Pain is the worst fear of women undergoing cesarean delivery.1 Postcesarean delivery pain hinders the mother's ability to care for and feed her newborn infant. Systemic and neuraxial opioid medications, nonsteroidal anti-inflammatory drugs, and acetaminophen, often in combination, are used to treat pain in this population; however, they do not completely relieve postcesarean delivery pain, and have the potential for debilitating and serious adverse reactions.2 The perioperative use of gabapentin has been shown to decrease acute pain after various surgical procedures,3 including total abdominal hysterectomy,4 a surgical procedure similar to cesarean delivery. Gabapentin is thought to exert its analgesic effect by binding to presynaptic voltage gated Ca2+ channels in the dorsal root ganglia of the spinal cord, where it inhibits the release of excitatory neurotransmitters.3
Chronic pain is being recognized as a complication of cesarean delivery, with one study finding that 6 months after cesarean delivery, 12.3% of patients experience pain severe enough to affect infant care.5 Patients experiencing persistent postcesarean delivery pain are more likely to recall higher levels of pain on the first postoperative day.6 Severe acute postcesarean delivery pain has been associated with persistent pain and postpartum depression 8 weeks after delivery.7 In nonobstetric patients, gabapentin may decrease persistent postsurgical pain, but the results of the small studies conducted to date have been conflicting.3 In one study, gabapentin was shown to reduce the incidence of persistent pain after abdominal hysterectomy.8
Gabapentin use in the pregnant population has been documented, and although the transplacental transfer of gabapentin does occur, no increased risks for adverse fetal or neonatal outcomes have been attributed to its use. In a study of 6 epileptic parturients receiving daily doses of 900 to 2100 mg of gabapentin, there were no neonatal adverse effects from the in utero or breast milk exposure to gabapentin.9 The Gabapentin Pregnancy Registry reviewed 39 women who had received gabapentin during pregnancy, of which 36 had received gabapentin throughout the entire gestation, and concluded that the exposure did not lead to an increased risk for adverse maternal or fetal events.10 There is also a case report describing the administration of gabapentin to a neonate for management of pain.11
This randomized, double-blind, placebo-controlled study was conducted to assess the efficacy of gabapentin, when used as part of a multimodal regimen, to prevent and treat postcesarean delivery pain. We hypothesized that a single dose of gabapentin administered before the spinal anesthetic would significantly reduce acute postoperative pain.
This trial was registered with the United States National Institutes of Health at www.clinicaltrials.gov, under number NCT00573664, and received a No Objection Letter from Health Canada. After institutional research ethics board approval, full-term pregnant women 18 years of age or older, undergoing scheduled cesarean delivery, were approached to participate in this study. We excluded patients with moderate to severe systemic illness as evidenced by an ASA Physical Status score of 3 or higher, patients with contraindications to neuraxial anesthesia or any of the study medications, patients with infectious diseases including human immunodeficiency virus infection and hepatitis, and patients with uncontrolled hypertension or diabetes mellitus. Known IV drug users, women with fetuses having known congenital abnormalities, and women who had taken pain medication in the past week were also excluded.
The only dose-finding study of gabapentin for postoperative pain, performed in patients undergoing lumbar discectomy, showed that increasing the preoperative gabapentin dose to >600 mg does not further decrease postoperative pain scores.12 It is thought that gabapentin exerts its analgesic effects by preemptively decreasing the spinal cord excitation caused by surgical trauma, and is more effective for pain reduction if given 1 to 2 hours preoperatively.3 Patients at our institution do not arrive >1.5 hours before their scheduled cesarean delivery. For these reasons we chose to administer 600 mg of oral gabapentin 1 hour before the start of surgery.
To ensure blinding, the hospital pharmacy department, which was not otherwise involved in the study, placed doses of gabapentin 600 mg, or lactose placebo, in identical blue capsule covers. According to a computer-generated randomization table known only to the pharmacy department, the gabapentin or placebo capsules were then placed in sequentially numbered envelopes. The randomization was done in 4 blocks of 26. After obtaining written informed consent, the investigators assigned each patient an ascending sequential study number. One hour before the scheduled start of their surgeries, each subject was given the medication in the envelope corresponding to their study number, and instructed to take it with a small sip of water. The medications were administered by the study personnel, who were blinded to the patient assignment.
