Ropivacaine 0.2% versus bupivacaine 0.1% with fentanyl: a double blind comparison for analgesia during labour

British Journal of Anaesthesia 85 (6): 826±9 (2000)

Ropivacaine 0.2% versus bupivacaine 0.1% with fentanyl: a double blind comparison for analgesia during labour M. Dresner*, J. Freeman, C. Calow, A. Quinn and J. Bamber Department of Anaesthetics, Leeds General In®rmary, Great George Street, Leeds LS1 3EX, UK *Corresponding author We have performed a randomized, double-blind comparison of two epidural drug regimens for analgesia in labour. In the bupivacaine group (BUPIV), 101 healthy parturients received 0.1% bupivacaine with fentanyl 2 mg ml±1. In the ropivacaine group (ROPIV), 102 women received 0.2% ropivacaine. Both groups received an initial loading dose of 15 ml, a continuous infusion of 8 ml h±1, and top-ups of 10 ml. Breakthrough pain not responding to a routine top-up was treated with an `escape' top-up of 10 ml 0.25% bupivacaine. The two groups were compared for complete analgesia at 30 min, routine and `escape' top-up requirements, midwife assessment of analgesic ef®cacy, delivery mode, patient visual analogue scores (VAS) for ®rst and second stage analgesia, overall satisfaction, and patient assessment of motor blockade. Patients receiving ropivacaine received fewer routine top-ups (median 1.0 vs. 2.0, P=0.001) and fewer escape top-ups (9.8% vs. 21.8%, P=0.02). The ropivacaine group was more likely to be pain free in the ®rst stage (51% vs. 33.7%, P=0.01). There were no signi®cant differences in patients' assessment of motor block or mode of delivery between the groups. Pain relief and satisfaction scores from midwives and patients were consistently better in the ropivacaine group, but did not reach statistical signi®cance. Br J Anaesth 2000; 85: 826±9 Keywords: anaesthetic techniques, regional; anaesthetic techniques, epidural; anaesthetics local, ropivacaine; anaesthesia, obstetric Accepted June 28, 2000

The most popular method of maintaining epidural analgesia for labour in the UK is a continuous infusion of low-dose bupivacaine with an opioid.1 The addition of the opioid allows the dose of bupivacaine to be reduced without adversely affecting the quality of analgesia.2 Reducing the dose of bupivacaine signi®cantly reduces the incidence and severity of maternal motor neural blockade.3 Increased mobility during epidural analgesia has been associated with greater maternal satisfaction.3 4 However, the use of opioids in epidural infusions is not without adverse effects. Pruritus is the most common of these,3 and maternal hypoxaemia during the second stage of labour has been reported.5 Ropivacaine has been introduced into clinical practice with the claim that it causes less motor block than bupivacaine.6 7 If true, this raises the possibility of producing the same high quality analgesia, modest motor blockade and high maternal satisfaction with plain ropivacaine as with a low-dose bupivacaine/opioid mixture, thereby avoiding the opioid side effects. The most common bupivacaine/opioid mixture used in the UK is 0.1% bupivacaine with fentanyl 2 mg ml±1.1 From minimum local analgesic concentration (MLAC) studies, it can be

inferred that 0.2% ropivacaine is the analgesic equivalent of 0.1% bupivacaine with fentanyl 2 mg ml±1.8±10 This concentration of ropivacaine has been used effectively in labour epidural analgesia.11 Therefore, we decided to compare 0.2% ropivacaine with 0.1% bupivacaine plus fentanyl 2 mg ml±1 for quality of analgesia, mode of delivery, patient assessment of motor blockade, midwife and maternal satisfaction in a randomised double blind trial.

Methods A power calculation for this study is dif®cult because of the many outcomes under consideration. However, an audit of 570 epidurals prior to this study showed that 76% of women receiving a bupivacaine/fentanyl epidural recorded `excellent satisfaction' (the best of a four point descriptive scale). To ®nd a difference of 20% in the percentage of women recording this score between the study groups we therefore needed to study 200 women to achieve a power of 0.8. Using data from other studies we calculated that this sample size was also suf®cient to detect a difference 20% for women recording minimal motor block.3 12 The study was approved

Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2000

Ropivacaine versus bupivacaine with fentanyl during labour

by the local Clinical Research (Ethics) Committee. Patient information sheets were then distributed in antenatal clinics, parentcraft classes, and all delivery rooms. All women requesting epidural analgesia were considered eligible for the study, excluding those refusing consent or having allergies to any of the study drugs. After providing informed, written consent, randomization was performed by sealed envelope into one of two treatment groups. Using blinded syringes, patients in the BUPIV group received 0.1% bupivacaine with fentanyl 2 mg ml±1, and the ROPIV group received 0.2% ropivacaine. All patients received an initial loading dose of 15 ml, with the ®rst 5 ml acting as a test dose for intrathecal catheter placement. This was followed by an infusion of 8 ml h±1 with top-ups of 10 ml as required. `Escape' top ups of 10 ml 0.25% bupivacaine were prescribed for pain that did not respond to top-ups of the study drugs. Patient characteristics, procedural, and outcome data were collected using the Wansbeck Epidural Audit System as previously described.12 This computer program generates epidural charts, midwife assessment forms, and patient follow-up questionnaires, whilst simultaneously feeding a database. Mode of delivery, midwife assessment of analgesia, `yes/no' for complete analgesia at 30 min after the loading dose, and top-up requirements were recorded contemporaneously. Patient visual analogue scores for ®rst Table 1 Patient characteristics and duration of labour after epidural siting as mean (SD or range) or number. *P value by two-sided unpaired Student t test. ²P value by chi-square test with 1 df.

Age (yr) Height (cm) Weight (kg) Primiparous (n) Multiparous (n) Duration of labour (min)

BUPIV group (n=101)

ROPIV group (n=102)

P

28 (16±44) 159.7 (6.8) 77.5 (15.1) 75 26 413.4 (186.2)

29 (16±43) 161.0 (4.9) 77.0 (11.1) 71 31 400.1 (217.0)

0.54* 0.11* 0.79* 0.45² 0.45² 0.64²

Table 2 Mode of delivery in the BUPIV and ROPIV groups (number). SVD=spontaneous vaginal delivery. OR=odds ratio; *Chi-square test with 1 df

SVD Instrumental Caesarean section

BUPIV group (n=101)

ROPIV group (n=102)

OR (95% CI)

Chi statistic*

P

56 24 21

50 30 22

1.29 (0.75±2.25) 0.75 (0.40±1.40) 0.95 (0.49±1.87)

0.84 0.83 0.02

0.37 0.36 0.89

and second stage pain, their assessment of motor block, and overall satisfaction with the epidural were obtained 24 h after delivery by an anaesthetist blinded to the treatment regimen. Data were collated using Microsoft Access Version 2.0 and Excel Version 5.0. Parametric data were compared using two sided, two sample t-tests and non-parametric data were compared using the Mann-Whitney U test. Discrete data were compared by Chi-Square, with odds ratios and con®dence intervals calculated where appropriate. Signi®cance level was taken at P<0.05. Statistical analysis was performed on SPSS Version 6.0 for Windows.

Results Complete data were obtained for 203 of the 212 patients who were recruited to the study. No patients experienced inadvertent dural puncture, required an epidural resite, or were withdrawn from the study. There were no signi®cant differences between the groups for patient characteristics (Table 1) or mode of delivery (Table 2). Patients in the BUPIV group required signi®cantly more epidural top-ups (median 2.0 vs. 1.0, P=0.001) and were twice as likely to require an escape top-up (21.8% vs. 9.8%, P=0.02) compared with the ROPIV group (Table 3). Whilst there were no differences between the two groups for the proportion of women who were pain-free at 30 min, signi®cantly more women remained pain-free throughout the ®rst stage of labour in the ROPIV group compared with the BUPIV group (51% vs. 33.7%, p=0.01, Table 4). There were no differences between the two groups for pain scores in the second stage of labour. There were no differences between the two groups for midwife and patient overall satisfaction with pain relief (Tables 5 and 6). Whilst a higher percentage of women in the BUPIV group had minimal motor block compared with the ROPIV group, this difference was not statistically signi®cant (Table 7).

Discussion There are several aspects of our study design that warrant discussion. Many of the outcomes are of a subjective nature and were recorded retrospectively. However, the fact that several large UK units use the same clinical audit software indicates that many practitioners consider these outcomes of interest. In practice, patients seem very comfortable with the follow up questionnaire and have no dif®culty in making their choice from the descriptive scales.

