British Journal of Anaesthesia 1991; 66: 645-650
ADRENALINE, FENTANYL OR ADRENALINE AND FENTANYL AS ADJUNCTS TO BUPIVACAINE FOR EXTRADURAL ANAESTHESIA IN ELECTIVE CAESAREAN SECTION D. W. NOBLE, L. M. MORRISON, M. S. BROCKWAY AND J. H. McCLURE
We have compared the effects of extradural fentanyland fentanylplus adrenaline with adrenaline alone as adjuncts to extradural bupivacaine in patients undergoing elective Caesarean section. Forty-five patients were allocated randomly to receive 0.45% bupivacaine 20 ml with adrenaline 4.5 fig mh1, fentanyl 4.5 fig ml'1 or adrenaline plus fentanyl (4.5 fig ml'1 of each) as supplements. The main outcome measures were time to bilateral analgesia of T6 or higher, need for intraoperative analgesic supplements, observer rating of intraoperative analgesia and patient assessment of analgesia using a 10-cm visual analogue scale. The time to onset of analgesia to T6 was reduced insignificantly by the fentanyl solutions compared with adrenaline only. The quality of analgesia as assessed by the need for analgesic supplements was superior for the patients given fentanyl. An observer rating of pain and visual analogue pain scoring by the patient also indicated superior analgesia with fentanyl supplementation. Two patients experienced respiratory depression after extradural fentanyl and were given naloxone. Two neonates were also given naloxone. Close supervision is therefore recommended in the early postoperative period when this technique is used. KEY WORDS Anaesthetic techniques: extradural. Anaesthetics, bupivacaine. Analgesics: fentanyl.
which may preclude its use for urgent procedures, and visceral pain, which is experienced by approximately 50% of patients [4]. Adrenaline has been shown to improve die efficacy of extradural block [5] and until recently has been the only adjunct available for this purpose. More recently, the role of extradural opioids has been investigated in enhancing the analgesic effect of local anaesthetics. Because of its rapid onset of action and lipid solubility, which theoretically reduces rostral spread in cerebrospinal fluid, fentanyl is considered to be a suitable opioid for extradural administration [6]. Peak plasma concentrations of fentanyl are smaller after extradural administration compared with the i.m. route [6] and bolus doses of 50-200 ug administered extradurally have been reported to cause little and only clinically insignificant respiratory depression [6-8]. The addition of fentanyl to local anaesthetic solutions administered extradurally has been shown to improve analgesia for parturients in labour and during Caesarean section [9-13]. In one open study, not only was patient comfort improved, but a significant reduction in onset time was observed when fentanyl 100 ug was added to 0.5% bupivacaine [14]. The object of this randomized, double-blind controlled clinical trial was to ascertain if a useful reduction in onset time may be obtained with fentanyl or fentanyl and adrenaline compared with adrenaline sup-
local: D. W. NOBLE, B.MED.BIOL., F.C.ANAES., Department of Anaes-
thetics, Royal Infirmary, Foresterhill, Aberdeen AB9 2ZB.
The advantages of extradural anaesthesia for Caesarean section are well documented [1-3]. However, disadvantages include slow onset time,
L. M. MORRISON, B.SC.(HONS), F.F.A.R.C.S.I. ; M. S. BROCKWAY, F.C.ANAES.; J. H. MCCLURE, B.SC.(HONS), F.C.ANAES.; Depart-
ment of Anaesthetics, Royal Infirmary, Lauriston Place, Edinburgh EH3 9YW. Accepted for Publication: January 5, 1991.
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
SUMMARY
BRITISH JOURNAL OF ANAESTHESIA
646
plementation only, and to compare the analgesic effectiveness of these three combinations.
PATIENTS AND METHODS
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
Following Ethics Committee approval, we studied 48 healthy mothers undergoing elective Caesarean section under extradural anaesthesia, after obtaining their informed consent. Patients who were less than 152 cm in height, or more than 100 kg in weight were excluded in order to reduce the risks of excessive spread from a standardized single dose of local anaesthetic. Those with pregnancy-induced hypertension were excluded also because of the potential adverse effects of adrenaline on maternal haemodynamics and uterine blood flow. Based on the results of an open study [14], taking a clinically significant reduction in onset time to be 10 min, and a statistically significant difference to be P < 0.05, it was calculated that 15 patients per group would achieve a power of greater than 0.80. Patients were premedicated orally on the evening before, and 2 h before the procedure, with ranitidine 150 mg and were given also sodium citrate 30 ml on arrival in the anaesthetic room. An i.v. infusion of Hartmann's solution 10 ml kg"1 was commenced. Patient monitoring included ECG, non-invasive arterial pressure measurement (Dinamap) and pulse oximetry. Patients were allocated randomly to receive 20 ml of 0.45 % bupivacaine with adrenaline 4.5ugml~1 (group I), 0.45% bupivacaine with fentanyl 4.5 ug ml"1 (group II) or 0.45% bupivacaine with adrenaline 4.5 \ig ml"1 plus fentanyl 4.5 ug ml"1 (group III), in a double-blind design, via a lateral-eyed extradural catheter inserted 4 cm into the extradural space between the second and third lumbar vertebrae. The solutions were constituted by adding fentanyl 100 ug or 0.1 ml of adrenaline 1 mg ml"1, or both, to 0.5% bupivacaine 20 ml; saline was used as required to give a total volume of 22 ml of solution. After insertion of the catheter, and with the patient in the left lateral position, 5 ml of the solution was injected as a test dose 5 min before injection of the main dose. The main dose was injected in three 5-ml aliquots, each aliquot being administered over 15 s followed by a pause of 45 s between aliquots. Onset of block was timed from the end of the last injection.
