Upper airway reflex sensitivity following general ... - Wiley Online Library

6 downloads 5400 Views 475KB Size Report
60 min but was similar to baseline values 120 min after recovery. Upper airway ..... services appeared to benefit from the patients' greater independence in managing their symptoms. Keywords Pain: ..... Failed lumbar disc surgery and repeat.
Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

F O RU M

Upper airway re¯ex sensitivity following general anaesthesia for day-case surgery R. Caranza,1 N. Nandwani,2 J. P. Tring,2 J. P. Thompson3 and G. Smith4 1 Consultant, 2 Specialist Registrar, 3 Senior Lecturer and 4 Professor, University Department of Anaesthesia, Leicester Royal In®rmary, Leicester LE1 5WW, UK Summary

We have studied changes in upper airway re¯ex sensitivity following general anaesthesia using dilute ammonia vapour as a chemical stimulant in 16 patients undergoing elective laparoscopic gynaecological surgery. We measured the threshold concentration of ammonia vapour required to elicit a transient reduction of inspiratory ¯ow caused by glottic closure, de®ned as a glottic stop. Measurements of upper airway re¯ex sensitivity and auditory reaction time were obtained before surgery, and at 60 and 120 min after recovery. Auditory reaction time was depressed signi®cantly at 60 min but was similar to baseline values 120 min after recovery. Upper airway re¯ex sensitivity remained signi®cantly reduced at 60 and 120 min despite the return of auditory reaction time to normal. The lack of correlation between upper airway re¯ex sensitivity and auditory reaction time suggests that central nervous system depression alone does not explain the delayed recovery in airway reactivity. Keywords Measurement techniques: airway re¯exes. Airway: re¯exes. Anaesthesia: general; day case. ...................................................................................... Correspondence to: Dr R. Caranza. Present address: Department of Anaesthesia, Pilgrim Hospital, Boston, Lincolnshire PE21 9QS, UK. Accepted: 27 September 1998

In recent years, there has been a great increase in day-case surgery. It is considered by some to be the `best option for 50% of all elective surgical procedures' [1]. This has been accompanied by the development of anaesthetic techniques that aim to ensure prompt recovery and enable early patient discharge. Discharge criteria are usually determined by locally agreed guidelines but there is no information on the return of upper airway re¯exes after general anaesthesia and consequently when to allow patients to start eating and drinking. The time taken for return of laryngeal re¯exes after anaesthesia is important, as this is the period during which the patient may be at risk from aspiration of food and gastric contents. This pilot study was designed to investigate upper airway re¯ex sensitivity in the postoperative period using dilute ammonia vapour as a chemical stimulus [2], and to determine whether changes in upper airway re¯ex sensitivity relate to changes in auditory reaction times [3]. Q 2000 Blackwell Science Ltd

Methods

After obtaining Local Ethics Committee approval and written informed consent, we studied 16 unpremedicated ASA grade I and II female patients, aged 26±41 years undergoing day-case laparoscopic gynaecological surgery under general anaesthesia. Patients were not studied if they were smokers, gave a history of asthma or had suffered an upper respiratory tract infection in the previous 6 weeks. Patients with a history of gastro-oesophageal re¯ux or anticipated dif®culty in airway maintenance or tracheal intubation were also not studied. Patients were allocated to one of two groups according to the type of airway management, either laryngeal mask airway (LMA) or tracheal tube. All patients received a standardised anaesthetic consisting of fentanyl 1 mg.kg 1, propofol 2±4 mg.kg 1 and vecuronium 0.05±0.1 mg.kg 1. Anaesthesia was maintained with 33% oxygen in nitrous oxide and iso¯urane 0.5±2%. Additional analgesia was 367

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

provided with rectal diclofenac 100 mg, given at the end of surgery. Ondansetron 4 mg was given intravenously. Residual neuromuscular block was antagonised with neostigmine and glycopyrronium at the end of surgery. Adequate reversal was con®rmed with the use of a percutaneous nerve stimulator. The tracheal tube or LMA was removed when the patient had begun to reject it and an adequate respiratory pattern had been re-established. Patients were assessed in the recovery area and the time at which they were ®rst able to communicate and respond to verbal commands was de®ned as time zero. After familiarisation with the equipment, baseline upper airway re¯ex sensitivity and auditory reaction time were assessed before surgery as described below. Additional measurements were repeated at 60 and 120 min after recovery from anaesthesia. Measurement of upper-airway-re¯ex sensitivity Upper airway re¯ex sensitivity was measured using low concentrations of ammonia vapour as described previously [2]. The subject's upper airway was exposed to single intermittent breaths containing increasing concentrations of ammonia vapour and the threshold concentration of ammonia required to elicit glottic closure was recorded. A glottic closure was de®ned as a rapid decrease in the inspiratory ¯ow of at least 25% of the peak value, followed by a swift recovery, the whole event lasting less than 0.5 s (Fig. 1). Low values for ammonia vapour threshold concentration are associated with increased sensitivity of the

upper airway, high values re¯ect a depression of upper airway re¯ex sensitivity [2]. Measurement of auditory reaction time Auditory reaction time was recorded as a measure of recovery from general anaesthesia [3]. At random intervals, the device used emits single bleeps delivered via headphones. A counter measures the time (in ms) until a button is pressed. The time taken for the patient to press the button in response to the auditory stimulus has been de®ned as auditory reaction time. At each assessment point, the mean value of ®ve auditory reaction time readings was obtained. Data were analysed by nonparametric tests, using analysis of variance and Wilcoxon matched-pairs signedrank tests. A probability value < 0.05 was considered to be signi®cant. Results

Sixteen female patients who underwent gynaecological laparoscopic surgery completed the study (eight subjects in each group). Patient characteristics and duration of surgery were similar in both groups (Table 1). LMA insertion or tracheal intubation was performed by experienced anaesthetists. No technical dif®culties were encountered in airway maintenance or intubation. The LMA was inserted in a conventional manner and ventilation was assessed by capnography. All patients underwent either diagnostic

Figure 1 An example of a glottic stop

recording. GS, glottic stop; BV, switching of the pneumatic valve, allowing the subject to take a single breath of ammonia vapour.

368

Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

Table 1 Patient characteristics in the tracheal tube and laryngeal

mask groups. Values are expressed as mean (range or SD).

Age; years Weight; kg Duration of surgery, min

Tracheal tube group nˆ8

Laryngeal mask group nˆ8

34.6 (26±41) 61.7 (8) 18 (4)

32.3 (29±37) 65.1 (9) 17 (4)

laparoscopy or laparoscopic sterilisation. There were no differences between the groups in the type or duration of surgery. We found that there was a signi®cant reduction in upper airway re¯ex sensitivity, shown by an increase in ammonia vapour threshold concentration, at 60 min and 120 min (Fig. 2). Auditory reaction time was increased signi®cantly in both groups at 60 min postoperatively, but had returned to baseline values by 120 min (Fig. 3). All

patients were discharged from hospital uneventfully after 120 min. Discussion

In this pilot study, we found a reduction in airway reactivity at 60 and 120 min after recovery from anaesthesia. In addition, we observed that auditory reaction time, although prolonged at 60 min, returned to baseline levels at 120 min. This study was performed in our Day Care Unit and we were unable to perform measurements at a later stage as patients were discharged from the hospital. Despite the reduction in upper airway re¯ex sensitivity during the ®rst 2 h after surgery, patients were allowed to eat and drink before discharge as is customary in our unit. The use of dilute ammonia vapour as a chemical stimulant of the upper airway is an objective method of measuring upper airway re¯exes by determining the threshold concentration required to elicit glottic closure.

Figure 2 Mean ammonia threshold concentration (NH3TR) measured before, and 60 and 120 min after recovery from anaesthesia (i.e.

time zero, Tzero ). Error bars indicate SEM. *p < 0.05. Left, tracheal tube; right, laryngeal mask airway.

Figure 3 Mean auditory reaction time (ART) measured before, and 60 and 120 min after recovery from anaesthesia (i.e. time zero,

Tzero ). Error bars indicate SEM. *p < 0.05. Left, tracheal tube; right, laryngeal mask airway.

Q 2000 Blackwell Science Ltd

369

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

Previous studies in healthy volunteers have demonstrated that the mean threshold concentration of ammonia lies between 363 and 642 p.p.m. [2]. The aim of this pilot study was to assess upper airway re¯ex sensitivity in the postoperative period in relation to clinical measures of recovery from anaesthesia and auditory reaction time. The exact relationship between a decrease in the sensitivity of upper airway re¯exes and the incidence of aspiration is unknown, but it could be of signi®cance in debilitated patients at risk of aspiration [4, 5]. Although there was a reduction in upper airway re¯ex sensitivity after surgery in both groups, comparisons between groups with regard to the type of airway management would be inappropriate in a study of this size. The mechanism responsible for glottic closure has not been elucidated fully. Subepithelial receptors in the upper airway responsive to chemical irritants relay to the brain stem via vagal and sympathetic afferent nerves [6]. The lack of correlation between upper airway re¯ex sensitivity and auditory reaction time suggests that central nervous system depression alone does not explain the delayed recovery in airway reactivity. The reduction in upper airway re¯ex sensitivity may be related to mechanical stimulation as a result of LMA or tracheal tube placement but it is of note that similar changes occurred in both groups. Alternatively, as upper airway receptors that respond to ammonia are thought to be rapidly adapting [7], it is possible that such changes are a result of altered receptor sensitivity after general anaesthesia. In summary, we have demonstrated that upper airway re¯ex sensitivity as assessed by ammonia vapour threshold concentration was impaired 2 h after general anaesthesia. Auditory reaction time had returned to baseline by this time and there was no clinical evidence of residual anaesthesia or sedation. The signi®cance of this ®nding

is unclear, but it is possible that patients may be at increased risk of aspiration despite being conscious with normal auditory reaction times. Further work is required to determine the time required for the upper airway re¯ex sensitivity to return to baseline values and the effect of different types of airway management on upper airway re¯ex sensitivity. Acknowledgment

We thank Dr E. Pallett for technical support. References 1 Royal College of Surgeons of England. Report of the Working Party on Guidelines for Day Case Surgery. London: Royal College of Surgeons, 1992. 2 Langton JA, Murphy PJ, Barker P, Key A, Smith G. Measurement of the sensitivity of upper airway re¯exes. British Journal of Anaesthesia 1993; 70: 126±30. 3 Gale GD. Recovery from methohexitone, halothane and diazepam. British Journal of Anaesthesia 1976; 48: 691±7. 4 Erskine RJ, Murphy PJ, Langton JA, Smith G. Effect of age on the sensitivity of upper airway re¯exes. British Journal of Anaesthesia 1993; 70: 574±5. 5 Pontoppidan H, Beecher HK. Progressive loss of protective re¯exes in the airway with the advanced age. Journal of the American Medical Association 1960; 174: 2209±13. 6 Nishino T, Tanaka A, Ishikawa T, Hiraga K. Respiratory, laryngeal and tracheal responses to nasal insuf¯ation of volatile anesthetics in anesthetized humans. Anesthesiology 1991; 75: 441±4. 7 Widdicombe JG. Re¯exes from the upper respiratory tract. In: Widdicombe JG, ed. Handbook of Physiology ± The Respiratory System, Vol. II. Bethseda, MD: American Physiology Society, 1986; 363±94.

