Proceedings of the Anaesthetic Research Society Meeting

British Journal of Anaesthesia 94 (3): 399–411P (2005)

doi:10.1093/bja/aei032

ABSTRACTS Proceedings of the Anaesthetic Research Society Meeting Weetwood Hall Hotel, Otley Road, Leeds December 2–3, 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.)

Effect of table tilt on maternal cardiac output using suprasternal Doppler ultrasound P. T. Selvan*, J. Bray*, R. Fernando and M. O. Columb Department of Anaesthesia, Royal Free Hospital, Pond Street, London NW3 2QG, UK

Our data suggest that the 15 L position, the position most commonly used for Caesarean section, provides a CO similar to the full lateral positions without hypotension. We recommend adopting this position routinely in full-term pregnant patients. Keywords: heart, cardiac output; measurement techniques, Doppler ultrasound; pregnancy, position

Reference

Aortocaval compression reduces maternal cardiac output (CO) and leads to the supine hypotension syndrome, an event that is mostly concealed but symptomatic only in 10% of the pregnant population.1 Although this is thought to be minimized by using table tilt during Caesarean section, very little is known about the effect of varying degrees of table tilt on maternal CO. Therefore, we used suprasternal Doppler ultrasound (SupraQ) to measure various linear and volumetric Doppler indices, including maternal CO and corrected flow time (FTc), a measure of intravascular volume, in different table tilt positions to find out the optimal position to adopt during Caesarean section. After ethics approval and informed consent, we recruited 16 healthy term singleton pregnant volunteers who were asked to adopt four different positions for 5 min in random order; full left lateral (90 LL), full right lateral (90 RL), and supine with either 7.5 (7.5 L) or 15 (15 L) degrees of left table tilt. During this period the following observations were recorded; heart rate (HR), systolic blood pressure (SBP), oxygen saturation (SpO2) and Doppler measures including CO, FTc, stroke volume (SV), and stroke distance (SD). Statistical analyses included repeated measures analysis of variance and Tukey–Kramer post-tests with P<0.05 defined as significant. CO, SV, SD, and FTc were significantly lower in the 7.5 L position compared with other positions, which were essentially similar (Table 1). The CO in the 15 L position was similar to the full lateral positions. The 15 L position also had the highest SBP.

1 Bamber JH, Dresner M. Anesth Analg 2003; 97: 256–8

Investigation of buprenorphine-induced respiratory depression in anaesthetized patients and its reversibility V. Mehta*, J. P. Phillips*, A. C. Wantman*, S. H. Ratcliffe*, P. A. van Raders* and R. M. Langford Anaesthetic Laboratory, St Bartholomew’s Hospital, London EC1A 7BE, UK

Buprenorphine administered transdermally has created renewed interest in its receptor binding and respiratory profile. This double blind, randomized controlled study was conducted in anaesthetized, spontaneously breathing patients before surgery. With ethics committee approval and written informed consent, 54 ASA grade I–II adults for elective surgery were randomized to receive buprenorphine 1.4, 2.8, or 5.6 mg kg 1 (12 patients each), morphine 0.07 mg kg 1 (12 patients) or placebo (six patients). Following propofol, anaesthesia was maintained with oxygen, nitrous oxide, and sevoflurane 1.5 MAC, and a laryngeal mask airway was inserted, with no clinically discernible leak. Once breathing spontaneously, the patient was connected to a breathing system incorporating a purpose-built, previously validated wet wedge spirometer.1 After observing steady respiratory function for 10 min, the

Table 1 Cardiovascular changes during table tilt. Data are presented as mean (SD). See text for abbreviations. *P<0.05 compared with 7.5 L lateral tilt and **compared with 90 LL and 90 RL positions Position

HR (beats min 1) P = 0.06

SBP (mm Hg) P<0.0001

CO (litre min 1) P = 0.0037

SV (ml) P = 0.0008

SD (cm) P = 0.0013

FTc (ms) P = 0.00014

90 RL 7.5 L 15 L 90 LL

83.8 80.7 81.9 80.1

100.9 (13.61) 114.6** (14.34) 117.5** (15.82) 100.7 (16.32)

4.77* (0.777) 4.06 (0.712) 4.62* (1.159) 4.53 (1.034)

59.8* (8.40) 51.1 (8.64) 58.2* (11.82) 59.4* (11.44)

17.4* (2.94) 14.7 (2.30) 16.9* (3.19) 17.2* (3.14)

372.7* (32.10) 334.4 (29.37) 370.5* (40.66) 378.1* (26.27)

(10.62) (9.29) (10.97) (9.95)

# The Board of Management and Trustees of the British Journal of Anaesthesia 2005. All rights reserved. For Permissions, please e-mail: [email protected]

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

VERBAL PRESENTATIONS

Proceedings of the Anaesthetic Research Society

25

n=5, P=0.099 (Placebo) n=7, P=0.005 (Morphine)

20 10 0 –10

n=7, P=0.099 (Bup.No.Nalox.) n=8, P=0.002 n=9, P=0.002 n=11, P=0.014

–20 –30 –40 –50

Bup. 2.8 µg kg–1 Morphine Bup. No Nalox.

VDalv/VTalv (%)

Change in minute volume (%)

Bup. 1.4 µg kg–1 Bup. 5.6 µg kg–1 Placebo

0

20 10 Time after study drug (min)

20

30% shunt 20% shunt 10% shunt 0% shunt

15 10 5 0 150

200 250 VCO2 (ml min–1)

300

Fig 2 The influence of VCO2 on alveolar deadspace fraction (small VQ defect).

30

study drug was injected. Respiratory depression (ventilatory frequency decrease >33%, or end-tidal CO2 increase >0.75 kPa from baseline values) within the first 10 min was treated with up to three incremental doses of naloxone 0.4, 0.8, and 1.2 mg, given 5 min apart, until respiratory function was better than the criteria detailed above. Recordings were continued for 20 min after opioid study drug administration. All four opioid groups showed a statistically significant decrease in minute volume (P<0.01, t-test, paired two-tail) (Fig. 1). Thirtyfive patients received naloxone (mean time 6.05 min), after which all groups showed significant (P<0.05) increase in minute volume, albeit partial recovery in the buprenorphine patients. Buprenorphine patients not given naloxone showed a further significant decrease after 6.05 min in minute volume (P<0.05). Keywords: analgesics opioid, buprenorphine; analgesics opioid, morphine; ventilation, respiratory depression

(VQ) defect was added to the modelled subject, mean (SD) VQ ratio 1.38 (0.77). We increased VCO2 from 150 to 300 ml min 1 while keeping VO2 constant. We repeated this protocol with the addition of Qs/Qt (10, 20, and 30%). For each increase in Qs/Qt we increased cardiac output to keep VQ configuration constant. The experiment was repeated with further configurations typical of critically ill patients’ VQ defects – mean (SD) VQ 2.15 (1.81), and 3.08 (2.86). Increases in VCO2 were associated with increases in VDalv/VTalv when an anatomical shunt or a VQ defect was present (Fig. 2). The increase in VDalv/VTalv associated with increasing VCO2 was directly related to the magnitude of Qs/Qt. When Qs/Qt was zero, and no VQ defect was present, VCO2 had no effect on VDalv/VTalv. The dependency of VDalv/VTalv upon Qs/Qt appears to be a result of the transmission of carbon dioxide past ventilated alveoli without gas-exchange. This increases the disparity between expired and arterial PCO2. Future work requires the production of a MIGET (Multiple Inert Gas Elimination Technique) model to assess the impact of shunting on alveolar deadspace measured using MIGET. Keywords: airway, deadspace; carbon dioxide; lung, shunt

References 1 Banks A, Moppett IK, Fletcher R, Hardman JG. Br J Anaesth 2004; 93: 160–1P 2 Hardman JG, Wills JS, Aitkenhead AR. Anesth Analg 2000; 90: 619–24

Reference 1 O’Connor SA, Wilkinson DJ, Dickson GR, et al. Med Biol Eng Comput 1985; 23: 258–62

The interaction of carbon dioxide production and anatomical shunt on the Bohr–Fowler alveolar deadspace fraction A. Banks* and J. G. Hardman University Department of Anaesthesia, University Hospital, Nottingham NG7 2UH, UK