Before delivery of the infant, oxygen was administered via a 60 % Ventimask; thereafter, it was discontinued. Hypotension, defined as a systolic arterial pressure of less than 90 mm Hg or a reduction of more than 30 % from baseline, was treated with ephedrine and fluids. Nausea and vomiting were treated with ephedrine (if hypotension-related) or metoclopramide. Sensory block was assessed by a blinded observer using loss of sensation of pinprick by the blunt end of a 27-gauge dental cartridge needle. Motor power was assessed using the Bromage scale. These assessments were made at 2, 5, 10, 15, 20, 25 and 30 min. The times for sensory block to develop bilaterally to T10 and T6 were noted, with the maximum height of block and time to maximum block. If a bilateral sensory block of T6 or higher was not present at 30 min, further 5-ml aliquots of bupivacaine were administered until an adequate height of block was achieved. Increments of local anaesthetic, or i.v. diamorphine, were administered to treat moderate pain or persistent discomfort as indicated clinically. In general, anatomically localized pain was treated with additional local anaesthetic and poorly localized pain or failure of additional local anaesthetic was an indication for i.v. diamorphine after delivery. The times from skin incision to uterine incision and uterine incision to delivery were noted. The Apgar scores at 1 and 5 min were recorded also. At the end of the procedure, the blinded observer rated the quality of anaesthesia according to the following scale: excellent; good = minor or transient discomfort; fair = moderate or persistent pain or discomfort requiring supplementation with an i.v. opioid; poor = severe pain requiring induction of general anaesthesia. At this time the patient was asked also to indicate on a 10-cm visual analogue scale, with "no pain" and "worst pain imaginable" at either end, how much pain they experienced during the procedure. The patient was questioned about other adverse effects (nausea, vomiting, lightheadedness, itching, shivering or any other symptoms she wished to report) shortly after returning to the recovery room and the following day. Differences between groups were analysed using analysis of variance or die Kruskal-Wallis test followed by Student's t test or MannWhitney test as appropriate for interval or ordinal data. Differences in proportions were analysed with chi-square or Fisher's exact test.
EXTRADURAL BUPIVACAINE WITH FENTANYL AND/OR ADRENALINE
647
TABLE I. Patient characteristics (mean (range or SD)). *P < 0.05 group I vs group II
Group I (n = 15)
Group II in = 15)
Group III (n = 15)
Supplement
Adrenaline
Fentanyl
Age (yr) Weight (kg) Height (cm) Pre-op. SAP (mm Hg)
28.5 (18-36) 74.9(12.4) 165(6.1) 123 (10.4)
28.9(18-35) 73.1(11.3) 161 (4.3)* 121 (10.6)
Fentanyl + adrenaline 29.1 (22-36) 72.4(7.5) 162 (4.1) 124(13.4)
TABLE II. Onset of analgesia (mean (SD) or median (quartiles)) and motor block (Bromage scale). •f Patients requiring supplementary local analgesia to reach 7*6 or higher. *P = 0.02 group I vs group III
Additional analgesiaf Time (min) to reach T10 T6
No. of analgesic dermatomes above LI Time to maximum spread (min) Motor block
3
1
Group III 1
10.6(6.0) 17.7(7.5) 10.0(8.0-11.0)
7.8(3.9) 16.3 (6.7) 10.0(7.0-13.0)
6.2 (3.4)* 14.3(7.0) 10.0(9.0-11.0)
23.3 (4.5)
22.7 (4.6)
24.0 (4.7)
1 5 8 1
0 1 2 3
Group II
RESULTS
Three patients were excluded after randomization : one procedure was interrupted to allow an emergency Caesarean section to proceed; one procedure was abandoned because of unilateral analgesia with no analgesia above LI on one side; and another procedure was abandoned after a bloody tap was followed by evidence of i.v. injection on reinsertion at another level. All three patients had been allocated to group II. Patient characteristics were similar, with the exception that patients in group I were taller than those in the other groups (table I). Three patients in group I required additional local anaesthetic to produce an adequate spread of analgesia, compared with one patient each in the other two groups (table II). Sensory block to T10 was most rapid in group III and the difference was statistically significant from group I. A similar pattern was observed when assessing time to sensory block to T6, but was less marked and was not statistically significant. It is noteworthy that, if the time for the test dose is included seven, six and four patients from groups I, II and III took
3 7 4 1
0 4 11 0
25 min or longer to achieve an analgesic block to T6 or higher. The maximum number of analgesic dermatomes above LI and the times to reach maximum spread were similar in all three groups and maximum motor block was not significantly different between groups (table II). Three patients in group I and three in group II required further increments of local anaesthetic during the procedure, compared with only one in group III. In addition, nine patients in group I required i.v. diamorphine, compared with two in group II and none in group III (table III). I.v. diamorphine was given after delivery for poorly localized pain or failure of additional local anaesthetic to control pain. The quality of analgesia as rated by the observer was superior in the fentanyl groups compared with the adrenaline only group, with 13 patients in group III being rated as having "excellent" analgesia and the remaining two patients as having "good" analgesia. In contrast, eight patients in group I were rated as having only "fair" analgesia (table III). The visual analogue scale completed by the parturients reflected this pattern, the lower scores being observed in the fentanyl groups.
