- Email: [email protected]
Ondansetron compared with metoclopramide in the treatment of PONV
British Journal of Anaesthesia 1998; 80: 406–410
CORRESPONDENCE
Anaesthesia after exhaustive exercise Sir,—There are no reports on the effect of preceding exhaustive exercise, such as marathon running, on the course of anaesthesia. We report an extreme case. A healthy 21-yr old (65 kg,173 cm) active jogger (40–60 km/week) presented for day-case tonsillectomy under general anaesthesia. Three days earlier he had run his first marathon and told the anaesthetist that his legs were painful. He had difficulty in lying down on the operating table. After operation, he told us that after 30 km of the marathon he had become so exhausted that he could remember little of the rest of the race. An i.v. anaesthetic was used, supplemented with 70% nitrous oxide in oxygen. The drugs given and their total doses were as follows: glycopyrrolate 0.2 mg, propofol 280 mg, fentanyl 300 g and rocuronium 50 mg. Residual neuromuscular block was antagonized at the end of operation with neostigmine 2.5 mg and glycopyrrolate 0.5 mg. During surgery, 2500 ml of crystalloid i.v. solutions were given. Postoperative analgesia was achieved with ketoprofen. After operation the patient was able to breathe satisfactorily but despite normal responses to peripheral nerve stimulation he appeared extremely weak, that is he was unable to lift his head or even his hand. Because his condition had not improved after 6 h, he was moved to the local university hospital. Thirteen hours after operation the patient had to be escorted to the toilet by two people. Next morning he was able to walk but still complained of muscle pain, especially in his legs. The muscle pain disappeared gradually over the next 3 days. Eight hours after operation total blood count, C-reactive protein concentrations, serum concentrations of creatinine, sodium, potassium and chloride, and blood-gas analysis, serum lactate, ammonium ion and insulin concentrations were normal. The only abnormal value was serum creatine kinase (1572 u. litre91, normal value :285 u. litre91), which was still high 2 days after operation (5 days after the marathon it was 338 u. litre91), but was within normal limits by 10 days after operation (114 u. litre91). The patient underwent 31P-magnetic resonance spectroscopy (MRS) 33 h after the beginning of anaesthesia to assess the status of skeletal muscle energy metabolism.1 2 He was re-examined 5 months later for comparison. The patient resumed his usual training schedule 1 month after surgery. From the MRS studies we found no evidence of altered energy metabolism of striated muscle to anaesthesia after exhaustive exercise. Variables assessing the main energy producing pathways (phosphocreatinine hydrolysis, glycolysis or aerobic oxidative phosphorylation) were similar immediately after the incident compared with 5 months later. In contrast, resting concentrations of phosphodiesters were substantially higher immediately after the incident (4.5 compared with 2.7 mmol litre91). Phosphodiesters are hypothesized to be indicators of release of membrane molecules in muscle cell damage.3 But muscle overuse is associated with damage to the muscle contractile elements.4 This damage to the contractile elements could also be an explanation for our phosphodiester results. After exhaustive exercise, a general anaesthetic could cause some type of interaction between the exhausted muscles and anaesthetic agents. Importantly, significant rhabdomyolysis may occur if succinylcholine is used.
M. ERKINTALO M. KOMU K. MATTILA Department of Diagnostic Radiology J. KANTO Department of Anaesthesiology Turku University Hospital, Turku, Finland 1. Mattila KT, Komu M, Karsikas R, Hatakka P, Heinonen OJ, Kormano M. Knee extension dynamometer: a new device for dynamic isokinetic MRS experiments. MAGMA 1996; 2: 1–8. 2. Järvi J, Nyman S, Komu M, Forsström J. A PC-program for automatic analysis of NMR spectrum series. Computer Methods and Programs in Biomedicine 1997; 52: 213–222.
3. Cozzone PJ, Bendahan D. 31P-NMR spectroscopy of metabolic changes associated with muscle exercise: physiopathological applications. In: Gillies RG, ed. NMR in Physiology and Biomedicine. San Diego: Academic Press, 1994; 389–402. 4. Kuipers H. Exercise-induced muscle damage. International Journal of Sports Medicine 1994; 15: 132–135.
Pain after laparoscopic cholecystectomy Sir,—We were interested in the review article by Alexander1 and wish to make some comments regarding local anaesthesia for pain after laparoscopic cholecystectomy. First, Alexander1 pointed out that analgesia tended to be given more readily during studies of pain or analgesics and that nearly all patients had received some form of intraoperative analgesia. We strongly support this view. It is difficult to compare postoperative pain when patients received various amounts of analgesics in the perioperative period. For this reason in our published study2 no opioid or non-opioid analgesics were used for premedication or during the perioperative period. Patients received only halothane and pancuronium during anaesthesia. For postoperative analgesia, patients received metamizol on request, as was usual in our hospital, by nurses who were blinded to the treatment. Postoperative pain was measured using a VAS score. The VAS score was not used as a threshold for administrating metamizol. Second, the timing of i.p. local anaesthetic (IPLA) is crucial, as it is for an NSAID. In the studies quoted in the review1 and in our study,2 IPLA was used after surgery, but in our study, in addition to that of Pasqualucci and colleagues,3 IPLA was given before and after surgery. Administration before surgery is important for pre-emptive analgesia. Third, the site of administration is important. The quoted studies reported that IPLA was ineffective when given only into the subdiaphragmatic space. In contrast, IPLA was effective for pain relief when given onto the gallbladder bed4 or into the subdiaphragmatic space and onto the operative area.2 3 Fourth, the local anaesthetic concentration should be important. The quoted studies reported that IPLA was not effective when a low concentration of IPLA (0.15–0.25% bupivacaine) was given in contrast with those2–4 in which IPLA was effective for pain relief when given in high concentrations (0.5% bupivacaine). Finally, Alexander1 concluded that local anaesthetic techniques appear to be more successful for pelvic laparoscopy than for laparoscopic cholecystectomy. We disagree. When IPLA is given at the correct time, on the right site and at the right concentration, it is also effective for pain relief after laparoscopic cholecystectomy. No side effects from IPLA have been reported. Therefore, we recommend routine administration of 0.5% plain bupivacaine 15 ml2 or 0.5% bupivacaine 20 ml with epinephrine3 into the subdiaphragmatic space and on the operative area before and after surgery.
B. MRAOVIĆ Department of Anaesthesia and Intensive Care General Hospital Šibenik Šibenik, Croatia V. MAJERIĆ-KOGLER Department of Anaesthesia and Intensive Care University Hospital for Chest Diseases “Jordanovac” Zagreb, Croatia 1. Alexander JI. Pain after laparoscopy. British Journal of Anaesthesia 1997; 79: 369–378. 2. Mraović B, Jurišić T, Majerić-Kogler V, Šustić A. Intraperitoneal bupivacaine for analgesia after laparoscopic cholecystectomy. Acta Anaesthesiologica Scandinavica 1997; 41: 193–196. 3. Pasqualucci A, De Angelis V, Contardo R, Colo F, Terrosu G, Donnini A, Pasetto A, Bresadola F. Preemptive analgesia: Intraperitoneal local anesthetic in laparoscopic cholecystectomy. Anesthesiology 1996; 85: 11–20. 4. Chundriger T, Morris R, Hedges AR, Stamatakis JD. Intraperitoneal bupivacaine for effective pain relief after laparoscopic cholecystectomy. Annals of the Royal College of Surgeons of England 1993; 40: 437–401.
Correspondence Sir,—Thank you for the opportunity to reply to Mraović and Majerić-Kogler, and to acknowledge a material addition to the evidence. In an adequate double-blind, randomized, controlled study,1 they demonstrated the lesser need for additional analgesics and lower pain scores in those who had bupivacaine injected between the liver and diaphragm and above the hepato-duodenal ligament. Their study was not reviewed because the article was published at the same time as the review2 was submitted. Their method was similar to that described in an earlier article by Pasqualucci and colleagues.3 This was primarily a study of preemptive analgesia and either bupivacaine or saline was sprayed above the liver and on the right subphrenic surface either before or after removal of the gallbladder, or both. Distension of the peritoneum occurred before application of local anaesthetic or saline. In those in whom bupivacaine was used at the end of intra-abdominal surgery, pain scores and ratings were lower than those in the placebo group at the end of surgery and 8 h after operation, but similar to those in the placebo group at 4 h (the time at which pain is most intense).1 4 Intraoperative i.v. fentanyl was used in a dose which was stated to be 15 mg kg91! Whether or not effective local anaesthesia or systemic analgesia provides pre-emptive analgesia during continuing nociception, as distinct from requirements for effective analgesia at the time of nociception, still appears to be in dispute. Mraović and Majerić-Kogler disagree only with my statement2 that “local anaesthetic techniques appear more successful for pelvic laparoscopy than for, say, laparoscopic cholecystectomy”. However, as they pointed out in their discussion, “six studies compared intraperitoneal bupivacaine for pain relief after laparoscopic cholecystectomy, four of which suggested that intraperitoneal bupivacaine does not reduce pain and two of which suggested that it does.” Nevertheless, the studies of Chundrigar and colleagues,5 Mraovic and colleagues1 and Pasqualucci and colleagues3 suggest with increasing evidence that to which I merely alluded: that local anaesthetic techniques are more likely to be effective when the local anaesthetic is placed accurately at the site of origin of nociception in effective concentrations, rather than relying on movement or diffusion of local anaesthetic to the target area from a convenient site in the peritoneal cavity, especially when such movement is against gravity in a cavity partly filled with gas.
