Pregabalin for the treatment of postherpetic neuralgia : A

randomized, placebo-controlled trial

R.H. Dworkin, A.E. Corbin, J.P. Young, Jr., et al.
Neurology 2003;60;1274
DOI 10.1212/01.WNL.0000055433.55136.55

This information is current as of December 16, 2012

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://www.neurology.org/content/60/8/1274.full.html

Neurology ® is the official journal of the American Academy of Neurology. Published continuously
since 1951, it is now a weekly with 48 issues per year. Copyright © 2003 by AAN Enterprises, Inc. All
rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
CME Pregabalin for the treatment
of postherpetic neuralgia
A randomized, placebo-controlled trial
R.H. Dworkin, PhD; A.E. Corbin, MS; J.P. Young Jr., MS; U. Sharma, PhD; L. LaMoreaux, MPH;
H. Bockbrader, PhD; E.A. Garofalo, MD; and R.M. Poole, MD

Abstract—Objective: To evaluate the efficacy and safety of pregabalin in the treatment of postherpetic neuralgia (PHN).
Methods: The authors conducted a multicenter, parallel-group, double-blind, placebo-controlled, 8-week, randomized
clinical trial in PHN, defined as pain for 3 or more months following herpes zoster rash healing. Patients (n ⫽ 173) were
randomized to treatment with pregabalin or placebo. Patients randomized to pregabalin received either 600 mg/day
(creatinine clearance ⬎ 60 mL/min) or 300 mg/day (creatinine clearance 30 to 60 mL/min). The primary efficacy measure
was the mean of the last seven daily pain ratings. Secondary endpoints included additional pain ratings, sleep interfer-
ence, quality of life, mood, and patient and clinician ratings of global improvement. Results: Pregabalin-treated patients
had greater decreases in pain than patients treated with placebo (endpoint mean scores 3.60 vs 5.29, p ⫽ 0.0001). Pain
was significantly reduced in the pregabalin-treated patients after the first full day of treatment and throughout the study,
and significant improvement on the McGill Pain Questionnaire total, sensory, and affective pain scores was also found.
The proportions of patients with ⱖ30% and ⱖ50% decreases in mean pain scores were greater in the pregabalin than in
the placebo group (63% vs 25% and 50% vs 20%, p ⫽ 0.001). Sleep also improved in patients treated with pregabalin
compared to placebo (p ⫽ 0.0001). Both patients and clinicians were more likely to report global improvement with
pregabalin than placebo (p ⫽ 0.001). Given the maximal dosage studied, pregabalin had acceptable tolerability compared
to placebo despite a greater incidence of side effects, which were generally mild to moderate in intensity. Conclusions:
Treatment of PHN with pregabalin is safe, efficacious in relieving pain and sleep interference, and associated with greater
global improvement than treatment with placebo.
NEUROLOGY 2003;60:1274 –1283

The most common complication of herpes zoster in
immunocompetent patients is postherpetic neuralgia
(PHN).1 PHN has been defined as pain persisting at
least 3 months after healing of the herpes zoster
rash,2-4 and physical disability, social withdrawal,
and psychological distress are common sequelae.1,2,5
Tricyclic antidepressants were the first agents to
demonstrate efficacy in randomized controlled trials
in PHN and were considered first-line therapy for
many years.6,7 Recently, anticonvulsant,8,9 opioid,10,11
and topical analgesics12 have been found to have sig-
nificant beneficial effects. Many patients, however,
are refractory to these and other treatments because
of inadequate pain relief or intolerable side effects.1,13
Thus, additional safe and effective treatments are
needed for patients with PHN.
Pregabalin [CI-1008 or (S)-3-isobutyl GABA, (S)-
(⫹)-3-(aminomethyl)-5-methylhexanoic acid] is an
␣2-␦ ligand that has been found effective in a variety
of animal models of neuropathic and nociceptive
pain.14,15 It has a predictable pharmacokinetic pro-
file, few drug interactions, and a rapid onset of ac-
tion,16,17 and it has been found to be safe and
efficacious in patients with painful diabetic neuropa-
thy.18 The objective of the current study was to test
the hypothesis that pregabalin has an analgesic ef-
fect in patients with PHN. In addition, the effects of
treatment on sleep interference, quality of life, mood,
and patient and clinician ratings of the patients’
global improvement were examined.
Methods. Patients. Study participants included men and
women of any race who were at least 18 years of age and had PHN
defined as pain present for more than 3 months after healing of a
herpes zoster skin rash. Patients were considered eligible if their
pain was at least 40 mm on the 100 mm visual analog scale (VAS)
of the Short-Form McGill Pain Questionnaire (SF-MPQ)19 at base-
line and randomization visits and they also completed at least
four daily pain diaries and had a minimum mean daily pain rating
of 4 on an 11-point numerical pain rating scale during the base-

From the University of Rochester School of Medicine and Dentistry (Dr. Dworkin), Rochester, NY; and Pfizer Global Research and Development (A.E. Corbin,
J.P. Young, Dr. Sharma, L. LaMoreaux, Dr. Bockbrader, and Dr. Garofalo), Ann Arbor, MI, and (Dr. Poole) New London, CT.
Supported by Pfizer Global Research and Development, Ann Arbor, MI. R.H.D. received a research grant from Pfizer for participating in the clinical trial
described in this article but was not compensated for article preparation; he has received research grants, consulting fees, or speakers’ bureau honoraria in
the past year from Akros Pharma, AstraZeneca, Elan Pharmaceuticals, Endo Pharmaceuticals, GlaxoSmithKline, King Pharmaceuticals, NeurogesX,
Novartis, Ortho-McNeil Pharmaceutical, Pfizer, Reliant Pharmaceuticals, and UCB Pharma; consulting fees and speakers’ bureau honoraria in excess of
$10,000 were received from Novartis and Pfizer.
Received February 19, 2002. Accepted in final form December 18, 2002.
Address correspondence and reprint requests to Dr. Robert H. Dworkin, University of Rochester School of Medicine and Dentistry, 601 Elmwood Avenue, Box
604, Rochester, NY 14642; e-mail: robert_dworkin@urmc.rochester.edu

