British Journal of Anaesthesia 97 (3): 36570 (2006)

doi:10.1093/bja/ael182

Advance Access publication July 21, 2006

The effect of addition of intrathecal clonidine to hyperbaric bupivacaine on postoperative pain and morphine requirements after Caesarean section: a randomized controlled trial
I. van Tuijl*, W. A. van Klei, D. B. M. van der Werff and C. J. Kalkman
Department of Anesthesiology, University Medical Center, Utrecht, The Netherlands
*Corresponding author: Department of Anesthesiology, University Medical Center Utrecht/Wilhelmina Childrens Hospital, PO Box 85090, 3508 AB Utrecht, The Netherlands. E-mail: I.vanTuyl@umcutrecht.nl
Background. Intrathecal clonidine prolongs spinal anaesthesia. We investigated the effect of the addition of clonidine (75 mg) to hyperbaric bupivacaine on postoperative morphine consumption after Caesarean section in a randomized controlled double-blind trial. Methods. A group of 106 women received spinal anaesthesia using either bupivacaine 0.5% (2.2 ml) heavy with 0.5 ml normal saline 0.9% (B) or bupivacaine 0.5% (2.2 ml) heavy with clonidine (75 mg) in 0.5 ml normal saline 0.9% (BC). The primary outcome was the total morphine consumption in the first 24 h after surgery. Secondary outcomes were the duration of postoperative analgesia, postoperative pain scores, the need for alfentanil during surgery, block regression, clonidine side-effects and morphine side-effects. Results. Total morphine consumption was similar in both study groups. The mean time to the first analgesic request in the BC group was 129 (SD 13.8) min, compared with 55 (14.2) min in the B group [mean difference (95% CI) 75 (106 to 44) min]. In the BC group 22 (42%) patients had a complete motor block 1 h after surgery compared with 4 (8%) patients in the B group [RR (95% CI) 0.18 (0.070.49)]. Side-effects of intrathecal clonidine were not detected. Conclusions. The addition of clonidine (75 mg) to hyperbaric bupivacaine prolongs spinal anaesthesia after Caesarean section and improves early analgesia, but does not reduce the postoperative morphine consumption during the first 24 h. No clinically relevant maternal or neonatal side-effects were detected. Br J Anaesth 2006; 97: 36570 Keywords: analgesics opioid, morphine, consumption; anaesthesia, obstetric, Caesarean section; anaesthetic techniques, regional, spinal; sympathetic nervous system, clonidine Accepted for publication: June 1, 2006
Downloaded from bja.oxfordjournals.org by guest on March 20, 2011

Spinal anaesthesia with hyperbaric bupivacaine is performed in most Caesarean sections. When only local anaesthetic is used, high doses of postoperative morphine are required to provide adequate postoperative pain relief.1 The high incidence of undesirable morphine side-effects such as pruritus or postoperative nausea and vomiting is well known.2 Similarly, the addition of opioids to the local anaesthetic solution to enhance analgesia has these disadvantages, especially pruritus.2 Several studies have shown that clonidine administered in the epidural space or intrathecally has a substantial antinociceptive effect by its action on the a2-receptor in the dorsal horn of the spinal cord.3 4 Intrathecal clonidine produces dose-dependent analgesia; it has been used successfully as a sole analgesic for pain relief in labour5 and for postoperative

pain treatment after Caesarean section.6 Adding clonidine to intrathecal bupivacaine improves intraoperative analgesia and might prolong the duration of spinal analgesia, resulting in a sparing effect on postoperative morphine consumption.3 7 8 The benet of adding clonidine to bupivacaine to improve postoperative pain relief after Caesarean section has been studied previously.9 Clonidine was added to a combination of bupivacaine and fentanyl and compared with a second combination of bupivacaine, fentanyl, clonidine and morphine. It was not possible to draw inferences about the effects of clonidine as the sole additive to intrathecal bupivacaine from this study. To investigate whether the addition of clonidine (75 mg) to hyperbaric bupivacaine prolongs the duration of spinal analgesia and reduces the need for postoperative

The Board of Management and Trustees of the British Journal of Anaesthesia 2006. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

van Tuijl et al.

self-administration of i.v. morphine after Caesarean section, we conducted a randomized controlled double-blind trial. We hypothesized that adding clonidine (75 mg) to bupivacaine would reduce the postoperative morphine consumption after Caesarean section. This hypothesis has not been directly tested in previous studies.6 8 10

Methods
Patients
The study was approved by the hospital ethics committee of the Wilhelmina Childrens Hospital, University Medical Centre Utrecht and included healthy women (ASA I or II) presenting for an elective Caesarean section. The patients were recruited between November 2003 and August 2005 from a group of 209 possible candidates. Inclusion criteria were 150195 cm height and 50120 kg weight. Patients were excluded if they had conditions that preclude spinal anaesthesia, had a psychiatric disorder, had chronic pain, were on antihypertensive medication or when they were unable to communicate in the Dutch language.