All patients were managed as per the standard institutional protocol for cesarean delivery. Standard intraoperative monitors were used. The patients were acutely administered 10 mL/kg of Ringer's lactate solution and, subsequently, spinal anesthesia was initiated with intrathecal 0.75% hyperbaric bupivacaine 12 mg, fentanyl 10 μg, and morphine 100 μg. Hypotension was treated with phenylephrine or ephedrine at the discretion of the anesthesiologist. Intraoperative analgesia was provided, upon patient request or if the treating anesthesiologist judged it necessary, with IV fentanyl, to a maximum of 100 μg. Intraoperative fentanyl use was recorded, but did not affect patient follow-up. After cord clamping, cefazolin 1 g IV or clindamycin 600 mg IV was given, and 20 U of oxytocin, diluted in 1 L of Ringer's lactate solution, was infused. At the end of the surgery, ketorolac 30 mg IV and acetaminophen 1 g PR were administered, and the patients were transferred to the postanesthesia care unit (PACU) for 2 to 3 hours. In the PACU, pain was treated with IV morphine 2 mg every 5 minutes upon patient request. Postoperative pain management on the ward included oral diclofenac 50 mg every 8 hours and oral acetaminophen 1 g every 6 hours for 72 hours. Breakthrough pain was treated, upon patient request, with nurse-administered subcutaneous morphine 2 mg every 4 hours for the first 24 hours, and then oral morphine 5 mg every 4 hours. Nausea was treated with IM dimenhydrinate 50 mg every 6 hours, and pruritus was treated with IM diphenhydramine 25 mg every 4 hours.
A respiratory therapist is present at all deliveries at our institution and is responsible for the initial evaluation, including assignment of Apgar scores and resuscitation, if required, of the neonate. After birth, the neonates were placed on warmers in the operating room, stimulated and dried. Suctioning of the nasopharynx was performed if necessary. The neonatal team was notified if respiratory support was deemed necessary and determined whether intensive care unit admission was indicated. If the clinical status was stable, the infant was swaddled and given to the parents. Breastfeeding, if chosen by the parents, was initiated in the PACU, with the aid of a trained nurse. On the ward, the parents had access to lactation consultants and breastfeeding information sessions. All the babies were assessed by a pediatrician within the first 48 hours of birth.
Maternal demographics—including age, height, weight, gravidity, parity, systemic illness, and reason for the cesarean delivery—were recorded. Patients were assessed at 6, 12, 24, and 48 hours after induction of spinal anesthesia. The study personnel performed the 6-hour assessment with the patient, and instructed the patient when and how to perform the 12-hour assessment. If the patient was asleep at the 12-hour assessment, she was instructed to perform the assessment as soon as she awoke. The 24- and 48-hour assessments were performed by study personnel, and patients were awakened if asleep. Postoperative pain was assessed using a visual analog scale (VAS) of 0 to 100 mm, with 0 being no pain and 100 being the worst possible pain. Patients rated their pain both at rest, and when asked by study personnel to move from supine to sitting. Satisfaction with pain management was assessed using a numerical rating scale (NRS) from 0 to 10, with 0 being not satisfied and 10 being completely satisfied. Patients were asked by the study personnel to rate their sedation, pruritus, nausea, vomiting, and dizziness on a 4-point scale (0 = absent, 1 = mild, 2 = moderate, or 3 = severe). The use of supplemental opioids and medication for the treatment of nausea and pruritus were recorded. At each assessment point, patients were asked whether they had experienced any breast-feeding difficulties. Patients were contacted by telephone by one of the study personnel 3 months after the surgery, and asked whether they had any pain or abnormal sensation (burning, itchiness, or numbness) at the incision site (yes/no). If the answer was yes, they were asked to rate the pain intensity using a verbal NRS of 0 (no pain) to 10 (worst pain imaginable). If patients answered no, the NRS was assigned to be 0. Patients were also asked whether the pain limited their functioning (yes/no), whether they had required any medication for pain at the incision site in the preceding week (yes/no), and whether they were still breastfeeding (yes/no).
Neonatal demographics—including gestational age, birth weight, gender, Apgar scores at 1 and 5 minutes, umbilical arterial blood gas variables, neonatal intensive care unit (NICU) admission, and admission reason—were recorded.