1274 Copyright © 2003 by AAN Enterprises, Inc.

line week preceding randomization. Women were neither preg-
nant nor lactating, and those of childbearing age had a confirmed
negative serum pregnancy test at baseline and practiced an appro-
priate method of contraception throughout the study. A normal
chest X-ray within the preceding 2 years was required for partici-
pation. All participating clinical sites received investigational re-
view board approval of the study protocol before patient
enrollment, and all patients provided written informed consent
before study participation.
Patients were generally free of serious or unstable medical
conditions and were excluded if they had any other severe pain
that might confound assessment or self-evaluation of pain due to
PHN or had previous neurolytic or neurosurgical therapy for
PHN. Patients who had failed to respond to previous PHN treat-
ment with gabapentin at dosages ⱖ1,200 mg/day were also ex-
cluded from participation in the trial. In addition, patients were
excluded if their baseline serum creatinine clearance was ⱕ30
mL/min, white blood cell count was ⬍2,500/mm3, neutrophil count
was ⬍1,500/mm3, or platelet count was ⬍100 ⫻ 103/mm3. Patients
were also excluded if they had participated in any other pregaba-
lin clinical trial, or if they had participated in any other clinical
trial of an investigational drug within 30 days before screening.
Certain existing medications were permitted, provided dosing
was stable for at least 30 days before baseline and remained
stable throughout the study. Permitted medications included nar-
cotic and non-narcotic analgesics, acetaminophen (not to exceed 4
g/day), nonsteroidal anti-inflammatory drugs, aspirin, and antide-
pressants, including selective serotonin reuptake inhibitors. Pro-
hibited medications, requiring at least a 7-day washout period
before baseline, included benzodiazepines, skeletal muscle relax-
ants, orally administered steroids, local and topical agents for
relief of PHN, and anticonvulsants. Patients using gabapentin
were required to discontinue its use at least 7 days before the
start of study medication. Injected local anesthetics or steroids
were not permitted within 1 month of baseline.
Study design. The study was a multicenter (29 centers),
parallel-group, double-blind, randomized clinical trial of pregaba-
lin vs placebo in the symptomatic treatment of PHN that con-
sisted of a 1-week baseline phase and an 8-week treatment phase.
To achieve equivalent pregabalin exposure, treatment was strati-
fied on baseline creatinine clearance. Patients with a creatinine
clearance ⬎60 mL/min received pregabalin 600 mg/day (200 mg
three times daily) or placebo, whereas patients with a creatinine
clearance ⬎30 and ⱕ60 mL/min received pregabalin 300 mg/day
(100 mg three times daily) or placebo. Based on pharmacokinetic
modeling studies, this dosage adjustment results in comparable
steady-state concentrations of pregabalin, allowing these strata to
be combined as a single pregabalin treatment arm for the purpose
of analyzing pain relief.20
Baseline phase. During the 1-week baseline phase, patients
rated the intensity of their PHN pain over the previous 24 hours
using an 11-point numerical pain rating scale (0 ⫽ no pain, 10 ⫽
worst possible pain). At the end of the baseline phase, those pa-
tients who had completed at least four entries in the daily pain
diary, who had a mean daily pain score of four or greater during
the preceding 7 days, and who met all other criteria were random-
ized to treatment.
Treatment phase. Patients were randomized on a double-
blind basis to either pregabalin or placebo and titrated to a stable
target dosage. All patients assigned to the pregabalin arm were
administered 150 mg/day (50 mg three times daily) for the first 3
days of treatment and 300 mg/day (100 mg three times daily) for
the remainder of the first treatment week. Beginning at the start
of the second week, those patients assigned to pregabalin who
were in the higher creatinine clearance group (⬎60 mL/min) were
administered 600 mg/day (200 mg three times daily). Clinic visits
occurred at baseline, at the start of randomized treatment, and at
the end of weeks 1, 3, 5, and 8. Study medication dosage remained
stable after the first week until the end of the 8-week treatment
phase. Following this double-blind treatment phase, or after early
termination, patients were given the option to continue treatment
with pregabalin in an open-label extension study at a starting
dosage of 300 mg/day. A follow-up visit was scheduled for patients
not continuing with open-label treatment.
Study medications. All medications were administered orally
and on a double-blind basis. Placebo capsules were identical in
appearance to capsules containing pregabalin. Capsules were sup-
plied in bottles, and patients were instructed to take one capsule
three times daily. Study medication was packaged and labeled
with sequential randomization numbers according to a random-
ization schedule generated with a block size of four. An investiga-
tor could break the blind for an individual patient if a medical
emergency occurred and the Pfizer Clinical Research Department
staff could not be reached. The code was to be broken only in cases
where it was necessary for proper treatment of the patient. One
patient was hospitalized with a stroke on study day 36 and the
code was subsequently broken for this patient. No other code
breaks occurred. The blind was maintained until after the study
was completed and all decisions regarding data evaluability had
been made.
Efficacy measurements. The primary efficacy measure was
pain reduction, as recorded by patients in a daily diary using the
11-point numerical pain rating scale described above. Each morn-
ing, patients rated the intensity of their PHN pain during the
preceding 24-hour period.
In addition, there were several secondary measures of efficacy.
SF-MPQ. The SF-MPQ19 was completed by the patient at
baseline, at the start of treatment, and at the end of treatment
weeks 1, 3, 5, and 8. This questionnaire has three parts: the first
assesses pain quality and yields a sensory score (sum of 11 adjec-
tives—throbbing, shooting, stabbing, sharp, cramping, gnawing,
hot-burning, aching, heavy, tender, and splitting— each rated on
an intensity scale with 0 ⫽ none, 1 ⫽ mild, 2 ⫽ moderate, and 3 ⫽
severe), an affective score (sum of four adjectives—tiring-
exhausting, sickening, fearful, and punishing-cruel—rated on the
same intensity scale), and a total score (sum of the sensory and
affective scores). The second part of the SF-MPQ consists of a
100-mm VAS of pain intensity that patients used to rate their
pain during the preceding week. The third part of the SF-MPQ is
a measure of present pain intensity (PPI) using a six-point scale
(0 ⫽ none, 1 ⫽ mild, 2 ⫽ discomforting, 3 ⫽ distressing, 4 ⫽
horrible, 5 ⫽ excruciating).
Daily sleep interference score. This measure was used to as-
sess the degree to which pain interfered with sleep during the
preceding 24-hour period on an 11-point numerical rating scale
(0 ⫽ did not interfere with sleep, 10 ⫽ completely interfered with
sleep). It was included in the daily diary and was completed by the
patient each day after awakening.
Medical Outcomes Study (MOS) Sleep Scale. The MOS Sleep
Scale21 is a patient-rated questionnaire that assesses key aspects
of sleep and provides an overall sleep problem index; higher scores
on this measure reflect worse sleep. The MOS Sleep Scale was
administered at the randomization and termination visits.
SF-36 Health Survey. The SF-36 Health Survey22 is a self-
administered, 36-item measure of quality of life that assesses the
patient’s level of functioning in eight health domains: physical
functioning, role-physical (role limitations caused by physical
problems), social functioning, bodily pain, mental health, role-
emotional (role limitations caused by emotional problems), vital-
ity, and general perception of health. This questionnaire was
administered at the randomization and termination visits and
was scored in the standard manner.22 For each domain of function-
ing, higher scores reflect better quality of life.
Profile of Mood States (POMS). The POMS23 consists of a list
of 65 descriptors of mood that are each rated on a five-point scale
by patients (0 ⫽ applies not at all, 4 ⫽ applies extremely) and that
yield six scores—tension-anxiety, depression-dejection, anger-
hostility, vigor-activity, fatigue-inertia, and confusion-bewilder-
ment—as well as an overall mood disturbance score. Except for
the vigor-activity scale, lower scores indicate better mood. The
POMS was administered at the randomization and termination
visits.
Patient Global Impression of Change (PGIC). The PGIC24
consists of a self-evaluation by the patient of his or her overall
change since the beginning of the study rated on a seven-point
scale (1 ⫽ very much improved; 2 ⫽ much improved; 3 ⫽ mini-
mally improved; 4 ⫽ no change; 5 ⫽ minimally worse; 6 ⫽ much
worse; 7 ⫽ very much worse). The PGIC was administered at the
termination visit.
Clinical Global Impression of Change (CGIC). The CGIC24
consists of an evaluation by the clinician of the patient’s overall
change since the beginning of the study, rated on the same seven-
point scale as the PGIC at the termination visit.
April (2 of 2) 2003 NEUROLOGY 60 1275
