Optimization of desflurane administration in morbidly obese patients: a comparison with sevoflurane using an ‘inhalation bolus’ technique

L. E. C. De Baerdemaeker*,1, M. M. R. F. Struys1, S. Jacobs1, N. M. M. Den Blauwen1, G. R. P. J. Bossuyt1, P. Pattyn2 and E. P. Mortier1

Departments of 1 Anaesthesiaand 2 Surgery, Ghent University Hospital, De Pintelaan 185, B-9000, Gent, Belgium

Corresponding author. E-mail: luc.debaerdemaeker@UGent.be

Accepted for publication: July 8, 2003


    Abstract
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 References
 
Background. The concept of an ‘inhalation bolus’ can be used to optimize inhaled drug administration. We investigated the depth of anaesthesia, haemodynamic stability, and recovery time in morbidly obese patients resulting from bispectral indexTM (BISTM)-guided sevoflurane or desflurane administration and BIS-triggered inhalation boluses of sevoflurane or desflurane combined with titration of remifentanil.

Methods. Fifty morbidly obese patients undergoing laparoscopic gastroplasty received either BIS-guided sevoflurane or desflurane anaesthesia in combination with a remifentanil target-controlled infusion. Intraoperative haemodynamic stability and BIS control were measured. Immediate recovery was recorded.

Results. Intraoperatively, the BIS was between 40 and 60 for a greater percentage of time in the sevoflurane (78 (13)% of case time) than in the desflurane patients (64 (14)% of case time), owing to too profound anaesthesia in the desflurane patients at the start of the procedure. However, fewer episodes of hypotension were found in the desflurane group, without the occurrence of more hypertensive episodes. During immediate recovery, eye opening, extubation, airway maintenance, and orientation occurred sooner in the desflurane group.

Conclusions. Immediate recovery was significantly faster in the desflurane group. Overall hypnotic controllability measured by BIS was less accurate with desflurane. Overall haemodynamic controllability was better when using desflurane. Fewer episodes of hypotension were found in the desflurane group. The use of the inhalation bolus was found to be appropriate in both groups without causing severe haemodynamic side effects. Minimal BIS values were significantly lower after a desflurane bolus.

Br J Anaesth 2003; 91: 638–50

Keywords: anaesthetics volatile; anaesthetics volatile, bolus; analgesics opioid, remifentanil


    Introduction
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 References
 
The concept of an ‘inhalation bolus’ has been introduced in an attempt to optimize inhaled drug administration.1 Adapted from the traditional concept of overpressure,2 the purpose of this bolus technique is to increase the end-tidal concentration of the volatile agent as quickly as possible to block an analgesic and/or hypnotic arousal effect, leading to a more rapid clinical effect.1 36

New methods of hypnotic-effect monitoring, such as the bispectral index (BIS, Aspect Medical Systems, Newton, MA, USA), might by used to guide the titration of inhaled anaesthetics.7 However, more data are required on the intraoperative control of BIS-guided, inhaled anaesthetics, particularly when using an inhalation bolus. Desflurane is characterized by having the most rapid pharmacokinetics because it is the least soluble of the halogenated volatile agents. An inhaled anaesthetic with a low blood-gas solubility coefficient will in theory always offer faster hypnotic control, but, unfortunately, desflurane causes sympathetic stimulation specifically during a rapid increase in end-tidal concentration to greater than approximately 6 vol. % end-tidal, making it perhaps less suitable for the ‘inhalation bolus technique’.8

Although in general the use of short-acting drugs such as desflurane, sevoflurane, and remifentanil seems obvious for hastening recovery,9 there are few data published on the intraoperative hypnotic and haemodynamic stability of anaesthesia in morbidly obese patients when using these short-acting drugs. In addition to physiological challenges, pharmacological changes associated with obesity might lead to alterations in the distribution, binding, and elimination of many drugs. The net pharmacokinetic effect in these patients is often uncertain, making drug titration even more difficult and unpredictable.10

In this study, our hypothesis was that the use of an inhalation bolus during sevoflurane, or desflurane-maintained anaesthesia, could be used to optimize the inhaled drug administration, resulting in improved overall control of depth of anaesthesia without causing haemodynamic instability or prolonged recovery times, even in morbidly obese patients. We compared the overall depth of hypnosis and the haemodynamic stability of anaesthesia in morbidly obese patients given BIS-guided sevoflurane or desflurane anaesthesia with BIS-triggered inhalation boluses of both agents in combination with remifentanil. We also compared the detailed pharmacodynamic characteristics of the inhalation bolus.


