1 Montréal, Canada 2 Gent, Belgium
LMA® is the property of Intavent Ltd.
EditorWe read with interest the article by De Baerdemaeker and colleagues1 comparing desflurane with sevoflurane in morbidly obese patients. There are several methodological and scientific points in it that stimulate controversy, and we cannot support the overall implication that desflurane may be the volatile anaesthetic of choice for these patients.
(i) This is not a double-blind study. In a study comparing recovery profiles or haemodynamic profiles of two volatile anesthetics, this surely is a source of bias. Why did the authors not mention this in their discussion? As the study was designed to titrate the volatile anaesthetic to a pre-defined BIS value, blinding would not have been difficult.
(ii) Anybody who routinely uses BIS monitoring knows of the problems of interaction with other devices in the operating room. Our experience with BIS monitoring in gastric banding surgery confirms this. However, in the Results section, there is no mention of the amount of time when BIS monitoring was impossible. The A-2000 BIS monitor has several ways of controlling for the validity of the BIS values: BIS values that have been recorded with the EMG column indicating facial muscle activity and BIS values that are not indicated in bold are an indication of interference or poor recording and should be questioned. Interference from electrocautery can easily be identified on the raw EEG. A study that uses all of these values and tries to determine the superiority of hypnosis with any volatile anaesthetic by recording the percentage of time when BIS is within a certain limit should also mention the percentage of time when BIS readings were not possible.
(iii) In their conclusions, the authors state that fewer episodes of hypotension were found in the desflurane group. Although this is true regarding the percentage of time when the MAP was <60 mm Hg (17% in the sevoflurane group vs 9% in the desflurane group), they also state in their results that an episode of MAP <50 mm Hg was found in 16 of the 25 patients in the desflurane group, but only 8 of the 25 patients in the sevoflurane group. What exactly is one episode of hypotension? How justified is it to conclude that desflurane gives superior haemodynamic control?
(iv) The authors state that recovery was significantly faster with desflurane than with sevoflurane. As determination of recovery time is highly subjective, the lack of even a single-blinded study design should be discussed as it might have influenced these results. Even if recovery is faster after desflurane, how significant clinically is a mean difference in recovery time of only 2 min? This is especially important with respect to the potential risks of desflurane in morbidly obese patients, which the authors fail to mention.
(v) The authors cite one study,2 and state that morbidly obese patients are at high risk of both aspiration and upper-airway obstruction after tracheal extubation. Why then do they not mention the potential danger of desflurane in these patients? Several studies have found that desflurane increases airway resistance,3 4 whereas sevoflurane has a bronchodilatory effect.5 That is why there is controversy over using desflurane with the laryngeal mask airway (LMA).6 7 The authors have excluded patients with significant cardiopulmonary disease and obstructive lung disease. Why? In morbidly obese patients who have high incidence of sleep apnoea syndrome, and are at risk from even slight increases in airway resistance during or after surgery, desflurane does not seem to be the volatile anaesthetic of choice.
Overall, this article has several flaws in its design (not double blind), interpretation of the results (haemodynamic control better with desflurane?), and the discussion (no mention of the pulmonary risks of desflurane), which questions the validity of its conclusions. The authors have shown that recovery with desflurane is marginally faster than with sevoflurane. This alone does not favour its use in morbidly obese patients.
T. M. Hemmerling
J. F. Olivier
Montréal, Canada
EditorI would like to thank Drs Hemmerling and Olivier for their interest in our paper and for the opportunity to reply to them. I will address their points in turn:
(i) We agree that setting up this study in a double-bind way would have reduced bias, but since we used bolus administration of volatile anaesthetics this is not easily done. This would require an additional, blinded anaesthetist who has to be blinded from the monitor screen, the vaporizer, the sounds of the vaporizer, and even the smell of the anaesthetic gases. This was not feasible in this study. For the immediate recovery data, observations were done by nurses who had received no information on the study protocol nor on the aims of the study. The Rugloop© data manager allows very accurate registration of events such as recovery variables. The nurse made the observations while the investigator marked the events on the Rugloop file. So in this respect these data were not biased.
(ii) The decision to alter the setting of the vaporizer or to give an inhalation bolus was based on valid BIS readings in the absence of any interference. I would like to stress the fact that the BIS triggerpoints had to be exceeded for >30 s, which also reduces the influence of artefactual readings. In this study, the most recent BIS 2000 XP electrodes were used. As claimed by Aspect Medical, the makers of the BIS monitor, and also in our own experience, these electrodes give few invalid readings. Moreover, the periods of electrocautery interference are short-lived and mainly occur when small skin incisions for trochar placement at the beginning of surgery are made. During laparoscopy, coagulation did not interfere with the recordings. The Rugloop data collector only registers valid BIS readings, and excludes invalid readings. During surgery, <1% of data were invalid on the raw datafiles.
(iii) As blood pressure was measured non-invasively every 2.5 min, an episode is a period >2.5 min. The fact that the overall duration of hypotension was longer in the sevoflurane group, but with a smaller number of hypotensive episodes, means that the episodes of hypotension lasted longer in the sevoflurane group. Figures 3 and 4 show that in the sevoflurane group, hypotensive episodes were longer, resulting in greater haemodynamic instability.