We showed previously that alveolar deadspace (measured using the Bohr–Fowler techniques, and expressed as a fraction of alveolar tidal volume: VDalv/VTalv) is influenced by pulmonary shunt fraction (Qs/Qt).1 However, the mechanism of this interaction is unclear. To elucidate the impact of Qs/Qt on VDalv/VTalv, we used the Nottingham Physiology Simulator2 to examine the effect of varying carbon dioxide production (VCO2) on VDalv/VTalv in the presence of varying Qs/Qt. The simulator was configured with constant ventilatory frequency, tidal volume, and inspired oxygen fraction (0.21). Hypoxic pulmonary vasoconstriction was disabled. A small ventilation-perfusion

Acute breathlessness measured by clamping the end-tidal carbon dioxide E. Pickering1*, M. S. Nazir2*, A. Guz2*, S. Semple2*, K. Murphy2* and A. Holdcroft1 1

Magill Department of Anaesthetics and the 2Department of Respiratory Medicine, Imperial College London, London, UK

Breathlessness is a common sensation that depends on central processing through respiratory and non-respiratory neural activity. In the search for methods to measure drug effects on breathlessness we determined to maintain spontaneous ventilation and to avoid the use of exercise as a stimulus to breathing. A breathing system was therefore chosen that holds arterial blood gases constant while at the same time ventilation is changing.1 The aim of the present study was to determine if this system could be developed in order to establish two different end-tidal PCO2 without maximum ventilatory effort and maintaining normal oxygenation. We hypothesized that the two levels of increased end tidal PCO2 could elicit stable scores of breathlessness and that these scores would be repeatable at different times for the same PCO2. In a pilot study four volunteers were monitored for oxygen saturation, ECG, and heart rate while breathing in through a mouthpiece

400P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

Fig 1 The mean percentage changes in minute volumes for the patients in placebo, morphine, and buprenorphine groups who required naloxone and for the buprenorphine patient group, which did not require naloxone. P values are for post- compared with pre-naloxone, and for the buprenorphine group not in receipt of naloxone, for 20 vs 6.05 min.

Proceedings of the Anaesthetic Research Society A

60

CO2 (mm Hg)

50 40 30 20 10 0 2566 2721 2863 3010 3157 3305 3454 Time (s) B

60 40 30 20 10 0 4024 4161 4284 4413 4538 4653 4775 Time (s)

Fig 3 Typical first and second traces in one subject with 10 min rest between. (A) The upper trace shows the end tidal PCO2 and (B) the lower, the inspired PCO2.

a filtered mixture of oxygen, carbon dioxide, and nitrogen in concentrations determined by a blender and breathing out through a large tube acting as an alveolar gas reservoir followed by a one-way expiratory valve in parallel with a loaded inspiratory valve. Ventilation was monitored using a pneumotachograph in the expiratory limb. The inspired and expired PCO2 was measured at the mouth. Once the subject was familiar with the breathing system they were introduced to an electronic breathlessness score. A light was moved between seven verbal scores from 0 = no breathlessness to 7 = extreme breathlessness.2 Scores were measured every 30 s throughout the study. Measures were recorded for two PCO2 levels after an initial trial run. The breathlessness scores were similar at the same PCO2 but significantly different between the two PCO2 levels (P<0.01 [Mann–Whitney U-test]). In this small group of subjects (Fig. 3) the stability of the breathing system and the repeatability of the scores at the different PCO2 levels indicate that the method may be applicable to testing in patients.

Acknowledgement: The authors wish to thank the Department of Anatomy at University of Liverpool for providing the mandible specimens. Keywords: model, mandible

Acknowledgement: We thank the Wellcome Trust for a Vacation Scholarship for MN.

References

Keywords: carbon dioxide; partial pressure, end-tidal; respiration, breathlessness

1 Rodrigues MAF, Gillies DF, Charters P. Comput Methods Biomech Biomed Engineer 2001; 4: 127–48 2 Lam YF, Gillies DF, Rueckert D, et al. Br J Anaesth 2003; 91: 467–8P

References 1 Banzett RB, Garcia RT, Moosavi SH. J Appl Physiol 2000; 88: 1597–1600 2 Lansing RW, Moosavi SH, Banzerr RB. Resp Phys Neurobiol 2003; 134: 77–83

Use of a multi-scenario questionnaire to investigate management option considerations by anaesthetists for difficult airway/difficult laryngoscopy

A shape model of the human mandible

A. Cherian* and P. Charters

Y. F. Lam1*, D. D. Boswell1*, D. F. Gillies1*, P. Edwards1*, D. Rueckert1*, P. Charters2 and P. Groom2

Department of Anaesthesia, University Hospital Aintree, Liverpool, UK

1

Department of Computing, Imperial College London, UK. Department of Anaesthesia, University Hospital Aintree, Liverpool, UK

2

We have developed a shape model of the human mandible in which the main modes of variation can be visualized in three dimensions and studied. The model is controlled by a small number of

Anaesthetists are concerned about anaesthetized patients with a difficult airway. For difficult laryngoscopy, perhaps what matters most is whether the larynx can be visualized. Unambiguous communication can be a problem when patients have had problems with either. A scenario questionnaire was designed to explore how

401P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

CO2 (mm Hg)

50

parameters. It will be of use in obtaining a better understanding of the geometry of the mandible and the shape factors that contribute to difficult intubation. We hope to combine the model with our earlier work on finite element modelling of the tongue,1 to produce a full model of the human upper airway, which can be incorporated in computer simulations for medical education, training and preoperative prediction of difficult laryngoscopy. CT scans of 10 dry human mandible specimens were used to build the model. The images were segmented by hand to remove imaging artifacts and to eliminate the internal air spaces. One of the 10 sets was then selected as a base, and the other nine were registered to it.2 The registration process deforms an image set so that it matches the base set. The correspondence between each surface point of the image set and the base image set can then be found. The surface geometry of the base image set was found by first using the marching cubes algorithm, which provides a fine triangulation of the surface. The set of triangles produced by this method was then decimated using standard techniques to find a small set of triangles that represents the geometry accurately. Each coordinate of each vertex of this set was taken as a variable and principal component analysis was used to find the main modes of variation in the set. The principal components can be used to reconstruct any mandible shape that is a linear combination of the others. The largest mode of variation in the data sets was a result of the different sizes of the specimens. However, shape changes, such as elongation, could also be observed. Some smaller modes of variation were caused by the shape, position, or absence of teeth. The work clearly demonstrates the potential of shape models to encode the geometry of the human mandible in a quantitative form suitable for use in accurate simulation of the human airway. However, the results are preliminary and further work is being undertaken on a number of aspects. These include increasing the sample size, which will increase the degree to which the model can represent the whole population and segmenting the teeth from the image sets to prevent their masking more interesting modes of variation. In the long term we plan to investigate whether there are simple external measurements that can be used to determine the mandible shape with reasonable accuracy without the need for scanning a patient.

Proceedings of the Anaesthetic Research Society

Keywords: anaesthesia, audit, questionnaire; complications, difficult airway

1.0

Female Male

Probablity pain

0.8 0.6 0.4 0.2 0.0 40

42

44 46 Temperature (°C)

48

Fig 4 Gender differences in probability of reporting pain at different temperature stimuli.

applicable to clinical interpretation by using classical dose–effect techniques that are appropriate to biological systems. In choosing a well-researched stimulus known to produce the weakest effect for sex differences we tested if this new technique would be sufficiently robust. Using a heat stimulus our objective was to determine the dose–response curves to a standardized random sequence of heat stimuli in males and females using a fixed choice method. The aim was to determine if sex and site specific differences could be detected. Twenty male and twenty female unpaid subjects were recruited; their inclusion criteria checked (18–40 yr, healthy not on medication, and excluding any sex-related illness or positive psychiatric history) and informed written consent was obtained. Subjects were asked to refrain for 24 h from consuming alcohol and nicotine before the study. The tests were carried out in a dedicated psychophysics laboratory. Thermal heat pain has been extensively studied, is highly controllable, objective, repeatable, and safe. Using a computercontrolled thermode (Somedic) with a Peltier-based stimulator (20·25-mm probe), temperatures were selected in 1 C intervals from 40 to 49 C. A forced choice was made to detect sensitivity for pain, that is pain ‘yes’ or ‘no’. The stimuli were repeated three times at six sites (right and left ventral forearms, lower abdomen, and lower back). Models were fitted to the pooled response for each stimulus with site as a within group factor. The data were analysed using logistic regression. A significant difference between sexes was observed at higher temperatures (48–49 C) for all sites (P<0.001). Women were more likely to report pain at these higher levels than men. The graph (Fig. 4) shows the probability of reporting pain (with 95% CI) over the range of 40–49 C when the stimulus is applied to the arms. Acknowledgement: We thank the ARS for a vacation scholarship for Aakta Patel. Keywords: pain, gender differences