Maternal and umbilical arterial and venous gabapentin levels were measured in the initial 26 cases, unless umbilical cord blood was being collected for stem cell purposes. One to 2 mL of blood were drawn from the mother with a fresh venipuncture immediately after delivery of the neonate, and from the umbilical artery and vein after double cord clamping. The samples were allowed to clot, and then centrifuged for 15 minutes at 5000 revolutions per minute (rpm). Plasma was removed from the samples and stored at −70°C. Batched samples were sent for analysis to an off-site laboratory (Clinical Research and Clinical Trials Laboratory, Hamilton General Hospital, Hamilton, Ontario, Canada). The high-performance liquid chromatographic assay for gabapentin that was used, based on a published method,13 was able to detect gabapentin concentrations between 1.75 and 234 μmol/L. The laboratory coefficient of variation for a control measurement of 3.5 μmol/L was 0.067.
The primary end point for this study was maternal pain on movement at 24 hours. A 2-week survey in our institution demonstrated that patients who underwent scheduled cesarean delivery and received standard postoperative pain management had a 24-hour mean (SD) pain VAS score on movement of 45 (24) mm. A study of posthysterectomy pain found that adding gabapentin to postoperative morphine and nonsteroidal anti-inflammatory drugs reduced pain on movement by 31%.14 With α of 0.05 and β of 0.8 and using the SD from the above-mentioned survey for each group, a 2-sided t test sample size calculation showed that 50 women per group would be necessary to show a reduction in pain scores by 30% (mean VAS from 45 to 31 mm). We randomized 4 blocks of 26 patients, for a total of 104 patients. Interim analyses were planned after each block. We did not plan any statistical compensation for multiple interim analyses.
VAS scores were compared using a repeated-measures analysis of variance (ANOVA) with group (gabapentin and placebo) as the between factor and time as the repeated-measures factor. The model used constant and equal correlations to estimate between time correlations, i.e., a compound symmetry correlation matrix. Contrasts between the treatment groups were then constructed at each time; hence, the analysis used all the data (not just the 24-hour data) to estimate the variance. This is a more robust estimate of variance. The VAS P values were not adjusted for multiple testing because 24 hours was considered the primary end point. Means and medians were used to estimate central values of continuous data, and proportions were reported for categorical data. Confidence intervals and standard deviations were used to estimate the spread of the estimates and data, respectively. Categorical data were analyzed by χ2 or Fisher exact tests as appropriate and continuous data by F or t tests. Maternal demographic, morphine dosage, and fetal nonparametric continuous data were analyzed using the Mann–Whitney U test. The satisfaction NRS scores were compared at each time point by using the 2-sample Wilcoxon test. Because no time point was planned a priori for testing the secondary outcomes such as satisfaction, and because outcomes were assessed between groups at multiple times, adjustment of the significance level was made to account for multiple testing; P values were set at 0.01 (0.05/4). However, we chose to accept a P value of 0.05 for severe side effects because of their clinical importance.
To assess whether there was an association of sedation with pain and satisfaction scores in the gabapentin group, patients with sedation scores of 2 or 3 at any time (high-sedation group) were compared with patients having scores of 0 or 1 (low-sedation group) using a repeated-measures ANOVA. Sedation scores were dichotomized into high versus low because of the ordinal nature of the sedation responses, and therefore were treated as a fixed effect in the analysis. The VAS and satisfaction scores were compared between the high- and low-sedation groups.
The first interim analysis showed a mean (95% confidence interval) 24-hour movement VAS of 41 (26–56) in the placebo group and 18 (8–29) in the gabapentin group. This 23-mm difference between groups in movement VAS scores was more than expected, because we had hypothesized a reduction in VAS of only 14 mm. The study was terminated after 46 patients because a very low recruitment rate led to prolongation of the study. With the results of all 46 patients—which showed a mean 24-hour VAS of 41 mm for the gabapentin group and of 21 mm for the placebo group, and an SD of 24 mm—the conditional power, i.e., the chance of rejecting the null hypothesis of no difference between the gabapentin and placebo groups assuming that the trend seen in the initial patients would have continued, was 0.99. With the original hypothesis of a 14-mm decrease in 24-hour movement VAS scores in the gabapentin group, an SD of 24 mm, and with the planned 52 patients in each group, the conditional power was 0.98.15 SAS version 9.1 was used in the analysis.
This study was conducted from November 2007 to November 2008. One hundred sixty-six women were approached for this study, and 46 were randomized (Fig. 1). Two patients were excluded from analysis. Because of an extremely low recruitment rate, the study took much longer than anticipated, and because of the movement of study personnel away from the study institution, the study was discontinued after 46 patients.
