Postoperative Recovery After Desflurane, Propofol, or Isoflurane Anesthesia Among Morbidly Obese Patients: A Prospective, Randomized Study

Abstract

Recovery from anesthesia might be compromised in obese patients. Because of its pharmacological properties, desflurane might allow rapid postoperative recovery for these patients. We compared postoperative recovery for 36 obese patients randomized to receive either desflurane, propofol, or isoflurane to maintain anesthesia during laparoscopic gastroplasties. Anesthesia was induced with propofol and succinylcholine IV and was maintained with rocuronium, alfentanil, inhaled nitrous oxide, and the study drug. Immediate recovery (i.e., times from the discontinuation of anesthesia to tracheal extubation, eye opening, and the ability to state one’s name) was measured. At the time of postanesthesia care unit (PACU) admission, arterial saturation and the ability of patients to move were recorded. In the PACU, intermediate recovery was measured by using sedation and psychometric evaluations, 30, 60, and 120 min postoperatively. Data were compared between groups by using the Kruskal-Wallis and χ² tests. Results were reported as means ± sd. P < 0.05, compared with desflurane, was considered significant. Immediate recovery occurred faster, and was more consistent, after desflurane than after propofol or isoflurane (times to extubation were 6 ± 1 min, 13 ± 8 min [P < 0.05, compared with desflurane], and 12 ± 6 min [P < 0.05, compared with desflurane], respectively). At PACU admission, Spo₂ values were significantly higher and patient mobility was significantly better after desflurane than after isoflurane or propofol. Sedation was significantly less pronounced with desflurane at 30 and 120 min postoperatively. In morbidly obese patients, postoperative immediate and intermediate recoveries are more rapid after desflurane than after propofol or isoflurane anesthesia. This advantage of desflurane persists at least for 2 h after surgery and is associated with both an improvement in patient mobility and a reduced incidence of postoperative desaturation.

Implications 

In morbidly obese patients, postoperative immediate and intermediate recoveries are more rapid and consistent after desflurane than after propofol or isoflurane anesthesia.

The choice of an anesthetic for general anesthesia maintenance in morbidly obese patients has not yet been the focus of in-depth studies and remains controversial (^1,2). Obese patients are at high risk of both aspiration and acute upper airway obstruction after tracheal extubation (³). Rapid recovery is therefore desirable, to ensure early efficient coughing and to decrease the rate of postoperative respiratory complications. Because of its pharmacological properties, desflurane may yield a rapid recovery in obese patients. However, this theoretical advantage has not been demonstrated. The aim of this study was to compare the characteristics of postoperative recovery among morbidly obese patients who had received either desflurane, isoflurane, or propofol for anesthesia maintenance.

Methods

With institutional review board agreement and informed consent, 36 morbidly obese patients (body mass index > 35 kg/m²) scheduled for laparoscopic gastroplasty were included in this randomized, prospective study. All patients were operated on by the same surgeon, who used the adjustable gastric banding technique (⁴). Exclusion criteria consisted of a preoperative hematocrit value of <25%, significant coronary artery disease, β-blocker treatment, chronic alcohol or drug abuse, renal failure (serum creatinine > 120 μmol/L), hepatic dysfunction (aspartate aminotransferase or alanine aminotransferase > 1.5 N), a personal history of allergy to an anesthetic, or a personal or family history of malignant hyperthermia. Criteria for premature study withdrawal were the failure of surgery to proceed as planned or the development of complications hindering the assessment of study variables. In cases of withdrawal, postoperative data collected up to the time of discontinuation of the study were retained for analysis.

On the day before surgery, the patients underwent a baseline psychometric evaluation by using the Mini-Mental State examination (temporospatial orientation, anterograde and retrograde memory, attention, word fluency, and concentration) (⁵). Premedication was with oral hydroxyzine 1 mg/kg and oral cimetidine 800 mg, administered 1 h before the induction of anesthesia. Standard monitoring was used, including bispectral index (BIS) recording (Aspect Medical Systems) via two bipolar electroencephalographic leads (FpZ-At1 and FpZ-At2).

