Proceedings of the Anaesthetic Research Society Meeting

British Journal of Anaesthesia 93 (4): 604–18P (2004)

doi:10.1093/bja/aeh495

ABSTRACTS Proceedings of the Anaesthetic Research Society Meeting Liverpool Medical Institution, Liverpool July 8–9, 2004 (The name of the author presenting the paper is shown in bold type. *Indicates non-member. All authors have certified that, where appropriate, studies have been conducted with the approval of the relevant Human Ethics Committee or Animal Experimental Review Committee.)

Table 1 ‘Intervention’: inotropes, vasodilators, diuretics CI ‘normal’ CI low

Continuous monitoring of the cardiac output, systemic vascular resistance, and extravascular lung water in patients undergoing cardiopulmonary bypass B. J. Donnelly*, P. G. Bedford*, P. C. Braidley* and J. J. Ross Chesterman Unit, Northern General Hospital, Herries Road, Sheffield S5 7AU, UK

Clinicians treating patients in the perioperative phase of cardiac surgery require real-time data on alterations of the cardiovascular and respiratory systems’ status. Cardiopulmonary bypass can induce myocardial stunning, low systemic vascular resistance syndrome, pulmonary capillary leak, and pulmonary oedema, which may be difficult to identify clinically. Cardiac preload, cardiac output, systemic vascular resistance, heart rate, and ventilatory support are manipulated for a successful surgical outcome. Devices commercially available to provide parts of these data include TOE, PAFC, and arterial waveform analysers. We assessed the clinical usefulness of the Pulsion Medical Systems Pulse Invasive Continuous Cardiac Output (PiCCO) equipment, which gives beat-to-beat cardiac output, extravascular lung water, intrathoracic blood volume, systemic vascular resistance in addition to the peripheral arterial pressure, giving, in one device, most clinically relevant information in these patients.1 2 We obtained Ethics Committee approval, then informed, written consent in 40 elective cardiac surgery patients undergoing cardiopulmonary bypass. The PiCCO arterial line was placed in one femoral artery before induction of anaesthesia, in addition to all other clinical invasive and non-invasive monitoring deemed necessary by the anaesthetist conducting the case. The anaesthetist was blinded to the PiCCO data. There were no standardized anaesthetic, surgical, ‘bypass nor postoperative care protocols’. We analysed retrospectively the PiCCO data for the first 24 h care against clinical progress in each case, to determine whether PiCCO data would have influenced the clinical management had it been available. Clinical progress was determined from the operative notes, anaesthetic charts, postoperative ICU charts, and nursing records. Cardiac index (<2 litre min
1 m
2), SVRI (<1500 cm dyne
5 s
2), EVLWImorethan7mlkg
1,PaO2/FI 2 ratioslessthan25weredeemed clinically relevant. Recognition of such values was termed an ‘event’ as these abnormal values should trigger an ‘intervention’. We compared clinical progress with PiCCO data, and assessed whether the PiCCO ‘events’ reflected accurately clinical progress. O

#

SVRI high 11 SVRI low 37 (event) SVRI normal 166

PaO2/F IO2 >25

PaO2/F IO2 <25

30 (event) EVLWI >7 83 (event) 17 (event) 0 EVLWI <7 82 32 (event) 3 (event)

The results are given in Table 1. The number of events: 202, 70 cardiac (42 treated and 28 not deemed clinically relevant) and 132 respiratory (32 ‘events’ were hypoxaemic with a raised EVLWI). The PiCCO system in these 40 patients gave 100 (76%) clinically irrelevant respiratory data, and 28 (40%) clinically irrelevant cardiac data during their perioperative care. Keywords: heart, cardiac output; monitoring, PiCCO surgery, cardiopulmonary bypass

References 1 Sakka SG, Bredle DL, Reinhart K, et al. J Crit Care1999; 14: 78–83 2 Enberg PR, Hansbrough JR, Anderson D, et al. Am Rev Resp Dis1987; 136: 662–74

Responses of the porcine isolated coronary artery; influence of temperature of incubation and presence of antibiotics J. X. Wei , V. G. Wilson1 and R. P. Mahajan University Department of Anesthesia and Intensive Care, 1 School of Biomedical Sciences, Queen’s Medical Centre, Nottingham NG7 2UH, UK

For in vitro studies of vascular responses in pre-incubated blood vessels, some workers routinely include antibiotics in the incubation medium,1 while others do not.2 During preliminary experiments on the porcine isolated coronary artery (PCA) we noted that following overnight storage at 37 C (but not at 4 C) the incubation medium was visibly cloudy possibly as a result of microbial contamination. Therefore, we examined the effects of temperature of incubation and presence of antibiotics on the vascular responsiveness of PCA. Paired 5 mm segments of the PCA were dissected from pig hearts and incubated in Krebs Henseleit (K-H) solution at 4 or 37 C. After 16–18 h the segments were prepared for isometric tension recording.3 A sub-maximally effective concentration of the thromboxane mimetic U46619 was used to induce a sustained contraction, followed by exposure to 10 nM Substance P to assess the integrity of the endothelium. The segments were then exposed to cumulatively

The Board of Management and Trustees of the British Journal of Anaesthesia 2004

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

VERBAL PRESENTATIONS

Proceedings of the Anaesthetic Research Society

Table 2
Log EC50 values for KCl and U46619 following overnight incubation under various conditions. Mean (SEM). *P<0.05 Temperature KCl

U46619 Dexameth L-NAME

4 C 37 C 37 C+AB

1.38 (0.03) 8.13 (0.07) 8.17 (0.09) 8.32 (0.10) 1.39 (0.02) 7.85 (0.07)* 7.54 (0.11) 8.12 (0.13) 8.17 (0.05) 7.89 (0.08)*

Fig 1 Mean (95% CI) peak inspiratory pressure.

respiratory failure related to bronchospasm.2 However, there are no data on the effects of using a helium–oxygen mixture during HFJV in upper airway obstruction. HFJV was administered using a Bromsgrove Humidified Jet Ventilator (Penlon Ltd.) at a frequency of 150 min
1 (driving pressure 28 PSI, inspiratory time 30%) to a modified trachea–lung model1 to simulate ventilation through varying degrees of fixed laryngotracheal stenosis (2.5–8.5 mm). HFJV was applied from above, through and below the level of stenosis to simulate supraglottic, transglottic, and infraglottic administration. Measurements were repeated via each route at steady state for each stenosis diameter using both 100% oxygen and heliox (O2 50%, He 50%). The gas mixture was delivered from a Sechrist Model 3500HL gas mixer calibrated for helium–oxygen. Oxygen concentration was determined using a Datex-Ohmeda S/5 anaesthetic gas analyser module. Data are presented as mean (95% CI) (Fig. 1). Peak inspiratory pressure generated via the transglottic route at 2.5 and 3.0 mm stenosis exceeded the ventilator cut-off but otherwise peak, mean, and end-expiratory pressures for the heliox and oxygen 100% mixtures were very similar. The data suggest that using a helium–oxygen 50% mixture during HFJV in the presence of airway stenosis may have little advantage in reducing generated airway pressures.

Keywords: arteries, porcine coronary; heart, vascular reactivity NG-nitro-L-arginine methyl ester (L-NAME)

Keywords: airway, stenosis; gas, helium; ventilation, high frequency

References

References

1 O’Brien AJ, Wilson AJ, Sibbald R, et al. Br J Pharmacol 2001; 133: 351–60 2 Mathewson AM, McPhaden AR, Wadsworth RM. J Immunol Methods 2003; 279: 163–71 3 Lawrence RN, Dunn WR, Wilson VG. J Pharm Pharmacol 1998; 50: 885–90

1 Ng A, Russell W, Harvey N, Thompson JP. Anesth Analg 2002; 95: 764–9 2 Tobias JD, Grueber RE. Anaesthesia 1999; 9: 451

Further studies on the acute effect of bacterial ‘quorum-sensing’ molecules on the porcine isolated coronary artery

Effects of helium on high frequency jet ventilation in a model of airway stenosis P. W. Buczkowski*, F. N. Fombon , W. C. Russell and J. P. Thompson

J. Richards*, J. Mok*, D. Pritchard1*, R. Chaabra1* and V. G. Wilson*

University Department of Cardiovascular Sciences, Division of Anaesthesia Critical Care and Pain Management, Leicester Royal Infirmary, Leicester, UK

The School of Biomedical Sciences and 1The School of Pharmaceutical Sciences, University of Nottingham, Nottingham, UK

We have investigated previously the effect of different routes of administration of high frequency jet ventilation (HFJV) in a benchtop model of airway stenosis.1 In clinical practice, the addition of helium to the inspired gas may facilitate ventilation in the presence of airway stenosis and has been used to treat a child with progressive

Bacteria-derived molecules (e.g. LPS) are well known to exert effects on the immune and cardiovascular system. We have recently described the ability of N-3-oxo-dodecanoyl homoserine lactone (OdDHL), a quorum-sensing molecule, to inhibit contractions of the pig coronary and pulmonary arteries.1 In the present study we

605P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

increasing concentrations of either U46619 or KCl and the responses expressed as percentage of their own maximum. For comparisons
log [EC50] was calculated and Student’s paired t-test was used. Paired experiments investigated the effects of co-incubation with an antibiotic mixture (60 mg ml
1 penicillin benzyl and 20 mg ml
1 streptomycin sulphate) or 10
6 M dexamethasone, or adding 100 mM NG-nitro-L-arginine methyl ester (L-NAME) before U46619. The K-H solution stored at 37 C was cloudy; dexamethasone had no visible effect. However, K-H solution maintained at 37 C in the presence of the antibiotic mixture was clear. The response to KCl (60 mM) of segments stored at 4 C was greater than that at 37 C (16.9 (1.1) vs 13.8 (0.8) g; P=0.004, n=32), but the sensitivity was unchanged (Table 2). In contrast, U46619 was 2-fold more potent in preparations stored at 4 C (n=12, Table 2); this difference in temperature-related sensitivity was not observed following co-incubation with dexamethasone (n=9) or addition of L-NAME (n=8) (Table 2). Co-incubation with antibiotics during overnight storage at 37 C significantly increased the sensitivity of preparations to U46619 (Table 2, n=8). Relaxations to Substance P (10 nM) were significantly greater in segments stored at 4 C compared with 37 C (63.6 (4.2) vs 40.2 (5.0)%, n=32, P<0.001); a difference not observed following co-incubation with dexamethasone at 4 and 37 C (63.5 (5.1) vs 56.2 (5.5)%, n=9, P=0.43). Finally, the antibiotic mixture improved relaxations to Substance P in vessels stored at 37 C (66.7 (4.7) vs 46.3 (7.3); P=0.005, n=16). We suggest that impairment in responses of PCA stored at 37 C, as compared with 4 C, may be partly as a result of unintentional microbial growth, co-incubation with antibiotics improved both vascular and endothelial responses. Similar effects by dexamethasone and L-NAME suggest the involvement of inducible nitric oxide synthase.