The maternal characteristics were similar in both groups (Table 1). One patient in the placebo group and no patients in the gabapentin group required fentanyl for intraoperative analgesia. The pain on movement was significantly lower in the gabapentin group at all times. At 24 hours, the mean (95% confidence interval) VAS scores were 21 mm (13–29) and 41 mm (32–51) in the gabapentin and placebo groups, respectively (P = 0.001) (Fig. 2). Pain at rest was significantly less in the gabapentin group at 6 and 12 hours. The satisfaction with pain management was significantly higher in the gabapentin group at 6 and 12 hours (Fig. 3). In the first 24 hours, 5 patients in each group required supplemental parenteral morphine (P = 1.0), and for those patients the mean (95% confidence interval) morphine dose was 3.2 (1.0–5.4) mg for the placebo group and 4.2 (1.1–7.2) mg for the gabapentin group (P = 0.58). From 24 to 48 hours, 3 patients in each group required supplemental oral morphine, and for those patients the mean (95% confidence interval) oral morphine dose was 8 (5–10) mg for the placebo group, and 5 (5–5) mg for the gabapentin group (P = 0.19). The 48-hour incidence of sedation was similar in both groups; however, the incidence of severe sedation was higher in the gabapentin group (19% vs. 0%, P = 0.04) (Table 2). The severe sedation occurred early, because there were no reports of severe sedation in either group at or after the 24-hour assessment time. The 48-hour incidence of all other side effects was similar in both groups. The statistical analysis of patients in the gabapentin group who experienced high sedation scores in comparison with those with low sedation scores demonstrated no significant effect of sedation on VAS or satisfaction. One patient in the gabapentin group and 2 patients in the placebo group did not attempt to breastfeed, and the number (percentage) of patients reporting breastfeeding difficulties was 3 (15%) in the gabapentin group and 5 (24%) in the placebo group (P = 0.7) in the first 24 hours, and was 5 (25%) in the gabapentin group, and 7 (33%) in the placebo group (P = 0.7) in the first 48 postoperative hours.
At 3 months after delivery, 20 patients in the placebo group and 16 patients in the gabapentin group were successfully contacted (Table 3). There was no difference between groups in the incidence of persistent pain or abnormal sensation at the incision site, incidence of pain limiting daily functions, use of medications for pain at the incision site, or the number of patients still breastfeeding.
One neonate from the gabapentin group was admitted to the NICU with the diagnosis of transient tachypnea of the newborn. There were no differences in the Apgar scores, umbilical artery pH, or requirement for neonatal intervention between the groups (Table 4). Maternal and umbilical blood samples were drawn from 24 of the first 26 patients. We did not collect blood from the other 2 because they were collecting umbilical blood for stem cell purposes. The mean (SD) plasma concentrations of gabapentin were 7.5 (3.7), 6.4 (3.3), and 5.1 (3.1) μmol/L for the maternal vein, umbilical vein, and umbilical artery, respectively. The mean (SD) umbilical vein to maternal vein ratio was 0.86 (0.12).
The results of our study suggest that even in the context of a multimodal regimen that includes intrathecal fentanyl and morphine, oral acetaminophen and diclofenac, and systemic opioids for breakthrough pain, a single dose of gabapentin 600 mg given 1 hour before cesarean delivery significantly improves pain scores in the first 48 hours postpartum, and increases patient satisfaction.
We observed a larger than expected reduction in pain scores in this study. We calculated our sample size on the basis of the previously shown reduction of posthysterectomy pain produced by gabapentin in the presence of postoperative NSAID and systemic morphine.14 The reason for our observed greater-than-expected improvement in pain scores may be related to a synergistic effect of gabapentin with intrathecal morphine. The combination of morphine and gabapentin has been shown to enhance analgesia in humans.16 The proposed site of action of gabapentin and intrathecal opioids is at receptors in the dorsal horn of the spinal cord, and their interaction at this site may be synergistic. Indeed, the combination of intrathecal morphine and gabapentin enhances antinociception in rats.17 It may be that the combination of intrathecal morphine and systemic gabapentin provides superior analgesia over systemic morphine and gabapentin.
This study did not demonstrate a reduction in patients requiring supplemental opioids in the gabapentin group. From our initial survey of postcesarean delivery pain in our institution, we had found the incidence of postoperative morphine consumption was very low, and we did not design the study to show a difference in postoperative morphine requirements. Indeed, the use of supplemental opioids in both groups was low.
The differences in the incidence of nausea, vomiting, pruritus, persistent pain, and persistent abnormal wound sensation at 3 months were not different between the groups. However, the sample size was too small to draw definitive conclusions.
