British Journal of Anaesthesia 90 (5): 659±64 (2003) DOI: 10.1093/bja/aeg123 Comparison of intrathecal isobaric bupivacaine± and ropivacaine± for Caesarean delivery ² C. OÈ.OÈ gæuèn 1 *, E. N. Kirgiz 1, A. Duman 1,S.OÈ kesli 1 and C. AkyuÈrek 2 1 Department of Anaesthesiology and Intensive Care and 2 Department of Obstetrics and Gynaecology, Faculty of Medicine, SelcËuk University, Konya, Turkey *Corresponding author: SelcËuk UÈ niversitesi Tõp FakuÈltesi Anestezi ve Reanimasyon AD. Akyoku, 42080 Konya, Turkey. E-mail: coztin@hotmail.com Background. This study was designed to evaluate the effects of intrathecal isobaric bupivacaine 0.5% plus and isobaric ropivacaine 0.5% plus combinations in women undergoing Caesarean deliveries. Method. Twenty- ve parturients received ropivacaine 15 mg and 150 mg (RM ) and twenty- ve parturients received bupivacaine 15 mg and 150 mg (BM ) for spinal anaesthesia. Sensory and motor block, haemodynamics, postoperative analgesia, fetal outcomes, and side-effects were evaluated. Results. Intrathecal bupivacaine± and ropivacaine± provided effective sensory anaesthesia and motor block. Time to reach complete motor block was shorter and time to complete recovery from motor block was longer in the BM than the RM (P<0.05). The time to regression of two dermatomes and time for the block to recede to the S2 dermatome were similar in both s (P>0.05). Time to rst complaint of pain and the mean total consumption of tenoxicam were similar in both s (P>0.05). APGAR scores at 1 and 5 min were similar in the two s, as were mean umbilical blood ph values (P>0.05). Hypotension and pruritus were the most common side-effects in both s during the operation. Conclusion. Intrathecal isobaric ropivacaine 0.5% 15 mg plus 150 mg provides suf cient anaesthesia for Caesarean delivery. The ropivacaine± combination resulted in shorter motor block, similar sensory and postoperative analgesia. Br J Anaesth 2003; 90: 659±64 Keywords: anaesthesia, obstetric; anaesthetic techniques, epidural; anaesthetics local, bupivacaine; anaesthetics local, ropivacaine; analgesics opioid, Accepted for publication: November 11, 2002 Ropivacaine is a new long-acting, enantiomerically pure (Senantiomer), amide local anaesthetic with a high pka and low lipid solubility. It is considered to block sensory nerves to a greater degree than motor nerves. 1 The analgesic potency of ropivacaine is reported as 0.60 (0.47±0.75) relative to bupivacaine for labour analgesia. 2 The drug is less cardiotoxic than equivalent concentrations of racemic bupivacaine in vitro and has a signi cantly higher threshold for central nervous system toxicity than racemic bupivacaine. 1 Although systemic toxicity of local anaesthetics is not a problem for intrathecal administration, block characteristics such as onset and duration of analgesia, the quality of muscle relaxation, haemodynamic stability, and sideeffects are important considerations during anaesthesia for Caesarean section (CS). 34 The potential advantages of using ropivacaine compared with bupivacaine remain to be determined. In previous studies, intrathecal ropivacaine was either used for nonobstetric surgery 5±10 or labour analgesia combined with sufentanyl or fentanyl. 11±13 Hyberbaric ropivacaine was used by Chung and colleagues 14 and a dose±response study with isobaric ropivacaine was conducted by Khaw and colleagues 15 in CS anaesthesia. The optimal safe dose and ² Presented at the 2nd World Congress of the World Institute of Pain, IÇstanbul, Turkey, June 2001. Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2003
