PLAIN BUPIVACAINE: AN UNPREDICTABLE SPINAL ANAESTHETIC AGENT

Br. J. Anaesth. (1986), 58, 292-296

PLAIN BUPIVACAINE: AN UNPREDICTABLE SPINAL ANAESTHETIC AGENT M. R. LOGAN, J. H. McCLURE AND J. A. W. WILD SMITH SUMMARY Plain bupivacaine 0.5% or 0.75% was injected intrathecally in four groups of 10 patients. Group A received 0.75% bupivacaine 2.7ml at L3J4, group B 0.5% bupivacaine 4 ml at L3/4, group C 0.75% bupivacaine 2.7 ml at L2/3 and group D 0.75% bupivacaine 2ml at L3J4. A very wide range of height of block was found in each group and because of this there were few statistically significant differences between the groups. The mean height of anaesthetic blockade was significantly higher when 20 mg {group C) was injected compared with 15mg (group D). Decreasing the volume of injection from 4ml (group B) to 2.7 ml (group A) did not decrease this variability. By injecting at L2\3 spinal space (group C) the mean maximum level of anaesthesia (77) was four segments higher than group A (77 7) injected at L3/4, and this was reflected in a greater incidence of arterial hypotension.

PATIENTS AND METHODS

Forty patients (ASA I and II) who were to undergo elective surgery to the lower limb or perineum under spinal anaesthesia consented to take part in the study. Each patient was premedicated with temazepam 20 mg by mouth. Before the lumbar puncture was performed a peripheral venous cannula was inserted, but no infusion was started until the end of the period of assessment. Patients were randomly allocated to one of four groups (n = 10) with specific variations in the level of injection, or in the dose, volume and concentration injected (table I). Lumbar puncture was performed with the M. R. LOGAN, B-SC., M.B., CH.B., F.F.A.R.C.S.; J. H. MCCLURB, B.SC., M.B., CH.B., F.F.A.R.C.S.; J. A. W . WUJJSMITH, M.D.,

F.F.A.R.CS. ; Department of Anaesthetics, Royal Infirmary, Lauriston Place, Edinburgh EH3 9YW.

patient in the lateral horizontal position using a 25-gauge spinal needle and a standard midline approach. The chosen spinal space was identified by accepting that a line joining the iliac crests approximated to L3/4. All subarachnoid injections were given at a rate of 1 ml per 7.5 s, immediately after which the patient was gently turned supine. Nerve blockade was assessed by an observer who was unaware of the volume and concentration of the injected solution, or the spinal space used. The cephalad spread of analgesia and anaesthesia, and the degree of motor weakness in the legs were assessed at 5-min intervals for 20 min. Analgesia was denned as the inability to appreciate pinprick, and anaesthesia as the inability to appreciate touch. Motor blockade was determined according to a modified "Bromage Scale" (Bromage, 1965): 0 = ability to raise extended leg against gravity;

Downloaded from http://bja.oxfordjournals.org/ at Université Laval on July 1, 2015

Recently, McClure, Brown and Wildsmith (1982) demonstrated that, irrespective of posture, the subarachnoid administration of a low volume of isobaric local anaesthetic solution resulted in a highly predictable block suitable for lower limb and perineal surgery. Of the solutions readily available in the United Kingdom (and many other countries), the closest to being isobaric is glucose-free plain bupivacaine. However, several authors (Cameron et al., 1981; Chambers, Edstrom and Scott, 1981; Pitkanen et al., 1984) have commented on the variability in the extent (height) of blockade associated with the intrathecal administration of 0.5 % plain bupivacaine. In an attempt to categorize some of the factors that might influence the intrathecal spread of plain bupivacaine, the present study was undertaken to compare the effects of injecting different volumes containing the same dose of plain bupivacaine, and the effects of injecting the same volume at different spinal spaces.

UNPREDICTABILITY OF SPINAL BUPIVACAINE

293

TABLB I. Volume and spinal space of injection in four groups of patients receiving plain bupivacaine: mean ± SD age, weight and height, segmental levels of analgesia and anaesthesia, and motor blockade plus percentage with complete motor block. Heart rate and systolic arterial pressure before spinal injection and after 20 mm are compared

Group

No. of patients Volume injected (ml) Solution % Dose(mg) Spinal interspace

B

C

D

10 2.7 0.75 20.25 L3/4

10 4 0.5 20.00 L3/4

10 2.7

10 2.0 0.75 15.00 L3/4

38±20 70±ll 170±7 T9±4.4 Tll±6.2 80

49 ±10 83 ±14

178±8 T8±3.5 T10±3.7

0.75 20.25 L2/3 47 ±16 77±12 170±12 T6±4.4 T7±3.9

48±17 70 ±14 170±6 Tll±3.1 Ll±4.1

100

100

70

83 ±10 74 ±12

73 ±10 71±9

81±17 70±15

82±8

120±12 100±17

129±14 101 ±25

133 ±24 97 ±26

125±18 109±18

77±13

Group A 0.75 Z 2.7 ml at L3/4

Group B (XSX 4.0mrat L3/4

0.75Z 2.7 ml at L2/3

Group D 0.75 Z 2.0 ml at L 3/4

TO

T8

T6

Ttl

Mean

Group C

FIG. 1. Spread of blockade resulting from the intrathecal injection of plain bupivacaine. The solid vertical bar representi the standard deviation and the broken vertical bar represents the range within the group.