A known side effect of gabapentin is sedation, and we used a patient-based assessment to measure sedation, which is a limitation of our study. Patient-based assessments of sedation have been used to assess sedation in other studies of gabapentin for postoperative pain14,18 and studies of other drugs with sedative effects,19 but lack validation. Observer-based sedation scores, such as the observer assessment of analgesia/sedation (OAA/S) scale,20 or the Ramsay scale,21 could have been used. The Ramsay scale is used most often to guide sedation management in the intensive care unit. The OAA/S has been validated, and provides a composite sedation score after assessment of patients' responsiveness, speech, facial expression, and eyes.20 Although the patient-based assessment of sedation that we used lacks studies of validity, its benefits include its ease of use, and its ability to assess the subjective experience of the patient.
The overall incidence of sedation was similar in both groups; however, the patients receiving gabapentin were more likely to report their sedation as severe (19%) than were patients receiving placebo (0%). The severe sedation did not last >24 hours and did not seem to affect the outcome of the patients. Standard nursing care in our institution includes regular monitoring of respiration and pulse oximetry for the first 24 postoperative hours, and no patients in the study had respiratory depression or oxygen desaturation. The sedation that was experienced by patients in the gabapentin group did not seem to interfere with their ability to care for their neonates, as was evidenced by the similar rates of breastfeeding success between groups, even at 3 months. We do not know whether the sedation affected the patients' ability to ambulate. We did not assess ambulation, because in our institution, patients who have undergone a cesarean delivery remain in bed for the first postoperative evening, and therefore the time to first ambulation may be artificially prolonged. The sedation did not interfere with maternal satisfaction, because patients in the gabapentin group were more satisfied with their pain management, suggesting that there may be value in studying smaller doses of the drug. Although increased sedation may have affected the patients' ability to report satisfaction, statistical analysis demonstrated no association between satisfaction and sedation scores.
The increased incidence of severe sedation may have caused patients in the gabapentin group to report lower pain scores, because patients who are more sedated may have reduced ability or reduced motivation to report pain. However, we specifically assessed movement pain by asking patients to move from lying to sitting, which should have limited the amount of bias secondary to sedation. In addition, statistical analysis to test the relationship of pain with sedation showed no correlation between pain and sedation.
Gabapentin crosses the placenta readily and is transferred to the fetus. In a study of pregnant patients on 900 to 2100 mg of daily gabapentin, the umbilical cord to maternal plasma ratio was found to be 1.7.9 This ratio was higher than that found in our study, which may be due to the difference in dose and timing of the administration of the gabapentin. As quoted in a review article, an umbilical to maternal ratio of 0.7 ± 0.1 was reported by the manufacturer after a single dose of 400 mg of gabapentin,22 which is similar to our results.
We did not find any evidence of deleterious effects of gabapentin on the neonates. The Apgar scores, umbilical artery pH, NICU admission, and breastfeeding difficulties were similar in both groups. We did not perform neonatal behavioral assessments because the 2 main tests used for this purpose, the Brazelton Neurobehaviour Assessment Score and the Neonatal Neurologic and Adaptive Capacity Score, have shown limited efficacy as assessment tools.23 Although our results suggest that maternal administration of gabapentin is safe for neonates, our study was not powered to permit firm conclusions.
Although our results are encouraging, many questions remain. This study does not provide any information on the optimal dose and timing of gabapentin administration in this population. Early termination of the study resulted in a small sample size, making the study underpowered to assess the side effects of gabapentin and its effects on chronic pain. The use of a nonvalidated score to assess sedation may have made the sedation results of the study unreliable.
In summary, a single dose of 600 mg of oral gabapentin, given preoperatively, significantly decreases acute postcesarean delivery pain and increases patient satisfaction. Gabapentin in this dosage increases maternal sedation; however, it does not adversely affect the neonate. Our results are encouraging in regard to efficacy, and further studies are warranted to optimize the dose, extent of treatment, and safety of gabapentin in this patient population.
The authors thank Kristi Downey, MSc, Perinatal Anesthesia Research Coordinator, Department of Anesthesia and Pain Management, Mount Sinai Hospital, Toronto, for her valuable help with patient recruitment and follow-up and organization of the study database.