F O RU M

The ef®cacy of surgically placed epidural catheters for analgesia after posterior spinal surgery A. Turner,1* J. Lee,2 R. Mitchell,3 J. Berman,4 G. Edge4 and M. Fennelly4 1 Specialist Registrar, 2 Research Fellow, 3 Consultant Radiologist and 4 Consultant Anaesthetist, Department of Anaesthesia, The Royal National Orthopaedic Hospital NHS Trust, Brockley Hill, Stanmore, Middlesex HA7 4LP, UK Summary

Posterior spinal fusion for correction of scoliosis is a major procedure for which the provision of satisfactory, safe postoperative analgesia is often a problem. One possible solution involves the 370

Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

placement of epidural catheters under direct vision by the surgeon at the end of the procedure, followed by an epidural infusion of local anaesthetic with or without an opioid. Despite its simplicity, this technique has not been reported as being consistently successful. We report an observational study of the analgesia achieved with surgically placed epidural catheters and of the reasons for the failure of the technique. Fourteen consecutive patients undergoing posterior spinal fusion had epidural catheters placed by the surgeon and had radio-opaque dye injected down the catheter 15 min before their routine postoperative chest X-ray. Analgesia was assessed at 0, 6, 12 and 24 h after surgery using visual analogue scores. Five patients had inadequate pain control; none of these patients had dye visible in the epidural space. Seven patients had dye visible in the epidural space; all of these cases had satisfactory analgesia. In two cases, dye was observed in the paravertebral gutters; both of these patients had satisfactory postoperative analgesia. This small pilot study suggests that correctly placed `surgical' epidural catheters are capable of providing good analgesia after posterior spinal fusion and that misplaced catheters, seen in a large proportion of patients, are associated with inadequate analgesia. Keywords Analgesia: postoperative. Surgery: spinal. Analgesia, techniques: epidural. ...................................................................................... Correspondence to: Dr A. Turner Accepted: 10 April 1999

Providing safe and effective postoperative analgesia remains problematic for patients undergoing posterior spinal fusion. This type of surgery is typically the second part of a two-stage procedure. The ®rst stage comprises an anterior soft tissue release through a thoracotomy, followed a week later by a posterior spinal fusion. The latter procedure is performed with the patient in the prone position and requires an extensive incision exposing most of the thoraco-lumbar spine. The spinous processes of the vertebrae involved in the scoliosis are then removed to allow access for the positioning of the spinal rods that are used to correct the curvature. The vertebral laminae, ligamenta ¯ava and meninges are not directly involved in the surgery. Posterior spinal fusion is most frequently performed in children, adolescents and young adults who have a range of neuromuscular, skeletal and cardiorespiratory pathology. These patients frequently have a degree of developmental delay. There is usually signi®cant postoperative pain. Epidural analgesia would appear to be an elegant solution, with the catheter being inserted through the ligamentum ¯avum under direct vision by the surgeon towards the end of surgery. Several small studies have investigated this technique using epidural opioids, with or without local anaesthetic agents [1±3]. However, between 7 and 100% of patients had inadequate analgesia with this technique. The design of these studies is open to criticism. In one study [1], supplementary intravenous or intramuscular opioids were given routinely to patients in both the epidural and nonepidural groups. In another, there was a broad case-mix, including anterior releases with percutaneously placed epidural catheters and posterior fusions with the epidural catheters positioned by Q 2000 Blackwell Science Ltd

the surgeon [2]. The same study also combined prospective and retrospective data and did not use a standardised epidural solution. The aim of this prospective observational study was to investigate the contribution of catheter placement and segmental spread to the quality of postoperative analgesia observed in patients with surgically placed epidural catheters. We do not believe that this has been investigated previously and we are continuing with a randomised prospective study investigating the success rates with different methods of ensuring correct catheter placement. Methods

Following ethics committee approval, sequential patients undergoing elective posterior spinal fusion for correction of scoliosis were invited to participate in the study. Patients weighing > 25 kg who were able to understand the nature of a visual analogue score (VAS) for pain, as well as the use of an intravenous patient-controlled analgesia (PCA) device, and in whom epidural analgesia was appropriate, were included. Informed verbal consent was obtained from all patients. Written consent was then obtained from either the patient or from the parent or legal guardian in the case of minors. Temazepam 0.5 mg.kg 1 was given orally 90 min before surgery. Induction of anaesthesia comprised fentanyl 1 mg.kg 1 and propofol 3±5 mg.kg 1, with vecuronium 0.1 mg.kg 1 given to provide neuromuscular blockade. Central venous and arterial cannulation preceded the start of surgery. Variation of technique was then allowed according to the preference of the anaesthetist. Anaesthesia 371

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

was maintained with a volatile agent and nitrous oxide in oxygen. Analgesia during surgery was provided with either boluses of fentanyl or a remifentanil infusion. Just before wound closure, the surgeon inserted 4±5 cm of a 16 G epidural catheter (Portex, UK) into the epidural space. An epidural bolus of plain bupivacaine 0.25% 10 ml could be given at this stage at the discretion of the anaesthetist. At the end of surgery, any residual muscle relaxation was reversed. Anaesthesia was discontinued and, after tracheal extubation, the patients were transferred to the recovery ward. As soon as the patients were able to communicate, pain was assessed using a standard VAS. Five millilitres of X-ray contrast medium (Omnipaque 300) were then injected down the epidural catheter. Fifteen minutes later, an anterior± posterior portable chest X-ray was performed to check the spread of contrast and to con®rm correct central venous catheter placement. The X-ray was assessed by a consultant radiologist (RM) with special attention to the presence, site and spread of contrast (Fig. 1). Postoperative analgesia was provided with an epidural infusion of a mixture of bupivacaine 0.1% and fentanyl 5 mg.ml 1 at a rate of 0±15 ml.h 1. If satisfactory epidural analgesia was not established, two `rescue' boluses of bupivacaine 0.25% 5±10 ml could be given at least 30 min apart. Paracetamol and diclofenac were given as required. Visual analogue pain assessments were made at 6, 12 and 24 h after surgery. The absolute VAS scores were divided into equal thirds and analgesia was de®ned as inadequate if more than one of the recordings was in the top third. The rate of epidural infusion, additional analgesia prescribed and any other adverse events were documented. Epidural infusions were discontinued if patients were experiencing signi®cant postoperative pain. These patients were given intravenous morphine PCA. Results

Fourteen patients (nine females) aged 12±22 years were entered into the study. Three patients underwent singlestage spinal fusion and the remainder underwent the second of a two-stage procedure. The epidural catheters were all inserted between vertebral levels T5 and T11. In seven patients, the catheter was inserted at the T8 level. Contrast medium was seen in the spinal canal of seven patients (50%; Table 1). In four of these, there was evidence of unilateral spread of contrast. Despite variable spread, all of these patients had satisfactory analgesia. Of the remaining seven with no epidural contrast visible on the chest X-ray, two had contrast seen within the paravertebral gutter (patients 4 and 8). Nine patients (64%) had satisfactory analgesia provided via the epidural catheter, seven of whom received additional oral medication and none needed `rescue' epidural boluses. Of the ®ve patients 372

Figure 1 Postoperative chest X-ray demonstrating contrast

within the epidural space (arrows).

(36%) who had inadequate analgesia despite `rescue' boluses of bupivacaine 0.25%, none had contrast within the spinal canal or paravertebral gutter. Twelve patients received fentanyl boluses during surgery and two had remifentanil infusions. Nine patients were given epidural boluses soon after siting of the catheter, six of whom went on to have good postoperative analgesia. The mean (SD) [range] postoperative bupivacaine and fentanyl solution infusion rate was 7.7 (2.9) [3±14] ml.h 1. The patients who received satisfactory epidural analgesia had their infusions continued for up to 48 h after surgery. Table 1 Spread of contrast and adequacy of analgesia Patient no.

Adequate analgesia?

Contrast seen in spinal canal?

Contrast spread; no. of vertebrae

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

Yes Yes Yes Yes Yes Yes Yes Yes Yes No No No No No

Yes Yes No Yes Yes Yes Yes Yes No No No No No No

1 5 8 4 5 5 5

Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

Two patients suffered nausea and vomiting, one of these also suffering from itching. Three patients reported paraesthesiae in one or both hands. These resolved in two of the patients during the study period without the need for reduction in epidural infusion rate and were probably due to positioning during surgery. One patient developed hypotension that was associated with postoperative haemorrhage. There were no reports of signi®cant motor block and the surgical and nursing staff were able to carry out their routine postoperative assessments. No patient required urinary catheterisation, which is not a routine procedure in this group of patients. Discussion

There is a considerable theoretical attraction in the use of surgically placed epidural catheters for analgesia after posterior spinal fusion surgery. However, the experience of others [1±3] has shown limited success, although their techniques were different to ours. In a report comparing epidural bupivacaine 0.0625% with PCA morphine, no difference in the incidence of effective analgesia and time to feeding and ambulation was found [4]. We have also experienced variable success with this technique at our institution. It is unclear whether the failures are attributable to misplacement of the catheter or to poor spread of the epidural solution as a result of abnormal anatomy. We have found that the surgical placement of epidural catheters is a feasible technique, with radiographic evidence of delivery of medication to the intended site. The spread of the contrast medium did not appear to affect the analgesia obtained. This may have been due to varying degrees of contrast spread at the time of X-ray imaging. Absence of contrast in the spinal canal may be secondary to incorrect initial catheter placement or to subsequent migration of the catheter. All of the patients who had contrast seen in the spinal canal reported adequate analgesia. Our data suggest that the use of a correctly sited epidural catheter will result in successful pain relief. There has been

Q 2000 Blackwell Science Ltd

understandable concern that this method of analgesia may interfere with postoperative assessments aimed at detecting cord compression or ischaemia. Purnell [5] describes a patient in whom the use of epidural local anaesthetic during surgery made it impossible to exclude a surgical cause for persistent postoperative paraparesis, necessitating the removal of the Harrington rods some 9 h after the end of surgery. In our series, there were no cases in which neurological assessment was compromised, although a larger study will be necessary to conclude that this is a rare occurrence. In conclusion, we believe that the surgical placement of catheters for postoperative epidural analgesia is a technique worthy of consideration in this group of patients, many of whom will be unable to use a PCA as they do not have the skills necessary to understand the concepts involved.

