BJA Advance Access originally published online on January 8, 2007
British Journal of Anaesthesia 2007 98(2):263-269; doi:10.1093/bja/ael335
Autonomic effects of epidural and intravenous fentanyl
1 Department of Anesthesia and Perioperative Care, University of California, San Francisco, USA
2 Royal Free Hospital, Pond Street, London NW3 2QG, UK
3 Department of Anaesthesia, Royal Melbourne Hospital, Parkville, Victoria, Australia
4 Department of Outcomes Research, The Cleveland Clinic, Outcomes Research Institute, University of Louisville, KY, USA
* Corresponding author: Department of Anesthesia and Perioperative Care, University of California, Box 0648, San Francisco 94143-0648, CA, USA. E-mail: larsonm{at}anesthesia.ucsf.edu
Accepted for publication October 27, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
BACKGROUND: We tested the hypothesis that there is greater suppression of autonomic reflexes during general anaesthesia when fentanyl is administered epidurally than when it is given intravenously.
METHODS: Ten volunteers were anaesthetized with desflurane. Noxious stimuli of variable intensity were then delivered by tetanic electrical stimuli. Heart rate, arterial pressure, and pupillary dilation in response to these stimuli defined nociception. Seven of these volunteers participated twice using a crossover design: they received i.v. fentanyl on one study day and epidurally on the other. Autonomic responses to alternative tetanic stimuli at L4 and C5 dermatomes were measured every 5 min for 3 h after fentanyl administration.
RESULTS: After a brief redistribution period, plasma fentanyl concentrations were virtually identical on both days. After stimulation of the L4 dermatome only, block of pupillary reflex dilation was greater by 47 (22)% after epidural fentanyl compared with i.v. fentanyl. Time to maximal depression of reflex dilation after L4 stimulation was 41 (13) min. Arterial pressure and heart rate decreased after fentanyl by either route but there were no differences observed between L4 and C5 stimulations.
CONCLUSION: We conclude that during general anaesthesia, epidural fentanyl enhances antinociception by a spinal mechanism which can be detected by pupillary dilation but not by changes in arterial pressure or heart rate.
Keywords: anaesthetic techniques, epidural; analgesics, opioids, fentanyl; pain; parasympathetic nervous system; pupils, pupillary reflex dilation; sympathetic nervous system
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Opioids are often given during general anaesthesia to control autonomic reflexes. Well-known postoperative side-effects of opioids include respiratory depression, nausea, and sedation. Additionally, opioids cause acute tolerance and opioid-induced hyperalgesia.1 2 Because of concerns about side-effects, anaesthesiologists now try to restrict perioperative opioid administration.3 4 One way to reduce the total dose is to inject the opioids in the epidural space. Bolus injections of epidural fentanyl, for example, are known to enhance analgesia in unanaesthetized subjects compared with the same dose given intravenously.5–8
Spinal opioid receptors that suppress nociception from lumbar dermatomes lie in the spinal cord segments below the lower thoracic epidural space.9 10 We thus asked whether these opioid receptors could be activated by administration of epidural fentanyl during desflurane anaesthesia. If epidural fentanyl does suppress autonomic reflexes to a greater extent than i.v. fentanyl, this might be a method to reduce the total opioid dose during surgical procedures in which an epidural catheter is in place.
Accordingly, we studied anaesthetized volunteers to test the hypothesis that epidural administration of fentanyl deposited in the lower thoracic epidural space would suppress autonomic measures of nociception more than the same dose administered intravenously, that this suppression of autonomic responsiveness would be a segmental effect, and that it would mimic the effect caused by decreasing the intensity of the stimulus.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
We designed the following study to compare the effects of i.v. fentanyl to epidural fentanyl during general anaesthesia. We measured heart rate and arterial pressure after noxious stimulation and observed suppression of these responses after administration of epidural or i.v. fentanyl. Infrared pupillometry was added as an additional measure of autonomic function because pupillary reflex dilation (PRD) is a sensitive measure of nociception during general anaesthesia.11 12 If deposition of fentanyl in the lower thoracic epidural space does enhance analgesia at a lower lumbar dermatome compared with i.v. administration, it should have the same effect on autonomic nervous system responses as that obtained by decreasing the intensity of the noxious stimulus. Specifically, the previously described shape of PRD13 should be altered in a similar manner by these two interventions because they would both decrease stimulus-evoked activity in the secondary neuron of the nociceptive pathway.