J. I. ALEXANDER Sir Humphry Davy Department of Anaesthesia Bristol Royal Infirmary, Bristol 1. Mraović B, Jurisic V, Kogler-Majerić V, Sustic A. Intraperitoneal bupivacaine for analgesia after laparoscopic cholecystectomy. Acta Anaesthesioligica Scandinavica 1997; 41: 193–196. 2. Alexander JI. Pain after laparoscopy. British Journal of Anaesthesia 1997; 79: 369–378. 3. Pasqualucci A, De Angelis V, Contardo R, Colo F, Terrosu G, Donini A, Pasetto A, Bresadola F. Preemptive analgesia: intraperitoneal local anesthetic in laparoscopic cholecystectomy. Anesthesiology 1996; 85: 11–20. 4. Fredman B, Jedeikin R, Olsfanger D, Flor P, Gruzman A. Residual pneumoperitoneum: a cause of postoperative pain after laparoscopic cholecystectomy. Anesthesia and Analgesia 1994; 79: 152–154. 5. Chundrigar T, Morris R, Hedges AR, Stamatakis JD. Intraperitoneal bupivacaine for effective pain relief after laparoscopic cholecystectomy. Annals of the Royal College of Surgeons in England 1993; 75: 437–439.
Ondansetron compared with metoclopramide in the treatment of PONV Sir,—We wish to comment on the study on postoperative nausea and vomiting (PONV) by Diemunsch and colleagues.1 This is the largest study to date comparing ondansetron with a commonly used antiemetic in the treatment of established PONV, and showed improved outcome for patients treated with the 5-HT, antagonist. However, we were struck by the fact that even this drug, the newest and possibly most highly promoted antiemetic in the pharmacopoeia, still failed to control nausea in 56% and vomiting in 41% of patients. A recent review in the BMJ by Tramer and colleagues2 concluded that there is little information in relation to the effectiveness of intervention in established PONV, and that a 25% success rate is perhaps the best that can be achieved. The study of
407 Diemunsch and colleagues goes someway to addressing this insufficiency in the literature and, in addition, had a greater success rate than Tramer and colleagues felt may be the best possible. However, we feel that the study poses several questions: (1) Can we identify a group at high risk of PONV, and should we make more of an effort to do so using scoring systems such as that devised by Koivuranta and colleagues?3 (2) If we can identify these patients, is prophylaxis effective? (3) Are there non-drug interventions that can help, such as reducing the length of the preoperative fast,4 or the use of pre- or intraoperative i.v. fluids?5 (4) If this study shows the best that can be achieved with single agent therapy, should future research be conducted on double or even triple agent regimens compared with other active comparators? In a recent audit of PONV and pain in 198 sequential patients in our day-case unit, two (1%) were admitted to hospital, both as a result of vomiting. This is a small but significant number and brings to our attention the economic consequences of PONV and the distress caused by intractable vomiting and the inconveninence to the patient. It is hard to believe that we have optimized our management of PONV; perhaps more co-ordinated research will provide some answers.
C. M. HARPER J. P. BARKER Wellhouse NHS Trust Barnet General Hospital Barnet, Hertfordshire 1. Diemunsch P, Conseiller C, Clyti N, Mamet JP and the French Ondansetron Study Group. Ondansetron compared with metoclopramide in the treatment of established postoperative nausea and vomiting. British Journal of Anaesthesia 1997; 79: 322–326. 2. Tramer MR, Moore RA, Reynolds DJM, McQuay HJ. A quantitative systematic review of ondansetron in treatment of established postoperative nausea and vomiting. British Medical Journal 1997; 314: 1088–1092. 3. Koivuranta M, Laara E, Snare L, Alahuhta S. A survey of postoperative nausea and vomiting. Anaesthesia 1997; 52: 443–449. 4. Smith AF, Vallance H, Slater RM. Shorter preoperative fluid fasts reduce postoperative emesis. British Medical Journal 1997; 344: 1486. 5. Bakhshi KN, Iqbal S, Wilkinson DJ. An open randomised study of the effects of intravenous fluid replacement during day case anaesthesia. Ambulatory Surgery 1994; 2: 49–51. Sir,—Harper and Barker ask the question: (1) Can we identify a group at high risk of postoperative nausea and vomiting (PONV), and should we make an effort to do so using scoring systems such as that devised by Koivuranta and colleagues? Several prognostic factors are known to be associated with an increased risk of nausea and vomiting1 2: female sex, previous history of PONY, use of postoperative opioids and previous history of motion sickness. In our study, we identified the prognostic factors for the absence of emetic episodes after administration of the study drug. Sex appeared to influence the antiemetic response, as 72% of male patients experienced no emetic episodes between 15 min and 24 h after drug administration compared with 48% of female patients. Other factors influencing the antiemetic response include: history of PONV (55% of patients with a history of PONV experienced at least one emetic episode compared with 47% with no history); postoperative opioid analgesia (55% of patients who received opioid postoperative analgesics experienced at least one emetic episode compared with 45% who did not); and benzodiazepine premedication (54% of patients who received a benzodiazepine premedicant experienced at least one emetic episode compared with 40% who did not). Induction of anaesthesia with propofol had no influence on the antiemetic response, as 49% of patients who received propofol experienced at least one emetic episode compared with 50% of patients who did not. The logistic regression analysis carried out for the absence of an emetic episode between 15 min and 24 h after drug administration showed that: (a) significant prognostic factors, adjusted for treatment, were sex, history of PONV, benzodiazepine premedication and postoperative opioid analgesia. The odds ratio (with 95% confidence intervals) for the absence of an emetic episode in the ondansetron group compared with the metoclopramide group was 2.12 (1.57–2.87), adjusted for the above mentioned prognostic factors (P:0.0001). (b) The adjusted odds ratio associated with sex indicated that males were 2.76 times more likely to experience no emetic episode than females (P:0.0006). Patients having one of
408
British Journal of Anaesthesia
the following prognostic factors experienced more frequent emetic episodes: previous history of PONV (odds ratio:0.70; P:0.0237), benzodiazepine premedication (odds ratio:0.56; P:0.0008), postoperative opioid analgesia (odds ratio:0.64; P:0.0040). No significant effect from treatment of the prognostic factors was found. (2) If we can identify these patients, is prophylaxis effective? Yes, clinical studies confirmed that ondansetron was effective in both the prophylaxis and treatment of PONV.3–5 Moreover, there are also other benefits from reducing PONV, such as patient global satisfaction,6 quality of recovery and prevention of delayed recovery.7 (3) Are there non-drug interventions that can help, such as reducing the length of the postoperative fast, or the use of i.v. fluids before or after operation? Apart from drugs which are known to have antiemetic properties, such as lorazepam,8 hydroxyzine9 and propofol,10 others factors could enhance the incidence of PONV, such as postoperative pain11 or movement. (4) If this study shows the best that can be achieved with simple agent therapy, should future research be conducted on double or even triple agent regimens compared with other active comparators? The incidence of PONV is approximately 60% in both the prevention and treatment studies. The number of complete responders in our treatment study does not reflect the real benefits to the patient. The non-surrogate criteria (for example, global satisfaction of the patient) reflect other benefits which are not sufficiently taken into account. These results could probably be improved with the additional use of drugs such as corticosteroids,12 as was demonstrated in clinical studies in chemotherapy.
N. CLYTI Laboratoire Glaxo Wellcome Paris, France P. DIEMUNSCH Hôpital Civil, Strasburg 1. Palazzo M, Evans R. Logistic regression analysis of fixed patient factors for postoperative sickness: A model for risk assessment. British Journal of Anaesthesia 1993; 70: 135–140. 2. Toner CC, Broomhead CJ, Littlejohn IH, Samra GS, Powney JG, Palazzo MGA, Evans SJW, Strunin L. Prediction of postoperative nausea and vomiting using a logistic regression model. British Journal of Anaesthesia 1996; 76: 347–351. 3. Kovac A, McKenzie R, O’Connor T. Prophylactic intravenous ondansetron in female outpatients undergoing gynaecological surgery: a multicentre dose-comparison study. European Journal of Anaesthesiology 1992; 9 (Suppl. 6): 37–47. 4. Pearman MH. Single dose intravenous ondansetron in the prevention of postoperative nausea and vomiting. Anaesthesia 1994; 49 (Suppl.): 11–15. 5. Rust M, Cohen LA. Single oral dose ondansetron in the prevention of postoperative nausea and emesis. The European and US Study Groups. Anaesthesia 1994; 49 (Suppl.): 16–23. 6. Davis PJ, Mc Gowan FX jr, Landsman I, Maloney K, Hoffmann P. Effect of antiemetic therapy on recovery and hospital discharge time. A double-blind assessment of ondansetron, droperidol, and placebo in paediatric patients undergoing ambulatory surgery. Anesthesiology 1995; 83: 956– 960. 7. Diemunsch P, Conseiller C, Clyti N, Mamet JP and the French Ondansetron Study Group. Ondansetron compared with metoclopramide in the treatment of established postoperative nausea and vomiting. British Journal of Anaesthesia 1997; 79: 322–326. 8. Levine JM. Antiemetic effects of preoperative lorazepam on postoperative nausea and vomiting. Anesthesiology 1992; 77 (3A): abstract A410. 9. Garbin GS. Efficacy of ephedrine and hydroxyzine in preventing post operative nausea and vomiting. Anesthesiology 1992; 77 (A3): abstract A3. 10. Watcha MF. Effect of propofol on the incidence of post operative vomiting after strabismus surgery in pediatric outpatients. Anesthesiology 1991; 75: 204–209. 11. Andersen R, Krohg K. Pain as a major factor of postoperative nausea. Canadian Anaesthetists Society Journal 1976; 23: 366– 369. 12. Wrench IJ, Ward JE, Walder AD. The prevention of postoperative nausea and vomiting using a combination of ondansetron and droperidol. Anaesthesia 1996; 51: 776–778.