Proceedings of the Anaesthetic Research Society

Table 3
Log IC50 values for the effect of OdDHL on U46619-induced contractions. The values shown are the mean (SEM) of 8–12 separate observations Treatment Denuded Control prep. Treated prep.

L-NAME

Indomethacin TEA

Ap/IBTX

5.37 (0.11) 4.80 (0.07) 4.89 (0.11)

5.03 (0.08) 4.95 (0.12)

5.03 (0.10)* 5.06 (0.09)* 5.19 (0.13)*

4.77 (0.09)* 5.03 (0.11)

Keywords: arteries, porcine coronary; molecules, quorum-sensing

References 1 Lawrence RL, Dunn WR, Bycroft B, et al. Br J Pharmacol 1999; 128: 845–8 2 Sheih C-C, Coghlan M, Suliivan JP, et al. Pharmacol Rev 2000; 52: 557–93

Keywords: chemoattractant, fMLP; heart, vascular reactivity

References 1 Kerr SW, Yu R, Stearn CD, et al. J Pharmacol Exp Ther 1998; 287: 640–7 2 Keitou M, Kohzuki M, Katoh H, et al. J Mol Cell Cardiol 1997; 29: 881–94 3 Lawrence RN, Clelland C, Beggs D, et al. Br J Pharmacol 1998; 125: 1128–37

Preliminary experience with a system to score scientific merit in case reports

fMLP produces vasoactive effects in human and porcine blood vessels

A. Cherian and P. Charters Department of Anaesthesia, University Hospital Aintree, Liverpool, UK

R. Lawrence*, W. R. Dunn*, C. Clelland1* and V. G. Wilson School of Biomedical Sciences, University of Nottingham and 1 Department of Histopathology, City Hospital, Nottingham, UK

Bacteria-derived molecules are known to exert effects on the immune and cardiovascular system (e.g. LPS and pertussis toxin). The formylated tripeptide methionine-leucine-phenylalanine (fMLP) is an established chemoattractant for neutrophils during

Many authorities stress the need to maximize the scientific contribution in case reports but, despite this, guidelines1 seemed to us limited. We present a novel system to describe merit or otherwise in reporting cases and case series based on scoring for: noteworthiness of the event(s); context (extent to which generalization is feasible); practice points and a biophysical index (i.e. the relative

606P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

have investigated the role of the vascular endothelium and potassium channels in the vasodilator action of OdDHL. Segments of the porcine coronary artery were dissected from pig hearts and incubated overnight in Krebs Henseleit (K-H) solution at 4 C. The following day paired 5 mm length segments were prepared for isometric tension recording with or without an intact endothelium.1 A sub-maximally effective concentration of the thromboxane mimetic U46619 was used to induce a sustained contraction, followed by exposure to 10 nM Substance P to assess the integrity of the endothelium. The segments were again exposed to U46619 and the effect of cumulatively increasing concentrations of OdDHL determined. In some experiments, 100 mM NG-nitro-L-arginine methyl ester (L-NAME), 3 mM indomethacin, 2 mM tetraethylammonium (TEA), or a combination of 0.3 mM apamin and 0.1 mM iberiotoxin (Ap/IBTX) was added to one of the paired segment. The effect of OdDHL has been assessed on the basis of the negative logarithim of the concentration required to cause a 50% reduction in U46619-induced tone (pIC50) and comparisons made using a Student’s paired t-test (P<0.05 was considered statistically significant*). OdDHL (0.3–30 mM) caused a slow developing (15 min to equilibrium), concentration-dependent inhibition of U46619-induced tone, with the highest concentration causing greater than 90% inhibition. As shown in Table 3, removal of the vascular endothelium reduced the potency of OdDHL by 2-fold, while the presence of either 100 mM L-NAME or 3 mM indomethacin was associated with a 2-fold increase in potency. TEA (2 mM) caused a reduction in the potency of OdDHL, but this was not mimicked by the presence of 0.3 mM apamin with 0.1 mM iberiotoxin (Table 3). We have demonstrated that endothelial denudation and inhibition of potassium channels produce a qualitatively similar effect on the relaxant activity of OdDHL. However, these effects do not directly involve either nitric oxide or a prostanoid, as inhibitors of these mediators increased the potency of OdDHL. The failure of apamin and iberiotoxin to influence responses to OdDHL excludes a role for either small or large conductance Ca2+ activated K+ channels.2

bacterial infection and acts at specific G protein-coupled receptors.1 Previous studies have shown evidence of variable contractions to fMLP in human isolated umbilical veins and coronary arteries.2 This study investigated whether fMLP can modify vascular tone in human and porcine pulmonary arteries. Isolated segments of human and porcine pulmonary arteries (5 mm length, 2–3 mm diameter) were prepared for isometric tension recording and pre-contracted with the thromboxane-mimetic, U46619.3 A sub-maximally effective concentration of U46619 was used to induce a sustained contraction, followed by exposure to 0.1 mM acetylcholine to assess the integrity of the endothelium. The segments were again exposed to U46619 and the effect of noncumulatively increasing concentrations of fMLP determined following endothelium-denudation or in the presence of 3 mM flurbiprofen, an inhibitor of cycloxygenase. Responses have been expressed as a percentage of the U46619-induced contraction and differences between groups compared using a Mann–Whitney U-test (P<0.05 considered statistically significant). Non-cumulative exposure to fMLP (0.1–30 mM) caused variable responses in human pulmonary arteries, comprising either a small contraction (10 min duration), followed by a prolonged relaxation of U46619-induced tone, or simply a large contraction. The maximum response was produced by 10 mM fMLP and caused a 111.1 (39)% (n=8) increase in tone. In contrast, fMLP (0.1–30 mM) caused only a concentration-dependent relaxation of U46619-induced contractions in the porcine pulmonary artery; maximum effect equivalent to 69.6 (9.4)% (n=7) inhibition. The tripeptide MLP (30 mM) failed to elicit a response in either artery (<10%, n=6–7). A fMLP receptor antagonist, BPLPLP (10 mM), significantly inhibited responses to 30 mM fMLP in human (23.9 (10)%, n=8) and porcine (30.9 (3.2)%, n=7) vessels. The fMLP-induced contractions in human pulmonary arteries were significantly inhibited by endothelium-denudation (29.5 (12.6)%, n=6), 3 mM flurbiprofen (10.6 (2.6)%, n=7) and 750 u ml
1 superoxide dismutase (SOD) (32.1 (10.1)%, n=7), while relaxations in porcine arteries were unaffected (n=7–11). Specific fMLP receptors are present on human and porcine pulmonary vessels and mediate contraction and relaxation, respectively, but require concentrations 300-fold greater than that involved in the chemotaxis of neutrophils.1 Contractions in human pulmonary arteries arise from the generation of endogenous prostanoids and free-radicals by the endothelium, as has been noted in other human vessels.2 In contrast, fMLP-induced relaxations in porcine pulmonary arteries are mediated by direct effects on the smooth muscle.

Proceedings of the Anaesthetic Research Society

Keywords: case reports; medicine, evidence-based

Table 4 Times in minutes (range) to waking, discharge home, and activity scores. *P<0.0001 (KruskallWallis) Group 1; Atr/Sev Time to spontaneous response Time to orientation Time to discharge home Median activity next day 3 (Scores of 1,2,3,4)

3.2 (1–11)* 18.7 (10–28) 220 (89–443) (1–4) (3,3,18,3)

Group 2; Prop/Remi 8.3 (4–13) 7.3 (3–13) 20.5 (13–39) 196 (88–317) 3 (1–4)

Group 3; Remi/Sev

15.1 (9–23)* 192 (96–362) 3 (1–4)

(2,7,12,6)

(2,4,13,7)

Gynacological patients, ASA I and II, presenting for diagnostic day-case laparoscopic procedures were recruited and divided into three groups. Group 1 (A/S) were induced with propofol, and maintenance used atracurium and sevoflurane. Group 2 (P/R) were induced and maintained with target-controlled propofol (4 mg litre
1) and a bolus of 1 mg kg
1 of remifentanil followed by a continuous infusion of 0.25 mg kg min
1. Group 3 (R/S) were given a propofol and 1 mg kg
1 remifentanil bolus followed by sevoflurane and infusion of 0.25 mg kg min
1 remifentanil. All patients were ventilated via a laryngeal mask, using nitrous oxide/oxygen for Group 1 and oxygen/air for Groups 2 and 3. Additionally all received fentanyl 100 mg, tramadol 100 mg, tenoxicam 20 mg and cyclizine 50 mg. Observations included haemodynamics, ventilation pressure, surgical conditions, and recovery profiles. On the first postoperative day, patients were contacted by telephone for pain level, activity (1, still in bed to 4, normal activity) and nausea and vomiting as in previous studies.1 Out of 81 patients (27 in each group), one patient (P/R group) presented a high intra-abdominal pressure for the surgeon and another (R/S group) required an additional small incision. Atracurium/sevoflurane patients showed a more rapid return of spontaneous respiration, but time to orientation was faster in the remifentanil/ sevoflurane patients (Table 4). There was no difference in time to discharge from recovery, or discharge home. Five patients (one A/S, two P/R, two R/S) were kept overnight because of pain or vomiting. On the following day, there were no differences in recovery profiles, activity, or pain between groups. This is in contrast with a previous study in minor gynaecological procedures.1 Perhaps the recovery conditions at home are not influenced by anaesthetic technique because they are masked by surgical morbidity for laparoscopy patients, although the use of remifentanil without neuromuscular blocking agents can provide acceptable conditions for gynaecological laparoscopy. Keywords: surgery, daycase; surgery, laparoscopy

Reference 1 Jenicek M. Clinical Case Reporting in Evidence-Based Medicine, 2nd Edn. London: Arnold, 2001

Reference

Gynaecological day-case laparoscopy without neuromuscular blocking agents; any benefit postoperatively?