With ethical and procedural approval from the Committee on Human Research at the University of California at San Francisco, the Research Advisory Panel of the State of California, and written informed consent from each volunteer, we studied 10 ASA I volunteers, some of whom simultaneously participated in a previously published study.13 Exclusion criteria included BMI > 30 m kg–2, history of eye or neurological disease, contraindication to epidural analgesia, medications other than oral contraceptives, and age <18 or >35 yr.
Using a randomized crossover design, volunteers consented to participate on 2 days separated by at least 48 h. The volunteers were given general anaesthesia and fentanyl on both study days. However, fentanyl was given intravenously on one study day and epidurally on the other. Treatment order was randomly assigned.
On the epidural study day, a catheter was inserted via the T10–T11 interspace using standard technique. No premedication or test dose was given. On both study days, general anaesthesia was induced with propofol (3 mg kg–1) and vecuronium bromide (0.15 mg kg–1); the trachea was intubated, and the lungs mechanically ventilated to maintain end-tidal carbon dioxide near 4.6 kPa. Anaesthesia was maintained with desflurane 5% end-tidal in oxygen 50% and nitrogen 50%. Lactated Ringer's solution was given at approximately 3 ml kg–1 h–1 through a catheter in the left arm. Vecuronium was infused to provide one-to-two mechanical twitches in response to supra-maximal electrical stimulation of the ulnar nerve at the wrist.
PRD was induced by noxious electrical stimulation at two sites: the L4 and C5 dermatomes. The L4 dermatome, the distal site, is located on the medial aspect of the right leg. Nerve fibres from the L4 dermatome enter the spinal cord between the thoracic spines of T10 and T12,14 so we assumed that enhanced antinociception, if any, would apply to this site. The C5 dermatome, the proximal site, is located on the lateral aspect of the right shoulder. To stimulate the dermatomes, stainless steel needle electrodes (1.5 cm long) were inserted subcutaneously 3 cm apart at each site. Stimulation started 30 min after induction of anaesthesia and continued for 180 min after i.v. or epidural fentanyl administration. Each site was stimulated for 3 s with a 100 Hz electric current (Digistim II, Neurotechnology, Dallas, TX, USA). The two sites were stimulated alternately, with 5 min separating each stimulus. The initial stimulation site was assigned randomly.
Before fentanyl was given, the electrical stimulation was adjusted to deliver variable intensities of stimulation. We categorized stimulation intensities into two categories, a low intensity stimulus (20–70 mA) and a high intensity stimulus (50–90 mA). A post hoc analysis suggested that the difference between the peak and end diameters (P–E diameters) would quantitate the altered shape of PRD produced by high and low intensity currents.
After this period of variable intensity stimulation, but before fentanyl administration, the stimulation current was set to approximately 65 mA and then adjusted slightly, as necessary, to provide equal pupillary dilations from the C5 and L4 sites.
After baseline pupillary measurements (see below), fentanyl (3 µg kg–1 diluted in 10 ml of saline) was given i.v. or epidurally, as randomly designated, in a 5 min period. After a 3 h measurement period, naloxone 400 µg was administered i.v. within 5 min. Ten minutes after the start of naloxone administration, two more measurements were taken, one from each stimulation site. Following these measurements, the neuromuscular block was antagonized and anaesthesia discontinued.
Approximately 1 h after emergence from anaesthesia on the epidural study day, lidocaine 20%, 10 ml was injected into the epidural catheter in a 5 min period. The extent of the sensory block was subsequently evaluated by sensory response to pinprick after 10, 15, and 20 min, and the maximum extent of the block was recorded.
Core temperatures were monitored in the distal oesophagus (Tyco-Mallinckrodt Anesthesiology Products, St Louis, MO, USA), and body temperature was kept near 37°C with active surface warming. Venous arterial blood for fentanyl analysis was sampled from the right arm before and 5, 10, 15, 30, 60, 90, 120, and 180 min after i.v. or epidural fentanyl was given. Plasma specimens were frozen at –20°C for subsequent gas chromatographic analysis.15 The limit of detection was near 0.2 ng ml–1.