Figure 1 Patient No. 1. A 6-yr-old boy with neurofibromatosis who presented for tracheostomy and debulking of a large intraoral tumour.
Sevoflurane for difficult intubation in children Sir,—We read with interest the case reports by Mostafa and Atherton1 in which sevoflurane was used to induce anaesthesia in three adult patients in whom difficult tracheal intubation was anticipated. We wish to report two paediatric patients with a difficult airway where sevoflurane was used as the induction agent. The first patient was a 6-yr-old boy with neurofibromatosis who presented for tracheostomy and debulking of a large intraoral tumour (fig. 1). He also had a 6 month’ history of progressive difficulty in breathing and stridor. Inhalation induction was planned because of airway obstruction. EMLA cream was applied to both hands 1 h before induction. I.v. access was established. Anaesthesia was induced with sevoflurane in oxygen. The concentration of sevoflurane was increased gradually until an end-tidal concentration of 5.5% was achieved. Direct laryngoscopy was performed and revealed a grade III larynx which was grossly deviated to the left. The larynx was intubated uneventfully with a 5.0-mm plain Portex tube orally. Oxygen saturation was maintained at 99% throughout induction. The second patient was a 11-yr-old girl with a fixed right temporal mandibular joint and a very receded mandible. Her inter-incisor distance was only 1 cm at full mouth opening. Fibreoptic intubation under inhalation induction was planned. I.v. access was established before induction. Anaesthesia was induced with incremental concentrations of sevoflurane in oxygen. A nasal airway was inserted via the left nostril and attached to a T-piece. When the end-tidal sevoflurane concentration reached 5.5%, the trachea was intubated with a 5.5-mm plain Portex tube nasally via the flexible fibreoptic laryngoscope. Both surgery and anaesthesia were uneventful. A difficult airway in paediatric patients presents a challenge to anaesthetists. Inhalation induction is the technique of choice because it is often difficult to perform awake fibreoptic intubation or tracheostomy under local anaesthesia in children. We agree with Mostafa and Atherton that sevoflurane has many advantages compared with halothane in the management of the difficult airway in both children and adults. The low blood:gas solubility of sevoflurane and consequent rapid induction and rapid recovery are reassuring features in the management of difficult intubation.
C. Y. WANG C. L. CHIU A. E. DELILKAN Department of Anaesthesia University of Malaya Kuala Lumpur, Malaysia 1. Mostafa SM, Atherton AMJ. Sevoflurane for difficult tracheal intubation. British Journal of Anaesthesia 1997; 79: 392–393.
Caudal tramadol for postoperative analgesia in hypospadias surgery Sir,—Prolongation of caudal anaesthesia is indeed a laudable aim. However, I have some concern with the recent article describing the use of tramadol in this role.1 The evidence that tramadol may be effective in this regard is scanty. There is little evidence that it is safe or effective by the extradural route in adults.
Correspondence
409
We do not know the incidence of neurological damage that may be caused by this drug, and just one child with a painful or disabling neuropathy would be disastrous. Until evidence emerges of its safety and efficacy in adults, I do not believe that studies of this nature should be carried out in children.
W. RUSSELL Intensive Care Unit Leicester Royal Infirmary Leicester 1. Prosser DP, Davis A, Booker PD, Murray A. Caudal tramadol for postoperative analgesia in paediatric hypospadias surgery. British Journal of Anaesthesia 1997; 79: 293–296. Sir,—Thank you for the opportunity to respond to Dr Russell about our caudal tramadol study.1 Although we agree in principle that studies should not involve children if the relevant data can be obtained from adults, this is rarely possible. In this instance, the significant age-related differences in opioid pharmacokinetics and pharmacodynamics,2 and vascularity, fat content and volume of the extradural space3 4 make extrapolation of clinical data from adults to young children rather suspect. Although some initial data may be gleaned from studies of drugs injected into the lumbar extradural space in adults, our results supported the concept that data achieved in this way cannot be used to predict accurately drug pharmacodynamics after caudal injection in children. All but one of 10 adult studies and the one paediatric study published in the past decade that have examined the effects of tramadol after extradural administration have recorded encouraging results. Our findings would not have been anticipated from meta-analysis of these studies. Furthermore, for various well recognized reasons, administration of drugs into the caudal extradural space is much less common in adults than in the very young. If studies examining caudal administration of drugs in adults were required before paediatric studies could proceed, we would often have to rely solely on data from non-randomized, unblinded, uncontrolled studies carried out in developing countries. The second point about drug safety is a thorny issue but one that is always raised at Paediatric Ethics Committee meetings when a study of a “new” drug is being reviewed. To prove the safety of any drug is almost impossible, but a thorough examination of all published (and any available unpublished) clinical and experimental evidence offers the best possible compromise between ensuring optimal safe care for our patients and advancement of our knowledge base. In this instance we were able to point out that tramadol, which has been widely available in the rest of Europe for more than 20 yr, has an exceptional safety record5; to the best of our knowledge no long-term adverse side effects attributable to its use have been reported. Finally, we did not perform this study because it was possible or because we wished to write another article; we believed that there was a problem (postoperative pain in children undergoing day-case hypospadias surgery) that could be prevented or alleviated. On the basis of previously published clinical work, we hypothesized that the analgesic effect produced by bupivacaine could be usefully prolonged by the addition of tramadol. Similar studies using more conventional opioids have shown that delayed respiratory depression may, occasionally, be a problem.6 7 Tramadol has the important and significant advantage over other commonly used opioids of having a marked lack of respiratory depressant effects.8 9 If we had found tramadol to be clinically useful in inpatients, then we would have extended our study to day-case patients. In the light of our initial findings, we did not believe that a study extension was justifiable. Many of these same arguments were presented and examined carefully by our Paediatric Ethics Committee before commencement of this study. One of the guiding principles used by this Committee in determining whether or not a study should be allowed to proceed is: “Children are not small adults and research should only be done in this group of patients if comparable research on adults cannot answer the same questions”. We believe, as did the Ethics Committee, that this principle was strictly adhered to in the conduct of this study.
D. P. PROSSER A. J. M. DAVIS P. D. BOOKER A. MURRAY Alder Hey Royal Liverpool Children’s NHS Trust Liverpool
1. Prosser DP, Davis A, Booker PD, Murray A. Caudal tramadol for postoperative analgesia in paediatric hypospadias surgery. British Journal of Anaesthesia 1997; 79: 293–296. 2. Olkkola KT, Hamunen K, Maunuksela EL. Clinical pharmacokinetics and pharmacodynamics of opioid analgesics in infants and children. Clinical Pharmacokinetics 1995; 28: 385– 404. 3. Hogan QH. Epidural anatomy examined by cryomicrotome section. Influence of age, vertebral level and disease. Regional Anesthesia 1996; 21: 395–406. 4. Pullerits J, Holzman RS. Pediatric neuraxial blockade. Journal of Clinical Anesthesia 1993; 5: 342–354. 5. Gibson TP. Pharmacokinetics, efficacy and safety of analgesia with a focus on tramadol HCL. American Journal of Medicine 1996; 101: 47S–53S. 6. Tyler DC, Krane EJ. Epidural opioids in children. Journal of Pediatric Surgery 1989; 24: 469–473. 7. Cook B, Doyle E. The use of additives to local anaesthetic solutions for caudal epidural blockade. Paediatric Anaesthesia 1996; 6: 353–359. 8. Vickers MD, O’Flaherty D, Szekely SM, Read M, Yoshizumi J. Tramadol: pain relief by an opioid without depression of respiration. Anaesthesia 1992; 47: 291–296. 9. Houmes RJM, Voets MA, Verkaaik A, Erdmann W, Lachmann B. Efficacy and safety of tramadol versus morphine for moderate and severe postoperative pain with special regard to respiratory depression. Anesthesia and Analgesia 1992; 74: 510–514.
Bair Hugger active patient warming system Sir,—I read with interest the case report of Ayala and Coe regarding softening of tracheal tubes when used in conjunction with the Bair Hugger patient warming system.1 As a result of their case report and subsequent study, they recommended the avoidance of uncut PVC tubes, unless adequately supported, with this warming system. Having personally experienced a similar problem I have a simple and effective solution. This is to make a small hole in the clear plastic region of the Bair Hugger (designed to cover the patient’s head) and to pass the tracheal tube through this, thus effectively removing it from the warmed environment. I have encountered no recurrence of the problem since adopting this method.