Does gas under diaphragm behave differently following laparosopic cholecystectomy compared with gynaecological laparoscopy?

N. Mahmoud1, R. A. Martlew2 and A. S. Laurence2

V. Sharma1 , A. S. Laurence1, J. Hill2 and B. Kocaman1

1

1 Mahmoud N, Rose DJA, Laurence AS. Anaesthesia 2001; 56: 171–4

Departments of Anaesthetics, Royal Albert Edward Infirmary, Wigan, UK and 2Royal Preston Hospital, Preston, UK

1 Department of Anaesthetics and 2Department of Radiology, Royal Preston Hospital, Preston, UK

Gynaecological laparoscopy traditionally involves muscle relaxation, although some anaesthetic techniques can allow abdominal surgery without neuromuscular blocking drugs. However, rapid recovery for early discharge is desirable. Can the newer anaesthetic agents offer any benefit?

Previous work has shown that carbon dioxide appears to be absorbed from the abdomen over several days in an approximately exponential manner following gynaecological laparoscopy1 and the degree of shoulder-tip pain correlates with the volume of the bubble.2 Does the same happen following laparoscopic cholecystectomy? Patients

607P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

physical vs biological contribution for the process described). Categories are scored 1–5 with the help of a ‘score guide’ developed by the authors as a result of five cycles of testing, comparison, and resolution of conflicts. (Each cycle was an arbitrarily chosen month of case reports in different anaesthesia journals.) To test the system with anaesthetic colleagues a specific set of 10 recently published case reports (from the April 2004 issue of Anesthesia and Analgesia) was used. Initially colleagues were asked to independently assess all 10 reports on the basis of a simple rating (1–5, highest best) equivalent to how they would judge a paper normally. In addition they were asked how much they agreed (1–5, higher less-so) with the assertion ‘Case Reports have little value as far as Evidence Based Medicine is concerned’. After an interval of not more than 2 weeks they were then asked to repeat the exercise using the same case reports and the new scoring system. They were given minimal tuition on the scoring guide to determine whether it was self-evident. At the end of the exercise they were also asked to report whether they thought the new system had value (1–5, higher less-so) by the question ‘‘Did you feel this system has any place in helping to assess ‘evidence’ in case reports?’’ Five of six trainees and four consultants who agreed to be involved returned forms in the time limit set. In the preliminary simple rating the mean value overall was 2.87 and no individuals were considered seriously out of line with the rest. (Mean (SD) for trainees was 2.74 (1.03) and consultants 3.03 (1.03).) The rating for the value of case reports was 2 (n=2), 3 (n=4), and 4 (n=3). Our initial measure of efficacy for the scoring system was consistency with ‘target’ scores, the average of preliminary independent scoring of the same case reports by the authors. (Of a total of 40, 25 were the same; differed by one in 13 and by two in just 2.) For the trial subjects, ‘hit’ was when the score was on target or within 0.5, ‘near’ within 1–1.5, and ‘miss’ was out by more than 1.5. The mean category scores for the 10 studies overall was: events hit 5.2, near 2.7, miss 1.1; contexts hit 2, near 4, miss 3; practice points hit 3.3, near 3.9, miss 1.8, and biophysical indices hit 1.9, near 3.4, miss 3.7. The answer to the question whether the system was regarded as helpful was encouraging (scores were 2 (n=5), 3 (n=4)). Consistency of scoring is obviously not the same as accuracy and many factors would be expected to influence both. In our view a scoring system supported by a generally agreed scoring guide should be feasible and warrants further exploration. A more prescriptive scoring guide should not become unduly long or unwieldy. Although the biophysical index performed badly in this study free text comments (which were allowed alongside the numerical scores) indicated that this was the least intuitive to score and may have performed better with a more full explanation than was used.

Proceedings of the Anaesthetic Research Society

ASA I and II, presenting for laparoscopic cholecystectomy on a morning list at the beginning of the week were recruited and gave signed informed consent for three X-rays following their operation, if their operation did not proceed to open cholecystectomy. During the afternoon following the operation, the patient was visited and assessed to ensure that he/she was happy for the journey to the X-ray department, accompanied by one of the researchers. For the X-ray on the second and third day, the normal portering service was used for the journey. In the X-ray department, patients were propped upright on their bed with wedges and a limited horizontal X-ray of the diaphragm area taken. The volume of gas under the left and right hemi-diaphragm was determined by measuring the arc height and length of the bubble and using the formula from previous studies.1 2 The minimum detectable gas volume was considered to be 0.1 ml, corresponding to a bubble size of 0.2·0.5 cm. Eighteen patients were X-rayed on at least two occasions (Fig. 2). Only eight patients had detectable gas at some time, which was absorbed slowly, but not as consistently as found following gynaecological procedures. The remaining 10 patients had no detectable gas, even on the day of operation. Gas bubble volumes were generally greater under the right hemi-diaphragm. Gas volumes seem less than found previously following gynaecological procedures, despite the operation being considerably longer. Over half the patients had no detectable gas, whereas gas was detected in all our gynaecological patients. General surgeons seem to be more careful than their gynaecological colleagues, or are there other factors? Acknowledgement: Lancashire Teaching Hospitals Trust Seedcorn Research Fund for costs of X-rays. Keywords: surgery, cholecystectomy, laparoscopic

Acknowledgement: Funding via grant from research and development fund, Broomfield Hospital, Chelmsford, Essex CM1 7ET. Keywords: analgesics opioid, diamorphine; equipment, pre-mixed syringes

References 1 Cowan CM, Kendall JB, Barclay PM. Wilkes RG. Br J Anaesth 2002; 89: 452–8 2 Abuzaid H, Prys-Roberts C, Wilkins DG, Terry DM. Anaesthesia 1993; 48: 492–5

Synchronization and directionality in cardio-respiratory oscillations in anaesthesia: a preliminary observational study in human males M. Entwistle1*, A. Bandrivskyy2*, B. Musizza3*, A. Stefanovska4*, P. McClintock2* and A. Smith1*

References

1

Department of Anaesthesia, Royal Lancaster Infirmary, UK. Department of Physics, Lancaster University, UK. 3Jozef Stefan Institute, Ljubljana, Slovenia. 4Faculty of Electrical Engineering, University of Ljubljana, Slovenia

1 Stanley I, Laurence AS, Hill J. Anaesthesia 2002; 57: 57–60 2 Jackson SA, Laurence AS, Hill JC. Anaesthesia1996; 51: 485–7

2

Stability of pre-mixed syringes of diamorphine and heavy bupivacaine for spinal anaesthesia S. J. Hudson1*, M. F. Jones2*, S. Nolan2*, H. Ellis1*, R. Duncombe1* and M. Alexander-Williams1 1

Departments of Anaesthesia and Pharmacy, Broomfield Hospital, Chelmsford CM1 7ET, UK. 2Quality Control North West, Stepping Hill Hospital, Stockport, UK

The aim of this study was to assess whether diamorphine 100 mg ml
1 is stable in solution with heavy bupivacaine 0.5%, to allow production of pre-filled syringes for use in spinal anaesthesia.

Previous analyses of cardiorespiratory coupling in anaesthesia have used linear methods.1 A different approach based on wavelet transform2 and synchronization indices revealed a marked increase in synchronization between the rhythmic oscillations of respiration and the ECG3 in rats given ketamine and xylazine. The direction of interaction also changed with changing depth of anaesthesia. Do these phenomena also occur in humans? We studied 10 healthy men aged between 20 and 40 undergoing elective surgery. Standard clinical monitoring was supplemented by forehead BIS Quatro electrodes (Aspect Medical Systems, Leiden) for EEG capture and a TSD201 respiratory effort transducer (BIOPAC Systems Inc., Goleta, CA) for ventilatory

608P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

Fig 2 Volume of gas under diaphragm (L+R), ml (minimum detectable volume 0.1 ml).

It is a common clinical practice to add diamorphine to heavy bupivacaine when performing spinal anaesthesia for either obstetric or general surgical procedures.1 2 Current practice involves the anaesthetist preparing this solution just before injection. If pre-filled syringes were available potential problems arising as a result of the wrong mixture being administered could be eliminated, whilst also providing greater assurances of sterility and accuracy of dosage. Previous work has shown diamorphine to be stable in plain bupivacaine but no data have been reported regarding stability in heavy bupivacaine. It is therefore necessary to establish the stability of such solutions. Diamorphine hydrochloride was dissolved in water for injection, and added to heavy bupivacaine to give a mixture with a diamorphine concentration of 100 mg ml
1 and bupivacaine w/v of 0.4%, then stored in 5 ml syringes. Eleven syringes were stored at each of 40, 25, and 7 C for 90 days. Samples were taken at several time points for measurement of diamorphine and bupivacaine levels using high performance liquid chromatography. We found that diamorphine concentrations fell over the study period. The rate of decline was related to the temperature at which the sample was stored. No significant changes were observed in pH of 25 C/60% RH sample and in 7 C sample, or in the bupivacaine content of samples. Ten per cent degradation of diamorphine occurred after 4 days at 40 C, 7 days at 25 C, and 26 days at 7 C. Therefore, we conclude that diamorphine is stable in heavy bupivacaine for long enough to allow manufacture and storage of prefilled syringes by hospital pharmacy aseptic units for use in spinal anaesthesia.

Proceedings of the Anaesthetic Research Society

Fig 3 Cardiorespiratory synchrogram for anaesthetized patient showing the phase relationship of heartbeat relative to a fixed point in the breathing cycle (C relative phase). Horizontal (flat) portions of the trace reveal synchronization.