Pupillary responses were measured with an infrared pupillometer (Fairville Medical Optics, Buckinghamshire, UK), programmed to scan the pupil at the rate of 20 Hz for 10 s from the start of the tetanic electric stimulus. PRD was quantified by integrating the area above the baseline pupil size for each entire 10 s scan. We have previously described this methodology16 and used it to quantify opioid effect.13 17 Ambient light was maintained near 150 lux, and the contralateral eye was covered during measurements.
Systolic and diastolic arterial pressures were measured oscillometrically (Dinamap TM 1846 SX; Critikon Inc., Tampa, FL, USA) 2 to 3 min after each stimulus. Heart rate and oxyhaemoglobin saturation were monitored continuously using three-lead electrocardiography and a Nellcor N200 pulse oximeter (Hayward, CA, USA), respectively. Resting heart rate and arterial pressure were recorded prior to the start of stimulations (‘before stimulation’). Heart rate and arterial pressure after stimulations were the highest values recorded after each stimulus and before the next stimulus. End-tidal desflurane and end-tidal carbon dioxide were monitored continuously (infrared analyzer, Datex, Finland).
Data analysis
Depression of PRD, heart rate, and systolic arterial pressure for the 3 h study period were each expressed as the AAC as determined by the trapezoid rule and comparisons were made using paired tests. For all P < 0.05, 95% confidence intervals were calculated.
To analyse the effects of i.v. and epidural fentanyl on the shape of PRD, we calculated the average of all 18 measurements at each stimulation site. We then calculated the peak-minus-end diameters (P–E diameter) as previously described and compared these values on the epidural and i.v. study days at both stimulation sites using paired t-tests. We also used two-tailed paired t-tests to compare peak depressant effect and time-to-peak effect between the epidural and i.v. study days.
Analysis of variance was used to compare ‘before stimulation’ values to ‘before fentanyl’ values and also to compare ‘peak effect’ values to ‘before fentanyl’ values. All data are reported as mean (SD) unless otherwise indicated; P < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Participants were 27 (5) yr old, weighed 69 (8) kg, and were 176 (11) cm tall. There were six men and four women. Of the 10 volunteers participating on the first study day, three did not return for the second study day. In one subject, the epidural produced a unilateral block following lidocaine 2%; this volunteer returned for a second epidural study day and results from this second day were used in our analysis. Paired data on fentanyl administration were, therefore, collected from the seven volunteers who completed both arms of the study.
PRD in response to low- and high-intensity stimulating currents was obtained from all 10 volunteers before fentanyl administration. The PRD (Fig. 1, Table 1) was smaller in response to the low-intensity stimulus, predominately because the late phase of PRD was depressed (Fig. 1, Table 1). Changes in heart rate, but not arterial pressure, were significantly smaller in response to the lower current (Table 1).
|
|
Stimulating currents were between 60 and 70 mA before fentanyl and did not differ between the two study days or between the L4 and C5 stimulation sites (Table 2). Before fentanyl administration, pupillary responses were similar after noxious stimulation at the L4 and C5 dermatomes (Table 2) on both the i.v. and epidural study days. P–E diameter did not differ before fentanyl administration (Table 2).
|
Plasma fentanyl concentrations were initially three-fold greater in the i.v. group; but within 15 min, plasma concentrations after epidural and i.v. administration were comparable and remained so for the rest of the measurement period (Fig. 2).
|
The 3 h reduction of PRD (AAC) was greater after L4 stimulation on the epidural study day than on the i.v. study day (Table 2, Fig. 3). In contrast, the suppression of PRD after C5 stimulation was similar with both i.v. and epidural administration of fentanyl (Table 2, Fig. 3). Fentanyl administration decreased heart rate and blood pressure by both routes of administration (Fig. 4, Table 3). No differences were observed in arterial pressure or heart rate between the epidural and i.v. study days and no segmental effects were noted (Fig. 4, Table 3).
|
|
|
The curves representing PRD following epidural and i.v. fentanyl diverged after 50 min with L4 stimulations (Fig. 3), but these curves did not diverge after C5 stimulations (Fig. 3). Time to maximum effect was longer after epidural administration than after i.v. administration at both stimulation sites for PRD reduction, heart rate, and arterial pressure (Tables 2 and 3).