After adequate denitrogenation (⁶), anesthesia was induced using a propofol target-controlled infusion (TCI) (Diprifusor^®; Astrazeneca, Rueil-Malmaison, France) to reach an initial blood concentration of 8 μg/mL, which was modified immediately after tracheal intubation as described below. After loss of consciousness was obtained, succinylcholine 1.2 mg/kg was given IV to facilitate orotracheal intubation. After tracheal intubation, ventilation with 50% nitrous oxide in oxygen was delivered via a closed system, with a fresh gas flow of 1 L/min, and was controlled to achieve an end-tidal carbon dioxide pressure of 30–35 mm Hg. Patients were then randomly allocated to receive either desflurane (12 patients), propofol (12 patients), or isoflurane (12 patients). The propofol TCI used for induction was then stopped in both the desflurane and isoflurane groups and continued in the propofol group. Desflurane, propofol, and isoflurane were titrated continuously to maintain the BIS between 45 and 55 throughout the surgical procedure. Immediately after tracheal intubation, patients were given a 50-mg bolus of rocuronium and an infusion of alfentanil. Additional rocuronium boluses were used to keep the train-of-four below 2 throughout the surgical procedure. The alfentanil infusion was delivered via a computer-controlled pump by using a program based on pharmacokinetic data reported by Maître et al. (⁷). A constant plasma target level of alfentanil (50 ng/mL) was maintained from tracheal intubation to gastric banding fixation (⁴).

Vital signs were recorded before the induction of anesthesia. They were also recorded, as were alfentanil and propofol plasma concentrations or volatile end-tidal concentrations (Capnomac Ultima^®; Datex, Helsinki, Finland), every 2 min until skin incision, every 1 min for 5 min after skin incision, and then every 15 min until the dressing was completed. BIS values were recorded continuously from the preoxygenation period until transfer to the postanesthesia care unit (PACU).

After gastric banding fixation (approximately 45 min before the dressing), the alfentanil target concentration was set at zero, and propacetamol 2 g IV and nefopam 20 mg IV were administered. When the dressing was completed, the administration of anesthetics and nitrous oxide was discontinued. The times from study drug discontinuation to eye opening, extubation, and orientation (giving one’s name on request) were recorded. Neostigmine 3 mg IV and atropine 1 mg IV were given systemically before tracheal extubation, 3 to 5 min after discontinuation of the anesthetic drugs.

After extubation, the ability of the patients to move themselves was evaluated during transfer from the operating table to the bed (0 = needs help, no movement; 1 = needs help, moves only head; 2 = needs help, moves head and one leg; 3 = needs help, moves head and both legs; 4 = does not need help, able to move alone). Patients were then placed in a semirecumbent position and given oxygen 3 L/min, via a face mask, during transfer to the PACU. Vital signs were recorded at the time of PACU admission. Intermediate recovery was assessed 30, 60, and 120 min after extubation. This assessment included psychometric evaluation (Mini-Mental State examination), sedation scoring using a modified Observer’s Assessment of Alertness/Sedation Scale (0 = asleep, not arousable; 1 = asleep but arousable; 2 = drowsy; 3 = awake but calm; 4 = awake, very aware) (⁸), pain scoring (10-cm visual analog scale, with 0 = no pain and 10 = worst pain), and recording of any nausea or vomiting. Postoperative analgesic requirements and the time to eligibility for PACU discharge, as determined by using the scoring system described by Aldrete and Kroulik (⁹), were also recorded. Postoperative recovery (i.e., from PACU admission to PACU discharge) was assessed by a single investigator blinded to the patient groups, whereas the single anesthesiologist who performed anesthesia (from denitrogenation to transfer to the PACU) was not blinded. Laboratory values, including serum electrolyte, hematocrit, creatinine, and liver function levels, were measured preoperatively and within 24 h after surgery.

Before study initiation, the required sample size was calculated. Because elderly patients share a number of similarities with obese patients, including an increased proportion of fat in the total body weight, we anticipated that the time from anesthesia discontinuation to eye opening for our obese patients would be at least equal to that reported for elderly patients (¹⁰). On the basis of our previous observations in elderly patients, we calculated the sample size such that times to eye opening of 5.6 ± 3.4 min (mean ± sd) and 11.9 min after desflurane and propofol, respectively, would produce a Type 1 error rate of one-tailed α = 0.05 and, under the alternate hypothesis, would retain the null hypothesis with a Type 2 error of β = 0.05 (yielding a power of 0.95). Therefore, we estimated that a sample size of nine patients per group would be required. To make allowances for the possibility that unpredictable events such as surgical complications might lead to the premature withdrawal of some patients, we included 12 patients in each group. Data were reported as mean ± sd or as median and range. Comparisons among groups were performed by using the Kruskal-Wallis and χ² tests. P values of <0.05 were considered significant.