Acknowledgements: J. Petrovcˇicˇ, Y. Shiogai and A. Pershakova (equipment and measurements) and Mr I. Crighton (flexibility in surgical scheduling). Keywords: anaesthesia, depth; cardiovascular system, cardio-respiratory oscillations

References 1 Galletly DC, Larssen PD. Br J Anaesth1997; 79: 35–40 2 Stefanovska A, Bracˇicˇ M. Contemp Physics1999; 40: 31–55 3 Stefanovska A, Haken H, McClintock PVE, et al. Phys Rev Lett 2000; 85: 4831–4

Fig 4 Distribution of heart rate observations (n=4219).

were defined by the 5th and 95th percentiles of the sample’s distribution. The resulting upper and lower limits for this zero score range are respectively, 59 and 107 beats min
1, these are similar to Morgan’s original instruction respectively, 51 and 100 beats min
1.2 We used natural logarithms to set the other score range limits: the limits less than score=2 (bigger dashed line in Fig. 4) were defined by the percentiles, respectively, ln(5)=1.61 and 100
ln(5)=98.39 and, the limits less than score=3 (smaller dashed line in Fig. 4) were defined by the percentiles, respectively, ln(1.61)=0.48 and 100
ln(1.61)=99.52. Using distribution-derived values to determine cut-off points has an intuitive appeal, as it appears that the assumptions of symmetry implicit in commonly used EWSS algorithms may be justified. We considered the data here to be offset, towards higher PRs, as a result of the dynamics of the group. Therefore, we are seeking a new dataset of unselected hospital patients with which to validate this approach. With these data we also plan to model the effect of these new ranges on trigger thresholds. Acknowledgement: This project was funded by the UK Department of Health. Keywords: monitoring, early warning scoring system

The effect of using aggregated patient data to determine cut-off points in an early warning scoring system

References 1 Department of Health. Comprehensive Critical Care. London: DH, 2000 2 Morgan R, Williams F, Wright. Clin Intens Care1997; 8: 100

R. Oakey1*, A. Harry1* and A. F. Smith2 1 Research and Development and 2Anaesthesia, Royal Lancaster Infirmary, Lancaster, UK

Early warning scoring systems (EWSS) have been advocated to aid referral of deteriorating hospital ward patients to higher levels of care.1 Abnormal primary physiologic observations are assigned scores (between 1 and 3, in both directions) depending on the severity of the patient’s condition. The cut-off points between score limits are apparently based on clinical experience; these are however poorly validated and the assumptions underlying them have not been made explicit. Here we present a new method that selects statistically valid score ranges. In this study we have utilized the primary physiologic data from 238 patients admitted with suspected Legionnaires’ disease to Furness General Hospital (FGH) during the outbreak of 2002. An EWSS that was modelled on Morgan’s2 was introduced at FGH in the early stages of the outbreak. The data for each of the primary observations used in the EWSS were aggregated. Then, from the resulting distributions, score ranges were derived and compared with their respective scoring instruction. In the analogy with descriptive statistics, the ‘normal’ limits of the score ranges

Cannabis in postoperative pain: a pharmacokinetic study A. Holdcroft, C. Dore´*, P. Phillips*, M. Maze and M. Hanna Magill Department of Anaesthesia, Chelsea and Westminster Hospital, Imperial College London, King’s College Hospital and the MRC Clinical Trials Unit, London, UK

Cannabis extracts contain a mixture of cannabinoids such as tetrahydrocannabinol (THC) and cannabidiol (CBD). When taken orally low, slow, and erratic bioavailability has been measured from 10 to 20%.1 Maximal plasma concentrations of THC peaked between 1 and 2 h but could take up to 6 h and more than one peak was observed in several subjects.2 We aimed to measure absorption and metabolism of an oral extract of cannabis plant material the day after surgery in patients while standardizing for lack of previous cannabis use, starvation and dose. This study was conducted as part of the CANPOP multicentre dose finding study of Cannador. Ethics Committee approval was given for patients who had an i.v. cannula in place. Informed written consent was taken before surgery and reaffirmed

609P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

frequency (f ). A standard anaesthetic (midazolam 2 mg, fentanyl 1.5 mg kg
1, sleep-dose propofol and isoflurane 1% in 40% oxygen in air) was administered by the same anaesthetist. Recordings were made for 30 min in the anaesthetic room before surgery. Waveforms were digitalized at 1000 Hz and stored on a PC for later analysis using the above non-linear techniques. There appears to be an increase in cardiorespiratory synchronization in these anaesthetized men. The direction of interaction was such that respiration was ‘driving’ the heart (Fig. 3). We now plan to monitor patients throughout surgery and into recovery to explore how synchronization and directionality alter with changing depth of anaesthesia.

Proceedings of the Anaesthetic Research Society

Table 5 The area under the curve (AUC) and the maximum concentration (Cmax) for 5 and 10 mg oral doses of Cannador AUC ng ml
1 min
1

Cmax ng ml
1

Dose n Median Minimum Maximum n Median Minimum Maximum (mg)

THC

5 10 CBD 5 10 11-OH 5 10

7 99 2 842 7 22 2 210 7 506 2 1513

5 103 1 18 22 604

1021 1581 161 402 825 2423

7 4 7 4 7 4

0.62 0.96 0.15 0.32 3.18 4.51

0.14 0.03 0.04 0.01 0.87 0.12

11.2 12.0 2.0 3.0 6.2 17.5

Acknowledgement: This study was part funded by the Medical Research Council and the Westminster Medical School Research Trust. We thank the Institute for Clinical Research (IKF Berlin) for the donation of Cannador.

Keywords: protein, drotrecogin alfa, guidelines

References 1 Angus DC, Linde-Zwirble WT, Lidicker J, et al. Crit Care Med 2001; 29: 1303–10 2 Bernard GR, Vincent JL, Laterre PF, et al. N Engl J Med 2001; 344: 699–709

Keywords: analgesics, cannabis; pain, postoperative

References 1 Wall ME, Sadler BM, Brine D, Taylor H, Perez-Reyes M. Clin Pharmacol Ther 1983; 34: 352 2 Ohlsson A, Lindgren JE, Wahlen A, et al. Clin Pharmacol Ther 1980; 28: 409–16

Action of ropivacaine on cloned cardiac Kv4.3/KChIP2.2 complexes P. Friederich1* and A Solth1 2* 1 Department of Anaesthesiology and 2Centre for Molecular Neurobiology, University Hospital Hamburg, Germany

Guideline applied practice: evaluating the clinical introduction of drotrecogin alfa M. Stallwood1*, J. Nolan1*, R. Wenstone1*, G. Masterson1* and G. Marx2 1

Department of Intensive Care, Royal Liverpool University Hospital and 2University Department of Anaesthesia, University of Liverpool, Liverpool, UK

Sepsis is a major health problem with a high mortality rate. A recent US epidemiology survey of severe sepsis estimated that there are 751 000 cases per yr in the US with a mortality of 28.6%.1 The therapeutic potential of activated protein C was evidenced in a recent phase 3 trial, in which the administration of human recombinant activated protein C (drotrecogin alfa (activated)) (DA) to patients with severe sepsis resulted in reduced mortality.2 An institutional guideline for DA use was developed and approved by the local drugs and therapeutic committee. The aim of this audit was to determine compliance with guidelines and evaluate characteristics and

The transient outward current Ito is an important repolarizing K current in human myocardium.1 The magnitude of Ito is reduced in various pathological states such as ventricular failure, myocardial infarction, and atrial fibrillation. Ito has been suggested to constitute a molecular target involved in cardiotoxic action of bupivacaine.2 The novel amino-amide local anaesthetic S(–) ropivacaine has been developed as a safer alternative to bupivacaine with regard to cardiotoxic side effects. However, the effects of S(–) ropivacaine on Ito have so far not been studied. In human ventricular myocardium Ito is formed by Kv4.3 and KChIP2.2 subunits. The aim of this study was to establish the effects of S(–) ropivacaine on human Kv4.3/ KChIP2.2 channels. Kv4.3/KChIP2.2 cDNA cloned from human heart was transiently transfected in Chinese Hamster Ovary cells. The pharmacological effects of ropivacaine were investigated with the patch clamp method.3

610P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

postoperatively. Following oral fluid intake without nausea or vomiting and agreement to stop the morphine patient-controlled analgesia Cannador was given as 2.5 mg capsules in a single dose of 5 mg or 10 mg. Blood samples for plasma were taken at 11 times: 15, 30, 45, 60, 90, 120, 150, 180, 240, 300 and 360 min after oral administration. The samples were analysed using specific gas chromatography-mass spectrometry (GC-MS) with a limit of quantification of 0.1 ng ml
1. The samples were tested for THC, CBD and the active metabolite 11-OH THC. Seven patients were recruited into the 5 mg dose and four at 10 mg. For each patient two summary measures were calculated: the area under the curve (AUC) and the maximum concentration (Cmax). A Mann–Whitney U-test was performed to compare the median value of each summary measure in each dose group. There were no significant differences between the two doses but the numbers were small and there was great variability between patients (Table 5).

outcomes of patients treated with DA on our 13-bedded general adult intensive care unit. The institutional guideline for DA use incorporates the presence of three or more criteria of the systemic inflammatory response syndrome, at least two-system organ failure, an APACHE II score of more than/equal to 25 and prescription by a consultant intensivist within 48 h after onset of sepsis-induced organ failure. Clinicians were provided electronically with the guideline and an information manual. A research nurse retrospectively reviewed records of the patients receiving the drug. The results are presented as median and interquartile range (IQR). Between March 2003 and December 2003, 11 patients received DA. Complete records were available for all patients. One hundred per cent met guideline criteria for administration of DA. Five female and six male patients received DA with a median age of 66 yr (IQR: 26) and an APACHE II score of 31 (IQR: 6). All patients required mechanical ventilation and the use of vasopressors. Six patients had three-system and five patients two-system organ failure. Seventythree per cent had a positive blood culture. Time after onset of sepsis-induced organ failure until start of DA was 9 h (IQR: 10 h; range: 4–48 h). The duration of infusion was 96 h (IQR: 73 h; range: 2–96 h). Overall 28-day mortality was 55% with a predicted APACHE II mortality of 74%. Notably, 64% (7/11) of the patients completed the 96-h infusion. Of those completing the full course, mortality was 29% (2/7) whereas all four patients not completing the 96-h course died. One non-intervention related major haemorrhage occurred (gastrointestinal) and the drug was discontinued for haemostatic concerns in three more patients. One of these events was intervention-related (bronchoscopy performed without stopped DA administration—protocol violation), one as a result of a decrease of platelets to 6000 mm3 after DA administration was initiated and finally, one haemorrhage at a catheter-insertion side. Guideline applied practice could be successfully used for local introduction of drotrecogin alfa treatment. Our data suggest a possible association between treatment duration and mortality.