 Safety assessments. The safety of pregabalin treatment was
assessed by comparing the groups with regard to the incidence
and intensity of adverse events and their relationship to study
drug as assessed by the investigator, and by comparing treatment
groups with respect to changes in physical and neurologic exami-
nations, vision testing (visual field screening, dilated ophthalmos-
copy, and visual acuity), 12-lead electrocardiogram, and clinical
laboratory evaluations throughout the study. Adverse events in-
cluded any concurrent illness, sign, or symptom reported by the
patient, the patient’s family, or the investigator. An adverse event
that was not present at baseline or one that increased in intensity
or frequency relative to baseline was considered treatment emer-
gent. All patients randomized to treatment were included in the
safety analyses.
Statistical analyses. Power analysis. Treatment effect was
estimated from the results of published studies of gabapentin in
the treatment of PHN8 and painful diabetic neuropathy,25 both of
which used the same primary efficacy measure as the current
study. The difference in endpoint mean pain scores between
groups was 1.8 in the PHN study and 1.3 in the painful diabetic
neuropathy study; in both studies, the overall SD was 2.35. As-
suming similar results for the current study, and employing two-
sided testing at the 0.05 level, it was determined that 76 patients
per treatment group would provide ⬎90% power to detect a differ-
ence in endpoint mean pain scores ⱖ1.3.
Statistical analysis. All analyses were carried out using the
intent-to-treat population, defined as any patient who received at
least one dose of study medication. Patients randomized to pre-
gabalin were analyzed as a single pregabalin treatment arm. Pa-
tients with no data for a given efficacy measure at baseline or at
the timepoint analyzed were treated as missing rather than using
imputed values.
The primary efficacy analysis was conducted on the endpoint
mean pain scores; specifically, the means of the last seven daily
diary pain ratings recorded by patients while taking study medi-
cation. This analysis was conducted with the last observations—
seven daily pain ratings— carried forward for those patients who
did not complete the trial. Supplemental analyses of the primary
efficacy measure examined pain scores for each day during the
first treatment week, weekly mean pain scores, and the propor-
tions of patients whose endpoint mean pain scores were reduced
ⱖ30% and ⱖ50% compared with their baseline week mean pain
scores.
Secondary efficacy analyses were conducted of the SF-MPQ
sensory, affective, total, VAS, and PPI scores at endpoint and at
weeks 1, 3, 5, and 8; mean sleep interference scores at endpoint
and at each week separately; and the MOS Sleep Scale sleep
problem index, SF-36 domain scores, POMS scale scores, and
PGIC and CGIC ratings at the end of the study. For outcome
measures collected in daily diaries (pain and sleep interference),
the endpoint mean score was the mean of the last seven diary
entries while on study medication, including the entry on the day
after the last dose. If fewer than seven scores were recorded by
endpoint, the available scores were used to determine a mean.
All testing was two-sided and performed using SAS procedures
without adjustment for testing multiple measures.26 The mean
daily pain scores, SF-MPQ scale scores, mean daily sleep interfer-
ence scores, MOS Sleep Scale sleep problem index, eight SF-36
domain scores, and seven POMS scale scores were analyzed using
an analysis of covariance (ANCOVA) main effects model, with
treatment, center, and creatinine clearance strata (a dichotomous
indicator variable) in the model, and the corresponding baseline
measure as the covariate.27 In each case, adjusted (least squares)
means were obtained from the model and 95% CI of the difference
in least-squares means between pregabalin and placebo were con-
structed. The proportions of responders with ⱖ30% and ⱖ50%
reduction from baseline to endpoint mean pain score were ana-
lyzed using the Cochran-Mantel-Haenszel (CMH) test with table
scores,28 adjusting for center and creatinine clearance strata. The
PGIC and CGIC were analyzed using the CMH test with modified
ridit scores28 adjusting for center and creatinine clearance strata.
Pharmacokinetic measures were derived from two venous
blood samples collected from each patient, at 2 weeks following
attainment of assigned dose and at the end of the study. Time of
blood sample collection relative to time of last dose was specified
in the protocol. All plasma samples from all treatment groups
were analyzed for pregabalin concentrations using a sensitive,
1276 NEUROLOGY 60 April (2 of 2) 2003
specific, and validated high-performance liquid chromatographic
method with ultraviolet detection.29 A population pharmacokinetic
analysis was performed using the nonlinear mixed effects model-
ing program NONMEM, version V.30 The analysis assumed a 6-,
6-, and 12-hour dosing interval for the three times daily dosage
regimen. Empirical Bayesian estimates of individual pharmacoki-
netic values (apparent oral plasma pregabalin clearance [CL/F]
and apparent pregabalin volume of distribution [Vd/F]) were ob-
tained using the POSTHOC option in the NONMEM estimation
step. As blood sampling for determining plasma pregabalin con-
centrations was not performed at the same time for all patients,
predicted steady-state morning trough concentrations (Cminss)
and average steady-state concentrations (Cavg) were also esti-
mated using the POSTHOC option to permit comparisons among
patients. Mean, standard error, 95% CI, and percent relative SD
for the estimates were calculated using WinNonlin Pro version 3,
a validated pharmacokinetic estimation program.31
The intent-to-treat population was evaluated for safety and the
data were analyzed using descriptive statistical methods. The
number and percentage of patients in each treatment group who
experienced a treatment-emergent adverse event were summa-
rized. Associated adverse events were also summarized, and in-
clude those events the investigator considered possibly, probably,
or definitely related to study medication, and those for which the
relationship was unknown.
Results. Patient characteristics. There were 173 patients ran-
domized to treatment: 84 to placebo and 89 to pregabalin. Patients
were predominantly white (95%), generally above 65 years of age
(mean age ⫽ 71.5 years; SD ⫽ 10.9; 82% ⱖ65 years), and there
were slightly more women than men (53% vs 47%). Creatinine
clearance was ⬎60 mL/min for 68% of the patients in the study.
Thoracic dermatomes were the most commonly affected site of
PHN (48%), and the mean duration of PHN was 33.8 months.
Overall, the two treatment groups were similar with respect to
demographic and clinical variables (table 1).
The flow of patient participation is depicted in figure 1. Seventy-
six percent of all patients randomized to treatment completed the
study (88% of patients in the placebo group and 65% of patients in
the pregabalin group). Concurrent pain medications were used by
118 patients (68%). The most frequently used pain medications were
acetylsalicylic acid (32%) and acetaminophen (14%). At study termi-
nation, 62 pregabalin-treated patients (70%) and 63 placebo patients
(75%) entered the open-label phase of the study.
A total of 29 patients had possibly important variations from
the protocol. One patient was randomized to treatment with a
baseline mean pain score that was less than four and two patients
had VAS scores ⬍40 at randomization. Two patients had CLcr
values ⬍60 mL/min but were randomized to the CLcr ⬎60 mL/
minute stratum, and two patients had normal CLcr values but
were randomized to the CLcr ⬍60 mL/min stratum. Of these four
patients, three were randomized to placebo and only one was
randomized to pregabalin treatment; this patient did not report
any adverse events. All other protocol variations involved the use
of prohibited medications or failure to maintain a stable regimen
of allowed medications; none was expected to affect the results of
the study.
Primary efficacy measure. There was a decrease in mean pain
scores at endpoint for patients treated with pregabalin compared
to placebo (p ⫽ 0.0001; table 2). This improvement in endpoint
mean pain scores in pregabalin-treated patients compared to
placebo-treated patients remained significant when the patients
in each of the two creatinine clearance strata were analyzed
separately.
Significantly greater pain relief was found with pregabalin
compared to placebo on the second day of the trial (figure 2),
which was the first full day of treatment during the titration
phase because not all patients had their full dosage on the day
treatment was initiated. The superior pain relief with pregabalin
treatment relative to placebo continued to be significant through-
out all 8 weeks of double-blind treatment (figure 3A).
Sixty-three percent of the pregabalin-treated patients vs 25%
of placebo-treated patients reported reductions in pain of 30% or
more (p ⫽ 0.001), a clinically important degree of pain relief.32
Fifty percent of patients treated with pregabalin and 20% of pa-
tients treated with placebo had a ⱖ50% decrease in their pain,
and this difference between the two groups was also significant.
Table 1 Patient demographics and baseline disease characteristics

