Proceedings of the Anaesthetic Research Society Meeting, Cardiff Meeting, July 11–12, 2002

British Journal of Anaesthesia 89 (4): 670±81P (2002)

ABSTRACTS

Proceedings of the Anaesthetic Research Society Meeting University of Wales College of Medicine, Cardiff July 11±12, 2002 (The name of the author presenting the paper is shown in bold type. *Indicates non-member. All authors have certi®ed that, where appropriate, studies have been conducted with the approval of the relevant Human Ethics Committee or Animal Experimental Review Committee.)

VERBAL PRESENTATIONS Comparison of two Macintosh laryngoscope blades T. Asai, S. Matsumoto* and K. Shingu* Department of Anaesthesiology, Kansai Medical University, Moriguchi City, Osaka, 570-8507, Japan

Sir Robert Macintosh stated that `the precise shape or curve of the blade does not seem to matter much'1 and that `there is so little difference in the performances of the various blades'.2 Nevertheless, two major types of Macintosh laryngoscope blades (standard and `English' type (E type)) are available: compared with the standard blade, the length of the E type blade is longer, its curve is more continuous across the entire length of the blade, and the height of the blade ¯ange is shorter. In addition, the ¯ange of the E type blade continues close to the blade tip, whereas the ¯ange of the standard blade ends more abruptly and further away from the tip. We felt that, compared with the standard blade, the E type blade might provide a better glottic view, and thus assessed the ease of laryngoscopy between the two Macintosh blades. We compared the two types of laryngoscope blades manufactured by Welch Allyn (NY, USA) in 300 patients, 116 males and 184 females, mean age 49 (range: 18±82) yr, mean height 160 (130±180) cm and mean weight 58 (33±94) kg. In a random crossover design, after induction of general anaesthesia and neuromuscular blockade, the two blades were inserted in turn in each patient. The view of the glottis at laryngoscopy was graded according to Cormack and Lehane for each blade. The McNemar test was used to compare the incidence of dif®cult laryngoscopy (grade 3 or 4) between the two blades. P<0.05 was considered signi®cant. There was no difference in the view of the glottis between the two laryngoscope blades in 220 of 300 patients, whereas there was a difference in the remaining 80 patients. Of these 80 patients, the view was better for the E type in 63 patients whereas the standard was better in the remaining 17 patients. Laryngoscopy was dif®cult (grade 3 or 4) at least for one laryngoscope blade in 42 of 300 patients (14%). Of these 42 patients, there was no difference in the grade between the two blades in 14 patients (33%), the view was better for the E type in 25 patients (60%), and the standard blade better in the remaining three patients (7%). The McNemar test showed that in patients in whom the laryngoscopy was dif®cult (grade 3 or 4), the view of the glottis was signi®cantly

better for the E type blade than for the standard blade (P<0.001). The 95% con®dence interval for the possibility of the type E being better than the standard blade in patients with dif®cult laryngoscopy was 45±74%. We have found that the type E laryngoscope blade provided a better glottic view signi®cantly more frequently than the standard blade in patients in whom laryngoscopy was unexpectedly dif®cult. Keywords: intubation tracheal, dif®cult; equipment, laryngoscopes

References

1 Macintosh RR. Lancet 1943; i: 205 2 Macintosh RR. Lancet 1944; i: 485

Effect of repeated exposure on the incidence of adverse airway events during brief nasal inhalations of anaesthetics A. R. Wilkes, N. Raj* and J. E. Hall* Department of Anaesthetics and Intensive Care Medicine, University of Wales College of Medicine, Cardiff CF14 4XN, UK

We investigated whether the different levels of humidi®cation provided by various breathing system ®lters affected the incidence of adverse airway events during brief nasal inhalations of volatile anaesthetics. Twelve volunteers who had airways that responded adversely to 2.0 MAC des¯urane were recruited. Each volunteer came to the laboratory on four separate occasions, once for each of the four anaesthetics: sevo¯urane, halothane, iso¯urane and des¯urane. The volunteer breathed nasally through a facemask attached to one of three different ®lters and, after 3 min, inhaled a single breath of 0.5, 1.0 or 2.0 MAC anaesthetic, in air. The volunteer then breathed air only until the expired MAC level was zero, before the next breath of anaesthetic. When all three levels of the anaesthetic had been inhaled once with each ®lter (nine breaths), each combination of anaesthetic level and ®lter was repeated twice more, each time in a different order. Although the ®lter used, and hence humidi®cation level, was not a signi®cant explanatory variable,1 further analysis has shown that repeated exposure was (Table 1). Repeated exposure to anaesthetics led to reduced irritancy to the airways even when the expired level of the anaesthetic was zero between breaths. This may be due to desensitization of, or the

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

Proceedings of the Anaesthetic Research Society Table 1 Number of breaths during which an adverse airway event (reduction or cessation of inspiratory or expiratory ¯ow or coughing) occurred. Summed for all humidi®cation levels. Results from GLIM (Generalized Linear Interactive Modelling) logistic regression analysis: anaesthetic, P<0.001; anaesthetic level, P<0.001; humidi®cation, P=0.09; `repeat', P<0.001 Anaesthetic

Sevo¯urane Halothane Iso¯urane Des¯urane

MAC level

0.5 1.0 2.0 0.5 1.0 2.0 0.5 1.0 2.0 0.5 1.0 2.0

Incidence of adverse airway events (n=36) First exposure

Second exposure

Third exposure

4 4 8 6 6 19 5 6 24 6 14 34

1 2 7 2 5 17 3 6 19 5 11 28

2 3 5 3 3 12 2 4 13 5 9 27

Total (n=108)

7 9 20 11 14 48 10 16 56 16 34 89

result of very low dose effect of anaesthetic on, the chemoreceptors in the upper airways. Acknowledgement: We gratefully acknowledge the funding provided by the Association of Anaesthetists of Great Britain and Ireland. Keywords: anaesthetic techniques, inhalation; potency, anaesthetic, tolerance

Table 2 Peak force (N) that could be exerted when pressing the multipleuse and single-use bougies against the force transducer when held at different distances Distance held from tip (cm)

Median peak force (N) [IQR] Multiple-use

Single-use

Mann±Whitney U test P-value

10 20 30 40

4.4 1.7 0.9 0.4

10.8 [10.1±11.5] 4.8 [4.1±5.0] 2.3 [2.1±2.4] 1.2 [1.2±1.3]

<0.0001 <0.0001 <0.0001 <0.0001

[3.7±5.4] [1.5±2.2] [0.7±1.2] [0.4±0.5]

The peak force that could be exerted on the disc at every position by the single-use bougies was two to three times greater than that which could be exerted by the multiple-use bougies (Table 2). The peak force decreased with increasing distance from the tip. As the force exerted by the tip of the bougie is over a relatively small area, the peak pressure that could be applied to the tissue is considerable. This study suggests that the single-use bougie is more likely to cause trauma to tissue during placement. A recent in vitro study demonstrated that successful placement of the single-use bougie was much less likely than the multiple-use bougie.3 Although the contamination risk may be less, using the single-use bougie in place of the multiple-use bougie may cause an increase in patient morbidity. Keywords: equipment, gum elastic bougie; intubation tracheal

References

1 Latto IP, Stacey M, Mecklenburgh JS, et al. Anaesthesia 2002; 57: 379±84 2 Cupitt JM. Anaesthesia 2000; 55: 466±8 3 Annamaneni R, Hodzovic I, Wilkes AR, et al. Eur J Anaesthesiol 2002; 19: A141

