1 Rhyl, UK 2 Wirral, UK 3 Manchester, UK
EditorWe were surprised at the conclusion reached by Moore and colleagues1 in their evaluation of sevoflurane for paediatric day-case anaesthesia, as the study data do not support it. The reported adverse events during induction were no more than would be expected during a typical gas induction. In addition, the significance of agitation and distress in the recovery area is difficult to gauge without details of an appropriate scoring system to include concomitant pain. Furthermore, though a statistical significance is derived, no clinical significance was identified or reported. The primary outcome measure was postoperative nausea and vomiting (PONV). An ideal study to evaluate this should be randomized and double-blind, where all compounding factors are evenly distributed between study groups.2 We accept that the mode of induction could not have been blinded, but no attempt appears to have been made either to eliminate or to stratify factors that would influence this primary outcome measure.3 4 These include, in order of the anaesthetic process:
Preoperative:
Previous PONV
Motion sickness
Anxiety, and the potential adverse influence of the anxious parent present at induction
Degree of hydration as reflected by fasting time
Intraoperative:
Gastric distension associated with hand ventilation following administration of a muscle relaxant
Intraoperative fluid administration
Duration of surgery
Opioid administration: fentanyl or alfentanil, not at a standard dose
Neostigmine for reversal of non-depolarising neuromuscular block
Postoperative:
Postoperative pain, the level of which may or may not have been influenced by regional local anaesthetic blockade
All aspects of the recovery environment including personnel deployed
The significance of the emetic episodes recorded in the study that, in any event, did not require an overnight stay, is therefore difficult to assess. In addition, propofol, administered to the group showing a lower incidence of nausea and vomiting, is itself considered to have antiemetic properties.5 Nausea is a subjective phenomenon and it is accepted that the smaller child may not be able to describe or gauge the severity of symptoms. An appropriate study end-point therefore allows a milder degree of nausea to be acceptable, provided that no antiemetic is needed.
Sevoflurane may not be ideal as a sole agent for paediatric day-case anaesthesia but this paper,1 arising from the Cost Effectiveness Study in Anaesthesia, in which it was not properly evaluated, does not allow such a conclusion to be made.
A. P. J. Lake
M. Khater
Rhyl, UK
EditorWe thank Drs Lake and Khater for their interest in our study. However, we disagree with their assertion that no clinical significance was identified or reported. The difference in postoperative nausea and vomiting (PONV) is highly significant between the sevoflurane/sevoflurane (S/S) and the propofol/halothane (P/H) groups; the significant differences between the groups in both PONV (Table 1) and in episodes of vomiting (Table 2) is demonstrated.
|
|
The large sample size, and meticulous randomization, further stratified by sex and by surgical specialty, resulted in recognized confounding factors being evenly distributed between groups.6 The final report and paper were prepared according to CONSORT guidelines.7
The paper is important for clinicians not because there is a single major difference between the two groups, but because the difference in both primary (PONV) and secondary outcome measures all point in the same direction. The results clearly suggest that sevoflurane is not ideal as a sole agent for this group of patients. Given that the practice of anaesthesia is moving rapidly towards the use of sevoflurane,6 this is highly relevant to clinicians. In addition, although not presented in this paper, the sevoflurane/sevoflurane arm was the more costly.8 It therefore has disadvantages on both clinical-effectiveness and cost-effectiveness grounds.
A further finding of importance to clinicians was the relatively low absolute level of PONV, despite the withholding of prophylactic antiemetics. The routine use of prophylactic antiemetics is increasing, with concomitant clinical and economic implications, and the CESA study should lead clinicians to question their practice in this regard.
J. K. Moore1
R. A. Elliott2
1Wirral, UK
2Manchester, UK
References
1 Moore JK, Moore EW, Elliott RA, St Leger AS, Payne K, Kerr J. Propofol and halothane vs. sevoflurane in paediatric day-case surgery: induction and recovery characteristics. Br J Anaesth 2003; 90: 4616
2 Rose JB, Watcha MF. Postoperative nausea and vomiting in paediatric patients. Br J Anaesth 1999; 83: 10417
3 Lerman J. Surgical and patient factors involved in postoperative nausea and vomiting. Br J Anaesth 1992; 69 [Suppl.]: S24S32[Medline]
4 Kenny GNC. Risk factors for postoperative nausea and vomiting. Anaesthesia 1994; 49[Suppl.]: 610[ISI][Medline]
5 Kim SI, Han TH, Kil HY, Lee JS, Kim SC. Prevention of postoperative nausea and vomiting by continuous infusion of subhypnotic propofol in female patients receiving intravenous patient-controlled analgesia. Br J Anaesth 2000; 85: 898900
6 Elliott RA, Payne K, Moore JK et al. Which anaesthetic agents are cost effective in day surgery? Literature review, national survey of practice and randomised controlled trial. Health Technol Assess 2002; 6: 1264[Medline]
7 Altman DG, Schulz KF, Moher D et al. The revised CONSORT statement for reporting randomised trials: explanation and elaboration. Ann Int Med 2001; 134: 66394
8 Elliott RA, Payne K, Moore JK, et al. Clinical and economic choices in anaesthesia for day surgery: a prospective randomised controlled trial. Anaesthesia 2003; 58: 41221[ISI][Medline]