pain at baseline, its metabolite norketamine was reached at the end of the treatment
sex and age) were included to determine the effect of treatment period after which the concentrations showed a brisk decline.
on pain at the end of the study period, while controlling for other
factors influencing pain perception. Since it is conceivable that
3.3. Primary and secondary outcomes
some covariates may modulate the effect of others, interaction
terms of clinically relevant combinations were tested for signifi-
Mean (SD) baseline pain scores were 7.20 (1.16) for ketamine
cance. Statistical analyses were performed using SPSS 16.0 (Chi-
and 6.87 (1.43) for placebo (P = 0.32). The lowest NRS scores were
cago, IL). P-values <0.05 were considered significant. Data are
observed at the end of week 1 (ketamine 2.68 ± 0.51 and placebo
presented as mean ± standard error of the mean (SEM) unless
5.45 ± 0.48). Ketamine modulated the course of pain during the
otherwise stated.
12-week study period more favorably than placebo (P < 0.001,
3. Results Fig. 3A). Significant differences in pain reduction between keta-
mine and placebo were maintained till week 11; at week 12, keta-
3.1. Patients mine’s treatment effect lost significance (P = 0.07). Compared to
placebo treatment, none of the secondary outcome measures im-
Sixty patients underwent randomization (see Fig. 1 for patients’ proved significantly on ketamine treatment (Table 2). Univariate
flowchart and Table 1 for patients’ characteristics). The eleven pa- analysis showed that baseline pain (P = 0.004) and McGill pain rat-
tients who refused to participate did not differ in characteristics ing index (P = 0.042) predicted pain at week 12. Multivariate anal-
from the patients who underwent randomization. The majority of ysis revealed that the McGill pain rating index was the only
patients were female (80%) and the median (range) disease dura- significant determinant (B = 0.07, 95% CI: 0.01–0.13, P = 0.027) of
tion was 7.4 (0.1–31.9) years. Treatment groups differed in depres- NRS pain at week 12. Particularly, the ketamine effect was inde-
sion and anxiety scores, and in health status measurements (HADS pendent of disease duration, which ranged from 0.1 to 31.9 years.
308 M.J. Sigtermans et al. / PAINÒ 145 (2009) 304–311

Fig. 3. (A) Weekly pain scores observed in the ketamine and placebo treatment groups (n = 30 in each group). S(+)-ketamine treatment had a significantly more favorable
effect on pain scores than placebo treatment with a main effect of P < 0.001. (B) Twenty patients initially treated with placebo took the opportunity to receive ketamine in an
open fashion after unblinding of the data. These patients had a similar reduction in NRS pain in week 1 as observed after blinded ketamine treatment. However, the
subsequent course of NRS pain towards baseline was slower compared to the blinded study with a 1–2 NRS difference. For comparative reasons the blinded ketamine data are
included (n = 30). Values are means ± SEM.

3.4. Side effects assumptions were correct in 26 patients receiving ketamine and 27
patients receiving placebo (j = 0.77, 95% CI: 0.60–0.93, P < 0.001).
During drug infusion most patients experienced side effects,
and, with the exception of headache, the incidence was greater 3.6. Open treatment
in subjects receiving ketamine: nausea 63% versus 17% in placebo
group (P < 0.001, v2-test), vomiting 47% versus 10% in placebo Despite an increased analgesic effect during the 12-week treat-
group (P = 0.004), psychomimetic effects 93% versus 17% in placebo ment course (Fig. 3B), open ketamine treatment was not associated
group (P < 0.001) and headache 37% versus 33% in placebo group with an increased incidence of side effects compared to blinded
(P = 0.78). Liver functions and blood pressure remained unaffected treatment: nausea 71%, vomiting 48%, psychomimetic effects 76%
by ketamine administration. and headache 43%.

3.5. Blinding
4. Discussion
At the end of infusion, 28 patients receiving ketamine correctly
indicated treatment assignment against 18 patients receiving pla- Ketamine has multiple sites of action but its effect on the
cebo (j = 0.53, 95% CI: 0.33–0.74, P < 0.001). The investigator’s NMDAR is generally considered to be the basis for its modulatory
 M.J. Sigtermans et al. / PAINÒ 145 (2009) 304–311 309

Table 2
Effect of ketamine treatment on secondary outcome parameters.