OÈ gæuèn et al. baricity of ropivacaine for obstetric anaesthesia is still to be determined. An isobaric ropivacaine± combination has not been previously studied and compared with bupivacaine± combination. The aim of this prospective, double-blind, randomized study was to compare the analgesia and spinal block characteristics of isobaric ropivacaine 0.5% plus and isobaric bupivacaine 0.5% plus given intrathecally for CS. Materials and methods After obtaining institutional ethical committee approval and written informed consent, 50 ASA physical status I or II parturients undergoing elective CS of single babies at term were included in the study. Parturients with cardiac disease, haematological disease, diabetes, eclampsia, bleeding or coagulation test abnormalities, fetal distress or known fetal anomalies were excluded from the study. Patients were premedicated with oral sodium citrate 30 ml 30 min before spinal block. Fluid loading was with an i.v. infusion of hydroxyethylstarch 6% solution 10 ml kg ±1. Using a computer-generated randomization table, patients were randomized to one of two s. Group RM received isobaric ropivacaine (Naropin, Astra Zeneca) 0.5% 15 mg plus preservative-free hydrochloride 150 mg. Group BM received isobaric bupivacaine (Marcain, Astra Zeneca) 0.5% 15 mg plus preservative-free hydrochloride 150 mg at room temperature. The study solution was prepared by an anaesthetist not involved in data collection. The investigators who performed the spinal block and who collected the intraoperative and postoperative data, as well as the patients, were blinded to the contents of the study solutions. After entering the operating theatre, patients were monitored (Poet II, Criticare Systems Inc., USA) for heart rate (HR), mean arterial pressure (MAP), peripheral oxygen saturation (Sp O2 ) and respiratory rate (RR). Lactated Ringer's solution was infused at 10 ml kg ±1 h ±1 during surgery. Spinal anaesthesia was performed with the patient in the sitting position, using a 25-gauge Quincke needle (Spinocan, Braun) at the L3±4 or L4±5 interspace. The study solution (3.7 ml) was administered within 30 s (approximately 0.12 ml s ±1 ) with the opening of the needle facing cephalad. The position of the needle was con rmed by aspiration and re-injection of 0.2±0.3 ml cerebrospinal uid before and after administration of the study solution. The patient was turned gently and placed supine with left uterine displacement and a pillow was placed under the head. A fetal heart rate monitor was used until delivery. After the spinal block, HR, RR, Sp O2 and MAP were measured every min until delivery and then every 2 min. Hypotension was de ned as 20% decrease from baseline MAP (baseline MAP was calculated from three measurements taken on the ward before surgery) and was treated with incremental i.v. boluses of ephedrine 5±10 mg. Total ephedrine requirements were recorded. Urine output was also monitored. Supplementary oxygen was given through a facemask. The level of sensory anaesthesia, de ned as the loss of pin-prick sensation (20G hypodermic needle) at midclavicular level, was measured every min until it reached the T5 dermatome level and then every 10 min during surgery. The following variables were recorded: time to initial onset of analgesia, time for sensory block to reach the T5 dermatome, time to maximum cephalad spread of analgesia, highest level of analgesia, time to two-segment regression of analgesic level, and regression of analgesic level to the S2 dermatome. Anaesthesia was considered adequate for surgery if pain sensation was lost at the T5 level. Time to motor block was assessed every min using the Bromage scale (0=no motor block, 3=complete motor block of lower limbs) until complete motor block and then every 30 min until the return of normal motor function. The times to complete motor block and complete recovery were recorded. The motor block score when the sensory level reached T5 was also recorded. Time to rst complaint of pain and total requirements for postoperative (48 h) analgesia (tenoxicam 20 mg i.v.) were recorded. The time from skin incision to delivery was also recorded. After delivery, umbilical blood samples were taken for measurement of umbilical venous blood gases. APGAR scores at 1 and 5 min were evaluated and recorded. In the intraoperative period, side-effects and requirement for sedation or general anaesthesia were noted. Nausea and vomiting were treated with metoclopramide 10 mg i.v. Diphenhydramine 20 mg i.v. was