Reference

1 Wilkes AR, Raj N, Hall JE. Eur J Anaesthesiol 2002; 19: A157

A clinical appraisal of the Bon®ls Intubating Fibrescope

Comparison of the peak forces that can be exerted by multiple-use and single-use bougies in vitro

M. Halligan*, B. Weldon* and P. Charters University Hospital Aintree, Longmoor Lane, Liverpool, UK

A. R. Wilkes, I. Hodzovic* and I. P. Latto* Department of Anaesthetics and Intensive Care Medicine, University of Wales College of Medicine, Cardiff CF14 4XN, UK

The multiple-use gum elastic bougie is widely used as an aid during dif®cult intubation.1 However, it is dif®cult to ensure that the multiple-use bougie is cleaned adequately before re-use.2 A single-use bougie is also available, but is less widely used. The single-use bougie appears to be stiffer than the multiple-use version. This might cause an increase in the risk of trauma to tissue during placement. The peak forces that could be exerted by the tips of the two types of bougies were therefore investigated in the laboratory. Five samples of both multiple-use (Eschmann Healthcare Tracheal Tube Introducer) and single-use bougies (Portex Tracheal Tube Introducer) were each held at four different positions (10, 20, 30 and 40 cm from the tip) and pressed three times onto a disc attached to a force transducer with a resolution of 0.1 N (Mecmesin PFI200N). A depression in the disc prevented the tip of the bougie from slipping as the bougie bent. The peak force was recorded for each test and the results from the two types of bougie were compared for each position using the Mann± Whitney U test. A P-value of <0.01 was considered signi®cant because of the multiple comparisons.

The Bon®ls Intubating Fibrescope is a rigid stylet-type laryngoscope. It enables a tracheal tube to be advanced beyond the vocal cords under direct vision. Our aim was to con®rm its use as safe, reliable and atraumatic. Sixteen patients requiring tracheal intubation for routine ENT surgery were recruited. All were ASA I or II and aged 17±64 yr. Exclusion criteria were a risk of aspirating gastric contents and inter-incisor distance <3 cm. One anaesthetist performed all the intubations. Patients were given midazolam 2 mg intravenously then preoxygenated for 3 min. Next fentanyl 1 mg kg±1, propofol 2.5 mg kg±1 and atracurium 0.5 mg kg±1 were administered and after a further 3 min, intubation attempted. If after a further 2 min there was no success an assistant performed a sustained jaw thrust. Three minutes was the maximum time allowed for a successful intubation and this was the primary outcome measure. Secondary outcomes included time taken (including need for jaw thrust) and evidence of trauma as assessed by post intubation conventional direct laryngoscopy. Subjective dif®culty was scored with rating scale (0±10) with relative contribution of patient versus instrument/user factors. Intubation was achieved in all cases. Average time taken was 69 (range 34±180) s. Jaw thrust was used in 2/16. There was no evidence of traumatic intubation. The Graphical Representation of Instrumentation Dif®culty (GRID) is a novel method for

671P

Proceedings of the Anaesthetic Research Society

Fig 2 BIS scores during stages of induction. Fig 1 Graphical representation of instrumentation dif®culty with individual patients.

Keywords: anaesthesia, depth; children representing dif®culty in terms of time taken, rating score and relative patient vs instrument/user contribution (Fig. 1). These results suggest that the Bon®ls Intubating Fibrescope can be useful when used for routine orotracheal intubations. Jaw thrust appeared to be a helpful manoeuvre. Acknowledgement: Karl Storz Endoscopy (UK) loaned the Bon®ls Intubating Fibrescope. Keywords: equipment, laryngoscopes ®breoptic

References

1 Sigl JC, Chamoun NC. J Clin Monitoring 1994; 10: 392±404 2 Degoute CS, Macabeo C, Dubreuil R, et al. Br J Anaesth 2001; 86: 209± 12

A preliminary CoMFA model for chemically diverse intravenous general anaesthetics J. C. Sewell and J. W. Sear Nuf®eld Department of Anaesthetics, University of Oxford, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK

BISÐcautionary interpretation of data in children M. S. Chawathe*, J. E. Hall*, J. S. Mecklenburgh and R. M. Jones* University of Wales College of Medicine, Cardiff, UK

Bispectral index or BIS is a processed EEG variable, derived from the EEG by a computer algorithm, scaled from 0±100, which is stated to re¯ect a patient's state of hypnosis and responsiveness.1 Thus it claims to improve titration of anaesthetic agents and monitor risk of awareness. Experience to date suggests that BIS monitoring in children and infants should provide similar bene®ts to those in adults.2 However, validation of BIS may present more of a challenge in children compared to adults, because the EEG in infants and children changes as the brain matures. The aim of our study was to analyse the interpretation of the BIS data collected in a study for gaseous induction in 60 ASA I and II children. In this preliminary work with BIS (using the Aspect 1000) we used data from a previous study of 60 ASA I±II children aged 5± 10 yr undergoing gaseous induction with sevo¯urane. BIS scores were measured preinduction, at loss of eyelash (LOE) re¯ex, at pupil size 2, and at 2 min and 4 min after LOE re¯ex. Regression analysis was used to determine any value of BIS in predicting heart rate and arterial pressure. Mann±Whitney test was used to determine any difference in BIS between children experiencing apnoea or no apnoea. A P<0.05 was considered signi®cant. Excessive range of values of BIS, from 75 to 8 at pupil size 2, was detected in this study (Fig. 2). At 2 min after LOE re¯ex some individual values were as low as 5 and as high as 88, when the depth of anaesthesia should have approached stability in all children. The median BIS scores in the children with and without apnoea differed only by a non-signi®cant 6%. There was no signi®cant relationship between BIS values and heart rates and arterial pressures recorded at the corresponding time (R2=0.01 and 0.09 respectively). Handedness of the individual and right- or leftsided placement of the BIS sensor were not noted during this study. We conclude that the algorithm deemed suitable for adults appears not to be suitable for use in children in this retrospective study.

We have previously demonstrated that molecular shape and electrostatic potential (a measure of charge distribution) are important determinants of activity in i.v. general anaesthetics.1 2 The aim of this study was to derive an activity model based on the spatial distribution of these molecular properties, using comparative molecular ®eld analysis (CoMFA). A group of 14 chemically diverse i.v. general anaesthetics were considered. The free plasma concentrations (log EC50) that abolish movement to a noxious stimulus were obtained from the literature. The anaesthetics were divided into a training set of nine compounds (eltanolone, minaxolone, Org 21465, pentobarbital, thiopental, methohexital, R-ketamine, propofol, chlormethiazole) used to formulate the activity model; and a test set of ®ve agents (alphaxalone, thiamylal, Org 25435 (an alpha amino acid phenolic ester derivative), S-ketamine and R-etomidate) used to independently evaluate the model's predictability. Computer-based structures of the anaesthetics were generated, and low energy conformers identi®ed using a random search (SYBYL 6.7, Tripos Inc.). The conformers were aligned so as to maximize their similarity in shape and electrostatic potential to the most active agent in the group, eltanolone. The conformers with the highest similarity, calculated as Carbo indices (ASP 3.22, Accelrys Inc.), were retained. The aligned conformers were placed in a rectangular grid consisting of a lattice of points separated at Ê intervals. The steric and electrostatic interaction energies 1A between the anaesthetics and a probe atom (sp3 carbon, +1 charge) at each grid point were evaluated, and correlated with in vivo potency (SYBYL). Thus an activity model was formulated, based on the spatial distribution of steric and electrostatic features of the anaesthetic molecules. The ®nal model explained 92.8% of the variance in the observed activities of the training set compounds (n=9, P<0.001). The model accurately predicted the potencies of three of the ®ve test set agents: Org 25435 predicted log EC50=5.44 (observed log EC50=5.42), thiamylal 4.59 (4.81) and S-ketamine 5.28 (5.37). However, the model was less effective at predicting the activities of R-etomidate 5.08 (5.99) and alphaxalone 6.69 (5.06). Analysis of the relative contributions of the molecular properties at