    Methods and materials
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 References
 
Clinical protocol
After institutional ethics committee (Ghent University Hospital, Gent, Belgium) approval, written informed consent was obtained from 50 morbidly obese patients (BMI >30 kg m–2), aged 18–70 yr, undergoing laparoscopic gastroplasty. Exclusion criteria included repeat surgery, a history of drug abuse, the use of ß-blockers, significant cardiopulmonary disease, renal failure (serum creatinine >120 µmol litre–1), hepatic failure (transaminases >1.5 N), or a history of allergy to anaesthetics. All patients were operated on by the same team of surgeons, using the same surgical technique (Swedish Adjustable Gastric Band; Obtech Medical, Baar, Switzerland).

Patients were allocated randomly to one of two groups, the sevoflurane group and desflurane group (see later). All the patients received midazolam 1 mg i.v. 10 min before surgery as premedication. After accurate preoxygenation (oxygen face mask with O2 10 litre min–1), a remifentanil infusion was started 2.5 min before induction via a computer-assisted continuous infusion device (RUGLOOP; see below for details) to a target plasma concentration using a three-compartment model as described by Minto and colleagues11 and clinically validated in morbidly obese patients by Egan and colleagues12 The initial target plasma concentration of remifentanil was set at 4 ng ml–1. Anaesthesia was induced with a bolus of propofol, administered at 300 ml h–1 until loss of consciousness. At loss of consciousness, rocuronium 0.9 mg kg–1 ideal body weight (IBW) (following the Table of Desirable Weights, Metropolitan Life Insurance, 1983) was given. At the same moment, pressure was applied to the cricoid; tracheal intubation was performed 60 s later. After intubation, positive-pressure ventilation with a mixture of oxygen/air (initial FIO2=0.5) was started, using an ADU ventilator (Datex-Ohmeda, Helsinki, Finland), with a fresh gas flow of 6 litres min–1 during the first 5 min and 2 litres min–1 thereafter. Tidal volume was set at 10 ml kg–1 IBW and peak airway pressure was kept below 35 cm H2O. The frequency was adjusted to achieve an end-tidal carbon dioxide pressure of 30–35 mm Hg. If required, the FIO2 was adjusted to maintain oxygen saturation above 95%.

In the sevoflurane group, the initial fresh gas flow was 6 litres min–1, with the vaporizer (Aladin; Datex-Ohmeda) set at 2 vol% (FD sevoflurane). After 5 min, the fresh gas flow was lowered to 2 litres min–1 and the FD sevoflurane was adjusted to maintain a BIS of between 45 and 55. If the BIS became less than 45 for more than 30 s, the FD sevoflurane was lowered by 25%. If the BIS exceeded 55 for more than 30 s, an inhalation bolus of sevoflurane was administered by setting the vaporizer to 8% in a fresh gas flow of 4 litres min–1 for 30 s. The fresh gas flow was then returned to 2 litres min–1 but with vaporizer setting 25% greater than the previous FD sevoflurane.

For the desflurane group, the initial fresh gas flow was 6 litres min–1, with the vaporizer set at 6 vol% (FD desflurane). After 5 min, the fresh gas flow was lowered to 2 litres min–1 and the FD desflurane was adjusted to maintain a BIS of between 45 and 55. If the BIS became less than 45 for more than 30 s, the FD desflurane was lowered by 25%. If BIS exceeded 55 for more than 30 s, an inhalation bolus of desflurane was administered by setting the vaporizer to 16% in a fresh gas flow of 4 litres min–1 for 30 s. The fresh gas flow was then returned to 2 litres min–1 but with a vaporizer setting 25% greater than the previous FD desflurane.

The target plasma concentration of remifentanil was adjusted according to clinical signs of the adequacy of analgesia. The baseline arterial pressure and heart rate were measured 5 min after tracheal intubation.

Inadequate analgesia was defined as:

a rise in systolic arterial pressure more than 15 mm Hg above the baseline;

heart rate >90 beats min–1 in the absence of hypovolaemia;

autonomic signs (e.g. sweating, salivation, flushing);

somatic signs (e.g. movement, swallowing).

If any of the above were present, the target plasma concentration of remifentanil was increased by 25%. Excessive analgesia was defined as:

a mean arterial pressure (MAP) below 60 mm Hg;

heart rate below 50 beats min–1.

In this case, the plasma concentration of remifentanil was decreased by 25%.

After every change in drug titration, a ‘lock-out’ period of 2.5 min was respected. An additional bolus of rocuronium (25% of the initial dose) was given only if required. All drug delivery was stopped after the surgical dressing had been applied. Residual muscle relaxation was assessed by ‘train of four’ (TOF) counts and if required antagonized with neostigmine/atropine (2.5 mg per 0.5 mg i.v.).