Baseline Week 1 Week 3 Week 6 Week 12 P-value
Affected upper extremity
Sensibilitya (g/mm2) Ketamine 2.7 ± 0.2 2.8 ± 0.2 2.8 ± 0.1 2.7 ± 0.2 2.7 ± 0.2 0.92
Placebo 2.4 ± 0.3 2.6 ± 0.3 2.5 ± 0.2 2.6 ± 0.3 3.1 ± 0.5
Temperatureb (oC) Ketamine 0.8 ± 0.3 0.7 ± 0.2 0.4 ± 0.1 1.9 ± 0.8 1.0 ± 0.5 0.12
Placebo 1.2 ± 0.7 0.7 ± 0.2 1.0 ± 0.3 0.6 ± 0.2 0.6 ± 0.2
c 3
Volume (cm ) Ketamine 35 ± 7 54 ± 13 39 ± 13 38 ± 9 28 ± 7 0.34
Placebo 53 ± 11 38 ± 12 41 ± 15 41 ± 5 48 ± 8
d
AROM (%) Ketamine 46 ± 9 30 ± 8 33 ± 7 32 ± 10 37 ± 11 0.48
Placebo 34 ± 7 29 ± 9 35 ± 9 38 ± 7 35 ± 8
RASQe Ketamine 3.3 ± 0.2 * 2.8 ± 0.3 2.8 ± 0.2 3.3 ± 0.2 0.51
Placebo 3.5 ± 0.3 * 3.1 ± 0.3 3.4 ± 0.3 3.3 ± 0.3
Affected lower extremity
Sensibilitya (g/mm2) Ketamine 4.0 ± 0.2 3.8 ± 0.2 3.6 ± 0.1 4.0 ± 0.2 4.0 ± 0.2 0.11
Placebo 3.7 ± 0.1 3.7 ± 0.2 3.6 ± 0.1 3.4 ± 0.2 3.7 ± 0.2
b o
Temperature ( C) Ketamine 1.1 ± 0.3 0.8 ± 0.3 1.3 ± 0.9 1.3 ± 0.7 1.8 ± 0.7 0.79
Placebo 2.9 ± 1.1 1.5 ± 0.4 2.0 ± 0.4 2.3 ± 0.8 3.3 ± 1.5
Volumec (cm3) Ketamine 116 ± 20 146 ± 51 115 ± 55 79 ± 31 55 ± 12 0.40
Placebo 83 ± 22 110 ± 24 78 ± 16 70 ± 19 139 ± 44
AROMd (%) Ketamine 39 ± 12 28 ± 17 48 ± 13 37 ± 17 22 ± 17 0.87
Placebo 28 ± 8 20 ± 9 23 ± 10 24 ± 14 11 ± 12
f
WAQ Ketamine 6.5 ± 0.5 * 5.1 ± 0.8 4.7 ± 1.1 5.8 ± 0.7 0.32
Walking inside Placebo 5.6 ± 0.8 * 5.2 ± 0.9 5.5 ± 0.9 6.1 ± 0.8
WAQf Ketamine 7.4 ± 0.6 * 6.9 ± 1.4 6.4 ± 1.1 7.1 ± 0.9 0.48
Walking outside Placebo 6.6 ± 0.7 * 6.0 ± 0.9 6.9 ± 0.9 6.9 ± 0.8
WAQf Ketamine 7.1 ± 0.8 * 6.7 ± 1.1 6.3 ± 1.4 7.6 ± 0.8 0.35
Standing up Placebo 7.1 ± 0.7 * 6.8 ± 1.0 7.7 ± 1.0 7.0 ± 0.9

Values are mean ± SEM.

a
Sensibility is the threshold for touch assessed by Semmes–Weinstein monofilaments and calculated as the median threshold in g/mm2 of five different locations on the
hands or feet.
b
Temperature is the mean difference in skin temperature of five different locations of the affected side compared to the non-affected side measured by infrared
thermometry.
c
Volume is the volumetric measurement of the limb using the method of water displacement. The limb is immersed in a box of water to a fixed distance from the elbow or
the lateral malleolus. The amount of water displaced corresponds to the volume of the limb measured in cm3. The difference between the affected limb and the non-affected
limb is presented.
d
AROM is the active range of motion assessed by goniometric measurements. The range of motion of the wrist and the two most affected metacarpophalangeal and
proximal interphalangeal joints was measured and compared to that of the unaffected side and averaged to a percentage. For the AROM of the lower extremity the following
joints were measured: the knee, the ankle (dorsiflexion–extension and pronation–supination) and the metatarsophalyngeal joint.
e
The ability to use the upper limb in normal day-to-day activity assessed using the Radboud Skills Questionnaire. RASQ values range from 1 (no disability) to 5 (inability to
use the arm).
f
The ability to use the lower limb in normal day-to-day activity assessed using the Walking Ability Questionnaire (the WAQ consists of three separate measurements
relating to walking inside the house, outside the house and standing up). WAQ values range from 1 (no disability) to 10 (inability to use the leg).