672P

Proceedings of the Anaesthetic Research Society

individual lattice points identi®ed areas where the spatial distribution of molecular bulk and electrostatic potential are important in determining in vivo potency. The results demonstrate that a single CoMFA model with good predictability can be formulated for chemically diverse i.v. general anaesthetics. The mapping of the spatial arrangement of the key electrostatic and steric features enabled the derivation of a preliminary three-dimensional `pharmacophore' for i.v. anaesthetic activity. Acknowledgement: This work was supported by a project grant from the British Journal of Anaesthesia. Keywords: anaesthetics, i.v.; model, computer simulation

Table 3 Study parameters during induction with sevo¯urane in three different groups Group 3B (n=14) Time to induction (s) End-tidal sevo¯urane at 30 s % End-tidal carbon dioxide (kPa)

57 (42±72) 3.0 (2.7±3.3) 4.2 (3.8±4.6)

In¯uence of oxygen therapy on heart rate after abdominal surgery Department of Anaesthesia Critical Care and Pain Medicine, Royal In®rmary, 51 Little France Crescent, Edinburgh EH16 4SA, UK

1 Sewell JC, Sear JW, Halsey MJ. Br J Anaesth 1999; 82: 470P 2 Sewell JC, Sear JW. Br J Anaesth 2002; 88: 166±74

In¯uence of fresh gas and circuit on induction of anaesthesia with sevo¯urane C. L. Knaggs* and G. B. Drummond Department of Anaesthesia Critical Care and Pain Medicine, Royal In®rmary, 51 Little France Crescent, Edinburgh EH16 4SA, UK

Induction of anaesthesia with sevo¯urane can be done with normal breathing using a low ¯ow of oxygen as fresh gas.1 Although this method maintains ventilation well, it must limit the rate of supply of anaesthetic to the patient. We studied the effects of different gas supply rates and circuits on inhalation induction. Sevo¯urane was used to induce anaesthesia in otherwise healthy patients about to have gynaecological procedures. They were randomly allocated by sealed envelope to one of the following: oxygen 3 litre min±1 from a Bain circuit (group 3B), oxygen 6 litre min±1 from a Bain circuit (group 6B), or oxygen 9 litre min±1 from a Magill circuit (group 9M). We gave the sevo¯urane by doubling the concentration from 0.5%, every three breaths, to 8%, and then maintaining that fresh gas concentration until loss of consciousness. The patient held her arm up at about 45° during induction. We measured the time from ®rst giving 8% to the time for the arm to come down to the horizontal, as the induction time. We measured the concentration of sevo¯urane and carbon dioxide in the gas in the mask, and respired ¯ow, and recorded these with an AD converter and computer. We calculated the best-®t polynomial to give an end-tidal sevo¯urane concentration/time plot for the three groups. We compared the end-tidal concentrations at 30 s and the time to induction in the three groups by ANOVA and post hoc Bonferroni tests. Values are median (quartiles). We asked 52 suitable patients to participate and 39 agreed. Of these, 37 were studied and 35 yielded adequate data. No patient had hypoxaemia or other adverse events. Table 3 shows the results. There was no signi®cant difference between induction times, although the sevo¯urane concentration at 30 s was signi®cantly less in the low ¯ow group (P<0.001). End-tidal carbon dioxide was signi®cantly less with the Magill circuit and 9 litre min±1. We conclude that rebreathing may assist inhalation induction of anaesthesia.

In patients after major abdominal surgery, oxygen treatment has been reported to reduce heart rate, but the change in heart rate did not appear to correlate with the change in oxygen saturation.1 We were able to assess this effect in the course of analysis of data from a study comparing oxygenation at night in patients randomly allocated to receive epidural or patient-controlled analgesia (PCA). Patients were given oxygen or air by nasal cannula for alternate hours, for 6 h for the ®rst two nights after surgery. We continuously recorded the ECG, and oxygen saturation with a ®nger probe pulse oximeter (Ohmeda 3400) using commercial software (Cardas). We used the `rate' output from the oximeter, and excluded periods of artefact from both recordings, when the pulse records did not correspond with the ECG record and when the oximeter pulse wave was erratic. We recorded the mean heart rate in the last 40 min of each hour of treatment. We related the change in heart rate and change in oxygen saturation for each change in inspired gas. Patients who did not achieve satisfactory oxygen saturation while breathing air were automatically given oxygen so full records were not available in all cases. One patient was converted from epidural to PCA for the second night. We had records available for 18 patients. Table 4 summarizes the results. Oxygenation was more satisfactory on the ®rst night, and the effects of oxygen on oxygen saturation were more clear-cut. There was no signi®cant change in heart rate, either in the groups receiving epidural or PCA for analgesia, or overall. We conclude that there is no evidence, in these patients, that oxygen administration affects heart rate. Acknowledgement: The original study was funded by the Scottish Hospitals Endowment Trust. Keywords: oxygen, saturation; heart, heart rate

Reference

1 Rosenberg-Adamsen S, Lie C, Bernhard A, Kehlet H, Rosenberg J. Anesthesiology 1999; 90: 380±4

Table 4 Mean (SD) changes in all the eligible subjects (7 epidural, 11 PCA), for the ®ve changes from one inspired gas to the other Epidural

Keywords: induction, anaesthesia; anaesthetic techniques, inhalation 1 Guracha Boru K, Drummond GB. Br J Anaesth 1999; 83: 650±3

64 (47±92) 48 (43±66) 2.0 (1.8±2.4) 3.0 (2.8±3.2) 5.5 (5.0±5.9) 4.8 (4.6±5.2)

Group 9M (n=10)

T. P. Banerjee*, A. F. Nimmo and G. B. Drummond

References

Reference

Group 6B (n=11)

Change in oxygen saturation (%) Change in heart rate (beats min±1)

673P

PCA

Night 1

Night 2

Night1

Night 2

4 (2.8) 0 (1.5)

2.3 (2.7) 0 (1.2)

2.5 (2) 0 (1.3)

2.3 (2.5) ±1 (1.4)

Proceedings of the Anaesthetic Research Society

Haemodynamic effects of the prone position for spinal surgery: TIVA versus inhalation anaesthesia P. S. Sudheer*, B. Ateleanu*, J. E. Hall*, S. W. Logan* and M. Harmer Department of Anaesthetics, University Hospital of Wales, Cardiff CF14 4XN, UK