Sex differences in pain: a new methodological approach 1

2

1

A. Patel *, E. Keogh *, D. Stephens* and A. Holdcroft

1 Departments of Anaesthesia and Mathematics, Imperial College London, London, UK. 2Department of Psychology, University of Bath, Bath, UK

In humans, a wide range of experimental studies reveal that sex differences in nociception occur but their direction may differ with the type and site of stimulus.1 Female nociceptive sensitivity may be larger for pressure and electrical skin stimuli than for thermal stimuli. However, the effect size may depend on the site of stimulus, for example lower abdomen or arm, and end point, that is whether thresholds or tolerance are being measured. These types of psychophysical techniques are now being used to study therapeutic efficacy.2 It was our postulate that a refinement of the methodology used to detect sex differences in pain may make experiments more

References 1 Riley JL 3rd, Robinson ME, Wise EA, Myers CD, Fillingim RB. Pain 1998; 74: 181–7 2 Fillingim RB, Ness TJ, Glover TL, et al. Anesthesiology 2004; 100: 1263–70

Measurement of motor block: evaluation of a dedicated dorsi flexion device in patients undergoing neuraxial block A. P. Syndercombe*, N. Haslam, M. Broadhurst* and J. E. Duggan Department of Anaesthesia, Wansbeck General Hospital, Ashington, UK

Motor block is an important feature of regional anaesthesia, but is difficult to measure. We previously described a device measuring

402P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

different grades of anaesthetists would formulate outline management plans for a fictitious patient who appeared entirely normal on preoperative bedside testing despite having been told he was a ‘difficulty airway’ at his previous anaesthetic. The relevant anaesthetic record was presented as a set of alternatives. At laryngoscopy only the epiglottis was seen and described as: A, ‘floppy epiglottis’ or B, ‘epiglottis seemed fixed on the posterior pharyngeal wall’. The contemporaneous airway maintenance was noted to be either: 1, easy; 2, difficult made easy with airway aid; 3, difficult but made only ‘adequate’ even after use of airway aid, or 4, impossible. Overall, eight combinations were possible. There was no documentation as to how intubation was achieved. On the basis of this information (only) the anaesthetists were asked to consider how they would approach each situation if now an anaesthetic where tracheal intubation with muscle relaxation was required. They were to consider three management options: X, carry on and paralyse without further thought; Y, induce anaesthesia, test the ventilation with bag and mask, then paralyse only if airway adequately maintained, or Z, resort to another method to intubate and only paralyse when the patient is intubated. They also indicated whether immediately available specialist help would be required. The grade of anaesthetist (and years of experience for trainees) was noted. Personal clinical experience of laryngoscopy findings of A and B (described above) was reported as: up to 5, 6–10, 11–50, or more than 50. Free text comments about the study were encouraged. Twenty colleagues (10 trainees/10 consultants) participated. Four of the consultants regularly anaesthetize for head and neck surgery. The maximum turnaround time for questionnaire completion was 8 h. In general, the seemingly fixed epiglottis was recognized as more serious than the floppy epiglottis. The mode value for personal clinical experience of floppy epiglottis was greater (‘11–50’ in 7/20, four consultants) than for fixed (‘up to 5’ in 11/20, six consultants). However, four of 10 trainees reported ‘11–50’ experiences of fixed epiglottis. We were concerned that six of 10 trainees implied that they would proceed without help where a floppy epiglottis was associated with an airway described as ‘adequate’ at best and four of 10 would carry on in the same way if the epiglottis was fixed with similar airway compromise. In one instance the trainee reported that no help would be asked for under any of these circumstances. As a result of analysing the findings, this study was followed up by discussion within the department and counselling of relevant individuals. Use of the scenario questionnaire appeared to be worthwhile and with careful attention to detail, materials similar in structure may prove valuable as a teaching and learning aid.

Proceedings of the Anaesthetic Research Society

Simplified reconstruction of a typical TRAM flap

175

150 Simplified arterial network

Mean force (newtons)

125

100 Grp3P Grp6P

75

Grp3H

50

Fig 6 Streamlines showing flow through a 70% stenosis created by

an end-to-end anastomosis. 0 0

1

2

3

4 5 6 7 Time (min)

8

1.6E-01

9 10

1.2E-01

muscle strength at the ankle joint1 and in volunteers showed dorsi flexion has advantages over plantar flexion: involves fewer myotomes, muscles do not act across the knee joint, less affected by foot position, and the heel tends to self-locate. We have constructed a dedicated dorsi flexion device based on a deformation load cell (Amber Instruments SP41) incorporated into an orthopaedic immobilization boot (Pro-Medics Range Walker). We set out to evaluate the device in patients with evolving neuraxial block using three doseequivalent bupivacaine regimes in groups of 10: 3 ml hyperbaric bupivacaine 0.5% (Grp3H); 3 ml plain bupivacaine 0.5% (Grp3P); 6 ml plain bupivacaine 0.25% (Grp6P). We performed spinal injection under standardized conditions: sitting, L3/4; 24G Sprotte; single operator; injection rate 0.5 ml s 1. We measured muscle strength following spinal injection at 30-s intervals for 10 min. Offset of motor block was evaluated in Grp3P by measuring strength at 15-min intervals after surgery was complete. The strength of dorsi flexion (mean [SD]; Newton) before spinal injection was: Grp3H = 142 (51)N; Grp3P = 140 (50)N; Grp6P = 139 (36)N. The reduction in strength of dorsi flexion is shown in Figure 5. Onset followed a single-phase exponential decay function (R2>0.99). The strength of dorsi flexion 10 min after spinal injection was: Grp3H = 15.4 (13.5)N; Grp3P = 7.6 (8.6)N; Grp6P = 2.8 (4.4)N. Recovery of motor function began between 135–240 min and the time to 25 and 75% recovery ranged from 150 to 260 min and 178–300 min, respectively. Once recovery began the rate of recovery was fast and the time (mean [range]) taken for recovery from 25 to 75% was 53 [40–70] min. Our subjects could use the device successfully even with profound sensory and motor changes. There was considerable variability in the density of motor block between subjects, despite controlling for many factors affecting spread. We believe the device is useful in the study of central neural and sciatic nerve block.

Reference 1 Haslam N, Broadhurst M, Duggan J. Anaesthesia 2001; 56: 1174–7

Sheer stress

Fig 5 Mean (SE) dorsiflexion forces after 15 mg spinal bupivacaine.

Keywords: motor block, quantification

70% stenosis 20% stenosis No stenosis

1.4E-01

1.0E-01 8.0E-02 6.0E-02 4.0E-02 2.0E-02 0.0E+00 −2.0E-02

0

10

20

30

40

50

60

70

Position along upper wall Fig 7 Sheer stress distributions along the upper wall for various

degrees of stenosis.

Effect of anastomosis on fluid flow through the arterial network of a rectus abdominis free flap S. K. Pal1, S. Patel2 and D. Drikakis2* 1 St Andrew’s Centre for Plastic Surgery and Burns, Broomfield Hospital, Chelmsford, Essex, UK. 2Cranfield University, School of Engineering, Fluid Mechanics and Computational Science Group, Bedfordshire, UK

The deep inferior epigastric arteries (DIEAs) via rectus abdominis musclocutaneous perforators supply directly most of the abdominal wall.1 We have constructed a simplified model of an arterial network consisting of an axial perforator and its connecting axial flap vessels2 (Fig. 6) and have computationally investigated fluid flow through this network using advanced computational fluid dynamics methods3 based on high-resolution numerical schemes. Questions have been raised concerning whether the anastomosis performed during reconstructive surgery can affect the fluid flow within the arterial network to an extent in which partial or full necrosis of the tissue flap may occur. The focus of this investigation is to try and elaborate on this area of concern. We have investigated an end-to-end type anastomosis consisting of various degrees of stenosis at the point of connection. The investigation has shown that the effect on fluid flow

403P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

25

Proceedings of the Anaesthetic Research Society

as a result of this type of anastomosis is localized to the area behind the stenosis. Depending on the degree of stenosis separation bubbles form behind the stenosis on the upper and lower walls of the artery. Figure 7 shows an extreme case where the artery has a 70% stenosis. The figure is a snapshot of the fluid flow at the peak of the pulsatile cycle. Two asymmetric recirculation regions appear on the upper and lower walls. Figure 7 shows the calculated sheer stress along the upper wall corresponding to a 20 and a 70% degree of stenosis as well as for a channel without any stenosis. An increase in sheer stress occurs directly after the stenosis. The degree of this increase in sheer stress is directly related to the degree of stenosis. Finally, it was found that there was no effect from the stenosis on the sheer stress in the downstream network thus the effect of the stenosis is local. Keywords: blood, supply; muscle, free flap; surgery, anastomosis

We report essentially irreversible binding of UII to the UT receptor with subsequent G-protein activation and an increase in intracellular Ca2+. The implications of irreversibility of binding in vivo are that high circulating UII concentration produce high degrees of occupancy, which may desensitize the receptor, possibly making it functionally silent under normal physiological conditions. Further studies of the desensitization process are essential to address this question.