I. R. TAYLOR Department of Anaesthesia Queen Alexandra Hospital Portsmouth 1. Ayala JL, Coe A. Thermal softening of tracheal tubes: an unrecognised hazard of the Bair Hugger active patient warming system. British Journal of Anaesthesia 1997; 79: 543–545.
Inappropriate use of placebo Sir,—Fujii, Toyooka and Tanaka compared granisetron with placebo for reduction of postoperative nausea and vomiting (PONV) in middle ear surgery.1 First, when established treatments are available, research must be compared with existing treatments, not with placebo. I wish to know if granisetron is more or less effective than other antiemetics on offer. Second, the same authors had already proved the effectiveness of granisetron in PONV in gynaecological patients (including a placebo group).2 Third, by withholding antiemetic therapy in the placebo group until significant morbidity from PONV had occurred represents sub-standard medical care.
J. CLARKE Department of Anaesthesia, Queen Elizabeth Hospital Woodville, Adelaide, Australia 1. Fujii Y, Toyooka H, Tanaka H. Granisetron reduces the incidence of nausea and vomiting after middle ear surgery. British Journal of Anaesthesia 1997; 79: 539–540. 2. Fujii Y, Tanaka H, Toyooka H. Optimal anti-emetic doses of granisetron for preventing post operative nausea and vomiting. Canadian Journal of Anaesthesia 1994; 41: 794–797. Sir,—We have already studied the efficacy of granisetron, a selective 5-hydroxytryptamine type-3 receptor antagonist, in preventing postoperative nausea and vomiting (PONV) after gynaecologi-
410
British Journal of Anaesthesia
cal surgery.1 2 However, there is no study investigating its efficacy in preventing PONV in patients undergoing middle ear surgery. The incidence of PONV after this surgical procedure is high (62– 80%) when no prophylactic antiemetic is given.3 Therefore, we studied the efficacy of granisetron in the prevention of PONV in this population. Several investigators have previously compared the antiemetic efficacy of such new agents as ondanstron, granisetron, tropisetron and dorasetron with placebo for preventing PONV.4–7 Similarly, we compared the efficacy of granisetron and placebo for the prevention of PONV after middle ear surgery.
Y. FUJII Department of Anaesthesiology University of Tsukuba Institute of Clinical Medicine Tsukuba City, Ibaraki, Japan 1. Fujii Y. Reduction of postoperative nausea and vomiting with granisetron. Canadian Journal of Anaesthesia 1994; 41: 291–294. 2. Fujii Y. Optimal anti-emetic doses of granisetron for preventing postoperative nausea and vomiting. Canadian Journal of Anaesthesia 1994; 41: 794–797. 3. Honkevaara P. Prevention of nausea and vomiting with transdermal hyoscine in adults after middle ear surgery during general anaesthesia. British Journal of Anaesthesia 1994; 73: 763–766. 4. Honkavarra P. Effect of ondansetron on nausea and vomiting after middle ear surgery during general anaesthesia. British Journal of Anaesthesia 1996; 76: 316–318. 5. Wilson AJ. Single-dose i.v. granisetron in the prevention of postoperative nausea and vomiting. British Journal of Anaesthesia 1996; 76: 515–518. 6. Capouet V. Single dose i.v. tropisetron in the prevention of postoperative nausea and vomiting after gynaecological surgery. British Journal of Anaesthesia 1996; 76: 54–60. 7. Graczyk SG. Intravenous dorasetron for the prevention of postoperative nausea and vomiting after outpatient laparoscopic gynecologic surgery. Anesthesia and Analgesia 1997; 84: 325–330.
Bradycardia during rapid inhalation induction with sevoflurane in children colleagues,1
Sir,—In common with Sigston and we have found inhalation induction with 8% sevoflurane to be rapid and well tolerated in young children. However, the development of profound bradycardia in four children has caused us to reflect on its use. These were healthy children, aged 6 months to 2 yr, undergoing minor surgery. None was premedicated and induction was by inhalation of 8% sevoflurane and 66% nitrous oxide in oxygen, using a standard technique. In all cases the onset of bradycardia occurred during induction with no loss of airway or ventilation. One infant required treatment with atropine because of clinical evidence of a decrease in cardiac output. In the other children heart rate recovered spontaneously when the concentration of sevoflurane was reduced or we changed to using isoflurane. In the study of Sigston and colleagues, bradycardia occurred in one child who received halothane (age unknown) but not in the sevoflurane group. It is noteworthy that all of their patients received atropine premedication. Baum, Yemmen and Baum reported no arrhythmias when comparing 8% sevoflurane with incremental sevoflurane in unpremedicated children.2 Johannesson, Floren and Lindahl found an incidence of cardiac arrhythmias of 5% while using incremental sevoflurane in premedicated patients.3 Otherwise, incremental induction with sevoflurane in children seems to show an increase or have little effect on heart rate.4 Those familiar with halothane induction in infants may admonish us for not routinely prescribing atropine where an inhalation induction is planned, if not to dry secretions and thus reduce airway complications, then to offset the myocardial depression which can accompany high concentrations of a volatile agent. The cardiovascular depression produced by sevoflurane is much less than that with halothane, but not absent. The single-handed anaesthetist, anticipating difficult i.v. cannulation, would be well advised to consider atropine premedication when planning inhalation induction with high concentrations of sevoflurane in infants.
P. TOWNSEND M. A. STOKES Department of Anaesthetics Birmingham Children’s Hospital NHS Trust Birmingham
1. Sigston PE, Jenkins AMC, Jackson EA, Sury MRJ, Mackersie AM, Hatch DJ. Rapid inhalation induction in children: 8% sevoflurane compared with 5% halothane. British Journal of Anaesthesia 1997; 78: 362–365. 2. Baum VC, Yemmen TA, Baum LD. Immediate 8% sevoflurane induction in children: a comparison with incremental sevoflurane and incremental halothane. Anesthesia and Analgesia 1997; 85: 313–316. 3. Johannesson GP, Floren M, Lindahl SGE. Sevoflurane for ENT-surgery in children. A comparison with halothane. Acta Anaesthesiologica Scandinavica 1995; 39: 546–550. 4. Kern C, Erb T, Frei FJ. Haemodynamic responses to sevoflurane compared with halothane during inhalational induction in children. Paediatric Anaesthesia 1997; 7: 439–444.
Effect of temperature on diaphragmatic function Sir,—The article by Mills and colleagues described the effect of a 5⬚C temperature change on the evoked compound potential in the human diaphragm.1 They found a reduction in the EMG of approximately 60%. This is in agreement with our own findings using electromyography carried out in five children undergoing cardiac surgery under profound hypothermia. We found this effect to be almost completely reversed by edrophonium and concluded it was most likely a result of decreased production or release of acetylcholine (ACh).2 Further experiments in dogs confirmed this effect.3 The recent work of England added support for the view that in non-hibernating mammals, cold critically affects ACh production or release.4 In the past, much of our information on the effects of hypothermia has come from experiments carried out either in cold blooded animals or on tissues from rats which are genetically closely allied to hibernating animals. These species do not necessarily respond to cold in the same manner as humans and as a result conflicting findings have been reported. This article confirms the critical effect of cold in human neuromuscular conduction. If the effect of hypothermia in humans is caused by a reduction in ACh release, this also has implications for its effect on neuromuscular blocking drugs.5
S. FELDMAN London 1. Mills GH, Khan ZP, Moxham J, Desai J, Forsyth A, Ponte J. Effects of temperature on phrenic nerve and diaphragmatic function during cardiac surgery. British Journal of Anaesthesia 1997; 79: 726–732. 2. Feldman SA. Muscle Relaxants. Philidelphia: WB Saunders, 1979; 14–146. 3. Thornton RJ, Feldman SA, Blackeney C. The effect of hypothermia on neuromuscular conduction in dogs. Abstracts of the 5th World Congress of Anesthesiologists. Amsterdam: Excerpta Medica, 1975; 12. 4. England AJ. Effect of Hypothermia on Neuromuscular Conduction and Blockade, MD Thesis. London: University of London, 1995. 5. Feldman S. Neuromuscular Block. Oxford: Butterworth Heinemann, 1996; 128–138.
Sir,—We thank Professor Feldman for his comments on our results. We agree entirely that the most likely mechanism which may explain our observations is indeed impairment of production or release of acetylcholine (ACh). We regret that at the time this did not occur to us and therefore we did not perform the simple edrophonium test. Examination of the human data of Feldman[2] (quoted by Feldman) suggests that the effects of hypothermia on twitch height of arm muscles start at approximately 34.5⬚C. Interestingly, our data (especially during the transition from warm to cold) suggest that the effects of hypothermia on the diaphragm may be detectable at temperatures even closer to 36⬚C. One could speculate that there may be differences between the diaphragm and limb muscles in this respect, perhaps because the temperature within limb muscles is often hypothermic in inactive, normal subjects exposed to cold, whereas the diaphragm is not.