Recently, total intravenous anaesthesia (TIVA) using propofol has become a widely used anaesthetic technique. Recent work has suggested inhalation anaesthesia is associated with improved cardiovascular stability compared with propofol anaesthesia.1 In the prone position, cardiac output (CO) is decreased because of decreased preload.2 There are no data in the literature describing the in¯uence of intravenous anaesthesia on haemodynamic stability in the prone position.2 Consequently, a study was designed to ascertain relative cardiovascular stabilities of these two techniques in prone patients. Forty patients aged 18±75 yr having lumbar spinal surgery were recruited. Patients were randomized to receive either propofol TIVA or INH anaesthesia using iso¯urane together with fentanyl, morphine and vecuronium. The lungs were ventilated with tidal volumes of 7±10 ml kg±1, respiratory rate of 8±12 bpm, PEEP=0, to maintain a PE¢CO2 of 4.5±5 kPa. We measured the CO, our primary variable, using the NICO monitor. NICO calculates the CO using the technique of `differential Fick partial rebreathing'. The NICO also calculates the stroke volume and systemic vascular resistance. We also measured the mean arterial pressure and heart rate. A two-way ANOVA performed on haemodynamic variables showed CO to be signi®cantly more stable with inhalation than TIVA anaesthesia during patient rotation (Table 5). Table 5 Haemodynamic changes on changing body position in anaesthetized patients. Values are given as mean (SD) Changes (%)

Cardiac output Mean arterial pressure Heart rate SVR

Supine to prone

Prone to supine

Inhalation

TIVA

Inhalation

TIVA

±12 (10) ±4 (12) 2 (10) 15 (38)

±23 (9) 8 (25) 3 (8) 40 (55)

18 11 12 ±4

29 (21) ±9 (11) 12 (17) ±10 (22)

(15) (21) (22) (21)

We conclude that inhalation anaesthesia using iso¯urane is more cardiovascularly stable than propofol TIVA when patients are turned prone and when they are turned supine again. We cannot exclude changes in pulmonary haemodynamics on moving to the prone position as the cause of these ®ndings. Keywords: position, prone; heart, cardiac output

Twenty-four patients scheduled for either nephrectomy or radical cystectomy were consecutively enrolled into the study. Exclusion criteria were current history of angina, signi®cant chronic lung disease, morbid obesity, signi®cant psychiatric history, and the need for more extensive surgery. All patients had an arterial line and thoracic epidural placed prior to induction of anaesthesia, and received a standardized anaesthetic consisting of a sufentanil infusion supplemented by an inhalation agent and neuromuscular relaxation. In the intervention group, the NICO2 cardiac output monitor (Novametrix) was used to guide perioperative volume replacement. The goal was to maintain the cardiac index above 3 litre min±1 m±2 throughout the case with either volume replacement or a dobutamine infusion if volume failed. In the control group, arterial pressure, heart rate and, where possible, urine output were used to guide resuscitative efforts. In both groups, the haemoglobin was maintained between 9 and 11 g dl±1 with packed red cell transfusions. Between group statistical comparisons were performed using the unpaired Student's ttest. The NICO2 monitor performed reliably in all cases and no patient was excluded because of failure of the device to provide consistent recordings. There was no difference in mean age between the NICO2 (65 yr) and the control (57 yr) groups. The two outcome variables in which we were interested were speed of recovery of gastrointestinal function, as measured by the time to tolerating a soft diet, and time to hospital discharge. Overall, the difference in the times to diet toleration (84 vs 113 h) and hospital discharge (5.1 vs 6.6 days) in the NICO2 (n=12) vs control (n=12) group were not signi®cantly different. However, examining the radical cystectomy patients separately (n=12) showed the time to diet toleration (123 vs 176 h) and hospital discharge (6.8 vs 9.3 days) was signi®cantly different (P<0.05) between the NICO2 vs control group. In contrast, in the nephrectomy patients (n=12) there was no difference in either time to diet toleration (46 vs 50 h) or hospital discharge (3.5 vs 3.8 days) between the NICO2 vs control group. The mean duration of surgery was 9.0 h for a radical cystectomy vs 4.4 h for a nephrectomy (P<0.01), and the mean difference in intraoperative volume replacement was 15 vs 7.3 litres (P<0.01). In summary, use of ¯ow guided therapy to guide intraoperative resuscitative efforts signi®cantly reduced the time to recovery of gastrointestinal function and hospital stay in patients undergoing extensive procedures associated with major intraoperative ¯uid shifts but it did not alter the hospital course of patients undergoing briefer and less extensive surgical resections. Acknowledgement: funding Wallingford, CT, USA.

source:

Novametrix

Inc.,

Keywords: ¯uids, replacement; surgery

References

1 Kirkbridge DA, Parker JL, Buggy DJ. Anesth Analg 2001; 93: 1185±7 2 Yokohama M, Ueda W, Hirakawa M, et al. Acta Anaesthesiol Scand 1991; 35: 741±4

Comparison of ¯ow guided versus conventional volume replacement in major elective surgery S. M. Lowson* Department of Anesthesiology, University of Virginia Health System, Charlottesville, Virginia, USA

Fluid replacement during Picolaxq bowel preparation: the consequences for anaesthesia S. T. A. Vaughan*, A. MacKillop*, J. M. Fryer*, A. S. Laurence and S. Forshaw* Department of Anaesthetics, Preston, UK

Picolaxq contains magnesium citrate, which acts as an osmotic laxative and sodium picosulphate, which decreases water and electrolyte absorption while increasing colonic motility. When used for preoperative bowel preparation, it can result in

674P

Proceedings of the Anaesthetic Research Society

dehydration with consequent dif®culties during anaesthesia.1 Administration of preoperative ¯uid has been shown to mitigate this.2 We investigated whether this would improve the stability of anaesthesia. Fifty patients ASA I to III, presenting for major surgery necessitating bowel preparation were studied. Patients were weighed prior to administration of bowel preparation and again on the morning of surgery. All patients were premedicated with temazepam 20 mg and metoclopamide 10 mg orally. Patients were randomized after recruitment to control, or an intravenous preoperative ¯uid load of 2 litres of normal saline given on the night before surgery. A radial artery catheter was sited prior to induction of anaesthesia using thiopental to loss of eyelash re¯ex and vecuronium 0.1 mg kg±1, given by an anaesthetist blind to the patient group. Patients were monitored for change in arterial pressure after induction and prior to intubation. Vasopressors were given, on the clinical judgement of the blinded anaesthetist. Intraoperative blood loss, ¯uid requirements and urine output were noted. As expected, weight loss due to bowel preparation was signi®cantly less in the ¯uid group (Table 6). Mean arterial pressure drop during induction did not reach statistical signi®cance but the fall of systolic arterial pressure was highly signi®cant between groups, as was the number of patients needing vasopressors during induction. Despite the use of similar ¯uid volumes during surgery, intraoperative urine output was signi®cantly greater in the ¯uid group. Thiopental dose, duration of surgery and blood loss were similar in both groups. Table 6 Weight loss and intraoperative variables. Values are median (range). ²P<0.01 (Mann±Whitney); **P<0.01 (Chi squared)

Weight change (kg) Systolic arterial pressure (mm Hg) Mean arterial pressure (mm Hg) Number given vasopressor Volume intraoperative ¯uids (litre) Urine output (ml)

Control (n=25)

Fluid group (n=25)

±1.5 (±3.2 to 2.5) ² ±44 (±87 to ±17) ² ±22 (±54 to 3) ns 10 ** 4 (1.5±7.3) ns ² 150 (0±600)

0 (±1.5 to 1) ±31 (±80 to 10) ±15 (±49 to 15) 0 3.5 (2.1±7.3) 312 (90±870)

Our study con®rms that Picolaxq causes preoperative dehydration, which exaggerates the hypotensive effects of anaesthetic induction agents. This can be offset by the administration of preoperative crystalloid, which also improves perioperative urine production. Administration of preoperative intravenous ¯uids could be of bene®t to patients receiving Picolaxq bowel preparation.