Proceedings of the Anaesthetic Research Society

Acknowledgement: Ropivacaine was a kind gift from ASTRA/ ZENECA, So¨dertalje, Sweden. Supported by the Deutsche Forschungsgemeinschaft (FR 1625/1-1). Keywords: anaesthetics local, ropivacaine; ions, potassium, K-channels

Table 6 Diagnostic performance of markers to predict sepsis on inclusion day; *P<0.05 PCT

IL-6

Cut-off 2 ng ml
1 150 pg ml
1 value Sensitivity 52 90 (%) Specificity 86 40 (%) AUC 0.70* 0.80*

C5b-9

WBC

CRP

600 ng ml
1 12 000 mm
3 150 mm ml
1 31

41

55

65

65

60

0.51

0.54

0.64

(Lumitest PCT Brahms Diagnostica). The other parameters were measured with conventional methods (commercially available). Each patient was examined daily for signs and symptoms of infection. Sepsis and severe sepsis/septic shock were defined according to the criteria of the American College of Chest Physicians/Society of Critical Care Medicine. The severity of sepsisrelated organ failure was assessed by the sepsis-related organ failure assessment (SOFA) score. Values were given as median and interquartile range (IQR). Sensitivity and specificity were calculated. Diagnostic accuracy of the parameter determined at study entry was expressed as the area under the receiver-operating curve (AUC). Of 72 recruited patients 11 developed postoperative sepsis and 18 severe sepsis or septic shock. Infection was microbiology proven in 22 of 29 patients (76%). The patients enrolled had an age of 68 (21.5) yr and an initial SOFA score of 5 (4). After 28 days, 12 patients had died, 60 were survivors (Table 6). Our results indicate that PCT may be a specific indicator of sepsis in postoperative patients and IL-6 may be a sensitive predictive marker for sepsis. Early recognition of sepsis is of particular interest, because this could expedite the initiation of early specific treatment and potentially improve patient outcome. Acknowledgement: This study was supported by grants from the R&D RLBUH NHS Trust support fund and from the R&D fund of the University of Liverpool. Keywords: complications, sepsis, postoperative; monitoring, predictive markers

References 1 Dixon JE, Shi W, Wang HS, et al. Circ Res 1996; 79: 659–68 2 Castle NA. J Pharmacol Exp Ther 1990; 255: 1038–46 3 Hamill OP, Marty A, Neher E, et al. Pflu¨gers Arch 1981; 391: 85–100

Predictive markers for postoperative sepsis

Reference 1 Angus DC, Linde-Zwirble WT, Lidicker J, et al. Crit Care Med 2001; 29: 1303–10

A. Breen1*, G. Marx1, G. Ritchie1*, S. Swaraj*, C. H. Toh2*, A. Shenkin3* and M. Leuwer1

The dynamic response to mivacurium in healthy and tetraplegic patients

1 Departments of Anaesthesia, 2Haematology and 3Clinical Chemistry, University of Liverpool, Liverpool, UK

A. Axon*, J. Appiah-Ankam*, J. Watt*, S. Adejumo*, C. Schrimshaw* and J. M. Hunter

Sepsis is a major health problem with a high mortality rate. A recent US epidemiology survey of severe sepsis estimated that there are 751 000 cases per yr in the US with mortality of 28.6%.1 The aim of this study was to determine the value of procalcitonin (PCT), white blood cell count (WBC), interleukin-6 (IL-6), terminal complement complex (C5b-9), and C-reactive protein (CRP) in the prediction of postoperative sepsis before the onset of clinical deterioration. In a prospective clinical observational study 72 non-infected patients admitted postoperatively to an intensive care unit (ICU) who required more than 24 h ICU-treatment were recruited. Blood samples were collected on the first, second, third, fifth, seventh and tenth day. PCT was determined by an immuno-luminometric assay

Southport and Ormskirk Hospitals and University Department of Anaesthesia, Liverpool, UK

There is evidence that the response to neuromuscular blocking drugs is altered in paralysed muscle and resistance has been reported.1 2 We are undertaking a prospective study of the response to mivacurium in healthy and tetraplegic patients. Neuromuscular block is assessed by acceleromyography (TOFWatch SX, Organon) simultaneously in the adductor pollicis and trapezius muscles. Following induction with propofol and fentanyl, monitoring is commenced using the ‘train-of-four’ technique at 15 s intervals. Mivacurium 0.15 mg kg
1 is given and, following recovery of T1/T0 to 10%, an infusion of 3 mg kg
1 min
1 started.

611P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

Ropivacaine inhibited Kv4.3/KChIP2.2 channels in a concentrationdependent, stereospecific, and reversible manner. The IC50-value of S(–) ropivacaine for inhibition of the charge conducted by Kv4.3/KChIP2.2 channel was 117 (21) mM (mean (SEM), n=30). The local anaesthetic accelerated macroscopic current decline with an IC50-value of 77 (11) mM (mean (SEM), n=30). It shifted the midpoint of channel activation into the depolarizing direction, and it slowed recovery from inactivation without altering steady-state inactivation. Co-expression of KChIP2.2 alters the pharmacological effect of S(–) ropivacaine and R(+) ropivacaine on Kv4.3 channels. Inhibition of Q by either isomer differed between Kv4.3 and Kv4.3/ KChIP2.2 channels with Kv4.3 channels being more sensitive to the inhibitory effect (n=6 for Kv4.3, and n=7 for Kv4.3/KChIPP2.2, P<0.05). Block of Kv4.3 channels developed with a time constant significantly smaller than the respective time constant of the block of Kv4.3/KChIP2.2 channels (P<0.05). Inhibition of the charge conducted by Kv4.3 channels by ropivacaine was also stereo-specific with different time constants of the onset of block(tonset=2.3 (0.4) ms, for S(–) ropivacaine, vs tonset=2.6 (0.5) ms for R(+) ropivacaine, mean (SD), n=7 paired experiments, P<0.05). The results are consistent with the idea that ropivacaine by blocking Kv4.3/KChIP2.2 from the open state interferes with the gating modifying effects of KChIP2.2 on Kv4.3 channels. The inhibitory effects of S(–) ropivacaine on Kv4.3/KChIP2.2 channels may contribute to prolongation of the QTc interval of the ECG by altering the initial phase of the action potential waveform. Prolongation of initial action potential repolarization as a result of inhibition of Ito has furthermore been suggested to impair ventricular excitation contraction coupling. Suppression of Ito by S(–) ropivacaine may, thus, result in complex changes of myocardial function during events of intoxication. The difference in S(–) ropivacaine sensitivity between Kv4.3 channels and channel complexes formed by Kv4.3 and KChIP2.2 may imply that ventricular layers with a lower level of KChIP expression may be more vulnerable to the inhibitory action of S(–) ropivacaine.

Proceedings of the Anaesthetic Research Society

Table 7 Onset and recovery variables simultaneously monitored in the adductor pollicis (AP) and trapezius (Trap) muscles. *P<0.05 vs control; **P<0.05 vs adductor pollicis Control Mean (SD) (n=6)

AP 229.7 (124.9) 191.2 (99.7) 38.5 (67.7) 0.58 Trap 114.8 (42.9)** 93.8 (18.4) 21.0 (19.4) 0.31 22.3 (2.5)

22.2 (6.9)

16.3 (7.9)

20.2 (4.4)
4.0 (4.4)

0.41

13.1 (3.3) 12.3 (5.6)

11.0 (4.2) 8.8 (2.8)

0.40 0.21

AP (n=4) Trap (n=4) AP Trap

AP Trap

12.5 (4.1) 9.1 (5.2)

Median (range) UII pg ml
1

Differences P Mean (SED) Plasma CSF Cord

Control

n

PET

n

11.85 (6.71–15.6) 8.24 (2.11–31.30) 10.10 (4.10–17.30)

8 8 10

9.29 (3.86–13.80) 8.73 (4.21–17.00) 13.10* (5.61–21.30)

10 8 10

0.18 (3.7) 0.96

Acknowledgement: Organon, for a TOF-Watch SX and software.

With LREC approval and informed consent we recruited two groups of 10 patients; control (mean (range)). Age: 29 (22–43) yr, BMI: 25 (20–32); gestation: 273 (267–281) days, MAP on day of delivery: 81 (75–96) mm Hg) and PET (age: 34 (22–40) yr, BMI: 25 (21–46), gestation: 253 (203–289) days, MAP on day of delivery: 106 (88–128) mm Hg). Blood and CSF (collected during placement of the spinal anaesthetic needle) were sampled and prepared as described previously.2 UII was extracted and levels measured using a commercial kit as described2 with an intra-assay coefficient of variation of 5.6%. Two plasma and two CSF samples in the control and two CSF samples in the PET group were below the assay detection limits. There were no differences in maternal plasma or CSF or cord UII levels between the groups. However, there was a small (40%) but significant increase in cord UII levels when compared with paired plasma in the PET group, Table 8. There was a weak but significant negative correlation (r2=0.4, P=0.049) between cord UII levels and gestation in the PET group. In addition, we observed a significant positive correlation between plasma and CSF (r2=0.57, P=0.0009, n=16), plasma and cord (r2=0.43, P=0.0031, n=18) and CSF and cord (r2=0.32, P=0.022, n=16) UII levels for the whole data set. Collectively the data indicate that UII levels do not increase in PET compared with controls but in PET patients cord UII levels are elevated relative to paired plasma samples. Elevated cord UII concentrations may simply indicate reduced cord viability and possibly UII metabolism as a result of reduced blood flow or possibly that the cord is producing UII. These suppositions will require further experimentation to evaluate.

Keywords: complications, tetraplegia; muscle, relaxant; neuromuscular block, mivacurium

Acknowledgement: Funded by a research grant from UHL NHS Trust.

References

Keywords: complications, pre-eclampsia; hormones, urotensin

6.9 (3.4)* 5.7 (1.9)

2.1 (2.3) 3.5 (2.6)

5.7 (2.3) 0.04 3.4 (2.3) 0.20

This is discontinued towards the end of surgery and monitoring continued until recovery from block. The pharmacodynamic data (Table 7) were compared using Student’s two-tailed t-test (SPSS v8.0, SPSS, Inc.). The mean infusion rate of mivacurium was 3.2 mg kg
1 min
1 (SD 1.0) in the control group and 2.6 mg kg
1 min
1 (0.7) in the tetraplegic group. The mean duration of infusion was 77.6 (95.2) and 77.4 min (79.0), respectively. We have not demonstrated resistance to neuromuscular block with mivacurium in tetraplegic patients. However, there is some evidence that the final stages of recovery from block are more rapid in this population.