References 1 Amaranath L, Andrish JT, Gurd AR, Weiker GG, Yoon H. Ef®cacy of intermittent epidural morphine following posterior fusion in children and adolescents. Clinical Orthopaedics and Related Research 1989; 249: 223±6. 2 Adu-Gyam® Y. Epidural morphine plus bupivacaine for relief of postoperative pain following Harrington rod insertion for correction of idiopathic scoliosis. Journal of International Medical Research 1995; 23: 211±17. 3 Shaw BA, Watson TC, Merzel DI, Gerardi J, Birek A. The safety of continuous epidural infusion for postoperative analgesia in pediatric spine surgery. Journal of Pediatric Orthopedics 1996; 16: 374±7. 4 Cohen BE, Hartman MB, Wade JT, Miller JS, Gilbert R, Chapman TM. Postoperative pain control after lumbar spine fusion. Spine 1997; 22: 1892±7. 5 Purnell RJ. Scoliosis correction and epidural analgesia. Prolonged block following Harrington rod insertion. Anaesthesia 1982; 37: 1115±17. 6 Price D, McGrath P, Ra®i A, Buckingham B. The validation of visual analogue scales as ratio scale measures for chronic and experimental pain. Pain 1983; 17: 45±56.

373

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

F O RU M

An independent assessment of a supraregional pain management programme and comparison of patients' and general practitioners' perceptions of the effects S. Gupta, J. D. Francis, G. E. Porter and J. M. J. Valentine* Pain Management Centre, West Norwich Hospital, Bowthorpe Road, Norwich NR2 3TU, UK Summary

This study was designed to evaluate the therapeutic response, as perceived by patients and their general practitioners, to a supraregional residential pain management programme (INPUT). Twenty patients who completed the INPUT programme and their general practitioners, were sent locally designed questionnaires. Eighteen patients and 16 general practitioners responded. Patients were assessed for the effects of the programme on a wide variety of treatment targets re¯ecting quality of life. General practitioners were questioned about changes in dependence upon healthcare services. The mean time from attending INPUT to assessment was 11.4 (5.7) months. Thirteen patients indicated a moderate or maximal improvement in quality of life. General practitioners reported a reduced use of healthcare services and improvement in quality of life. The majority of our patients attending INPUT gained bene®ts that improved their quality of life. Healthcare services appeared to bene®t from the patients' greater independence in managing their symptoms. Keywords Pain: chronic. Psychological responses. ...................................................................................... Correspondence to: Dr J. M. J. Valentine Accepted: 23 June 1999

Chronic pain affects about 10% of the population in the developed world [1±3] and about 1% of the population are rendered severely disabled [4]. Coping with pain that persists beyond the expected period of tissue healing can be extremely dif®cult. This often leads to clinically relevant psychological and social problems [5±7]. Diagnostic tests often fail to identify de®nitive pathology; drugs provide limited relief and adverse effects are common [8±13]. Invasive treatments, although often tried, frequently do not help and have the potential to make the situation worse [5, 8, 14±16]. Multidisciplinary pain management programmes (PMPs), such as INPUT, address the distress and dysfunction secondary to chronic pain, rather than attempt to identify and cure a lesion [17]. Pain management programmes have been shown to lead to an improved quality of life in patients with chronic, nonmalignant pain [18, 19]. The aim of this investigation was to assess independently the medium-term effects, as perceived by the patients and their general practitioners 374

(GPs), of referral to a supraregional PMP. We also examined the relationship between the patients' and their GPs' perception of the effects. In view of the fact that our investigation was an independent survey, it may provide a more dispassionate assessment of `response' than follow-up by the INPUT staff themselves. Methods

Over a 2-year period, a total of 25 patients were referred to the INPUT programme at St. Thomas' Hospital in London, which treats patients suffering from chronic pain using cognitive-behavioural principles within an interdisciplinary approach. The staff are a team of doctors, psychologists, physiotherapists, nurses and occupational therapists who have extensive experience of dealing with the problems caused by chronic pain to individuals and their families. The residential programme is: 4 days a week (Monday to Thursday) for 4 weeks. Weekends are spent at Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

home to put into practice the skills that they have learnt in the unit. Spouses, partners, friends and family members are encouraged to attend on the last day of each week. The focus is on those aspects of function that are causing dif®culties for the patients, not on the site, cause or nature of the pain itself. The aims of the programme are to: 1 improve ®tness, mobility and posture by exercise and stretching, thus counteracting the effects of disuse; 2 return to more normal and satisfying activities by improved ®tness, through pacing and goal setting in chosen activities; 3 counteract unhelpful beliefs and improve mood and con®dence by cognitive principles and education about pain; 4 reduce the adverse effects of unhelpful medications by gradual withdrawal; 5 improve stress management and sleep by teaching relaxation techniques; 6 reduce the negative effects of chronic pain on the patient's immediate family by teaching negotiation skills and involving them in the programme. After assessment, all patients were considered suitable for the 4 week, residential, multidisciplinary PMP. Twenty patients who completed the programme were later sent our questionnaire. This questionnaire was designed locally

Table 1 The number of patients with different pain problems and

its duration

Pain problem

No. of patients

Mean (SD) duration of pain (years)

Low back pain (LBP) LBP ‡ lower limb pain LBP ‡ neck pain Generalised spinal pain Neck ‡ shoulder pain Neck ‡ upper limb pain

6 6 2 2 1 1

8 (8.1) 5 (2.9) 5.3 (5.3) 10.5 (2.1) 7 5

to assess the effects of the programme on a wide variety of treatment targets. Four patients voluntarily decided against attendance and one patient dropped out after 2 weeks attendance because he felt this programme was not helping him. We assessed the effects of the PMP on ability to cope with pain, understanding pain mechanisms, physical activity, social participation, out-of-house activities, pain medication, GP consultations and use of coping techniques during ¯are up. Each parameter was scored using a ®ve-point scale, one point indicating no bene®t through to ®ve points for maximum bene®t. The total score (maximum 40 points) was used to judge the overall response; > 32 points (i.e. >

Figure 1 Patients' responses to the various parameters in our questionnaire with one point indicating no bene®t through to ®ve points

indicating maximal bene®t from attendance of the pain management programme.

Q 2000 Blackwell Science Ltd

375

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

mean 4/5 points per question) indicating `maximal bene®cial effect'; 25±32 points (i.e. mean > 3/5 points per question) `moderate bene®cial effect'; 17±24 points (i.e. mean > 2/5 points per question) `minimal bene®cial effect' and < 17 points (i.e. mean < 2/5 points per question) `no bene®cial effect' from attending the PMP. To assess subjectively the change in overall quality of life, we used a simple numerical scale where `zero' indicated no improvement through to `10', which indicated maximum improvement. Patients were asked to identify the three most useful aspects of PMP and the relaxation techniques that they continued to ®nd most bene®cial. They were also asked questions about employment prospects and their opinion on the potential bene®ts of developing a PMP at Norwich (Appendix 1). General practitioners were sent a separate questionnaire designed to assess alterations in their patients' dependence upon healthcare services and reduction in use of analgesic medications. They were asked to indicate `yes', `no' or `same' as appropriate to the questions outlined in Appendix 2. Results

Eighteen patients (90%) and 16 GPs (80%) responded. There were 12 female and six male patients with a mean (SD) age of 40.1 (10.8) years. Mean (SD) duration from completing the PMP to assessment was 11. (5.8) months. The nature of the pain problems and the duration of the symptoms are shown in Table 1. Figure 1 shows the number of patients scoring each of one to ®ve points on our scale for the eight different parameters in the questionnaire. Figure 2 outlines the overall bene®cial response after attendance of the PMP. Thirteen of the 18 patients had a moderate to maximal overall bene®cial effect (> 3/5 points on each parameter or > 24 out of a total of 40 points) by attending the PMP.

Seventeen of the 18 patients responding indicated some improvement in quality of life, the mean (SD) `score' being 5 (2.4) with 11 patients scoring > 50% improvement. All patients used relaxation techniques, the three most common being breathing exercises, visualisation and music. The three aspects of PMP most useful to our patients were goal setting and pacing, exercise and stretching and learning to manage `¯are-ups' effectively. All except one patient felt that a local PMP, if available, would be a very useful adjunct to `pain management' in the region. The reasons given in support of a local PMP were: less travel, a more familiar environment, accessible quali®ed support, greater involvement for carers and family members, and that more patients could be offered this `treatment'. However, one patient felt that being away from home allowed him to concentrate on what was being taught and was not in favour of a local PMP. Three of the four patients who decided against attending the PMP indicated that travel was their main problem (the distance from our pain management unit to INPUT is < 120 miles). One patient discontinued after 2 weeks because he was unwilling to `suffer' the PMP routine any further. On the issue of return to work, nine patients were still not working due to their pain, four were planning to

Table 2 Responses given by the GP to the various parameters

questioned

Reduced number of consultations? Patient still searching for the `cure'? Reduction in analgesic medications? Referral for further medical opinions? PMP generally bene®cial?

Yes

No

Not Same answered

9 5 4 4 11

5 10 ± 10 3

± ± 9 ± ±

2 1 3 2 2

Figure 2 Overall bene®cial response after

attending the PMP as indicated by the patients.

376

Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

Table 3 Comparison of patients' and GPs'

responses to common parameters questioned

Parameters questioned

Yes ± Patients and GPs agree

No ± Patients and GPs agree

Patient and GP disagree

PMP generally bene®cial? Reduction in analgesic medication? Reduced number of consultations?

