ARTICLE IN PRESS

Pain xxx (2007) xxxxxx www.elsevier.com/locate/pain

Pregabalin in patients with central neuropathic pain: A randomized, double-blind, placebo-controlled trial of a exible-dose regimen
J.H. Vranken
a b

a,*

, M.G.W. Dijkgraaf b, M.R. Kruis a, M.H. van der Vegt a, M.W. Hollmann a, M. Heesen c

Department of Anesthesiology, Academic Medical Center, University of Amsterdam, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands Department of Clinical Epidemiology, Biostatistics & Bio-informatics, Academic Medical Center, University of Amsterdam, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands c Department of Anesthesiology, Bamberg Hospital, Bamberg, Germany Received 23 January 2007; received in revised form 14 June 2007; accepted 25 June 2007

Abstract The eective treatment of patients suering from central neuropathic pain remains a clinical challenge, despite a standard pharmacological approach in combination with anticonvulsants and antidepressants. A randomized, double-blinded, placebo-controlled trial evaluated the eects of pregabalin on pain relief, tolerability, health status, and quality of life in patients with central neuropathic pain caused by brain or spinal cord injuries. At baseline and 4 weeks after the start of treatment subjects were evaluated with standard measures of ecacy: pain intensity measured by visual analog scale, health status (Pain Disability Index and EQ-5D) and quality of life (SF-36). Forty patients received escalating doses of either pregabalin (150, 300, and 600 mg/day) or matching placebo capsules. In both groups, patients started with 1 capsule per day (either 150 mg of pregabalin or placebo). If pain relief was insufcient, patients were titrated to a higher dose. There was a statistically signicant decrease in mean pain score at endpoint for pregabalin treatment, compared with placebo (P = 0.016). Follow-up observation showed no signicant dierence in Pain Disability Index scores between the two groups. The pregabalin group, however, showed a statistically signicant improvement for the EQ5D. Pregabalin treatment led to a signicant improvement in the bodily pain domain of the SF36. In the other domains, more favorable scores were reported without reaching statistical signicance. Pregabalin, in a exible-dose regime, produced clinically significant reductions in pain, as well as improvements in health status in patients suering from severe central neuropathic pain. 2007 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Keywords: Pregabalin; Central neuropathic pain; Quality of life

1. Introduction Central neuropathic pain (pain associated with lesions of the central nervous system) has been estimated to occur in up to 8% of patients after stroke, and in about 40% of patients with spinal cord injury (Siddall
Corresponding author. Address: Pain Relief Unit, Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. Tel.: +31 20 5662533; fax: +31 20 6979441. E-mail address: j.h.vranken@amc.uva.nl (J.H. Vranken).
*

et al., 2003; Nicholson, 2004). The mechanisms underlying central neuropathic pain are not fully understood. A dominating feature of central pain, however, is an abnormal spinothalamic function with altered sensitivity to temperature and pinprick (Finnerup and Jensen, 2004). Disruption of the spinothalamic pathways may contribute to neuronal hyperexcitability, loss of descending inhibitory control mechanisms in the spinal cord, and alterations in the processing of incoming noxious and non-noxious stimuli resulting in an abnormal pain perception (Eide, 1998; Nicholson, 2004). In addition, loss of balance between noxious and non-noxious

0304-3959/$32.00 2007 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pain.2007.06.033