Characteristic Placebo, n ⫽ 84 Pregabalin, n ⫽ 89 All patients, n ⫽ 173

Sex, n (%)
Men 44 (52.4) 37 (41.6) 81 (46.8)
Women 40 (47.6) 52 (58.4) 92 (53.2)
Race, n (%)
White 82 (97.6) 82 (92.1) 164 (94.8)
Hispanic 1 (1.2) 6 (6.7) 7 (4.0)
Asian or Pacific Islander 1 (1.2) 1 (1.1) 2 (1.2)
Age, y, mean (SD) 70.5 (11.3) 72.4 (10.5) 71.5 (10.9)
Age, y, n (%)
18–64 17 (20.2) 15 (16.9) 32 (18.5)
ⱖ65 67 (79.8) 74 (83.1) 141 (81.5)
Weight, kg, mean (SD) 79.8 (16.7) 74.8 (13.4)* 77.2 (15.3)
Estimated creatinine clearance, mL/min, mean (SD) 80.3 (27.3) 72.9 (27.5) 76.5 (27.6)
Creatinine clearance strata, n (%)
Low (⬎30, ⱕ60 mL/min) 25 (29.8) 30 (33.7) 55 (31.8)
Normal (⬎60 mL/min) 59 (70.2) 59 (66.3) 118 (68.2)
Duration of PHN, mo, mean (SD) 34.4 (36.7) 33.3 (35.4) 33.8 (35.9)
Predominantly affected dermatome, n (%)
Trigeminal 22 (26.2) 18 (20.2) 40 (23.1)
Cervical 6 (7.1) 9 (10.1) 15 (8.7)
Thoracic 41 (48.8) 42 (47.2) 83 (48.0)
Lumbar 11 (13.1) 16 (18.0) 27 (15.6)
Sacral 4 (4.8) 4 (4.5) 8 (4.6)
Baseline mean pain score (SD) 6.4 (1.5) 6.3 (1.4) 6.4 (1.5)

* n ⫽ 88 for the pregabalin-treated group.

