Faster wash-out and recovery for desflurane vs sevoflurane in morbidly obese patients when no premedication is used

Br J Anaesth 2007;99:353-358

La Colla L, Albertin A, La Colla G, Mangano A



Purpose            The purpose of this study was to compare the kinetics and clinical end points of desflurane and sevoflurane in morbidly obese patients during wash-in and emergence.

Background            Morbidly obese individuals have a relatively lower proportion of lean body mass. Blood flow to fat tissue is lower in obese compared to non-obese individuals. Obesity alters cardiopulmonary dynamics which plays a role in the uptake and elimination of inhalation agents. The kinetics of many drugs, including potent inhalation agents, differs in morbidly obese and non-obese individuals. In particular, as total body weight and the duration of anesthetic administration increase both the blood : gas and oil : gas partition coefficients have a larger influence on the rate of emergence from anesthesia. Desflurane’s oil : gas partition coefficient is about 64% lower than that of sevoflurane. Consequently, desflurane has a smaller apparent volume of distribution than sevoflurane.

Methodology            This prospective, randomized, double-blind study included 28 ASA class II and III patients scheduled for elective open bilio-intestinal bypass surgery. Exclusion criteria included ASA physical status class > III, age < 20 years or > 65 years, a history of alcohol or other drug abuse, and myocardial dysfunction.

To avoid bias in emergence times, no patients received any premedication. All patients were intubated with a fiberoptic bronchoscope during a pharmacokinetic model-driven target controlled remifentanil infusion. General anesthesia was then induced with propofol 2 mg/kg while the remifentanil infusion continued. Patients were mechanically ventilated with 50% oxygen in air via a non-rebreathing circuit. Cis-atracurium was used for neuromuscular blockade. The anesthetic vaporizer was then turned directly to either 2% sevoflurane or 6% desflurane according to randomization. These concentrations were maintained for a 30 minute data collection period before surgery was begun. During surgery, the remifentanil infusion was titrated to keep heart rate and blood pressure within 10% of baseline values. Sevoflurane 1% to 2% or desflurane 3% to 4% was titrated to achieve a Bispectral Index (BIS) value of 40 to 50.

During wound closure all patients received morphine 0.05 mg/kg and ketorolac 30 mg Editors Note: the route of administration was not specified. During the last 20 minutes of surgery, inhalation agent concentration was decreased to achieve a BIS value of 60. After the last suture end tidal agent concentration was recorded and the vaporizers were turned off. End tidal agent concentration was then recorded every 30 seconds for 5 minutes. Mechanical ventilation was continued until the first spontaneous breath.

Result            There were no statistically significant differences in patient demographics between groups. The average age was 27.1 (sd 212.9) years. Average body Mass Index was 50.6 (sd 5.4); 48 in the sevoflurane group and 53 in the desflurane group (P=0.180). The duration of anesthesia was 180 minutes in the sevoflurane group and 193 minutes in the desflurane group (P=0.232).

The wash out of desflurane was significantly faster than sevoflurane at all time points (P<0.01). At five minutes the end tidal concentration of desflurane was abut 10% of what it was when the vaporizer was turned off. The sevoflurane concentration was about 20% of what is was when the vaporizer was turned off. Desflurane patients squeezed the investigator’s hand in 8 minutes vs. 15.8 minutes for sevoflurane patients. Desflurane patients were extubated in 9.4 minutes vs. 16.4 minutes for sevoflurane patients. Desflurane patients were discharged from recovery in 16.3 minutes vs. 27.0 minutes for sevoflurane patients. These and other clinical end points were all significant at a P<0.001.

Conclusion            Morbidly obese patients recover more quickly following desflurane than sevoflurane anesthesia following an approximately three hour anesthetic.



The blood : gas solubilities of desflurane and sevoflurane are relatively similar (desflurane 0.45, sevoflurane 0.65) and much lower than the agents we used before they became available (isoflurane 1.4). Although fat absorbs inhalation agents slowly, each agent’s solubility in fat largely determines how quickly it can be eliminated from fat. Desflurane’s fat : blood partition coefficient is 29. Sevoflurane’s fat : blood coefficient is 52, similar to that of isoflurane (50) and halothane (57).

This study tries to answer the question, “does the lower fat solubility allow morbidly obese patients to wake up more quickly after a desflurane anesthetic than a sevoflurane anesthetic?” It does a pretty good job of answering the question (“yes”). Much as in lean patients, once the vaporizer is turned off and the fresh gas flow increased to prevent rebreathing of inhalation agent, the fraction of sevoflurane exhaled is about twice as great at any given time, compared to the concentration just before the vaporizer was turned off, as the fraction of desflurane. Whether or not the faster wake up is clinically significant will depend upon each anesthetist’s practice environment and anesthetic technique. If you can take advantage of being able to extubated morbidly obese patients 7 minutes faster and get them out of recovery 11 minutes faster following a three hour case the difference will be clinically significant. Also, keep in mind that relatively low concentrations of inhalation agent were used during the maintenance phase of the case.

I do want to suggest one way in which these differences in recovery time might be an advantage for all our obese patients. A previous study has shown that two minutes after responding to command all desflurane patients had a complete return of airway reflexes and could swallow water without coughing or drooling.1 At that the same time, 55% of sevoflurane patients did not have a complete return of airway reflexes and 18% still did not at six minutes. If this delay in recovery of airway reflexes was related to the remaining concentration of agent, the faster elimination of desflurane may offer a wider margin of safety for airway complications during early recovery.

There were several aspects of this study design that may have resulted in biased results. The maintenance concentrations of desflurane and sevoflurane allowed (3% - 4% and 1% - 2% respectively) represent a lower MAC fraction of desflurane than sevoflurane. If a lower MAC equivalent of desflurane was administered the results would have been biased in favor of desflurane. The actual concentration of inhalation agent delivered during the maintenance phase of the cases was not reported so there is no way to know if the study results were biased. The target controlled remifentanil infusion was also titrated during the case. The actual rates of remifentanil infusion during maintenance were not reported. If one group received more remifentanil than the other it is likely that less inhalation agent would have been administered in that group. The study results would therefore have been biased in favor of the agent used in that group.


1. McKay RE, Large MJ, Balea MC, McKay WR. Airway reflexes return more rapidly after desflurane anesthesia than after sevoflurane anesthesia. Anesth Analg 2005;100:697-700.


Michael Fiedler, PhD, CRNA