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033

ARTICLE IN PRESS
2 J.H. Vranken et al. / Pain xxx (2007) xxxxxx

sensory inputs gives rise to neuronal reorganization in the thalamus contributing to the onward ow of nociceptive information to the postcentral gyrus of the cortex (Millan, 1999). Despite recent advances in identication of peripheral and central sensitization mechanisms related to central nervous system injury, the eective treatment of patients suering from central pain remains a clinical challenge. In spite of numerous treatment options (including opioids, anticonvulsants, antidepressants, baclofen, a-adrenergic agonists, and ketamine), some of these patients still experience severe neuropathic pain. In addition, the use of these agents is often limited by signicant side eects. Recently, pregabalin was reported to possess antihyperalgesic and antiallodynic properties in various animal models. Moreover, pregabalin was shown to be eective in randomized clinical trials of nonmalignant chronic neuropathic pain (including postherpetic neuralgia, diabetic peripheral neuropathy, and neuropathic pain following spinal cord injury) (Gilron and Flatters, 2006; Siddall et al., 2006). Additionally, this anticonvulsant has excellent bioavailability and a favorable safety prole with minimal concern for drug interactions and no interference with hepatic enzymes. Given the absence of other eective pharmacological treatments for central pain, any medication providing some benet in terms of symptom amelioration and quality of life improvement in patients with neuropathic pain has to be evaluated. Therefore, we tested, in a randomized, double-blind, placebo-controlled trial, the eects of pregabalin on pain relief, tolerability, health status, and quality of life in patients with central neuropathic pain.
2. Methods 2.1. Patients The study was approved by the Medical Ethical Committee of our institution. Eligible patients visiting our outpatient hospital clinic (between January, 2006 and March, 2006) were recruited. Randomization took place after all patients were recruited (see below). Inclusion criteria: age 18 years or older; written informed consent; patients suering from severe neuropathic pain, visual analog scale score more than six caused by lesion or dysfunction in the central nervous system. The pain had to persist continuously for at least 6 months, and had started after sustaining the lesion or dysfunction in the central nervous system. Neuropathic pain was described by at least one of the following: burning pain, paroxysmal episodes of shooting pain, or pain on light touch. Additionally, patients had to score above 12 on the Leeds Assessment of Neuropathic Symptoms and Signs questionnaire (LANSS) (Bennett, 2001). Patients were not eligible or not recruited, if they: were pregnant; had a history of intolerance, hypersensitivity, or known allergy to pregabalin; had a known history of signicant hepatic, renal, or psychiatric disorder; had a history of galactose intolerance, lactase deciency, or glucosegalactose malabsorption syndrome. Subjects with a calculated creatinine clear-

ance rate below 60 mL/min (estimated from serum creatinine using CockroftGault equation) were also excluded. Patients taking concomitant analgesic medication were allowed to enter the study if neuropathic pain treatment was on a stable regimen at least 90 days before start of the study. Patients who had been exposed previously to gabapentin, regardless of dose and treatment duration, were permitted to enter the study. However, treatment with gabapentin was to be discontinued at least 3 days before receiving study medication. No new analgesic therapies were to be initiated at any time during the trial. Full blood count, urea, electrolytes and liver function tests were measured before treatment. 2.2. Study design, randomization, and treatment The study consisted of a 4-week double-blind treatment in which a blinded adaptation in response to patients needs was applied. Patients who met all inclusion criteria were arranged into 8 categories of dierent size based on their sex, age (above and below the median of 55 years of age), and diagnosis (central or spinal pain). Within each category (e.g. male above 55 years of age with central pain, n = 8), patients were randomized at the Department of Clinical Epidemiology, Biostatistics and Bioinformatics based on an automated assignment system to receive either exible-dose placebo (14 capsules per day) or exible-dose pregabalin (14 capsules of 150 mg/day). The resulting randomization list was handed over to the hospital pharmacist. The hospital pharmacist prepared identical, coded medication bottles containing identical capsules of pregabalin or placebo. Patients received escalating doses of either pregabalin (150, 300, and 600 mg/day) or matching placebo capsules (daily orally intake of 1, 2, or 4 capsules) titrated at 3-day intervals based on response and tolerability. At baseline each coded medication bottle was supplied by hospital pharmacist to the blinded treating physician. In both groups, patients started with 1 capsule per day (containing either 150 mg of pregabalin or placebo). If the response (i.e. pain relief) was insucient (dened as a reduction of less than 1.8 on a visual analog scale), patients were given a higher dose (i.e. more capsules of either 150 mg of pregabalin or placebo). All patients received, respectively, active medication or matching placebo capsules and followed the same oral dosing schedule (daily dosages were split into two equivalent doses, one administered in the morning and one in the evening). A single downward dose titration after a 3-day interval was allowed, after which the patient remained on the nal dosage during the remainder of the study period. Outpatient clinic visits occurred at baseline, at the start of the randomized trial, and at the end of week 1 and week 4. At weeks 2 and 3, out-of-hospital consultations took place by telephone. 2.3. Measurements Before start of treatment, patient history and a physical examination provided information about the pain localization, pain characteristics, opioid use, and the need for adjuvant analgesics. The patients existing medical therapy (including analgesics) was continued during the study period. The primary ecacy parameter was a pain intensity score recorded by patients (at baseline, end of each week, and 4 weeks follow-