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
Group I
BRITISH JOURNAL OF ANAESTHESIA
648 TABLE III. Number of patients requiring supplementary analgesia and quality of analgesia for Caesarean section. Significant differences compared with group I: *P < 0.05; **P = 0.01; ***P = 0.005
Group I
TABLE IV. Number
Group III
3 9
3 2*
1 0**
5 2 8 0 6
11* 2 2 0 2
13*** 2 0 0 0
of patients with complications during operation
One or more episodes of hypotension Ephedrine given Nausea Vomiting Metoclopramide given Respiratory depression Naloxone given
Group I
Group II
Group III
8 9 5 2 7 0 0
9 10 5 2 2 2 2
13 13 5 0 0 0 0
These differences did not reach statistical significance (group I vs. group III, P = 0.051; group I vs group II, P = 0.13). Hypotension was most frequent in group III compared with the other groups and the incidence of nausea or vomiting was similar between groups (table IV). Two patients in group II were given naloxone for respiratory depression. The first patient has been reported in detail elsewhere [15]. The second patient, who was drowsy from about 10 min after extradural injection, demonstrated decreases in SpO2 from 96% to less than 90% after delivery. Rousing the patient resulted in larger breaths with quick return of 5pOz to previous values. Naloxone 0.4 mg i.v. antagonized the patient's somnolence and respiratory depression effectively and there was no recurrence in the recovery room. Maternal interview in the recovery period and the following day failed to demonstrate significant differences in the incidence of nausea, vomiting, lightheadedness, itching or shivering. Shivering, however, was most frequent in those receiving adrenaline supplementation (47%) and least fre-
DISCUSSION
The results of this study indicate that supplementation of bupivacaine with fentanyl or fentanyl and adrenaline was not associated with clinically useful reduction in onset time of analgesia to T6. These results conflict with those of an earlier open study which found a mean reduction in onset time of more than 10 min [14]. In this study, the mean reduction in analgesic onset was of the order of 3-4 min only. Other studies also have failed to show a reduction in onset time [11-13]. However, these have utilized a slow, multiple increment technique with lengthy periods between increments which might obscure any differences. Spinal anaesthesia to this level or higher can be achieved in about 5 min [17]. In comparison, if the time for the test dose is added to the times reported here, only two of 45 patients studied here could be considered ready for Caesarean section at 10 min and in 17 patients at least 25 min elapsed before bilateral analgesia to T6. The number of analgesic dermatomes was similar in all three groups, despite a tendency for the adrenaline-supplemented group to be taller compared with the other groups. However, statistically significant differences in patient characteristics may occur in correctly conducted randomized trials [18]. The combination of fentanyl and adrenaline appeared to provide the best analgesia with the least requirements for analgesic supplements, the best observer rating and the best visual analogue scores. The fentanyl only group ranked second for these assessments. Although the fentanyl plus adrenaline group ranked first for all analgesia assessments, there were no significant differences between it and the fentanyl only group. The finding of improved analgesia with fentanyl
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
Supplementation Local anaesthetic I.v. opioid Quality of analgesia Excellent Good Fair Poor VAS score $ 2 cm
Group II
quent in those receiving fentanyl (7 %)—a finding which is in accordance with others [16]. The Apgar scores at 1 and 5 min were generally satisfactory, with no significant differences between groups. Two neonates, both of whom had Apgar scores of nine at 1 and 5 min received naloxone. One (group II) was "very sleepy" 3 min after delivery (Apgar score of 6) and was given naloxone 20 ug i.m. The other (group III) was observed to have a "persistent blue tinge" at 10 min and received naloxone 60 ug i.m. However, this neonate had aspirated a large volume of liquor and there was no discernible improvement following naloxone.
EXTRADURAL BUPIVACAINE WITH FENTANYL AND/OR ADRENALINE
ation-PaCOj relationship; these effects are greater and of longer duration than those produced by i.m. fentanyl [8], but are probably not of clinical importance [7]. Respiratory arrhythmias have been documented also in a study in which fentanyl 100 ug was used as a supplement to bupivacaine for percutaneous nephrolithotripsy [24]. Interestingly, in that study, the addition of adrenaline to this mixture reduced the incidence of respiratory arrhythmias. In this study, the relatively rapid injection of solution may have produced a "longitudinal sleeve" of dilute fentanyl within the extradural space, which may have enhanced rostral spread of fentanyl. This effect may not be seen when a similar dose of fentanyl is administered slowly or in a smaller volume. Rostral spread may be the main mechanism of depression, as blood concentrations of fentanyl are small after extradural administration [22]. It has been demonstrated also that dilute solutions of fentanyl have a more rapid onset and more prolonged duration of action [25]. However, as a result of this study, we now believe that close observation is warranted for at least 3 h after administration of similar doses of extradural fentanyl. Further work to examine the efficacy and safety of extradural fentanyl given in smaller doses, or similar dose in a smaller volume is now required. Two neonates were given naloxone, although both had Apgar scores of 9 at 1 and 5 min. The neonatologists, knowing that the mothers had received an opioid, may have applied a lower threshold for administration of an antagonist, and it is uncertain if these two neonates suffered as a result of administration of opioid. Neonatal neurobehavioural effects were not observed with extradural fentanyl 1 ug kg"1 [26, 27] although they have been observed with high doses of extradural sufentanil (80 ug) [28]. However, high umbilical arterial concentrations are encountered occasionally after extradural fentanyl 100 ug and some neonates may be at risk of opioid effects [13]. In conclusion, this study has failed to demonstrate a clinically significant reduction in time to an acceptable level of analgesia for Caesarean section. However, fentanyl supplementation was associated with greater maternal comfort and a reduced requirement for analgesics during operation. Two cases of maternal respiratory depression highlight the need for close observation
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
supplementation accords with those of other workers [11-14]. Single doses of 50-200 ug and cumulative doses in excess of this have been reported to cause little or only clinically insignificant respiratory depression [6-8, 19]. Some consider parturients to be at very low risk of respiratory depression because of pregnancyassociated hyperventilation [20], and the dose of fentanyl chosen for this study (100 ug) was anticipated to be both safe and efficacious. The analgesia achieved with extradural fentanyl is superior to that after i.v. or i.m. administration and, perhaps because of sequestration of this lipophilic drug in adipose and neural tissue, the blood or plasma concentrations of fentanyl are smaller after extradural administration [6, 10, 21, 22]. Comparison of supplementary extradural with i.m. fentanyl given before local anaesthetic block for Caesarean section may be more valid scientifically. However, the known pharmacokinetic profile [6, 21], the time of peak plasma concentrations and subsequent placental transfer after i.m. fentanyl may render such a comparison unethical because of the risk of neonatal depression. Unsupplemented extradural bupivacaine for Caesarean section is associated with an incidence of visceral pain approaching 50 % [4] and so there is scope for improvement in maternal comfort. The addition of adrenaline to bupivacaine has been shown to improve its efficacy [5]. However, this study has demonstrated fentanyl to be more efficacious than adrenaline as a supplement to bupivacaine. There was a high incidence of maternal hypotension in this study which may reflect the modest i.v. infusion given to these patients, and the relatively rapid extension of block. This was treated promptly and effectively with ephedrine and additional fluids. However, the most worrying adverse effect seen in this small study was respiratory depression, which occurred in two patients. Although other previously published studies have not documented clinically severe respiratory depression, these studies have been small and the absence of side effects gives only limited reassurance [23]. For example, in one of the larger trials [13], 30 women were allocated randomly to receive extradural fentanyl and bupivacaine. The 95 % confidence intervals for any side effect in a trial of this size in which the side effect did not occur is 0-10%. Extradural fentanyl has been shown to produce increases in Pa co and flattening of the ventil-
649
BRITISH JOURNAL OF ANAESTHESIA
650
into the recovery period to avert possible adverse outcome. Neonatologists should be forewarned when extradural opioids are given to the mother before delivery.