1. Carvalho B, Cohen SE, Lipman SS, Fuller A, Mathusamy AD, Macario A. Patient preferences for anesthesia outcomes associated with cesarean delivery. Anesth Analg 2005;101:1182–7
2. Gadsen J, Hart S, Santos A. Post-cesarean delivery analgesia. Anesth Analg 2005;101:S62–9
3. Tiippana EM, Hamunen K, Kontinen VK, Kalso E. Do surgical patients benefit from perioperative gabapentin/pregabalin? A systematic review of efficacy and safety. Anesth Analg 2007;104:1545–56
4. Turan A, Karamanlioglu B, Memis D, Usar P, Pamukcu Z, Ture M. The analgesic effects of gabapentin after total abdominal hysterectomy. Anesth Analg 2004;98:1370–3
5. Nikolajsen L, Sorensen HC, Jensen TS, Kehlet H. Chronic pain following cesarean section. Acta Anaesthesiol Scand 2004; 48:111–6
6. Kainu JP, Sarvela J, Tiippana E, Halmesmaki E, Korttila KT. Persistent pain after cesarean section and vaginal birth: a cohort study. Int J Obstet Anesth 2010;19:4–9
7. Eisenach JC, Pan PH, Smiley R, Lavand'homme P, Landau R, Houle TT. Severity of acute pain after childbirth, but not type of delivery, predicts persistent pain and post-partum depression. Pain 2008;140:87–94
8. Fassoulaki A, Stamatakis E, Petropoulos G, Siafaka I, Hassiakos D, Sarantopoulos C. Gabapentin attenuates late but not acute pain after abdominal hysterectomy. Eur J Anesth 2006;23:136–41
9. Ohman I, Vitols S, Tomson T. Pharmacokinetics of gabapentin during delivery, in the neonatal period, and lactation: does a fetal accumulation occur during pregnancy? Epilepsia 2005;46:1621–4
10. Montouris G. Gabapentin exposure in human pregnancy: results from the gabapentin pregnancy registry. Epilepsy Behav 2003;4:310–7
11. Behm MO, Kearns GL. Treatment of pain with gabapentin in a neonate. Pediatrics 2001;108:482–4
12. Pandey CK, Navkar DV, Giri PJ, Raza M, Behari S, Singh RB, Singh U, Singh PK. Evaluation of the optimal pre-emptive dose of gabapentin for post-operative pain relief after lumbar diskectomy: a randomized, double-blind placebo-controlled study. J Neurosurg Anesthesiol 2005;17:65–8
13. Leusmeyer GL, Kempf T, Gidal BE, Wiebe DA. Optimized method for determination of gabapentin in serum by HPLC. Ther Drug Monit 1995;17:251–8
14. Gilron I, Orr E, Tu D, O'Neill P, Zamora JE, Bell AC. A placebo-controlled randomized clinical trial of perioperative administration of gabapentin, rofecoxib and their combination for spontaneous and movement-evoked pain after abdominal hysterectomy. Pain 2005;113:191–200
15. Lachin JM. A review of methods for futility stopping based on conditional power. Statist Med 2005;24:2747–64
16. Eckhardt K, Ammon S, Hofmann U, Riebe A, Gugeler N, Mikus G. Gabapentin enhances the analgesic effect of morphine in healthy volunteers. Anesth Analg 2000;91:185–91
17. Shimoyama M, Shimoyama N, Inturris CE, Elliot KJ. Gabapentin enhances the antinociceptive effects of spinal morphine in the rat tail-flick test. Pain 1997;72:375
18. Dierking G, Duedahl TH, Rasmussen ML, Fomsgaard JS, Moiniche S, Romsing J, Dahl JB. Effects of gabapentin on postoperative morphine consumption and pain after abdominal hysterectomy: a randomized, double-blind trial. Acta Anaesthesiol Scand 2004;48:322–7
19. Campagni MA, Howie MB, White PF, McSweeney TD. Comparative effects of oral clonidine and intravenous esmolol in attenuating the hemodynamic response to epinephrine injection. J Clin Anesth 1999;11:208–15
20. Chernik DA, Gillings D, Laine H, Hendler J, Silver JM, Davidson AB, Schwam EM, Siegel JL. Validity and reliability of the observer's assessment of alertness/sedation scale: study with intravenous midazolam. J Clin Psychopharm 1990;10:244–51
21. Ramsay AE, Savege TM, Simpson BRJ, Goodwin R. Controlled sedation with alphaxalone–alphadolone. BMJ 1974;2:656–9
22. Hagg S, Spigset O. Anticonvulsant use during lactation. Drug Saf 2000;22:425–40
23. Camann W, Brazleton TB. Use and abuse of neonatal neurobehavioural testing. Anesthesiology 2000;92:3–5