PHN ⫽ postherpetic neuralgia.

When the patients who were taking any of the four types of
medications for which there is evidence of efficacy in the treatment of
neuropathic pain—tricyclic antidepressants, opioid analgesics (in-
cluding tramadol), anticonvulsants, and topical analgesics13—were
excluded (n ⫽ 46), the benefit of treatment with pregabalin vs
placebo on endpoint mean pain scores remained (endpoint mean
scores 3.17 vs 5.14; p ⫽ 0.0001).
Because side effects may compromise the integrity of the
double-blind in clinical trials, the primary analysis of endpoint
mean pain scores was repeated after removing all patients who
had one or more of the five side effects reported by more than 10%
of pregabalin-treated patients at any time during the study (dizzi-
ness, somnolence, peripheral edema, amblyopia, and dry mouth;
table 3). In these patients, there was still greater improvement in
the endpoint mean pain scores in the patients treated with pre-
gabalin (3.60 vs 5.29 in placebo patients, p ⫽ 0.0001). When all
patients who reported any of the 11 side effects listed in table 3 at
any time during the study were removed, the greater improve-
ment in endpoint mean pain scores of the pregabalin-treated pa-
tients remained significant (3.62 vs 5.37 in placebo patients).
Secondary efficacy measures. Pregabalin-treated patients
were improved relative to the placebo-treated group at study end-
point with respect to the SF-MPQ sensory, affective, and total
pain scores (p ⬍ 0.005; see table 2). The SF-MPQ total score
demonstrated significant pain reduction for pregabalin vs placebo
at all time points during the study (figure 4A); significant im-
provements in the sensory and affective scores were also observed
by the end of week 1 for pregabalin, and these effects were main-
tained throughout the study. In addition, the pregabalin group
had significantly lower SF-MPQ pain scores on the VAS and PPI
at study endpoint (see table 2), and significant differences be-
tween the treatment groups for these two measures were also
observed at each time point during the study (weeks 1, 3, 5, and 8;
figure 4, B and C).
Mean sleep interference scores were improved in the pregaba-
lin vs the placebo group at study endpoint (p ⫽ 0.0001; see table
2). In addition, as observed with the pain measures, the mean
sleep interference score in pregabalin-treated patients demon-
strated significant improvement beginning with the first week of
treatment, and this remained significant compared to placebo
throughout the study (figure 3B). There was also a significant
improvement on the MOS Sleep Scale sleep problem index at
study endpoint (see table 2).
More favorable mean scores were found in the pregabalin
group compared with the placebo group for several of the SF-36
quality of life domains at the end of the trial; the differences were
significant for the SF-36 bodily pain and general health perception
scales (see table 2). Pregabalin-treated patients improved more
than placebo-treated patients on the POMS depression-dejection
scale (pregabalin endpoint mean score ⫽ 6.70 vs 8.47; p ⫽ 0.051);
the differences between the two treatment groups were not signif-
icant for the other POMS scales.
Significant differences between pregabalin and placebo were
seen in both the PGIC and the CGIC. Global assessments of
change showed that at the end of the trial more patients receiving
pregabalin reported global improvements of very much improved,
much improved, and minimally improved (84% vs 26%; figure 5),
and fewer pregabalin-treated patients reported global worsening
of minimally worse, much worse, and very much worse (4% vs
14%; see figure 5). Clinician assessments of global change closely
paralleled patients’ self-assessments.
April (2 of 2) 2003 NEUROLOGY 60 1277

Figure 1. Patient disposition, indicated numerically at
each level of study progress. Dashed line reflects the fact
that patients who were withdrawn from the trial were gen-
erally not allowed to enter the open-label extension.
Pharmacokinetics. Pregabalin dosage adjustments for creati-
nine clearance used in this study produced comparable plasma
pregabalin concentrations in the two renal function strata of the
pregabalin treatment group. Likewise, the predicted morning
Cminss and Cavg values were comparable in the two renal function
strata of the pregabalin treatment group (table 4). Estimates of total
body clearance were somewhat lower and volume of distribution
somewhat higher in this study population of older individuals
than previously reported,20 as would be expected due to age-
related diminished renal function.
Several subjects had low plasma pregabalin concentration at
study endpoint but the low values can be explained by the time
since the last pregabalin dose, which may reflect either poor com-
pliance or subjects having exhausted their drug supply before the
termination visit. At visit four, only two subjects appeared non-
compliant based on low plasma pregabalin concentrations and the
corresponding post-dose time; both were compliant based on their
study endpoint results. Thus, the two subjects who can be consid-
ered noncompliant at the first visit in which there was pharmaco-
kinetic sampling were deemed compliant at the second visit in
which there was sampling. Only one subject in the placebo group
had a significant plasma pregabalin concentration. This occurred
at study endpoint and it is assumed that the patient received his
open-label extension dose before the blood sample because the
visit four plasma concentration was zero. Therefore, no subjects in
the placebo group appeared to receive pregabalin.
Safety. A majority of patients in both treatment groups— 87%
in the pregabalin group and 63% in the placebo group—reported
an adverse event during the double-blind treatment phase. Re-
gardless of treatment, most patients with adverse events reported
1278 NEUROLOGY 60 April (2 of 2) 2003
a maximum intensity of mild or moderate (81% in the pregabalin
group, and 92% in the placebo group).
More of the adverse events experienced by pregabalin-treated
patients were considered to be related to study medication (73% vs
37% in the placebo group). Of the most commonly reported ad-
verse events related to pregabalin, dizziness, somnolence, periph-
eral edema, dry mouth, and various CNS-related adverse events
occurred more frequently in the pregabalin-treated patients than
in the placebo group (see table 3). More pregabalin-treated pa-
tients discontinued from the study because of an adverse event
(32%, n ⫽ 28) than did placebo-treated patients (5%, n ⫽ 4), but
no pregabalin-treated patients discontinued due to lack of efficacy,
whereas six placebo-treated patients did (7%; see figure 1). The
adverse event that most commonly led to discontinuation of pre-
gabalin treatment was somnolence (11.2%, n ⫽ 10).
The median time to onset of any adverse event was 6 days for
the pregabalin-treated patients and 8 days for the placebo-treated
patients. The median time to onset for somnolence or dizziness
was 2 to 3 days for both treatment groups. In pregabalin-treated
patients who completed the study, the median durations of somno-
lence (53 days) and dizziness (33 days) were longer than in
placebo-treated patients (17 days for somnolence, 3 days for
dizziness).
Peripheral edema was reported for 19 patients: 17 who re-
ceived pregabalin and 2 who received placebo. All cases were mild
to moderate in severity, and in no case was edema related to
worsening of an underlying cardiovascular condition. Median time
to onset of edema in the pregabalin-treated patients was 24 days,
and the median duration of edema was 31 days. Two patients
treated with pregabalin discontinued treatment due to edema.
There were no clinically significant findings involving the hema-
tology, blood chemistry, or urinalysis evaluations in the patients
exposed to pregabalin. Similarly, there were no clinically signifi-
cant findings in the electrocardiogram, visual examinations, or
physical and neurologic examinations.
Discussion. The results of this study demon-
strated that pregabalin is efficacious in relieving
pain and improving sleep in patients with PHN. A
consistent pattern of results among the different out-
come measures demonstrated that there is a rapid
and durable beneficial effect of pregabalin treat-
ment. Beginning with the first full day of treatment
and continuing to the end of the trial 8 weeks later,
pain was significantly improved in the patients
treated with pregabalin compared to those treated
with placebo. In addition, sleep interference scores
as well as the five major components of the SF-
MPQ—total, sensory, affective, VAS, and PPI
scores—were all significantly improved relative to
placebo by the end of the first week of treatment and
they remained so until the end of the trial.
Many of the patients enrolled in the study were
elderly, the mean duration of their PHN was almost
3 years, and although 68% were taking concomitant
pain medications, they still met the enrollment crite-
ria, which required that pain was moderate or se-
vere in intensity.33 These features are characteristic of
patients with PHN, as many as half of whom are par-
tially or totally refractory to all existing treatments.34
Because the individuals we studied appear represen-
tative of the general population of patients with PHN,
our results suggest that pregabalin will be effective in
treating PHN in clinical practice and that it provides
an important therapeutic option for patients with this
chronic neuropathic pain syndrome.
It was recently concluded that a 30% decrease in
pain, which corresponds to PGIC ratings of much im-
Table 2 Summary of efficacy analyses at study endpoint