Study procedures
Written informed consent was obtained from all patients. To allow for sufcient time for informed consent, the patients were provided with written information at the outpatient preoperative evaluation clinic a few days before the actual operation. Before induction of spinal anaesthesia, all patients received an i.v. infusion of Ringers lactate (1500 ml) or normal saline and standard intraoperative monitoring was used [ECG, pulse oximetry and non-invasive blood pressure (NIBP)]. Patients were randomly allocated to receive spinal anaesthesia using either bupivacaine 0.5% (2.2 ml) heavy with 0.5 ml normal saline 0.9% (total 2.7 ml) (Group B) or bupivacaine 0.5% (2.2 ml) heavy with clonidine (75 mg) in 0.5 ml normal saline 0.9% (total 2.7 ml) (Group BC). Spinal anaesthesia was performed using a 25-gauge pencil point needle with the patient in the sitting position at the lumbar 34 interspace. Immediately after the intrathecal injection, ephedrine (10 mg) was administered by i.v. in both groups to prevent hypotension. Between the intrathecal injection and the delivery of the child, NIBP was measured every minute and every 3 min thereafter until the end of the procedure. Any subsequent decrease in mean arterial pressure (MAP) of >20% from baseline was treated with an extra dose of 5 mg i.v. ephedrine. If the patient complained of insufcient intraoperative analgesia, alfentanil was administered by i.v. in incremental doses of 250 mg. All recorded physiological variables and intraoperative medications were entered in a case record form. The Apgar scores of the newborn were recorded as usual at 1 and 5 min after birth. Umbilical cord blood gas analysis was routinely performed.

Starting at the end of the Caesarean section (i.e. just before leaving the operating room), each patient received acetaminophen (1000 mg) rectally every 6 h. Immediately after the operation the patient was instructed in the use of an i.v. patient-controlled analgesia (PCA) system (Gemstar, ABBOTT Laboratories) for self-administration of i.v. morphine. The PCA was set with a bolus dose of 1 mg morphine and a lockout interval of 10 min. There was no continuous background infusion. If a patient rated her pain more than 4 on a visual analogue scale (VAS) slide ruler for pain during the recovery period of 60 min, she received a loading dose of morphine (5 mg) by i.v. This dose was repeated once if the VAS did not decrease below 4 within 20 min. The number of demands (D), the number of morphine boluses (MB) the patient actually received (i.e. honoured requests) and the total amount of morphine consumed (TMC) during 24 h after the operation were recorded. Postoperative assessments were performed 1 h after the operation in the recovery room (T1), and at 24 (plus or minus 2 h) postoperatively (T2) on the ward. At each assessment VAS scores were recorded on a 11-point scale using a 10 cm VAS slide ruler. The patient was asked to rate her maximal experienced pain during the rst postoperative day (VASmax). In addition, we asked for the occurrence of nausea, vomiting and pruritus during the rst postoperative day at T2 and recorded them. The level of sedation was recorded using the 5-level observers assessment of alertness/sedation scale (OAA/S scale) at both assessment times.11 In this scale responsiveness, speech, facial expression and glance are judged with a maximum score level of 5 points for each item.

Downloaded from bja.oxfordjournals.org by guest on March 20, 2011

Outcomes
The primary outcome was the total morphine consumption 24 h after surgery (TMC). Secondary outcomes were the duration of postoperative analgesia in minutes (time to rst analgesic request), postoperative pain scores, block regression at T1 (any sign of regression, i.e. could the patient move any muscle in the lower extremities), possible clonidine side-effects (total amount of ephedrine needed to keep the MAP within a range of 20% from baseline and sedation), the need for alfentanil during surgery, morphine side-effects (pruritus, nausea and vomiting) and side-effects in the newborn (Apgar scores and umbilical cord blood gas analysis).

