Burns due to pulse oximetry

We pesume that a small mucosal tear deither with insertion or movement of the nasal catheter and that continual oxygen flow caused the surgical emphysema. Surgical emphysema of this origin could well compromise the airway and is another potentially serious complication of oxygen therapy delivered by this route.

lbe -’fle Although the aims of prwedication may be disputed, whether they be anxiolytk, amnesic,sedative or ‘narcotic’, or indeed whether premeditation is necessary at all, a large body of clinical opinion recognises a ‘well-premedicated‘ patient The possibility of repeated anaesthesia really necessitates that the ‘&ectiveness’ of premedication be recorded objectively. The following Scoring system has been elabor-

Queen Mary’s Hospital for Children, Carshalton, SurreySM54NR

P. NEWMAN

P. RADFORD B.J. EPPEL

of premedcation ated over the years and has been found to be most useful: 0, no effect; I , subjective effect only; 2, subjective and objective effect but awake; 3, sleeping but responds to normal conversation; 4, not responsive to normal conversation. The Royal Surrey County Hospital, Guildford GU2 5 X X

J.M. LEIGH

Smbmtmmm opioids: the painles approach

Dr Teillol-Foo’s observations on the use of an indwelling intramuscular cannula for postoperative analgesia (Anaesthesia 1991; 46c 897) are interesting and should certainly be consided before committing patients to repeated intramuscular injections. If one of the aims of the anaesthetist is to relieve pain, we should not accept that the barbaric idlidion of a 20 gauge needle (or 21 gauge as we would use) is justified by its end result. In central Nottinghamshire we have recently introduced an acute pain service and, as in many other centres, we make wide use of patient controlled analgesia, but as we have a paucity of equipment we run a subcutaneous regimen alongside. This involves an indwelling cannula, usually a 20 gauge Venflon, which is sited subcutaneously during anaesthesia, either on the chest wall or the anterior thigh (especially ifa regional block has been used). Nurses are requested to give a variable dose of opioid (usually morphine) according to a rigid protocol. The success of this method depends upon the enthusiasm of the nurses, who willingly assess patients’ pain hourly and give appropriate quantities of drug, despite some difficulties with the Controlled Drug Register. Motivation is enhanced by their observing a great improvement in postoperative pain relief. All risk of &stick injury at the patient site is avoided as the injection is given via the V e d o n port; the small ‘deadspace’ is usually only relevant at the first administration and, as this dose is usually the largest, say 1 ml, this should not be significant. Nurses are instructed not to dilute drugs not to flush the cannulae. It has been shown that absorption from both subcuta-

neous and intramuscular sites follows a similar time course and that the absorption profile in a particular patient is more reproducible via the same subcutaneous site than via several different intramuscular ones. The actual injection may not be entirely painless and some patients do describe a mild irritation at the cannula tip, although all questioned would agree that the regimen is far superior to the intramuscular method. Unfortunately, we have not found that the regimen produces any significant reduction in nausea and vomiting and, as antiemetics cause a local skin reaction when given subcutaneously, such injections cannot be combined with the opioid, except perhaps when using low dose droperidol. It may, of course, be necessary to administer an occasional bolus dose of opioid to a patient who does not have an indwelling cannula. We believe that such injections could also be prescribed subcutaneously and, therefore, we can see no further role for intramuscular opioids, unless they are to be combined with an antiemetic. We freely admit that we are not the first to advocate subcutaneous injections of opioids and perhaps not the first to suggest the use of an indwelling cannula, but, from our recent experience we would confidently recommend this technique be used more widely, even by our more sceptical immediate colleagues. Comparison of this regimen with that of Dr Teillol-Foo would be worthwhile. ManSJield Hospitals, Nottinghamshire

P.T. BULL M.J. MOWBRAY S.J. MARKHAM

Burns due to pulse oximetry

A 4-month-old baby, whose lungs were ventilated for 14 days for ventilatory failure secondary to bronchiolitis, suffered burns to his big toe related to the site of application of a pulse oximeter probe (Fig. I). The probe involved was a Nellcor N25 oxisensor and was combined with a Hewlett Packard HP Model 66s monitor and module. Prior to the discovery of the bum, the patient had no periods of poor peripheral perfusion and there had been no problems with the pulse oximeter trace. On the 13th day he was noted to have blistering beneath the pulse oximeter probe on two-thirds of the circumference of his left big toe, and a similar but smaller lesion on the right big toe. These

had not been present during examination earlier in the day and had not been noticed during routine rotation of the position of the probe. This consists of movement of the probe to another limb every 3 h and is standard practice for the nurses on the intensive care unit. The pulse oximeter probe was removed and the light source felt to be hotter than usual. The lesions have been managed conservatively and are resolving satisfactorily. The pulse oximeter probe was returned to the manufacturer for assessment. Following this episode, the heat production from a number of similar probes was tested. Six pulse oximeter probes were suspended in a Drager 8000 incubator main-