Acknowledgement: Funded in part by a research grant from UHL NHS Trust. Keywords: cardiovascular system; pharmacology; urotensin

References

References

1 Douglas SA, Ohlstein HA. Trends Cardiovasc Med 2000; 10: 229–37 2 Douglas SA. Curr Opin Pharmacol 2003; 3: 159–67 3 Douglas SA, Dhanak D, Johns DG. Trends Pharmacol Sci 2004; 25: 76–85 4 McDonald J, Barnes TA, Okawa H, et al. Br J Pharmacol 2003; 140: 61–70 5 Hirst RA, Harrison C, Hirota K, Lambert DG. Methods Mol Biol 1999; 114: 31–9

Characterization of recombinant human urotensin II receptors W. Song1*, J. McDonald1*, J. P. Thompson1, D. J. Rowbotham1, G. Calo2 and D. G. Lambert1

Effect of lipopolysaccharide on the response of the rat extra-splenic microvasculature to atrial natriuretic peptide in vitro

1

Department of Cardiovascular Sciences (Pharmacology and Therapeutics Group), Division of Anaesthesia, Critical Care and Pain Management, LRI, Leicester LE1 5WW, UK. 2Department of Pharmacology, University of Ferrara, Ferrara, Italy

A. Mansart1*, J. J. Ross2, C. S. Reilly1 and Z. L. S. Brookes1

Urotensin II (UII) is a potent endogenous vasoconstrictor. Activation of the UII receptor (UT) increases intracellular Ca2+ in vascular smooth muscle cells to induce contraction. Plasma UII concentrations are variable but generally high indicating that under physiological conditions UT receptor occupancy will be high.1–3 In this study using Chinese hamster ovary cells stably expressing the human UT (CHOhUT) we have characterized the receptor binding profile of [125I]UII and compared affinity estimates with: (i) G-protein activation and (ii) intracellular Ca2+ responses to UII. CHOhUT membranes (10–15 mg) were incubated in buffer (Tris– HCl 50 mM, MgSO4 5 mM) containing bovine serum albumin 0.5%, pH 7.4 for up to 4 h at 37 C with approximately 10 pM [125I]UII. Non-specific binding was defined in the presence of 10 6 M UII. Bound and free radioactivities were separated by vacuum filtration. Receptor density (Bmax) and equilibrium dissociation constant (Kd) were determined by isotope dilution of [125I]UII with unlabelled UII. G-protein activation was determined by assessing agonist stimulated GTP[g-35S] binding (in the presence of 0.1 mM GDP) and increases in intracellular Ca2+ were assessed in Fura 2-loaded cells as described previously.4 5 All data are mean (SEM) (n). The binding of [125I]UII was time-dependent reaching equilibrium at 4 h. Interestingly, when an excess of unlabelled UII was added after 4 h to initiate dissociation [125I]UII binding appeared irreversible (for a further 2 h of the experiment). Isotope dilution studies at 4 h revealed a concentration-dependent decrease in binding as dilution increased. Scatchard transformation of these data yielded Bmax and Kd values of 1110 (70) fmol mg 1 protein and 742 (63) pM (n = 5), respectively. UII stimulated GTP[g-35S] binding was concentration-dependent and saturable with pEC50 (potency) and Emax (efficacy) of 8.69 (0.12) and 5042 (468) disintegrations per minute (DPM) (n = 7), respectively. In addition UII produced a concentration-dependent biphasic increase in intracellular Ca2+ with pEC50 and Emax of 8.38 (0.17) and 841 (59) nM (n = 11), respectively.

1 Academic Anaesthesia Unit, University of Sheffield, Royal Hallamshire Hospital, Sheffield S10 2JF, UK. 2Northern General Hospital, Sheffield S5 7AU, UK

Detrimental responses to sepsis include hypotension, hypovolaemia, and extravasation of fluid from the splenic microvasculature and increase during sepsis.1 Atrial natriuretic peptide (ANP) is an important physiological regulator of plasma volume, causing generalized vasodilatation, except in mesenteric, coronary, renal, and extrasplenic vessels, where ANP causes constriction. The aim of the present study was to determine the effects of low-dose lipopolysaccharide (LPS) on extra-splenic vessels during the early phase of sepsis (<4 h) and determine whether LPS alters microvascular responses to ANP. Male Wistar rats (180–240 g) (n = 44) were used to obtain extrasplenic arterioles (270–380 mm) and venules (480–700 mm). Isolated vessels were then mounted on the pressure myograph to determine changes in diameter in the following experimental groups: (i) control, HEPES for 4 h (n = 10); (ii) LPS, LPS (50 mg ml 1) for 4 h (n = 12); (iii) control+ANP, HEPES for 4 h + ANP (EC50) at 10, 60, 120, 180, and 240 min (n = 10); (iv) LPS+ANP, LPS (50 mg ml 1) for 4 h and ANP (EC50) at 10, 60, 120, 180, and 240 min (n = 12). Responses to phenylephrine (10 4 mol litre 1) were also determined at the end of control and LPS groups to assess vascular tone. At 4 h, LPS constricted the venules (45.7 [5.2]%, P<0.05), but not the arterioles (2 [6.1]%). Constriction to ANP at 10, 60, 120, 180, and 240 min was reduced in the presence of LPS (Table 2). LPS also reduced the response to phenylephrine (decrease of 89 [17.4]% arterioles and 89.9 [29]% venules) (P<0.05). ANP caused contraction of extra-splenic vessels that was inhibited by LPS. The dysfunction of physiological mechanisms that normally regulate blood volume could contribute to hypovolaemia during sepsis. However, the implications and mechanisms of our findings will be investigated in future studies.

404P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

1 Moon HK, Taylor GI. Plast Reconstr Surg 1988; 82: 815–29 2 Pal S, Patel S, Drikakis D. Br J Anaesth 2004; 93: 167–8 3 Drikakis D, Rider W. High-Resolution Methods for Incompressible and Low-Speed Flows. Berlin: Springer, 2004

Proceedings of the Anaesthetic Research Society

Effects of ANP and LPS on microvascular constrictive response. Values are mean [SEM]. *P<0.05, control+ANP vs LPS+ANP (arterioles). {P<0.05, control+ANP vs LPS+ANP (venules) Table 2

10 min

60 min

120 min

180 min

240 min

36.7 33.3 2.1 2.3

36.7 (7.1) 29.4 (5) 5.8 (0.5)* 13 (3.3){

34.9 27.3 8.2 3.2

34.5 30.3 3.1 2.3

35.3 (1.8) 29.6 (0.4) 5.1 (3.1)* 0{

Table 3

Categories and frequencies of events, with examples

Category

Example

n

‘Lack of smoothness’

Cough on induction. Movement on incision Failed venous cannulation. Failed spinal Inappropriate sharps disposal Re-use of i.v. fluid bag Alarms sound inappropriately ECG electrode falls off Capnometer not functioning No assistant available No handover in recovery room Trip over wires. I.V. cannula accidentally pulled out Near miss—i.v. connection of local anaesthetic infusion (averted)

28

Procedural difficulty Control+ ANP LPS+ ANP

Arterioles Venules Arterioles Venules

(5.8) (3) (0.7)* (1.2){

(3.7) (1.6) (3.5)* (1.3){

(3.3) (0.8) (1.9)* (1.2){

‘Failure to follow protocol’ Monitoring/equipment difficulties Organizational/staff-related

Acknowledgement: This work was funded by the Anaesthetic Cardiothoracic Research Fund, Northern General Hospital, Sheffield and the Association of Anaesthetists of Great Britain and Ireland.