Sample size calculation

Based on published data, women receiving PCA after Caesarean section were expected to consume 35 (SD 19) morphine per 24 h.1 12 13 Using power analysis for continuous variables, and assuming an a level of 0.05 and b error of 0.8, 57 patients were needed per trial arm to detect a clinically meaningful reduction in morphine consumption of 10 mg (i.e. from 35 to 25 mg per 24 h) when clonidine was added to bupivacaine. To account for possible loss to follow-up, it was decided to include 60 patients per trial arm.

366

Addition of clonidine to hyperbaric bupivacaine

Study medication, randomization, blinding and concealment

The study medication supplement, clonidine or normal saline, was packed in labelled boxes at the hospital pharmacy and stored in the refrigerator adjacent to the operating room where the surgery would take place. Every vial and box was labelled with a randomization number according to a computer generated random number list. The hospital pharmacy was not able to repackage the yellow colour coded glass ampoules of clonidine (Boehringer Ingelheim) into unlabelled new glass ampoules, as there was uncertainty regarding the stability of the medication after resterilization. Because the vials containing the study medication therefore were recognizable, the study medication was prepared by a non-involved nurse under sterile conditions.

Table 1 Patient characteristics. Values are mean (SD) or absolute numbers (%). Group B, intrathecal bupivacaine; Group BC, intrathecal bupivacaine plus clonidine Group B (n=53) Age (yr) BMI (kg m2) Twins and triplets (%) Repeat Caesarean sections (%) Duration of surgery (min) 35 30 5 25 41 (1943) (5.7) (9.4) (47) (14.8) Group BC (n=53) 34 29 4 19 40 (2241) (4.1) (7.5) (36) (7.7)

Analysis
SPSS release 12.0.1. was used for statistical analyses. An intention to treat analysis was planned. Data are presented as means and standard deviations (SD) or as counts with percentages. To estimate differences in normally distributed continuous outcome variables, the Students t-test for independent samples was used. Outcomes were reported as a mean difference (MD) with 95% condence interval (95% CI). CIs not including zero were considered statistically signicant. For time-based data, a KaplanMeier survival analysis was done. For binary outcomes, odds ratios (OR) were used with 95% CI. For outcomes with a high occurrence rate we used risk ratios (RR) instead of ORs. CIs not including unity were considered statistically signicant.

Table 2 Postoperative morphine consumption. Values are mean (SD), unless otherwise specied. Group B, intrathecal bupivacaine; Group BC, intrathecal bupivacaine plus clonidine; PCA, patient-controlled analgesia Group B (n=53) Total morphine 28.0 (14.2) consumption (mg) First analgesic 55 (39) request (min) Number of PCA 32 (27) demands Number of PCA 23 (13) boluses Group BC (n=53) 24.7 (13.8) 129 (107) 32 (27) 24 (13) Mean difference (95% CI) 3.3 (2.1 to 8.7) 74.6 (105.6 to 43.5) 0.6 (9.7 to 10.8) 1.5 (6.4 to 3.5)

Downloaded from bja.oxfordjournals.org by guest on March 20, 2011

1.0

Results
In total 209 patients were scheduled for Caesarean section during the study period, of which 89 either did not meet the inclusion criteria or refused to participate, leaving 120 patients for randomization. After randomization, in each trial arm the data of 7 patients could not be analysed leaving 53 patients in each group. One patient withdrew consent after randomization (Group B), in eight patients technical problems with the PCA resulted in missing the data (ve in Group BC and three in Group B) and spinal anaesthesia was converted to general anaesthesia in ve patients (two in Group BC and three in Group B). Reasons for adding general anaesthesia were as follows: placenta praevia with excessive blood loss (n=3), total spinal (n=1) (BC group) and restlessness as a result of severe anxiety (n=1). All blocks were tested before starting the procedure and were deemed adequate for surgery. Patient characteristics are summarized in Table 1. Continuous outcome data were normally distributed. Total morphine consumption, number of PCA demands and number of PCA doses were similar in both study groups (Table 2). The KaplanMeier survival plot for time to rst analgesic request is shown in Figure 1. The time to the rst analgesic request in the BC group was 129 (13.8) min,
Cumulative % of patients that did an analgesic request 0.8

0.6

0.4

0.2

Mean time in minutes (95% CI) clonidine 129 (100158) normal saline 55 (4465)

0.0 0 600 120 240 360 480 Time in minutes to first analgesic request 720

Fig 1 KaplanMeier plot of time to rst analgesic request. In patients in whom clonidine (Group BC) or saline (Group B) was added to the spinal injection.

