549

CORRESPONDENCE

vention and treatment of surgical pain. Patients with early renal failure NSAIDs are best avoided, or at least renal funciton should be carefully monitored.

William Code MD FRCPC Department of Anaesthesia Cowichan District Hospital REFERENCES

pressure may further compromise perfusion particularly if the arterial pressure is labile. A tube that fits well at the start of a case may become too tight. It is the duration and the age of the patient that seem to be the major determinants of tracheal stenosis. Thus the ex-premature neonate whose trachea has been intubated for some days is at particular risk.S I know of no study of leak pressure in this population.

1 Murray MD, Brater D C Renal toxicity of the nonsteroi2 3 4 5

dal anti-inflamatory drugs. Ann Rev Pharmacol Toxicol 1993; 33: 435-65. Day RO, Henry DA, Muirden KD, et al. Non-steroidal anti-inflammatory drugs induced upper gastrointestinal haemorrhage and bleeding. Med J Aust 1992; 157: 810-2. Moote CA. Complication associated with NSAIDs use for acute pain management. Perspectivesin Pain Management 1993; 3: 3-6. Kehlet H, Dahl JB. Are perioperative nonsteriodal antiinflammatory drugs ulcerogeniein the short term? Drugs 1992; 5: 38-41. Aitken HA, Burns JVg McArdle CS, Kenny GN.. Effects of ketorolac trometamol on renal function. Br J Anaesth 1992; 68: 481-5.

J . H . S m i t h MB MRCP FRCA

Newcastle-upon-Tyne England REFERENCES 1 Koka BV, Jeon IS, Andre JM, MacKay I, Smith RM.

2

3

Tracheal tube leak test To the Editor: The paper on the leak test by Schwartz et al. was interesting. The conclusion that it is unreasonable to set an upper limit of leak pressure for changing all endotracheal tubes is contradicted by their results. Of the 242 patients studied 30 (12%) were excluded because one or both examiners found that no leak occurred at 50 cm H20. Does this mean the authors would consider a leak at 50 (4-18) cm H20 unacceptable? Does the finding that the leak measured does not vary between observers invalidate the test? The need to determine fit of the endotracheal tube in the trachea and at the level of the cricoid ring derives from the fear that too tight a tube will cause pressure changes of the tissues at the level of the cricoid. These changes may cause transient problems such as stridor after extubation ! or may be directly related to later tracheal stenosis. 2 The fact that a leak is detected at approximately 25 (4-9.5) em H20 means that a pressure equal to the leak pressure is being exerted on the tracheal mucosa at the cricoid ring. This pressure should not exceed the perfusion pressure of the tissues at that level. In adults this has been estimated as 30 cm H20 (22 mmHg). 3 If we no longer measure the leak pressure, there is no other way we can easily estimate how tight the tube is. The leak pressure may increase appreciably 4 in children that undergo cardiac surgery, this may be due to subglottic oedema as a reflection of generalised oedema. The increase in the

4 5

Post intubation croup in children. Anesth Analg 1977; 56: 501-5. Minnigerode B, Richter HG. Pathophysiology of subglottic traeheal stenosis in childhood. Prog Pediatr Surg 1987; 21: 1-7. Seegobin RD, van Hasselt GL Endotracheal cuff pressure and tracheal mucosal blood flow: endoscopic study of effects of four large volume cuffs. BMJ 1984; 288: 965-8. Smith JH, O'Kelly SW. Brain swelling after coronary artery surgery (Letter). Lancet 1993; 342: 1370-1. Weber TR, Connors RH, Tracy TF Jr. Acquired tracheal stenosis in infants and children. J Thorac Cardiovasc Surg 1991; 102: 29-34.

REPLY We thank Dr. Smith for his response to the paper entitled "Tracheal Tube Leak Test- Is There Inter-observer Agreement?" We stated that we believe it is unreasonable to set a rigid upper limit of leak pressure for the changing of all endotracheal tubes. We also suggested that otherfactors such as duration of surgery, difficulty of intubation, risk of aspiration of gastric contents, surgical positoin, and history of croup or other airway abnormalities be considered when deciding if tight-fitting endotracheal tubes should be changed. 1 The optimal limits of leak pressure used to avoid laryngeal injury are unknown. Koka et aL suggested using 25 cm HzO, Litman and Keon use 40 cm 1-120~ In this study it was found there is considerable variation between two experienced observers in assessing leak pressure. Rather than set a rigid upper limit, the peak test shouM be used as one of multiple factors employed to determine if the endotracheal tube should be changed. We agree with Dr. Smith that the plan to undergo cardiopulmonary bypass should be added to the list of factors used when deciding to change tight-fitting endotracheal tubes. Dr. Smith questions why we excluded patients in which one or both examiners found no leak at 50 cm of water. We were reluctant to raise the intrathoracic pressure to greater levels for fear of barotrauma. Since no specific leak pressure was assigned by one or both of the examiners in these patients, no interobserver difference could be calculated. In summary, we did not intend to suggest that the leak pressure no longer be measured. We suggest that the absolute

550 number determined for the leak pressure should be one of many factors used in deciding when to change a tight-fining endotracheal tube.