Table 3 Epidural top-ups required in the BUPIV and ROPIV groups (number or median (range)). OR=odds ratio; *Chi-square test with 1 df; ²Mann-Whitney U test

No. of patients who required escape top-up No. of top-ups given per patient

BUPIV group (n=101)

ROPIV group (n= 102)

OR (95% CI)

Chi statistic*

P

22 2 (0±14)

10 1 (0±7)

2.56 (1.14±5.73) ±

5.48 ±

0.02* 0.001²

827

Dresner et al. Table 4 Quality of analgesia during labour in the BUPIV and ROPIV groups (number). OR=odds ratio; *Chi-square test with 1 df BUPIV group (n=101)

ROPIV group (n=102)

OR (95% CI)

Chi statistic*

P

Pain free at 30 min

84

81

1.28 (0.63±2.60)

0.47

0.50

Analgesia during 1st stage of labour: VAS=0 mm VAS <10 mm VAS <30 mm Missing data

34 52 77 3

52 61 84 2

0.49 (0.28±0.86) 0.71 (0.41±1.24) 0.72 (0.36±1.43)

6.23 1.42 1.16

0.01 0.24 0.27

Analgesia during 2nd stage of labour: VAS=0 mm VAS <10 mm VAS <30 mm Missing data or sectioned prior to 2nd stage

33 41 61 11

40 46 69 12

0.75 (0.42±1.34) 0.83 (0.48±1.45) 0.73 (0.41±1.30)

0.94 0.42 1.16

0.32 0.53 0.27

Table 5 Midwife assessment of pain relief in the BUPIV and ROPIV groups (number). Chi-square trend=0.81 (1 df), P=0.32. OR=odds ratio; *Chi-square test with 1 df

Excellent Satisfactory Somewhat unsatisfactory Unsatisfactory Missing data

BUPIV group (n=101)

ROPIV group (n=102)

OR (95% CI)

Chi statistic*

P

47 32 10 6 6

54 27 8 3 10

0.77 1.29 1.29 2.1

0.83 0.67 0.26 1.07

0.37 0.40 0.63 0.32

(0.45±1.34) (0.70±2.37) (0.49±3.42) (0.51±8.57)

Table 6 Patient overall assessment of pain relief in the BUPIV and ROPIV groups (number). Chi-square trend=0.18 (1 df), P=0.65. OR=odds ratio; *Chisquare test with 1 df

Excellent Satisfactory Somewhat unsatisfactory Unsatisfactory Missing data

BUPIV group (n=101)

ROPIV group (n=102)

OR (95% CI)

Chi statistic

P

68 23 5 2 3

75 21 1 4 1

0.74 1.14 5.26 0.49

0.94 0.14 2.78 0.67

0.33 0.70 0.10 0.41

(0.40±1.36) (0.58±2.22) (0.60±45.9) (0.10±2.76)

Table 7 Patient assessment of motor block in the BUPIV and ROPIV groups (number). Chi-square trend=0.88 (1 df), P=0.34. OR=odds ratio; *Chi-square test with 1 df Motor block

BUPIV group (n=101)

ROPIV group (n=102)

OR (95% CI)

Minimal Able to walk Mobile in bed Moderate Needed some help to move Dependent on help to move Severe Could just move legs No leg movement at all Missing data

52 8 44 41 29 12 5 2 3 3

43 2 41 52 40 12 5 2 3 2

1.46 (0.84±2.50)

1.77

0.18

0.66 (0.38±1.15)

2.21

0.14

1.01 (0.28±3.60)

<0.01

0.99

828

Chi statistic*

P

Ropivacaine versus bupivacaine with fentanyl during labour

Contemporaneous, hourly visual analogue scores for pain with formal modi®ed Bromage scale motor block assessments might be the academic ideal, but are not feasible in the context of a relatively large study in a busy unit. We argue that subjective assessments of analgesia and motor block, however ¯awed scienti®cally, are how patients judge their experience and therefore should be of primary concern. The focus of this study was to compare the two techniques from the perspective of midwives and patients in the context of a standard technique used in a busy unit. It was never our aim to formally assess pharmacokinetic and dynamic variables such as speed of onset, block heights, and objective motor block. Similarly, we justify the recruitment of all women into this study, regardless of parity and stage of labour, because it is our clinical practice to use a single starting regimen for all women requesting epidural analgesia. No formal comparisons were made of side effects such as pruritis, nausea, vomiting, and hypotension between the groups. This decision was made because it can be safely assumed that the bupivacaine/fentanyl combination will cause pruritis in a percentage of women, and plain ropivacaine will not. An audit of the incidence of clinically signi®cant hypotension using these treatments revealed such a low incidence that a much larger study would be required to separate the treatments on haemodynamic grounds. Similarly, given that the incidence of nausea and vomiting at some time during childbirth approaches 100%, it would be dif®cult to assess the causative contribution of the epidurals. One outcome not considered was the performance of epidurals when converted to anaesthesia for Caesarean section. This is because we have a very low threshold for using spinal anaesthesia despite the presence of an existing epidural in our unit. This policy follows a three year audit showing that all cases of inadequate regional anaesthesia involved topped-up epidurals. Conversion to spinal anaesthesia is not therefore a meaningful measure of epidural performance in our unit. The principal ®ndings of this study are that 0.2% ropivacaine produces better ®rst stage analgesia than 0.1% bupivacaine with fentanyl 2 mg ml±1 and requires fewer routine and escape top-ups. This may be due to the longer duration of action of ropivacaine.13 Whatever the explanation, this challenges our earlier assumption that an opioid is necessary to minimise missed segments, perineal pressure, and to maximise ef®cacy and satisfaction. The increased motor block in the ropivacaine group was not statistically signi®cant, but the study was probably underpowered for this point. For the size of difference in minimal motor block found in this study, a two sided test to detect a signi®cant difference at P<0.05 with a power of 0.8 would have required the recruitment of at least 800 women. However, even if the ropivacaine regimen did cause more motor block, this did not adversely affect patient satisfaction. Thus another widely held assumption, that satisfaction is strongly related to mobility, is challenged.