ACKNOWLEDGEMENT We thank Drs A. Buchan, G. Sharwood-Smith and A. Whitfield for their assistance with this project.
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
REFERENCES 1. Moir DD, Thorburn J. Obstetric Anaesthesia and Analgesia, 3rd Edn. London: Bailliere-Tindall, 1986. 2. Moir DD. Extradural analgesia for Caesarean section. British Journal of Anaesthesia 1979; 51: 79-80. 3. Carrie LES. Extradural, spinal or combined block for obstetric surgical anaesthesia. British Journal of Anaesthesia 1990; 65: 225-233. 4. Alahuta S, Kangas-Saarela T, Hollmen AI, Edstrom HH. Visceral pain during Caesarean section under spinal and epidural anaesthesia with bupivacaine. Ada Anaesthesiologica Scandinavica 1990; 34: 95-98. 5. Laishley RS, Morgan BM. A single dose epidural technique for Caesarean section. A comparison between 0.5% bupivacaine plain and 0.5% bupivacaine with adrenaline. Anaesthesia 1988; 43: 100-103. 6. Lomessy A, Magnin C, Viale J, Motin J, Cohen R. Clinical advantages of fentanyl given epidurally for postoperative analgesia. Aneslhesiology 1984; 61: 466-469. 7. Morisot P, Dessanges JF, Regnard J, Lockhart A. Ventilatory response to carbon dioxide during extradural anaesthesia with lignocaine and fentanyl. British Journal of Anaesthesia 1989; 63: 97-102. 8. Negre I, Gueneron J, Ecoffey C, Penon C, Gross JB, Levron J, Samii K. Ventilatory response to carbon dioxide after intramuscular and epidural fentanyl. Anesthesia and Analgesia 1987; 66: 707-710. 9. Jones G, Paul DL, Elton RA, McClure JH. A comparison of bupivacaine and bupivacaine with fentanyl in continuous extradural analgesia in labour. British Journal of Anaesthesia 1989; 63: 254-259. 10. Vella LM, Willats DG, Knott C, Linton DJ, Justins DM, Reynolds F. Epidural fentanyl in labour. An evaluation of the systemic contribution to analgesia. Anaesthesia 1985; 40: 741-747. 11. Gaffud MP, Bansal P, Lawton C, Velasquez N, Watson WA. Surgical analgesia for Cesarean delivery with epidural bupivacaine and fentanyl. Anesthesiology 1986; 65: 331-334. 12. King MJ, Bowden MI, Cooper GM. Epidural fentanyl and 0.5% bupivacaine for elective Caesarean section. Anaesthesia 1990; 45: 285-288.
13. Paech MJ, Westmore MD, Speirs HM. A double-blind comparison of epidural bupivacaine and bupivacaine— fentanyl for Caesarean section. Anaesthesia and Intensive Care 1990; 18: 22-30. 14. Milon D, Bentue-Ferrer D, Noury D, Reymann JM, Sauvage J, Allain H, Saint-Marc C, van den Driessche J. Epidural bupivacaine-fentanyl anaesthesia for Caesarean section. Annales Francoises a"Anesthe'sie et de Re'animation 1983;2: 273-279. 15. Brockway MS, Noble DW, Sharwood-Smith GH, McClure JH. Profound respiratory depression after extradural fentanyl. British Journal of Anaesthesia 1990; 64: 243-245. 16. Shehabi Y, Gatt S, Buckman T, Isert P. Effect of adrenaline, fentanyl and warming of injectate on shivering following extradural analgesia in labour. Anaesthesia and Intensive Care 1990; 18: 31-37. 17. Hunt CO, Naulty JS, Bader AM, Hauch MA, Vartikar JV, Datta S, Hertwig LM, Ostheimer GW. Perioperative analgesia with subarachnoid fentanyl—bupivacaine for Cesarean delivery. Anesthesiology 1989; 71: 535-540. 18. Altman DG, Dore CJ. Randomisation and baseline comparisons in clinical trials. Lancet 1990; 335: 149-153. 19. Gant NF, Wallace DH. Anaesthetic management in pregnancy induced hypertension. Clinics in Anaesthesiology 1986; 4: 321-350. 20. Writer WDR. Epidural morphine for post-Caesarean analgesia. Canadian Journal of Anaesthesia 1990; 37: 608-612. 21. Justins DM, Knott C, Luthman J, Reynolds F. Epidural versus intramuscular fentanyl. Analgesia and pharmacokinetics in labour. Anaesthesia 1983; 38: 937-942. 22. Gourlay GK, Murphy TM, Plummer JL, Kowalski SR, Cherry DA, Cousins MJ. Pharmacokinetics of fentanyl in lumbar and cervical CSF following lumbar epidural and intravenous administration. Pain 1989; 38: 253-259. 23. Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. Journal of the American Medical Association 1983; 249: 1743-1745. 24. Bromage P, El Faqih S, Husain I, Naquib M. Epinephrine and fentanyl as adjuvants to 0.5 % bupivacaine for epidural analgesia. Regional Anesthesia 1989; 14: 189-194. 25. Birnbach DJ, Johnson M, Arcario T, Datta S, Naulty JS, Ostheimer GW. Effect of diluent volume on analgesia produced by epidural fentanyl. Anesthesia and Analgesia 1989; 68: 808-810. 26. Preston PG, Rosen MA, Hughes SC, Glosten B, Ross BK, Daniels D, Shnider S, Dailey P. Epidural anaesthesia with fentanyl for Cesarean section: Maternal effects and neonatal outcome. Anesthesiology 1988; 68: 938-943. 27. Capogna G, Tomassetti M, Gostantino P, DiFeo G, Nisini R. Epidural versus intravenous fentanyl for Cesarean delivery: Neonatal neurobehavioural effects. Regional Anesthesia 1988; 13 (Suppl): 17. 28. Capogna G, Celleno D, Tomassetti M. Maternal analgesia and neonatal effects of epidural sufentanil for Cesarean section. Regional Anesthesia 1989; 14: 282-287.