After intubation, all patients received a prophylactic i.v. antibiotic dose of 2 g cefazoline (Cefacidal®; Bristol Myers Squibb, Sermoneta, Italy), 2 g propacetamol (Prodafalgan®; Bristol Myers Squibb, Sermoneta, Italy) and 40 mg tenoxicam (Tilcotil®; Roche, Milano, Italy). To facilitate the surgical procedure, all patients were placed in a semi-recumbent position after having received a crystalloid loading dose of 10 ml kg–1 IBW. They were put in a 45° sitting position, with both arms resting horizontally on supports. Their buttocks were well supported by a special seat and the legs placed in padded supports. Adhesive tape was used to fix the head in a neutral position to prevent brachial plexus injury as a result of abnormal flexion of the heavy head.

One disadvantage of this position is hypotension from venous pooling, hence the i.v. pre-load. Where hypotension persisted during positioning, a bolus of phenylephrine (0.1 mg) was given rather than changing the remifentanil dosage. If hypotension or bradycardia occurred during surgery despite lowering the remifentanil concentration, an escape vasoactive medication could be used.

The heart rate, SpO2, end-tidal carbon dioxide concentration, and inspiratory and end-tidal anaesthetic drug concentrations were measured continuously with an S5 monitor (Datex-Ohmeda, Helsinki, Finland). The arterial pressure was measured non-invasively every 2.5 min using the same monitor. The BIS (version 3.4) was derived from the frontal EEG (At-Fpzt) and calculated by the A-2000 BIS monitor, using a BIS sensor (Aspect Medical Systems, Inc., Newton, MA, USA). The smoothing time of the BIS monitor was set at 15 s. All data were continuously recorded automatically, using the RUGLOOP data manager (see below). Induction of anaesthesia was defined as the time period from the start of propofol injection until loss of consciousness. At loss of consciousness, the BIS, induction time and amount of propofol used were recorded, as was the time until intubation. The maintenance phase of anaesthesia was defined as the time between loss of consciousness and the moment of drug discontinuation.

RUGLOOP is a general infusion-pump control and data-management system, written by T. De Smet and M. Struys, Ghent University, Belgium (more information: http://allserv.rug.ac.be/~mstruys). It is written in Visual C++ (Microsoft, Redmond, WA, USA) for Windows 2000TM operating systems. It is able to deliver a computer-controlled infusion, targeting either the plasma or effect-site concentrations by using a combination of compartmental pharmacokinetic and effect-site models.

After all drug delivery had been stopped, the fresh gas flow was returned to 6 litres min–1 with an FIO2 of 1. Thereafter, the following recovery variables were recorded: time from drug discontinuation until spontaneous breathing, eye opening, extubation, free airway (being able to breathe unobstructed without the help of airway devices or external manoeuvres), and orientation (saying name, date and location on request, assessed every 20 s). To obtain a non-biased return of spontaneous ventilation and other recovery variables, the end-tidal carbon dioxide must be controlled. Therefore, 2 min after drug discontinuation, ventilation was stopped and manual-breathing support was used (one breath every 30 s until the return of spontaneous ventilation except if the end-tidal carbon dioxide became higher than 60 mm Hg, in which case manual support was continued until the carbon dioxide returned to less than 50 mm Hg).

Off-line detailed analysis of the inhaled bolus
Each bolus gave rise to a specific BIS pattern. Therefore, we studied the following characteristics for each bolus and compared these between the two groups. For each bolus, we defined three theoretical time-points as illustrated in Figure 1: (i) the trigger point, defined as the moment that the BIS increased to 55 for 30 s; (ii) the cross point, defined as the moment at which the BIS returned to 55 (and lower); and (iii) the recovery point, defined as the time at which the BIS returned to 45 (and higher) after having overshot to below 45. As a result, we ended up with two time periods: (i) time A, defined as the time between the trigger point and the cross point; and (ii) time B, defined as the time between the cross point and the recovery point. We also defined two BIS values during the bolus period: the maximum BIS during time A and the minimum BIS during time B.



View larger version (13K):
[in this window]
[in a new window]
 
Fig 1 Schematic representation of how variables were defined in the detailed analysis of the BIS profiles during an inhalation bolus.

 
Statistics
A sample size of 22 patients was determined using a power analysis (alpha=0.05 and beta=0.2), designed to detect a difference of >=25% in the intraoperative period of accurate control (% of anaesthesia time with BIS between 40 and 60) between the groups. Our assumptions for the power calculation were extrapolated from a previous report on BIS-guided, manual propofol administration from which propofol and sevoflurane kinetics were assumed to be similar.13

Statistical significance was set at 5% unless otherwise stated. For all data-sets, a Gaussian distribution was tested using the Kolomogorov–Smirnov test. Between groups, continuous data were analysed using the independent samples t-test or the Mann–Whitney test, where appropriate. Categorical data were analysed using Fisher’s exact test. Within-group statistics were evaluated using RMANOVA. All statistical tests were performed with SPSS v 10.0 (SPSS Inc., Chicago, IL, USA).