Physical hazards Drug-related events

19 18 15

14 5 4

References

Reference 1 Andrew P, Deng Y, Kaufman S. Am J Physiol Regul Integr Comp Physiol 2000; 278: R60–5

1 Crombie I, Davies HTO. J Eval Clin Pract 1998; 4: 31–8 2 Smith AF, Goodwin D, Mort M, et al. Br J Anaesth 2003; 91: 319–28

Advances in the genetic basis and diagnosis of susceptibility to malignant hyperthermia

Categorization and significance of non-conformities in anaesthetic practice: pointers from the Lancaster anaesthetic expertise study

R. Robinson*, P. J. Halsall*, M. Hogg* and P. M. Hopkins MH Investigation Unit, Academic Unit of Anaesthesia, St James’s University Hospital, Leeds LS9 7TF, UK

A. F. Smith1, D. Goodwin2*, M. Mort2* and C. Pope3* 1

Royal Lancaster Infirmary, Institute for Health Research and Lancaster University, School of Nursing and Midwifery, 3Southampton University 2

Measuring the quality of anaesthetic care is difficult. Previous approaches have tended to focus on outcomes while process measures have received little attention.1 With this in mind, we analysed the transcripts from our observation of anaesthetists and other staff at work gathered as part of the Lancaster expertise study,2 which provided a large and unique repository of observations of ‘everyday’ anaesthetic practice. Aberrations in process, no matter how minor, were noted and categorized as below. We had over 130 h of observation of anaesthetists at work. Whilst we witnessed no serious mishaps, we noted 103 ‘events’, which were categorized as in Table 3. We also noted five instances of actions intended to reduce some of the above hazards. Minor deviations from what might be called protocolized practice are very common in anaesthesia. There is thus a potential problem with what these events should be called. Possible terms include variances, non-routine events, deviations, and non-conformities, but the choice must be made carefully to avoid the implication that such events come about as a result of some failing on the anaesthetist’s part. However, some unconventional events may actually make practice safer—rigid adherence to protocol may in some circumstances cause more problems. If, though, these events are preventable by a change of process (consistent with the Royal College of Anaesthetists’ definition of a critical incident), further study and analysis of process measures such as these may be a fruitful approach to measuring the quality of anaesthetic care. Acknowledgement: NHS North West Regional R&D Fund (RDO/ 28/3/05) funded the study. Keywords: anaesthesia, professional expertise; anaesthesia, quality of care

Malignant hyperthermia (MH) is an autosomal dominant disorder with estimated prevalence of 1 in 8500. For approximately 30 yr the only method of clinical diagnosis was by the invasive in vitro contracture test (IVCT). In the early 1990s the first MH susceptibility locus was identified, RYR1 on chromosome 19q13.1. To date, it remains the major genetic locus for MH, although a further five susceptibility loci have been identified. In 2001 guidelines for genetic diagnosis of MH were published.1 Since 2000 the MH unit, in partnership with the Yorkshire Regional Genetics Service has been developing a DNA screen, which now benefits approximately 36% of families. A major research focus is to further increase the number of patients who may benefit from such testing. Here we report our progress with this research. The MH Unit holds the National MH Register, which comprises nearly 700 families. The methods used include the following. Genome wide approaches (i) Family linkage studies to estimate the proportion of families where there is chromosome 19, and therefore likely RYR1 gene involvement. (ii) Association analyses to assess the involvement of multiple loci in MH susceptibility in single families, rather than a single susceptibility locus. Gene-specific analysis (i) Mutation scanning of functionally relevant regions of RYR1, and whole gene sequencing to identify mutations associated with MH susceptibility. (ii) Mutation frequency assessment in the UK MH population to assess the potential benefit for use in a DNA test. (iii) Functional characterization of mutations identified by in vitro assessment. Our linkage results indicate that RYR1 predisposes to MH susceptibility in approximately 80% of UK MH families. Ten families with no linkage to chromosome 19 have been investigated for linkage to the other known susceptibility loci and no linkage was found;

405P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

Keywords: antigen, lipopolysaccharide; blood, plasma, atrial natriuretic peptide; cardiovascular system

Proceedings of the Anaesthetic Research Society

140

Summary of RYR1 mutation analyses in UK MH families

No. families with functionally characterized mutation No. families with non-functionally characterized mutation No. families screened for known mutations with none found No. families where DNA available but no screening conducted No. families where no DNA available for screening Total no. families on national database

Peak contraction or Ca transient (% of control)

Table 4

181 44 231 29 198 683

this suggests either further novel loci remain to be characterized or the operation of a more complex genetic model in some families.2 Mutation analysis has so far detected 34 RYR1 mutations in 46% of families where DNA is available (see Table 4). Fourteen of thirty-four mutations detected are accepted by the European MH Group for use in diagnostic testing. Ten mutations are unique to our families and require further characterization before they may be used for genetic diagnosis.

120 100 80 60

∗ 40 20 0 0

60

120

180

Incubation time (min)

Fig 8 Mean data (SEM) describing the effects of TNF-a 0.05 ng ml

1

and IL-1b 2 ng ml 1 on the magnitude of cell shortening and the Ca2+ transient (expressed as per cent of control). *P<0.05 vs control. n>12 for each step.

References 1 Urwyler A, Deufel T, McCarthy T, et al. Br J Anaesth 2001; 86: 283–7 2 Robinson R, Hopkins P, Carsana A, et al. Hum Genet 2003; 112: 217–8

The effect of TNF-a and IL-1b on contractility and calcium handling of rat ventricular myocytes 1

M. Shaw*, M. D. Graham*, P. M. Hopkins and S. M. Harrison*

The results of this study indicate that TNF-a and IL-1b depress cell shortening secondary to reduced myofilament Ca2+ sensitivity. The increase in time to 50% relaxation of both contraction and the Ca2+ transient observed after 3 h incubation may be suggestive of a decrease in the activity of SR Ca2+ uptake (via SERCA2) or Ca2+ efflux mechanisms. Keywords: calcium; complications, TNF-a; heart, muscle; polypeptides, cytokines, interleukins, IL-1b

School of Biomedical Sciences and 1Academic Unit of Anaesthesia, University of Leeds, Leeds, UK

The aim of this study was to investigate the effect of pathophysiologically relevant concentrations of TNF-a and IL-1b on contractility (measured as unloaded shortening) and the cytosolic Ca2+ transient in isolated rat ventricular myocytes. Ventricular myocytes were isolated by enzymatic dispersion from Wistar rats (killed by a schedule 1 technique). The cells were incubated with TNF-a 0.05 ng ml 1 and IL-1b 2 ng ml 1 for 60, 120, or 180 min. Control cells were incubated in normal Tyrode’s solution. Cells were loaded with fura 2-AM to assess intracellular Ca2+ concentration. During the experiments the cells were maintained at 30 C in a constantly perfused chamber mounted on the stage of an inverted microscope and stimulated at 1 Hz. Changes in cell length were measured using an optical edgedetection system and simultaneous recordings of Ca2+ transient derived from ratiometric fura 2 fluorescence emission. The cells were superfused with Tyrode’s solution or TNF-a 0.05 ng ml 1 and IL-1b 2 ng ml 1. Data (mean [SEM]) from incubation and control groups were compared using ANOVA with post-hoc Dunn’s test. Incubation with TNF-a and IL-1b followed by superfusion with Tyrode’s solution did not affect contractility. Incubation and superfusion with TNF-a and IL-1b caused a time-dependent depression of cell shortening apparent after 1 h. The maximum depression occurred after 3 h when cell shortening was 38 (7)% of control (P<0.05). No change in the amplitude of the Ca2+ transient was observed (Fig. 8). Prolongation of cell relaxation was accompanied by a decrease in the rate of decline of the Ca2+ transient. After 3 h, the time taken for 50% cell relaxation and Ca2+ decline was 165 (5) and 136 (4)% (respectively) of control. Myofilament Ca2+ sensitivity, assessed from plots of cell length vs fura-2 fluorescence, was reduced in cells that had been incubated and superfused with TNF-a and IL-1b. Significant depression of sensitivity was apparent after 1 h (61 [16]% of control, P<0.05) and 3 h of incubation (42 [10]% of control, P<0.001) although data after 2 h were not significantly different from control.

