1 Anaesthetic Department, Memorial Hospital, Darlington DL3 6HX, UK. 2 Anaesthetic Department, University Hospital of Wales and University of Wales College of Medicine, Cardiff CF14 4XW, UK *Corresponding author
Accepted for publication: March 6, 2002
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Mothers booked for elective Caesarean section were given various intrathecal doses of bupivacaine with fentanyl during routine combined spinalepidural anaesthesia.
Results. Using sequential allocation we found that the ED50 for S1 motor block 10 min after intrathecal injection was bupivacaine 7 mg with fentanyl 14 µg (95% CI, 6.27.8 mg). We then used intrathecal bupivacaine 13 mg to look for the ED95. We found the calculated ED97.5 to be bupivacaine 9.7 mg with fentanyl 19.4 µg (95% CI, 8.711.4).
Conclusion. We conclude that testing for S1 motor block 10 min after epidural injection of bupivacaine 10 mg is a reliable test to detect accidental intrathecal injection in the obstetric population.
Br J Anaesth 2002; 88: 4425
Keywords: anaesthesia, obstetric; anaesthetic techniques, epidural; anaesthetic techniques, test doses; anaesthetics local, bupivacaine; spinal cord, motor block
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
This study was designed to evaluate the use of S1 motor block as a reliable and easily reproducible clinical sign that will differentiate between an epidural injection and an accidental subarachnoid injection for labour analgesia.
We used S1 motor block as our clinical test because during epidural anaesthesia there is a sequence of motor blockade, starting with an early lumbar motor block involving hip flexion, and then extending to a sacral motor block involving the plantar flexion of the foot. In spinal anaesthesia, early sacral anaesthesia is characteristic. In other words, S1 blockade appears early after an intrathecal injection and is the last dermatome to be affected after an epidural injection. Yarnell and colleagues1 showed that after 10 ml of 0.25% bupivacaine, only 3% of mothers had S1 motor blockade.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
I.v. access and routine monitoring were established in the anaesthetic room. The non-invasive arterial pressure cuff was set for a 2-min recording cycle. I.v. fluids were infused before and during the anaesthetic (Hartmanns solution 5001000 ml).
The CSE technique was performed in the sitting position at the L3-4 interspace, a standard 16 gauge Tuohy needle was inserted into the epidural space using loss of resistance to air and a 119 mm 27 gauge Becton-Dickinson pencil-point spinal needle was inserted through the epidural needle once the epidural space was located. After injection of the study mixture, the spinal needle was withdrawn and the epidural catheter was inserted. The Tuohy needle was then removed and the mother placed in the full right lateral position. After 5 min, the block was assessed and the mother turned supine with a 15° left lateral tilt. At 10 min, the block was tested again.
The loss of resistance to air was used to increase the reliability of detecting that the intrathecal dose was injected into the subarachnoid space, rather than being misplaced in the epidural space. The CSE technique had to be technically perfect with free flow of cerebrospinal fluid seen at the first pass of the spinal needle and no spillage of the spinal anaesthetic on injection, for the patient to remain in the study.
Arterial pressure was monitored non-invasively on a 2-min cycle and ephedrine was used in 6-mg i.v. boluses if the mother felt light-headed or nauseous or if her arterial pressure fell by 20% from her pre-block base line.
The study was divided into two parts. The first part was to determine the ED50 of S1 motor block, 10 min after spinal injection. The drugs used were 0.25% bupivacaine with fentanyl in a fixed ratio of bupivacaine 1 mg to fentanyl 2 µg. We chose this drug combination to mimic the bupivacainefentanyl mixture that is in common use as an epidural solution for labour analgesia. The first patient in the study was given intrathecal bupivacaine 3 mg with fentanyl 6 µg. At 10 min there was no S1 motor block (non-responder), and the next patient in the study received 4 mg and so on until the next patient did have S1 motor block (responder) (Table 1). The next patient after a responder received bupivacaine 1 mg and fentanyl 2 µg less, while the next patient after a non-responder received bupivacaine 1 mg and fentanyl 2 µg more. From this updown sequential allocation, the ED50 for S1 motor block at 10 min was calculated after 20 patients.