9 4 6

2 4 2

1 2 4

work, two were actively searching for a job and three were in part-time employment. The responses by the GPs are summarised in Table 2. Comparison of the responses by the patients and the GPs on the common parameters questioned is summarised in Table 3. Discussion

Establishments running PMPs often publish favourable outcomes; however, it is dif®cult to be certain that there is absolutely no bias in their reports [18, 19]. Our survey was different in the sense that we, as purchasers of the service, have attempted to quantify the bene®ts derived by our patients. Our survey indicates that most of our patients who attended the INPUT programme gained signi®cant short- to medium-term bene®ts, including improvement in their general quality of life. The survey of GPs indicated that their patients do bene®t from attending the PMP and that there is an apparent reduction in the use of healthcare services. Eleven of the 14 GPs who responded to the question, indicated that the PMP had an overall bene®cial effect on their patients. Luscombe et al. [19] reported a similar ®nding in which 13 of the 18 GPs responding suggested that patients who attended their PMP obtained `much bene®t'. In our study, nine of the 14 GPs responding to the question, believed that their patients consulted them less frequently for pain-related problems following the PMP. This is similar to the results published by Luscombe et al. and indicates that healthcare services do bene®t. However, in terms of overall cost to local healthcare providers it is not yet clear whether referral to a supraregional PMP is truly cost-effective. The 4-week residential programme costs our health authority approximately £3500 per patient. It is dif®cult to quantify the true cost savings after the PMP attendance in terms of reduced GP consultations, reduced specialist consultations, better quality of life and return to work in the small group of patients that we studied. However, Pither et al. [20] reported a cost saving of £240 per patient per year on drugs alone in addition to considerable improvements in function, psychological distress and quality of life after attendance of the INPUT programme. In our small sample of patients, Q 2000 Blackwell Science Ltd

only four out of 18 (22%) had a reduction in analgesic consumption following the PMP at the time of assessment, an average 11.4 (5.8) months after the programme. This is disappointing, as reduced dependence upon `unhelpful' medications is one of the key aims of the PMP. Previous studies have reported a signi®cant decrease in analgesic consumption following PMP [20±22], while others have suggested that overall failure of a PMP may be linked to failure to reduce analgesic consumption [22]. Twelve patients were, at the time of the survey, leading a life improved in quality by more than 50%. This indicates that even though the patients had pain that needed analgesics, they had learned to cope with their pain more effectively, which in fact, was the main aim of the PMP. Patients who were not taking analgesics before starting the programme would not report a decrease in analgesic consumption. A supplementary question about whether or not patients were taking any analgesics would have been helpful. Methodologically, the study would have been better had our patients been offered the chance to register worsening, not just staying the same or improving, on the assessment parameters. However, since only one patient reported no change on one item (Fig. 1), the results in this study are unlikely to have been signi®cantly in¯uenced by this. It would have been useful to have an equivalent untreated control group; however, such control patients are not willing to remain untreated over such a long period [23]. The intention of a PMP is to enable patients to change the way they deal with pain forever and to enable them to become more independent, active and satis®ed with their lives. The longer the follow-up, the more convincing the changes seen. Nevertheless, vulnerability to relapse into old habits is greatest in the ®rst few months after completing the programme, especially when adverse events impinge on the patients' ability to cope. General practitioners have an important role in helping patients who have completed a PMP to sustain their gains at such points. A local multidisciplinary pain management centre could offer important back-up facilities for patients at times of need. In our study, 11 (61%) of the 18 patients indicated that they were leading a quality of life that had been improved by more than 50% following PMP. Little change is apparent when comparing follow-up at 1 month and 377

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

1 year after PMP in this population [24]. Studies by other groups who have compared a treated group with patients who were denied treatment for ®nancial reasons, report an improvement in quality of life in 60±70% of patients attending the PMP, compared with 5% of patients who were unable to attend [21, 22, 25]. Finally, when we compared the response to questions sent to both patients and their GPs, we found that there was close agreement on the major issue of perceived improvement in quality of life (Table 3). This provides further subjective evidence to support the bene®cial role of a multidisciplinary PMP in helping patients optimally manage a life with intractable pain in an optimal way. Conclusions

Our survey shows that the majority of the patients from our region, who have attended the INPUT programme, gained signi®cant short- to medium-term bene®ts including improvement in their general quality of life. Healthcare services also appear to bene®t from this approach. The GPs also concur with these ®ndings when assessed independently. References 1 Rideout E, Browne G. The prevalence of pain complaints in a general population. Pain 1994; 18: 299±314. 2 Sternbach RA. Survey of pain in the United States: the Nuprin pain report. Clinical Journal of Pain 1986; 1: 49±53. 3 Magni G, Caldieron C, Rigatti-Luchini S, Merskey H. Chronic musculoskelatal pain and depressive symptoms in the general population. An analysis of the 1st national health and nutrition examination survey data. Pain 1990; 43: 299±307. 4 Von Korff M, Dworkin SF, Le Resche L. Graded chronic pain status: an epidemiologic evaluation. Pain 1990; 40: 279±91. 5 Fordyce WE. Behavioural Methods for Chronic Pain and Illness. St. Louis, MO: CV Mosby, 1976. 6 Sternbach RA. Psychological factors in pain. In: Bonica JJ, Albe-Fessard D, eds. Advances in Pain Research and Therapy, Vol . 1. New York: Raven Press, 1976. 7 Bonica JJ. Neurophysiologic and pathologic aspects of acute and chronic pain. Archives of Surgery 1977; 112: 750±61. 8 Flor H, Turk DC. Etiological theories and treatment for chronic back pain. I. Somatic models and interventions. Pain 1984; 19: 105±21. 9 Parry CBW. The failed back. In: Wall PD, Melzack R, eds. Textbook of Pain, 2nd edn. Edinburgh: Churchill Livingstone, 1989; 293±9. 10 Rosomoff HL, Fishbain DA, Goldberg M, Santana R, Rosomoff RS. Physical ®ndings in patients with chronic intractable benign pain of the neck and/or back. Pain 1989; 37: 279±87. 378

11 Turner JA, Calsyn DA, Fordyce WE, Ready LB. Drug utilization patterns in chronic pain patients. Pain 1982; 12: 357±63. 12 McNairy SL, Maruta T, Ivnik RJ, Swanson DW, Ilstrup DM. Prescription medication dependence and neuropsychology function. Pain 1984; 18: 169±77. 13 King SA, Strain JJ. Benzodiazepine use by chronic pain patients. Clinical Journal of Pain 1990; 6: 143±7. 14 Waddell F, Kummell EG, Lotto WN, Graham JD, Hall H, McCulloch JA. Failed lumbar disc surgery and repeat surgery following industrial injuries. Journal of Bone and Joint Surgery (Am) 1979; 61: 201±7. 15 Linton SJ, Bradly LA, Jensen I, Spangfort E, Sundell L. The secondary prevention of low back pain: a controlled study with follow-up. Pain 1989; 36: 197±207. 16 Pither CE, Nicholas MK. The identi®cation of iatrogenic factors in the development of chronic pain syndromes; abnormal treatment behaviour. In: Bond MR, Charlton JE, Woolf CJ, eds. Proceedings of the 6th World Congress on Pain. Amsterdam, The Netherlands: Elsevier, 1991. 17 Bradley LA. Cognitive-behavioural therapy for chronic pain. In: Gatchel RJ, Turk DC, eds. Psychological Approaches to Pain Management ± A Practitioner's Handbook. London: Guilford Press, 1996; 131±47. 18 Williams AC de C, Nicholas MK, Richardson PH, et al. Evaluation of a cognitive behavioural programme for rehabilitating patients with chronic pain. British Journal of General Practice 1993; 43: 513±18. 19 Luscombe FE, Wallace L, Williams J, Grif®ths DPG. A district general hospital pain management programme: ®rst year experiences and outcomes. Anaesthesia 1995; 50: 114±17. 20 Pither CE, Ralphs J. Behavioural treatment not drugs for chronic pain. British Medical Journal 1993; 306: 1687±8. 21 Guck TP, Skultety FM, Meilman PW, Dowd ET. Multidisciplinary pain center follow-up study: evaluation with a no-treatment control group. Pain 1985; 21: 295±306. 22 Roberts AH, Reinhardt L. The behavioural management of chronic pain: long-term follow up with comparison group. Pain 1980; 8: 151±62. 23 Peters JL, Large RG, Elkind G. Follow-up results from a randomised controlled trial evaluating in- and outpatient pain management programme. Pain 1992; 50: 42±50. 24 Williams AC de C, Richardson PH, Nicholas MK, et al. Inpatient vs. outpatient pain management: results of a randomised controlled trial. Pain 1996; 66: 13±22. 25 Morley S, Eccleston C, Williams A. Systematic review and meta-analysis of randomised controlled trials of cognitive behaviour therapy and behaviour therapy for chronic pain in adults, excluding headaches. Pain 1999; 80: 1±13.

Appendix 1

Audit of patients who attended the INPUT pain management programme Please circle the appropriate word to indicate how much you agree or disagree with the following statements. Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

(The statement `strongly disagree' scores one point through to `strongly agree' scoring ®ve points) 1. You are able to cope with your pain better now because you attended the pain management programme. strongly disagree disagree uncertain agree strongly agree 2. After attending the pain management programme I now understand the way pain occurs and that helps me to cope with my pain better. strongly disagree disagree uncertain agree strongly agree 3. Attending the pain management programme has improved my physical activities and performance. strongly disagree disagree uncertain agree strongly agree 4. Attending the pain management programme has improved my social participation. strongly disagree disagree uncertain agree strongly agree 5. Attending the pain management programme has improved my out-of-house activity (going for a walk, shopping, etc.). strongly disagree disagree uncertain agree strongly agree 6. The number and different types of tablets I take for pain has decreased since I attended the pain management programme. strongly disagree disagree uncertain agree strongly agree 7. The number of times I visit my GP due to my pain problem has decreased after attending the pain management programme strongly disagree disagree uncertain agree strongly agree 8. I use one or more coping techniques to cope with increased pain which occurs at times. strongly disagree disagree uncertain agree strongly agree 9. Do you use any relaxation techniques to help you cope with your pain? Yes/No If Yes, Please name them below 10. Do you think attending the pain management programme has helped to improve the quality of your life? Yes/No If Yes, On a scale of 0 to 10, `0' being no improvement and `10' being the maximum improvement you can think of, please indicate on the scale below the improvement in the Q 2000 Blackwell Science Ltd

quality of your life after attending the pain management programme. 0 1 2 3 4 5 6 7 8 9 10 no maximum improvement improvement If Yes, What three aspects of pain management programme you learnt are most helpful to you (Please write the most useful one ®rst and then the next according to your priority) 1. 2. 3. 11. Please list the medications that you currently take for pain and how many of each have you taken in the last 24 h. Medication Number of tablets If you can remember, could you please list the medications you were taking on an average for 24 h for your pain before attending the pain management programme. Medication Number of tablets for 24 h on an average 12. Have you been able to go back to work Yes/No (Please circle as appropriate) Unable to work due to pain Planning to work in future Trying to ®nd a job Working part-time Working full-time 13. Do you foresee any advantages or disadvantages of a pain management programme at Norwich? Yes/No Please give reasons for your answer Advantages Disadvantages

Appendix 2

Audit of patients who have attended the INPUT pain management programme ± questionnaire to general practitioners 1. Has the number of visits by your patient due to painrelated problems decreased after attending the pain management programme? Yes/No 2. Do you think this patient has been bene®ted by attending the pain management programme? Yes/No 379

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

3. Has the patient been referred to other speciality (orthopaedics, rheumatology, etc.) with pain-related problems after completing the pain management programme? 4. Do you think this patient is still looking for a cure for the pain problem? Yes/No 5. If possible, could you please tell us what medication the patient was taking for pain before attending and after attending the pain management programme?