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033

ARTICLE IN PRESS
J.H. Vranken et al. / Pain xxx (2007) xxxxxx 3

ing treatment), using a visual analog scale (VAS). The mean endpoint pain score (in both groups) was based on the average of three (VAS) pain scores measured the last 24 h of treatment. Health status and quality of life (QOL) questionnaires (secondary outcomes) were completed before start of treatment and 4 weeks following start of treatment. Health status and QOL measurements included the Pain Disability Index (PDI), the EQ-5D, and the Medical Outcomes Short-form Health Survey questionnaire 36 (SF36) (EuroQol, 1990; McHorney et al., 1992; Chibnall and Tait, 1994). The PDI asks subjects to rate the degree to which chronic pain interferes with activities in each of seven domains: family/home responsibilities, recreation, social activity, occupation, sexual behavior, self-care, and life-supporting activities. The total range is 070 points with higher scores indicating more perceived disability. The EQ-5D consists of two sections. The rst section (EQ5D health status prole) contains a description of the health state in 5 dimensions: mobility, self-care, usual activities, pain/complaints, and anxiety/depression. Patients indicate for each dimension whether they experience no, some, or serious health problems. An individual health status prole results, e.g. no problems with mobility and with self-care, some problems with usual activities and with anxiety/depression, and serious problems with pain/discomfort. Each health status prole is converted into a single utility score using a previously determined scoring algorithm based on a time trade-o elicitation technique during interviews with non-institutionalized adults from the general UK population (Kind et al., 1998). The utility score ranges from as low as 0.594 indicating serious problems on all dimensions to unity indicating no problems at all. By convention, death takes the value of zero. The second section (EQ-5D VAS) indicates the perception by the patient of his overall health on a 100 mm VAS (zero denoting worst imaginable health state, and 100 best imaginable health). The SF-36 questionnaire compromises 36 items, which fall into eight domains including physical functioning (Physical), role-physical functioning (Role), bodily pain (Pain), general health perception (Health), energy/vitality (Vitality), social functioning (Social), role-emotional functioning (Emotional), and mental health (Mental). Responses are summed and then transformed into a scale from 0 (for poor health) to 100 (good health) for each domain. The safety of pregabalin was assessed based on the occurrence, nature, and severity of adverse events.

Normally distributed data are expressed as mean values (SD). Non-normally distributed data are expressed as medians with the interquartile range (IQR). Group dierences after 4 weeks following the start of treatment were assessed using a two-group t-test. In case of signicant baseline dierences between groups or in case of non-signicant baseline dierences pointing in the same direction as signicant dierences at 4 weeks, correction for baseline scores was performed by analysis of covariance or by two-group t-tests or non-parametric MannWhitney U-tests of change scores from baseline, depending on data distributions. The frequency of reported adverse events was analyzed by Chi-square test or Fishers exact test where appropriate. Statistical analyses were performed without prior knowledge of which group represented the placebo group. The group code was unveiled just before the test results were interpreted.

3. Results 3.1. Patient characteristics Of a total of 76 patients, 40 patients were randomized and received study medication (Fig. 1). Thirty-ve screened patients were excluded for failing to meet the eligibility criteria. One patient withdrew, after randomization, before taking any study medication and was not included in the ITT population. Patient characteristics, cause of central pain, and analgesic treatment of the ITT population are shown in Table 1. The two groups (n = 20 per group) were similar with respect to baseline demographic, and disease characteristics (diagnosis, presence of allodynia, baseline pain intensity, and analgesic treatment). The etiology causing central neuropathic pain in these patients was stroke (n = 12, 30%), thalamus lesion (n = 4, 10%), brainstem infarction (n = 3, 7.5%), and spinal cord lesion (complete spinal cord lesion: n = 11, 27.5%; incomplete spinal cord lesions: n = 10, 25%). Before start of the study, patients were treated with a combination of opioids, antidepressants (30% of patients), carbamazepine (10% of patients), and baclofen (15% of patients). Overall, 82.5% of patients randomized to treatment completed the entire course of study, including 85% of the pregabalin-treated group (n = 17), and 80% of the placebo group (n = 16). In each group (pregabalin and placebo), 3 patients discontinued treatment due to adverse events (dizziness and somnolence) (Fig. 1). Additionally, one patient receiving placebo discontinued treatment due to perceived lack of ecacy. 3.2. Pain intensity scores Weakly mean pain scores (SD) for both groups are displayed in Fig. 2. Mean pain intensity VAS scores (SD) before and after 4 weeks of treatment (end of trial) changed from 7.4 (1.0) to 7.3 (2.0) in the pla-