367

van Tuijl et al.

Table 3 Secondary outcomes. Values are mean (SD) unless otherwise specied. Group B, intrathecal bupivacaine; Group BC, intrathecal bupivacaine plus clonidine; VAS, visual analogue pain score; T1, 1 h after surgery; T2, 24 h after surgery; OAA/S, observers assessment of sedation scale; n.a., not applicable Group B (n=53) Group BC (n=53) Mean difference or odds ratio (95% CI) 2.8 (2.1 to 3.4) 0.1 (0.8 to 0.7) 0.3 (0.5 to 1.1) 1.2 (1.6 to 4.1) 12.1 (8.0 to 16.2) 2.0 (0.2 to 23.3) 3.1 (1.2 to 7.4) n.a. n.a. 0.2 (0.1 to 0.8) 0.6 (0.2 to 1.6) 0.7 (0.1 to 4.1)

Table 4 Neonatal outcomes. Values are mean (SD) or numbers (%). n.a., not applicable Group B (n=53) Apgar score at 1 min Apgar score at 5 min Neonates with Apgar 1 <7 Neonates with Apgar 2 <7 Umbilical artery pH Umbilical artery base excess 8.9 (1.0) 9.8 (0.5) 4 (8) 1 (2) 7.3 (0.1) 5.3 (2.0) Group BC (n=53) 9.0 (0.7) 9.9 (0.3) 0 (0) 0 (0) 7.3 (0.1) 5.4 (2.2) Mean difference (95% CI) 0.1 (0.2 to 0.3) 0.1 (0.1 to 0.3) n.a. n.a. 0.01 (0.02 to 0.03) 0.07 (0.8 to 0.9)

VAS at T1 VAS at T2 Maximal VAS Total dose ephedrine (mg) Mean arterial pressure at T1 (mm Hg) Patients requiring alfentanil during surgery (%) Heart rate T1 (beats min1) OAA/S at T1 OAA/S at T2 Patients with pruritus during rst 24 h (%) Patients with nausea during rst 24 h (%) Patients with vomiting during rst 24 h (%)

3.7 2.6 6.4 18 82

(2.3) (2.0) (2.1) (8.4) (11.7)

0.9 2.6 6.1 17 70

(1.4) (2.0) (2.3) (7.4) (9.5)

2 (4) 80 (11.7) 20 (0) 20 (0) 10 (19) 11 (21) 3 (6)

1 (2) 7 (10.6) 20 (0) 20 (0) 2 (4) 7 (13) 2 (4)

pain scores and prolongs the duration of analgesia and of the motor block, but does not result in a reduced consumption of morphine during the rst postoperative day. In the present study we detected no deleterious sideeffects in mother or child.

Downloaded from bja.oxfordjournals.org by guest on March 20, 2011

Clonidine effects and side-effects

Immediate postoperative analgesia was better with the combination of bupivacaine and clonidine as demonstrated by a signicantly later rst request for analgesia, less need for morphine top-ups in the recovery period and lower VAS scores in the BC group. Approximately 12 h after surgery the analgesic effect of clonidine seems to wear off, which is shown by the average time to rst analgesic request in the BC group of 2 h and a similar total morphine consumption and PCA morphine requests in the two groups. Furthermore, the VAS scores at 24 h after surgery were identical, as were the maximal VAS scores during the rst 24 h. These ndings conrm the results of previous studies. Paech and colleagues9 found that clonidine added to a combination of intrathecal bupivacaine and fentanyl signicantly prolongs the duration of analgesia (time to rst analgesic request), but did not decrease postoperative morphine consumption or VAS scores. In this four arm trial, the addition of a mixture of morphine (100 mg), clonidine (60150 mg) and fentanyl (15 mg) to bupivacaine was found to be the optimal combination to improve postoperative pain relief at the expense, however, of sedation and pruritus in 30% of the patients. Of the patients who received intrathecal morphine, 43% received antipruritic drugs. In all patients and especially in those who received clonidine, drowsiness was seen, which was considered a disadvantage. We assume that this side-effect resulted from the combination of intrathecal opioid (morphine or fentanyl) and clonidine,10 as we did not detect any drowsiness as measured with the OAA/S. Benhamou and colleagues7 studied 78 women scheduled for Caesarean section comparing the intraoperative analgesic effect and the time to rst analgesic request of the addition of 75 mg clonidine or 75 mg clonidine plus fentanyl to hyperbaric bupivacaine. This study also showed that addition of clonidine prolonged the postoperative pain-free