It is common practice at St Andrew's Centre to perform axillary brachial plexus blocks in patients undergoing hand surgery. We employ a transarterial approach with a two-point injection technique. In an adult we use a mixture of 30 ml of 1% lidocaine and 20 ml of 0.5% bupivacaine with epinephrine 1:200 000. The maximum safe dose of lidocaine with epinephrine is 7 mg kg±1 and that for bupivacaine1 with or without epinephrine is 2 mg kg±1. However, there are few references in the literature to the maximum safe dose of a mixture of local anaesthetics.2 We would like to introduce the concept of a Theoretical Toxic Dose Index (TTDI). It is the dose of lidocaine (mg kg±1) divided by 7 plus the dose of bupivacaine (mg kg±1) divided by 2. A safe index would be <1. The data of 571 patients undergoing brachial plexus block were analysed. Any complications were recorded. The TTDI were calculated and correlated with rate of complications attributable to local anaesthetics. The results are shown in Table 7. Our overall success rate for the block was 87% for all grades of anaesthetists, rising to 95% with local anaesthetic supplementation. When the TTDI data were categorized (0±0.5; 0.5±1.0; 1.0±1.5; 1.5±2.0) it was found that the proportion of complication increased (linear regression, n=4, P=0.06) with a high predictability (r2=0.88). Table 7 Relationship of TTDI with success and complication rates TTDI

Total

Success rate (%)

Success with LA

Complication

Complication rate (%)

<1 1.0±1.5 >1.5 Total

83 436 52 571

82 88 90 87

91 97 96 95

1 6 1 8

1.2 1.4 1.9 1.4

We conclude that in our series of patients our current practice is associated with a good success rate and low number of complications. The complications were minor and occurred at the time of injection. It would be prudent to obtain more data on the relationship between TTDI and complication. Keywords: anaesthetics local; anaesthetic techniques, regional, brachial plexus

References

1 Pinnock C, Lin T, Smith T. Fundamentals of Anaesthesia. London: Greenwich Medical Media Ltd., 1999 2 De Jong RH, Bonin JD. Anesthesiology 1981; 54: 177±81

Keywords: ¯uids, replacement

References

1 Barker P, Hanning C, Trotter T. Ann R Coll Surg Engl 1992; 74: 318±19 2 Sanders G, Mercer S, Saeb-Parsey, et al. Br J Surg 2001; 88: 1363±5

Evaluation of personal, environmental and biological exposure of paediatric anaesthetists to sevo¯urane N. Raj*, K. A. Henderson*, J. E. Hall*, B. Williams* and M. Harmer

Correlation between theoretical toxic dose of a mixture of local anaesthetics and complication rate in patients undergoing brachial plexus block: a prospective analysis of 571 cases

Department of Anaesthetics and Intensive Care Medicine, University of Wales College of Medicine, Cardiff CF14 4XN, UK

R. V. Martin*, P. Smith* and S. K. Pal St Andrew's Centre for Plastic Surgery and Burns, Broom®eld Hospital, Chelmsford, Essex CM1 7ET, UK

Paediatric anaesthetists have a higher degree of exposure than adult anaesthetists to waste anaesthetic gases especially during induction1 and in non-theatre venues.2 This study investigated the acute exposure of paediatric anaesthetists in theatre and other venues to sevo¯urane and any association between the exposure and anaesthetic technique. Five lists each in four venues were targeted: MRI Suite, Day Surgery Unit, Oncology Unit and Dental

675P

Proceedings of the Anaesthetic Research Society

Theatre. Blood samples were collected at baseline and immediately following each induction throughout each list. All samples were measured using headspace analysis by gas chromatography and the concentration in blood calculated. A note was made of type of induction, airway management, circuit used and presence or absence of scavenging. The exposure indices for blood sevo¯urane were generally very low in all venues and did not exceed the Time Weighted Average limit of 20 ppm set by Abbot. Unpaired t-test with Bonferroni correction was applied to determine if there was any signi®cant relationship between blood sevo¯urane levels and the type of induction technique, airway control, breathing system or scavenging used (Table 8). Induction type and airway control made no difference to blood sevo¯urane level, whereas predictably, breathing system type and presence of scavenging did have a signi®cant effect (P<0.001). Acknowledgement: funding provided by Anaesthetists of Great Britain and Ireland.

Association

of

Keywords: anaesthetics volatile, sevo¯urane; anaesthetics volatile, atmospheric pollution

References

1 Henderson KA, Matthews IP. Eur J Anaesthesiol 2000; 17: 149±51 2 Hoerauf KH, Wallner T, Akca O, et al. Anesth Analg 1999; 88: 925±9

Table 9 Results are given as numbers (out of 10) or mean distances (mm). LOS=line of sight Position External rotation

Direct LOS to sciatic n. Direct LOS through fem. art Direct LOS through fem n. Distance from skin to sciatic n. Distance from LOS to fem. art Distance from LOS to fem n.

Neutral position

Internal rotation

Left

Right

Left

Right

Left

Right

1

0

6

6

9

8

3

0

3

1

1

0

4

4

1

0

1

1

106

108

107

106

109

109

7

10

10

11

13

14

5

5

5

4

8

8

The mean age was 30 (range 21±44) yr and the mean body mass index was 24 (19.5±28). Other results are shown in Table 9. These results con®rm that, as the thigh is moved from an externally to an internally rotated position, the sciatic nerve becomes more accessible by the anterior approach, and the risk of femoral artery or nerve puncture is reduced, but not eliminated. Acknowledgement: CSM was supported by AstraZeneca Pharmaceuticals.

Thigh rotation and the anterior approach to the sciatic nerve; an MRI study

Keywords: anaesthetic techniques, regional, sciatic

C. S. Moore, J. A. W. Wildsmith and D. E. Sheppard* University Department of Anaesthesia, Ninewells Hospital, Dundee, UK

The anterior approach to the sciatic nerve block may be associated with a high failure rate because the nerve lies posterior to the lesser trochanter of the femur at the level of needle insertion.1 However, previous work using cadavers2 3 demonstrated that internal rotation of the leg renders the nerve more accessible to the anterior approach. We wished to examine this relationship in vivo. After ethics committee approval, ®ve male and ®ve female volunteers consented to undergo magnetic resonance imaging. A vitamin E marker was placed on the surface where a needle would have been inserted for an anterior approach to the sciatic nerve.4 Three scans were then performed, one with both legs in the neutral position, the next with maximal bilateral internal rotation at the hip and the third with maximal bilateral external rotation at the hip. The scans were then examined with a consultant radiologist. The line of sight from the insertion point to the sciatic nerve was examined, and the distance from the skin to the nerve measured, as were the distances from the line of sight to both the femoral artery and nerve.

