| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Pharmacology and Anaesthesiology Unit, School of Medicine and Pharmacology, The University of Western Australia; Department of Anaesthesia and Pain Medicine, King Edward Memorial Hospital for Women; and Research Design and Biostatistics Unit, Women and Infants Research Foundation, Perth, Western Australia, Australia.
Address correspondence to Prof. Michael Paech, Department of Anaesthesia and Pain Medicine, King Edward Memorial Hospital for Women, 374 Bagot Rd., Subiaco, Western Australia 6008, Australia. Address e-mail to michael.paech{at}health.wa.gov.au
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: A randomized, double-blind, parallel-group, placebo-controlled trial was performed in 90 women having minor gynecological surgery involving the uterus. Patients received either oral pregabalin 100 mg (Group PG) or placebo (Group C) approximately 1 h before surgery. The primary outcome was pain score in the recovery unit and patients were followed for 24 h.
RESULTS: There was no significant difference between groups for pain experienced in the recovery room (median, interquartile range 16, 0–36 vs 10, 6.5–36 for Groups PG and C, respectively, P = 0.80) or thereafter; nor for recovery room fentanyl requirement (42% Group PG versus 27% Group C, P = 0.12) or the quality of recovery at 24 h postoperatively (median, interquartile range score 17, 17–18 Group PG versus 18, 16.5–18 Group C, P = 0.75). The incidence of posthospital discharge light-headedness, visual disturbance, and difficulty with walking was significantly higher in the pregabalin group.
CONCLUSIONS: A single preoperative dose of 100 mg pregabalin does not reduce acute pain or improve recovery after minor surgery involving only the uterus.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
-aminobutyric acid, which shows analgesic, anticonvulsant, and anxiolytic effects. In many countries, it is approved for the treatment of neuropathic pain, the pharmacological basis of which is presynaptic binding to the 
-2-
subunit of voltage-dependent calcium channels that are widely distributed in the spinal cord and brain. By altering calcium currents, pregabalin reduces or modulates the release of several excitatory neurotransmitters, including glutamate, norepinephrine, substance P, and calcitonin gene-related peptide, producing inhibitory modulation of "over-excited" neurons and returning them to a "normal" state (l,2). Pregabalin is several times more potent than the similar drug, gabapentin. It is rapidly absorbed orally with more than 90% bioavailability, achieves peak plasma levels within 30 min to 2 h, and shows linear pharmacokinetics with low intersubject variability (3,4). The side effect profile is good, with the most common adverse events being dizziness and somnolence (1), and pregabalin has no effect on arterial blood pressure or heart rate (5).
Gabapentin has been found to have antinociceptive activity that translates to a role in the management of postoperative pain (6). Animal models of surgical pain and clinical studies of inflammatory pain in adults demonstrate that these conditions provoke allodynia and hyperalgesia, which are modified by gabapentin, independent of opioid receptor activation (6). Some clinical trials of gabapentin, given before a variety of surgical procedures producing visceral and somatic injury, have found significant reduction in postoperative analgesic requirements and others a reduction in early and late postoperative pain (6). Two recent placebo-controlled studies suggest that pregabalin also has a significant analgesic effect on acute postoperative pain (5,7).
Visceral inflammation and peritoneal irritation result in the release of excitatory amino acids (8), and visceral uterine pain during active labor induces significantly increased concentrations of these excitatory amino acids in lumbar cerebrospinal fluid (9). We postulated that pregabalin might, therefore, show analgesic effects and be a suitable drug for preoperative administration for women having day-case surgery involving cervical and uterine injury.
The objective of this randomized clinical trial was to investigate the effects of a single preoperative dose of pregabalin on acute postoperative pain after minor gynecological surgery. Our alternative hypothesis was that pregabalin would show greater analgesic efficacy than placebo in this setting.
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Using a computer-derived random number sequence, 90 women were allocated by means of sealed opaque envelopes into one of two groups to receive either pregabalin 100 mg (Group PG) or placebo (Group C), prepared by the hospital pharmacy as identical capsules to maintain blinding. The study drug was administered orally approximately 1 h before the estimated time of surgery and no other premedicant drugs were permitted. No investigator or data collector was aware of the group assignment until after all patients had been randomized.
Before surgery in the day-surgical unit, consenting patients were instructed in the use of a visual analog scale (VAS) for pain (0 mm = no pain, 100 mm = worst pain imaginable). A baseline Quality of Recovery score range 0–18 (10) was obtained by interview.