Neutrophil migration as a marker of EPCR polymorphism functionality—a pilot study D. A. Lowes*, A. M. Watt*, H. F. Galley and N. R. Webster Academic Unit of Anaesthesia and Intensive Care, University of Aberdeen, Scotland, UK

Mortality from sepsis remains high even with intensive medical care. Activated protein C (APC) therapy has been shown to reduce mortality in patients with severe sepsis, but there is some evidence of resistance to APC treatment in certain patients, which may be related to expression of the endothelial protein C receptor (EPCR). In addition to endothelial cells, neutrophils also express EPCR and neutrophil migration towards a chemoattractant is inhibited by APC.1 This inhibition is ablated by antibodies to EPCR and thus neutrophil migration in the presence of APC provides a useful functional measure of EPCR activity. Using data from the human genome project we identified a single nucleotide substitution (G/A) polymorphism in exon IV of the gene for EPCR, such that the amino acid serine is replaced with glycine in the synthesized protein. We hypothesized that the polymorphism may influence EPCR structure, function, or expression and used neutrophil migration as a functional measure of EPCR activity. Following ethical committee approval and informed consent, 10 healthy Caucasian subjects were recruited (six female). DNA was isolated from EDTA blood using standard techniques and PCR was performed using appropriate primers. Genotype of the polymorphism was determined using the restriction enzyme PstI and alleles were identified under UV light after gel electrophoresis in the presence of ethidium bromide. Neutrophils were isolated using single-density gradient centrifugation and cells were placed in a microchemotaxis chamber in which a 5 mm pore-sized cellulose nitrate filter separated the upper from the lower chamber. Migration in response to formyl-met-leu-phe (FMLP, 1 mM) and interleukin-8

406P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

Keywords: complications, malignant hyperthermia; genetics

Contraction Ca transient

Proceedings of the Anaesthetic Research Society

(IL-8, 1 nM) with and without pre-treatment with APC (100 ng ml) was determined in triplicate. After 30 min, filters were washed and stained with giemsa, then fixed and counted manually (3 fields well 1) by an observer blinded to the genotype. We found that, unexpectedly, eight subjects were heterozygous (A/G) for the EPCR polymorphism, but only two subjects were homozygous for the G form (glycine) and none were homozygous for the serine form (A/A). Migration towards both FMLP or IL-8 was reduced with prior exposure to APC in all subjects (P = 0.005 and P = 0.007, respectively) but the frequency of the polymorphism precluded any conclusion as to the functionality in terms of neutrophil migration. The technique is cheap and easy to perform as a simple measure of EPCR activity but is extremely time consuming and tedious. Further studies will necessitate automation of the counting process.

Acknowledgement: We are grateful to Lilly for financial support for this study. Keywords: blood, neutrophils; protein, activated protein C

Reference 1 Bernard GR, Vincent JL, Laterre PF, et al. N Engl J Med 2001; 344: 699–709

Acute liver failure: neuroexcitation and oxygen metabolism M. C. Bellamy, S. Turner, S. Murdoch, C. E. Millson and R. Harding Intensive Care Unit, St James’s University Hospital, Leeds, UK

Reference 1 Sturn DH, Kaneider NC, Feistritzer C, et al. Blood 2003; 102: 1499–505

The effect of activated protein C (APC) on neutrophil function N. E. ElSakka, H. F. Galley, N. R. Webster and B. H. Cuthbertson Academic Unit of Anaesthesia and Intensive Care, University of Aberdeen, Scotland, UK

Mortality from sepsis remains high even with intensive medical care. Activated protein C (APC) therapy has been shown to reduce mortality in patients with severe sepsis, but there is little information as to the possible immunological effects of APC. Neutrophils are key cells involved in early host defence mechanisms in sepsis. We hypothesized that APC may have effects on neutrophil function. In this pilot study we measured early apoptosis, respiratory burst activity, and cytokine expression in the presence of increasing concentrations of APC in neutrophils from healthy volunteers. Following ethical committee approval and with informed consent, 10 healthy subjects were recruited. Neutrophils were isolated using single density gradient centrifugation and incubated for 24 h at 37 C in air 95%/carbon dioxide 5% in the presence of 2 mg ml 1 lipopolysaccharide (LPS) with and without incubation with 0–200 mg ml 1 recombinant human APC (rhAPC) for 60 min before the addition of the LPS. Apoptosis was determined using Annexin-V staining with flow activated cell sorting (FACS) analysis. Respiratory burst activity was also determined using FACS, after treatment of cells with phorbylmyristate acetate in the presence of dihydrorhodamine. Cytokine bead array with FACS analysis was used to measure six key cytokines simultaneously: interleukin (IL)-1b, -6, -8, -10, and -12p70 and tumour necrosis factor (TNF). Exposure to LPS resulted in a significant increase in respiratory burst activity, IL-1b, IL-6, and IL-8 expression (all P<0.0001) and a significant decrease in the number of apoptotic cells (P<0.0001). IL10, IL-12p70 and TNF were not detectable. Pre-incubation with APC had no significant effect on respiratory burst, apoptosis, IL8 or IL-1b at any concentration, but resulted in a significant and dosedependent decrease in IL-6 expression (P = 0.04, Friedman analysis of variance). We have shown in this small pilot study that APC treatment of human neutrophils in vitro results in decreased IL-6 expression using cytokine bead array analysis, without affecting other cytokines, apoptosis, or respiratory burst activity. This confirms clinical data

We have previously demonstrated that neuroexcitatory phenomena in human acute liver failure as shown by glutamate release can produce brain death in the absence of raised intracranial pressure. In the present study we have therefore investigated the relationship between neuroexcitation and oxygen metabolism in this patient group. Seven patients with hyperacute liver failure secondary to paracetamol poisoning were studied. Intracranial and cerebral perfusion pressures (ICP and CPP) were monitored by a Camino fibre-optic system. Cerebral microdialysis was performed using the CMA70 microdialysis system via a catheter placed in the right parietofrontal region. Mixed venous oxygen saturation was measured by a continuos SjO2 catheter placed in the left jugular bulb. All patients were sedated with propofol and alfentanil and ventilated to normocapnea. Haemodynamic stability was maintained by infusion of norepinephrine as required. All patients received N-acetylcysteine by infusion. Data were analysed hourly and considered by case as well as pooled. 300 data points were analysed. Cerebral lactate/pyruvate ratio (a marker of cellular hypoxia) was positively correlated with glutamate but not with CPP or ICP (r = 0.4, P = 0.0001). There was a negative correlation in individual patients between glutamate and SjO2 (i.e. positive correlation with oxygen extraction) (see Fig. 9). There was no overall relationship between CPP and either glutamate or oxygen extraction. These results suggest that neuroexcitatory mediators result in increased oxygen extraction and cellular anaerobic metabolism in

100 Glutamate (µmol litre–1)

Keywords: blood, neutrophils; genetics, EPCR polymorphism

r 2=0.8, P<0.0001

50

0 60

70

80

90

100

110

SjO2 Fig 9 Relationship between glutamate and oxygen saturation in the blood drawn from jugular bulb.

407P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

Acknowledgement: We are grateful to Baxter for financial support for this study.

where treatment of patients with severe sepsis with APC also resulted in decreases in circulating IL-6.1 Further studies are needed to investigate the mechanism of this effect.

Proceedings of the Anaesthetic Research Society

hyperacute liver failure. These processes appear independent of traditional measures such as CPP and ICP. Keywords: complications, liver failure; liver, neuroexcitation; oxygen, metabolism

Reference 1 Milligan L, Harding RJ, Buglass H, et al. Br J Anaesth 2003; 90: 818P

Lack of evidence for the prevention of cardiac dysfunction from reperfusion injury during orthotopic liver transplantation by the Ca2+ sensitizer levosimendan

hypertension, cardiac or pulmonary disease, were taking Ca2+ channel antagonists, or b-blockers. Groups were similar in patient characteristics and surgically. There were no differences in diastolic function as assessed by the rEDVI/pressure relationship. Systolic performance differed significantly during drug infusion but was lost after graft reperfusion (Fig. 10). Levosimendan improved systolic function during the infusion period before graft reperfusion but did not affect diastolic function appreciably. The effect was lost rapidly following reperfusion. Keywords: complications, reperfusion injury; pharmacology, levosimendan; surgery, liver transplant

Protection against hepatic reperfusion injury by thoracic epidural anaesthesia?