compared with 55 (14.2) min in the B group [MD (95% CI) 75 (106 to 44) min]. Furthermore, patients who received clonidine had a considerably lower probability of receiving an i.v. morphine top-up [RR (95% CI) 0.09 (0.040.33)], resulting in a signicantly lower dose of morphine administered in the recovery room (Table 2). Although the patients in Group B received more morphine in the recovery period, the mean VAS score after 1 h was still signicantly higher compared with the BC group [3.7 and 0.9, respectively, MD (95% CI) 2.8 (2.13.4)] (Table 3). This difference disappeared at T2 (after 24 h). In the BC group 22 (42%) patients had a complete motor block (i.e. had no signs of block regression) 1 h after surgery compared with 4 (8%) patients in the B group [RR (95% CI) 0.18 (0.070.49)]. In the BC group, 31 (58%) patients had a mean arterial pressure below 70 mm Hg, compared with 5 patients (9%) in the B group. Mean arterial pressure at T1 in the BC group was 70 mm Hg compared with 82 mm Hg in the B group [MD (95% CI) 12 (816)], but the amount of ephedrine administered did not differ between the two groups (Table 3). Sedation scores as assessed with the OAA/S at either T1 or T2 did not differ (Table 3). Pruritus was observed more frequently in the B group, whereas the incidence of nausea and vomiting did not differ (Table 3). There were no differences in the outcomes of the newborns (Table 4).

Discussion
Clonidine added to bupivacaine for spinal anaesthesia in Caesarean section improves the immediate postoperative

368

Addition of clonidine to hyperbaric bupivacaine

period. However, addition of fentanyl increased the incidence of sedation and pruritus. We found that the incidence of pruritus was signicantly lower in the BC group compared with the B group (4 and 19%, respectively). This could result from lower peak plasma concentrations of morphine resulting from the much lower need for morphine top-ups in the BC group. The intrathecal clonidine dosage of 75 mg used in our study had no measurable deleterious side-effects in either mother or child. Although MAP was lower in the BC group, this apparently was not considered clinically important by the attending anaesthesiologists, as the mean doses of ephedrine used did not differ between both groups. Furthermore, the average MAP did not decrease >20% from baseline. A higher dose of hyperbaric bupivacaine may have resulted in comparable effects. However, the increased risk of side-effects for mother and child, such as hypotension or total spinal, makes this option less attractive. Intraoperative analgesia was adequate in both groups as only two patients in the B group and one patient in the BC group required additional analgesia (alfentanil).

spinal or epidural analgesia. The intrathecal and epidural use of clonidine has not been associated with any irreversible side-effects as it was rst used in humans in 1984.4 Adding clonidine to bupivacaine does not result in a clinically meaningful reduction in self-administered morphine or improved pain scores. However, it prolongs the postoperative pain-free period. This is a benet that all patients will appreciate. These effects of clonidine are also valuable in prolonged procedures such as repeat Caesarean section. When a procedure unexpectedly takes more time, it is reassuring to use a technique that results in an extended duration of analgesia and motor block, without serious side-effects. This study has resulted in the addition of 75 mg to hyperbaric bupivacaine in every scheduled Caesarean section in ASA I and II patients under spinal anaesthesia in our hospital.
Downloaded from bja.oxfordjournals.org by guest on March 20, 2011

Conclusion
We have demonstrated that addition of 75 mg clonidine to hyperbaric bupivacaine prolongs spinal analgesia and the motor block after Caesarean section and improves early analgesia. In this study, this effect was obtained without clinically relevant maternal or neonatal side-effects. This single 75 mg intrathecal clonidine dose did not reduce the amount of subsequent self-administered i.v. morphine during the rst 24 h after the operation.