Results

The data for two patients were excluded from the analysis. One of these patients was in the propofol group and developed an anaphylactic reaction immediately after the succinylcholine injection. The anesthesia was stopped, and the surgical procedure was not performed. No data for this patient were used. The second patient was in the isoflurane group. Just before extubation, but after eye opening, the patient developed bronchospasm requiring sedation, mechanical ventilation, and transfer to the intensive care unit. Postoperative assessments were therefore discontinued. For this patient, only the data collected up to the time of discontinuation (i.e., to eye opening) were used. Therefore, complete data were available for 12 desflurane-treated patients, 11 propofol-treated patients, and 11 isoflurane-treated patients. The three groups were comparable with respect to age, sex, weight, body mass index, physical status, time from premedication to the induction of anesthesia, anesthesia duration, and doses of rocuronium and alfentanil (Table 1). The doses of propofol used in the three groups are noted in Table 1. The BIS values throughout anesthesia were similar among the groups. The BIS values and alfentanil plasma concentrations when anesthesia was discontinued and at extubation were similar among the groups (Table 1). At the end of anesthesia, the mean propofol blood concentration was 2.5 ± 0.96 μg/mL, and the mean end-tidal concentrations of desflurane and isoflurane were 2.56% ± 1.37% and 0.53% ± 0.21%, respectively. There were no conversions to open surgery or other surgical complications.

Table 1: Description of Experimental Groups and Peroperative Data

The times from the end of administration of the study drug to extubation, eye opening, and orientation were significantly shorter in the desflurane group than in the propofol and isoflurane groups (Table 2). The ability of the patients to move was also significantly better with desflurane than with the other two medications; the postoperative mobility scores were 3 (range 2–4), 1 (range 0–3) (P < 0.05, compared with desflurane), and 1 (range 0–3) (P < 0.05, compared with desflurane) after desflurane, isoflurane, and propofol, respectively. At PACU admission, none of the desflurane-treated patients, compared with five of the propofol-treated patients and five of the isoflurane-treated patients, had Spo₂ values of <95% (P < 0.05, compared with the desflurane group). The median (minimum–maximum) values of Spo₂ were 97.5% (95%–99%), 95.5% (86%–98%), and 96% (84%–99%) after desflurane, isoflurane, and propofol, respectively. In the PACU, sedation was less pronounced after desflurane than after isoflurane or propofol, 30 and 120 min after surgery (Figure 1). The results of postoperative psychometric evaluations (Figure 2) were similar in the three groups. The times to eligibility for PACU discharge (126 ± 56, 180 ± 72, and 198 ± 109 min after desflurane, isoflurane, and propofol, respectively) showed a trend toward shorter PACU stays for the desflurane group, compared with the two other groups, despite a lack of statistical significance.

Table 2: Early Recovery Parameters, PONV, and Analgesic Requirements in the PACU

Figure 1: Postoperative sedation levels at 30, 60, and 120 min postoperatively. The postoperative sedation level (as assessed by using the Observer’s Assessment of Alertness/Sedation Scale [OAAS]) was significantly less pronounced after desflurane (○) than after isoflurane (▪) or propofol (▴) at 30 and 120 min postoperatively. Individual values are noted at each of the study time points. *P < 0.05, when the isoflurane and propofol groups are compared with the desflurane group.

Figure 2: Mini-Mental State examination results at 30, 60, and 120 min postoperatively. No statistically significant differences among the desflurane (□), isoflurane (▨), and propofol (▪) treatment groups were noted. Values are given as mean ± sd.

At none of the time points were there any differences among the three groups with respect to pain scores (Figure 3), PACU analgesic requirements, or postoperative nausea and vomiting (PONV) incidences (Table 2). The amounts of IV fluids given during and after surgery and the incidences of the use of vasoactive drugs to maintain hemodynamic stability during the procedure were similar for the three groups (Table 1).

Figure 3: Visual analog scores at 30, 60, and 120 min postoperatively. No statistically significant differences among the desflurane (□), isoflurane (▨), and propofol (▪) treatment groups were noted. Values are given as mean ± sd.

Discussion

This study demonstrates that, in morbidly obese patients, postoperative recovery occurs faster and is more constant after desflurane anesthesia than after isoflurane or propofol anesthesia. In agreement with previous studies (^10–16), faster early recovery was demonstrated after desflurane, compared with isoflurane, in the present study. However, the magnitude of the advantage of desflurane over isoflurane was often smaller in those previous studies than in the present one (^11–16). We also demonstrated that early recovery was faster and more predictable after desflurane than after propofol anesthesia, whereas previous studies, except one (¹⁰), found no difference in early recovery between these two drugs (^11,17–21). However, most of those studies were performed in lean patients anesthetized for short times (^17–21). The fact that our patients were obese and were anesthetized for more than two hours may explain our results. First, because of its low solubility, less desflurane needs to be released from the tissues and eliminated from the body at the end of prolonged anesthesia (²²). This pharmacological advantage was suggested by clinical studies, in which anesthesia duration seemed to have little influence on recovery time after desflurane anesthesia (²³). Second, propofol is a lipid-soluble anesthetic and may therefore have a prolonged effect in obese patients, whose proportion of fat in their total body weight is increased. Desflurane, in contrast, has very low solubility. This explanation remains controversial, because a pharmacokinetic study suggested that the propofol elimination half-life is not prolonged in obese patients, compared with lean subjects (²⁴). The only study in which early recovery was also faster after desflurane than after propofol and isoflurane was conducted in our institution and involved prolonged anesthesia in elderly patients, who share with obese patients a high proportion of fat in their total body weight (¹⁰).