1 Graham DG. Anesthesiology 1980; 52: 74–5 2 Iwasaki H, Namiki A, Omote K, Omote T, Takahashi T. Anesth Analg 1985; 64: 864–6

Urotensin II concentrations are not elevated in pre-eclampsia

References 1 Douglas SA Curr. Opin Pharmacol 2003; 3: 159–67 2 Thompson JP, Watt P, Sanghavi S, Strupish J, Lambert DG. Anesth Analg 2003; 97: 1501–3

Use of a physical model to explore limitations in use of the Bonfils laryngoscope

E. Cowley, J. Waugh, N. Ali, P. Sharpe, J. P. Thompson and D. G. Lambert

M. Halligan and P. Charters

Department of Cardiovascular Sciences, Division of Anaesthesia, Critical Care and Pain Management, LRI, Leicester LE1 5WW, UK

Department of Anaesthesia, University Hospital Aintree, Liverpool, UK

Urotensin II (UII) is the most potent endogenous vasoconstrictor identified to date1 and plasma concentrations are increased in heart failure and renal failure.2 The aetiology of pre-eclampsia is uncertain and we hypothesized that in diseases where MAP increases then UII levels might also be elevated. In this study we measured maternal plasma and cerebrospinal fluid (CSF) UII levels along with those in cord plasma from pre-eclamptic (PET) and normotensive patients undergoing Caesarean section under spinal or combined spinal-epidural anaesthesia.

The Bonfils intubation laryngoscope is a rigid fibre-optic system previously suggested to have specific advantages for patients with supraglottic laryngeal tumours.1 It is not recommended for use in nasal intubations, but little attention otherwise has been paid to the conditions which limit efficacy. A physical model was developed to address this question. The model was designed to incorporate variations in relevant bony features and to make visibility and access straightforward. Omission of the tongue, epiglottis, and stylo-hyoid apparatus from the model was justified on the basis of how this

612P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

Time to maximum T1/T0 depression (s) Time for T1/T0 recovery to 10% before infusion (min) Time for T1/T0 recovery 0.5!0.9 after infusion (min) Time for T4/T1 recovery 0.7!0.9 after infusion (min)

Tetra Mean (SD) (n=6)

Table 8 Plasma, CSF and cord UII concentrations (*P=0.002 increased compared with paired plasma, Wilcoxon signed rank)

Proceedings of the Anaesthetic Research Society

Table 9 Mean and standard deviation (SD) of peak forces and duration of laryngoscopy. *P<0.05 (ANOVA) of Mac3

Disposable plastic blade Penlon crystal ‘green’ Truphatek Lite-blade II bulb type Heine XP Truphatek Lite-blade slim Rusch Bioblad Vygon code 691.03 Timesco Freeway Welch-Allyn one piece moulded blade Rusch Truphatek Lite-blade II Penlon crystal polycarbonate clear plastic Vital Signs Vital View ‘straight’ Vital Signs Vital View ‘curved’ Inter-surgical Flexicare Venticaire Disposable metal blade Proact green spec Timesco Europa Proact metal max 90 Timesco Callisto Proact metal Max 100 Re-usable metal blade Penlon standard Mac3

Blade Force (n) number

Time (s)

5 12 13 6 11 7 10 3 14 4 9 8 1 2

33.2 (16.1) 33.7 (18.1) 34.2 (17.1) 34.3 (16.8) 34.6 (16.5) 34.9 (16.5) 35.3 (17.7) 35.3 (18.1) 35.4 (16.5) 36.5 (16.6) 36.5 (17.8) 37.3 (17.8) 38.3 (16.8) 38.6 (14.9)

4.3 (0.7) 3.5 (0.5) 3.3 (0.6) 4.5 (0.7) 3.7 (0.6) 4.2 (0.8) 3.8 (0.5) 4.9 (1.0)* 4.2 (0.6) 4.6 (0.8)* 3.8 (0.5) 4.4 (0.8) 5.2 (1.1)* 6.6 (1.2)*

19 15 18 17 16

32.1 (16.8) 32.7 (16.2) 33.1 (18.2) 33.7 (16.1) 35.2 (19.1)

3.4 (0.5) 3.5 (0.5) 3.4 (0.5) 3.4 (0.4) 3.2 (0.5)

20

33.0 (15.7) 3.5 (0.4)

transducer respectively to produce one peak force as a vector. Data were stored on a computer. Comparing disposable blades to the re-usable Mac 3 blade, ANOVA showed that the use of some plastic blades increased the duration of laryngoscopy significantly (Table 9). This study suggests that not all plastic blades can be regarded as equal to metal blades. Keywords: equipment, laryngoscopes; larynx, laryngoscopy

Reference 1 Evans A, Vaughan RS, Hall JE, et al. Anaesthesia 2003; 58: 869–73

Keywords: equipment, Bonfils laryngoscope; modelling, physical

References 1 Halligan M, Charters P. Presented to ARS Meeting in April 2004, Aberdeen 2 Horton WA, Fahy L, Charters P. Br J Anaesth1989; 62: 6–12 3 Charters P. Br J Anaesth1996; 77: 309–11

A comparison of the forces exerted by 20 laryngoscope blades during laryngoscopy S. Rassam*, J. E. Hall, J. Mecklenburgh and A. R. Wilkes University of Wales College of Medicine, Cardiff CF14 4XN, UK

Tracheal intubation is commonly used to secure the airway during anaesthesia. In recent years concerns that re-usable equipment may be a source of contamination have encouraged the increased use of disposable laryngoscope blades. It has been recommended that all new blades should be compared against standard equipment before their introduction.1 In a randomized crossover study, 20 laryngoscope blades were assessed in a manikin by 50 anaesthetists. Peak force generated and time required to achieve a grade I Cormack and Lehane view were investigated. Vertical and horizontal forces generated at laryngoscopy were measured by a mass balance and force

Assessment of 20 laryngoscope blades in intubating manikin S. Rassam*, J. E. Hall, J. Mecklenburgh and A. R. Wilkes University of Wales College of Medicine, Cardiff CF14 4XN, UK

Direct laryngoscopy is the most common technique used to visualize the larynx and intubate the trachea. In recent years, with the increasing awareness of the risks of cross infection of variant CreutzfeltJacob disease, different blades have been developed in an attempt to find an alternative to non-disposable blades. There is little evidence demonstrating the safety and efficacy of these devices compared with the non-disposable laryngoscopes blades. It has been recommended that all new blades should be compared against standard equipment before their introduction.1 In a randomized crossover study, 20 laryngoscope blades were assessed in a manikin comparing: ease of attachment of the blade to the handle, illumination, view of the larynx, and satisfaction for clinical use. Fifty anaesthetists participated and were asked to give a visual analogue score (VAS) for these attributes. Ease of attaching the handle to the blade and illumination were assessed before laryngoscopy. A manikin (Laerdal Airway Management Trainer) was used and anaesthetists were asked to perform direct laryngoscopy to achieve grade I Cormack and Lehane view. At laryngoscopy satisfaction with view (primary outcome) was

613P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

instrument is manoeuvred in normal practice. (Manipulation of the epiglottis with the instrument is less important than with a conventional Macintosh blade.) The head was represented by a simple rectangular block from which a narrow hinged strip was suspended to represent the posterior pharyngeal wall. The larynx was a simple card fixed (with ‘Blu-Tack’) onto the strip. (Variables facilitated: head/neck angulation; pharyngo-laryngeal distance and laryngeal inlet size.) Conventional dental prosthetic upper and lower jaw sets were used to complete the model. (Different sets allow for dental variations, e.g. ‘buck teeth’, empty tooth sockets, etc.) An incomplete posterior midline vertical slit cut into the upper palate allowed vertical translation when attached to the head (i.e. variation in maxillary prominence). The lower palate was set on a supporting bar, itself connected to symmetrical interchangeable fixtures to simulate variation in body and condyle length plus relative angulation. The paired temporo-mandibular joint equivalents consisted of two-position sockets attached so as to allow alterations in respect of orientation, A-P vertical height and caudal/rostral positioning relative to the head. Each joint was completed on its underside with a strap to hold the mandibular condyles in position and allow gliding between the upper and lower joint sockets. Alternative fixtures for the temporomandibular joints allowed variation in jaw protrusion and asymmetrical articulations. Overall dimensional size variations for the model were taken from historical radiological data within the department of 50 lateral X-rays of subjects who were in a standardized intubation position, all of whom were known to be normal tracheal intubations at conventional Macintosh laryngoscopy.2 It has been suggested previously that difficulty or otherwise is the net result of factors worsening vs facilitating conventional laryngoscopy.3 The model showed that the Bonfils laryngoscope should be expected to have limitations in subjects with a receded jaw, limited mouth opening, limited head extension and particularly with combinations of these factors. On the other hand, as expected, the edentulous state, certain empty tooth socket configurations and maximized head/neck extension tended to counter these effects. Detailed analyses in respect of the consequences of missing teeth combinations require only limited modifications to the model. While the tongue, epiglottis, and stylo-hyoid apparatus were omitted from this model it should be possible to add these at a latter stage to facilitate further applications including rigid blade comparisons.

Proceedings of the Anaesthetic Research Society

Table 10 Changes in flow rates and temperature output when fluid warming device is added to the fluid giving set at atmospheric pressure. Values are mean (range) percentages and  C Device

No heating Bair Hotline device Hugger

Standard Fluido Ranger

Flow rate 185 68 152 153 141 (ml/min) (184–186) (68–70) (152–154) (152–154) (140–142) Fluid temperature 23 26 31 35 35 ( C) Change in flow (%) 0
63
16
16
23

assessed. Anaesthetists’ satisfaction to use the blades on patients was also indicated. VAS for attachment was generally lower in metal blades while VAS for illumination appears to be independent of blade composition. All VAS for view of the larynx and satisfaction were however higher in metal than plastic blades (Fig. 5). From these initial data, plastic blades cannot generally be regarded as equal to metal blades and anaesthetists are more satisfied with metal blades.