References

1 Charlton JE, Nicholls BJ, White E. Br J Anaesth 1987; 59: 127P 2 Hadzic A, Reiss W, Dilberovic F, et al. Reg Anesth Pain Med 2001; 23 (May±June supplement): 38 3 Vloka JD, Hadzic A, April E, et al. Anesth Analg 2001; 92: 460±2 4 Wildsmith JAW, Armitage EN (eds) Principles & Practice of Regional Anaesthesia, 2nd edn, 1993; 195±7

TNFa cleaving enzyme (TACE) activity assessed by TIMP-3 binding H. J. Robertshaw* and F. M. Brennan* Kennedy Institute of Rheumatology Division, Imperial College School of Medicine, 1 Aspenlea Road, London W6 8LH, UK

Circulating soluble tumour necrosis factor a (TNFa) is formed by the cleavage of membrane-bound TNFa by a membrane-bound metalloproteinase (TNFa cleaving enzyme, TACE).1 The natural inhibitor of TACE, tissue inhibitor of metalloproteinase-3 (TIMP-

Table 8 Blood sevo¯urane levels expressed by induction technique for all staff at every induction, for each venue and percentage of technique used at each site. Mean (SD) ppm for total cases and mean [range] ppm per list. IV=intravenous; IH=inhalational induction; LMA²=Laryngeal Mask Airway; ETT=tracheal tube; Other breathing system=circle or scavenged Baine. *P<0.001. (²LMA is the property of Intavent Limited) Blood sevo¯urane (all staff) (ppm) Induction (IV:IH) Airway control (Mask:LMA/ETT) System used (T piece:other) Scavenging (present:absent) Blood sevo¯urane each venue

3 (3): 3 (3) 3 (3): 3 (3) 5 (3): 1.5 (3)* 1.5 (3): 4.5 (3)*

Percentage technique used MRI suite (n=5)

Day surgery (n=5)

Oncology (n=5)

Dental (n=5)

37%: 63% 0%: 100% 100%:0% 0%: 100% 4.6 [1±10]

47%: 53% 42%: 58% 64%: 36% 36%: 64% 4.2 [0±10]

45%: 55% 100%:0% 11%: 89% 89%: 11% 2.8 [0±13]

94%: 6% 0%: 100% 0%: 100% 100%: 0% 1.4 [0±13]

676P

Proceedings of the Anaesthetic Research Society

3), has been shown to bind selectively to active TACE on the cell surface and the inhibitory activity of TIMP-3 lies within the Nterminus.2 We used these properties of TIMP-3 to develop an assay for TACE catalytic activity using ¯ow cytometry and so examine TACE activity following cell stimulation. Elutriated human monocytes (13106 cells well±1) were stimulated with lipopolysaccharide (LPS 10 ng ml±1) and harvested at time points shown. At harvest, in one group TACE surface protein expression was measured by ¯ow cytometry (using optimized antibody combination). To the other group, histidinelabelled N-TIMP-3 was added and detected by ¯ow cytometry using anti-histidine antibodies. Flow cytometry data are expressed as mean ¯uorescence intensity (MFI) 6SD from seven experiments (10 000 cells in each experiment). Upon LPS stimulation, TACE was rapidly lost from the cell surface (Fig. 3A), con®rming previous work which examined TACE using confocal microscopy.3 This protein down-regulation at the cell surface coincides with maximal catalytic activity (Fig. 3B). These ®ndings support the hypothesis that inactive TACE is expressed on the cell surface and then rapidly removed from the cell surface following cell activation. Acknowledgement: This work was funded by a grant from the Arthritis Research Council, UK. Keywords: enzymes, tumour necrosis factor a cleaving enzyme

References

1 Black RA, Rauch CT, Kozlosky CJ, et al. Nature 1997: 385; 729±3 2 Amour A, Slocombe PM, Webster A, et al. FEBS Lett 1998: 435; 39±44 3 Doedens JR, Black RA. J Biol Chem 2000: 275: 14598±607

contrast, transthoracic impedance changes are extremely sensitive to patient movement, demonstrate slow response times and do not recognize apnoea rapidly enough. Human observation of large numbers of patients is a near impossible task. In 1999 a new simple respiratory rate monitor based on the pyroelectric polymer sensor (PEP) was described and evaluated.1 It was concluded that, in the laboratory, the PEP sensor performed well and appeared to have many desirable features. If these advantages were maintained in a clinical setting such a system would compare favourably with existing methods and may be valuable in cases of acute illness, major injury and perioperative care. We received on loan a prototype of the PIPPA Breathing Monitor. After approval of the local ethics committee and the medical devices agency, we chose to study four groups of patients as follows: a control group, patients receiving supplemental oxygen therapy, patients receiving nebulized medication and patients receiving humidi®ed oxygen treatment. The measurements made by the PIPPA were compared with those obtained simultaneously by the use of transthoracic impedance from the ECG monitor and human observation by nurses as part of their regular monitoring. The target number of patients in each group is 50. The results are only being presented for the control group and the nebulized medication group as the others are not yet complete. Results are summarized in Table 10. The PIPPA monitor correlates well with both human observation and transthoracic impedance. Keywords: monitoring, ventilation

Reference

1 Dodds D, Purdy J, Moulton C. The PEP transducer: A new way of measuring respiratory rate in the non-intubated patient. J Acc Emerg Med 1999; 16: 26±8

Respiratory monitoring of non-intubated patients using the `PIPPA' breathing monitor D. Sankar* and K. J. Kini* Rochdale Health Care NHS Trust, Rochdale In®rmary, Rochdale, UK

Increasingly, with the development of newer protocols in clinical practice, there is a need for monitoring respiration. The most commonly used technique for this purpose is human observation. Capnography and changes in transthoracic impedance are potential alternatives but have not found widespread acceptance. With capnography a nasal cannula is required, the equipment is expensive and cumbersome and regular calibration is needed. In

Table 10 Coef®cient of correlation between PIPPA and the ECG monitor (r1), between PIPPA and the nurse observer (r2) and between ECG monitor and nurse observer (r3) Group

Number of patients

Number of observations

r1

Control Nebulized medication

50 50

100 241

0.73 0.84

Fig 3 Time course of TACE protein (A) and TACE activity (B) in LPS-stimulated monocytes.

677P

r2 0.91 0.90

r3 0.75 0.85

Proceedings of the Anaesthetic Research Society

Actigraphic measurements of sleep variables and circadian rhythm in chronic pain patients compared with healthy controls

Change in snoring as a predictive factor for pre-eclampsia

M. Schupp*, N. A. Hall*, P. Sharpe and D. J. Rowbotham

University Department of Anaesthesia, Critical Care and Pain Management, University Hospitals of Leicester, Leicester Royal In®rmary, Leicester LE1 5WW, UK

M. Dolling*, P. Sharpe and C. D. Hanning

University of Leicester and University Hospitals of Leicester NHS Trust, Leicester, UK