Roy E. Schwartz MD Stephen A. Stayer MD Caroline A. Pasquariello MD St. Christopher's Hospital for Children Philadelphia, Pa. REFERENCES

1 Schwartz RE, Stayer SA, Pasquariello CA. Tracheal tube

leak test - Is there inter-observer agreement?Can J Anaesth 1993; 40: 1049-52. 2 Koka BF, Jeon IS, Andre JM, MacKay I, Smith RM. Postintubation croup in children. Anesth Analg 1977; 56: 501-5. 3 Litman RS, K_eon TP. Postintubation croup in children (Letter). Anesthesiology1991; 75:1122-3.

Clonidine and postoperative myocardial ischaem& To the Editor: Quintin et al. recently suggested that clonidine, given as premeditation, might decrease the incidence of myocardial ischaemia in patients at risk. As the number of patients of this study was limited, firm conclusions depend on the results of other similar studies. We recently conducted a double-blind study to assess the effect of epidur:al clonidine on postoperative myocardial ischaemia. Forty-five A S A 3 patients scheduled for abdominal aortic surgery were included after informed consent and Ethics Committee approval. Surgery was performed under combined general anaesthesia (using propofol, fentanyl and isoflurane) and thoracic epidural anaesthesia (using 0.5% bupivacaine). Postoperatively, patients were randomly assigned to receive a continuous infusion of either fentanyl (0.4 Isg-kg -I- hr -I) (Group F (fentanyl group)) or fentanyl (0.4 ~tg. kg -I. hr -l) plus clonidine (0.3 ~g. kg -I. hr -l) (Group F + C (fentanyl + clonidine group)). Patients were monitored during the f ~ t 24 hr with an ST segment analyser (Merlin | Hewlett Packard) (leads II, V4, Vs) and pulmonary and radial artery catheters. Myocardial ishaemia was defined as an ST-segment decrease or increase > 1 mm from the base line for at least one minute. Hypotension was defined as a systolic arterial pressure <90 mmHg, hypertension as a systolic arterial pressure > 160 mmHg and tachycardia as an increase in heart rate > 110 bts/min. Haemodynamic events and myocardial ischaemia were treated according to the judgment of the physician in charge of the patient who was aware of the analgesic treatment given. Statistical analysis

CANADIAN JOURNAL OF ANAESTHESIA TABLE I Cardiovascular risk factors and previous treatments given to the patients

Angina Previousmyocardialinfarction Hypertension Diabetes mellitus ECG abnormalitiesindicativeof isehaemia Nitrates Betablockers Calciumchannelblockers Convertingenzymeinhibitor Diuretics Anticoagulants

Fentanyl

Fentanyl + clonidine

3 3 I1 1 7 2 6 7 2 2

4 1 9 3 5 1 2 7 4 4

6

7

TABLE II Number of episodes of ST changes associated with changes in heart rate and arterial blood pressure

Group F Group F+C

Hypotension

Hypertension Tachycardia

1 0

9 1

2 3

used the Mann-Whitney U test and the Fisher exact test. Results are expressed as mean + SD. Twenty-two patients were given fentanyl and 23 fentanyl + clonidine. Patients in the two groups were comparable: age (years) 65 5:2 vs 66 5: 9, sex ratio: 1 W/ 21M vs 3W/20M respectively in Groups F and F+C. The cardiovascular risk factors and previous treatments are reported in Table I. The duration of surgery (Group F : 210 + 114 rain, Group F + C : 238 + 118 min) and the time of aortic cross-clamping (Group F : 55 + 34 min, Group F + C : 78 + 40 min) were comparable in the two groups. Signiticant ST changes indicative of ischaemia were noticed in 11 patients in group F and 6 in group F + C (NS). The cumulative durations of ishaemia were respectively 87 5:31 min (range 3-503 min) and 23 + 25 min (5-60 min) in Groups F and F+C. Twenty-one and I1 ST changes occurred respectively in Groups F and F + C without simultaneous changes in blood pressure and/or heart rate. The distribution of ST-changes episodes associated with haemodynamic events is reported in Table II which demonstrates that ST changes associated with hypertension were more frequent in Group F ( P < 0.05). Clonidine has been previously documented to decrease sympathetic activity2 cateeholamines plasma levels 3 and to produce postoperative analgesia. 4 Clonidine may prevent myocardial ishaemia related to increased oxygen consumption induced by increased cardiac workload. This study therefore confirms previous data from Quin-