Current medication for pain (including nonconventional analgesics, such as amitriptyline, anticonvulsants, etc.) Medication Approximate dosage 1. 2. Medication for pain before attending the pain management programme Medication Approximate dosage 1. 2. Any other information you would like to add:

F O RU M

Intra-ocular pressure and haemodynamic changes after tracheal intubation and extubation: a comparative study in glaucomatous and nonglaucomatous children R. Madan,1* P. Tamilselvan,2 S. Sadhasivam,2 D. Shende,3 V. Gupta2 and H. L. Kaul4 1 Additional Professor, 2 Senior Resident, 3 Associate Professor, and 4 Professor and Head of Department, Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India Summary

This prospective, controlled study was conducted to compare the effects of tracheal intubation and extubation on intra-ocular pressure changes and haemodynamic parameters in paediatric patients with and without glaucoma. The children were scheduled for intra-ocular surgery. Twenty children with normal intra-ocular pressure and 15 with glaucoma were studied. A standardised general anaesthetic was administered to both groups. After 5 min of anaesthesia, intra-ocular pressure, heart rate and noninvasive blood pressure were measured. These measurements were repeated 30 s and 2 min after tracheal intubation. Further measurements were taken before, and 30 s and 2 min after extubation. The increase in intra-ocular pressure after intubation was greater in the glaucomatous group than in the normal group. The increase in intra-ocular pressure was greater after extubation than intubation in both groups, but was similar in the two groups. However, because of the already increased intra-ocular pressure in glaucomatous children, they may be at an increased risk of visual damage after intubation and extubation. Keywords Ophthalmology: glaucoma; intra-ocular pressure. Airway: intubation; extubation. ...................................................................................... Correspondence to: Dr R. Madan Accepted: 14 July 1999

Airway manipulation during general anaesthesia can be associated with an increase in intra-ocular pressure (IOP) due to an increase in blood pressure and an increase in blood ¯ow to the eye [1]. Tracheal intubation is associated 380

with a marked increase in IOP [1±3] in adults. These changes also occur at the time of tracheal extubation during recovery from anaesthesia [4]. Most studies have highlighted the haemodynamic and IOP changes during Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

intubation only [1, 2]. In one study, the magnitude of IOP changes during both intubation and extubation in adult patients with normal eyes was compared. The authors concluded that IOP increases to a signi®cantly greater extent during extubation [4]. The increase in IOP is greater in glaucomatous eyes than in the normal eyes whenever there is an increase in blood pressure and conditions that impede the venous return, e.g. coughing [5]. Therefore, we anticipated that glaucomatous patients might have an exaggerated IOP response during intubation and extubation. After an extensive literature search, we could ®nd only one study in which IOP changes in glaucomatous eyes during intubation and laryngeal mask airway insertion were studied in adult patients [6]. We could not ®nd a study involving children that compared IOP changes during intubation and extubation in normal eyes with those found in glaucomatous eyes. Therefore, this trial was designed to compare the changes in the IOP and haemodynamic responses to intubation and extubation in normal and glaucomatous eyes in children undergoing intraocular surgery under general anaesthesia. Methods

After institutional approval and parental informed consent, 35 ASA I children between the ages of 1 and 6 years, who were scheduled for routine unilateral intra-ocular surgery, e.g. lens aspiration, intra-ocular lens implantation, penetrating keratoplasty or trabeculectomy, were enrolled in the study. All the patients suffering from glaucoma were receiving antiglaucoma therapy. Twenty children with normal eyes were used as a control group and the remaining 15 children with glaucomatous eyes comprised the glaucoma group. All children were premedicated with diazepam syrup 0.2 mg.kg 1 given 2 h before surgery. Monitoring comprised heart rate, oxygen saturation (SpO2), noninvasive blood pressure (NIBP), end-tidal carbon dioxide (FE0 CO2) (Datex, Cardiocap-II, Datex Instruments Inc., Finland), end-tidal concentration of inhalational agents (Datex Capnomac), and IOP (Perkin's hand-held applanation tonometer). General anaesthesia was induced with nitrous oxide 66% in oxygen along with concentrations of halothane increased progressively to 3±4%, until surgical anaesthesia was achieved. The concentration of halothane was then reduced to achieve an end-tidal concentration of 0.5%. This level was maintained for 5 min. Lidocaine 4% was then instilled into the eye that was not to be operated upon. Heart rate, NIBP and IOP were measured at this time. This was taken as a baseline reading as we could not measure IOP in awake children. In all cases, a single blinded investigator measured IOP. After baseline measurements Q 2000 Blackwell Science Ltd

had been taken, vecuronium 0.1 mg.kg 1 was given to facilitate tracheal intubation in both groups. The IOP, heart rate and NIBP were measured at 30 s and 2 min after intubation. Patients who presented dif®culty in intubation and required repeated attempts at intubation were not studied. General anaesthesia was maintained with an end-tidal halothane concentration of 0.5±1.0% in 66% nitrous oxide in oxygen. The FE0 CO2 was maintained between 4.5 and 5.5 kPa with controlled ventilation at a peak airway pressure < 25 cmH2O. Analgesia was provided with intravenous pethidine 1 mg.kg 1. At the end of the surgery, the preextubation values for heart rate, NIBP and IOP were measured. The residual neuromuscular blockade was reversed with neostigmine 50 mg.kg 1 and glycopyrronium 10 mg.kg 1. At the onset of spontaneous breathing, the tracheal tube was removed after gentle pharyngeal suctioning. After extubation, the children were allowed to breathe spontaneously with an end-tidal concentration of halothane of 0.5% in 66% nitrous oxide in oxygen. Heart rate, NIBP and IOP were recorded 30 s and 2 min after extubation. The incidences of coughing, breath-holding and laryngospasm after extubation were also noted. The data were analysed using Statistix version 4.0 and Microstat statistics software. The demographic data were analysed using Student's t-test. The HR, NIBP and IOP measurements were analysed using analysis of variance within each group at different times and by using Student's t-test between the groups. The incidences of coughing, breath-holding and laryngospasm were compared using the Chi-squared test and Fisher's exact test where appropriate. Data are presented as mean (SD). The null hypothesis was rejected at p < 0.05. Results

Patient characteristics are summarised in Table 1. There were no statistically signi®cant differences between the two groups. The operations performed are listed in Table 2. The IOP increased signi®cantly after intubation in both groups (Table 3). In the normal group, IOP increased signi®cantly from a pre-intubation level of 12.7 (3.7) mmHg to 15.5 (5.7) mmHg (p ˆ 0.003) and 15.5 (5.5) mmHg (p ˆ 0.0002), 30 s and 2 min after intubation, respectively. In the glaucoma group, the IOP increased signi®cantly from the baseline value of 24.5 (7.5) mmHg to 31.6 (9.8) mmHg (p ˆ 0.0001) and 31.5 (10.0) mmHg (p ˆ 0.00005), 30 s and 2 min after intubation, respectively. The increase in IOP 30 s and 2 min after intubation was greater in the glaucoma group than in the normal group (p ˆ 0.0118 and p ˆ 0.0093, respectively; Table 4). At extubation, there was a marked increase in IOP in both groups (Table 3, Fig. 1). In the normal group, the IOP 381

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

Table 1 Demographic data. Values are given as mean (SD) where

appropriate.

Glaucoma

(n ˆ 20)

(n ˆ 15) 4.1 (5.1) 9:6 12.8 (7.4) 12.8 (7.4)

Table 2 Operations performed.

Lensectomy Optical iridectomy Trabeculectomy Keratoplasty Hyphema drainage

Normal group (n ˆ 20)

Glaucoma group (n ˆ 15)

14 1 2 2 1

0 0 15 0 0

24.5 (7.5) [14±40] 31.6 (9.8) [22±46]* 31.5 (7.7) [20±44]*

Extubation Baseline 30 s after extubation 2 min after extubation

15.5 (6.3) [8±30] 22.9 (7.6) [13±44]* 22.8 (7.4) [12±34]*

31.5 (7.7) [18±40] 38.7 (7.6) [28±50]* 38.6 (9.1) [26±50]*

* Signi®cantly different from baseline value, p < 0.01.

30 s after intubation 2 min after intubation 30 s after extubation 2 min after extubation

Normal group (n ˆ 20)

Glaucoma group (n ˆ 15)

p value

2.8 (3.9) 3.1 (3.3) 7.0 (7.9) 7.5 (5.4)

7.7 (5.5) 7.5 (5.0) 7.4 (4.3) 8.5 (4.6)

0.0118 0.0093 0.4634 0.4333

intubation and extubation (Table 5). However, there were no statistically signi®cant differences in absolute values or relative increases between the two groups. Tracheal intubation and ventilation were not dif®cult in any of the patients. None of the patients coughed during intubation. The incidence of complications, such as coughing, breath-holding, straining and laryngospasm, was comparable in the two groups (Table 6).

GLAUCOMA

*

*

*

*

*

* 2 in min tu ba af tio ter n

* 3 in 0 s tu a ba fte tio r n

12.7 (3.7) [7±16] 15.5 (5.7) [8±24]* 15.5 (5.5) [7±16]*

mmHg) after intubation and extubation.