administered for pruritus. Maternal sedation scores were evaluated and recorded during and after surgery until the block wore off, using a 0±3 grading score (0=no sedation, 3=severe sedation, dif cult to arouse). The patient remained in a semi-recumbent position in the post-anaesthesia care unit until full recovery from spinal block. During the rst 48 h after surgery, side-effects, including pruritus, headache, backache, respiratory depression, nausea and vomiting, were recorded. The quality of anaesthesia (judged by the anaesthetists), the quality of muscle relaxation (judged by the surgeon) and the degree of intraoperative comfort (judged by the patient) were recorded as excellent, good, fair or poor. Statistical analysis Data from previous studies, 78 including SD of time to recovery from motor block and time to regression of two segments with ropivacaine, were used for power analysis. To detect a 30-min difference in mean duration of motor block or a 10-min difference in mean duration of twosegment regression between the s (two-sided a of 5% and b of 20%), a size of 16 or 17 was necessary. Oneway and two-way ANOVA for repeated measures, and unpaired and paired t-tests were used for quantitative 660
Isobaric ropivacaine± for Caesarean delivery nominal data, as appropriate. Categorical data were compared using the c 2 -test. P<0.05 was considered signi cant. Results Patient characteristics were similar in the two s (Table 1). Spinal block was successfully performed in all cases and all patients completed the study. No haemorrhage or paraesthesia was observed and none of the patients required supplementary analgesia, sedation or general anaesthesia. Sensory and motor block properties are presented in Table 2. There were no differences between the s in onset time, time to reach the T5 sensory level, maximum level of sensory cephalad spread, time to maximum cephalad spread, and the motor block score when the sensory level reached T5. The time to reach complete motor block was shorter and time to complete recovery from motor block was longer in the BM than the RM (P<0.05). The time to regression of two dermatomes and time for the block to recede to the S2 dermatome were similar in the two s (P >0.05). The mean time to rst complaint of pain (170.6 (SD 34.0) min in the BM and 168.0 (42.0) min in the RM ) and the mean total Table 1 Patient characteristics. Data are mean (SD). There were no signi cant differences between the s Bupivacaine± Ropivacaine± Height (cm) 160.5 (5.7) 160.6 (6.0) Weight (kg) 77.4 (10.2) 76.1 (16.5) Age (years) 29.1 (19±43) 29.4 (21±38) Number of pregnancy 2.4 (1.7) 2.1 (0.8) Number of Caesarean section 0.3 (0.5) 0.6 (0.6) Weeks of gestation 39.7 (2.1) 38.4 (1.3) Operation duration (min) 27.3 (11.5) 29.0 (13.3) Delivery time (min) 3.5 (1.9) 4.1 (2.0) consumption of tenoxicam (46.4 (16.0) mg and 48.8 (14.2) mg, respectively) were similar in the two s (P>0.05). Haemodynamic and respiratory data and sedation scores are presented in Table 3. MAP decreased signi cantly in the BM when compared with the RM 1 min after spinal block (P<0.05). Total ephedrine requirements were higher in the BM than the RM (P<0.05) but the number of patients who required ephedrine was similar in the two s (P>0.05). RR, Sp O2 and sedation scores were similar in both s. Changes in MAP were similar between the s throughout the study. However, MAP 1 min after the block was signi cantly lower in the BM (Table 3). APGAR scores and umbilical venous ph values were within the normal range in both s. The 1-min APGAR scores were 7.8 (1.1) and 7.6 (0.8) in the BM and RM s, respectively, and 5-min APGAR scores were 9.9 (0.2) and 9.9 (0.3) (P>0.05). The umbilical blood ph values of the BM and RM s were 7.29 (0.04) and 7.31 (0.03), respectively (P>0.05). During surgery, vomiting was observed in two patients in the BM and was not observed in any patient in the RM (P>0.05); similar numbers of patients in each experienced nausea. Shivering was not observed in any patient (Table 4). Other side-effects such as pruritus and