*
RASQ and WAQ were not obtained at the end of the in-hospital treatment week.

effect on pain responses [36,44]. The current observation of a ben- these changes are the primary or secondary phenomena and if they
eficiary effect of ketamine on pain suggests an important role for are related to functional improvement. In our study, the functional
the sensitized NMDAR in the maintenance of pain in CRPS-1. Our status of patients did not improve despite significant pain relief,
findings show that 4 days of ketamine infusion may result in a clin- even in the initial weeks following treatment. Although relief of pain
ically significant pain reduction over a prolonged period of with an NMDAR antagonist is likely to be associated with changes in
10 weeks. Interestingly, the effect of ketamine on acute experi- inhibitory/excitatory state of the nervous system, this may have
mental pain and EEG slowing is directly linked to its plasma con- been insufficient in duration and intensity in our study to allow
centration and disappears rapidly upon termination of infusion for amelioration of the patients’ physical status. More prolonged
[37,39]. The differential findings of ketamine on spontaneous pain treatment possibly in combination with physical therapy or rehabil-
in CRPS-1 may suggest that these effects are obtained by long-term itation strategies may be required to obtain functional benefit.
desensitization of the NMDAR in the central nervous system (for Zarate et al. [43] described a long-lasting antidepressant effect
example at the spinal cord level). Further proof for this could be from ketamine. It may therefore be argued that, at least part, of
found in a reduction of hyperalgesia and allodynia. The current the pain relief observed in our patients was due to an improvement
study, however, provided insufficient information on these mea- of depression- and/or anxiety-related symptoms. However, depres-
sures of NMDAR sensitization. sion and anxiety scores in the majority of patients indicated the ab-
Brain imaging studies indicate various abnormalities encom- sence of clinically relevant symptoms of depression and/or anxiety
passing regions involved in emotional, autonomic, sensorimotor (HADS, Table 1). And although the statistical analysis did not show
and pain processing [8,11,20,21,31,32]. Reorganization of the an effect of initial depression and anxiety scores or mental health
somatosensory cortex contralateral to the CRPS affected side ap- status (SF-36 MHS) on pain relief, we cannot exclude some effect
peared to be linked to the severity of pain [20,31]. Additionally, of ketamine on pain perception occurring via an improvement of
neurophysiological studies revealed motor cortex disinhibition the patients’ mental health.
[11]. In patients who recovered from CRPS the reduction of pain The observation that pain relief dissipated slowly (Fig. 3) may
was associated with regaining of the cortical map size and restora- be related to the relatively short ketamine infusion. Some case
tion of impaired tactile discrimination [21,32]. Together these find- studies allowing a longer duration of infusion show a more pro-
ings suggest that the development and resolution of CRPS are longed effect of ketamine [5]. We restricted the duration of the
associated with cortical changes, but the question remains whether ketamine infusion to 4 days (100 h) in light of the concern from
310 M.J. Sigtermans et al. / PAINÒ 145 (2009) 304–311

animal data that high-dose and long-term ketamine infusion may and genetics on Complex Regional Pain Syndrome Type 1. The con-
be associated with neurotoxicity [5,27]. Further follow-up studies sortium aims to develop concepts on disease mechanisms that oc-
are needed to explore this important issue. The available data in cur in response to tissue injury, its assessment and treatment.
humans suggest that high infusion rates (50 mg/h) for durations TREND is supported by a government grant (BSIK03016).
of up to 7 days are without long-term neurotoxic effects [17,18].
We used the S(+) enantiomer of ketamine in our study, which
has the advantage of a better cardiovascular profile along with a References
possible neuroprotective potential over racemic ketamine [10].