(iv) Because of the pharmacokinetic differences between sevoflurane and desflurane, desflurane will be the agent producing a faster recovery, and our data confirm this even when both groups were kept at a similar depth of hypnosis as measured by BIS. Statistically, this difference was significant. Whether these 2 min are clinically relevant is, we accept, questionable.
In studying recovery after desflurane, propofol or isoflurane anaesthesia in morbidly obese patients, Juvin and colleagues8 concluded that even slight improvements in early recovery may be beneficial. From his study, we have adapted the use of immediate and intermediate data in the evaluation of postoperative recovery. Our postoperative data included mobility scores, SpO2 while breathing air, the Aldrete and Kroulik score, VAS Score, PONV incidence, OAA/S scores, amount of antiemetic and piritramide consumption. These data are still in the process of analysis.
(v) The focus of this study was mainly on the dynamic use of the inhalation bolus as a tool to control depth of hypnosis and its effects on BIS profiles, haemodynamic stability and immediate recovery. We did not focus on the respiratory effects of the inhalation boli. Dikmen and colleagues3 subjected their patients to 2 MAC desflurane for 5 min. In our study, high inspiratory concentrations were used for only 30 s. Desflurane end-tidal concentrations were on average <1 MAC. We agree that in the literature there are studies suggesting an increase in airway resistance at a concentration of 2 MAC, but Drs Hemmerling and Olivier must agree that most results are conflicting. For instance, Goff and colleagues4 found that desflurane-induced bronchoconstriction was primarily noted in smokers. There is a time factor involved: the author states that desflurane lacks the ability to cause early bronchodilation. Maybe desflurane requires more time to cause it. Our postoperative data on SpO2 showed no difference between groups. These data were double blinded, the PACU nurses were blinded, and patients were breathing air.
Clinically, we did not observe any pulmonary events attributable to bronchoconstriction during anaesthesia with desflurane, even when using 30 s boli of high inspiratory concentrations equal to or exceeding 2 MAC inspiratory concentration.
After induction, all patients received 1 MAC of sevoflurane or desflurane for 5 min. As Dikmen and colleagues3 have shown, in these concentrations desflurane does have bronchodilatory effects, albeit less then sevoflurane.
What about the range between 1 and 2 MAC? Mercier and colleagues9 reported in their study on relaxation of isolated human bronchi that the volatile anaesthetics, halothane, isoflurane, and desflurane all showed dose-dependent bronchodilation at 0.66, 1.33, and 2 MAC. Effects on proximal bronchi were similar, but halothane was more potent than desflurane on the distal bronchi, with isoflurane being intermediate.
Morbidly obese patients with COPD, or sleep apnoea disorders with a potential for right ventricular or biventricular failure or even pulmonary hypertension are, in my opinion, a totally different study population with a higher anaesthetic risk. As there are no data on the use of the concept of inhalation bolus in the morbidly obese, we decided to exclude these patients.
We feel that our conclusions reflect our results. We did not state that desflurane is the best or the sole agent for use in the morbidly obese patient. The pharmacokinetic concept of inhalation bolus administration of volatile agents was the main purpose of this study. Currently, we are investigating the respiratory effects of inhalation bolus administration.
L. De Baerdemaeker
Gent, Belgium
References
1 De Baerdemaeker LE, Struys MM, Jacobs S, et al. Optimization of desflurane administration in morbidly obese patients: a comparison with sevoflurane using an inhalation bolus technique. Br J Anaesth 2003; 91: 638502
2 Rose DK, Cohen MM, Wigglesworth DF, DeBoer DP. Critical respiratory events in the postanesthesia care unit. Patient, surgical, and anesthetic factors. Anesthesiology 1994; 81: 41018[ISI][Medline]
3 Dikmen Y, Eminoglu E, Salihoglu Z, Demiroluk S. Pulmonary mechanics during isoflurane, sevoflurane and desflurane anaesthesia. Anaesthesia 2003; 58: 7458[CrossRef][ISI][Medline]
4 Goff MJ, Arain SR, Ficke DJ, Uhrich TD, Ebert TJ. Absence of bronchodilation during desflurane anesthesia: a comparison to sevoflurane and thiopental. Anesthesiology 2000; 93: 4048[CrossRef][ISI][Medline]
5 Rooke GA, Choi JH, Bishop MJ. The effect of isoflurane, halothane, sevoflurane, and thiopental/nitrous oxide on respiratory system resistance after tracheal intubation. Anesthesiology 1997; 86: 12949[CrossRef][ISI][Medline]
6 Ebert T. Airway responses to sevoflurane vs desflurane: comparing apples to oranges? Anesth Analg 2003; 97: 1206
7 Eshima RW, Maurer A, King T, et al. A comparison of airway responses during desflurane and sevoflurane administration via a laryngeal mask airway for maintenance of anesthesia. Anesth Analg 2003; 96: 7015
8 Juvin P, Vadam C, Malek L, Dupont H, Marmuse JP, Desmonts JM. Postoperative recovery after desflurane, propofol, or isoflurane anesthesia among morbidly obese patients: a prospective, randomized study. Anesth Analg 2000; 91: 71419
9 Mercier FJ, Naline E, Bardou M, et al. Relaxation of proximal and distal isolated human bronchi by halothane, isoflurane and desflurane. Eur Respir J 2002; 20: 28692