2.4. Sample size, calculation, and statistical analysis The sample size calculation was based on data from a recent trial in patients suering from central neuropathic pain (Vranken et al., 2005). The average score on a visual analog scale for pain was expected to be 7.3 for patients treated with placebo. With a sample size of 20 in each group the study will have 80% power to detect a clinically relevant 25% decrease in mean VAS score for pain from 7.3 (placebo group) to 5.4 (pregabalin group), assuming that the common standard deviation is 2.0 and using a two-group t-test with a 0.05 two-sided significance level. Data were analyzed on an intention-to-treat basis (ITT) including the patients who received at least one dose of randomized study medication.

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033

ARTICLE IN PRESS
4 J.H. Vranken et al. / Pain xxx (2007) xxxxxx

Screened (n = 76)

Patients did not meet criteria (n = 35) Patient withdrew informed consent (n = 1)

Randomized/ ITT (n = 40)

Placebo (n = 20)

Pregabalin (n = 20)

Completed study (n = 16) 4 capsules per day (n = 14) 2 capsules per day (n = 2) Withdrawn (n = 4). Adverse events (n = 3) Lack of efficacy (n = 1)

Completed study (n = 17) 600 mg per day (n = 9) 300 mg per day (n = 8) Withdrawn (n = 3). Adverse events (n = 3)

Fig. 1. Summary of participant ow during the study.

Table 1 Patient characteristics (intention-to-treat population) Characteristics Demography Age in years Weight Gender Calculated creatinine cleareance Etiology Stroke Thalamus lesion Brainstem pathology Spinal cord (complete) lesion Spinal cord (incomplete) lesion Site of pain Hemibody Lower half of the body (both legs) Arm and/or leg on one side Both arms Presence of allodynia Baseline pain score (VAS) Concomitant therapy (patients may be taken > 1) Groups Mean SD Mean SD F:M Mean SD Number of patients 6 2 2 6 4 Number of patients 5 5 7 3 Number of patients Median (IQR) Mean SD) Opioidsa Anti-inammatoriesb Antidepressantsc Carbamazepine Baclofen 17 7.0 (7.08.0) 7.4 (1.1) 11 4 3 2 2 4 4 11 1 17 7.5 (7.08.0) 7.6 (0.8) 10 2 5 2 2 6 2 1 5 6 Placebo 54.7 (9.7) 79.1 (15.8) 10 F:10 M 132.9 (71.5) Pregabalin 54.2 (9.4) 84.3 (11.7) 9 F:11 M 128.1 (31.8)

a b c

Includes tramadol, buprenorphine, oxycodone, and fentanyl. Includes aspirin, diclofenac, and ibuprofen. Includes paroxetine and mirtazepine.

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033

ARTICLE IN PRESS
J.H. Vranken et al. / Pain xxx (2007) xxxxxx
VAS score Mean pain score (VAS for pain intensity)
10 9 8 7 6 5 4 3 2 1 0 baseline week1 week 2 week 3 week 4
*