Pregabalin Placebo Endpoint comparison

Least Least
squares squares
Measure No. mean SE No. mean SE Difference 95% CI p Value

Endpoint mean pain score* 88 3.60 0.24 84 5.29 0.24 ⫺1.69 (⫺2.33, ⫺1.05) 0.0001
Short-form McGill Pain Questionnaire
Sensory score 89 8.15 0.73 84 11.90 0.74 ⫺3.75 (⫺5.71, ⫺1.79) 0.0002
Affective score 89 1.75 0.27 84 2.82 0.28 ⫺1.07 (⫺1.81, ⫺0.33) 0.0047
Total score 89 9.85 0.95 84 14.72 0.96 ⫺4.87 (⫺7.41, ⫺2.34) 0.0002
Visual analogue scale 89 38.68 2.90 84 56.30 2.93 ⫺17.62 (⫺25.37, ⫺9.86) 0.0001
Present pain intensity 89 1.58 0.12 84 1.98 0.12 ⫺0.40 (⫺0.71, ⫺0.09) 0.0127
Endpoint mean sleep interference score*† 88 1.93 0.23 84 3.51 0.23 ⫺1.58 (⫺2.19, ⫺0.97) 0.0001
MOS Sleep Scale sleep problem index† 85 26.63 1.77 82 36.43 1.75 ⫺9.80 (⫺14.49, ⫺5.11) 0.0001
SF-36 Health Survey‡
Physical functioning 85 62.25 1.96 83 61.41 1.90 0.84 (⫺4.25, 5.93) 0.7449
Physical role limitations 86 47.04 4.12 83 44.43 4.09 2.60 (⫺8.27, 13.48) 0.6369
Social functioning 86 78.15 2.44 83 74.68 2.44 3.47 (⫺3.02, 9.96) 0.2920
Bodily pain 86 55.14 2.13 83 46.14 2.13 9.00 (3.33, 14.66) 0.0021
Mental health 86 77.53 1.54 83 73.73 1.54 3.81 (⫺0.28, 7.89) 0.0676
Emotional role limitations 86 73.55 3.97 83 64.62 3.97 8.93 (⫺1.60, 19.46) 0.0960
Vitality 86 49.99 1.89 83 48.94 1.88 1.05 (⫺3.96, 6.05) 0.6798
General health perception 85 67.61 1.58 81 63.40 1.58 4.21 (0.02, 8.40) 0.0488

* Endpoint mean pain and mean sleep interference measures were derived from patients’ last 7 days of diary entries while taking
study drug.
† Higher values reflect greater sleep impairment.
‡ Higher values reflect better quality of life.

MOS ⫽ Medical Outcomes Study; SF ⫽ Short Form.

proved or very much improved, should be considered
clinically important in evaluating the results of chronic
pain trials.32 Using this criterion, almost two-thirds of
the pregabalin-treated patients reported clinically im-
Figure 2. Daily diary pain scores for the first 7 days of
treatment as rated on an 11-point numerical pain rating
scale from 0 (no pain) to 10 (worst possible pain). Solid
curve ⫽ placebo-treated patients; dashed curve ⫽
pregabalin-treated patients; PBO ⫽ placebo; PGB ⫽
pregabalin. *p ⬍ 0.01.
portant improvement of their average daily pain. A
50% decrease in pain ratings between baseline and
study endpoint—a level of pain relief that corresponds
to the highest degree of improvement on the PGIC, a
rating of very much improved32— has also been consid-
ered evidence of clinically important improvement. In
the current study, 50% of the patients in the
pregabalin-treated group but only 20% of those in the
placebo-treated group had a 50% decrease in mean
pain score.
There were no serious adverse events related to pre-
gabalin treatment; no clinically significant electrocardio-
gram, ophthalmologic, physical, or neurologic
examination findings; and the results of laboratory evalu-
ations related to pregabalin treatment were unremark-
able. The tolerability of pregabalin treatment was
acceptable, given the protocol’s brisk titration in a mostly
elderly sample of patients to a maximum dosage of 600
mg daily of pregabalin that patients were not allowed to
adjust during the trial. Indeed, one-third of the
pregabalin-treated patients who withdrew from the study
because of adverse events entered the open-label exten-
sion, in which improved tolerability would be expected
because of the lower dosages permitted. Of those
pregabalin-treated patients with adverse events, more
April (2 of 2) 2003 NEUROLOGY 60 1279

Figure 3. (A) Weekly mean pain score as rated on an 11-
point numerical pain rating scale from 0 (no pain) to 10
(worst possible pain). (B) Weekly mean score of sleep inter-
ference score during the past 24 hours as rated on an 11-
point scale from 0 (did not interfere) to 10 (unable to sleep
due to pain). Solid curves ⫽ placebo-treated patients;
dashed curves ⫽ pregabalin-treated patients; PBO ⫽ pla-
cebo; PGB ⫽ pregabalin. *p ⬍ 0.01.
than 80% reported that the maximum intensity of these
events was mild to moderate. Dizziness and somnolence
were the most common side effects of pregabalin treat-
ment, but could be tolerated by patients even though they
may have persisted throughout the major portion of the
treatment period. Approximately 11% of patients receiv-
ing pregabalin discontinued treatment due to somno-
lence, which was not unexpected given the dosage tested.
Although more pregabalin-treated patients discontinued
the study because of an adverse event compared to pa-
tients administered placebo, fewer pregabalin-treated pa-
tients discontinued the study because of unrelieved pain.
Several methodologic aspects of this trial require
comment. The pregabalin treatment arm was strati-
fied based on renal function because the population
of PHN patients is likely to include significant num-
bers of elderly individuals with diminished renal
function. The steady-state levels of pregabalin result-
ing from the stratification suggest that therapeutic
1280 NEUROLOGY 60 April (2 of 2) 2003
Table 3 Most frequently occurring adverse events*
Preferred term Placebo, n ⫽ 84 Pregabalin, n ⫽ 89
Dizziness 10 (11.9) 25 (28.1)
Somnolence 6 (7.1) 22 (24.7)
Peripheral edema 2 (2.4) 17 (19.1)
Amblyopia 1 (1.2) 10 (11.2)
Dry mouth 2 (2.4) 10 (11.2)
Abnormal gait 1 (1.2) 7 (7.9)
Headache 7 (8.3) 7 (7.9)
Ataxia 0 6 (6.7)
Confusion 0 6 (6.7)
Diarrhea 4 (4.8) 6 (6.7)
Speech disorder 0 5 (5.6)
Values are n (%).
* Adverse events reported by at least 5% of patients in the pre-
gabalin treatment group. Events sorted by decreasing fre-
quency in the pregabalin group.
levels of pregabalin in both renal function strata were
achieved. These pharmacokinetic analyses as well as
the results of the strata subgroup analyses of end-
point mean pain scores provide support for combining
both strata into one pregabalin treatment arm.
One of the exclusion criteria was failure to re-
spond to previous gabapentin treatment of PHN at
dosages ⱖ1,200 mg/day. Although it could be argued
that this methodologic feature of the trial increased
the likelihood of finding a beneficial effect of pregaba-
lin, this assumes that therapeutic response to gabap-
entin predicts therapeutic response to pregabalin.
Unfortunately, history of prior gabapentin treatment
was not recorded, and so it is not possible to deter-
mine how many patients had a beneficial response to
gabapentin before enrollment in the trial and
whether these patients were more likely to respond
to treatment with pregabalin.
It has been proposed that there are distinct sub-
types of PHN, and that these subtypes differ in
patterns of symptoms and signs, underlying patho-
physiologic mechanisms, and treatment response.35,36
No attempt was made in the current study to collect
the extensive data that would be required to accu-
rately categorize patients into these putative PHN
subtypes, which include epidermal nerve fiber den-
sity, response to pharmacologic challenge, and quan-
titative sensory testing. An exploratory attempt was
made to assess allodynia and hyperalgesia at base-
line and study endpoint and to determine whether
resolution of these sensory abnormalities was more
common in the patients treated with pregabalin com-
pared to placebo. Resolution of allodynia at study
endpoint occurred in 48.8% of pregabalin-treated pa-
tients and 30.6% of placebo-treated patients who had
allodynia at baseline; resolution of hyperalgesia at
study endpoint occurred in 33.9% of pregabalin-
treated patients and 29.1% of placebo-treated pa-