The general anesthetic technique was standardized. All participants were permitted preinduction midazolam if required, were induced with IV propofol after 1 µg/kg of IV fentanyl, and were maintained using sevoflurane in oxygen and air. During surgery, IV acetaminophen 1 g was administered, with provision for further IV fentanyl 25 µg boluses if required, and IV ondansetron 4 mg given for prophylaxis of nausea and vomiting. The type and duration of surgery and total intraoperative fentanyl dose were documented.
In the postanesthesia care unit (PACU), patients were given boluses of 20 or 30 µg of fentanyl, based on an age-stratified IV analgesia protocol, if the pain score was 
3. If a pain score 3 persisted, IV tramadol 50 mg followed by oral diclofenac 50 mg was administered. After discharge, patients were advised to use acetaminophen if required. With prior permission from the patient, a telephone interview was conducted postdischarge at 24 h postoperatively.
The primary outcome measure was predischarge pain, with co-end-points of PACU VAS pain intensity at rest and the area under the curve for VAS pain score until 2 h postoperatively. Pain with movement (coughing) and the intensity of lower abdominal or pelvic cramping was also evaluated. Pain scores were recorded on arousal in the PACU, and at 1, 2, and 24 h postoperatively. The intensity of pain at 24 h was also rated by the patient as none, mild, moderate, or severe and a 0–10 score for all components of the modified Brief Pain Inventory short form (MBPI-SF) (11) was completed. The latter assessed any pain, worst pain, least pain, average pain, and pain relief over 24 h and current pain at 24 h; and interference of pain with activity, mood, walking, relations with people, sleep, coughing, deep breathing, and concentration. Secondary efficacy outcomes included the requirement for fentanyl and the quality of recovery at 2 h postoperatively. We noted analgesic requirements; 0–100 mm VAS sedation scores; the presence of nausea and vomiting; 0–100 mm VAS satisfaction scores; and asked specifically for the presence of possible side effects of headache, light-headedness, visual disturbance, and difficulty walking.
The sample size was based on a desire to detect a clinically significant difference in the pain score. A reduction in score of 20 mm was considered significant. A study of similar patients comparing dextromethorphan, ibuprofen, and placebo demonstrated a mean score of 50 mm (SD 30 mm) in the placebo group (12). The lack of normality of scores was anticipated, and nonparametric tests (Mann-Whitney tests) for the comparison of the primary outcome were planned. To detect a difference of 20 mm between the groups, with 90% power at an of 0.05, a sample size of 37 per group was required (PASS, Power and Sample Size Program 2005, Kaysville, UT). Allowing for 10% dropout rate, a total sample of 90 was recruited.
Descriptive statistics were based on means and standard deviations or nonparametric statistics for continuous data. Medians, interquartile ranges (IQR), and ranges were shown where appropriate. The majority of VAS scores were summarized using percentiles (e.g., 60th [45th-75th]) rather than median and IQR, as the former were better suited, given that an identical median, IQR, and range for each group was a common finding. The MBPI-SF data were highly skewed and were thus analyzed in discrete categories of score 0–3 or 4–10. Univariate comparisons between the groups were based on Mann-Whitney rank sum test for continuous data and 
2 or Fisher's exact test for categorical data. P < 0.05 was considered statistically significant. The analysis was performed using SPSS 12 for Windows (SPSS, Chicago, IL).
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The demographic and intraoperative characteristics of the groups were similar (Table 1).
|
The median (IQR) pain score in the recovery room was not significantly different between groups (16, 0–36 vs 10, 6.5–36 for Group PG and Group C, respectively, P = 0.80) and there was also no significant difference for scores at other assessment times (Table 2). Pain scores with movement or for specific visceral pelvic pain did not significantly differ at any in-hospital assessment or at the 24 h postoperative assessment and were of similar or lower magnitude to global pain scores. The use of postoperative analgesics, the postoperative dose of IV fentanyl, the area under the curve for pain scores to 2 h, and pain ratings at 24 h postoperatively did not differ between groups (Table 2). The MBPI-SF data did not significantly differ between groups (results not shown).
|
The incidence of nausea (20% vs 20%) or vomiting (10% vs 2%, P = 0.13) at 24 h in Groups PG and C, respectively, did not significantly differ, nor did the need for antiemetic treatment until discharge (22% vs 16%, Group PG and Group C, respectively, P = 0.45). There were no significant between-group differences for sedation scores, but Group PG had a significantly higher incidence of light-headedness, visual disturbance, and difficulty with walking at 24 h postoperatively (Table 3).
|
The Quality of Recovery Score at 24 h postoperatively did not significantly differ between groups (median score 17, IQR 17–18 vs 18, IQR 16.5–18 for Group PG and Group C, respectively, P = 0.75). Satisfaction scores also did not differ (completely satisfied 59% vs 64%, P = 0.63 and median scores 100 [IQR 95–100, range 40–100] and 100 [IQR 90–100, range 50–100], P = 0.43 for Group PG and Group C, respectively).