*

In Ba tub se ati lin on e

IOP (mmHg)

CONTROL

Intubation Baseline 30 s after intubation 2 min after intubation

Table 4 Mean increase from baseline in intra-ocular pressure (in

increased signi®cantly from 15.5 (6.3) mmHg just before extubation to 22.9 (7.6) mmHg (p ˆ 0.0002) and 22.8 (7.4) mmHg (p ˆ 0.00005) at 30 s and 2 min after extubation, respectively. In the glaucoma group, the increase in IOP was from 31.5 (7.7) mmHg to 38.7 (18.6) mmHg (p ˆ 0.00003) and 38.6 (9.1) mmHg (p ˆ 0.0009) at 30 s and 2 min after extubation, respectively. The difference in the increase in IOP after extubation between the normal group and the glaucoma group was not statistically signi®cant (Table 4). There were signi®cant increases in heart rate, systolic and diastolic blood pressure in both groups after both 50 45 40 35 30 25 20 15 10 5 0

Glaucoma group

2 ex min tu a ba ft tio er n

4.8 (4.5) 13 : 7 14.5 (9.3) 15.1 (4.00)

Normal group

3 ex 0 s tu af ba te tio r n

Age; years Sex ratio; M : F Weight; kg Duration of surgery; min

Normal group

extubation. Values are given as mean (SD) [range].

E Ba xtu se bat lin ion e

group

Table 3 Intra-ocular pressures before and after intubation and

Time of IOP measurement Figure 1 Intra-ocular pressures (IOP) measured 5 min after induction of general anaesthesia, 30 s and 2 min after intubation, before

extubation and 30 s and 2 min after extubation. Error bars indicate SD. *Signi®cantly different from respective baseline values, p < 0.01.

382

Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

Table 5 Haemodynamic parameters before and after intubation and extubation. Values are given as mean (SD) [range]. Intubation

Extubation 30 s after intubation

Baseline Normal group Heart rate; beat.min

1

Systolic blood pressure; mmHg Diastolic blood pressure; mmHg

Glaucoma group Heart rate; beat.min

1

Systolic blood pressure; mmHg Diastolic blood pressure; mmHg

2 min after intubation

Baseline

30 s after extubation

2 min after extubation

110.9 (9.8) [79±142] 106.4 (13.9) [68±130] 62.1 (12.3) [40±84]

130.7 (27.7)² [81±181] 121.7 (23.0)² [72±164] 74.5 (14.7)² [44±94]

113.6 (27.0) [84±172] 112.6 (15.2)* [86±152] 68.3 (14.5)* [48±91]

137.7 (18.9) [67±157] 112.7 (8.2) [88±129] 70.0 (9.9) [51±91]

129.9 (21.6)² [103±169] 129.9 (15.0)² [89±158] 87.1 (11.4)² [56±107]

[101±176] 123.9 (13.6)² [94±156] 82.0 (9.4)² [62±103]

116.1 (15.3) [88±150] 100.7 (24.1) [68±143] 68.6 (13.9) [45±86]

135.9 (14.3)² [120±170] 125.6 (17.4)² [102±148] 79.4 (13.0)² [54±96]

131.7 (15.1)² 106±166] 126.0 (19.1)² [101±167] 76.3 (14.0)² [60±100]

123.7 (27.9) [84±176] 119.6 (11.9) [96±137] 72.5 (13.1) [56±98]

139.7 (25.4)² [96±168] 141.3 (16.6)² [116±166] 89.1 (12.2)² [72±102]

148.2 (27.9)² [90±181] 139.7 (22.0)² [116±168] 84.5 (16.00)² [69±100]

*Signi®cantly different from baseline values, p < 0.05. ² Signi®cantly different from baseline values, p < 0.01.

Table 6 Airway complications occurring after extubation.

Coughing Straining Breath-holding Laryngospasm

Normal group (n ˆ 20)

Glaucoma group (n ˆ 15)

6 5 2 0

4 4 2 0

Discussion

Our main ®ndings were that the increase in IOP was signi®cantly greater after intubation in the glaucoma group than in the normal group, and that there was a greater increase in IOP in both groups after extubation than after intubation. The baseline values in this study were measured when anaesthesia had been maintained with an endtidal concentration of halothane of 0.5% for 5 min. Our results, including a lower absolute baseline IOP value, are in accordance with those of Watcha et al. [7] who used an identical anaesthetic technique, and measured baseline IOP after 10 min of anaesthesia. However, regardless of the time before the measurement of baseline IOP, the two studies are in agreement as there was a statistically signi®cant increase in IOP associated with intubation. The baseline IOP before extubation was close to the maximum IOP reached after intubation (Fig. 1). Since we did not measure IOP continuously during surgery, it is dif®cult to say whether it was steady at this level throughout. Normally, IOP decreases towards baseline values within 7 min of an increase associated with Q 2000 Blackwell Science Ltd

intubation and the use of suxamethonium [8]. In our study, the baseline IOP before extubation might have been increased because of the decreasing depth of anaesthesia at this time. In patients with normal IOP, Lamb et al. [4] observed that the most marked increase in IOP occurred after extubation. No previously published study has quanti®ed the increase of IOP in glaucomatous eyes at extubation. Glaucomatous patients were shown to have an exaggerated response to intravenous ketamine in the form of an increase in IOP when compared with normal patients [9]. Even a small increase in IOP for a short time, particularly in patients with severe closed-angle glaucoma, can compromise optic disc perfusion, and may result in disc ischaemia and loss of vision [4]. In this study, although the increase in IOP after extubation was comparable in the two groups, the glaucomatous patients would be more prone to ocular damage than the control group because of the higher baseline IOP values. This reinforces the need to use techniques that minimise increases in IOP when managing patients with glaucoma. Deep anaesthesia can reduce a high IOP by > 15 mmHg [5]. Extubation under deep anaesthesia may, therefore, be expected to prevent increases in IOP in glaucomatous patients. There are reports that suggest that awake extubation might not lead to a substantially different haemodynamic response to extubation under anaesthesia [11]. This has not been evaluated in glaucomatous patients. The hazards of extubation in a deep plane of anaesthesia are well known, and include upper airway obstruction, laryngospasm and pulmonary aspiration. The use of drugs, such as lidocaine [12], beta-blockers [13] and alfentanil [14], has 383

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

been tried to reduce the extubation pressor response. However, the IOP changes were not assessed in these studies. Further studies are required to ®nd out whether these techniques may be useful to attenuate the extubation response in glaucomatous patients. Coughing may increase IOP to > 50 mmHg [15] and should be avoided, particularly in glaucomatous patients. There was no difference in the incidence of coughing and breath-holding between the groups. In conclusion, the increase in IOP during intubation was signi®cantly exaggerated in the glaucomatous eyes compared with the normal eyes. The increase in IOP at extubation was more than at intubation both in normal and glaucomatous eyes. Even though these responses were similar in the two groups, there is a greater risk of damage to vision in the glaucomatous patients because of the preexisting intra-ocular hypertension. References 1 Holden R, Morsman CDG, Butler J, et al. Intra-ocular pressure changes using the laryngeal mask airway and tracheal tube. Anaesthesia 1991; 46: 922±4. 2 Brain AIJ. The laryngeal mask ± A new concept in airway management. British Journal of Anaesthesia 1983; 55: 801±5. 3 Abbott MA, Samuel JR. The control of intra-ocular pressure during induction of anaesthesia for emergency eye surgery. A high dose vecuronium technique. Anaesthesia 1987; 42: 1008±12. 4 Lamb K, James MFM, Janicki PK. The laryngeal mask airway for intra-ocular surgery: Effects on intra-ocular pressure and stress responses. British Journal of Anaestheisa 1992; 69: 143±7. 5 McGoldrick KE. Anesthesia and the eye. In: Barash PG, Cullen BF, Stoelting RK, eds. Clinical Anesthesia. Philadelphia, PA: Lippincott-Raven, 1997; 911±28.

384

6 Barclay K, Wall T, Wareham K, Asai T. Intra-ocular pressure changes in patients with glaucoma ± comparison between the laryngeal mask airway and tracheal tube. Anaesthesia 1994; 49: 159±62. 7 Watcha MF, White PF, Tychsen L, Stevens JL. Comparative effects of laryngeal mask airway and endotracheal tube insertion on intra-ocular pressure in children. Anesthesia and Analgesia 1992; 75: 355±60. 8 Pandey K, Badola RP, Kumar S. Time course of intraocular hypertension produced by suxamethonium. British Journal of Anaesthesia 1972; 44: 191±5. 9 Kaul HL, Gode GR. Effect of ketamine on intra-ocular pressure. Eastern Archives of Ophthalmology 1974; 2: 247±50. 10 Dillion FX. Safety considerations for otolaryngologic surgery: anaesthesia for otolaryngologic and head and neck surgery. Anesthesiology Clinics of North America 1993; 11: 637±49. 11 Patel RI, Hannallah RS, Norden J, Casey WF, Verghese ST. Emergence airway complications in children: a comparison of tracheal extubation in awake and deeply anaesthetized patients. Anesthesia and Analgesia 1991; 73: 266±70. 12 Lerman J, Kiskis AA. Effects of intravenous lidocaine and high dose pancuronium on intra-ocular pressure in children. Anesthesia and Analgesia 1985; 64: 245. 13 Dyson A, Isaac PA, Pennant JH, Giesecke AH, Lipton JM. Esmolol attenuates cardiovascular responses to extubation. Anesthesia and Analgesia 1990; 71: 675±8. 14 Crawford DC, Fell D, Achola KJ, Smith G. Effects of alfentanil on the pressor and catecholamine responses to tracheal intubation. British Journal of Anaesthesia 1987; 59: 707±12. 15 Guedes Y, Rakotoseheno JC, Leveque M, Mimouni F, Egreteau JP. Changes in intra-ocular pressure in the elderly during anaesthesia with propofol. Anaesthesia 1988; 43 (Suppl.): 58±60.

Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

F O RU M

The relationship between change in blood pressure, blood pressure and time M. J. Harrison,1* M. T. Kluger2 and N. N. Robertson1 Specialist Anaesthetists, 1 Department of Anaesthesia, Auckland Hospital and 2 Department of Anaesthesia, North Shore Hospital, Auckland, New Zealand Summary

Physiological homeostatic mechanisms and interventions by anaesthetists attempt to moderate excessive change in many biological variables during anaesthesia. These mechanisms may have fast or slow response times. This study describes how mean arterial blood pressure changes with time and how the change is dependent upon the pre-existing blood pressure. The results demonstrate the `regression towards the mean' concept; low arterial blood pressures increase and high pressures decrease. The data are the result of all interactions and have been used to produce an `envelope' into which 80% of all changes fall. Alarm systems using this envelope could warn of excessive changes that occur within short time intervals. Keywords Measurement, cardiovascular system: blood pressure. ...................................................................................... Correspondence to: Dr M. J. Harrison Accepted: 29 September 1999

Physiological changes during anaesthesia vary from patient to patient; furthermore, this variability changes from one phase of the operative procedure to another. For many years, investigators examining changes in arterial blood pressure during anaesthesia have used arbitrary changes from patients' baseline systolic arterial blood pressure and percentage changes in mean arterial blood pressure (MAP) [1] to indicate a signi®cant change or need for intervention. Arterial blood pressure is labile over short periods and this lability incorporates a diurnal rhythm [2, 3]; it is therefore dif®cult to say what is the patient's baseline arterial blood pressure. The variability of arterial pressure, in hypertensive patients, is important in the long-term as it is signi®cantly related to end-organ damage [4]. Basing `signi®cant change' on a proportion of the pre-operative arterial pressure is even more misleading when the preoperative blood pressure is commonly measured in a stressed and anxious patient. This study provides some data for the construction of a template that describes how blood pressure changes with time and with the recent pre-existing arterial blood pressure. Q 2000 Blackwell Science Ltd

Methods

With Auckland Ethics Committee approval, and informed consent from the patients, physiological data were collected at 10-s intervals using an AS3 Datex Anaesthesia Monitor (Datex-Ohmeda, Helsinki, Finland). The anaesthetists responsible for the anaesthetics were not involved in the collection of the data and their goal was to try to maintain cardiovascular stability. The patients were scheduled for craniotomy for nonaneurysmal surgery, and the data were collected between induction of anaesthesia and the raising of the skull ¯ap. Prior to the induction of anaesthesia a 20-g arterial cannula was inserted under local anaesthesia. A standard anaesthetic technique was used consisting of propofol 1.5± 2.5 mg.kg 1, fentanyl 2 mg.kg 1 and vecuronium 0.1 mg. kg 1. Alfentanil (1 mg) was given prior to intubation. Anaesthesia was maintained with sevo¯urane (end-tidal concentration 1.0±2.2%) in oxygen-enriched air; nitrous oxide was not used. The data were downloaded to a computer and analysed off-line. The collected data (MAP) were sorted to determine 385

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

changes in MAP over ®ve time intervals (10, 30, 60, 120 and 300 s). The 300-s period can be considered to be a moving window that is advanced at 10-s intervals throughout the data collection period, and the ®rst MAP value in this window is termed the `initial MAP' for that 300-s period. The data were further sorted to group the `initial' MAP values into separate 10 mmHg bands (50±59, 60± 69, 70±79, etc.). Once sorted in this way, the average change and the 10th and 90th percentiles of the change in each band were calculated. Results

Results were collected from 20 patients (®ve men, 15 women), with ages ranging from 26 to 66 years. Figures 1±3 are examples of three levels of initial MAP, low (50±59 mmHg), medium (90±99 mmHg) and high (130±139 mmHg), and they demonstrate the changes in MAP that occurred over time from these different initial MAPs. Table 1 lists the changes in MAP that occurred during the ®ve time intervals from different initial MAPs. Not all the patients had initial MAPs within all bands. The results show that high arterial blood pressures decrease and low blood pressures increase. If the MAP is between 90 and 99 mmHg then the decreases and increases are approximately equal. Discussion

This study considered the patient, anaesthetic and drug interaction complex as a `black box'. The data result from all interactions and have been used to produce an `envelope' into which 80% of all changes fall. The changes described are those associated with this particular group of patients having a speci®c type of neurosurgical anaesthesia. However, the concept is applicable to all forms of anaesthesia. Alarm systems using this `envelope' technique

Figure 1 Changes in MAP (mean, 10th and 90th percentiles),

from an initial MAP between 50 and 59 mmHg, over time (s).

386

Figure 2 Changes in MAP (mean, 10th and 90th percentiles),

from an initial MAP between 90 and 99 mmHg, over time (s).

could warn of excessive changes that occur within short time intervals. To say that a clinically signi®cant change is either a ®xed percentage change from `baseline' blood pressure, or a 20-mmHg absolute change, is no longer valid. An algorithm for assessing MAP lability has been described previously [5], but this involves a ®xed 2-min epoch. Our `envelope' may be more ¯exible as it can highlight both rapid- and slow-onset changes. In 1992, van Oostrom et al. [6] suggested that `knowing the clinical operating range (COR) of physiological variables is important in developing new monitor alarm strategies' and that `patients' vital signs change as a consequence of drugs, mechanical ventilation, body position, and surgical perturbations'. They asked resident anaesthetists for their COR for vital signs during the different phases of anaesthesia and then proceeded to investigate how often these CORs were transgressed. Transgressions were `common' for blood pressure and heart rate, and it

Figure 3 Changes in MAP (mean, 10th and 90th percentiles),

from an initial MAP between 130 and 139 mmHg, over time (s). Q 2000 Blackwell Science Ltd

Q 2000 Blackwell Science Ltd

The values for n are given for the 60-s interval data as an indicator of the number of blood pressure changes calculated from the patients. n changes slightly with the interval over which the measurement is made.

26.3 27.5 27.3 21.2 17.7 14.1 4.2 8.3 13.1 10.4 8.1 6.0 1.4 2.5 8.0 17.0 26.1 35.3 1.0 4.9 10.2 15.7 21.3 28.2 40.2 50.3 56.9 26.6 20.4 20.7 15.8 13.3 12.9 7.1 1.6 3.3 7.7 4.9 3.3 0.5 1.3 4.9 9.4 13.8 24.8 2.4 5.0 9.0 11.9 15.6 22.2 28.1 33.4 46.1 18.1 14.2 14.9 11.3 10.3 11.3 7.9 5.9 3.8 5.4 2.6 2.2 0.2 0.9 2.6 5.5 8.4 14.8 3.0 5.1 7.3 10.0 12.0 15.5 19.4 23.6 33.3 10.5 8.1 8.5 8.5 7.0 8.2 7.4 5.4 4.7 2.8 1.5 1.0 0.2 0.4 1.7 2.2 4.8 6.0 3.6 4.8 6.2 7.8 8.8 11.3 13.0 17.2 20.7 8.0 5.8 6.0 6.4 6.4 6.4 7.2 6.0 5.2 3.7 1.2 0.9 0.6 0.4 0.5 0.7 2.7 2.8 1.0 3.8 4.4 5.1 5.7 7.3 8.4 10.0 11.3 439 1937 2267 2228 1919 1151 648 297 135 14 15 20 20 20 19 19 17 12

50±59 60±69 70±79 80±89 90±99 100±109 110±119 120±129 130±139

Mean 10th % 10th %

Mean

90th %

30 s 10 s

Initial MAP (mmHg) n (60 s) No. of patients

Changes in MAP (mmHg) over time (s)

Table 1 Changes in MAP (mmHg) with time (s) for different levels of MAP

90th %

60 s

10th %

Mean

90th %

120 s

10th %

Mean

90th %

300 s

10th %

Mean

90th %

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

would follow that CORs would, therefore, not reduce the incidence of alarm activation. Although they used the word `change' in their article, they did not measure the size of the changes that took place, but rather the absolute values of the physiological variables. Our study addresses the change in MAP during anaesthesia. Variability of blood pressure has been cited in many studies as a cause of morbidity, although the validity of this conclusion has been questioned [7, 8]. The aim of our study was to manipulate the raw blood pressure data, such that uncommon changes in MAP could be highlighted and thus the adequacy of anaesthesia assessed. A signi®cant increase in MAP may indicate that the depth of anaesthesia is inadequate and more analgesia may be required. A decrease that transgresses the 10th percentile (or the ®fth, depending on the clinician's choice) has to be seen in the light of recent drug administration, changes in body posture or continuing blood loss. The monitors themselves could generate the database and the percentiles, which could be automatically updated after each measurement. In conclusion, this study has shown that it is possible to quantify changes in MAP and that the magnitude and direction of change is dependent on the recent past MAP. Any change highlighted by the database may help in the decision as to whether clinical intervention is necessary.

References 1 Rao TKL, Jacobs KH, El-Etr AA. Reinfarction following anaesthesia in patients with myocardial infarction. Anesthesiology 1983; 59: 499±505. 2 Kobrin I, Oigman W, Kumar A, et al. Diurnal variation of blood pressure in elderly patients with essential hypertension. Journal of the American Geriatrics Society 1984; 32: 896±9. 3 Gross M. Diurnal blood pressure variation in cerebrovascular disease. Annals of Internal Medicine 1970; 72: 823±33. 4 Parati G, Di Rienzo M, Ulian L, et al. Clinical relevance of blood pressure variability. Journal of Hypertension 1998; 16: S25±S33. 5 Reich DL, Osinski TK, Bodian, C, et al. An algorithm for assessing intraoperative mean arterial pressure lability. Anesthesiology 1997; 87: 156±61. 6 van Oostrom JH, Gravenstein C, Beneken JEW, Gravenstein JS. Improving alarm systems: How to de®ne acceptable vital signs during general anaesthesia. Journal of Clinical Monitoring 1992; 8: 159±60. 7 Reves JG, Smith LR. From monitoring to predicting outcome. Annals of Surgery 1990; 212: 559±60. 8 Roizen M. Diseases involving the cardiovascular system. In: Miller RD, ed. Anesthesia, 3rd edn. New York: Churchill Livingstone, 1990; 820±4. 387

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

F O RU M

Visible and occult blood contamination of laryngeal mask airways and tracheal tubes used in adult anaesthesia M. R. J. Parker1* and C. J. E. Day2 1 Locum Consultant in Anaesthesia, Department of Anaesthesia, Bristol Royal In®rmary, Bristol BS2 8HW, UK 2 Consultant in Intensive Care Medicine and Anaesthesia, Department of Anaesthesia, Royal Devon and Exeter Hospital, Exeter, UK Summary

The purpose of this study was to compare visible and occult blood contamination of 50 laryngeal mask airways and 50 tracheal tubes following routine anaesthesia for procedures not involving the oropharyngeal or nasal cavities. All airway devices were examined visually for the presence of blood before washing in 100 ml of water. A semiquantitative dipstick was used to test for the presence of blood in the washings. Laryngeal mask airways were examined visually by both authors to test agreement. The results show that occult blood contamination occurred in 78% of tracheal tubes and 76% of laryngeal mask airways, while visible blood contamination was 16% and 12%, respectively. Other studies reporting visible blood contamination of airway devices probably underestimate the true incidence of blood contamination. Oral secretions following the use of these devices should be considered as high risk for transmission of blood-born viruses. Anaesthetic and recovery staff should be protected against the risks of occupational exposure to oral secretions following the use of airway devices. Keywords Equipment: laryngeal mask airway; tubes; tracheal. Blood. Complications: infections. ...................................................................................... Correspondence to: Dr M. R. J. Parker Accepted: 20 October 1999