J. C. PONTE Academic Department of Anaesthetics King’s College School of Medicine and Dentistry London
CORRESPONDENCE
Anaesthesia after exhaustive exercise Sir,—There are no reports on the effect of preceding exhaustive exercise, such as marathon running, on the course of anaesthesia. We report an extreme case. A healthy 21-yr old (65 kg,173 cm) active jogger (40–60 km/week) presented for day-case tonsillectomy under general anaesthesia. Three days earlier he had run his first marathon and told the anaesthetist that his legs were painful. He had difficulty in lying down on the operating table. After operation, he told us that after 30 km of the marathon he had become so exhausted that he could remember little of the rest of the race. An i.v. anaesthetic was used, supplemented with 70% nitrous oxide in oxygen. The drugs given and their total doses were as follows: glycopyrrolate 0.2 mg, propofol 280 mg, fentanyl 300 g and rocuronium 50 mg. Residual neuromuscular block was antagonized at the end of operation with neostigmine 2.5 mg and glycopyrrolate 0.5 mg. During surgery, 2500 ml of crystalloid i.v. solutions were given. Postoperative analgesia was achieved with ketoprofen. After operation the patient was able to breathe satisfactorily but despite normal responses to peripheral nerve stimulation he appeared extremely weak, that is he was unable to lift his head or even his hand. Because his condition had not improved after 6 h, he was moved to the local university hospital. Thirteen hours after operation the patient had to be escorted to the toilet by two people. Next morning he was able to walk but still complained of muscle pain, especially in his legs. The muscle pain disappeared gradually over the next 3 days. Eight hours after operation total blood count, C-reactive protein concentrations, serum concentrations of creatinine, sodium, potassium and chloride, and blood-gas analysis, serum lactate, ammonium ion and insulin concentrations were normal. The only abnormal value was serum creatine kinase (1572 u. litre91, normal value :285 u. litre91), which was still high 2 days after operation (5 days after the marathon it was 338 u. litre91), but was within normal limits by 10 days after operation (114 u. litre91). The patient underwent 31P-magnetic resonance spectroscopy (MRS) 33 h after the beginning of anaesthesia to assess the status of skeletal muscle energy metabolism.1 2 He was re-examined 5 months later for comparison. The patient resumed his usual training schedule 1 month after surgery. From the MRS studies we found no evidence of altered energy metabolism of striated muscle to anaesthesia after exhaustive exercise. Variables assessing the main energy producing pathways (phosphocreatinine hydrolysis, glycolysis or aerobic oxidative phosphorylation) were similar immediately after the incident compared with 5 months later. In contrast, resting concentrations of phosphodiesters were substantially higher immediately after the incident (4.5 compared with 2.7 mmol litre91). Phosphodiesters are hypothesized to be indicators of release of membrane molecules in muscle cell damage.3 But muscle overuse is associated with damage to the muscle contractile elements.4 This damage to the contractile elements could also be an explanation for our phosphodiester results. After exhaustive exercise, a general anaesthetic could cause some type of interaction between the exhausted muscles and anaesthetic agents. Importantly, significant rhabdomyolysis may occur if succinylcholine is used.
M. ERKINTALO M. KOMU K. MATTILA Department of Diagnostic Radiology J. KANTO Department of Anaesthesiology Turku University Hospital, Turku, Finland 1. Mattila KT, Komu M, Karsikas R, Hatakka P, Heinonen OJ, Kormano M. Knee extension dynamometer: a new device for dynamic isokinetic MRS experiments. MAGMA 1996; 2: 1–8. 2. Järvi J, Nyman S, Komu M, Forsström J. A PC-program for automatic analysis of NMR spectrum series. Computer Methods and Programs in Biomedicine 1997; 52: 213–222.
3. Cozzone PJ, Bendahan D. 31P-NMR spectroscopy of metabolic changes associated with muscle exercise: physiopathological applications. In: Gillies RG, ed. NMR in Physiology and Biomedicine. San Diego: Academic Press, 1994; 389–402. 4. Kuipers H. Exercise-induced muscle damage. International Journal of Sports Medicine 1994; 15: 132–135.
Pain after laparoscopic cholecystectomy Sir,—We were interested in the review article by Alexander1 and wish to make some comments regarding local anaesthesia for pain after laparoscopic cholecystectomy. First, Alexander1 pointed out that analgesia tended to be given more readily during studies of pain or analgesics and that nearly all patients had received some form of intraoperative analgesia. We strongly support this view. It is difficult to compare postoperative pain when patients received various amounts of analgesics in the perioperative period. For this reason in our published study2 no opioid or non-opioid analgesics were used for premedication or during the perioperative period. Patients received only halothane and pancuronium during anaesthesia. For postoperative analgesia, patients received metamizol on request, as was usual in our hospital, by nurses who were blinded to the treatment. Postoperative pain was measured using a VAS score. The VAS score was not used as a threshold for administrating metamizol. Second, the timing of i.p. local anaesthetic (IPLA) is crucial, as it is for an NSAID. In the studies quoted in the review1 and in our study,2 IPLA was used after surgery, but in our study, in addition to that of Pasqualucci and colleagues,3 IPLA was given before and after surgery. Administration before surgery is important for pre-emptive analgesia. Third, the site of administration is important. The quoted studies reported that IPLA was ineffective when given only into the subdiaphragmatic space. In contrast, IPLA was effective for pain relief when given onto the gallbladder bed4 or into the subdiaphragmatic space and onto the operative area.2 3 Fourth, the local anaesthetic concentration should be important. The quoted studies reported that IPLA was not effective when a low concentration of IPLA (0.15–0.25% bupivacaine) was given in contrast with those2–4 in which IPLA was effective for pain relief when given in high concentrations (0.5% bupivacaine). Finally, Alexander1 concluded that local anaesthetic techniques appear to be more successful for pelvic laparoscopy than for laparoscopic cholecystectomy. We disagree. When IPLA is given at the correct time, on the right site and at the right concentration, it is also effective for pain relief after laparoscopic cholecystectomy. No side effects from IPLA have been reported. Therefore, we recommend routine administration of 0.5% plain bupivacaine 15 ml2 or 0.5% bupivacaine 20 ml with epinephrine3 into the subdiaphragmatic space and on the operative area before and after surgery.
B. MRAOVIĆ Department of Anaesthesia and Intensive Care General Hospital Šibenik Šibenik, Croatia V. MAJERIĆ-KOGLER Department of Anaesthesia and Intensive Care University Hospital for Chest Diseases “Jordanovac” Zagreb, Croatia 1. Alexander JI. Pain after laparoscopy. British Journal of Anaesthesia 1997; 79: 369–378. 2. Mraović B, Jurišić T, Majerić-Kogler V, Šustić A. Intraperitoneal bupivacaine for analgesia after laparoscopic cholecystectomy. Acta Anaesthesiologica Scandinavica 1997; 41: 193–196. 3. Pasqualucci A, De Angelis V, Contardo R, Colo F, Terrosu G, Donnini A, Pasetto A, Bresadola F. Preemptive analgesia: Intraperitoneal local anesthetic in laparoscopic cholecystectomy. Anesthesiology 1996; 85: 11–20. 4. Chundriger T, Morris R, Hedges AR, Stamatakis JD. Intraperitoneal bupivacaine for effective pain relief after laparoscopic cholecystectomy. Annals of the Royal College of Surgeons of England 1993; 40: 437–401.
Correspondence Sir,—Thank you for the opportunity to reply to Mraović and Majerić-Kogler, and to acknowledge a material addition to the evidence. In an adequate double-blind, randomized, controlled study,1 they demonstrated the lesser need for additional analgesics and lower pain scores in those who had bupivacaine injected between the liver and diaphragm and above the hepato-duodenal ligament. Their study was not reviewed because the article was published at the same time as the review2 was submitted. Their method was similar to that described in an earlier article by Pasqualucci and colleagues.3 This was primarily a study of preemptive analgesia and either bupivacaine or saline was sprayed above the liver and on the right subphrenic surface either before or after removal of the gallbladder, or both. Distension of the peritoneum occurred before application of local anaesthetic or saline. In those in whom bupivacaine was used at the end of intra-abdominal surgery, pain scores and ratings were lower than those in the placebo group at the end of surgery and 8 h after operation, but similar to those in the placebo group at 4 h (the time at which pain is most intense).1 4 Intraoperative i.v. fentanyl was used in a dose which was stated to be 15 mg kg91! Whether or not effective local anaesthesia or systemic analgesia provides pre-emptive analgesia during continuing nociception, as distinct from requirements for effective analgesia at the time of nociception, still appears to be in dispute. Mraović and Majerić-Kogler disagree only with my statement2 that “local anaesthetic techniques appear more successful for pelvic laparoscopy than for, say, laparoscopic cholecystectomy”. However, as they pointed out in their discussion, “six studies compared intraperitoneal bupivacaine for pain relief after laparoscopic cholecystectomy, four of which suggested that intraperitoneal bupivacaine does not reduce pain and two of which suggested that it does.” Nevertheless, the studies of Chundrigar and colleagues,5 Mraovic and colleagues1 and Pasqualucci and colleagues3 suggest with increasing evidence that to which I merely alluded: that local anaesthetic techniques are more likely to be effective when the local anaesthetic is placed accurately at the site of origin of nociception in effective concentrations, rather than relying on movement or diffusion of local anaesthetic to the target area from a convenient site in the peritoneal cavity, especially when such movement is against gravity in a cavity partly filled with gas.