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental models of neuropathic pain and inflammatory hyperalgesia have shown that -aminobutyric acid analogs, such as gabapentin and pregabalin, have antinociceptive and antihyperalgesic properties. It has been suggested that central neuronal sensitization may result in an amplification of postoperative pain (13), and that preoperative administration of gabapentin, before inflammatory trauma or surgical stimulation, may reduce the degree of central sensitization (14). Compared with gabapentin, pregabalin, which is substituted at the third-position to increase its lipid solubility and diffusion across the blood-brain barrier, has better pharmacokinetic properties and fewer drug interactions, due to an absence of hepatic metabolism (2). A study investigating pain relief after dental extraction showed that pregabalin was more effective than ibuprofen in attenuating acute postprocedural pain (7). For patients having lumbar laminectomy, pregabalin before and after surgery was as effective as celecoxib in reducing pain and morphine consumption, and the combination of both drugs was the most effective treatment (5).
In contrast to these trials, we did not find any analgesic effect. There are several possible explanations. First, pain scores in the placebo group in our study were substantially lower than we had predicted and of wide variability. We noted that these patients formed subsets of women experiencing high initial pain scores that reduced over time with treatment, low scores that remained stable or reduced over time, or zero scores initially but increasing scores over time despite treatment. Consequently, the study was under-powered to detect a significant difference in early pain. However, given that the pain experience of the two groups was very similar, based on multiple evaluation tools and at all assessments, we do not consider it likely that routine administration of pregabalin would produce a clinically useful effect in this specific population.
Second, the dose of pregabalin administered may have been subtherapeutic. Hill et al. (7) found no difference between pregabalin 50 mg and placebo, but a statistically significant reduction in pain and pain intensity difference after 300 mg. Reuben et al. (5) also used a larger dose of 150 mg, given on two separate occasions. When used for the treatment of refractory partial seizures, the side effects of gabapentin include somnolence, dizziness, diplopia, confusion, and ataxia (15). These patients were often receiving chronic administration of doses in excess of 600 mg/day. We chose to study a dose of 100 mg because the recommended starting dose is 150 mg/day in two divided doses and because dizziness and somnolence, the most common side effects, generally begin shortly after initiation of dosing. We particularly wished to avoid these side effects in an ambulatory population, but nevertheless found that these and similar effects were common among our patients, affecting their ability to ambulate. It appears that administration of a larger dose, which might further increase the incidence or severity of side effects, would not be appropriate in this particular surgical population. This could be tested in future research, which should include a cost versus benefit analysis, given that the acquisition cost of a starting dose of pregabalin in our hospital is currently 10 times more than that of gabapentin.
Finally, because of experimental support for a possible effect (9), we examined analgesia using a visceral pain model caused by cervical dilation and uterine contraction. Pregabalin has shown efficacy only against acute bone pain (5,7) and it may or may not be less effective against visceral pain than somatic pain, which was specifically avoided in this clinical trial. Further studies, using other patient populations who experience postoperative pain that has a visceral component, would be required to investigate this.
In conclusion, we found no better pain relief than placebo and an increase in side effects, after a single preoperative dose of pregabalin 100 mg in patients have day-case uterine surgery.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
Supported by a research grant from Pfizer Pty Ltd.
Reprints will not be available from the author.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. F. White, H. Kehlet, and S. Liu Perioperative Analgesia: What Do We Still Know? Anesth. Analg., May 1, 2009; 108(5): 1364 - 1367. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Agarwal, S. Gautam, D. Gupta, S. Agarwal, P. K. Singh, and U. Singh Evaluation of a single preoperative dose of pregabalin for attenuation of postoperative pain after laparoscopic cholecystectomy Br. J. Anaesth., November 1, 2008; 101(5): 700 - 704. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