pregabalin placebo

The SF36 domain scores at baseline and 4 weeks following start of treatment are presented in Table 3. Each of the eight domains of the SF36 was analyzed separately. Pregabalin treatment led to a signicant improvement in the bodily pain domain. In the other domains, higher scores (i.e. more favorable) were reported for the pregabalin group without reaching statistical signicance. 3.4. Safety Pregabalin was generally well tolerated with few reported side eects. The most frequently reported adverse events were central nervous system related (dizziness, decreased intellectual performance, and somnolence) and nausea. The incidence of these adverse events (mild or moderate in intensity), however, did not dier signicantly between treatment groups (Table 4). 4. Discussion The present double-blinded, placebo-controlled study demonstrated the analgesic ecacy of pregabalin for central neuropathic pain caused by brain or spinal cord injuries. The primary ecacy measure (i.e. reduction in mean endpoint pain score using a VAS score) at the end of the study changed signicantly between placebo and pregabalin. Additionally, there were signicant improvements in health status (EQ-5D) and in the bodily pain subscale of the SF36 (secondary pain-related outcome measures) following 4 weeks of pregabalin administration. The clinical relevance of the reduction in mean endpoint pain score may be illustrated by the numbers needed to treat (NNT) by pregabalin per additional responder to pain treatment. If successful treatment of individual patients is dened as a 30% or even a 50% reduction in pain level, then the NNT equals 4.0 (95% CI: 2.0328.1) and 3.3 (95% CI: 1.914.3), respectively, given that seven patients in the pregabalin group showed >50% pain relief against one patient with >50% and another patient with >30% pain relief in the placebo group.

Fig. 2. Mean pain score (error bars indicate the standard deviation) rated on a VAS scale at baseline, and each week during treatment with placebo (open circle and dashed line), and pregabalin (black square and solid line) for patients with central neuropathic pain. *p = 0.01; p-value is based on T-test for group dierences after 4 weeks.

cebo group, and from 7.6 (0.8) to 5.1 (2.9) in the pregabalin group. There was a statistically signicant decrease in mean pain score at endpoint for pregabalin treatment, compared with placebo (VAS-score dierence from placebo: 2.18 with 95% condence of interval: 0.573.80; P = 0.01). There was no dierence in pain relief following pregabalin treatment between the group of patients with neuropathic pain due to brain injury and the group with neuropathic pain due to spinal cord injury. 3.3. Health status and quality of life assessment The PDI, the EQ-5D utility, and EQ-5D VAS scores are shown in Table 2. Follow-up observation showed no signicant dierence in PDI scores between the two groups. The pregabalin group, however, showed a statistically signicant improvement (i.e. more favorable) for the EQ-5D utility score and EQ-5D VAS score compared with the placebo group.

Table 2 Pain Disability Index scores, EQ-5D health utility scores, and EQ-5D VAS scores by treatment group, at baseline and 4 weeks following start of treatment Groups Placebo Baseline Health status score Pain Disability Index EQ-5D utility EQ-5D VAS 41.7 (14.8) 0.16 (0.34) 50.1 (19.7) After 4 weeks 43.3 (14.7) 0.06 (0.17, 0.57) (#) 37.8 (18.5) Pregabalin Baseline 39.9 (13.2) 0.28 (0.32) 60.4 (17.0) After 4 weeks 35.7 (14.9) 0.59 (0.52, 0.67) (#) 65.7 (17.0) 0.111 <0.001 U <0.001 U P-Value

Outcomes of the PDI, EQ-5D health utility, and EQ-5D VAS. Data are presented as means (SD) or, if appropriate, as median (IQR) (#). Higher PDI scores indicate more perceived disability. Higher EQ-5D health utility scores and EQ-5D. VAS scores indicate improved health status. P-Values are based on t-tests for group dierences after 4 weeks, unless otherwise stated. U based on analysis of covariance to correct for baseline scores.

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033

ARTICLE IN PRESS
6 J.H. Vranken et al. / Pain xxx (2007) xxxxxx

Table 3 Baseline and end of trial outcomes data for the dierent domains of the SF36 quality of life survey Domains of the SF36 Placebo Baseline Physical Role (#) Pain Health Vitality Social Emotional Mental 30.5 (23.8) 0 (0, 0) 26.2 (15.4) 39.1 (21.8) 46.3 (18.0) 55.0 (23.4) 61.7 (47.5) 63.2 (22.0) After 4 weeks 30.0 (23.5) 0 (00) 27.8 (19.4) 37.3 (17.7) 41.0 (20.8) 58.1 (23.7) 40.0 (46.6) 62.0 (21.3) Pregabalin Baseline 31.5 12.5 30.7 46.2 45.8 51.3 60.0 67.5 (21.4) (043.8) (16.1) (23.6) (21.7) (20.6) (45.4) (19.2) After 4 weeks 34.0 12.5 46.3 49.9 50.3 63.8 55.0 70.3 (23.4) (087.5) (20.2) (21.5) (19.3) (20.2) (43.6) (18.8) 0.592 0.206 0.009 U 0.156 0.153 0.425 0.3 0.202 P-Value