ADRENALINE, FENTANYL OR ADRENALINE AND FENTANYL AS ADJUNCTS TO BUPIVACAINE FOR EXTRADURAL ANAESTHESIA IN ELECTIVE CAESAREAN SECTION D. W. NOBLE, L. M. MORRISON, M. S. BROCKWAY AND J. H. McCLURE
We have compared the effects of extradural fentanyland fentanylplus adrenaline with adrenaline alone as adjuncts to extradural bupivacaine in patients undergoing elective Caesarean section. Forty-five patients were allocated randomly to receive 0.45% bupivacaine 20 ml with adrenaline 4.5 fig mh1, fentanyl 4.5 fig ml'1 or adrenaline plus fentanyl (4.5 fig ml'1 of each) as supplements. The main outcome measures were time to bilateral analgesia of T6 or higher, need for intraoperative analgesic supplements, observer rating of intraoperative analgesia and patient assessment of analgesia using a 10-cm visual analogue scale. The time to onset of analgesia to T6 was reduced insignificantly by the fentanyl solutions compared with adrenaline only. The quality of analgesia as assessed by the need for analgesic supplements was superior for the patients given fentanyl. An observer rating of pain and visual analogue pain scoring by the patient also indicated superior analgesia with fentanyl supplementation. Two patients experienced respiratory depression after extradural fentanyl and were given naloxone. Two neonates were also given naloxone. Close supervision is therefore recommended in the early postoperative period when this technique is used. KEY WORDS Anaesthetic techniques: extradural. Anaesthetics, bupivacaine. Analgesics: fentanyl.
which may preclude its use for urgent procedures, and visceral pain, which is experienced by approximately 50% of patients [4]. Adrenaline has been shown to improve die efficacy of extradural block [5] and until recently has been the only adjunct available for this purpose. More recently, the role of extradural opioids has been investigated in enhancing the analgesic effect of local anaesthetics. Because of its rapid onset of action and lipid solubility, which theoretically reduces rostral spread in cerebrospinal fluid, fentanyl is considered to be a suitable opioid for extradural administration [6]. Peak plasma concentrations of fentanyl are smaller after extradural administration compared with the i.m. route [6] and bolus doses of 50-200 ug administered extradurally have been reported to cause little and only clinically insignificant respiratory depression [6-8]. The addition of fentanyl to local anaesthetic solutions administered extradurally has been shown to improve analgesia for parturients in labour and during Caesarean section [9-13]. In one open study, not only was patient comfort improved, but a significant reduction in onset time was observed when fentanyl 100 ug was added to 0.5% bupivacaine [14]. The object of this randomized, double-blind controlled clinical trial was to ascertain if a useful reduction in onset time may be obtained with fentanyl or fentanyl and adrenaline compared with adrenaline sup-
local: D. W. NOBLE, B.MED.BIOL., F.C.ANAES., Department of Anaes-
thetics, Royal Infirmary, Foresterhill, Aberdeen AB9 2ZB.
The advantages of extradural anaesthesia for Caesarean section are well documented [1-3]. However, disadvantages include slow onset time,
L. M. MORRISON, B.SC.(HONS), F.F.A.R.C.S.I. ; M. S. BROCKWAY, F.C.ANAES.; J. H. MCCLURE, B.SC.(HONS), F.C.ANAES.; Depart-
ment of Anaesthetics, Royal Infirmary, Lauriston Place, Edinburgh EH3 9YW. Accepted for Publication: January 5, 1991.
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
SUMMARY
BRITISH JOURNAL OF ANAESTHESIA
646
plementation only, and to compare the analgesic effectiveness of these three combinations.
PATIENTS AND METHODS
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
Following Ethics Committee approval, we studied 48 healthy mothers undergoing elective Caesarean section under extradural anaesthesia, after obtaining their informed consent. Patients who were less than 152 cm in height, or more than 100 kg in weight were excluded in order to reduce the risks of excessive spread from a standardized single dose of local anaesthetic. Those with pregnancy-induced hypertension were excluded also because of the potential adverse effects of adrenaline on maternal haemodynamics and uterine blood flow. Based on the results of an open study [14], taking a clinically significant reduction in onset time to be 10 min, and a statistically significant difference to be P < 0.05, it was calculated that 15 patients per group would achieve a power of greater than 0.80. Patients were premedicated orally on the evening before, and 2 h before the procedure, with ranitidine 150 mg and were given also sodium citrate 30 ml on arrival in the anaesthetic room. An i.v. infusion of Hartmann's solution 10 ml kg"1 was commenced. Patient monitoring included ECG, non-invasive arterial pressure measurement (Dinamap) and pulse oximetry. Patients were allocated randomly to receive 20 ml of 0.45 % bupivacaine with adrenaline 4.5ugml~1 (group I), 0.45% bupivacaine with fentanyl 4.5 ug ml"1 (group II) or 0.45% bupivacaine with adrenaline 4.5 \ig ml"1 plus fentanyl 4.5 ug ml"1 (group III), in a double-blind design, via a lateral-eyed extradural catheter inserted 4 cm into the extradural space between the second and third lumbar vertebrae. The solutions were constituted by adding fentanyl 100 ug or 0.1 ml of adrenaline 1 mg ml"1, or both, to 0.5% bupivacaine 20 ml; saline was used as required to give a total volume of 22 ml of solution. After insertion of the catheter, and with the patient in the left lateral position, 5 ml of the solution was injected as a test dose 5 min before injection of the main dose. The main dose was injected in three 5-ml aliquots, each aliquot being administered over 15 s followed by a pause of 45 s between aliquots. Onset of block was timed from the end of the last injection.