On behalf of the Leeds Liver Research Group, St James’s University Hospital Leeds, Leeds, UK

A. Quinn, S. L. Perry*, A. Ermenyi*, G. Bem*, J. P. A. Lodge*, G. J. Toogood* and M. C. Bellamy

Graft reperfusion during liver transplantation imposes major stress on the heart. Circulating mediators lead to myocardial depression and alteration in vascular reactivity with pulmonary hypertension and systemic vasodilatation. Changes in cardiac function resemble myocardial stunning with ventricular stiffening leading to systolic and diastolic dysfunction. Standard treatment with inotropes and vasopressors supports systolic function but leaves diastolic relaxation. The calcium sensitizing agent levosimendan has been claimed to have both inotropic and lusitropic properties. We have therefore studied this drug in liver transplantation. Twenty adult patients undergoing elective orthotopic liver transplantation were enrolled into this double-blind study. With written informed consent, patients were randomized to receive either 0.1 mg kg 1 min 1 levosimendan or placebo. Trial drug infusion was started following induction of anaesthesia and before commencement of surgery. It was discontinued at reperfusion of the donor liver graft. Patients were anaesthetized following standard local guidelines with continuous cardiac monitoring taking place using a pulmonary artery flotation catheter for the estimation of cardiac output, right heart pressure, and ejection fraction and followed up until 6 h after the operation. Patients were excluded from the study if they had renal failure, acute liver failure,

On behalf of the Liver Research Group, Departments of Anaesthesia and Surgery, St James’s University Hospital, Leeds, UK

Cardiac index litre min–1 m–2

P=0.047 between groups and P=0.0023 with time 8

Levo

Placebo

7 6 5 4 3 H1

H2

H3

H4

H5

H6

J

Time point Fig 10 Changes in cardiac index during different time intervals in

patients receiving placebo or levosimendan. Data are mean (SEM); H1, baseline; H2, pre-bypass; H3, 3 min prior to reperfusion; H4, 10 min after reperfusion; H5, 30 min after reperfusion; H6, skin closure; J, 6 h post-op.

Thoracic epidural anaesthesia (TEA) reduces infarct size following myocardial reperfusion injury.1 TEA has recently been reported to confer greater haemodynamic stability following the Pringle manoeuvre during liver resection. We propose that these effects are related to a phenomenon similar to protective ischaemic preconditioning and could be mediated by nitric oxide, which has been postulated as a key mediator for hepatic protection in ischaemia reperfusion (IR) injury.2 We have studied patients undergoing liver resection with varying durations of hepatic ischaemia in a pilot study to establish1 whether it is possible to measure changes in nitric oxide metabolites in this patient group, and second what the potential size of any protective effect would be for purposes of powering a future trial. Twenty patients undergoing liver resection were studied. Patients were allocated in a 2:1 ratio to receive TEA intraoperatively using bupivacaine in a dose at the clinical discretion of the anaesthetist, or to receive TEA for postoperative use only. An air–oxygen– desflurane–fentanyl anaesthetic sequence was used. The number of liver segments resected and the duration of hepatic ischaemia were recorded. Liver function tests were measured preoperatively, before and after ischaemia, and daily thereafter. Nitric oxide intraoperatively was estimated by total nitrate/nitrite (TNN) measured by the Griess reaction. Data are presented as mean (SE) and given for the epidural group followed by the non-epidural group. Statistical analysis was performed using Mann–Whitney test. Patients in the epidural (n = 13) and non-epidural (n = 7) group were well matched. Clamp times and the numbers of segments resected did not differ between groups. TNN values were similar between groups preoperatively (0.38[0.06] vs 0.33[0.09] mmol litre 1) but significantly higher in the epidural group at the point of Pringle clamping (0.42[0.06] vs 0.18 [0.03], P = 0.04). ALT values (a marker of hepatocellular injury) were similar preoperatively (42.7 [11.5] vs 36.5 [9.0], P = 0.78) but significantly less in the epidural group after ischaemia–reperfusion injury (183 [24] vs 388 [117], P = 0.027). These results confirm that it is possible to detect changes in TNN suggestive of changes in nitric oxide intraoperatively in patients receiving TEA. Our data suggest a protective effect of TEA and possibly through nitric oxide in hepatic ischaemia–reperfusion injury. A larger interventional trial is warranted on the basis of these results to investigate the clinical importance of these biochemical observations and the cellular mechanisms involved. Keywords: anaesthesia, reperfusion injury

408P

thoracic

epidural;

complications,

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

S. Turner, R. Harding, A. Ermenyi*, M. Bellamy and L. Christiansson*

Proceedings of the Anaesthetic Research Society

References 1 Groban L, Zvara DA, Deal DD, et al. J Cardiothorac Vasc Anesth 1999; 13: 579–85 2 Koti RS, Seifalian AM, Davidson BR. Dig Surg 2003; 20: 383–96

Epidural analgesia and coagulopathy following liver resection: when to remove the epidural catheter? D. Stamenkovic, Z. Jankovic, G. Toogood, J. P. A. Lodge and M. C. Bellamy

by the acute pain team and the criterion was an INR less than 1.2. The epidural catheters were removed on day 5 (5.03 [1.49] days) with minimum 1 day to maximum of 11 days stay. The mean time for epidural catheter removal in our patients was 5 days, although the coagulation disorders were persistent on the sixth postoperative day. The criterion INR for epidural removal was seldom achieved. Despite this no epidural or spinal haematoma was recorded. There does not seem to be clear consensus regarding current practice. Keywords: anaesthetic techniques, epidural; blood, coagulation; liver, resection

St James’s University Hospital, Beckett Street, Leeds LS9 7TF, UK

Table 5

The effect of iontophoresed norepinephrine and heat on skin vascular reactivity as assessed by the transient hyperaemic response in human volunteers M. Beed*, J. Kaur*, I. K. Moppett* and R. P. Mahajan University Department of Anaesthesia and Intensive Care, City Hospital, Nottingham, UK

The evaluation of vascular reactivity in different circulatory beds may be useful in disease states such as sepsis. Forearm skin vascular reactivity may be assessed using a transient hyperaemic response (THR) induced by 20 s of axillary artery compression. THR can be manipulated by the iontophoresis of vasodilators, but not vasoconstrictors, possibly because healthy skin has low baseline blood flow.1 Heating the skin whilst vasoconstrictors are applied may accentuate blood flow changes.2 Our experiment used locally applied heat to dilate skin before iontophoresis of norepinephrine in 15 healthy volunteers. Two laser Doppler probes measured forearm skin blood flow-flux, one a control laser, the other via a combined heating and iotophoresis chamber. Three THR tests were performed before heating the skin for 5 min, then three further THR tests were performed, followed by iontophoresis of 0.1% norepinephrine before three final THR tests. Iontophoresis was pulsed using 45-s periods of 75 mA and 0 current over 10 min. The three temperatures used were ambient skin temperature, 35 and 42 C. Baseline flow-flux was measured for 60 s before each set of THR tests. The THR ratio (THRR) was calculated by comparing baseline flow-flux immediately before arterial compression (F1) with the maximum increase after release (F2): THRR = F2/F1. The average values of each group of THRR results and baseline data were compared using the Kruskal– Wallis test. Mean arterial pressure (MAP) and skin temperature, were compared using ANOVA tests. No statistically significant changes in MAP or control site skin temperatures occurred. Iontophoresis of norepinephrine consistently decreased flow-flux (P<0.005) at all skin temperatures; it also decreased THRR significantly at ambient skin temperature or at 35 C. At 42 C, THR response was abolished and at this temperature norepinephrine partially restored THRR (Table 5). We have shown that iontophoresed norepinephrine

Flow-flux and THRR changes when iontophoresis of norepinephrine follows skin preheating

Temperature

Ambient temperature 35 C 42 C

Flow-flux median (range), n = 15

THRR median (range), n = 15

Before norepinephrine

After norepinephrine

P-value

Before norepinephrine

After norepinephrine

P-value

11.2 (7.3–24) 16 (7.9–50.8) 116.6 (46.2–257)

6.1 (5–10.7) 10.3 (5.1–69.5) 13.7 (8.4–34.2)

<0.005 0.007 <0.005

3.05 (1.52–5.65) 2.72 (1.56–3.79) 0.96 (0.62–2.34)

1.83 (1.09–3.69) 1.33 (1.03–3.53) 1.62 (1.23–3.86)