headache were observed similarly between the s (P>0.05). Quality of anaesthesia and muscle relaxation was similar in both s (P>0.05). Anaesthetist's and surgeon's scores for 24 patients were excellent and good for one patient in the BM and were excellent for all 25 patients in the RM. Patient satisfaction scores were excellent for all patients in the RM ; in the BM, scores were excellent in 23 patients, good in one patient and fair in one patient (P>0.05). Discussion The results of this study show that the ±ropivacaine combination provides similar sensory block and post- Table 2 Characteristics of spinal anaesthesia and postoperative analgesia times. Data are mean (SD). NS=not signi cant Bupivacaine± Ropivacaine± P Sensory block Onset time (min) 0.4 (0.9) 0.3 (1.0) NS Time to T5 (min) 4.9 (2.0) 6.1 (2.5) NS Time to maximum level (min) 7.3 (4.2) 7.2 (4.0) NS Maximum level (range) T4 (T3±5) T4 (T4±5) NS Time to two-segment regression (min) 74.8 (11.5) 72.8 (20.8) NS Time to regression to S2 (min) 170.0 (28.1) 165.2 (37.0) NS Motor block Time to complete block (min) 4.0 (2.0) 5.9 (3.3) <0.05 Complete block at T5 (%) 24 (96) 22 (88) NS Time to complete recovery (min) 220.0 (32.4) 200.2 (34.9) <0.05 Sympathetic block Ephedrine required (n) 10 10 NS Ephedrine dose (mg) 19.7 (8.6) 4.0 (5.2) <0.05 661
OÈ gæuèn et al. Table 3 Haemodynamic, respiratory data and sedation scores. Data are mean (SD). *P<0.05 compared with baseline; NS=not signi cant Bupivacaine± Ropivacaine± P Baseline Heart rate (beats min ±1 ) 102.2 (13.4) 100.8 (16.1) NS Mean arterial pressure (mm Hg) 97.7 (10.4) 97.8 (15.2) NS Sp O2 (%) 98.0 (0.9) 98.0 (0.9) NS Respiratory rate (bpm) 18.4 (4.8) 18.4 (2.1) NS Sedation score 0 (0) 0 (0) NS 1 min after block Heart rate (beats min ±1 ) 105.2 (14.9) 102.0 (14.9) NS Mean arterial pressure (mm Hg) 76.8 (19.7)* 88.7 (14.9)* <0.05 Sp O2 (%) 98.4 (0.8) 98.4 (0.7) NS Respiratory rate (bpm) 19.4 (3.0) 18.8 (2.2) NS Sedation score 0 (0) 0 (0) NS After incision Heart rate (beats min ±1 ) 100.0 (22.3) 99.9 (17.4) NS Mean arterial pressure (mm Hg) 78.6 (19.7)* 83.0 (18.1)* NS Sp O2 (%) 98.6 (0.6) 98.6 (0.5) NS Respiratory rate (bpm) 20.3 (3.8) 18.6 (2.3) NS Sedation score 0.3 (0.4) 0.4 (0.4) NS 10 min after incision Heart rate (beats min ±1 ) 100.8 (17.2) 95.1 (15.3) NS Mean arterial pressure (mm Hg) 86.6 (16.4) 93.7 (13.6) NS Sp O2 (%) 98.5 (0.7) 98.6 (0.5) NS Respiratory rate (bpm) 18.8 (3.4) 18.1 (2.4) NS Sedation score 0.4 (0.6)* 0.4 (0.8)* NS 20 min after incision Heart rate (beats min ±1 ) 95.4 (12.7) 96.7 (14.2) NS Mean arterial pressure (mm Hg) 90.4 (15.6) 93.5 (12.4) NS Sp O2 (%) 98.2 (1.1) 98.5 (0.7) NS Respiratory rate (bpm) 18.6 (3.1) 18.4 (2.5) NS Sedation score 0.4 (0.5)* 0.4 (0.7)* NS Table 4 Side-effects. Values are number of patients (%). There were no signi cant differences between the s Bupivacaine± Ropivacaine± Bradycardia 0 (0) 0 (0) Hypotension 10 (40) 10 (40) Nausea 6 (24) 3 (12) Vomiting 2 (8) 0 (0) Shivering 0 (0) 0 (0) Sedation 11 (44) 9 (36) Respiratory depression 0 (0) 0 (0) Pruritus 16 (64) 18 (72) Backache 0 (0) 0 (0) Headache 2 (8) 1 (4) operative analgesia, shorter duration of motor block, less ephedrine requirement, similar APGAR scores and umblical ph values and similar side-effects in CS operations when compared with the ±bupivacaine combination. The optimal dose and baricity of ropivacaine for Caesarean delivery is unknown. The only study investigating plain isobaric ropivacaine during Caesarean delivery was a dose±response study by Khaw and colleagues, 15 who estimated the ED 50 to be 16.7 mg (14.1±18.8 mg) and the ED 90 as 26.8 mg for CS anaesthesia. This is higher than previous non-obstetric studies. 6±9 It is known that spinal injections in full-term pregnant women produce higherthan-expected levels of anaesthesia. 