Downloaded from http://bja.oxfordjournals.org/ at Université Laval on July 1, 2015

Age(yr) Weight (kg) Height (cm) Level of analgesia Level of anaesthesia Motor blockade (% with complete block]) Heart rate (beat min"1) Before After 20 min Systolic AP (mm Hg) Before After 20 min

A

BRITISH JOURNAL OF ANAESTHESIA

294

1 = inability to raise extended leg; 2 = inability to flex the knee, 3 = inability to flex the ankle. Arterial pressure (by sphygmomanometry) and heart rate were recorded before the patient was positioned for lumbar puncture, and at 5-min intervals for 20 min after injection. Statistical analyses were made using t tests for paired and unpaired data, as appropriate. RESULTS

Group A

Group B

Group C

Group D

O.7S7. 2.7ml at L 3/4 0.5% 4.0 ml at L 3/4 O.7S7. 2.7ml at L2/3 0.75% 2.0 ml at L 3/4

S5 0

-r

10

20

0

10

10

20

Time after injection (min) FIG. 2. Mean onset of analgesia following intrathecal injection of plain bupivacaine. The solid vertical bar represents the SEM and the broken vertical bar represents the range within each group.

Downloaded from http://bja.oxfordjournals.org/ at Université Laval on July 1, 2015

The patient data are shown in table I. Although the groups were comparable in age and weight, the subjects in group B were significantly taller than those in group D (P < 0.05). The administration of glucose-free bupivacaine 20 mg in groups A (2.7 ml of 0.75% at L3/4), B (4 ml of 0.5 % at L3/4) and C (2.7 ml of 0.75 % at L2/3) produced mean maximum levels of analgesia to T9, T8 and T6, respectively (fig. 1). These differences were not statistically significant. However, bupivacaine 15 mg in group D produced a mean maximum block to T i l which was significantly lower than that obtained in group C (P < 0.02). The rate of onset of analgesia (fig. 2) was comparable in groups A, B and C, but was slower in group D than in group C at 5, 10 and 15 min (P < 0.01). By injecting the solution at

L2/3 (group C) rather than at L3/4 (group A) the onset of anaesthesia (fig. 3) was significantly more rapid 5 min after the injection (P < 0.05). However, by 10 min the two groups were indistinguishable. By decreasing the injected dose to 15 mg in group D, the spread of anaesthesia (fig. 3) was slowed at all stages (P < 0.01). All the blocks were symmetrical: no patient developed a block that varied by more than one segment on either side of the body. Groups A and D each had two patients in whom blockade was of insufficient height to permit surgery on the lower limbs. In these patients anaesthesia was achieved either by repeating the spinal anaesthetic or by the induction of general anaesthesia. All patients in groups B and C had complete motor blockade of the lower limbs by 20 min. Two patients in group A and 3 in group D retained the ability to flex the knee or ankle (fig. 4). Eight patients had decreases in systolic arterial pressure of more than 30%, giving symptoms of dizziness or nausea, or both. In four of these the injection was placed at the higher level of L2/3, and groups A and B each contained two of the remaining four patients. Hypotension was treated with ephedrine 6 mg i.v. Two patients with signs of mild vasovagal overactivity not sufficient to

UNPREDICTABILITY OF SPINAL BUPIVACAINE Group A

Group B

295

Group C

Group D

0.75% 2.7 ml at L3/4 0.5 % 4.0 ml at L 3/4 0.75^2.7 ml at L2/3 0.75% 2.0 ml at L 3/4 T2-

T6-

T10

20

0

10

20

FIG. 3. Mean onset of anaesthesia following intrathecal injection of plain bupivacaine. The solid vertical bar represents the SEM and the broken vertical bar represents the range within each group.

Group A

Group B

Group C

0.5% 4.0 ml at L 3/4

0.75% 2.7ml at L 3/4

Group D

0.75% 2.7 ml at L 2/3 0.75% 2.0 ml at L 3/4

3-

8

2 a 10

20

0

20

0

10

20

0

Onset time (min) FIG. 4. Mean onset of motor blockade ( ± range) of the lower limbs following intrathecal injection of plain bupivacaine.

warrant ephedrine, received midazolam 5 mg i.v. noted poor predictability of the spread of analgesia once the pinprick observations were complete, (Cameron etal., 1981; Chambers, 1982; Axelsson, Edstrdm and Widman, 1984; Pitkanen et al., with subsequent remission of the signs. 1984). Spinal anaesthesia using isobaric amethocaine has been shown to produce more predictable DISCUSSION blockade if a small volume of a concentrated Since the introduction of plain bupivacaine as a solution is injected slowly (McClure, Brown and spinal anaesthetic solution, several authors have Wildsmith, 1982). A preliminary study suggested