Sleep disturbance is a signi®cant problem for chronic pain patients. Furthermore insomnia may indicate severity of chronic pain symptoms and psychological disturbance.1 Actigraphy is the study of sleep via movement using a wristwatch containing a small accelerometer and data logger which is worn on the patient's non-dominant arm. Frequency and amplitude of movement are logged at 1-min intervals throughout the observation period. Data are processed using Actiwatch Sleepwatch software (Cambridge Neurotechnologyq). It may be worn for several weeks at a time. It monitors sleep in the home environment, reducing the impact of `®rst-night' effects seen in laboratory assessments of sleep.2 We used a combination of sleep diaries and activity monitoring (actigraphy) to assess the presence and magnitude of sleep disturbance in the patient groups. We studied 21 adult subjects, eight volunteers without any symptoms of chronic pain and 13 patients from the pain clinic. Following written informed consent they were asked to wear the monitor for a 2-week period and complete a daily sleep diary. The watch was well tolerated in all groups. There were no differences in actual sleep time, sleep ef®ciency or sleep bouts. Fragmentation index score was signi®cantly higher in the chronic pain group (P<0.001). Measures of circadian rhythm such as intradaily stability, intradaily variability and relative amplitude were measured with conventional sleep variables and were similar between groups. Data were analysed using SPSS v11.0. Sleep variables were analysed with Student's t-test and circadian rhythm with Mann±Whitney U test. The fragmentation index is the addition of the percentage time spent moving and the percentage of immobility phases of 1 min, this is used as an indicator of restlessness. The fragmentation index scores were signi®cantly higher in the pain group whilst other sleep variables remained the same (Table 11). It is possible that this index may relate to quality of sleep; however, more work is required to validate this thoroughly. Keywords: pain, chronic; sleep

References

1 Wittg RM, Zorick FJ, Blumer D, et al. J Nerv Ment Dis 1982; 170: 429± 31 2 Jean LG, von Gizycki H, Zizi F, et al. Precept Mot Skills 1997; 85: 219±16

The amount and quality of sleep changes with pregnancy.1 There is growing evidence of a link between snoring, sleep disordered breathing and pre-eclampsia.2 Pre-eclampsia is a major cause of maternal and neonatal morbidity and mortality.3 Patients were approached within 24 h of delivery and asked to complete a simple questionnaire, with the aid of a research midwife. The questionnaire was used to record demographic data (including date of birth, ethnicity, smoking history, alcohol and illicit drug consumption), previous medical and relevant family history, obstetric history, and sleep history (focusing on snoring, change in snoring, witnessed apnoeas, Epworth sleepiness score) and clinical data from case records. Data were collected from 609 patients of whom 53 had proteinuric hypertension and were treated by the obstetricians as suffering from pre-eclampsia. Several sleep-related variables were signi®cantly different in the pre-eclamptic group (Table 12). Logistic regression analysis was employed and demonstrated a highly signi®cant correlation between worsening snoring during pregnancy and pre-eclampsia. A prospective study is required to demonstrate the presence of upper airways resistance syndrome in this group of patients and subsequently measure any bene®cial effects of continuous positive airway pressure therapy. Table 12 Incidence of snoring in different groups Proteinuric Normal Hypertension hypertension P value New snoring in pregnancy No Sometimes Often Nightly Snoring worse during pregnancy Yes No Epworth sleepiness score 9 or less 10 or more

188 208 97 44

5 4 2 5

11 17 12 13

<0.001

116 168

5 4

25 13

0.01

344 176

9 7

24 27

0.02

Keywords: complications, pre-eclampsia; sleep

References

1 Feinsilver SH, Hertz G. Clin Chest Med 1992; 13: 637±44 2 Edwards N. Am Family Physician 1993; 60: 1792 3 Department of Health. Why Mothers Die. Report on con®dential enquiries into maternal deaths in the United Kingdom, 1994±6

Table 11 Actual sleep time (min), actual sleep (%), sleep bouts and fragmentation index (mean (SD))

AST (min) AS (%) SB FI

Control

Pain

387.8 90.69 19.55 25.89

386.0 81.10 29.72 59.21

(60.36) (4.84) (8.44) (8.87)

P (Student's) (100.25) (18.05) (20.26) (21.26)

0.96 0.16 0.19 0.001

Effect of remifentanil on postoperative pain in patients having fentanyl±propofol anaesthesia for cardiac surgery K. Rauf*, A. Vohra, P. Fernandez-Jimenez*, M. Forrest* and N. O'Keeffe* Department of Anaesthesia, Manchester Royal In®rmary, Manchester, UK

Remifentanil is an ultra-short acting opioid. During anaesthesia it provides intense analgesia and sympatholysis that is ideal for

678P

Proceedings of the Anaesthetic Research Society

cardiac surgery. Postoperative agitation and increased analgesic requirement after remifentanil infusion may be a problem. Studies in non-cardiac surgery have shown mixed results.1 2 This prospective double-blind randomized study assesses the effect of remifentanil on postoperative recovery expressed as increased analgesic requirement after off-pump coronary artery bypass surgery (OPCAB). Twenty patients undergoing elective OPCAB were randomly assigned to receive infusions of remifentanil (50 mg ml±1 at 0.12 ml kg±1 h±1 = 0.1 mg kg±1 min±1) (group R, n=10); or saline (at 0.12 ml kg±1 h±1) (group S, n=10). Anaesthesia was standardized with target controlled propofol infusion (1.8± 4 mg ml±1) and fentanyl (15 mg kg±1). After surgery, propofol (1.5±3 mg kg±1 h±1) and the study solution were continued until the patient was ready to be woken up (warm and cardiovascularly stable). Morphine infusion (0.015±0.02 mg kg±1) was also started immediately after operation with additional boluses (1±2 mg) as assessed by pain scores (VAS 0±10). Morphine requirements were noted as total morphine before stopping sedation (Mpre), morphine in the ®rst hour after stopping sedation (M1) and total morphine in 12 postoperative hours (M12). Patient data including height, weight, duration of surgery (op time) and of sedation (sedn time) were recorded. Student's t-test and Mann±Whitney U test were used for data analysis (*P<0.05 signi®cance level). In the ®rst hour after discontinuing sedation, morphine requirement was higher in group R as compared to group S, but similar in the period till sedation was stopped and in 12 h after surgery. There was no difference in pain scores (median (range) R=5 (2±9), S=3 (0±7)). Table 13 shows mean (SD). We conclude that remifentanil infusion is associated with an increased requirement of morphine in the ®rst hour after discontinuation, in spite of preloading by fentanyl. Keywords: analgesics opioid, remifentanil; surgery, cardiovascular

References

1 Guignard B, Bossard A, Coste C, et al. Anesthesiology 2000; 93: 409±17 2 Cortinez I, Brandes V, et al. Br J Anaesth 2001; 87: 866±9

N. Patel*, R. Fernando and V. Sodhi Department of Anaesthesia, Royal Free Hospital, London NW3 2QG, UK

Although the urgency of Caesarean section (CS) has recently been classi®ed,1 its relationship with fetal outcome is unclear. Our aim was to identify any association between the grade of urgency and markers of fetal condition at birth such as Apgar scores and umbilical cord arterial (UA) and venous (Uv) blood gases. The in¯uence of the time interval taken to establish regional anaesthesia (RA) as well as surgical delivery times were also assessed. This prospective observational study was conducted in our unit over a 4-month period, collecting data from 203 consecutive CSs performed under regional anaesthesia. Data included: grade of CS (1, Emergency; 2, Urgent; 3, Scheduled; 4, Elective), type of RA technique (spinal or epidural top up of a pre-existing labour epidural), RA injection to T4 sensory block time, injection to knife-to-skin (KTS) time, KTS to delivery time, uterine incision (UI) to delivery time, Apgar scores at 1 and 5 min, umbilical cord pH, [H+] and base excess. Statistical analysis included a modi®ed Levene equal variance test, analysis of covariance (ANCOVA) and linear trend analysis using a general linear model (P<0.05). RA technique was analysed on an intention to treat basis. Age, height, weight, gestation and parity did not differ between groups. There were two conversions to general anaesthesia after failed RA: one pre-delivery and one postdelivery. Apgar score at 1 min improved signi®cantly, as did Uv pH and [H+] with reducing urgency of CS (1 to 4) (Table 14). A corresponding trend for UA pH and [H+] was not identi®ed. The signi®cantly longer times to T4 block and KTS with reducing CS urgency were not confounding variables in terms of fetal outcome measures. Further veri®cation with larger patient numbers is required. Acknowledgement: Obstetric Anaesthetists' Association funded research fellowship (NP).