16 Therefore, the dose of ropivacaine should be kept at the minimum possible. When designing the study, we intended to use an equal dose of ropivacaine and bupivacaine (15 mg). Our ropivacaine dose 14 15 is similar to the ED 50 described by others. Prolonging effective postoperative analgesia after Caesarean delivery is considered to be an advantage of spinal anaesthesia. Chung and colleagues 14 stated that adding an opioid to ropivacaine may improve the quality of analgesia. In our study, the ropivacaine± combination produced similar two-segment regression and time to regression to S2 as the bupivacaine± combination. Although ropivacaine is considered to be less potent than bupivacaine, the similar duration of sensory and postoperative analgesia is interesting. The equipotent ratio between bupivacaine and ropivacaine is considered as 3:2 or 2:1. 257 Therefore, ropivacaine is expected to lead to less cephalad spread than similar bupivacaine doses. 57 Parlow and colleagues 17 demonstrated that adding opioid to local anaesthetics alters the density and spread of local anaesthetics in cerebrospinal uid. Our ndings suggest that cephalad spread and onset time of ropivacaine or bupivacaine with are similar, which is contradictory to 662
Isobaric ropivacaine± for Caesarean delivery our expectations and previous data. 14 Glucose-free ropivacaine 0.5% solution is slightly hypobaric at body temperature (density at 37 C=0.9988); 14 this may have resulted in a higher cephalad spread. Pharmacodynamic studies on ±ropivacaine combinations in spinal anaesthesia are not available, and we do not know why a ropivacaine± combination behaved similarly to the bupivacaine± combination. Further studies are therefore needed. Although motor block is a less important issue in CS anaesthesia, the quality of muscle relaxation during surgery, and early ambulation should be considered. Early studies on the motor block properties of ropivacaine showed less motor depressant effect when compared with bupivacaine. 18 Later clinical studies disputed this nding. 2810 Buggy and colleagues 19 observed unexpectedly high incidence and duration of motor block in patients receiving epidural ropivacaine for postoperative analgesia after spinal anaesthesia for CS. Studies of intrathecal ropivacaine also report similar results when ropivacaine was compared with bupivacaine. 81015 However, the studies comparing ropivacaine and bupivacaine were performed using doses with a 3:2 ratio. 2578 The properties of motor block have not been compared using similar doses. In studies of isobaric ropivacaine 15±25 mg, the intensity and duration of motor block was found to be related to the dose. 81015 In our study, complete motor block occurred in all patients. Although a bupivacaine± combination was associated with a more rapid onset and longer duration of motor block, the differences were not clinically signi cant (220 (32.4) min vs 200 (34.9) min, respectively). We did not record any postoperative neurological symptoms in any of our patients up to 48 h after administration of 20 21 intrathecal ropivacaine. Kristensen and colleagues demonstrated that ropivacaine decreased spinal cord blood ow without causing neurotoxicity in two animal studies. Likewise, in other clinical studies, intrathecal ropivacaine did not induce any neurological symptoms. 2 5±15 The available data regarding neurotoxicity suggest that ropivacaine is a suitable alternative spinal anaesthetic. Iida and colleagues 22 pointed out that the vasoconstrictive effect of ropivacaine on spinal pial vessels could eliminate the need for addition of a vasopressor such as epinephrine into the ropivacaine solutions. This natural vasopressor property of ropivacaine may contribute to the prolongation of analgesia by reducing its local clearance. Hypotension is the most commonly reported adverse event in studies in women undergoing CS with epidural ropivacaine 0.5% or 0.75% solutions, together with labour pain or postoperative pain. 23 24 The incidence of hypotension increases with increasing dose and it is most likely to be a consequence of sympathetic block. 