Limitations
First, in our study we chose a 75 mg dose of clonidine. A higher intrathecal dose of clonidine (e.g. 150 mg) might have resulted in a further prolongation of analgesia and thus in a clinically relevant morphine sparing effect. However, higher doses of clonidine have been reported to cause important decreases in arterial pressure and marked sedation.4,6,8,10 We considered these side-effects to be undesirable in this population.8 Second, in both study groups the data of 7 patients were incomplete unexpectedly, leaving 53 patients in each group for a per protocol analysis. As a result, the power to detect a clinically meaningful decrease in postoperative morphine consumption of 10 mg per 24 h decreased to 0.77. However, as we found a MD in morphine consumption of only 3.3 mg with the upper level of the 95% CI at 8.7 mg, the probability of overlooking a clinically meaningful difference is negligible. Finally, as in other analgesic trials,9 we expected patients to treat their postoperative pain to an acceptable level by means of the PCA pump, which is expressed by VAS scores. In a minority of patients, a correlation between VAS scores and morphine consumption cannot be detected. It has been reported that addition of opioid background infusions to PCA at different rates did not reduce pain intensity. Probably, the factors responsible for these phenomena are anxiety related.14 Furthermore, breast-feeding women may choose to limit their analgesic doses, and the demands of the new baby means that they choose to be more active after surgery than other surgical patients. For these reasons they cannot be considered to conform to standard post surgical norms.

References
1 Ngan Kee WD, Khaw KS, Wong EL. Randomised double-blind comparison of morphine vs. a morphine-alfentanil combination for patient-controlled analgesia. Anaesthesia 1999; 54: 62933 2 Slappendel R, Weber EW, Benraad B, van Limbeek J, Dirksen R. Itching after intrathecal morphine. Incidence and treatment. Eur J Anaesthesiol 2000; 17: 61621 3 Dobrydnjov I, Axelsson K, Samarutel J, Holmstrom B. Postoperative pain relief following intrathecal bupivacaine combined with intrathecal or oral clonidine. Acta Anaesthesiol Scand 2002; 46: 80614 4 Eisenach JC, De Kock M, Klimscha W. Alpha(2)-adrenergic agonists for regional anaesthesia. A clinical review of clonidine (19841995). Anesthesiology 1996; 85: 65574 5 Chiari A, Lorber C, Eisenach JC, et al. Analgesic and hemodynamic effects of intrathecal clonidine as the sole analgesic agent during rst stage of labor: a doseresponse study. Anesthesiology 1999; 91: 38896 6 Filos KS, Goudas LC, Patroni O, Polyzou V. Intrathecal clonidine as a sole analgesic for pain relief after cesarean section. Anesthesiology 1992; 77: 26774 7 Benhamou D, Thorin D, Brichant JF, Dailland P, Milon D, Schneider M. Intrathecal clonidine and fentanyl with hyperbaric bupivacaine improves analgesia during cesarean section. Anesth Analg 1998; 87: 60913 8 Filos KS, Goudas LC, Patroni O, Polyzou V. Hemodynamic and analgesic prole after intrathecal clonidine in humans. A doseresponse study. Anesthesiology 1994; 81: 591601

Clinical implications
Although it is still not registered for this purpose, clonidine is used routinely in The Netherlands for prolongation of

369

van Tuijl et al.

9 Paech MJ, Pavy TJ, Orlikowski CE, et al. Postcesarean analgesia with spinal morphine, clonidine, or their combination. Anesth Analg 2004; 98: 14606 10 Pan PM, Huang CT, Wei TT, Mok MS. Enhancement of analgesic effect of intrathecal neostigmine and clonidine on bupivacaine spinal anaesthesia. Reg Anesth Pain Med 1998; 23: 4956 11 Chernik DA, Gillings D, Laine H, et al. Validity and reliability of the Observers Assessment of Alertness/Sedation Scale: study with intravenous midazolam. J Clin Psychopharmacol 1990; 10: 24451

12 Montgomery JE, Sutherland CJ, Kestin IG, Sneyd JR. Infusions of subhypnotic doses of propofol for the prevention of postoperative nausea and vomiting. Anaesthesia 1996; 51: 5547 13 Ngan Kee WD, Khaw KS, Ma ML, Mainland PA, Gin T. Postoperative analgesic requirement after cesarean section: a comparison of anesthetic induction with ketamine or thiopental. Anesth Analg 1997; 85: 12948 14 Svedman P, Ingvar M, Gordh T. Anxiebo placebo, and postoperative pain. BMC Anesthesiol 2005; 5: 9

Downloaded from bja.oxfordjournals.org by guest on March 20, 2011

370