Correspondence

Fig. 1.

tained at 37.0"C, and switched on. Hewlett Packard (HP21078A) temperature probes were attached to the light source and the sensor of the pulse oximeter probe and recorded for 3 h using the HP66S temperature monitor. The temperature of the incubator remained constant between 36.9 and 37.0"C. The pulse oximeter probes warmed rapidly when brought in from room temperature to the incubator environment. The light sources quickly reached their working temperature, after which their temperature remained remarkably constant over the 3 h period. The working temperature varied between the indivi-

277

dual probes (range 39.1 to 39.7"C). The light sensors remained very close to the ambient temperature. The sixth probe was monitored for 24 h. The temperature of the light source remained constant, + - O.I"C, once it had reached its working level, for the whole 24 h. Pulse oximetry is not without its hazards. The baby in our report suffered burns due to short circuiting between the wires in the pulse oximeter probe lead. There was also some damage to the protective film over the light source. These defects were the result of the repeated use of a disposable probe, an unfortunate practice that had arisen due to the need to curb expenditure. We would suggest that pulse oximeter probes and connecting wires should be inspected frequently and although not seen in this case, any freezing of the pulse oximeter display should be investigated to aid the detection of short circuits in the wiring within the leads [I]. We would also suggest that the point of application should be regularly inspected and the position of the probe varied at regular intervals, especially during conditions of decreased skin perfusion. Finally, we would strongly recommend that disposable pulse oximeter probes are re-used only in strict compliance with the manufacturers recommendations. Royal Hallamshire Hospital, Gkossop Road, Shefleld

G.H. MILLS S.J. RALPH

Reference [ I ] SLOANTB. Finger injury by an oxygen saturation monitor probe. Anesthesiology 1988; 68:936-8.

Cardiovascular effects of nasotracheal intubation Drs Smith and Grewal (Anaesthesia 1991; 4 6 683-6) report that the cardiovascular changes of nasal intubation have not previously been compared with those of oral intubation. This is not true as we have already studied and published the results of such a study [I]. Another study investigated the metabolic responses of nasotracheal intubation [2]. In our study both nasal and oral intubation were accompanied by significant changes in systolic arterial pressure and heart rate. Within-group data were compared using analysis of variance, unlike Drs Smith and Grewal, who used multiple paired Student's t-tests. The latter is not appropriate for multiple comparisons since false statistically significant results may be obtained; ANOVA should be used instead. In contrast to our results, they found that arterial pressure associated with nasal intubation was significantly higher than that following oral intubation. Again, multiple unpaired Student's t-tests were applied for intergroup analysis, but this should have included Bonferroni's correction. Scheffe's method could also be applied for inter-group comparisons. With regard to heart rate following nasal intubation, we observed peak values 2-3 min after intubation, while peak values following oral intubation occurred 1 min after intubation [I]. Smith and Grewal showed that nasal intubation was associated with significantly less tachycardia than oral intubation. The authors assume that the slower heart rate during the first minute after nasal intubation may be mediated by baroreceptor reflexes in response to a more rapid increase in arterial pressure associated with nasal intubation. However, they ignore the fact that anaesthesia depresses baroreceptor function [3-71. The degree of such depression produced by 20mg of temazepam followed I h later by thiopentone 5 mg.kg-',

50% N,O in 0, and 1YOisoflurane is rather unpredictable! lnvolvement of the trigeminovagal reflex may be an explanation, although the influence of general anaesthesia on this reflex is probably unknown. We have demonstrated significantly higher VO, values during nasal intubation than during oral tracheal intubation [I]. These results are not in accordance with the lower heart rate after nasal intubation when compared with that 1 min after oral intubation. Before extrapolating Smith and Grewal's conclusions to ITU patients and to patients suffering from cardiovascular disease more studies are required to assess whether nasal intubation is a more potent stimulus than oral intubation. St Savas Hospital, 171 Alexandras Ave, Athens 11522, Greece

A. FASSOULAKI

References FASSOULAKI A, ANDREOPOULOU K , SALEHM, KITHARATZI D. Metabolic and cardiovascular responses following oral and nasal intubation of the trachea. Acta Anaesthesiologica Belgica 1990; 41: 281-6. FASSOULAKI A, EFORAKOPOULOU M. VASSILIOU M. Metabolic responses (VO,, VCO, and energy expenditure) - associated with nasal intubation of the trachea. Anesthesia and Analgesia 1989; 68: 112-5. BRISTOWJD, PRYS-ROBERTS C, FISHERA, PICKERING TG, SLEIGHT P. EtTects of anesthesia on baroreceptor control of heart rate in man. Anesthesiology 1969; 31: 422-8. SKOVSTED M D . PRICEML, PRICEHL. The effects of short acting barbiturates on arterial pressure, preganglionic synaptic activity. and barostatic reflexes. Anesthesiology 1970; 33: 10-8. PKYS-ROBERTS C. Regulation of the circulation; In: The