J. I. ALEXANDER Sir Humphry Davy Department of Anaesthesia Bristol Royal Infirmary, Bristol 1. Mraović B, Jurisic V, Kogler-Majerić V, Sustic A. Intraperitoneal bupivacaine for analgesia after laparoscopic cholecystectomy. Acta Anaesthesioligica Scandinavica 1997; 41: 193–196. 2. Alexander JI. Pain after laparoscopy. British Journal of Anaesthesia 1997; 79: 369–378. 3. Pasqualucci A, De Angelis V, Contardo R, Colo F, Terrosu G, Donini A, Pasetto A, Bresadola F. Preemptive analgesia: intraperitoneal local anesthetic in laparoscopic cholecystectomy. Anesthesiology 1996; 85: 11–20. 4. Fredman B, Jedeikin R, Olsfanger D, Flor P, Gruzman A. Residual pneumoperitoneum: a cause of postoperative pain after laparoscopic cholecystectomy. Anesthesia and Analgesia 1994; 79: 152–154. 5. Chundrigar T, Morris R, Hedges AR, Stamatakis JD. Intraperitoneal bupivacaine for effective pain relief after laparoscopic cholecystectomy. Annals of the Royal College of Surgeons in England 1993; 75: 437–439.
Ondansetron compared with metoclopramide in the treatment of PONV Sir,—We wish to comment on the study on postoperative nausea and vomiting (PONV) by Diemunsch and colleagues.1 This is the largest study to date comparing ondansetron with a commonly used antiemetic in the treatment of established PONV, and showed improved outcome for patients treated with the 5-HT, antagonist. However, we were struck by the fact that even this drug, the newest and possibly most highly promoted antiemetic in the pharmacopoeia, still failed to control nausea in 56% and vomiting in 41% of patients. A recent review in the BMJ by Tramer and colleagues2 concluded that there is little information in relation to the effectiveness of intervention in established PONV, and that a 25% success rate is perhaps the best that can be achieved. The study of
407 Diemunsch and colleagues goes someway to addressing this insufficiency in the literature and, in addition, had a greater success rate than Tramer and colleagues felt may be the best possible. However, we feel that the study poses several questions: (1) Can we identify a group at high risk of PONV, and should we make more of an effort to do so using scoring systems such as that devised by Koivuranta and colleagues?3 (2) If we can identify these patients, is prophylaxis effective? (3) Are there non-drug interventions that can help, such as reducing the length of the preoperative fast,4 or the use of pre- or intraoperative i.v. fluids?5 (4) If this study shows the best that can be achieved with single agent therapy, should future research be conducted on double or even triple agent regimens compared with other active comparators? In a recent audit of PONV and pain in 198 sequential patients in our day-case unit, two (1%) were admitted to hospital, both as a result of vomiting. This is a small but significant number and brings to our attention the economic consequences of PONV and the distress caused by intractable vomiting and the inconveninence to the patient. It is hard to believe that we have optimized our management of PONV; perhaps more co-ordinated research will provide some answers.
C. M. HARPER J. P. BARKER Wellhouse NHS Trust Barnet General Hospital Barnet, Hertfordshire 1. Diemunsch P, Conseiller C, Clyti N, Mamet JP and the French Ondansetron Study Group. Ondansetron compared with metoclopramide in the treatment of established postoperative nausea and vomiting. British Journal of Anaesthesia 1997; 79: 322–326. 2. Tramer MR, Moore RA, Reynolds DJM, McQuay HJ. A quantitative systematic review of ondansetron in treatment of established postoperative nausea and vomiting. British Medical Journal 1997; 314: 1088–1092. 3. Koivuranta M, Laara E, Snare L, Alahuhta S. A survey of postoperative nausea and vomiting. Anaesthesia 1997; 52: 443–449. 4. Smith AF, Vallance H, Slater RM. Shorter preoperative fluid fasts reduce postoperative emesis. British Medical Journal 1997; 344: 1486. 5. Bakhshi KN, Iqbal S, Wilkinson DJ. An open randomised study of the effects of intravenous fluid replacement during day case anaesthesia. Ambulatory Surgery 1994; 2: 49–51. Sir,—Harper and Barker ask the question: (1) Can we identify a group at high risk of postoperative nausea and vomiting (PONV), and should we make an effort to do so using scoring systems such as that devised by Koivuranta and colleagues? Several prognostic factors are known to be associated with an increased risk of nausea and vomiting1 2: female sex, previous history of PONY, use of postoperative opioids and previous history of motion sickness. In our study, we identified the prognostic factors for the absence of emetic episodes after administration of the study drug. Sex appeared to influence the antiemetic response, as 72% of male patients experienced no emetic episodes between 15 min and 24 h after drug administration compared with 48% of female patients. Other factors influencing the antiemetic response include: history of PONV (55% of patients with a history of PONV experienced at least one emetic episode compared with 47% with no history); postoperative opioid analgesia (55% of patients who received opioid postoperative analgesics experienced at least one emetic episode compared with 45% who did not); and benzodiazepine premedication (54% of patients who received a benzodiazepine premedicant experienced at least one emetic episode compared with 40% who did not). Induction of anaesthesia with propofol had no influence on the antiemetic response, as 49% of patients who received propofol experienced at least one emetic episode compared with 50% of patients who did not. The logistic regression analysis carried out for the absence of an emetic episode between 15 min and 24 h after drug administration showed that: (a) significant prognostic factors, adjusted for treatment, were sex, history of PONV, benzodiazepine premedication and postoperative opioid analgesia. The odds ratio (with 95% confidence intervals) for the absence of an emetic episode in the ondansetron group compared with the metoclopramide group was 2.12 (1.57–2.87), adjusted for the above mentioned prognostic factors (P:0.0001). (b) The adjusted odds ratio associated with sex indicated that males were 2.76 times more likely to experience no emetic episode than females (P:0.0006). Patients having one of
408
British Journal of Anaesthesia
the following prognostic factors experienced more frequent emetic episodes: previous history of PONV (odds ratio:0.70; P:0.0237), benzodiazepine premedication (odds ratio:0.56; P:0.0008), postoperative opioid analgesia (odds ratio:0.64; P:0.0040). No significant effect from treatment of the prognostic factors was found. (2) If we can identify these patients, is prophylaxis effective? Yes, clinical studies confirmed that ondansetron was effective in both the prophylaxis and treatment of PONV.3–5 Moreover, there are also other benefits from reducing PONV, such as patient global satisfaction,6 quality of recovery and prevention of delayed recovery.7 (3) Are there non-drug interventions that can help, such as reducing the length of the postoperative fast, or the use of i.v. fluids before or after operation? Apart from drugs which are known to have antiemetic properties, such as lorazepam,8 hydroxyzine9 and propofol,10 others factors could enhance the incidence of PONV, such as postoperative pain11 or movement. (4) If this study shows the best that can be achieved with simple agent therapy, should future research be conducted on double or even triple agent regimens compared with other active comparators? The incidence of PONV is approximately 60% in both the prevention and treatment studies. The number of complete responders in our treatment study does not reflect the real benefits to the patient. The non-surrogate criteria (for example, global satisfaction of the patient) reflect other benefits which are not sufficiently taken into account. These results could probably be improved with the additional use of drugs such as corticosteroids,12 as was demonstrated in clinical studies in chemotherapy.
N. CLYTI Laboratoire Glaxo Wellcome Paris, France P. DIEMUNSCH Hôpital Civil, Strasburg 1. Palazzo M, Evans R. Logistic regression analysis of fixed patient factors for postoperative sickness: A model for risk assessment. British Journal of Anaesthesia 1993; 70: 135–140. 2. Toner CC, Broomhead CJ, Littlejohn IH, Samra GS, Powney JG, Palazzo MGA, Evans SJW, Strunin L. Prediction of postoperative nausea and vomiting using a logistic regression model. British Journal of Anaesthesia 1996; 76: 347–351. 3. Kovac A, McKenzie R, O’Connor T. Prophylactic intravenous ondansetron in female outpatients undergoing gynaecological surgery: a multicentre dose-comparison study. European Journal of Anaesthesiology 1992; 9 (Suppl. 6): 37–47. 4. Pearman MH. Single dose intravenous ondansetron in the prevention of postoperative nausea and vomiting. Anaesthesia 1994; 49 (Suppl.): 11–15. 5. Rust M, Cohen LA. Single oral dose ondansetron in the prevention of postoperative nausea and emesis. The European and US Study Groups. Anaesthesia 1994; 49 (Suppl.): 16–23. 6. Davis PJ, Mc Gowan FX jr, Landsman I, Maloney K, Hoffmann P. Effect of antiemetic therapy on recovery and hospital discharge time. A double-blind assessment of ondansetron, droperidol, and placebo in paediatric patients undergoing ambulatory surgery. Anesthesiology 1995; 83: 956– 960. 7. Diemunsch P, Conseiller C, Clyti N, Mamet JP and the French Ondansetron Study Group. Ondansetron compared with metoclopramide in the treatment of established postoperative nausea and vomiting. British Journal of Anaesthesia 1997; 79: 322–326. 8. Levine JM. Antiemetic effects of preoperative lorazepam on postoperative nausea and vomiting. Anesthesiology 1992; 77 (3A): abstract A410. 9. Garbin GS. Efficacy of ephedrine and hydroxyzine in preventing post operative nausea and vomiting. Anesthesiology 1992; 77 (A3): abstract A3. 10. Watcha MF. Effect of propofol on the incidence of post operative vomiting after strabismus surgery in pediatric outpatients. Anesthesiology 1991; 75: 204–209. 11. Andersen R, Krohg K. Pain as a major factor of postoperative nausea. Canadian Anaesthetists Society Journal 1976; 23: 366– 369. 12. Wrench IJ, Ward JE, Walder AD. The prevention of postoperative nausea and vomiting using a combination of ondansetron and droperidol. Anaesthesia 1996; 51: 776–778.