The SF-36 questionnaire compromises 36 items, which fall into eight domains including physical functioning (Physical), role-physical functioning (Role), bodily pain (Pain), general health perception (Health), energy/vitality (Vitality), social functioning (Social), role-emotional functioning (Emotional), and mental health (Mental). Responses are summed and then transformed onto a scale from 0 (for poor health) to 100 (good health) for each domain. Data are presented as means (SD) or, if appropriate, as median (IQR) (#). P-Values are based on t-tests for group dierences after 4 weeks, unless otherwise stated. based on MannWhitney U-test for change scores from baseline. U based on analysis of covariance to correct for baseline score.

The abnormal pain perception in central pain syndromes results from multiple mechanisms including central neuronal hyperexcitability, loss of descending inhibitory control mechanisms in the spinal cord, and alterations in the processing of incoming noxious and non-noxious stimuli (Eide, 1998). One of the key neurological processes in the induction and maintenance of neuropathic pain, however, is the involvement of voltage-sensitive calcium channels (Yaksh, 2006). Individual case reports and several clinical studies have provided convincing evidence that pregabalin (binds potently to the a2d-subunit, an auxiliary protein associated with voltage-sensitive calcium channels) has analgesic activity in chronic pain syndromes and in a variety of peripheral neuropathic pain syndromes (Fink et al., 2002; Gilron and Flatters, 2006). Additionally, pregabalin is also eective in relieving neuropathic pain in patients with spinal cord injury (Siddall et al., 2006). In this study, pregabalin-treated patients experienced signicant pain relief compared with placebo. Because the reduction in pain scores was modest (patients in the pregabalin group still experienced pain

Table 4 Frequency of reported adverse events Adverse event Patients may experience > 1 Nausea Cognitive performance Somnolence Dizziness Confusion Peripheral edema
a

Placebo adverse event 4 8 9 6 4 4

Pregabalin 6 6 9 7 7 1

P-Valuea 0.507 0.507 1.0 0.736 0.288 0.342

Frequency of adverse events was analyzed by Chi-square test or Fishers exact test.

with a mean VAS score of 5 at the end of the trial), the clinical importance of the observed decrease from baseline might be debatable. However, the refractory nature of the pain and its duration in these patients should be kept in mind. These patients suered from severe pain with a baseline VAS score of 7.3 before start of the trial, despite combinations of opioids, antidepressants, baclofen, and carbamazepine. In addition, it has been demonstrated that an approximately 2030% reduction in pain intensity represents a patient determined, clinically important response regardless of disease type, age, sex, baseline pain, or treatment group (Farrar et al., 2001). By this criterion, patients experienced signicant pain relief following treatment with pregabalin. This improvement in pain score was reinforced by parallel improvements observed in secondary outcome measurements. Pregabalin-treated patients scored better than placebo treated patients for all eight domains of the SF36. For the bodily pain domain, this dierence was statistically signicant. Scores from the EQ-5D questionnaires following administration of pregabalin were also improved signicantly suggesting an increase in health status. The benets of treatment across specic domains in quality of life, however, were less consistent than the benets of treatment on pain intensity. Additionally, the PDI (assessment of disability) outcome did not dier between both groups at the end of the trial. The results from these secondary outcome parameters are not surprising because patients with central neuropathic pain experience impairments in both physical and psychological functioning (Haythornthwaite and Benrud-Larson, 2000). Patients distress also results from other factors (than pain) which add to suering, including altered emotional well being, irreversible disabilities due to brain