|
The plan for the second part of the study was to use a fixed dose equal to roughly twice the ED50, derived from part one of the study, in 60 patients. The choice of twice the ED50 was based on the rule-of-thumb that sensitivity to injected drugs amongst the population ranges from half the ED50 to twice the ED50. The number 60 was chosen on the grounds that if r patients did not respond, where r was zero or very small, the upper 95% confidence limit of non-responders2 would be between approximately r+3 and r+6 out of 60, i.e. an uncertainty of 510%. In fact, 62 mothers were given bupivacaine 13 mg (plain 0.5% bupivacaine with 26 µg of fentanyl) using the CSE technique. We chose 13 mg rather than 14 mg because we did not want to deviate too far from our routine practice of bupivacaine 12.5 mg and fentanyl 25 µg for elective Caesarean section. S1 motor block was assessed at 10 min, as before. Epidural morphine was given for postoperative analgesia.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Logistic regression analysis4 5 of the results for all 82 patients made use of all the data. This confirmed the Dixon Massey result, ED50 being 7.0 mg (95% CI, 6.27.8) with an estimated ED95 of 9.1 mg (95% CI, 8.210.6) and ED97.5 of 9.7 mg (95% CI, 8.711.4) (Fig. 2).
|
Fifty per cent of mothers in the first part required ephedrine with an average dose of 13.6 mg, whilst 75% of mothers in the second part of the study required on average 17.3 mg. Nine of 10 mothers who developed hypotension in part one of the study, did so after the spinal injection and then tended to need more ephedrine during the epidural top-ups to maintain arterial pressure. Only one mother needed a small 6-mg dose of ephedrine after her epidural top-up; she did not require ephedrine after her intrathecal injection. A wide dose range of epidural 0.5% bupivacaine was given to supplement the intrathecal injection (025 ml (0125 mg)), and the mean time to establish the final block height was 25 min (range 1550 min).
There were no excessive sensory blocks at final block assessment just before surgery commenced in the first part of the study, where a maximum intrathecal dose of bupivacaine 9 mg was used, followed by incremental epidural top-ups. There was, however, an 8% incidence of a sensory block to ice at C6 or above, and a 14% incidence of block to light touch at T2 and above, in part two of the study where only the higher dose intrathecal injection was given (Table 2).
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Historically, the test dose contained a mixture of low volume high dose local anaesthetic and a vasoconstrictor. The vasoconstrictor was used as an indicator of an accidental intravascular injection.6 A vasoconstrictor has never been shown to be very predictive of an intravascular injection and careful aspiration of the epidural catheter has now taken its place. The high concentration, low volume local anaesthetic that has traditionally been used, aimed to identify an accidental subarachnoid block before a high volume, high concentration local anaesthetic was given into the epidural space.
It was suggested that an epidural test dose should be mandatory7 and Kumar and colleagues8 recommended the use of 1.5 ml of 0.5% bupivacaine (7.5 mg). Prince and colleagues9 10 suggested 1.6 ml of 0.5% bupivacaine (8 mg) plus 10 min waiting time in order to avoid false negative tests. Others11 recommended more than 10 yr ago that 10 ml of 0.125% bupivacaine (12.5 mg) is a safe and effective test dose. A review of the literature shows that obstetric anaesthetists use a wide range of test doses on initial epidural placement12 and previous research has suggested a number of regimen ranging from 1.5% lidocaine 3 ml (45 mg) to 0.5% bupivacaine 3 ml (15 mg). No research has been able to demonstrate the safety and reliability profile of any of these regimen, or the exact test that should be performed.
It is now accepted that the main determinant of a so-called test dose is the actual dose of local anaesthetic injected in the epidural space. In other words, it has been shown that when injected intrathecally, 0.1% bupivacaine 10 ml (10 mg) results in the same spread and intensity as 2 ml of 0.5% bupivacaine (10 mg).13 Therefore, the use of low volume, high concentration test doses are not necessary.