Figure 1 Patient No. 1. A 6-yr-old boy with neurofibromatosis who presented for tracheostomy and debulking of a large intraoral tumour.
Sevoflurane for difficult intubation in children Sir,—We read with interest the case reports by Mostafa and Atherton1 in which sevoflurane was used to induce anaesthesia in three adult patients in whom difficult tracheal intubation was anticipated. We wish to report two paediatric patients with a difficult airway where sevoflurane was used as the induction agent. The first patient was a 6-yr-old boy with neurofibromatosis who presented for tracheostomy and debulking of a large intraoral tumour (fig. 1). He also had a 6 month’ history of progressive difficulty in breathing and stridor. Inhalation induction was planned because of airway obstruction. EMLA cream was applied to both hands 1 h before induction. I.v. access was established. Anaesthesia was induced with sevoflurane in oxygen. The concentration of sevoflurane was increased gradually until an end-tidal concentration of 5.5% was achieved. Direct laryngoscopy was performed and revealed a grade III larynx which was grossly deviated to the left. The larynx was intubated uneventfully with a 5.0-mm plain Portex tube orally. Oxygen saturation was maintained at 99% throughout induction. The second patient was a 11-yr-old girl with a fixed right temporal mandibular joint and a very receded mandible. Her inter-incisor distance was only 1 cm at full mouth opening. Fibreoptic intubation under inhalation induction was planned. I.v. access was established before induction. Anaesthesia was induced with incremental concentrations of sevoflurane in oxygen. A nasal airway was inserted via the left nostril and attached to a T-piece. When the end-tidal sevoflurane concentration reached 5.5%, the trachea was intubated with a 5.5-mm plain Portex tube nasally via the flexible fibreoptic laryngoscope. Both surgery and anaesthesia were uneventful. A difficult airway in paediatric patients presents a challenge to anaesthetists. Inhalation induction is the technique of choice because it is often difficult to perform awake fibreoptic intubation or tracheostomy under local anaesthesia in children. We agree with Mostafa and Atherton that sevoflurane has many advantages compared with halothane in the management of the difficult airway in both children and adults. The low blood:gas solubility of sevoflurane and consequent rapid induction and rapid recovery are reassuring features in the management of difficult intubation.
C. Y. WANG C. L. CHIU A. E. DELILKAN Department of Anaesthesia University of Malaya Kuala Lumpur, Malaysia 1. Mostafa SM, Atherton AMJ. Sevoflurane for difficult tracheal intubation. British Journal of Anaesthesia 1997; 79: 392–393.
Caudal tramadol for postoperative analgesia in hypospadias surgery Sir,—Prolongation of caudal anaesthesia is indeed a laudable aim. However, I have some concern with the recent article describing the use of tramadol in this role.1 The evidence that tramadol may be effective in this regard is scanty. There is little evidence that it is safe or effective by the extradural route in adults.
Correspondence
409
We do not know the incidence of neurological damage that may be caused by this drug, and just one child with a painful or disabling neuropathy would be disastrous. Until evidence emerges of its safety and efficacy in adults, I do not believe that studies of this nature should be carried out in children.
W. RUSSELL Intensive Care Unit Leicester Royal Infirmary Leicester 1. Prosser DP, Davis A, Booker PD, Murray A. Caudal tramadol for postoperative analgesia in paediatric hypospadias surgery. British Journal of Anaesthesia 1997; 79: 293–296. Sir,—Thank you for the opportunity to respond to Dr Russell about our caudal tramadol study.1 Although we agree in principle that studies should not involve children if the relevant data can be obtained from adults, this is rarely possible. In this instance, the significant age-related differences in opioid pharmacokinetics and pharmacodynamics,2 and vascularity, fat content and volume of the extradural space3 4 make extrapolation of clinical data from adults to young children rather suspect. Although some initial data may be gleaned from studies of drugs injected into the lumbar extradural space in adults, our results supported the concept that data achieved in this way cannot be used to predict accurately drug pharmacodynamics after caudal injection in children. All but one of 10 adult studies and the one paediatric study published in the past decade that have examined the effects of tramadol after extradural administration have recorded encouraging results. Our findings would not have been anticipated from meta-analysis of these studies. Furthermore, for various well recognized reasons, administration of drugs into the caudal extradural space is much less common in adults than in the very young. If studies examining caudal administration of drugs in adults were required before paediatric studies could proceed, we would often have to rely solely on data from non-randomized, unblinded, uncontrolled studies carried out in developing countries. The second point about drug safety is a thorny issue but one that is always raised at Paediatric Ethics Committee meetings when a study of a “new” drug is being reviewed. To prove the safety of any drug is almost impossible, but a thorough examination of all published (and any available unpublished) clinical and experimental evidence offers the best possible compromise between ensuring optimal safe care for our patients and advancement of our knowledge base. In this instance we were able to point out that tramadol, which has been widely available in the rest of Europe for more than 20 yr, has an exceptional safety record5; to the best of our knowledge no long-term adverse side effects attributable to its use have been reported. Finally, we did not perform this study because it was possible or because we wished to write another article; we believed that there was a problem (postoperative pain in children undergoing day-case hypospadias surgery) that could be prevented or alleviated. On the basis of previously published clinical work, we hypothesized that the analgesic effect produced by bupivacaine could be usefully prolonged by the addition of tramadol. Similar studies using more conventional opioids have shown that delayed respiratory depression may, occasionally, be a problem.6 7 Tramadol has the important and significant advantage over other commonly used opioids of having a marked lack of respiratory depressant effects.8 9 If we had found tramadol to be clinically useful in inpatients, then we would have extended our study to day-case patients. In the light of our initial findings, we did not believe that a study extension was justifiable. Many of these same arguments were presented and examined carefully by our Paediatric Ethics Committee before commencement of this study. One of the guiding principles used by this Committee in determining whether or not a study should be allowed to proceed is: “Children are not small adults and research should only be done in this group of patients if comparable research on adults cannot answer the same questions”. We believe, as did the Ethics Committee, that this principle was strictly adhered to in the conduct of this study.
D. P. PROSSER A. J. M. DAVIS P. D. BOOKER A. MURRAY Alder Hey Royal Liverpool Children’s NHS Trust Liverpool
1. Prosser DP, Davis A, Booker PD, Murray A. Caudal tramadol for postoperative analgesia in paediatric hypospadias surgery. British Journal of Anaesthesia 1997; 79: 293–296. 2. Olkkola KT, Hamunen K, Maunuksela EL. Clinical pharmacokinetics and pharmacodynamics of opioid analgesics in infants and children. Clinical Pharmacokinetics 1995; 28: 385– 404. 3. Hogan QH. Epidural anatomy examined by cryomicrotome section. Influence of age, vertebral level and disease. Regional Anesthesia 1996; 21: 395–406. 4. Pullerits J, Holzman RS. Pediatric neuraxial blockade. Journal of Clinical Anesthesia 1993; 5: 342–354. 5. Gibson TP. Pharmacokinetics, efficacy and safety of analgesia with a focus on tramadol HCL. American Journal of Medicine 1996; 101: 47S–53S. 6. Tyler DC, Krane EJ. Epidural opioids in children. Journal of Pediatric Surgery 1989; 24: 469–473. 7. Cook B, Doyle E. The use of additives to local anaesthetic solutions for caudal epidural blockade. Paediatric Anaesthesia 1996; 6: 353–359. 8. Vickers MD, O’Flaherty D, Szekely SM, Read M, Yoshizumi J. Tramadol: pain relief by an opioid without depression of respiration. Anaesthesia 1992; 47: 291–296. 9. Houmes RJM, Voets MA, Verkaaik A, Erdmann W, Lachmann B. Efficacy and safety of tramadol versus morphine for moderate and severe postoperative pain with special regard to respiratory depression. Anesthesia and Analgesia 1992; 74: 510–514.
Bair Hugger active patient warming system Sir,—I read with interest the case report of Ayala and Coe regarding softening of tracheal tubes when used in conjunction with the Bair Hugger patient warming system.1 As a result of their case report and subsequent study, they recommended the avoidance of uncut PVC tubes, unless adequately supported, with this warming system. Having personally experienced a similar problem I have a simple and effective solution. This is to make a small hole in the clear plastic region of the Bair Hugger (designed to cover the patient’s head) and to pass the tracheal tube through this, thus effectively removing it from the warmed environment. I have encountered no recurrence of the problem since adopting this method.