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033

ARTICLE IN PRESS
J.H. Vranken et al. / Pain xxx (2007) xxxxxx 7

or spinal cord injury, changes in body perception, changes in professional and social function, increased dependency, and distress from retrospection. These issues may have a profound adverse impact on patients quality of life despite improved pain control (Cassel, 1982; Haythornthwaite and Benrud-Larson, 2000). When interpreting these results, however, one should bear in mind that the study was powered on 25% pain relief as the primary outcome measure. With respect to the SF-36 health domains, the actual sample size suced to detect absolute improvements by 20 points on the 0100 scale, except for Role (30 points) and Emotional (40 points). In case of the PDI, the minimum detectable improvement was 14 points on a 070 scale. Treatment with pregabalin was well tolerated. The incidence of side eects did not dier between the placebo group and the pregabalin group. The most frequently occurring adverse events, somnolence and decreased intellectual performance, were mild in intensity. More patients in the pregabalin-treatment group completed the study than did patients in the placebo group. Because doses of pregabalin were individualized to achieve meaningful improvement in the patients neuropathic pain, not all patients received dosing of pregabalin at the upper end of its dosing range which may be responsible for a decreased incidence of adverse events and discontinuation of therapy (Freynhagen et al., 2005). Alternatively, pain relief may have been of more importance to the majority of treated patient than were the observed adverse events. Two methodological aspects of this study require comment. In this study, a exible-dose regimen of pregabalin was evaluated in patients with central neuropathic pain. Due to this design, it was not possible to draw any conclusions regarding a doseresponse eect of pregabalin in central neuropathic pain. This study design, however, aimed to reect the real-life management of neuropathic pain in which physicians tailor the dosage of prescribed drugs based on individual patients responses (i.e. adequate pain relief). In clinical practice, exible dosing of a drug optimizes the balance of ecacy and tolerability (Freynhagen et al., 2005). Adequate pain relief is achieved with lower doses of the drug together with a reduced incidence of adverse events, thus enhancing the tolerability of the drug. Because our patients experienced severe pain, concomitant analgesic medication was continued during the trial. Neuropathic pain treatment, however, was on a stable regimen and no new analgesic therapy was allowed to be initiated during the trial. Additionally, the proportion of patients taking additional pain medication was well balanced across treatment groups and analysis demonstrated a similar treatment eect of pregabalin regardless of whether patients used any concomitant analgesics.

A limitation of this study might be that patients in this trial did not receive the drug for a sucient length of time. Although most studies using pregabalin for other neuropathic pain syndromes have demonstrated pain improvements within 1 week, it is important to emphasise that longer duration of treatment may be necessary to conrm long term treatment eectiveness (clinical practice) and to demonstrate widespread benecial eects of pain relief on the various components of quality of life. Because the main purpose of this study was to assess the clinical benets of pregabalin in patients with central neuropathic pain, individual symptoms of neuropathic pain (including allodynia, burning pain, shooting pain, and hyperalgesia) were not scored following pregabalin treatment. There is, however, increased evidence that a symptom-based analysis should be performed because these specic symptoms may respond dierently to treatment (Jensen and Baron, 2003). Pregabalin was eective in relieving central neuropathic pain. Additionally, pregabalin was well tolerated. Direct comparisons between pregabalin and other drugs used to relieve central neuropathic pain, however, are not possible because comparator trials have not yet been performed. Although tricyclic antidepressants are also eective in relieving central neuropathic pain, treatment with these analgesics, however, may be compromised by their side eects (Saarto and Wien, 2005; Sindrup et al., 2005). Alternatively, opioids have also been recommended as treatment for neuropathic pain (Dworkin et al., 2003; Rowbotham et al., 2003). The prospect of commencing an analgesic, however, whose use may be complicated by analgesic tolerance, withdrawal reactions after discontinuation, and always a possibility for addiction, is not satisfactory (Ballantine and Mao, 2003). Given the poorly eective pharmacological treatments for neuropathic pain (in terms of ecacy and tolerability), our ndings and results from other studies regarding the use of pregabalin in other neuropathic pain syndromes support the view that pregabalin may be proposed as a rst-line pharmacological treatment of central neuropathic pain (Dworkin et al., 2003; Finnerup et al., 2005). In summary, pregabalin, in a exible-dose regime, produced clinically signicant reductions in pain, as well as improvements in health status in patients suffering from severe central neuropathic pain. The study protocol was designed to provide a regimen of treatment that is consistent with typical product utilization (every day clinical practice) and allows the individual to receive a dose that provides optimal ecacy and tolerability. Pregabalin has substantial promise as therapy for central neuropathic pain providing benets in terms of symptom amelioration and health improvement.