An unusual pupillary response was observed following stimulation of the L4 dermatome after epidural injection: there was an increased depression of PRD at L4 produced by a late reduction of the reflex that was not apparent on the scans following i.v. fentanyl (Fig. 5). This was statistically confirmed by comparing the effect of the two methods of fentanyl administration on the P–E diameters (Table 2). The similarity between this altered PRD reflex shape to that observed after a decrease in the intensity of the noxious stimulus was readily apparent (compare Figs 1 and 5). The change in the shape of the PRD curve and the decrease in PRD after epidural administration at L4 were both reversed by naloxone (Table 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
The autonomic effects of epidural fentanyl in anaesthetized human subjects remain poorly characterized, but are nonetheless important because these measures are often used to gauge nociception during anaesthesia. Ours is the first study to measure the common autonomic responses following nociceptive stimuli before and after bolus injections of epidural fentanyl in anaesthetized subjects and to compare the maximum effect and time course of action of epidural fentanyl with i.v. fentanyl.
Epidural fentanyl suppressed PRD 47 (22)% more than i.v. fentanyl over the 3 h study period. This enhanced effect was not caused by higher serum fentanyl concentrations on the epidural study day and it was a segmental effect, observed only at the L4 stimulation site. We thus confirmed our hypothesis that bolus injections of epidural fentanyl have a segmentally enhanced effect on nociception similar to that observed in awake subjects. Our results are in agreement with Inagaki and colleagues18 who demonstrated a greater reduction of halothane MAC with epidural compared with i.v. fentanyl and with Harakuni and colleagues19 who demonstrated less hormonal change during gastrectomy with epidural fentanyl.
We confirmed the spinal effect of epidural fentanyl by showing that this route of drug administration produced the same alteration in the shape of PRD as that brought about by decreasing the intensity of the stimulus. The mechanism of PRD is thought to be brought about through activation of a putative inhibitory neuron acting on the pupilloconstrictor nucleus.20 Because the initial phase of the dilation was minimally affected, it appears that the primary alteration brought about by decreasing the intensity of the stimulus, and by epidural fentanyl, is to shorten the duration and not the initial frequency of the neuronal spike train. Similar findings have been observed in the prefrontal cortex where the encoding of nociceptive stimulus intensity is related to spike train duration and not initial firing frequency.21
Fentanyl decreased heart rate and arterial pressure after either i.v. or epidural administration. There was no advantage to either injection method in suppressing these autonomic responses. Additionally, we did not find any segmental differences or any trends which would suggest that a larger sample size would show a segmental effect of epidural fentanyl on heart rate or arterial pressure.
The haemodynamic reflexes that we observed were small compared with PRD and this may account for our inability to detect advantages to epidural administration using arterial pressure and heart rate as our measurement tool. Other studies have also demonstrated greater sensitivity of PRD as a measure of nociception compared with circulatory reflexes.11 12 Possibly, a stronger stimulus, a larger sample size, analysis of heart rate variability, or invasive monitoring would have allowed us to demonstrate a segmental advantage of epidural compared with i.v. fentanyl by using haemodynamic data as our outcome measures.
On the other hand, heart rate and arterial pressure may not be a precise measure of nociception during anaesthesia. One study that used haemodynamic reflexes as a measure of nociception specifically concluded that there was no advantage to administration of epidural compared with i.v. fentanyl.22 Differences in the central mechanisms of pupillary and circulatory reflexes might explain these discrepancies. Whereas heart rate and arterial pressure responses to noxious stimulation are spinal or lower brain stem reflexes which can even be observed in brain-dead subjects,23 24 PRD is a parasympathetic reflex that requires a partially intact mesencephalic reticular formation.20 25 26 Clearly, nociception during general anaesthesia is difficult to define, but pupillary dilation indicates activation of more rostral brain regions than the circulatory responses.