In common with the Hotline and the Standard Ranger, the performance of the Bair Hugger declined further when the flow of fluid increased by pressurizing the giving set. Conversely, the Fluido appears to warm the fluid more effectively when the giving set is pressurized. However, when Fluido is tested at the minimum recommended flow rate (0.9 litre h
1) the fluid was only warmed to 31 C. The Bair Hugger offered the highest resistance to the fluid flow, effectively converting the 14 G i.v. cannula to an 18–20 G i.v. cannula. The anaesthetist should be aware that the commonly used fluid warming devices do not behave according to the manufacturer’s specifications. Keywords: equipment, fluid warming devices; fluids, i.v.

References 1 Frank SM, Fleisher LA, Breslow MJ, et al. JAMA1997; 277: 1127–34 2 Kirkbride DA, Buggy DJ. Br J Anaesth CEPD Rev 2003; 3: 24–8

Keywords: equipment, laryngoscope; larynx, laryngoscopy

Reference 1 Evans A, Vaughan RS, Hall JE, et al. Anaesthesia 2003; 58: 869–73

Psychomotor testing to detect changes in cerebral function V. Rybynok1*, P. A. Kyriacou1*, S. K. Pal2 and D. C. White2

Fluid warming devices: help or hindrance? Royal Gwent Hospital, Cardiff Road, Newport, UK

1 School of Engineering and Mathematical Sciences, City University, London, UK. 2St Andrew’s Centre for Plastic Surgery and Burns, Broomfield Hospital, Chelmsford, Essex, UK

General anaesthesia typically leads to mild core hypothermia, which has been shown to increase perioperative morbidity.1 Devices designed to warm i.v. fluids are used to prevent mild hypothermia.2 We were interested to see whether these devices warmed the fluid to body temperature and whether they offered resistance to the flow of fluid. Four fluid warming devices were investigated: Bair Hugger (Bair Hugger 241 Fluid Warming Set), Hotline (Hotline Warming System, Grasby Medical), Standard Ranger (Ranger Fluid Warming System, Augustine Medical), and Fluido (Datex-Ohmeda). An experiment based on the British Standard for determining fluid flow through i.v. cannulae was set up inthe laboratory. A standard 4 mm, 110cm giving set was connected to a 1000 ml bag of 0.9% saline suspended at a height of 100 cm. A 120 cm extension line was connected to a 14 G i.v. cannulae at the ‘patient end’. The warming devices were each introduced between the giving set and extension line. Three recordings of fluid flow were made with the bag of fluid at atmospheric pressure and at 300 mm Hg. The temperature of the fluid entering i.v. cannula was measured with a needle temperature sensor (Level 1, Rockland). None of the four devices warmed the administered fluid to body temperature (Table 10). All four devices offered resistance to fluid flow (Table 10). The fluid temperatures when the fluid bag was pressurized to 300 mm Hg were: 24 C for the Bair Hugger, 26 C for the Hotline, 31 C for the Standard Ranger, and 35 C for the Fluido. When the fluid bag was pressurized the resistance in all systems was overcome.

Previous work has suggested that atmospheric nitrogen may have a detectable effect on cerebral function.1 From the oil/gas solubility of nitrogen it can be theoretically predicted that the fractional MAC (f MAC) of nitrogen in air at 1 ATA to be 2–4% of 1 MAC. On the hypothesis that changes of cerebral function of this order may be detected by psychomotor testing a comprehensive computer program named ‘Psychom’ has been developed to perform this task based on responses to sound and visual stimuli. An IBM ·86 compatible personal computer (PC) was used for the development of the software, data acquisition and analysis. Modern PCs of this family allow the production of both sound and visual stimuli, using interface devices such as push buttons and measure the time with very high precision (up to processor frequency). The Windows NT operating system was chosen as the software platform. The programming language used for the software development was C++ with Borland C++ Builder 6, as interactive development environment (IDE). Visual component library (VCL) was used as the basis for the program. The hardware interaction between the user and the software has been accomplished via two custom-made push button switches. Technology from a standard USB/PC2 mouse controller has been used for the development of the switches. The developed system allows the operator to design, save and load different experimental protocols with windows based easy-to-use visual interface. The duration and nature (fixed or random time intervals) of both stimuli can be easily configured by the user using a drop down selection menu. During an experiment the system continually collects data

M. Turner* and I. Hodzovic

614P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

Fig 5 VAS: ease of attachment of blade to handle, illumination, satisfaction of view at laryngoscopy, satisfaction for clinical use. Blade 3 was a one-piece moulded blade to handle. Blade type (number) can be found in Table 9.

Proceedings of the Anaesthetic Research Society

Keywords: anaesthetics gases, nitrogen; brain, cerebral function; brain, psychomotor tests

regional or nerve block and eight patients received fentanyl. All patients were mechanically ventilated. Data were logged to a PC from a gas analyser (Datex Ultima-1) and two channels of a BIS A1000 monitor (Aspect Medical Inc., v3.3) with referential montage FP1, FP2, T3 (ground), T4 (reference). As a result of the known effect of surgical stimulation on BIS, we examined the recovery phase only. The 5-min epoch before the discontinuation of sevoflurane, comprising 60 measurements of BIS, was taken as a baseline and compared with successive epochs. Data were analysed using either Student’s t-test or Mann–Whitney U-test after testing for normality. All children recovered to BIS values greater than or equal to 80, with recovery times ranging from 2 min 30 s to 16 min 12 s. Seven subjects showed significant asymmetry at baseline (P<0.05) and all of these had coincident BIS values at recovery as shown in the example trace (Fig. 6). Two subjects did not show significant asymmetry at baseline or in any subsequent 5min epoch to recovery. BIS asymmetry shows a similar pattern in infants and small children at recovery as that seen previously in adults (Fig. 6). The clinical implication of this is uncertain. Keywords: children; monitoring, BIS

References

1 White DC, Lockwood GG. Br J Anaesth 2003; 91: 465P

1 Tomlinson S, Pomfrett CJD, Rolfe S, Pollard BJ. Br J Anaesth 2001; 86: 305P 2 Pomfrett CJD, Pollard BJ. Anesthesiology 2001; 86: 305 3 Bryan A, Pollard BJ, Nuwegi M, Pomfrett CJD. Br J Anaesth 2004; 92: 304–5P

BIS asymmetry during recovery from anaesthesia in children

Changes in physiological variables with propofol induction recorded by an anaesthetic information management system

B. Saha*, A. Bryan, S. Rolfe* and B. J. Pollard

N. J. N. Harper, A. Bryan, and B. J. Pollard

Department of Anaesthesia, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK

Department of Anaesthesia, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK

Using single channel BIS, adult-like readings have already been observed in children.1 BIS asymmetry, the difference between left- and right-sided recordings, has been reported in adult elective surgical patients2 and in adult patients receiving sedation in ITU.3 We have investigated whether asymmetry is also demonstrable in children. We recruited nine subjects (age range 2–56 months), undergoing elective surgery. Following inhalation induction with sevoflurane (n=5) or propofol (n=4), anaesthesia was maintained with sevoflurane in oxygen and nitrous oxide in all subjects. Six patients received

Large data sets may be used to describe the physiological effects of anaesthesia in the population. The RECALL Anaesthetic Information Management System (AIMS) automatically logs vital-signs data from the anaesthetic monitor. Previously published AIMS data have described the decrease in arterial pressure and heart rate consequent on induction with propofol in patients receiving tracheal intubation.1 We investigated the changes in heart rate (HR) and non-invasive arterial pressure (mean MBP, systolic SBP, and diastolic DBP) during propofol induction without the stimulus of tracheal intubation. HR was logged at 30-s intervals

Reference

Fig 6 Change in left- and right-sided BIS at recovery in a 3-month-old infant.

615P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

and performs basic statistical analysis corresponding to the subject’s visual and/or sound responses. The statistical results can be viewed at the end of the experiment using a specially designed adjustable viewer. Also, during acquisition, the data are saved in a file in a format that can be easily accessed and used for further offline processing and analysis by conventional commercial statistical packages. Preliminary laboratory tests were carried out on healthy volunteers to evaluate the functionality of the system. An experimental protocol has been suggested and implemented. This protocol allows the generation of sound and visual stimuli to run simultaneously at random intervals (1–3 s) for a period of 3 min. Two volunteers, one that is accustomed and one that is not accustomed to drinking alcohol, were selected to perform the tests before and after the intake of 100 ml of 40% alcohol. The performance of the system was successful and statistical results were automatically generated at the end of the test. The reaction times for visual stimuli before and after the intake of alcohol were 393 and 560 ms correspondingly. Similarly for the sound stimuli the reaction times were 433 and 602 ms. In the subject accustomed to alcohol no significant effect was found. These preliminary encouraging results confirm the successful operation of this new software system and suggest further evaluation on more rigorous clinical studies.

Proceedings of the Anaesthetic Research Society

and NIBP at a minimum of 3 min intervals. Anaesthesia was maintained with nitrous oxide and sevoflurane or isoflurane in oxygen via a laryngeal mask airway (LMA
){. Vital signs data from 447 patients were uploaded from the RECALL database via structured query language (SQL) to Microsoft Access. Pooled values during the 5-min period before induction and during the 5-min period after induction were analysed using Student’s t-test or Mann–Whitney U-test (Fig. 7). There was no surgical stimulation during this period and the patients remained in the anaesthetic room. There was a significant reduction in all parameters (P<0.0001): HR declined from 79.6 to 77.7 beats min
1; SBP from 139.4 to 128.18 mm Hg; MBP from 111.1 to 101.0 mm Hg and DBP from 82.9 to 75.9 mm Hg. The haemodynamic changes were not maximal until approximately 10 min after induction of anaesthesia. The heart rate demonstrated greater fluctuations than the arterial pressure: in particular, before and shortly after induction of anaesthesia. These findings may be important in establishing a benchmark against which haemodynamic changes in a particular patient or group of patients might be compared against the expected changes in the population. Keywords: anaesthetics i.v., propofol arterial pressure; heart rate

Reference 1 Benson M, Junger A, Fuchs C, et al. J Clin Monit Comput 2000; 16: 183–90

Microalbuminuria as a predictor of mortality in burns patients W. S. Yew* D. King*, R. Deroy*, P. Smith* and S. K. Pal St Andrew’s Centre for Plastic Surgery and Burns, Broomfield Hospital, Chelmsford, Essex CM1 7ET, UK

Microalbuminuria has been used as a reflection of the increased capillary permeability often seen in the systemic inflammatory response syndrome.1 Expressed as a ratio of urinary albumin-creatinine ratio (ACR), it has been found to be predictive of outcome in patients in intensive care units (ITU). In a study of 55 patients, De Gaudio found that ACR correlated well with Sepsis-related Organ Failure assessment scores.2 Similarly, Gosling showed that

LMA
is the property of Intavent Limited.