Intermediate recovery, as assessed by sedation scores, was more rapid after desflurane than after propofol or isoflurane anesthesia in obese patients. Some previous studies performed with nonobese patients found no such advantage of desflurane (^17,19,20). An advantage of desflurane was sometimes demonstrated but was small and short-lived (^10,12,13,21). In the present study, the advantage of desflurane over propofol and isoflurane, in terms of sedation levels, was marked and lasted at least two hours.

In previous studies performed with nonobese patients, the clinical relevance of the advantages of desflurane over other anesthetic regimens was not demonstrated (²⁵). In the present study, the increase in the rapidity of early and intermediate recoveries with desflurane may be clinically significant. First, obese patients are at high risk of both aspiration and acute upper airway obstruction after tracheal extubation (³). Rapid early recovery may decrease these risks by improving active airway control by the patient and by permitting early efficient coughing. Consistent with this possibility, we found less hypoxemia at the time of PACU arrival for the desflurane group, compared with the propofol and isoflurane groups. Second, in contrast to previous studies, the improvement in intermediate recovery after desflurane was prolonged, being still perceptible two hours after surgery. This prolonged improvement might also have reduced the risk of postoperative respiratory complications. Third, the desflurane-treated patients were more successful in actively transferring from the operating table to the bed than were the patients in the other two groups. Despite the fact that the scores used in the present study have not been validated in the literature, this difference may also be of clinical relevance, because transferring from the operating table to the bed is often difficult, and it can require the help of several people or special lifting devices. More generally, our observations from this study suggest that obese patients anesthetized with desflurane can participate more effectively in their care (including physiotherapy), compared with those anesthetized with propofol or isoflurane. Lastly, the times to eligibility for PACU discharge showed a trend toward shorter PACU stays in the desflurane group, in agreement with a previous study performed with lean patients (²⁶). However, the difference was not statistically significant in the present study, perhaps because of the small sample sizes.

Our study was carefully designed to provide a valid comparison of the postoperative period after three different anesthetic regimens. Previous studies pursuing similar goals involved bolus opioid administration at undefined times, which might have resulted in interindividual differences in plasma opioid levels at the time of anesthesia discontinuation. These differences might have interfered with the assessment of recovery times. In the present study, the use of a TCI system to administer alfentanil ensured that alfentanil levels at anesthesia discontinuation and at extubation were similar for the three groups. Although such alfentanil concentrations were theoretical, it can be presumed that the potential effects of alfentanil on recovery times were similar in the three groups. Another problem with comparisons of postoperative recovery after different anesthetic regimens is the potential confounding influence of differences in the depth of anesthesia. To our knowledge, no previous studies comparing postoperative recovery after desflurane, propofol, and isoflurane have monitored the depth of anesthesia. In the present study, the depth of anesthesia was monitored continuously. The BIS values were similar between groups at the time of anesthesia discontinuation, establishing that the more rapid recovery in the desflurane group was not attributable to a lesser depth of anesthesia (²⁷). However, this study can be criticized because of the lack of investigator blinding in the assessments of early recovery status (i.e., until PACU admission). However, all patients were undergoing identical surgical procedures performed by the same anesthesiologist and the same surgeon, the use of BIS values and the TCI system might have minimized investigator bias, and early recovery was assessed only by using objective end points. In addition, the postoperative period from PACU admission to PACU discharge was assessed by a single investigator blinded to the patient groups.

The incidences of PONV were similar in the three groups. Propofol has intrinsic antiemetic properties, but the incidence of PONV after propofol or volatile anesthetic treatment remains controversial (^10,17–21). In the present study, the concurrent administration of alfentanil and nitrous oxide might have blunted any beneficial effect of propofol in that respect. In addition, the incidence of PONV was low and, under these conditions, demonstration of a clinical advantage of one anesthetic over another is difficult.

It is generally thought that the choice of anesthetic technique has little effect on significant or long-term outcomes (e.g., postoperative morbidity or economic costs). However, obese patients are at particular risk of early postoperative respiratory complications, so even slight improvements in early or intermediate recovery may be beneficial. In this respect, the more predictable and rapid recovery after desflurane demonstrated for our patients might have a significant beneficial effect on postoperative morbidity in the obese population. The clinical relevance of this early advantage of desflurane in obese patient is strengthened by its duration (at least two hours after surgery) and by the fact that it is associated with both a reduction in postoperative hypoxemia and an increase in patient mobility at the time of PACU admission.