Could this epidural effect of fentanyl be used advantageously during surgery in which an epidural catheter is in place, as they commonly are for extensive abdominal operations? When these operations extend into the pelvis and perineum (prostatectomy, abdominal-perineal resection, hysterectomy) a low thoracic epidural catheter injected with only local anaesthetics may not provide adequate analgesia for noxious stimuli arising from caudal nociceptors. Epidural fentanyl would enhance antinociception in these caudal segments to a greater extent than i.v. fentanyl during a 3 h procedure. With our experimental design, we were unable to define the rostral extent of enhanced antinociception by epidural fentanyl.
Electrical stimulation is an unnatural stimulus because it bypasses the nociceptor. It also activates all nerve endings in the dermis, thus initiating central processes not associated with pain. Nevertheless, we believe that this form of stimulation is an appropriate method to study opioid suppression of intraoperative nociception. We have no evidence that pupillary dilation during general anaesthesia is brought about by activation of large afferent fibres that sub-serve proprioception and touch.16 We do not contend, however, that this form of stimulation is suppressed by opioids in the same manner as all of the stimuli occurring during surgical manipulations. Brennum27 showed, for example, that opioid suppression of pain elicited by pressure, electrical tetanus, pinch, or heat is different. Our study can only address the question of whether epidural fentanyl can produce segmental analgesia for this particular type of noxious stimulus, that is, electrical tetanic stimulation.
In summary, using pupillometry as a measure of autonomic nerve activity, we have shown enhanced segmental analgesia at a low lumbar dermatome following deposition of epidural fentanyl in the thoracic epidural space. We were unable to detect this difference when using haemodynamic responses as a measure of nociception. The enhanced analgesia effect of epidural fentanyl was not apparent until the second hour after administration, it was more intense than after i.v. fentanyl administration and dissipated within 3 h.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Supported by NIH Grant GM 061655 (Bethesda, MD), the Gheens Foundation (Louisville, KY), the Joseph Drown Foundation (Los Angeles, CA), and the Commonwealth of Kentucky Research Challenge Trust Fund (Louisville, KY). The study sponsors were not involved in the design of the study, the collection, analysis, or interpretation of the data, or the preparation of the manuscript. None of the authors has a personal financial interest in this research.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
1 Li X, Angst MS, Clark JD. (2001) Opioid-induced hyperalgesia and incisional pain. Anesth Analg 93:204–9.
2 Guignard B, Bossard AE, Coste C, Sessler DI, Lebrault C, Alfonsi P, Fletcher D, Chauvin M. (2000) Acute opioid tolerance: intraoperative remifentanil increases postoperative pain and morphine requirement. Anesthesiology 93:409–17.[CrossRef][Web of Science][Medline]
3 Steinlechner B, Dworschak M, Birkenberg B, Grubhofer G, Weigl M, Schiferer A, Lang T, Rajek A. (2006) Magnesium moderately decreases remifentanil dosage required for pain management after cardiac surgery. Br J Anaesth 96:444–9.
4 Kehlet H. (2005) Postoperative opioid sparing to hasten recovery: what are the issues? Anesthesiology 102:1083–5.[CrossRef][Web of Science][Medline]
5 Coda BA, Brown MC, Schaffer R, Donaldson G, Jacobson R, Hautman B, Shen DD. (1994) Pharmacology of epidural fentanyl, alfentanil, and sufentanil in volunteers. Anesthesiology 81:1149–61.[CrossRef][Web of Science][Medline]
6 George MJ. (2006) The site of action of epidurally administered opioids and its relevance to postoperative pain management. Anaesthesia 61:659–64.[CrossRef][Web of Science][Medline]
7 Liu SS, Gerancher JC, Bainton BG, Kopacz DJ, Carpenter RL. (1996) The effects of electrical stimulation at different frequencies on perception and pain in human volunteers: epidural versus intravenous administration of fentanyl. Anesth Analg 82:98–102.[Abstract]
8 Ginosar Y, Riley ET, Angst MS. (2003) The site of action of epidural fentanyl in humans: the difference between infusion and bolus administration. Anesth Analg 97:1428–38.