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033

ARTICLE IN PRESS
8 J.H. Vranken et al. / Pain xxx (2007) xxxxxx Haythornthwaite JA, Benrud-Larson LM. Psychological aspects of neuropathic pain. Clin J Pain 2000;16:S1015. Jensen TS, Baron R. Translation of symptoms and signs into mechanisms in neuropathic pain. Pain 2003;102:18. Kind P, Dolan P, Gudex C, Williams A. Variations in population health status: results from a United Kingdom national questionnaire survey. BMJ 1998;316:73641. McHorney CA, Ware Jr JE, Lu JF, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions and reliability across diverse patient groups. Med Care 1992;32:4066. Millan MJ. The induction of pain: an integrative review. Prog Neurobiol 1999;57:1164. Nicholson BD. Evaluation and treatment of central pain syndromes. Neurology 2004;62:S306. Rowbotham MC, Twilling L, Davies PS, Reisner L, Taylor K, Mohr D. Oral opioid therapy for chronic peripheral and central neuropathic pain. N Engl J Med 2003;348:122332. Saarto T, Wien PJ. Antidepressants for neuropathic pain. Cochrane Database Syst Rev 2005. doi:10.1002/14651858. Art. No.: CD005454. Siddall PJ, Cousins MJ, Otto A, Griesing T, Chambers R, Murphy TK. Pregabalin in central neuropathic pain associated with spinal cord injury. Neurology 2006;67:1792800. Siddall PJ, McClelland JM, Rutkowski SB, Cousins MJ. A longitudinal study of the prevalence and characteristics of pain in the rst 5 years following spinal cord injury. Pain 2003;103:24957. Sindrup SH, Otto M, Finnerup NB, Jensen TS. Antidepressants in the treatment of neuropathic pain. Basic Clin Pharmacol Toxicol 2005;96:399409. Vranken JH, Dijkgraaf MGW, Kruis MR, van Dasselaar NT, van der Vegt MH. Iontophoretic administration of S(+)-ketamine in Patients with intractable Central Pain: a Placebo-Controlled Trial. Pain 2005;118:22431. Yaksh TL. Calcium channels as therapeutic targets in neuropathic pain. J Pain 2006;7:S1330.

References
Ballantine JC, Mao J. Opioid therapy for chronic pain. N Engl J Med 2003;349:194353. Bennett M. The LANSS Pain Scale: the Leeds assessment of neuropathic symptoms and signs. Pain 2001;92:14757. Cassel EJ. The nature of suering. N Engl J Med 1982;306:63945. Chibnall JT, Tait RC. The Pain Disability Index: factor structure and normative data. Arch Phys Med Rehabil 1994;75:10826. Dworkin RH, Backonja M, Rowbotham MC, Allen RR, Argo CR, Bennett GJ, et al. Advances in neuropathic pain: diagnosis, mechanisms, and treatment recommendations. Arch Neurol 2003;60:152434. Eide PK. Pathophysiological mechanisms of central neuropathic pain after spinal cord injury. Spinal Cord 1998;36:60112. EuroQol group. EuroQol a new facility for the measurement of health-related quality of life. Health Policy 1990;16:199209. Farrar JT, Young Jr JP, LaMoreaux L, Werth JL, Poole M. Clinical importance of changes in chronic pain intensity measured on an 11point numerical pain rating scale. Pain 2001;94:14958. Fink K, Dooley DJ, Meder WP, Suman-Chauhan N, Duy S, Clusman H, et al. Inhibition of neuronal Ca2+-inux by gabapentin and pregabalin in the human neocortex. Neuropharmacology 2002;42:22936. Finnerup NB, Jensen TS. Spinal cord injury pain mechanisms and treatment. Eur J Neurol 2004;11:7382. Finnerup NB, Otto M, McQuay HJ, Jensen TS, Sindrup SH. Algorithm for neuropathic pain treatment: an evidence based proposal. Pain 2005;118:289305. Freynhagen R, Strojek K, Griesing T, Whalen E, Balkenohl M. Ecacy of pregabalin in neuropathic pain evaluated in a 12-week, randomized, double-blind, multicentre, placebo-controlled trial of exible- and xed-dose regimens. Pain 2005;115:25463. Gilron I, Flatters SJ. Gabapentin and pregabalin for the treatment of neuropathic pain: a review of laboratory and clinical evidence. Pain Res Manag 2006;11:16A29A.

Please cite this article in press as: Vranken JH et al., Pregabalin in patients with central neuropathic pain: A randomized, ..., Pain (2007), doi:10.1016/j.pain.2007.06.033