{

Table 11 A comparison of patient parameters with outcome. Values expressed as mean (SD), significance assessed using the non-parametric Mann–Whitney U-test % BURNS

ACMAX ACMIN ACADM Length of (mg mmol
1) (mg mmol
1) (mg mmol
1) ITU stay

Survived 38.0 (19.9) 19.3 (38.9) Mortalities 60.0 (27.7) 98.5 (78.8) Significance 0.034 0.020 (P)

1.4 (1.1) 1.1 (0.7) 0.722

8.4 (20.0) 22.0 (53.8) 0.528

24.6 (17.0) 17.0 (9.4) 0.622

elevated ACR on admission predicted death in surgical, trauma and burns patients as well as the Acute Physiological and Chronic Health Evaluation (APACHE II) scores.3 However, only 11 patients with burns were included in his study. Hence, we explored the utility of ACR as a predictor of outcome in burns patients. A retrospective audit was performed on patients with major burns admitted to the ITU over a period of 18 months. Inclusion criteria were adults with thermal burns; patients with skin loss as a result of non-thermal processes like Scalded Skin Syndrome were excluded. ACR was routinely measured daily in all patients and on admission. Admission (ACADM), maximum (ACMAX), and minimum (ACMIN) ACR were recorded. For this analysis the outcome was hospital mortality. Data were analysed using SPSS for Windows v11. Of the 50 patients analysed there were seven mortalities. Patient characteristics between survivors and mortalities were similar. We found that % BURNS and ACMAX but not ACADM, ACMIN, or ICUSTAY, significantly differed according to outcome (Table 11). The non-parametric correlation coefficient was calculated for each factor with respect to mortality. Significant correlations between mortality and ACMAX (P=0.003) and % Burns (P=0.033) were observed with the former having a greater coefficient (Spearman’s rho 0.415, compare with 0.302). We conclude that the importance of ACMAX is comparable, if not greater, with that of percentage burns in non-parametric correlation with mortality, based on this pilot retrospective study. Acknowledgements: We would like to thank Mr Keith Gibson and Ms Karen Cook for their invaluable help. Keywords: complications, albuminuria; complications, burns

616P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

Fig 7 Changes in arterial pressure and heart rate during propofol induction.

Proceedings of the Anaesthetic Research Society

References

References

1 Evans G, Greaves I. BMJ 1998; 318: 207–8 2 De Gaudio AR, Adembri C, Grechi S, et al. Intensive Care Med. 2003; 26: 1364–8 3 Gosling P, Brudney S, McGrath L, et al. Crit Care Med 2003; 31: 98–103

1 Tyndale RF, Sellers EM. Ther Drug Monit 2002; 24: 163–71 2 Sellers EM, Kaplan HL, Tyndale RF. Clin Pharmacol Ther 2000; 68: 35–43 3 Merkel U, Sigusch H, Hoffman A. Eur J Pharmacol 1994; 46: 175–7

Ethnic differences in thermal pain responses between volunteers of South Asian origin and white British

J. Berrington*, M. Grayling* and B. P. Sweeney

R. K. Latif*, P. J. Watson and D. J. Rowbotham

Poole Hospital NHS Trust, Longfleet Road, Poole, Dorset BH15 2JB, UK

Department of Health Sciences, Division of Anaesthesia, Critical Care and Pain Management, University of Leicester, Leicester Royal Infirmary, Leicester, UK

Genetic variants of the hepatic cytochrome P450 enzyme CYP2A6 (coumarin hydroxylase), which is responsible for the majority of nicotine inactivation to cotinine, are related to both propensity to nicotine addiction and the level of cigarette consumption.1 It is possible to mimic these genetic variants using enzyme blockers such as methoxsalen and tranylcypromine and thus decrease the requirements for nicotine.2 Grapefruit juice (GFJ) contains bio-flavinoids that inhibit a number of CYP enzymes including CYP3A4 but more importantly CYP2A6.3 We postulated that ingestion of GFJ might alter nicotine metabolism and thereby reduce cigarette consumption by addicts. With ethics committee approval and informed consent, four healthy non-smoking volunteers (1:1, M:F) were asked to refrain from drinking citrus fruits for 48 h before the study. After a 12-h fast, each volunteer drank 1 litre of water and then chewed 4 mg nicotine gum (Nicorette
) over a 3-h period. Blood samples were then taken at 0, 30, 60, 90, 120, and 180 min and analysed for nicotine and cotinine levels. The study was then repeated using the same subject as a control, using 1 litre of Tropicana
GFJ instead of water. No statistically significant difference was found (Wilkinson Signed Rank test) in either nicotine to cotinine ratios or cotinine plasma concentrations between the two groups; however, there was a significant reduction in nicotine plasma concentrations in the GFJ group (Fig. 8). GFJ does not appear to alter nicotine metabolism in healthy nonsmokers but does reduce nicotine plasma concentration. The mechanism for this unexpected finding is unclear but may represent altered buccal nicotine absorption. Acknowledgement: The Association of Anaesthetists of Great Britain and Ireland kindly provided a research grant for the purchasing of materials and laboratory testing. Keywords: nicotine, metabolism; grapefruit juice

The expression and report of pain is influenced by social environment and culture.1 Previous studies have suggested ethnically determined differences in report of pain threshold, intensity and affect.2 However, the influence of ethnic differences between white British and South Asians has remained unexplored. This study examines the influence of ethnicity on pain, comparing first generation South Asian professionals with white British professionals using thermal stimuli. Twenty age-matched, male volunteers were included in each group. The South Asian subjects were all born and educated in a South Asian country and the white British group were all born and educated in the UK. All were educated to at least degree standard. Following ethical approval, each group underwent evaluation of cold and warm perception, cold and heat pain threshold, magnitude estimation of thermal pain unpleasantness, and pain intensity using a computer controlled Peltier element with a 30·30 mm-contact surface probe. Cold and warm perception and cold and heat threshold were assessed using an ascending method of limits from the holding temperature of 32 C, the temperature increased or decreased at a rate of 1 C s
1 for cold and warm perception, and 1.5 C s
1 for cold and heat threshold. The subject responded by pressing the button to stop the stimuli. For thermal perception, the subject pressed the button as soon as he felt the temperature change and for thermal threshold as soon as the stimulus started to become painful. Magnitude estimation of pain unpleasantness and pain intensity was measured using an electronic numerical scale (0–100). Thermal stimuli of 46, 47, 48, and 49 C were delivered from the holding temperature of 32 C with a rate of temperature increase of 4 C s
1. Each stimulus was delivered for 5 s with inter-stimulus interval of 30 s. Data were analysed using analysis of variance and t-test as appropriate. There was a statistically significant difference between the two groups for heat pain threshold (P=0.004) and heat pain intensity at 46 (P=0.004), 47 (P=0.001), 48 (P=0.001), and 49 C (P=0.001). Although no group differences emerged for cold pain threshold and heat unpleasantness between the two groups, first generation South Asians demonstrated lower cold pain threshold and more unpleasantness at 46, 47, 48, and 49 C but this was not statistically significant. There was no difference in cold and warm perception between the two groups. Our study shows that ethnicity plays an important role in heat pain threshold and pain report, South Asian males demonstrated lower pain thresholds and higher pain report when compared with matched white British males. Keywords: ethnic factors; pain, threshold

References Fig 8 Box-plot of within-subject, between-group differences (water minus grapefruit) for plasma nicotine concentration. *P=0.002.

1 Zborowski M. J Soc Issues 1952; 8: 16–30 2 Zatzick DF, Dimsdale JE. Psychosom Med 1990; 52: 544–57

617P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

A pilot study of the effect of grapefruit juice on CYP2A6-related nicotine metabolism

Proceedings of the Anaesthetic Research Society

Efficacy of cyclizine in combination with dexamethasone in the prevention of postoperative nausea and vomiting

Table 12 Incidence of nausea in different groups

J. M. Alexander-Williams, J. Radhakrishnan* and S. Kemp* MidEssex Hospitals NHS Trust, Broomfield Hospital, Chelmsford, Essex, UK

Cyclizine (n=50)

Combination (n=52)

Placebo (n=27)

P

Nausea anytime Moderate to severe nausea Overnight admission

12 5 2

16 9 4

16 11 0

0.006 0.007 0.29

travel sickness, time since previous period, smoking history, or anaesthetic time. Patients receiving cyclizine had lesser incidence of nausea and overnight stay than the combination group and placebo. The latter two had comparable incidences of nausea. The NNT for cyclizine was 2.8 while that for combination drugs was 3.5 (Table 12). Acknowledgement: Grant awarded by CMERT (Chelmsford medical education and Research trust). Keywords: antiemetic, cyclizine; vomiting, PONV

References 1 Watts SA. Anaesth Int Care1996; 24: 546–51

618P

Downloaded from http://bja.oxfordjournals.org/ at University of Massachusetts Medical School on April 10, 2015

Up to 50% of patients undergoing day-case laparoscopic gynaecological procedures have postoperative nausea and vomiting (PONV).1 Cyclizine is commonly used as an antiemetic and has proved effective in this setting either alone or in combination with ondansetron.2 We chose to evaluate the effectiveness of cyclizine and dexamethasone combination on PONV in day-case laprascopic gynaecological procedures. In a randomized, double blind trial, the efficacy of cyclizine 50 mg was compared with a combination of cyclizine 50 mg and dexamethasone 8 mg and placebo in 150 patients undergoing day-case gynaecological surgery. The groups were comparable in terms of age and other risk factors. Patients were anaesthetized with propofol, fentanyl and inhaled agents. Nausea, vomiting, and pain were monitored during the first 2 h in recovery and after 24 h. Twenty-one patients were excluded due to: high analgesic requirement, postoperative protocal failure, loss of follow up. There was no difference between the groups in age, PONV history,

Variable