9 Sabbe MB and Yaksh TL. (1990) Pharmacology of spinal opioids. J Pain Symptom Management 5:191–203.[CrossRef][Medline]
10 Kitahata LM, Kosaka Y, Taub A, Bonikos K, Hoffert M. (1974) Lamina-specific suppression of dorsal-horn unit activity by morphine sulfate. Anesthesiology 41:39–48.[Web of Science][Medline]
11 Larson MD, Sessler DI, Washington DE, Merrifield BR, Hynson JA, McGuire J. (1993) Pupillary response to noxious stimulation during isoflurane and propofol anesthesia. Anesth Analg 76:1072–8.
12 Constant I, Nghe MC, Boudet L, Berniere J, Schrayer S, Seeman R, Murat I. (2006) Reflex pupillary dilatation in response to skin incision and alfentanil in children anaesthetized with sevoflurane: a more sensitive measure of noxious stimulation than the commonly used variables. Br J Anaesth 96:614–9.
13 Larson MD, Berry PD, May J, Bjorksten A, Sessler DI. (2004) Latency of pupillary reflex dilation during general anesthesia. J Appl Physiol 97:725–30.
14 Larsell O. (1953) In Schaeffer JP (Ed.). The Nervous System, Spinal Nerves, Morris' Human Anatomy(McGraw-Hill, London) pp. 1252.
15 Bjorkman S and Stanski D. (1988) Simultaneous determination of fentanyl and alfentanil in rat tissues by capillary column gas chromatography. J Chromatography 433:95–104.[Web of Science][Medline]
16 Larson MD, Sessler DI, Washington DE, Merrifield BR, Hynson JA, McGuire J. (1993) Pupillary response to noxious stimulation during isoflurane and propofol anesthesia. Anesth Analg 76:1072–8.
17 Larson MD, Kurz A, Sessler DI, Dechert M, Bjorksten AR, Tayefeh F. (1997) Alfentanil blocks reflex pupillary dilation in response to noxious stimulation but does not diminish the light reflex. Anesthesiology 87:849–55.[CrossRef][Web of Science][Medline]
18 Inagaki Y, Mashimo T, Yoshiya I. (1992) Segmental analgesic effect and reduction of halothane MAC from epidural fentanyl in humans. Anesth Analg 74:856–64.
19 Harukuni I, Yamaguchi H, Sato S, Naito H. (1995) The comparison of epidural fentanyl, epidural lidocaine, and intravenous. fentanyl in patients undergoing gastrectomy. Anesth Analg 81:1169–74.[Abstract]
20 Loewenfeld IE. (1958) Mechanism of reflex dilation of the pupil: historical review and experimental analysis. Doc Opthalmol 12:184–448.
21 Zhang R, Tomida M, Katayama Y, Kawakami Y. (2004) Response durations encode nociceptive stimulus intensity in the rat medial prefrontal cortex. Neuroscience 125:777–85.[CrossRef][Web of Science][Medline]
22 Guinard JP, Carpenter RL, Chassot PG. (1995) Epidural and intravenous fentanyl produce equivalent effects during major surgery. Anesthesiology 82:377–82.[CrossRef][Web of Science][Medline]
23 Yang LL, Niemann CU, Larson MD. (2003) Mechanism of pupillary reflex dilation in awake volunteers and in organ donors. Anesthesiology 99:1281–6.[CrossRef][Web of Science][Medline]
24 Wetzel RC, Setzer N, Stiff JL, Rogers MC. (1985) Hemodynamic responses in brain dead organ donor patients. Anesth Analg 64:125–8.
25 Larson MD, Tayefeh F, Sessler DI, Daniel M, Noorani M. (1996) Sympathetic nervous system does not mediate reflex pupillary dilation during desflurane anesthesia. Anesthesiology 85:748–54.[CrossRef][Web of Science][Medline]
26 Loewy AD, Araujo JC, Kerr FW. (1973) Pupillodilator pathways in the brain stem of the cat: anatomical and electrophysiological identification of a central autonomic pathway. Brain Res 60:65–91.[CrossRef][Web of Science][Medline]
27 Brennum J. (1996) Quantitative sensory examination of epidural anaesthesia and analgesia in man. Acta Physiol Scand Suppl 634:1–27.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||