Before delivery of the infant, oxygen was administered via a 60 % Ventimask; thereafter, it was discontinued. Hypotension, defined as a systolic arterial pressure of less than 90 mm Hg or a reduction of more than 30 % from baseline, was treated with ephedrine and fluids. Nausea and vomiting were treated with ephedrine (if hypotension-related) or metoclopramide. Sensory block was assessed by a blinded observer using loss of sensation of pinprick by the blunt end of a 27-gauge dental cartridge needle. Motor power was assessed using the Bromage scale. These assessments were made at 2, 5, 10, 15, 20, 25 and 30 min. The times for sensory block to develop bilaterally to T10 and T6 were noted, with the maximum height of block and time to maximum block. If a bilateral sensory block of T6 or higher was not present at 30 min, further 5-ml aliquots of bupivacaine were administered until an adequate height of block was achieved. Increments of local anaesthetic, or i.v. diamorphine, were administered to treat moderate pain or persistent discomfort as indicated clinically. In general, anatomically localized pain was treated with additional local anaesthetic and poorly localized pain or failure of additional local anaesthetic was an indication for i.v. diamorphine after delivery. The times from skin incision to uterine incision and uterine incision to delivery were noted. The Apgar scores at 1 and 5 min were recorded also. At the end of the procedure, the blinded observer rated the quality of anaesthesia according to the following scale: excellent; good = minor or transient discomfort; fair = moderate or persistent pain or discomfort requiring supplementation with an i.v. opioid; poor = severe pain requiring induction of general anaesthesia. At this time the patient was asked also to indicate on a 10-cm visual analogue scale, with "no pain" and "worst pain imaginable" at either end, how much pain they experienced during the procedure. The patient was questioned about other adverse effects (nausea, vomiting, lightheadedness, itching, shivering or any other symptoms she wished to report) shortly after returning to the recovery room and the following day. Differences between groups were analysed using analysis of variance or die Kruskal-Wallis test followed by Student's t test or MannWhitney test as appropriate for interval or ordinal data. Differences in proportions were analysed with chi-square or Fisher's exact test.
EXTRADURAL BUPIVACAINE WITH FENTANYL AND/OR ADRENALINE
647
TABLE I. Patient characteristics (mean (range or SD)). *P < 0.05 group I vs group II
Group I (n = 15)
Group II in = 15)
Group III (n = 15)
Supplement
Adrenaline
Fentanyl
Age (yr) Weight (kg) Height (cm) Pre-op. SAP (mm Hg)
28.5 (18-36) 74.9(12.4) 165(6.1) 123 (10.4)
28.9(18-35) 73.1(11.3) 161 (4.3)* 121 (10.6)
Fentanyl + adrenaline 29.1 (22-36) 72.4(7.5) 162 (4.1) 124(13.4)
TABLE II. Onset of analgesia (mean (SD) or median (quartiles)) and motor block (Bromage scale). •f Patients requiring supplementary local analgesia to reach 7*6 or higher. *P = 0.02 group I vs group III
Additional analgesiaf Time (min) to reach T10 T6
No. of analgesic dermatomes above LI Time to maximum spread (min) Motor block
3
1
Group III 1
10.6(6.0) 17.7(7.5) 10.0(8.0-11.0)
7.8(3.9) 16.3 (6.7) 10.0(7.0-13.0)
6.2 (3.4)* 14.3(7.0) 10.0(9.0-11.0)
23.3 (4.5)
22.7 (4.6)
24.0 (4.7)
1 5 8 1
0 1 2 3
Group II
RESULTS
Three patients were excluded after randomization : one procedure was interrupted to allow an emergency Caesarean section to proceed; one procedure was abandoned because of unilateral analgesia with no analgesia above LI on one side; and another procedure was abandoned after a bloody tap was followed by evidence of i.v. injection on reinsertion at another level. All three patients had been allocated to group II. Patient characteristics were similar, with the exception that patients in group I were taller than those in the other groups (table I). Three patients in group I required additional local anaesthetic to produce an adequate spread of analgesia, compared with one patient each in the other two groups (table II). Sensory block to T10 was most rapid in group III and the difference was statistically significant from group I. A similar pattern was observed when assessing time to sensory block to T6, but was less marked and was not statistically significant. It is noteworthy that, if the time for the test dose is included seven, six and four patients from groups I, II and III took
3 7 4 1
0 4 11 0
25 min or longer to achieve an analgesic block to T6 or higher. The maximum number of analgesic dermatomes above LI and the times to reach maximum spread were similar in all three groups and maximum motor block was not significantly different between groups (table II). Three patients in group I and three in group II required further increments of local anaesthetic during the procedure, compared with only one in group III. In addition, nine patients in group I required i.v. diamorphine, compared with two in group II and none in group III (table III). I.v. diamorphine was given after delivery for poorly localized pain or failure of additional local anaesthetic to control pain. The quality of analgesia as rated by the observer was superior in the fentanyl groups compared with the adrenaline only group, with 13 patients in group III being rated as having "excellent" analgesia and the remaining two patients as having "good" analgesia. In contrast, eight patients in group I were rated as having only "fair" analgesia (table III). The visual analogue scale completed by the parturients reflected this pattern, the lower scores being observed in the fentanyl groups.