I. R. TAYLOR Department of Anaesthesia Queen Alexandra Hospital Portsmouth 1. Ayala JL, Coe A. Thermal softening of tracheal tubes: an unrecognised hazard of the Bair Hugger active patient warming system. British Journal of Anaesthesia 1997; 79: 543–545.
Inappropriate use of placebo Sir,—Fujii, Toyooka and Tanaka compared granisetron with placebo for reduction of postoperative nausea and vomiting (PONV) in middle ear surgery.1 First, when established treatments are available, research must be compared with existing treatments, not with placebo. I wish to know if granisetron is more or less effective than other antiemetics on offer. Second, the same authors had already proved the effectiveness of granisetron in PONV in gynaecological patients (including a placebo group).2 Third, by withholding antiemetic therapy in the placebo group until significant morbidity from PONV had occurred represents sub-standard medical care.
J. CLARKE Department of Anaesthesia, Queen Elizabeth Hospital Woodville, Adelaide, Australia 1. Fujii Y, Toyooka H, Tanaka H. Granisetron reduces the incidence of nausea and vomiting after middle ear surgery. British Journal of Anaesthesia 1997; 79: 539–540. 2. Fujii Y, Tanaka H, Toyooka H. Optimal anti-emetic doses of granisetron for preventing post operative nausea and vomiting. Canadian Journal of Anaesthesia 1994; 41: 794–797. Sir,—We have already studied the efficacy of granisetron, a selective 5-hydroxytryptamine type-3 receptor antagonist, in preventing postoperative nausea and vomiting (PONV) after gynaecologi-
410
British Journal of Anaesthesia
cal surgery.1 2 However, there is no study investigating its efficacy in preventing PONV in patients undergoing middle ear surgery. The incidence of PONV after this surgical procedure is high (62– 80%) when no prophylactic antiemetic is given.3 Therefore, we studied the efficacy of granisetron in the prevention of PONV in this population. Several investigators have previously compared the antiemetic efficacy of such new agents as ondanstron, granisetron, tropisetron and dorasetron with placebo for preventing PONV.4–7 Similarly, we compared the efficacy of granisetron and placebo for the prevention of PONV after middle ear surgery.
Y. FUJII Department of Anaesthesiology University of Tsukuba Institute of Clinical Medicine Tsukuba City, Ibaraki, Japan 1. Fujii Y. Reduction of postoperative nausea and vomiting with granisetron. Canadian Journal of Anaesthesia 1994; 41: 291–294. 2. Fujii Y. Optimal anti-emetic doses of granisetron for preventing postoperative nausea and vomiting. Canadian Journal of Anaesthesia 1994; 41: 794–797. 3. Honkevaara P. Prevention of nausea and vomiting with transdermal hyoscine in adults after middle ear surgery during general anaesthesia. British Journal of Anaesthesia 1994; 73: 763–766. 4. Honkavarra P. Effect of ondansetron on nausea and vomiting after middle ear surgery during general anaesthesia. British Journal of Anaesthesia 1996; 76: 316–318. 5. Wilson AJ. Single-dose i.v. granisetron in the prevention of postoperative nausea and vomiting. British Journal of Anaesthesia 1996; 76: 515–518. 6. Capouet V. Single dose i.v. tropisetron in the prevention of postoperative nausea and vomiting after gynaecological surgery. British Journal of Anaesthesia 1996; 76: 54–60. 7. Graczyk SG. Intravenous dorasetron for the prevention of postoperative nausea and vomiting after outpatient laparoscopic gynecologic surgery. Anesthesia and Analgesia 1997; 84: 325–330.
Bradycardia during rapid inhalation induction with sevoflurane in children colleagues,1
Sir,—In common with Sigston and we have found inhalation induction with 8% sevoflurane to be rapid and well tolerated in young children. However, the development of profound bradycardia in four children has caused us to reflect on its use. These were healthy children, aged 6 months to 2 yr, undergoing minor surgery. None was premedicated and induction was by inhalation of 8% sevoflurane and 66% nitrous oxide in oxygen, using a standard technique. In all cases the onset of bradycardia occurred during induction with no loss of airway or ventilation. One infant required treatment with atropine because of clinical evidence of a decrease in cardiac output. In the other children heart rate recovered spontaneously when the concentration of sevoflurane was reduced or we changed to using isoflurane. In the study of Sigston and colleagues, bradycardia occurred in one child who received halothane (age unknown) but not in the sevoflurane group. It is noteworthy that all of their patients received atropine premedication. Baum, Yemmen and Baum reported no arrhythmias when comparing 8% sevoflurane with incremental sevoflurane in unpremedicated children.2 Johannesson, Floren and Lindahl found an incidence of cardiac arrhythmias of 5% while using incremental sevoflurane in premedicated patients.3 Otherwise, incremental induction with sevoflurane in children seems to show an increase or have little effect on heart rate.4 Those familiar with halothane induction in infants may admonish us for not routinely prescribing atropine where an inhalation induction is planned, if not to dry secretions and thus reduce airway complications, then to offset the myocardial depression which can accompany high concentrations of a volatile agent. The cardiovascular depression produced by sevoflurane is much less than that with halothane, but not absent. The single-handed anaesthetist, anticipating difficult i.v. cannulation, would be well advised to consider atropine premedication when planning inhalation induction with high concentrations of sevoflurane in infants.
P. TOWNSEND M. A. STOKES Department of Anaesthetics Birmingham Children’s Hospital NHS Trust Birmingham
1. Sigston PE, Jenkins AMC, Jackson EA, Sury MRJ, Mackersie AM, Hatch DJ. Rapid inhalation induction in children: 8% sevoflurane compared with 5% halothane. British Journal of Anaesthesia 1997; 78: 362–365. 2. Baum VC, Yemmen TA, Baum LD. Immediate 8% sevoflurane induction in children: a comparison with incremental sevoflurane and incremental halothane. Anesthesia and Analgesia 1997; 85: 313–316. 3. Johannesson GP, Floren M, Lindahl SGE. Sevoflurane for ENT-surgery in children. A comparison with halothane. Acta Anaesthesiologica Scandinavica 1995; 39: 546–550. 4. Kern C, Erb T, Frei FJ. Haemodynamic responses to sevoflurane compared with halothane during inhalational induction in children. Paediatric Anaesthesia 1997; 7: 439–444.
Effect of temperature on diaphragmatic function Sir,—The article by Mills and colleagues described the effect of a 5⬚C temperature change on the evoked compound potential in the human diaphragm.1 They found a reduction in the EMG of approximately 60%. This is in agreement with our own findings using electromyography carried out in five children undergoing cardiac surgery under profound hypothermia. We found this effect to be almost completely reversed by edrophonium and concluded it was most likely a result of decreased production or release of acetylcholine (ACh).2 Further experiments in dogs confirmed this effect.3 The recent work of England added support for the view that in non-hibernating mammals, cold critically affects ACh production or release.4 In the past, much of our information on the effects of hypothermia has come from experiments carried out either in cold blooded animals or on tissues from rats which are genetically closely allied to hibernating animals. These species do not necessarily respond to cold in the same manner as humans and as a result conflicting findings have been reported. This article confirms the critical effect of cold in human neuromuscular conduction. If the effect of hypothermia in humans is caused by a reduction in ACh release, this also has implications for its effect on neuromuscular blocking drugs.5
S. FELDMAN London 1. Mills GH, Khan ZP, Moxham J, Desai J, Forsyth A, Ponte J. Effects of temperature on phrenic nerve and diaphragmatic function during cardiac surgery. British Journal of Anaesthesia 1997; 79: 726–732. 2. Feldman SA. Muscle Relaxants. Philidelphia: WB Saunders, 1979; 14–146. 3. Thornton RJ, Feldman SA, Blackeney C. The effect of hypothermia on neuromuscular conduction in dogs. Abstracts of the 5th World Congress of Anesthesiologists. Amsterdam: Excerpta Medica, 1975; 12. 4. England AJ. Effect of Hypothermia on Neuromuscular Conduction and Blockade, MD Thesis. London: University of London, 1995. 5. Feldman S. Neuromuscular Block. Oxford: Butterworth Heinemann, 1996; 128–138.
Sir,—We thank Professor Feldman for his comments on our results. We agree entirely that the most likely mechanism which may explain our observations is indeed impairment of production or release of acetylcholine (ACh). We regret that at the time this did not occur to us and therefore we did not perform the simple edrophonium test. Examination of the human data of Feldman[2] (quoted by Feldman) suggests that the effects of hypothermia on twitch height of arm muscles start at approximately 34.5⬚C. Interestingly, our data (especially during the transition from warm to cold) suggest that the effects of hypothermia on the diaphragm may be detectable at temperatures even closer to 36⬚C. One could speculate that there may be differences between the diaphragm and limb muscles in this respect, perhaps because the temperature within limb muscles is often hypothermic in inactive, normal subjects exposed to cold, whereas the diaphragm is not.
J. C. PONTE Academic Department of Anaesthetics King’s College School of Medicine and Dentistry London