0.006 0.007 <0.005

409P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

Epidural analgesia has numerous well-known advantages and perceived risk of epidural haematoma formation secondary to the coagulopathy in liver resection surgery. The aim of this study was to detect the duration and severity of postoperative coagulopathy and to establish the optimal time for epidural catheter removal. This was a retrospective case note review. Data were collected from medical records of 115 patients who underwent major liver resections. Patients were identified from theatre records and an audit database and included on the basis of case-note availability. Surgical data were taken directly from an existing surgical audit database (MS Excel) and included all identified patients over a 2-yr period. The exclusion criteria were re-do liver resection and epidural catheters sited in other hospitals. A data collection proforma was designed to compile information from surgical and anaesthesia records. Data are mean (SD) or median (range). Multiple comparisons within the group were performed using ANOVA for repeated measures (GLM) (SPS5). An epidural catheter was present in 100 patients (89.3%) and used for intra and postoperative local anaesthetic and opioid infusion. Indications for hepatectomy were colorectal metastases (78.3%), hepatocellular carcinoma (3.5%), hemangioma (0.9%), and other (17.4%). The most frequent hepatic resections were non-anatomical metastasectomy (21.2%), right hepatectomy (18.6%), and right trisectionectomy (extended hemihepatectomy) (14.2%). The extent of resection was estimated by the number of liver segments resected (median 4 segments, range 0–7). The Pringle manoeuvre was used in 65.2%. The mean ischaemic time was 27.16 (18.17) min (0–104 min). The mean number of Pringle manoeuvres was 2.26 (1.58) per case. All patients were extubated immediately after surgery. A reduced platelet count (202.8 [84.5] 109 litre 1) was seen post operation with the nadir on the third postoperative day. This persisted even at the sixth postoperative day (300.8 [123.2] vs 250.9 [110.2] 10 9 litre 1, GLM, P = 0.013). There was a significant increase in PT, PTT, and INR values after the operation. Peak PT, PTT, and INR values were generally attained on day 2 (22.7 [5.5] s, 39.1 [6.8] s, and 2.01 [0.50], respectively). Significant elevation of PT, PTT, and INR persisted even at day 6 post operation (17.6 [4.3] s, 34.8 [5.5] s, and 1.53 (0.37), respectively) (GLM, P<0.001). The mean INR on day 5 (peak time for catheter removal) was 1.67 (0.38). Coagulopathy was corrected with FFP in five patients, though in none of these was any clinical indication recorded in the case notes. The decision for time of epidural catheter removal was made

Proceedings of the Anaesthetic Research Society

modulates vascular reactivity differently at different skin temperatures. Acknowledgement: RCA/AAGBI for providing Intavent Research Fellowship grant to Dr Beed. Keywords: drug delivery, iontophoresis; skin, vascular reactivity; sympathetic nervous system, norepinephrine

References

The effect of LPS on vasoconstrictor responses of the porcine coronary artery was quantitatively similar to that previously reported for unintentional microbial contamination.1 However, unintentional exposure to micro-organisms exerted greater effects than LPS alone on endothelium-dependent and endothelium-independent relaxations. Clearly, microbial factors other than LPS appear to be implicated in altering vascular responsiveness and they may have relevance for developing in vitro models of sepsis. Keywords: antigen, lipopolysaccharide; complications, sepsis; heart, vascular reactivity

1 Brown H, Moppett IK, Mahajan RP. Br J Anaesth 2003; 90: 446–51 2 Drummond PD. Br J Clin Pharmacol 2002; 54: 45–50

References 1 Wei J-X, Wilson G, Mahajan R. P. Br J Anaeth 2004; 93: 604–5P 2 Lawrence RN, Dunn WR, Wilson VG. J Pharm Pharmacol 1998; 50: 885–90

Clinical significance of mis-scoring in a ‘track and trigger’ system during an outbreak of Legionnaires’ disease

J. X. Wei, V. G. Wilson1 and R. P. Mahajan University Department of Anaesthesia and Intensive Care, 1School of Biomedical Sciences, Queen’s Medical Centre, Nottingham NG7 2UH, UK

R. J. Oakey1*, A. M. Harry1* and A. F. Smith2 1

Research and Development and 2Anaesthesia, Royal Lancaster Infirmary, Lancaster, UK

The Legionnaires’ outbreak in Barrow-in-Furness in 2002 allowed us to study the performance of a ‘track and trigger’ system (TTS) in practice. Written permission for use of data was obtained from patients with confirmed Legionnaires’ disease and from randomly selected patients admitted with similar symptoms during the outbreak. Primary observations and early warning scores (EWS) from nursing notes and charts were entered into an Excel spreadsheet. Recalculation of corrected or ‘true’ EWS allowed us to establish the overall incidence of missing data and miscalculated EWS.1 Here we present between-group comparisons and an estimate of the effect of EWS miscalculation. There were 95 Legionnaires’ positive (LP) and 108 Legionnaires’ negative (LN) patients. LP patients were significantly older (P = 0.003) but there was no difference in sex distribution. Significantly more primary observations were made daily on LP patients (P = 0.051), but there was no difference in the number of EWS recorded. There was no difference in the incidence of missing EWS between the groups (LP 34%, LN 30%), but there was a higher overall incidence of miscalculation in the LP group (20 vs 14%, P = 0.011). Further, the likelihood of miscalculation in both groups increased with the severity of illness. The direction of error is also relevant (see Fig. 11). At EWS = 1, less than 40% of errors were underscored and, at an EWS of >3, about 80% of errors were 100% Underscored - control group

We have recently reported that unintentional microbial contamination of porcine isolated coronary arteries impaired vasoconstrictor responses to U46619; an effect associated with the induction of nitric oxide synthase.1 We have compared the effect of exposure to lipopolysaccharide (LPS) and microbial contamination on vasoconstrictor and vasodilator responses of the porcine isolated coronary artery, in an attempt to understand changes in vascular reactivity in sepsis. Segments (5 mm) of the PCA were dissected from pig hearts and incubated in Krebs Henseleit (K-H) solution at 37 C in either the absence of an antibiotic mixture (60 mg ml 1 benzylpenicillin and 20 mg ml 1 streptomycin sulphate) or in the presence of LPS (100 mg ml 1) with the antibiotic mixture. After 16–18 h the segments were prepared for isometric tension recording2 and responses to 60 mM KCl elicited. 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 (SP) to assess the integrity of the endothelium, or increasing concentration of sodium nitroprusside. The segments were then exposed to cumulatively increasing concentrations of U46619 and the responses expressed as percentage of their own maximum or U46619-induced tone. For comparisons – log [EC50] was calculated and Student’s paired t-test was used. Data are mean–SEM. K-H solutions stored at 37 C without antibiotics were cloudy, indicating the presence of microbial contamination, while those stored in the presence of the antibiotics were clear. Following exposure to LPS (100 mg ml 1) in the presence of the antibiotic mixture contractions to 60 mM KCl (control 15.7 [1.0] g weight; LPS 15.0 [1.7] g weight, n = 9) and relaxation to 10 nM SP (control 60.4 [3.5]%; LPS 71.1 [4.6]%, n = 9) were unchanged. However, the sensitivity of the preparation to U46619 was reduced 2-fold (Control log EC50 7.87 [0.04]; LPS log EC50 7.54 [0.1], P<0.05, n = 9) following exposure to LPS (100 mg ml 1). Sodium nitroprusside caused concentration-dependent inhibition of U46619-induced tone and this effect was significantly (P<0.05) reduced 4-fold following exposure to 100 mg ml 1 LPS (control – log EC50 7.30 [0.22]; LPS log EC50 6.75 [0.19], n = 9). Inclusion of the antibiotic mixture in the incubation solution significantly increased SP-induced relaxations in vessels (antibiotics 66.7 [4.7]%; control 46.3 [7.3], P<0.005, n = 16) and increased 20-fold the sensitivity of the preparations to sodium nitroprusside (antibiotics – log EC50 7.04 [0.1]; control log EC50 5.79 [0.16], P<0.003, n = 8).

EWS=3 80%

EWS=2

60% 1:1 40%

EWS=1

20% 20%

40%

60%

80%

100%

Underscored - legionnaires' group

Fig 11 Percentages of miscalculations resulting in underscoring of the EWS.

410P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

Comparison of the effect of prolonged exposure to LPS or microbial organisms on responses of the porcine isolated coronary artery

Proceedings of the Anaesthetic Research Society

underscored. The proportion of errors where recorded EWS was less than ‘true’ EWS was always lower for LP patients (above 1:1 line in Fig. 11). There are many contributing factors to error in scoring.

Keywords: monitoring, early warning score; monitoring, track and trigger system

Reference Acknowledgement: Funding: UK Department of Health.

1 Oakey RJ, Harry AM, Smith AF. Br J Anaesth 2004; 92: 308P

Downloaded from http://bja.oxfordjournals.org/ at Serials Section, Dixson Library on June 1, 2015

411P