Table 13 Morphine requirement in the two groups

Group R Group S

Urgency of Caesarean section, time to delivery and fetal outcome: a prospective observational study

Op time (min)

Sed time (min)

Mpre (mg)

M1 (mg)

M12 (mg)

Keywords: anaesthesia, obstetric

196 (53) 198 (35)

264 (97) 226 (91)

9.3 (2.6) 8.7 (3.3)

8.15 (3.6)* 3.29 (2.3)*

27.1 (8.3) 24.0 (6.3)

Reference

1 Lucas DN, Yentis SM, Kinsella SM, et al. J R Soc Med 2000; 93: 346±50

Table 14 Data are number (n), mean (SD) or median [IQR]; *P<0.001, **P<0.0001 (linear trend)

Spinal/Epidural Topup (n) Injection to T4 block (min) Injection to KTS (min) KTS to delivery (min) UI to delivery (min) Apgar: 1 min Apgar: 5 min UA pH UA [H+] (nmol litre±1) UA Base Excess (mmol litre±1) Uv pH Uv [H+] (nmol litre±1) Uv base excess (mmol litre±1)

Grade 1 (n=18)

Grade 2 (n=83)

Grade 3 (n=19)

Grade 4 (n=83)

9/9 7.50 (2.31) 12.17 (6.40) 7.50 (4.73) 3.12 (3.66) 8.5 [5.25±9] 10 [9.25±10] 7.22 (0.09) 62.10 (13.76) ±8.00 (4.53) 7.26 (0.09) 56.40 (13.35) ±7.49 (3.47)

28/55 12.68 (6.95) 21.55 (10.39) 9.14 (6.37) 2.23 (1.86) 9 [8±9] 10 [10±10] 7.25 (0.06) 56.49 (8.66) ±6.98 (3.81) 7.30 (0.06) 50.31 (6.71) ±7.38 (2.99)

14/5 10.89 (5.92) 22.21 (9.10) 11.21 (6.48) 2.50 (1.47) 9 [8±9] 10 [9.75±10] 7.27 (0.05) 54.51 (6.72) ±4.17 (2.33) 7.31 (0.04) 48.21 (4.48) ±5.99 (2.29)

80/3 12.16 (6.56)* 26.30 (10.27)** 10.13 (7.68) 2.01 (1.20) 9 [9±9]** 10 [10±10] 7.27 (0.06) 54.47 (9.03) ±5.73 (3.05) 7.32 (0.04)** 48.14 (4.92)** ±5.56 (2.17)

679P

Proceedings of the Anaesthetic Research Society

methods. These ®gures were then used to calculate the limits of agreement between ethyl chloride and Neuropen (Fig. 4). The Neuropen provides a means of de®ning height of a sensory block within three dermatomes of ethyl chloride.

POSTER PRESENTATIONS The effectiveness of the Neuropen in predicting the dermatomal limits of sensory blockade in regional anaesthesia for Caesarean section as compared with ethyl chloride

Keywords: sensory block, anaesthetic techniques, regional

References

1 Aitkenhead AR, Smith G. Textbook of Anaesthesia, 3rd Edn. Edinburgh: Churchill Livingstone, 1996; 297 2 Greene NM. Anesthesiology 1958; 19: 45±50

M. Dolling*, N. Roberts* and P. Sharpe University Department of Anaesthesia, Critical Care and Pain Management, University Hospitals of Leicester, Leicester Royal In®rmary, Leicester LE1 5WW, UK

Ethyl chloride is commonly used to de®ne the limits of sensory block for regional anaesthesia. This volatile agent is a chlorocarbon that is both explosive and an atmospheric pollutant.1 Other methods have been tested for assessing sensory blockade but all possess their own limitations. It has been demonstrated that when using a needle the recorded dermatomal level can be about two dermatomes below that of ethyl chloride.2 In this study we have tested a device (the Neuropen) that is employed in medical practice as a means of assessing sensory neuropathy. It works by applying a constant (40 g) pressure via a blunt sprung disposable stylet, applied to the skin in a dabbing motion in the same way a needle might be used. Forty patients were recruited who were booked for elective Caesarean section. There was one participant who withdrew. Each patient received a standardized combined spinal±epidural anaesthetic. Two investigators then independently recorded the level of the block in a caudal to cephalic direction and then cephalic to caudal, one using the Neuropen and one ethyl chloride spray. These recordings were made at 15 min and 1.5 h after insertion of the spinal component. The data were analysed using a Bland Altman plot. Mean and standard deviations were calculated for the differences between

Recording postoperative pain in ambulatory surgery: further evaluation of a revised electronic data collection device P. Watt*, J. Williams*, A. Roscoe*, E. Pallett*, P. Maslowski*, J. P. Thompson* and D. Rowbotham University Department of Anaesthesia, Critical Care & Pain Management, Leicester Royal In®rmary, Leicester LE1 5WW, UK

The Visual Analogue Scale (VAS) is widely used in the assessment of postoperative pain. However, it may generate data storage problems and errors when data are transferred to a computer for analysis. We have re®ned a previous automatic data collection device to incorporate a slider and digital display (the EPIC slide). This device records VAS scores on a 0±100 mm scale and the data are automatically tagged with time and date. In addition, it can ask and store responses to other questions. Forty patients aged 18±65 yr undergoing ambulatory surgery were enrolled. The investigators made three postoperative visits at 0.5, 1 and 2 h. At each visit patients were asked to perform four VAS pain scores, two using the new EPIC slide device and two using a card VAS slider. Between readings a distraction question was asked regarding the patients preferred method of measurement, plus a verbal assessment of pain. The patient was blinded to

Fig 4 Limits of agreement (LA) using Bland Altman plot for the difference between ethyl chloride and Neuropen.

680P

Proceedings of the Anaesthetic Research Society

Fig 5 Limits of agreement using Bland Altman plot of mean VAS scores detremined using a VAS card or an EPIC slide device.

exact measurements of all readings. Data were analyzed using a Bland Altman plot (Fig. 5). Mean and standard deviations were calculated for the differences in VAS pain scores of the two measurement methods and for repeatability within each method. Analysis showed good agreement between the methods with a mean difference of 1.98 mm, limits of agreement 11.13 to ±7.2 mm (95% CI ±10.8 to 14.2 mm) (Fig. 5). The EPIC slide system was generally well received by patients, with two thirds expressing a preference for this method of data collection. The remaining one third of patients expressed no preference and no one expressed a preference for the card VAS. Keywords: pain, postoperative; surgery, day-case

References

1 DeLoach LJ, Higgins MS, Caplan AB, et al. Anesth Analg 1998; 86: 102±6 2 Anderson I, Dolling M, Pallet E, et al. Br J Anaesth 2002; 88: 328P

681P