The use of ketamine is often avoided by pain physicians due to [1] Albazaz R, Wong YT, Homer-Vanniasinkam S. Complex regional pain
syndrome: a review. Ann Vasc Surg 2008;22:297–306.
the fear of occurrence of psychomimetic side effects. We observed [2] Bell R, Eccleston C, Kalso E. Ketamine as an adjuvant to opioids for cancer pain.
drug high in about 90% of patients receiving ketamine; hallucina- Cochrane Database Syst Rev 2003: CD003351.
tions did occur in a minority of patients. Patient’s and investiga- [3] Brazier JE, Harper R, Jones NM, O’Cathain A, Thomas KJ, Usherwood T, Westlake
L. Validating the SF-36 health survey questionnaire: new outcome measure for
tor’s guesses of the administered treatment were correct in 74%
primary care. BMJ 1992;305:160–4.
and 88%, respectively. This is most likely explained by ketamine’s [4] Bruehl S, Harden RN, Galer BS, Saltz S, Bertram M, Backonja M, Gayles R,
psychomimetic side effects and suggests that in future trials an Rudin N, Bhugra MK, Stanton-Hicks M. External validation of IASP
diagnostic criteria for complex regional pain syndrome and proposed
active placebo and/or the addition of an active placebo or addi-
research diagnostic criteria. International Association for the Study of Pain.
tion of a benzodiazepine to the ketamine treatment group is war- Pain 1999;81:147–54.
ranted to preclude deblinding of treatment allocation. However, [5] Correll GE, Maleki J, Gracely EJ, Muir JJ, Harbut RE. Subanesthetic ketamine
our experience is such that complete blinding of ketamine treat- infusion therapy: a retrospective analysis of a novel therapeutic approach to
complex regional pain syndrome. Pain Med 2004;5:263–75.
ment is difficult due to the specific nature and coloring of the [6] De Mos M, De Bruijn AG, Huygen FJ, Dieleman JP, Stricker BH, Sturkenboom
psychomimetic effects. Except for two patients in whom treat- MC. The incidence of complex regional pain syndrome: a population-based
ment was terminated prematurely due to psychomimetic side ef- study. Pain 2007;129:12–20.
[7] Galer BS, Henderson J, Perander J. Jensen MP: course of symptoms and quality
fects, most patients considered the intensity of the side effects to of life measurement in complex regional pain syndrome: a pilot survey. J Pain
be mild to moderate and acceptable. The individualized stepwise Symptom Manage 2000;20:286–92.
titration of ketamine according to the pain status and side effects [8] Geha PY, Baliki MN, Herden RN, Bauer WR, Parrish TB, Apkarian AV. The brain
in chronic CRPS pain: abnormal gray-white matter interactions in emotional
allowed for optimizing the level of pain reduction while the and autonomie regions. Neuron 2008;60:570–81.
occurrence of side effects was kept at an acceptable level. Of [9] Goldberg ME, Domsky R, Scaringe D, Hirsch R, Dotson J, Sharaf I, Torjman MC,
interest is the observation that in the open treatment paradigm Schwartzman RJ. Multi-day low dose ketamine infusion for treatment of
complex regional pain syndrome. Pain Phys 2005;8:175–9.
the incidence of side effects did not increase despite a better
[10] Himmelseher S, Durieux ME. Revising a dogma: ketamine for patients with
analgesic course of the treatment. neurological injury? Anesth Analg 2005;101:524–34.
In spite of significant pain relief, ketamine did not improve the [11] Juottonen K, Gockel M, Silén T, Hurri H, Hari R, Forss N. Altered central
sensorimotor processing in patients with complex regional pain syndrome.
functional status of our patients. More prolonged or repetitive ket-
Pain 2002;98:315–23.
amine treatment may be required to induce longer periods of pain [12] Kemler MA, Barendse GAM, van Kleef M, de Vet HCW, Rijks CPM, Furnée CA,
relief that allows patients to increase their activity level, thus van den Wildenberg FAJM. Spinal cord stimulation in patients with chronic
enhancing opportunities for improvement in functional status. It reflex sympathetic dystrophy. N Engl J Med 2000;343:618–24.