To assess the differences in BIS between groups every 10 s during maintenance, defined as the interval from the beginning of loss of consciousness (0 s for all patients) up to 6000 s (insufficient data thereafter), the difference between the mean BIS values was plotted against time. The closer the differences between the means were to zero the less difference there was between the BIS values of the two groups. The 95% confidence intervals (CI) for the differences between the mean BIS values were calculated at each 10-s time point. When zero is included in the 95% CI, there is no statistically significant difference between the BIS values of the two groups.

For the measures of immediate recovery, a survival distribution function was generated, Kaplan–Meier survival curves were produced and the significance between groups was calculated using the log-rank test. These tests were analysed using SAS statistical software version 8.2 (SAS Institute Inc., Cary, NC, USA).


    Results
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 References
 
The characteristics of the 25 patients in each group were similar (Table 1). As shown in Table 2, the induction profiles (time until loss of consciousness, BIS at loss of consciousness, and amount of propofol used for induction) were similar in both groups. The time until intubation and the total duration of anaesthesia were also similar. Figure 2 shows the individual target plasma and effect-site concentrations of remifentanil (adjusted for IBW) and the individual inspiratory and end-tidal concentration of the inhaled anaesthetics. No significant difference between the groups was detected in the amount of remifentanil used during anaesthesia (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 1 Patient characteristics (mean (SD) (minimum–maximum))
 

View this table:
[in this window]
[in a new window]
 
Table 2 Induction, maintenance and recovery characteristics (time 0 is defined as the start of propofol induction). Values are mean (SD). *P<0.05 between groups (independent sample t-test)
 


View larger version (45K):
[in this window]
[in a new window]
 
Fig 2 Individual calculated remifentanil plasma (Cp) and effect-site (Ce) concentrations in the desflurane (A and B) and sevoflurane (E and F) groups. Individual inspiratory (FI) (C and G) and end-tidal (D and H) concentrations for desflurane and sevoflurane.

 
The individual haemodynamic profiles are plotted in Figures 3 and 4 for the sevoflurane and desflurane groups, respectively. There were some statistically significant differences in haemodynamic stability between the groups (Table 3). The percentage of time with hypotension (MAP below 60 mm Hg) was higher in the sevoflurane than in the desflurane group. Episodes of more pronounced hypotension were also recorded in both groups. An episode of MAP lower than 50 mm Hg was found in eight patients in the sevoflurane group and 16 patients in the desflurane group. Out of these, three patients in the sevoflurane group and five patients in the desflurane group had an episode of MAP lower than 40 mm Hg. Episodes of (severe) hypotension were mostly encountered when placing the patient in the semi-recumbent position and after releasing the pneumoperitoneum. No relation was found between hypotension and the delivery of the inhalation bolus. Concerning potentially harmful hypertension (defined as an episode of MAP above 135 mm Hg for longer than 2.5 min14), we found one episode in the sevoflurane group and none in the desflurane group. During the episode of severe hypertension, the patient received an inhalation bolus of sevoflurane and the target concentration of remifentanil was increased, resulting in MAP reduction to a safe range within 4 min.



View larger version (43K):
[in this window]
[in a new window]
 
Fig 3 Individual BIS (A) and haemodynamic profiles for all patients in the sevoflurane group. Heart rate (HR) is shown in (B); systolic (SYS) and mean (MAP) arterial pressure are shown in (C) and (D), respectively.

 


View larger version (37K):
[in this window]
[in a new window]
 
Fig 4 Individual BIS (A) and haemodynamic profiles for all patients in the desflurane group. Heart rate (HR) is shown in (B); systolic (SYS) and mean (MAP) arterial pressure are shown in (C) and (D), respectively.

 

View this table:
[in this window]
[in a new window]
 
Table 3 Haemodynamic and hypnotic stability during maintenance of anaesthesia. Values are mean (SD) (range). *P<0.05 between groups (Mann–Whitney test)
 
The individual BIS profiles are plotted in Figures 3 and 4 for the sevoflurane and desflurane groups, respectively. Overall, during maintenance, more periods with low BIS, defined as a BIS of less than 40, were found in the desflurane group than the sevoflurane group (Table 3). To assess whether these differences in BIS could be attributed to specific moments within the maintenance phase, a time-synchronized analysis of the differences in BIS between groups was plotted (Fig. 5). The specific periods at which the BIS differed significantly between the two groups (upper-limit CI 95% below 0) were found during the first part of the maintenance phase. Patients in the desflurane group tended to reach the target BIS of 50 more quickly than those in the sevoflurane group (median (min–max)): sevoflurane, 200 s after loss of consciousness (100–1160) vs desflurane, 180 s after loss of consciousness (40–1120)). However, at the end of surgery the BIS were 48 (5) in sevoflurane group and 51 (12) in desflurane group, with no statistical significance between the groups.