1 In a study by McNamee and colleagues, 10 the different doses of ropivacaine (18.75 and 25 mg) provided a high degree of cardiovascular stability. In our study, although the BM required more ephedrine than the RM, the number of patients requiring ephedrine was similar and the APGAR scores and umbilical blood ph values of the neonates were within normal ranges. References 1 McClellan KJ, Faulds D. Ropivacaine. An update of its use in regional anesthesia. Drugs 2000; 60: 1065±93 2 Capogna G, Celleno D, Fusco P, Lyons G, Columb M. Relative potencies of bupivacaine and ropivacaine for analgesia in labour. Br J Anaesth 1999; 82: 371±3 3 Russel IF. Inadvertent total spinal for Caesarean section. Anaesthesia 1985; 40: 199±200 4 Kestin IG. Spinal anaesthesia in obstetrics. Br J Anaesth 1991; 66: 596±608 5 McDonald S, Spencer SL, Kopacz DJ, Stephenson CA. Hyperbaric spinal ropivacaine. Anesthesiology 1999; 90: 971±7 6 Wahedi W, Nolte H, Klein P. Ropivacaine in spinal anaesthesia. Anaesthesist 1996; 45: 737±44 7 Gautier E, DeKock M, Van Steenberge A, et al. Intrathecal ropivacaine for ambulatory surgery. Anesthesiology 1999; 91: 1239±45 8 Malinovsky JM, Charles F, Kick O, et al. Intrathecal anesthesia: ropivacaine versus bupivacaine. Anesth Analg 2000; 91: 1457±60 9 Van Kleef J, Veering B, Burm AGL. Spinal anesthesia with ropivacaine: double-blind study on the ef cacy and safety of 0.5% and 0.75% solutions in patients undergoing minor lower limb surgery. Anesth Analg 1994; 78: 1125±30 10 McNamee DA, Parks L, McClelland AM, et al. Intrathecal ropivacaine for total hip arthroplasty: double-blind comparative study with isobaric 7.5 mg ml ±1 and 10 mg ml ±1 solutions. Br J Anaesth 2001; 87: 743±7 11 Levin A, Datta S, Camann WR. Intrathecal ropivacaine for labor analgesia: a comparison with bupivacaine. Anesth Analg 1998; 87: 624±7 12 Sony AK, Miller CG, Pratt SD, et al. Low dose intrathecal ropivacaine with or without sufentanyl provides effective analgesia and does not impair motor strength during labour: a pilot study. Can J Anaesth 2001; 48: 677±80 13 Shah MK, Sia AT, Chong JL. The effect of the addition of ropivacaine or bupivacaine upon pruritus induced by intrathecal fentanyl in labour. Anaesthesia 2000; 55: 1008±13 14 Chung CJ, Choi SR, Yeo KH, Park HS, Lee SI, Chin YJ. Hyperbaric spinal ropivacaine for cesarean delivery: a comparison to hyperbaric bupivacaine. Anesth Analg 2001; 93: 157±61 15 Khaw KS, Ngan Kee WD, Wong ELY, Liu JYW, Chung R. Spinal ropivacaine for cesarean section. Anesthesiology 2001; 95: 1346±50 16 Mulroy MF. Regional Anesthesia, 2nd Edn. Boston: Little Brown & Co., 1996; 259 17 Parlow JL, Money P, Chan PSL, et al. Addition of opioids alters the density and spread of intrathecal local anesthetics. An in vitro study. Can J Anaesth 1999; 46: 66±70 18 Bader AM, Datta S, Flanagan H, et al. Comparison of bupivacaine and ropivacaine induced conduction blockade in the isolated rabbit vagus nerve. Anesth Analg 1989; 68: 724±7 19 Buggy DJ, Hall NA, Shah J, Brown J, Williams J. Motor block during patient-controlled epidural analgesia with ropivacaine or ropivacaine-fentanyl after intrathecal bupivacaine for Caesarean section. Br J Anaesth 2000; 85: 468±70 20 Kristensen JD, Karlsten R, Gordh T. Spinal cord blood ow after 663
OÈ gæuèn et al. intrathecal injection of ropivacaine: a screening for neurotoxic effects. Anesth Analg 1996; 82: 636±40 21 Kristensen JD, Karlsten R, Gordh T. Spinal cord blood ow after intrathecal injection of ropivacaine and bupivacaine with or without epinephrine in rats. Acta Anaesthesiol Scand 1998; 42: 685±90 22 Iida H, Ohata H, Iida M, Watanabe Y, Nagase K, Dohi S. Attenuated additional hypocapnic constriction, but not hypercapnic dilation of spinal pial arterioles during spinal ropivacaine. Anesth Analg 1999; 89: 1510±13 23 Bjornestad E, Smedvig JP, Bjerkreim T, et al. Epidural ropivacaine 7.5 mg/ml for elective cesarean section: a double-blind comparison of ef cacy and tolerability with bupivacaine 5 mg/ ml. Acta Anaesthesiol Scand 1999; 43: 603±8 24 Cascio MG, Gaiser RR, Camann WR, et al. Comparative evaluation of four different infusion rates of ropivacaine (2 mg/ ml) for epidural labor analgesia. Reg Anesth Pain Med 1998; 23: 548±53 664