Figure 4. (A) Total pain descriptor score from the Short-
Form McGill Pain Questionnaire (SF-MPQ). The score is
the sum of the intensity ratings for pain intensity for 15
sensory and affective pain descriptors on a scale of 0
(none) to 3 (severe). (B) Visual analog scale score from the
SF-MPQ. Patients placed a slash on a 100-mm line from 0
(no pain) to 100 (worst possible pain). (C) SF-MPQ present
pain intensity score, reported on a scale of 0 (no pain), 1
(mild pain), 2 (discomfort), 3 (distressing), 4 (horrible),
and 5 (excruciating). Solid curves ⫽ placebo-treated pa-
tients; dashed curves ⫽ pregabalin-treated patients;
PBO ⫽ placebo; PGB ⫽ pregabalin. *p ⬍ 0.01.
Figure 5. Patient Global Impression of Change (PGIC),
rated by patients on a seven-point scale of overall change
experienced since beginning study medication (1 ⫽ very
much improved; 2 ⫽ much improved; 3 ⫽ minimally im-
proved; 4 ⫽ no change; 5 ⫽ minimally worse; 6 ⫽ much
worse; 7 ⫽ very much worse). *p ⫽ 0.001 for the overall
pregabalin vs placebo comparison using the Cochran-
Mantel-Haenszel test.
tients who had hyperalgesia at baseline. Neither of
these group differences was significant. Interpreta-
tion of these results is problematic, however, because
there were no specific guidelines provided to investi-
gators for determining the presence or absence of
allodynia and hyperalgesia and each investigator in-
dependently decided how to assess these sensory
abnormalities.
Chronic pain syndromes such as PHN are charac-
terized by a constellation of symptoms that extends
well beyond the experience of pain and that may
include profound disturbances of sleep, mood, and
quality of life.1,2,5,37,38 The consistent improvements
observed in this study in the pain and sleep diary
measures, which began early and were maintained
throughout treatment, suggest that pregabalin re-
duces not only the severity of daytime pain in PHN
but also the sleep disturbance that accompanies
chronic pain. Pregabalin-treated patients also re-
ported significantly greater improvements than
placebo-treated patients on the MOS Sleep Scale
sleep problem index as well as on the SF-36 bodily
pain and general health perception scales.
Despite these significant improvements in pain,
sleep, and overall perception of health, significant
benefits of treatment on other domains of mood and
quality of life were not observed. In previous trials of
neuropathic pain, the benefits of treatment across
specific domains of quality of life have also been
found less consistently than the benefits of treat-
ment on pain intensity.8,11,12,25,39,40 Possible explana-
tions for this are that most trials have used general
rather than disease-specific measures of quality of
life and that these trials may not have been of suffi-
cient duration to demonstrate widespread beneficial
effects of pain relief on the various components of
quality of life.
April (2 of 2) 2003 NEUROLOGY 60 1281
Table 4 Pharmacokinetic values for the two renal function strata of the pregabalin treatment group
Measure
Observed plasma pregabalin concentration, ␮g/mL, range
Time postdose for observed plasma pregabalin
concentration, h, range
Predicted Cavg concentrations, ␮g/mL (range)
Predicted morning CminSS concentrations, ␮g/mL (range)
Pregabalin clearance, mL/min, range
Pregabalin volume of distribution, L, range
CLcr ⫽ creatinine clearance; Cavg ⫽ average steady-state concentration; CminSS ⫽ predicted steady-state morning trough concentra-
tion.
Pregabalin is a structural analogue of the neuro-
transmitter gamma-aminobutyric acid (GABA), as is
gabapentin, but neither compound interacts with ra-
dioligand binding at either GABA-A or GABA-B re-
ceptors. Pregabalin, with a Ki close to 40 nM, avidly
displaces [3H]gabapentin from the ␣2-␦ subunit of
voltage-gated calcium channels, which are localized
heterogeneously in a number of regions of rat
brain.41-43 With both gabapentin and pregabalin, this
binding activity correlates with decreased calcium
channel function44,45 and decreased release of several
neurotransmitters, including glutamate, noradrena-
line, and substance P.46-48 This decreased neurotrans-
mitter release is associated with reduced
hyperexcitability in several neuronal models49 and
may provide the mechanism of action of the analge-
sic effects of these two medications.
In humans, the rate of gabapentin absorption is
relatively slow, with peak plasma concentrations oc-
curring around 3 hours postdose, and the extent of
absorption is saturable.50,51 These observations are
consistent with gabapentin being a substrate for Sys-
tem L transporter; this transporter is most likely the
predominant pathway for the absorption of gabapen-
tin.52 For pregabalin, the rate of absorption is rapid,
with peak plasma concentrations occurring within 1
hour postdose, and the extent of absorption is pro-
portional to dose.53,54 Pregabalin is also a substrate
for the System L transporter. Studies using in situ
intestine perfusion or brush-border membrane vesi-
cles suggest that multiple amino acid transport sys-
tems may be involved in the small intestinal
absorption of pregabalin.55 It is most likely the mul-
tiple amino acid transport systems that provide pre-
gabalin with dose-proportional absorption and linear
pharmacokinetics.
Gabapentin has demonstrated efficacy in the treat-
ment of PHN,8,9 as have tricyclic antidepressants,6,7
opioid analgesics,10,11 and lidocaine patch 5%.12 To com-
pare the efficacy of different medications used in treat-
ing PHN, number-needed-to-treat (NNT) analyses
have been conducted.56-60 The NNT for pregabalin-
associated decreases in endpoint mean pain scores
ⱖ30% was 2.7 and for decreases in endpoint mean pain
scores ⱖ50% it was 3.4. These NNT are comparable to
what has been reported in previous studies of gabapen-
1282 NEUROLOGY 60 April (2 of 2) 2003
Pregabalin 300 mg/day, Pregabalin 600 mg/day,
CLcr ⬎30 and ⱕ60 mL/min CLcr ⬎60 mL/min
2.44–4.8 0.244–18.6
1.00–6.67 0.75–17.8
6.64 (4.05–14.4) 8.36 (3.45–15.7)
4.69 (2.97–13.1) 5.27 (1.31–11.4)
42.7–52.2
29.8–34.2
tin, tricyclic antidepressants, and opioid
anagesics.11,56-60 Likewise, although few systematic
comparisons of differences in side effects among the
medications used in the treatment of PHN have been
conducted,57 the number needed to harm (NNH) for
pregabalin based on all side effects was 4.3, which is
comparable to what has been reported in previous
studies of gabapentin, tricyclic antidepressants, and
opioid analgesics.10,57,59 However, it is important to note
that except for the studies of gabapentin, previous ran-
domized controlled trials in PHN have been relatively
small studies that have typically used a two-period
crossover research design. Comparison of NNT and
NNH among these different treatments is therefore
problematic, especially given the use of intent-to-treat
analyses in the gabapentin trials and in the current
study but not in the crossover trials of tricyclic antide-
pressants, opioid analgesics, and lidocaine patch 5%.
The results of the current study have demon-
strated that pregabalin is safe and efficacious in the
treatment of PHN. The use of oral medications in
PHN and other neuropathic pain syndromes has long
been guided by the following rule of thumb: start low
and go slow. The results therefore also suggest that
the advantage of pregabalin in comparison to exist-
ing treatments is that it provides clinically impor-
tant, rapid, and durable pain relief without the
necessity of a slow and lengthy titration to an effec-
tive dosage.
Appendix
Participating investigators: Charles Bernick, MD, Las Vegas, NV; E. Rich-
ard Blonsky, MD, Chicago, IL; Paul K. Brownstone, MD, Boulder, CO;
Michael Z. Chesser, MD, Little Rock, AR; Steven Cohen, MD, St. Peters-
burg, FL; Ghassan Kanazi, MD, Robert Dworkin, PhD, Rochester, NY;
Bradley S. Galer, MD, New York, NY; Daniel T. Garber, MD, Asheville, NC;
Gary Gerard, MD, Toledo, OH; Joseph S. Gimbel, MD, Phoenix, AZ; James
Storey, MD, Albany, NY; Sassan Hassassian, MD, Baltimore, MD; Everett
Heinze, MD, Austin, TX; Ronica Kluge, MD, Fort Myers, FL; Montie Rom-
melman, MD, Paducah, KY; W. Alvin McElveen, MD, Bradenton, FL; Anne
L. Oaklander, MD, PhD, Boston, MA; Michael C. Rowbotham, MD, San
Francisco, CA; Jay J. Rubin, MD, Ocala, FL; Richard Singer, MD, Pem-
broke Pines, FL; Brett R. Stacey, MD, David Sibell, MD, Portland, OR;
Michael M. Tuchman, MD, Palm Beach Gardens, FL; Jeanette Wendt, MD,
Tucson, AZ; Naynesh R. Patel, MD, Kettering, OH; Richard Pellegrino, MD,
PhD, Hot Springs, AR; Ruth Fredericks, MD, Jackson, MS.
References
1. Cluff RS, Rowbotham MC. Pain caused by herpes zoster infection. Neu-
rol Clin 1998;16:813– 832.
 2. Dworkin RH, Portenoy RK. Pain and its persistence in herpes zoster.
Pain 1996;67:241–251.
3. Arani RB, Soong S-J, Weiss HL, et al. Phase specific analysis of herpes
zoster associated pain data: a new statistical approach. Stat Med 2001;
20:2429 –2439.
4. Desmond R, Weiss HL, Arani RB, et al. Clinical applications for change-
point analysis of herpes zoster pain. J Pain Symptom Manage 2002;23:
510 –516.
5. Dworkin RH, Schmader KE. The epidemiology and natural history of
herpes zoster and postherpetic neuralgia. In: Watson CPN, Gershon
AA, eds. Herpes zoster and postherpetic neuralgia, 2nd ed. New York:
Elsevier Press, 2001;39 – 64.
6. Max MB. Treatment of post-herpetic neuralgia: antidepressants. Ann
Neurol 1994;35(suppl):S50 –S53.
7. Watson CPN, Vernich L, Chipman M, Reed K. Nortriptyline versus
amitriptyline in postherpetic neuralgia: a randomized trial. Neurology
1998;51:1166 –1171.
8. Rowbotham M, Harden N, Stacey B, Bernstein P, Magnus-Miller L.
Gabapentin for the treatment of postherpetic neuralgia: a randomized
controlled trial. JAMA 1998;280:1837–1842.
9. Rice ASC, Maton S, Postherpetic Neuralgia Study Group. Gabapentin
in postherpetic neuralgia: a randomised, double blind, placebo con-
trolled study. Pain 2001;94:215–224.
10. Watson CPN, Babul N. Efficacy of oxycodone in neuropathic pain: a ran-
domized trial in postherpetic neuralgia. Neurology 1998;50:1837–1841.
11. Raja SN, Haythornthwaite JA, Pappagallo M, et al. Opioids versus
antidepressants in postherpetic neuralgia: a randomized, placebo-