View larger version (39K):
[in this window]
[in a new window]
 
Fig 5 Time-synchronized (beginning at loss of consciousness) analysis of the differences in BIS between groups. The mean is plotted as a continuous line; the upper and lower 95% CI are plotted as dotted lines.

 
As the ultimate aim of this study was to investigate the hypnotic and haemodynamic characteristics of the inhalation bolus in both groups, multiple analyses were done. In the sevoflurane group, 41 boluses were administered. Of these boluses, nine of 41 were associated with a simultaneous increase in the target concentration of remifentanil. In the desflurane group, 34 boluses were administered; of these, 11 of 34 were associated with an increase in the target concentration of remifentanl. There were no statistically significant differences between groups in that respect (Fisher’s exact test). As shown in Figure 2, a similar, randomized temporal distribution of the inhalation boluses was observed in both groups. All boluses were considered effective in controlling BIS over time.

As plotted in Figure 1 and shown in Table 4, we studied the pharmacodynamic behaviour of the BIS during the bolus periods. Similar control of the BIS was achieved in both groups without a statistically significant difference in time, defined as time A and B. In contrast, a lower minimum BIS was found in the desflurane group. Figure 6 shows the changes in MAP between the baseline (at the moment of the start of the inhalation bolus) and the lowest or highest value within 8 min after the start. Of the 41 boluses administered in the sevoflurane group, 10 were associated with an increase in MAP (18 (14) mm Hg; range 1–38 mm Hg) and 29 resulted in a decrease in MAP (11 (9) mm Hg; range 1–32 mm Hg). For two boluses, the MAP did not change. Of the 34 boluses administered in desflurane group, 10 resulted in an increase in MAP (25 (9) mm Hg; range 8–36 mm Hg) and 23 in a decrease (12 (10) mm Hg; range 1–43 mm Hg). For one bolus, MAP did not change. After the bolus, a MAP of less than 50 was found three times in sevoflurane group and once in desflurane group; among these, there was no recorded episode of MAP lower than 40. No differences in episodes of increasing MAP (more than 15 mm Hg higher than the baseline) were found between the two groups. The overall incidence of episodes of bradycardia (heart rate less than 50 bpm) and tachycardia (higher than 90 bpm) was low and similar between the groups.


View this table:
[in this window]
[in a new window]
 
Table 4 Bolus pharmacodynamic characteristics. Values are mean (SD) (range). *P<0.05 between groups (Mann–Whitney test)
 


View larger version (23K):
[in this window]
[in a new window]
 
Fig 6 Changes in MAP (mean arterial pressure) for all patients after an inhalation bolus of volatile anaesthetics for S and D, sevoflurane and desflurane, respectively. The baseline measure was taken at the start of the inhalation bolus and marked as ‘Baseline’ and the highest or lowest measure within 8 min after the start of the inhalation bolus is marked as ‘Trend’.

 
The immediate recovery variables measured at the end of the procedure are presented in Table 2. The mean times until eye opening, extubation, orientation, and free airway were significantly shorter when using desflurane than when using sevoflurane. The time until recovery of spontaneous breathing was similar in both groups. Figure 7 shows the percentage of patients within a group that had not yet met the endpoint at a particular time point for the five endpoints of recovery: with the exception of recovery of spontaneous breathing, those in the sevoflurane group took longer to respond than those in the desflurane group.



View larger version (23K):
[in this window]
[in a new window]
 
Fig 7 Survival distribution function for different recovery measures. Sevoflurane group (sevoflurane) data plotted as dotted lines and desflurane group data as straight lines.