This study cannot be used in isolation to make comparative judgements about plain ropivacaine and bupivacaine in labour epidurals. To answer questions about relative motor block effects and overall ef®cacy requires a comparison of MLAC equivalent plain solutions, such as 0.2% ropivacaine and 0.12% bupivacaine. So, despite the encouraging results in this study, a ®nal judgement on the value of ropivacaine in obstetric practice is not yet possible. In summary, the results from our study support other work11 suggesting that 0.2% ropivacaine is a suitable choice for labour epidurals. The assumption that an added opioid is always necessary to achieve good analgesia, high maternal satisfaction, and acceptable motor block is challenged and warrants further investigation.

References

829

1 Burnstein R, Buckland R, Pickett JA. A survey of epidural analgesia for labour in the United Kingdom. Anaesthesia 1999; 54: 634±640 2 Chestnut DH, Owen CL, Bates JN, Ostman LG, Choi WW, Geiger NW. Continuous infusion epidural analgesia during labor: a randomized, double-blind comparison of 0.0625% bupivacaine/ 0.0002% fentanyl vs. 0.125% bupivacaine. Anesthesiology 1988; 68: 754±759 3 Russel R, Reynolds F. Epidural infusion of low-dose bupivacaine and opioid in labour. Anaesthesia 1996; 51: 266±273 4 Collis E, Davies DWL, Aveling W. Randomised comparison of combined spinal-epidural and standard epidural analgesia in labour. Lancet 1995; 345: 1413±16 5 Porter JS, Bonello E, Reynolds F. The effect of epidural opioids on maternal oxygenation during labour and delivery. Anaesthesia 1996; 51: 899±903 6 Brockway MS, Bannister J, McClure JH, McKeown D, Wildsmith JAW. Comparison of extradural ropivacaine and bupivacaine. Br J Anaesth 1991; 66: 31±37 7 Zaric D, Axelsson K, Nydahl PA, Philipson L, Samuelsson L. Sensory and motor blockade during continuous epidural infusion of ropivacaine 0.1%, 0.2%, 0.3% in volunteers Ð a double blind study. Reg Anesth 1994; 19 (Suppl. 2S): 60 8 Capogna G, Cellena D, Fusco P, Lyons G, Columb M. Relative potencies of bupivacaine and ropivacaine for analgesia in labour. Br J Anaesth 1999; 82; 371±3 9 Polley LS, Columb MO, Naughton NN, Wagner DS, Cosmas JM. Relative analgesic potencies of ropivacaine and bupivacaine for epidural analgesia in labor. Anesthesiology 1999; 90: 944±50 10 Lyons G, Columb M, Hawthorne L, Dresner M. Extradural pain relief in labour: bupivacaine sparing by extradural fentanyl is dose dependent. Br J Anaesth 1997; 78: 493±497 11 Benhamou D, Hamza J, Eledjam J-J, Dailland P, Palot M, Seebacher J, Milon D, Heeroma K. Continuous extradural infusion of ropivacaine 2 mg ml±1 for pain relief during labour. Br J Anaesth 1997; 78: 748±750 12 Dresner M, Bamber J, Calow C, Freeman J, Charlton P. Comparison of low-dose epidural with combined spinalepidural analgesia for labour. Br J Anaesth 1999; 83: 756±760 13 Parpaglioni R, Capogna G, Celleno D. A comparison between low-dose ropivacaine and bupivacaine at equianalgesic concentrations for epidural analgesia during the ®rst stage of labour. Int J Obstet Anesth 2000; 9: 83±86