controlled trial. Neurology 2002;59:1015–1021.
12. Galer BS, Rowbotham MC, Perander J, Friedman E. Topical lidocaine
patch relieves postherpetic neuralgia more effectively than a vehicle topi-
cal patch: results of an enriched enrollment study. Pain 1999;80:533–538.
13. Kanazi GE, Johnson RW, Dworkin RH. Treatment of postherpetic neu-
ralgia: an update. Drugs 2000;59:1113–1126.
14. Field MJ, Oles RJ, Lewis AS, McCleary S, Hughes J, Singh L. Gabap-
entin (Neurontin) and S-(⫹)-3-isobutyl GABA represent a novel class of
selective antihyperalgesic agents. Br J Pharmacol 1997;121:1513–1522.
15. Field MJ, Holloman EF, McCleary S, Hughes J, Singh L. Evaluation of
gabapentin and S-(⫹)-3-isobutylGABA in a rat model of postoperative
pain. J Pharmacol Exp Ther 1997;282:1242–1246.
16. Bockbrader HN, Burger PJ, Kugler AR, Knapp LE, Garofalo EA, La-
londe RL. Population pharmacokinetics (PK) analyses of commonly pre-
scribed antiepileptic drugs (AEDs) coadministered with pregabalin
(PGB) in adult patients with refractory partial seizures. Epilepsia 2001;
42(suppl 7):84.
17. Bockbrader HN, Strand JC, Wesche DL, et al. Drug interaction studies
of pregabalin (CI-1008, PGB) in patients with epilepsy maintained on
either valproate (VPA), lamotrigine (LTG), phenytoin (PHT), or car-
bamazepine (CBZ). Epilepsia 2001;42(suppl 7):86.
18. Richter RW, Dworkin RH, Sharma U, et al. Pregabalin is effective in
patients with painful diabetic neuropathy as evaluated in three clinical
trials. Presented at the 10th World Congress on Pain; August 17–22,
2002; San Diego, CA.
19. Melzack R. The Short-Form McGill Pain Questionnaire. Pain 1987;30:
191–197.
20. Bockbrader HN, Burger P, Corrigan BW. Population pharmacokinetics
of pregabalin in healthy volunteers, renally impaired patients, and
patients with chronic pain. Presented at the meeting of the American
Pain Society; March 14 –17, 2002; Baltimore, MD.
21. Hayes RD, Stewart AL. Sleep measures. In: Stewart AL, Ware JE Jr,
eds. Measuring functioning and well-being. Durham, NC: Duke Univer-
sity Press, 1992;235–259.
22. Ware JE Jr, Snow KK, Kosinski M, Gandek B. SF-36-item Short Form
Health Survey (SF-36). I. Conceptual framework and item selection.
Med Care 1992;30:473– 483.
23. McNair DM, Lorr M, Droppleman LF. Profile of Mood States: manual.
San Diego: Educational and Industrial Testing Service, 1981.
24. Guy W. ECDEU assessment manual for psychopharmacology, revised.
Rockville, MD: US Department of Health, Education, and Welfare, 1976.
25. Backonja M, Beydoun A, Edwards KR, et al. Gabapentin for the symp-
tomatic treatment of painful neuropathy in patients with diabetes mel-
litus: a randomized controlled trial. JAMA 1998;280:1831–1836.
26. SAS Institute Inc. SAS/STAT user’s guide, version 6, fourth edition,
volumes 1 and 2. Cary, NC: SAS Institute, 1989.
27. Steel R, Torrie J. Principles and procedures of statistics. New York:
McGraw-Hill Book Company, 1960;308 –310.
28. Berry D, ed. Statistical methodology in the pharmaceutical sciences.
New York: Marcel Dekker, 1990;464:403– 408.
29. Windsor BL, Radulovic LL. Measurement of a new anticonvulsant,
(S)-3-(aminomethyl1)-5-methylhexanoic acid, in plasma and milk by
high-performance liquid chromatography. J Chromatogr (Biomed Appl)
1995;674:143–148.
30. Beal SL, Sheiner LB, eds. NONMEM user’s guides, part I–VII. San
Francisco: University of California San Francisco NONMEM Project
Group, 1988 –1992.
31. Pharsight. WinNonlin version 3.0 user’s guide. Mountain View, CA:
Pharsight Corporation, 1998 –1999.
32. Farrar JT, Young JP, LaMoreaux L, Werth JL, Poole M. Clinical impor-
tance of changes in chronic pain intensity on an 11-point numerical
pain rating scale. Pain 2001;94:149 –158.
33. Serlin RC, Mendoza TR, Nakamura Y, Edwards KR, Cleeland CS.
When is cancer pain mild, moderate, or severe? Grading pain severity
by its interference with function. Pain 1995;61:277–284.
34. Watson CPN. Postherpetic neuralgia: the importance of preventing this
intractable end-stage disorder. J Infect Dis 1998;178(suppl 1):S91–S94.
35. Fields HL, Rowbotham M, Baron R. Postherpetic neuralgia: irritable
nociceptors and deafferentation. Neurobiol Dis 1998;5:209 –227.
36. Rowbotham MC, Petersen KL, Fields HL. Is postherpetic neuralgia
more than one disorder? Pain Forum 1998;7:231–237.
37. Haythornthwaite JA, Benrud-Larson LM. Psychological aspects of neu-
ropathic pain. Clin J Pain 2000;16:S101–S105.
38. Smith MT, Perlis ML, Smith MS, Giles DE, Carmody TP. Sleep quality
and presleep arousal in chronic pain. J Behav Med 2000;23:1–13.
39. Eisenberg E, Luria Y, Braker C, Daoud D, Ishay A. Lamotrigine re-
duces painful diabetic neuropathy: a randomized, controlled study.
Neurology 2001;57:505–509.
40. Finnerup NB, Sindrup SH, Bach FW, Johannesen IL, Jensen TS. Lam-
otrigine in spinal cord injury pain: a randomized controlled trial. Pain
2002;96:375–383.
41. Taylor CP, Vartanian MG, Yuen PW, Bigge C, Suman-Chauhan N, Hill
DR. Potent and stereospecific anticonvulsant activity of 3-isobutyl
GABA relates to in vitro binding at a novel site labeled by tritiated
gabapentin. Epilepsy Res 1993;14:11–15.
42. Gee NS, Brown JP, Dissanayake VU, Offord J, Thurlow R, Woodruff GN. The
novel anticonvulsant drug, gabapentin (Neurontin), binds to the alpha2delta
subunit of a calcium channel. J Biol Chem 1996;271:5768–5776.
43. Hill DR, Suman-Chauhan N, Woodruff GN. Localization of (3-
H)gabapentin to a novel site in rat brain: autoradiographic studies. Eur
J Pharmacol Mol Pharmacol 1993;244:303–309.
44. Fink K, Dooley DJ, Meder WP, et al. Inhibition of neuronal Ca2⫹ influx
by gabapentin and pregabalin in the human neocortex. Neuropharma-
cology 2002;42:229 –236.
45. Martin DJ, McClelland D, Herd MB, et al. Gabapentin-mediated inhi-
bition of voltage-activated Ca-2⫹ channel currents in cultured sensory
neurones is dependent on culture conditions and channel subunit ex-
pression. Neuropharmacology 2002;42:353–366.
46. Dooley DJ, Mieske CA, Borosky SA. Inhibition of K⫹-evoked glutamate
release from rat neocortical and hippocampal slices by gabapentin.
Neurosci Lett 2000;280:107–110.
47. Maneuf YP, Hughes J, McKnight AT. Gabapentin inhibits the sub-
stance P-facilitated K⫹-evoked release of [3H]-glutamate from rat cau-
dal trigeminal nucleus slices. Pain 2001;93:191–196.
48. Dooley DJ, Donovan CM, Pugsley TA. Stimulus-dependent modulation
of [3H]norepinephrine release from rat neocortical slices by gabapentin
and pregabalin. J Pharmacol Exp Ther 2002;296:1086 –1098.
49. Kaneko M, Mestre C, Sanchez EH, Hammond DL. Intrathecally admin-
istered gabapentin inhibits formalin-evoked nociception and the expres-
sion of Fos-like immunoreactivity in the spinal cord of the rat.
J Pharmacol Exp Ther 2000;292:743–751.
50. Bockbrader HN. Clinical pharmacokinetics of gabapentin. Drugs Today
1995;31:613– 619.
51. Bockbrader HN, Breslin EM, Underwood BA, Posvar EL, Sedman AJ.
Multiple-dose, dose-proportionality study of Neurontin® (gabapentin)
in healthy volunteers. Epilepsia 1996;37(suppl 5):159.
52. Stewart BH, Kugler AR, Thompson PR, Bockbrader HN. A saturable
transport mechanism in the intestinal absorption of gabapentin is the
underlying cause of the lack of proportionality between increasing dose
and drug levels in plasma. Pharm Res 1993;10:276 –281.
53. Busch JA, Strand JA, Posvar EL, Bockbrader HN, Radulovic LL. Pre-
gabalin (CI-1008) multiple-dose pharmacokinetics and safety/tolerance
in healthy volunteers. Pharm Sci 1999;1(suppl 4):2033.
54. Alvey CW, Bockbrader HN, Busch JA, et al. Clinical pharmacokinetics
of pregabalin in healthy volunteers. Presented at the 10th World Con-
gress on Pain; August 17–22, 2002; San Diego, CA.
55. Piyapolrungroj N, Li C, Bockbrader H, Liu G, Fleisher D. Mucosal
uptake of gabapentin (Neurontin) vs pregabalin in the small intestine.
Pharm Res 2001;18:1126 –1130.
56. Sindrup SH, Jensen TS. Efficacy of pharmacological treatments of neu-
ropathic pain: an update and effect related to mechanism of drug ac-
tion. Pain 1999;83:389 – 400.
57. Collins SL, Moore RA, McQuay HJ, Wiffen P. Antidepressants and anti-
convulsants for diabetic neuropathy and postherpetic neuralgia: a quanti-
tative systematic review. J Pain Symptom Manage 2000;20:449 – 458.
58. Jensen TS. Anticonvulsants in neuropathic pain: rationale and clinical
evidence. Eur J Pain 2002;6(suppl A):61– 68.
59. McQuay HJ. Neuropathic pain: evidence matters. Eur J Pain 2002;
6(suppl A):11–18.
60. Rice ASC, Maton S. Response to comment on Rice ASC, Maton S, the
Postherpetic Neuralgia Study Group (UK), Gabapentin in postherpetic
neuralgia: a randomized, double-blind, placebo-controlled study, Pain
2001;94:215–224 (letter to the editor). Pain 2002;96:411– 412.
April (2 of 2) 2003 NEUROLOGY 60 1283
 Pregabalin for the treatment of postherpetic neuralgia : A randomized,
placebo-controlled trial
R.H. Dworkin, A.E. Corbin, J.P. Young, Jr., et al.
Neurology 2003;60;1274
DOI 10.1212/01.WNL.0000055433.55136.55
This information is current as of December 16, 2012

Updated Information & including high resolution figures, can be found at:
Services http://www.neurology.org/content/60/8/1274.full.html
References This article cites 44 articles, 7 of which can be accessed free at:
http://www.neurology.org/content/60/8/1274.full.html#ref-list-1

Citations This article has been cited by 35 HighWire-hosted articles:

Subspecialty Collections
Permissions & Licensing
Reprints
http://www.neurology.org/content/60/8/1274.full.html#related-u
rls
This article, along with others on similar topics, appears in the
following collection(s):
All Clinical trials
http://www.neurology.org/cgi/collection/all_clinical_trials
All Pain
http://www.neurology.org/cgi/collection/all_pain
Antiepileptic drugs
http://www.neurology.org/cgi/collection/antiepileptic_drugs
Clinical trials Randomized controlled (CONSORT
agreement)
http://www.neurology.org/cgi/collection/clinical_trials_randomi
zed_controlled_consort_agreement
Neuropathic pain
http://www.neurology.org/cgi/collection/neuropathic_pain
Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://www.neurology.org/misc/about.xhtml#permissions
Information about ordering reprints can be found online:
http://www.neurology.org/misc/addir.xhtml#reprintsus