 

    Discussion
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 References
 
An inhalation bolus can be used to optimize inhaled drug administration, leading to improved control over depth of anaesthesia without causing haemodynamic instability or prolonged recovery times, even in morbidly obese patients. We compared the inhalation agents desflurane and sevoflurane, using BIS guidance and inhalation boluses to gauge the depth of anaesthesia, in combination with titration of remifentanil. We used a plasma compartment-controlled TCI system to facilitate remifentanil administration15 and to obtain a continuous estimate of the calculated plasma and effect-site concentrations.16 Instead of using MAC equivalent doses of sevoflurane or desflurane, we used the BIS control as an endpoint. Although the MAC is useful in comparing the relative potency of volatile anaesthetics, multiple confounding factors can affect the MAC of individual patients. In clinical practice, the use of MAC as a guide to titrate volatile anaesthetics can result in either under- or over-dosing.7 Further, end-tidal anaesthetic monitoring does not improve intraoperative haemodynamic stability or decrease emergence times from general anaesthesia.17

Intraoperative stability can be defined as the maintenance of an adequate level of hypnosis/anaesthesia without either awareness or excessive anaesthesia, in combination with adequate haemodynamic and respiratory stability. We used BIS monitoring to measure and guide the administration of the inhaled anaesthetics in order to maintain the desired hypnotic/anaesthetic effect. The BIS is a quantifiable measure of the sedative and hypnotic effects of inhaled anaesthetics.7 The delay between a change in the vaporizer setting and the resulting change in hypnotic/anaesthetic effect was minimized by using an inhalation bolus of volatile anaesthetics. Matute and colleagues1 defined this as the dynamic use of the vaporizer and fresh gas flow to control the haemodynamic responses to stress caused during surgery. In contrast, we used the inhalation bolus to control the hypnotic/anaesthetic component of anaesthesia as measured by the BIS. If the BIS became too low, the vaporizer setting was decreased by 25%.

It is well documented that the range of BIS for adequate hypnotic levels of anaesthesia is between 40 and 60.13 Therefore, our statistical analysis focused on this range. In our methods however we aimed at maintaining BIS at 50 with decision thresholds of 55 as the upper limit and 45 as lower limit. There is an inherent time delay of 15 s between EEG changes and display of the BIS calculation increasing the risk of awareness if the upper decision threshold is set at a BIS value of 60.

As shown in Figures 3 and 4, the individual BIS profiles were found to be within an adequate range, although there were differences between the two groups. In the desflurane group, more periods of too low a BIS were observed (Table 3), making the overall hypnotic control for sevoflurane better than for desflurane. However, this difference in control might be attributable to the pharmacokinetic differences between sevoflurane and desflurane at the beginning of anaesthesia. As seen in Figure 5, desflurane produced a lower BIS significantly faster than did sevoflurane at a similar moment in the initial part of the process. This difference was probably caused by the initial setting of 1 MAC for both volatile anaesthetics for the first 5 min (using a high fresh gas flow) after propofol induction, launching the patient into a more profound hypnotic trajectory with the volatile anaesthetic (desflurane) of faster onset.18 The initial settings were derived from the original work of Baum and colleagues19 on low-flow anaesthesia, which provided us with the opportunity of having both anaesthetics ‘on-line’ after 5 min and made it easier to begin titration control with BIS from that point on. After this initial overshoot, similar BIS profiles were observed in both groups. As seen in Table 3, similarly high levels of overall control were found in both groups in preventing the BIS from exceeding the limit of 60.

When looking at the inhalation boluses in detail (Fig. 1; Table 4), we found differences in the pharmacodynamic behaviour of the BIS during the bolus periods for both groups. No statistically significant difference in time characteristics, defined as times A and B, was found between these groups, possibly because of large individual variability. In contrast, a significant lower minimal BIS was found in the desflurane group, not leading to a prolonged time B, because of the faster kinetics of desflurane. This overshoot, resulting in a lower minimal BIS, might be prevented by using a shorter bolus time for desflurane, but because of a lack of published information, we used a bolus time of 30 s for both agents.

Intraoperative haemodynamic stability was compared in both groups. Overall, better stability was found in the desflurane group, resulting in fewer and shorter episodes of hypotension. One could argue that the use of the inhalation bolus might result in episodes of severe hypotension, but no hypotensive episodes could be attributed to the bolus technique. Detailed observation of the hypotensive episodes revealed a closer correlation with two specific interventions, the semi-recumbent positioning of the patient after induction and the release of the pneumoperitoneum, both of which are known to cause hypotension.20 The overall better stability in the desflurane group might be attributed to sympathetic activation when using desflurane in rapidly increasing concentrations greater than 6 vol% end-tidal.21 22 As shown in Figure 2D, we reached these concentrations during the inhalation bolus phase. However, no differences in hypertension or tachycardia were observed between the groups. This finding could be a result of the use of remifentanil, which can blunt this sympathetic activation by lowering the MAC-BAR, as described for other opiates.2328 Our findings may represent a combination of sympathetic stimulation and simultaneous blunting of the sympathetic response,22 29 resulting in an overall better haemodynamic profile when using desflurane than sevoflurane. Further research is needed to clarify this phenomenon. Also, no harmful hypertension or tachycardia could be attributed to the use of desflurane.