Downloaded from http://bja.oxfordjournals.org/ at Université Laval on July 1, 2015

10

Time after injection (min)

296

produce greater CSF displacement in some patients and, therefore, contribute to the unpredictability of blockade. This unpredictability with a 4-ml intrathecal injection has previously been reported with isobaric amethocaine (McClure, Brown and Wildsmith, 1982). We conclude that the unpredictability of plain bupivacaine is of clinical importance, since 10% of our patients required supplementary anaesthesia to permit surgery on the lower limbs. Injection at L2/3 may reduce this "failure rate", but at the expense of a higher incidence of hypotension requiring haemodynamic support.

REFERENCES Axelsson, K. H., Edstrom, H. H., and Widman, G. B. (1984). Spinal anaesthesia with glucose-free 0.5% bupivacaine: Effects of different volumes. Br. J. Anaesth., 56, 271. Bromage, P. R. (1965). A comparison of the hydrochloride and carbon dioxide salts of lidocaine and prilocaine in epidural analgesia. Acta Anaesthesiol. Scand., (Suppl. XVI), 55. Brown D. T., Wildsmith, J. A. W., Covino, B. G., and Scott, D. B. (1980). Effect of baricity on spinal anaesthesia with amethocaine. Br. J. Anaath., SI, 589. Cameron, A. E., Arnold, R. W., Ghoris, M. W., and Jamieson, V. (1981). Spinal analgesia using bupivacaine 0.5 per cent plain. Variations in the extent of block with patient age. Anaesthesia, 36, 318. Chambers, W. A. (1982). Intrathecal bupivacaine. Br. J. Anaesth., 54, 799. Edstrom, H. H., and Scon, D. B. (1981). Effect of baricity on spinal anaesthesia with bupivacaine. Br. J. Anaesth., 53, 279. McClure, J. H., Brown, D. T., and Wildsmith, J. A. W. (1982). Effect of injected volume and speed of injection on the spread of spinal anaesthesia with isobaric amethocaine. Br. J. Anaesth., 54, 917. Pitkanen, M., Haapaniemi, L., Tuominen, M., and Rosenberg, P. H. (1984). Influence of age on spinal anaesthesia with isobaric 0.5 % bupivacaine. Br. J. Anaesth., 56, 279. Tuominen, M., Kalso, E., and Rosenberg, P. H. (1982). Effects of posture on the spread of spinal anaesthesia with isobaric 0.75% or 0.5% bupivacaine. Br. J. Anaesth., 54, 313. Wildsmith, J. A. W., McClure, J. H., and Brown, D. T. (1983). (Correspondence) Ann. R. Col. Surg. Lond., 65, 278. Scott, D. B. (1981). Effects of posture on the spread of isobaric and hyperbaric amethocaine. Br. J. Anaesth., 53, 273.

Downloaded from http://bja.oxfordjournals.org/ at Université Laval on July 1, 2015

that 0.75 % plain bupivacaine might produce more predictable blockade than 0.5% bupivacaine (Wildsmith,McClureandBrown,1983).However, this has proved not to be the case in the present study. Indeed, the range of segmental blockade with bupivacaine 20 mg was marginally narrower when the volume of injection was increased to 4 ml. The outstanding feature of the present doubleblind study is the wide range of spread of anaesthesia and analgesia, despite the use of a very strictly controlled anaesthetic technique. Using 2.7 ml of 0.5% plain bupivacaine at L3/4, it was not possible to predict whether anaesthesia would result at all or if it would spread as far as T2. A number of factors may contribute to this unpredictability. First, plain bupivacaine is relatively hypobaric (specific gravity 1.004 at 20 °C and 0.998 at 37 °C) compared with cerebrospinal fluid (specific gravity 1.0063-1.0075 at 25 °C). These may be thought to be minor differences, but plain bupivacaine does spread to a higher level in the erect patient (Tuominen, Kalso and Rosenberg, 1982), whereas the spread of a truely isobaric solution is not affected by gravity (Wildsmith et al., 1981). The height of the blockade produced by hypobaric amethocaine has greater variability than an equipotent isobaric solution, and is markedly more variable than hyperbaric amethocaine (Brown et al., 1980). Aqueous hypobaric solutions are less viscous than their isobaric or hyperbaric equivalents prepared in saline or dextrose (Brown et al., 1980). This may increase the diffusibility of the drug, permitting more rapid dilution by the cerebrospinal fluid (CSF). Furthermore, CSF currents such as those initiated during subarachnoid injection or by positioning the patient may be an important factor in potentiating this diffusion and may be responsible for further dilution of the drug and the occasional unexpected high block. Since plain bupivacaine is a relatively insoluble base and precipitates in CSF in concentrations greater than 1% (Chambers, 1982), a relatively large volume of injection is required to achieve an anaesthetic dosage. This larger volume may

BRITISH JOURNAL OF ANAESTHESIA