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
Group I
BRITISH JOURNAL OF ANAESTHESIA
648 TABLE III. Number of patients requiring supplementary analgesia and quality of analgesia for Caesarean section. Significant differences compared with group I: *P < 0.05; **P = 0.01; ***P = 0.005
Group I
TABLE IV. Number
Group III
3 9
3 2*
1 0**
5 2 8 0 6
11* 2 2 0 2
13*** 2 0 0 0
of patients with complications during operation
One or more episodes of hypotension Ephedrine given Nausea Vomiting Metoclopramide given Respiratory depression Naloxone given
Group I
Group II
Group III
8 9 5 2 7 0 0
9 10 5 2 2 2 2
13 13 5 0 0 0 0
These differences did not reach statistical significance (group I vs. group III, P = 0.051; group I vs group II, P = 0.13). Hypotension was most frequent in group III compared with the other groups and the incidence of nausea or vomiting was similar between groups (table IV). Two patients in group II were given naloxone for respiratory depression. The first patient has been reported in detail elsewhere [15]. The second patient, who was drowsy from about 10 min after extradural injection, demonstrated decreases in SpO2 from 96% to less than 90% after delivery. Rousing the patient resulted in larger breaths with quick return of 5pOz to previous values. Naloxone 0.4 mg i.v. antagonized the patient's somnolence and respiratory depression effectively and there was no recurrence in the recovery room. Maternal interview in the recovery period and the following day failed to demonstrate significant differences in the incidence of nausea, vomiting, lightheadedness, itching or shivering. Shivering, however, was most frequent in those receiving adrenaline supplementation (47%) and least fre-
DISCUSSION
The results of this study indicate that supplementation of bupivacaine with fentanyl or fentanyl and adrenaline was not associated with clinically useful reduction in onset time of analgesia to T6. These results conflict with those of an earlier open study which found a mean reduction in onset time of more than 10 min [14]. In this study, the mean reduction in analgesic onset was of the order of 3-4 min only. Other studies also have failed to show a reduction in onset time [11-13]. However, these have utilized a slow, multiple increment technique with lengthy periods between increments which might obscure any differences. Spinal anaesthesia to this level or higher can be achieved in about 5 min [17]. In comparison, if the time for the test dose is added to the times reported here, only two of 45 patients studied here could be considered ready for Caesarean section at 10 min and in 17 patients at least 25 min elapsed before bilateral analgesia to T6. The number of analgesic dermatomes was similar in all three groups, despite a tendency for the adrenaline-supplemented group to be taller compared with the other groups. However, statistically significant differences in patient characteristics may occur in correctly conducted randomized trials [18]. The combination of fentanyl and adrenaline appeared to provide the best analgesia with the least requirements for analgesic supplements, the best observer rating and the best visual analogue scores. The fentanyl only group ranked second for these assessments. Although the fentanyl plus adrenaline group ranked first for all analgesia assessments, there were no significant differences between it and the fentanyl only group. The finding of improved analgesia with fentanyl
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
Supplementation Local anaesthetic I.v. opioid Quality of analgesia Excellent Good Fair Poor VAS score $ 2 cm
Group II
quent in those receiving fentanyl (7 %)—a finding which is in accordance with others [16]. The Apgar scores at 1 and 5 min were generally satisfactory, with no significant differences between groups. Two neonates, both of whom had Apgar scores of nine at 1 and 5 min received naloxone. One (group II) was "very sleepy" 3 min after delivery (Apgar score of 6) and was given naloxone 20 ug i.m. The other (group III) was observed to have a "persistent blue tinge" at 10 min and received naloxone 60 ug i.m. However, this neonate had aspirated a large volume of liquor and there was no discernible improvement following naloxone.
EXTRADURAL BUPIVACAINE WITH FENTANYL AND/OR ADRENALINE
ation-PaCOj relationship; these effects are greater and of longer duration than those produced by i.m. fentanyl [8], but are probably not of clinical importance [7]. Respiratory arrhythmias have been documented also in a study in which fentanyl 100 ug was used as a supplement to bupivacaine for percutaneous nephrolithotripsy [24]. Interestingly, in that study, the addition of adrenaline to this mixture reduced the incidence of respiratory arrhythmias. In this study, the relatively rapid injection of solution may have produced a "longitudinal sleeve" of dilute fentanyl within the extradural space, which may have enhanced rostral spread of fentanyl. This effect may not be seen when a similar dose of fentanyl is administered slowly or in a smaller volume. Rostral spread may be the main mechanism of depression, as blood concentrations of fentanyl are small after extradural administration [22]. It has been demonstrated also that dilute solutions of fentanyl have a more rapid onset and more prolonged duration of action [25]. However, as a result of this study, we now believe that close observation is warranted for at least 3 h after administration of similar doses of extradural fentanyl. Further work to examine the efficacy and safety of extradural fentanyl given in smaller doses, or similar dose in a smaller volume is now required. Two neonates were given naloxone, although both had Apgar scores of 9 at 1 and 5 min. The neonatologists, knowing that the mothers had received an opioid, may have applied a lower threshold for administration of an antagonist, and it is uncertain if these two neonates suffered as a result of administration of opioid. Neonatal neurobehavioural effects were not observed with extradural fentanyl 1 ug kg"1 [26, 27] although they have been observed with high doses of extradural sufentanil (80 ug) [28]. However, high umbilical arterial concentrations are encountered occasionally after extradural fentanyl 100 ug and some neonates may be at risk of opioid effects [13]. In conclusion, this study has failed to demonstrate a clinically significant reduction in time to an acceptable level of analgesia for Caesarean section. However, fentanyl supplementation was associated with greater maternal comfort and a reduced requirement for analgesics during operation. Two cases of maternal respiratory depression highlight the need for close observation
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
supplementation accords with those of other workers [11-14]. Single doses of 50-200 ug and cumulative doses in excess of this have been reported to cause little or only clinically insignificant respiratory depression [6-8, 19]. Some consider parturients to be at very low risk of respiratory depression because of pregnancyassociated hyperventilation [20], and the dose of fentanyl chosen for this study (100 ug) was anticipated to be both safe and efficacious. The analgesia achieved with extradural fentanyl is superior to that after i.v. or i.m. administration and, perhaps because of sequestration of this lipophilic drug in adipose and neural tissue, the blood or plasma concentrations of fentanyl are smaller after extradural administration [6, 10, 21, 22]. Comparison of supplementary extradural with i.m. fentanyl given before local anaesthetic block for Caesarean section may be more valid scientifically. However, the known pharmacokinetic profile [6, 21], the time of peak plasma concentrations and subsequent placental transfer after i.m. fentanyl may render such a comparison unethical because of the risk of neonatal depression. Unsupplemented extradural bupivacaine for Caesarean section is associated with an incidence of visceral pain approaching 50 % [4] and so there is scope for improvement in maternal comfort. The addition of adrenaline to bupivacaine has been shown to improve its efficacy [5]. However, this study has demonstrated fentanyl to be more efficacious than adrenaline as a supplement to bupivacaine. There was a high incidence of maternal hypotension in this study which may reflect the modest i.v. infusion given to these patients, and the relatively rapid extension of block. This was treated promptly and effectively with ephedrine and additional fluids. However, the most worrying adverse effect seen in this small study was respiratory depression, which occurred in two patients. Although other previously published studies have not documented clinically severe respiratory depression, these studies have been small and the absence of side effects gives only limited reassurance [23]. For example, in one of the larger trials [13], 30 women were allocated randomly to receive extradural fentanyl and bupivacaine. The 95 % confidence intervals for any side effect in a trial of this size in which the side effect did not occur is 0-10%. Extradural fentanyl has been shown to produce increases in Pa co and flattening of the ventil-
649
BRITISH JOURNAL OF ANAESTHESIA
650
into the recovery period to avert possible adverse outcome. Neonatologists should be forewarned when extradural opioids are given to the mother before delivery.