At the end of surgery, the BIS did not differ between the groups, allowing similar baseline conditions for studying recovery profiles. Several immediate recovery variables were recorded and compared between groups (Table 2). Except for the return of spontaneous breathing, all recovery variables were significantly shorter when using desflurane than when using sevoflurane. Also, more patients in the desflurane group reached recovery endpoints significantly faster as plotted in Figure 7 using Kaplan–Meier survival curves. These findings are in agreement with those of Juvin and colleagues,9 who compared postoperative recovery after desflurane, propofol, or isoflurane anaesthesia in morbidly obese patients. They concluded that immediate recovery was significantly quicker after BIS-guided desflurane anaesthesia in combination with a constant target plasma concentration of 50 ng ml–1 of alfentanil using TCI technology. Similar results for immediate recovery were also found when comparing desflurane and sevoflurane anaesthesia in non-obese patients undergoing laparoscopic tubal ligation.30 Morbidly obese patients are at high risk of both aspiration and upper-airway obstruction after tracheal extubation.31 Therefore, rapid recovery is desirable to ensure early efficient coughing and to decrease the rate of postoperative respiratory complications.9

In studying the concept of an inhalation bolus of sevoflurane or desflurane used to optimize inhaled drug administration, we found that immediate recovery was significantly faster in the desflurane group even though intraoperative hypnotic control as measured by the BIS was less accurate in the early maintenance phase. Intraoperative haemodynamic control was found to be better when using desflurane. Overall, fewer episodes of hypotension were found in the desflurane group. Although detailed analysis of the inhalation bolus revealed a larger overshoot when using desflurane, the application of the inhalation bolus was found to be appropriate in both groups, and did not cause severe haemodynamic side-effects.


    Acknowledgements
 
The authors thank Mrs Dorine Dyzers, CN (Department of Anaesthesia, University Hospital Ghent, Gent, Belgium) and Geert Byttebier and Mieke De Bosschere (General Biometrics Consulting and Services, Gent, Belgium) for their help during the preparation of the manuscript. Support for this study was partially funded by a non-restrictive educational grant from Baxter World Trade (Brussels, Belgium) and partially by departmental and institution funding.


    References
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 References
 
1 Matute E, Alsina E, Roses R, Blanc G, Perez-Hernandez C, Gilsanz F. An inhalation bolus of sevoflurane versus an intravenous bolus of remifentanil for controlling hemodynamic responses to surgical stress during major surgery: a prospective randomized trial. Anesth Analg 2002; 94: 1217–22[Abstract/Free Full Text]

2 Eger EI. Uptake and distribution. In: Miller RD, ed. Anesthesia. Philadelphia: Churchill Livingstone, 2000; 74–95

3 Carpenter RL, Eger EI, 2nd, Johnson BH, Unadkat JD, Sheiner LB. Pharmacokinetics of inhaled anesthetics in humans: measurements during and after the simultaneous administration of enflurane, halothane, isoflurane, methoxyflurane, and nitrous oxide. Anesth Analg 1986; 65: 575–82[Abstract]

4 Karzai W, Haberstroh J, Muller W, Priebe HJ. Rapid increase in inspired desflurane concentration does not elicit a hyperdynamic circulatory response in the pig. Lab Anim 1997; 31: 279–82[ISI][Medline]

5 Munoz HR, Gonzalez JA, Concha MR, Palma MA. Hemodynamic response to tracheal intubation after vital capacity rapid inhalation induction (VCRII) with different concentrations of sevoflurane. J Clin Anesth 1999; 11: 567–71[CrossRef][ISI][Medline]

6 Struys M, Mortier E. Target-controlled administration of inhaled anaesthesics. Best Pract Res Clin Anaesthesiol 2001; 15: 35–50[CrossRef]

7 Song D, Joshi GP, White PF. Titration of volatile anesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology 1997; 87: 842–8[ISI][Medline]

8 Daniel M, Larson MD, Eger EI, 2nd, Noorani M, Weiskopf RB. Fentanyl, clonidine, and repeated increases in desflurane concentration, but not nitrous oxide or esmolol, block the transient mydriasis caused by rapid increases in desflurane concentration. Anesth Analg 1995; 81: 372–8[Abstract]

9 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: 714–9[Abstract/Free Full Text]

10 Adams JP, Murphy PG. Obesity in anaesthesia and intensive care. Br J Anaesth 2000; 85: 91–108[Free Full Text]

11 Minto CF, Schnider TW, Egan TD, et al. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology 1997; 86: 10–23[ISI][Medline]