The high prevalence of previous exposure to hepatitis B virus among anaesthetic personnel before the introduction of a safe and ef®cient vaccine is partly explained by the high incidence of needle stick injury [1, 2]. There are guidelines to reduce the risk of such injuries [3]. Despite recommendations [4], anaesthetists do not always wear gloves for procedures that risk skin contamination by blood or saliva even when they have evidence of a clear breach in the integrity of the skin. A previous study demonstrated that during 100 anaesthetic sessions, there were 25 episodes in which potential contamination of an open skin lesion by oral secretions occurred [5]. The prevalence of previous hepatitis B virus positivity was suggested to be higher in anaesthetists who wear protective gloves only while performing sterile procedures [6]. The occupational risks of hepatitis C and human immuno388

de®ciency virus (HIV) are important and the modes of occupation transmission are similar to hepatitis B; however, at present, no effective vaccine exists. Tracheal tubes and laryngeal mask airways (LMAs) are used routinely in anaesthetic practice. Visible contamination of these devices with blood has been observed with a frequency of 3±22% [7±9], and they may contribute to contamination of other anaesthetic equipment if they are not disposed of appropriately. Anaesthetic equipment, especially if it is likely to be touched by the anaesthetist, commonly shows evidence of blood contamination [10]. Hepatitis B virus continues to be infectious for at least 1 week in dried blood on external surfaces [11]. There is evidence that anaesthetists alter their practice when they are aware of an increased occupational risk of exposure to blood-born hazards [12, 13]. Q 2000 Blackwell Science Ltd

Anaesthesia, 2000, 55, pages 367±390 Forum ................................................................................................................................................................................................................................................

We hypothesised that the true incidence of blood contamination of LMAs and tracheal tubes is much higher than described, and that the visible presence of blood has a low sensitivity for the actual presence of blood contamination. We therefore proposed the null hypothesis that there is no difference between visible and true blood contamination of LMAs and tracheal tubes. The purpose of this study was to investigate the incidence of both visible and invisible blood contamination of LMAs and tracheal tubes, and whether the method and ease of insertion or removal, and the experience of the anaesthetist, in¯uence contamination of LMAs. Methods

This was an observational study, not affecting normal patient care, therefore informed patient consent was not required. Consecutive tracheal tubes and LMAs were studied following removal from patients undergoing general anaesthesia for surgery which did not involve the mouth, the nose or insertion of a nasogastric tube, and in which the criteria described below were met. LMAs were inserted and removed according to the normal practice of the anaesthetist. The ease of insertion was graded as either easy or dif®cult. We de®ned a dif®cult insertion as the necessity to either remove and replace the LMA for any reason, reposition of an LMA because of an inadequate airway or administration of further boluses of induction agent to depress the upper airway protective responses. Techniques to assist insertion were de®ned as any method of insertion other than that described by the instructions for use [14]. Removal by the anaesthetist or patient was recorded. Tracheal tubes were inserted following non-rapid sequence induction of anaesthesia with a normal intubating dose of a non-depolarising neuromuscular blocker. Laryngoscopy was performed using a standard length Macintoch bladed laryngoscope. All patients were grade 1 or 2 laryngoscopies. At the end of the surgical procedure, tubes were removed by the anaesthetist according to their normal practice. All tubes and LMAs were immediately stored in a sealed clear plastic bag to prevent drying of the secretions and were examined for blood contamination within 2 h of removal from the patient. Visible blood contamination of the tracheal tube or LMA and was con®rmed by semiquantitative dipstick for haemoglobin (see below). All LMAs were examined separately by both authors to test agreement. Following visual examination, tubes were washed in 100 ml of water for not less than 2 min to ensure haemolysis of adherent blood. A semiquantitative dipstick for Q 2000 Blackwell Science Ltd

Table 1 Visible and invisible blood contamination of tracheal

tubes and laryngeal mask airways

Visible blood Dipstick positive Dipstick negative

Tracheal tubes (n ˆ 50)

Laryngeal mask airways (n ˆ 50)

n

%

n

%

8 39 11

16 78 22

6 38 12

12 76 24

For each category of contamination, there was no signi®cant difference between tracheal tubes and laryngeal mask airways. Within each group, there was a signi®cant (p < 0.0001) difference between visible and Haemostick positive blood contamination.

haemoglobin (Haemostick) was used to detect or con®rm blood contamination in the washing ¯uid. The lower limit of sensitivity of the Haemostick was 0.125 ml of whole blood (unpublished data using blood taken from a volunteer) and an unused LMA and tracheal tube were tested to con®rm negative controls. Results are expressed as number and percentage where appropriate. A Chi-squared test with Yates' correction was used to measure statistical signi®cance and p < 0.05 was considered signi®cant. Results

Over an 8-week period, 50 LMAs and 50 tracheal tubes were studied. Table 1 summarises the incidence of visible blood contamination, Haemostick positive and Haemostick negative tubes and LMAs. There was a signi®cant (p < 0.0001) difference between visible and Haemostick positive contamination of both LMAs and tubes. However, there was no difference between tubes or LMAs for either visible or Haemostick positive contamination. There was complete agreement between examiners for the visible blood contamination of LMAs. Table 2 summarises blood contamination of LMAs with grade of anaesthetist, ease of insertion and method of removal. Only one anaesthetist used a technique to aid insertion of an LMA (two gloved ®ngers in the patient's mouth to guide the LMA into position). The grade of anaesthetist, ease of insertion and method of removal did not appear to in¯uence contamination. Discussion

We rejected our null hypothesis on the basis of the results. This study demonstrates not only that blood contamination of both tracheal tubes and LMAs is extremely common (< 75%), but that it is more likely to be invisible to the 389

Forum Anaesthesia, 2000, 55, pages 367±390 ................................................................................................................................................................................................................................................

Table 2 Visible and invisible blood contamination of 50 laryngeal mask airways by grade of anaesthetist, ease of insertion and method of

removal

Visible blood Dipstick positive Dipstick negative

Grade of anaesthetist

Insertion

Consultant (n ˆ 29)

Trainee (n ˆ 21)

Easy (n ˆ 43)

n

%

n

%

n

%

n

%

n

%

n

%

3 21 8

10 72 28

3 17 4

14 81 19

5 33 10

12 77 23

1 5 2

14 71 29

0 2 3

0 40 60

6 36 9

13 80 20

naked eye. The frequency of visible blood contamination in this study was similar to the results of other studies [7±9]. However, since our study has demonstrated that the visible presence of blood is an insensitive guide to true contamination, it is likely that these other studies of various airway devices [7±9] also underestimate the true risk of blood contamination. It is possible that the differences between devices may simply re¯ect a low sensitivity of the observer to identify contamination. In view of these ®ndings, even in the absence of visible blood, all oral secretions following the use of an LMA or tracheal tube must be treated as blood contaminated. It has been suggested that saliva may represent a lower infection risk for HIV [15]. However, it is clear that blood should always be considered as high risk. Therefore, all oral secretions following the use of a LMA or tracheal tube should be considered as a high infection risk. This has important implications for both anaesthetic and recovery staff. We suggest that protective gloves, mouth and eye protection should be worn during the removal of a tube, LMA or other airway device and this should probably be extended into the immediate recovery period when a patient may cough and produce a ®ne spray of oral secretions. It is also important not to contaminate other anaesthetic equipment by improper disposal of a used tube or LMA or any suction device used to aspirate the patient's mouth. References 1 Berry A, Isaacson I, Hunt D, Kane M. The prevalence of hepatitis B viral markers in anesthesia personnel. Anesthesiology 1984; 60: 6±9. 2 Sinclair M, Ashby M, Kurtz J. The prevalence of serological markers for hepatitis B virus infection amongst anaesthetists in Oxford region. Anaesthesia 1987; 42: 30±2. 3 UK Health Departments. Guidelines for Clinical Care Workers Protection Against HIV and Hepatitis Viruses. London: Her Majesty's Stationery Of®ce, 1990.

390

Removal Dif®cult (n ˆ 7)

Patient (n ˆ 5)

Anaesthetist (n ˆ 45)

4 Association of Anaesthetists. HIV and Other Blood Born Viruses: a Guidance for Anaesthetists. London: Association of Anaesthetists, 1992. 5 Parker M. The use of protective gloves, the incidence of ampoule injury and the prevalence of hand laceration amongst anaesthetic personnel. Anaesthesia 1995; 50: 726±9. 6 Berry A, Isaacson I, Kane M, et al. A multi centre study of the epidemiology of hepatitis B in anesthesia residents. Anesthesia and Analgesia 1985; 64: 672±6. 7 Brimacombe JR, Brimacombe JC, Berry A, et al. A comparison of the laryngeal mask airway and cuffed oropharyngeal airway in anesthetized adult patients. Anesthesia and Analgesia 1998; 87: 147±52. 8 Cork R, Depa R, Standen J. Prospective comparison of use of the laryngeal mask and endotracheal tube for ambulatory surgery. Anesthesia and Analgesia 1994; 79: 719±27. 9 Dingley J, Whitehead M, Wareham K. A comparative study of the incidence of sore throat with the laryngeal mask airway. Anaesthesia 1994; 49: 251±4. 10 Hall J. Blood contamination of anesthesia equipment and monitoring equipment. Anesthesia and Analgesia 1994; 78: 1136±9. 11 Bond W, Favero M, Peterson N, Gravelle C, Ebert J, Maynard J. Survival of hepatitis B virus after drying and storage for one week. Lancet 1981; 1: 550±1. 12 O'Donnell N, Asbury A. The occupational hazard of human immunode®ciency virus and hepatitis B infection. 1. Perceived risks and preventive measures adopted by anaesthetists: a postal survey. Anaesthesia 1992; 47: 923±8. 13 O'Donnell N, Asbury A. The occupational hazard of human immunode®ciency virus and hepatitis B virus infection. II. Effect of age, sex and region of employment on perceived risks and preventive measures adopted by anaesthetists. Anaesthesia 1992; 47: 929±35. 14 Brimacombe J, Brain A, Berry A. The Laryngeal Mask Airway Instruction Manual, 4th edn, pp. 14±18. Reading, UK: Intravent Research Limited, 1999. 15 Shugars D, Wahl S. The role of the oral environment in HIV 1 transmission. Journal of the American Dental Association 1998; 129: 851±7.

Q 2000 Blackwell Science Ltd