[13] Kemler MA, De Vet HC. Health-related quality of life in chronic refractory
is of interest to remark that the observed pain relief from ketamine reflex sympathetic dystrophy (complex regional pain syndrome type I). J Pain
was not dependent on the dose given. Since most patients were al- Symptom Manage 2000;20:68–76.
most pain free during ketamine infusion, it can be suggested that [14] Khan SA, Apkarian AV. The characteristics of cyclical and non-cyclical
mastalgia: a prospective study using a modified McGill Pain Questionnaire.
the cessation of nociceptive input (irrespective of mechanisms) Breast Cancer Res Treat 2002;75:147–57.
could account for the long-term pain relief. Fifty-four of our 60 pa- [15] Kiefer R-T, Rohr P, Ploppa A, Altemeyer K-H, Schwartzman RJ. Complete
tients had chronic CRPS-1 (disease duration longer than 6 months) recovery from intractable complex regional pain syndrome, CRPS-type 1,
following anesthetic ketamine and midazolam. Pain Pract 2007;7:
and disease duration did not predict pain reduction by ketamine. 147–50.
This indicated that the duration of NMDAR sensitization appar- [16] Kiefer R-T, Rohr P, Ploppa A, Dieterich H-J, Grothisen J, Koffler S, Altemeyer K-
ently plays no major role in ketamine’s efficacy in reducing spon- H, Unertl K, Schwartzman RJ. Efficacy of ketamine in anesthetic dosage for the
treatment of refractory complex regional pain syndrome: an open-label phase
taneous pain in CRPS-1. This is encouraging for patients with
II study. Pain Med 2008;9:1173–201.
refractory chronic CRPS-1. [17] Kiefer RT, Rohr P, Ploppa A, Unertl K, Schwartzman RJ. Is high dosed ketamine a
We conclude that CRPS-1 patients with severe pain trea- therapeutic option for severe intractable complex regional pain syndrome?
ted with 4 days of continuous infusion of low-dose ketamine Anesthesiology 2003;99:A1008.
[18] Kiefer RT, Rohr P, Unertl K, Altmeyer KH, Grothusen J, Schwartzman RJ.
using an individualized stepwise tailoring of dosage have a Recovery from intractable complex regional pain syndrome type-I (RSD) under
clinically relevant reduction in pain lasting for 11 weeks. high dose intravenous ketamine–midazolam sedation. Neurology
Although the incidence of side effects in the treatment week 2002;58:A474.
[19] Linn BS, Linn MW, Gurel L. Cumulative illness rating scale. J Am Geriatr Soc
was high, they were relatively mild and well accepted by 1968;16:622–6.
the patients and disappeared upon the termination of infu- [20] Maihöfner C, Handwerker HO, Neundörfer B, Birklein F. Patterns of cortical
sion. In contrast to pain, there was no improvement in func- reorganization in complex regional pain syndrome. Neurology 2003;61. 17-7-
15.
tional status. [21] Maihöfner C, Handwerker HO, Neundörfer B, Birklein F. Cortical reorganization
during recovery from complex regional pain syndrome. Neurology
2004;63:693–701.
Conflict of interest
[22] Mercadante S, Arcuri E, Tirelli W, Casuccio A. Analgesic effect of intravenous
ketamine in cancer patients on morphine therapy: a randomized, controlled,
The authors of this study report no conflict of interest. double-blind, crossover, double-dose study. J Pain Symptom Manage
2000;20:246–52.
[23] Merskey H, Bogduk K. Classification of chronic pain: definitions of chronic pain
syndromes and definition of pain terms. Seattle: IASP Press; 1994.
Acknowledgments [24] Oerlemans HM, Cup EH, DeBoo T, Goris RJ, Oostendorp RA. The Radboud skills
questionnaire: construction and reliability in patients with reflex sympathetic
This study is part of TREND (Trauma RElated Neuronal Dysfunc- dystrophy of one upper extremity. Disabil Rehabil 2000;22:233–45.
[25] Oerlemans HM, Goris RJ, Oostendorp RA. Impairment level sumscore in reflex
tion), a Dutch Consortium that integrates research on epidemiol- sympathetic dystrophy of one upper extremity. Arch Phys Med Rehabil
ogy, assessment technology, pharmacotherapeutics, biomarkers 1998;79:979–90.