12 Egan TD, Huizinga B, Gupta SK, et al. Remifentanil pharmacokinetics in obese versus lean patients. Anesthesiology 1998; 89: 562–73[ISI][Medline]

13 Struys MM, De Smet T, Versichelen LF, Van De Velde S, Van den Broecke R, Mortier EP. Comparison of closed-loop controlled administration of propofol using Bispectral Index as the controlled variable versus ‘standard practice’ controlled administration. Anesthesiology 2001; 95: 6–17[ISI][Medline]

14 Joint National Committee on the detection eathbp. The sixth report of the joint National Committee on detection, evaluation and treatment of high blood pressure. Arch Intern Med 1997; 157: 2413–2446[Abstract]

15 Milne SE, Kenny GN. Future applications for TCI systems. Anaesthesia 1998; 53 (Suppl. 1): 56–60[ISI][Medline]

16 Hoymork SC, Raeder J, Grimsmo B, Steen PA. Bispectral index, predicted and measured drug levels of target-controlled infusions of remifentanil and propofol during laparoscopic cholecystectomy and emergence. Acta Anaesthesiol Scand 2000; 44: 1138–44[CrossRef][ISI][Medline]

17 Wang J, Liu J, White PF, Klein KW, Browne RH. Effects of end-tidal gas monitoring and flow rates on hemodynamic stability and recovery profiles [see comments]. Anesth Analg 1994; 79: 538–44[Abstract]

18 Rehberg B, Bouillon T, Zinserling J, Hoeft A. Comparative pharmacodynamic modeling of the electroencephalography-slowing effect of isoflurane, sevoflurane, and desflurane. Anesthesiology 1999; 91: 397–405[ISI][Medline]

19 Baum JA. Control of inhalational anaesthesia. In: Baum JA, ed. Low Flow Anaesthesia. Oxford: Butterworth-Heinemann, 2001; 73–87

20 Joris J. Anesthesia for laparoscopic surgery. In: Miller R, ed. Anesthesia. New York: Churchill Livingstone, 2000; 2003–23

21 Ebert TJ, Muzi M. Sympathetic activation with desflurane in humans. Adv Pharmacol 1994; 31: 369–78[Medline]

22 Weiskopf RB, Moore MA, Eger EI 2nd, et al. Rapid increase in desflurane concentration is associated with greater transient cardiovascular stimulation than with rapid increase in isoflurane concentration in humans. Anesthesiology 1994; 80: 1035–45[ISI][Medline]

23 Pacentine GG, Muzi M, Ebert TJ. Effects of fentanyl on sympathetic activation associated with the administration of desflurane. Anesthesiology 1995; 82: 823–31[ISI][Medline]

24 Yonker-Sell AE, Muzi M, Hope WG, Ebert TJ. Alfentanil modifies the neurocirculatory responses to desflurane. Anesth Analg 1996; 82: 162–6[Abstract]

25 Tonner PH, Scholz J, Krause T, Paris A, von Knobelsdorff G, Schulte an Esch J. Administration of sufentanil and nitrous oxide blunts cardiovascular effects of desflurane but does not prevent an increase in middle cerebral artery blood flow velocity. Eur J Anaesthesiol 1997; 14: 389–96[ISI][Medline]

26 Daniel M, Weiskopf RB, Noorani M, Eger EI, 2nd. Fentanyl augments the blockade of the sympathetic response to incision (MAC-BAR) produced by desflurane and isoflurane: desflurane and isoflurane MAC-BAR without and with fentanyl. Anesthesiology 1998; 88: 43–9[ISI][Medline]

27 Avramov MN, Griffin JD, White PF. The effect of fresh gas flow and anesthetic technique on the ability to control acute hemodynamic responses during surgery. Anesth Analg 1998; 87: 666–70[Abstract]

28 Katoh T, Kobayashi S, Suzuki A, Iwamoto T, Bito H, Ikeda K. The effect of fentanyl on sevoflurane requirements for somatic and sympathetic responses to surgical incision. Anesthesiology 1999; 90: 398–405[ISI][Medline]

29 Weiskopf RB, Eger EI, 2nd, Noorani M, Daniel M. Repetitive rapid increases in desflurane concentration blunt transient cardiovascular stimulation in humans. Anesthesiology 1994; 81: 843–9[ISI][Medline]

30 Song D, Joshi GP, White PF. Fast-track eligibility after ambulatory anesthesia: a comparison of desflurane, sevoflurane, and propofol. Anesth Analg 1998; 86: 267–73[Abstract]

31 Rose DK, Cohen MM, Wigglesworth DF, DeBoer DP. Critical respiratory events in the postanesthesia care unit. Patient, surgical, and anesthetic factors. Anesthesiology 1994; 81: 410–18[ISI][Medline]