Another potential danger of epidural analgesia in labour is that accidental intrathecal injection has been reported many hours after an apparently uneventful epidural block. Obstetric anaesthetists use epidural boluses of 550 mg of bupivacaine for labour pain.11 This wide range of doses would produce widely differing clinical pictures if the catheter migrated into the subarachnoid space. The use of 10-mg top-ups can therefore be recommended for use throughout labour as they give reliable analgesia14 with a margin of safety. Routine assessment of the mothers ankle plantar flexion can be taught to anyone giving the top-ups, allowing early detection of catheter migration. We feel that it would remain a valuable test even if there were some hip flexion weakness.
A spin off from our study design was a comparison between low dose spinal anaesthetics followed by epidural top-ups and high dose spinal anaesthesia for Caesarean section. There was improved haemodynamic stability in the low dose group and fewer high blocks. The spinal blocks were easily supplemented by 0.5% epidural bupivacaine in the epidural space and in most cases there was only a few minutes increase in preparation time. We now use the low dose spinal anaesthetic technique with epidural top-ups, if reducing the risk of hypotension in the mother is important and there is no hurry to establish anaesthesia.
This study looked at only one element, i.e. S1 motor block, to allow reliable early detection of accidental intrathecal injection. Although we do not know what the incidence of S1 motor block is after bupivacaine 10 mg in the epidural space, we can assume it is very low as only 3% had S1 blockade after 25 mg of bupivacaine epidurally.1 We found that there was a very steep dose response curve for S1 motor block. The ED97.5 of 9.7 mg (95% CI, 8.711.4) approaches the result required for a reliable predictive test. Our initial guess of an ED95 from the ED50 of approximately double the dose of bupivacaine grossly over-estimated the result. Individual doses of 13 mg, used in the second part of the study, would be more reliable, but there was a high requirement for ephedrine with a high incidence of excessive sensory and motor blockade. We recognize that the addition of fentanyl may have altered our results and that bupivacaine alone may have a different ED97.5, but we felt justified in our study design because an opioid is very commonly added to bupivacaine for labour analgesia.
We therefore recommend that epidural analgesia can be initiated with an epidural dose of bupivacaine 10 mg in a 0.1% concentration with fentanyl, followed by testing for S1 motor block at 10 min. Assessing the sensory spread of the block and the mothers arterial pressure also remains important. If the test dose is negative, the same dose can be used to establish and then maintain full analgesia in labour as required.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Davies OL, Goldsmith PL, eds. Statistical Methods in Research and Production, 4th Edn. London: Longman Group, 1972; 4457
3 Dixon WJ, Massey FJ jr. Introduction to Statistic Analysis, 4th Edn. Boston, MA: McGraw Hill, 1983; 42641
4 Francis B, Green M, Payne C, eds. The GLIM System-Release 4 Manual. Oxford: Clarendon Press, 1993
5 Healy MJR. GLIM: An Introduction. Oxford: Clarendon Press, 1988
6 Mulroy M, Glosten B. The epinephrine test dose in obstetrics: note the limitations. Anesth Analg 1998; 86: 9235
7 Morgan B. Is an epidural test dose mandatory? Eur J Obstet Gynecol Reprod Biol 1995; 59 (suppl): S578
8 Kumar CM, Dennison B, Panchal HI. Epidural test dose. Anaesthesia 1985; 40: 1023
9 Prince G, McGregor D. Obstetric epidural test doses. A reappraisal. Anaesthesia 1986; 41: 124050
10 Prince GD, Shetty GR, Miles M. Safety and efficacy of low volume extradural test dose of bupivacaine in labour. Br J Anaesth 1989; 62: 50340
11 Van Zundert A, Vaes L, Soetens M et al. Every dose given in epidural analgesia for vaginal delivery can be a test dose. Anesthesiology 1987; 67: 43640
12 Burnstein R, Buckland R, Pickett JA. A survey of epidural analgesia for labour in the United Kingdom. Anaesthesia 1999; 54: 63440
13 Vucevic M, Russell IF. Spinal anaesthesia for Caesarean section 0.125% plain bupivacaine 12 ml compared with 0.5% plain bupivacaine 3 ml. Br J Anaesth 1992; 68: 590595
14 Collis RE, Davies DW, Aveling W. Randomised comparison of combined spinal-epidural and standard epidural analgesia in labour. Lancet 1995; 345: 141316