ACKNOWLEDGEMENT We thank Drs A. Buchan, G. Sharwood-Smith and A. Whitfield for their assistance with this project.
Downloaded from http://bja.oxfordjournals.org/ at New York University on May 23, 2015
REFERENCES 1. Moir DD, Thorburn J. Obstetric Anaesthesia and Analgesia, 3rd Edn. London: Bailliere-Tindall, 1986. 2. Moir DD. Extradural analgesia for Caesarean section. British Journal of Anaesthesia 1979; 51: 79-80. 3. Carrie LES. Extradural, spinal or combined block for obstetric surgical anaesthesia. British Journal of Anaesthesia 1990; 65: 225-233. 4. Alahuta S, Kangas-Saarela T, Hollmen AI, Edstrom HH. Visceral pain during Caesarean section under spinal and epidural anaesthesia with bupivacaine. Ada Anaesthesiologica Scandinavica 1990; 34: 95-98. 5. Laishley RS, Morgan BM. A single dose epidural technique for Caesarean section. A comparison between 0.5% bupivacaine plain and 0.5% bupivacaine with adrenaline. Anaesthesia 1988; 43: 100-103. 6. Lomessy A, Magnin C, Viale J, Motin J, Cohen R. Clinical advantages of fentanyl given epidurally for postoperative analgesia. Aneslhesiology 1984; 61: 466-469. 7. Morisot P, Dessanges JF, Regnard J, Lockhart A. Ventilatory response to carbon dioxide during extradural anaesthesia with lignocaine and fentanyl. British Journal of Anaesthesia 1989; 63: 97-102. 8. Negre I, Gueneron J, Ecoffey C, Penon C, Gross JB, Levron J, Samii K. Ventilatory response to carbon dioxide after intramuscular and epidural fentanyl. Anesthesia and Analgesia 1987; 66: 707-710. 9. Jones G, Paul DL, Elton RA, McClure JH. A comparison of bupivacaine and bupivacaine with fentanyl in continuous extradural analgesia in labour. British Journal of Anaesthesia 1989; 63: 254-259. 10. Vella LM, Willats DG, Knott C, Linton DJ, Justins DM, Reynolds F. Epidural fentanyl in labour. An evaluation of the systemic contribution to analgesia. Anaesthesia 1985; 40: 741-747. 11. Gaffud MP, Bansal P, Lawton C, Velasquez N, Watson WA. Surgical analgesia for Cesarean delivery with epidural bupivacaine and fentanyl. Anesthesiology 1986; 65: 331-334. 12. King MJ, Bowden MI, Cooper GM. Epidural fentanyl and 0.5% bupivacaine for elective Caesarean section. Anaesthesia 1990; 45: 285-288.
13. Paech MJ, Westmore MD, Speirs HM. A double-blind comparison of epidural bupivacaine and bupivacaine— fentanyl for Caesarean section. Anaesthesia and Intensive Care 1990; 18: 22-30. 14. Milon D, Bentue-Ferrer D, Noury D, Reymann JM, Sauvage J, Allain H, Saint-Marc C, van den Driessche J. Epidural bupivacaine-fentanyl anaesthesia for Caesarean section. Annales Francoises a"Anesthe'sie et de Re'animation 1983;2: 273-279. 15. Brockway MS, Noble DW, Sharwood-Smith GH, McClure JH. Profound respiratory depression after extradural fentanyl. British Journal of Anaesthesia 1990; 64: 243-245. 16. Shehabi Y, Gatt S, Buckman T, Isert P. Effect of adrenaline, fentanyl and warming of injectate on shivering following extradural analgesia in labour. Anaesthesia and Intensive Care 1990; 18: 31-37. 17. Hunt CO, Naulty JS, Bader AM, Hauch MA, Vartikar JV, Datta S, Hertwig LM, Ostheimer GW. Perioperative analgesia with subarachnoid fentanyl—bupivacaine for Cesarean delivery. Anesthesiology 1989; 71: 535-540. 18. Altman DG, Dore CJ. Randomisation and baseline comparisons in clinical trials. Lancet 1990; 335: 149-153. 19. Gant NF, Wallace DH. Anaesthetic management in pregnancy induced hypertension. Clinics in Anaesthesiology 1986; 4: 321-350. 20. Writer WDR. Epidural morphine for post-Caesarean analgesia. Canadian Journal of Anaesthesia 1990; 37: 608-612. 21. Justins DM, Knott C, Luthman J, Reynolds F. Epidural versus intramuscular fentanyl. Analgesia and pharmacokinetics in labour. Anaesthesia 1983; 38: 937-942. 22. Gourlay GK, Murphy TM, Plummer JL, Kowalski SR, Cherry DA, Cousins MJ. Pharmacokinetics of fentanyl in lumbar and cervical CSF following lumbar epidural and intravenous administration. Pain 1989; 38: 253-259. 23. Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. Journal of the American Medical Association 1983; 249: 1743-1745. 24. Bromage P, El Faqih S, Husain I, Naquib M. Epinephrine and fentanyl as adjuvants to 0.5 % bupivacaine for epidural analgesia. Regional Anesthesia 1989; 14: 189-194. 25. Birnbach DJ, Johnson M, Arcario T, Datta S, Naulty JS, Ostheimer GW. Effect of diluent volume on analgesia produced by epidural fentanyl. Anesthesia and Analgesia 1989; 68: 808-810. 26. Preston PG, Rosen MA, Hughes SC, Glosten B, Ross BK, Daniels D, Shnider S, Dailey P. Epidural anaesthesia with fentanyl for Cesarean section: Maternal effects and neonatal outcome. Anesthesiology 1988; 68: 938-943. 27. Capogna G, Tomassetti M, Gostantino P, DiFeo G, Nisini R. Epidural versus intravenous fentanyl for Cesarean delivery: Neonatal neurobehavioural effects. Regional Anesthesia 1988; 13 (Suppl): 17. 28. Capogna G, Celleno D, Tomassetti M. Maternal analgesia and neonatal effects of epidural sufentanil for Cesarean section. Regional Anesthesia 1989; 14: 282-287.