 M.J. Sigtermans et al. / PAINÒ 145 (2009) 304–311 311

[26] Oerlemans HM, Oostendorp RA, de Boo T, Goris RJ. Pain and reduced mobility [35] Schwartzmann RJ. New treatments for reflex sympathetic dystrophy. N Engl J
in complex regional pain syndrome I: outcome of a prospective randomised Med 2000;343:654–6.
controlled clinical trial of adjuvant physical therapy versus occupational [36] Schwartzman RJ, Grothusen J, Kiefer TR, Rohr P. Neuropathic central pain:
therapy. Pain 1999;83:77–83. epidemiology, etiology, and treatment options. Arch Neurol
[27] Olney JW, Labruyere J, Wang G, Wozniak DF, Price MT, Sesma A. NMDA 2001;58:1547–50.
antagonist neurotoxicity: mechanisms and prevention. Science [37] Schüttler J, Stanski DR, White PF, Trevor AJ, Horai Y, Verotta D. Sheiner LB:
1991;254:1515–8. pharmacodynamic modeling of the EEG effects of ketamine and its
[28] Perez RS, Oerlemans HM, Zuurmond WW, De Lange JJ. Impairment level enantiomers in man. J Pharmacokinet Biopharm 1987;15:241–53.
sumscore for lower extremity complex regional pain syndrome type I. Disabil [38] Shirani P, Salamone AR, Schulz PE, Edmondson EA. Ketamine treatment for
Rehabil 2003;25:984–91. intractable pain in a patient with severe refractory complex regional pain
[29] Perez RS, Roorda LD, Zuurmond WW, Bannink II, Vranken JH, De Lange JJ. syndrome: a case report. Pain Phys 2008;11:229–42.
Measuring perceived activity limitations in lower extremity complex regional [39] Sigtermans MJ, Olofsen E, Sarton EY, Kleef van JW, Dahan A. PK/PD modeling of
pain syndrome type 1 (CRPS I): test–retest reliability of two questionnaires. the analgesic and side effects of low-dose S(+)-ketamine in healthy volunteers.
Clin Rehabil 2002;16:454–60. Anesthesiology 2008; ASA annual meeting, abstract #A1584.
[30] Perez RS, Zuurmond WW, Bezemer PD, Kuik DJ, van Loenen AC, De Lange [40] Stomp-van den BSG, Vlaeyen JW, van Breukelen G, Stomp SG, ter Kuile MM,
JJ, Zuidhof AJ. The treatment of complex regional pain syndrome type I Kole-Snijders AM. Meetinstrumenten chronische pijn: deel 2. Pijn Coping en
with free radical scavengers: a randomized controlled study. Pain Cognitielijst (PCCL). Pijn Kennis Centrum; 2001.
2003;102:297–307. [41] Veldman PH, Reynen HM, Arntz IE, Goris RJ. Signs and symptoms of reflex
[31] Pleger B, Ragert P, Schwenkreis P, Förster A-F, Wilimzig C, Dinse H, Nicolas V, sympathetic dystrophy: prospective study of 829 patients. Lancet
Maier C, Tegenthoff M. Patterns of cortical reorganization parallel impaired 1993;342:1012–6.
tactile discrimination and pain intensity in complex regional pain syndrome. [42] Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic
Neuroimage 2006;32:503–10. musculoskeletal pain: a state of the art. Pain 2000;85:317–32.
[32] Pleger B, Tegenthoff M, Ragert P, Förtser AF, Dinse HR, Schwenkreis P, Nicolas [43] Zarate Jr CA, Singh JB, Carlson PJ, Brutsche NE, Ameli R, Luckenbaugh DA,
V, Maier C. Sensorimotor returning in complex regional pain syndrome Charney DS, Manji HK. A randomized trial of an N-methyl-D-aspartate
parallels pain reduction. Ann Neurol 2005;57:425–9. antagonist in treatment-resistant major depression. Arch Gen Psychiatry
[33] Rommel O, Malin JP, Zenz M, Janig W. Quantitative sensory 2006;63:856–64.
testing, neurophysiological and psychological examination in patients with [44] Zhou QQ, Imbe H, Zou S, Dubner R, Ren K. Selective upregulation of the flip-
complex regional pain syndrome and hemisensory deficits. Pain flop splice variants of AMPA receptor subunits in the rat spinal cord after
2001;93:279–93. hindpaw inflammation. Brain Res Mol Brain Res 2001;88:186–93.
[34] Sang CN. NMDA-receptor antagonists in neuropathic pain: experimental [45] Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta
methods to clinical trials. J Pain Symptom Manage 2000;19:S21–5. Psychiatr Scand 1983;67:361–70.