Share this article on:

Metabolic and Hemodynamic Changes During Recovery and Tracheal Extubation in Neurosurgical Patients: Immediate Versus Delayed Recovery

Bruder, Nicolas MCH-PH*; Stordeur, Jean-Marc CCA*; Ravussin, Patrick MD, PU§; Valli, Marc MCU†; Dufour, Henri PHU‡; Bruguerolle, Bernard MD, PU†; Francois, Georges MD, PU*

doi: 10.1213/00000539-199909000-00027
Pediatric Anesthesia: Neurosurgical Anesthesia

Delayed recovery has been advocated to limit the postoperative stress linked to awakening from anesthesia, but data on this subject are lacking. In this study, we measured oxygen consumption (V̇O2) and plasma catecholamine concentrations as markers of postoperative stress. We tested the hypothesis that delayed recovery and extubation would attenuate metabolic changes after intracranial surgery. Thirty patients were included in a prospective, open study and were randomized into two groups. In Group I, the patients were tracheally extubated as soon as possible after surgery. In Group II, the patients were sedated with propofol for 2 h after surgery. V̇O2, catecholamine concentration, mean arterial pressure (MAP), and heart rate (HR) were measured during anesthesia, at extubation, and 30 min after extubation. V̇O2 and noradrenaline on extubation and mean V̇O2 during recovery were significantly higher in Group II than in Group I (V̇O2 for Group I: preextubation 215 ± 46 mL/min, recovery 198 ± 38 mL/min; for Group II: preextubation 320 ± 75 mL/min, recovery 268 ± 49 mL/min; noradrenaline on extubation for Group I: 207 ± 76 pg/mL, for Group II: 374 ± 236 pg/mL). Extubation induced a significant increase in MAP. MAP, HR, and adrenaline values were not statistically different between groups. In conclusion, delayed recovery after neurosurgery cannot be recommended as a mechanism of limiting the metabolic and hemodynamic consequences from emergence from general anesthesia.

Implications: In this study, we tested the hypothesis that delayed recovery after neurosurgery would attenuate the consequences of recovery from general anesthesia. As markers of stress, oxygen consumption and noradrenaline blood levels were higher after delayed versus early recovery. Thus, delayed recovery cannot be recommended as a mechanism of limiting the metabolic and hemodynamic consequences from emergence after neurosurgery.

*Département d’Anesthésie-Réanimation, †Laboratoire de pharmacologie clinique, and ‡Service de Neurochirurgie, Hôpital Timone, Marseille; France; and §Département d’Anesthésiologie et de Réanimation, Hôpital de Sion, Sion, Switzerland

May 18, 1999.

Address correspondence and reprint requests to Dr. Nicolas Bruder, UNSIN, Hôpital Timone, 13385 Marseille Cedex 05, France. Address e-mail to

Recovery from general anesthesia is a stress period characterized by sympathetic stimulation. This period is associated with increased oxygen consumption (V̇O2) (1), catecholamine secretion (2,3), tachycardia, and hypertension. These metabolic and cardiovascular responses may adversely affect the balance between myocardial oxygen supply and demand. Thus, modulation of the postoperative sympathetic response may afford hemodynamic stability and decrease morbidity in high-risk surgical patients. Several techniques may achieve this goal. Clinical trials have shown that the perioperative use of sympatholytic drugs (4) decreased sympathetic activity, tachycardia, hypertension, and myocardial ischemia during emergence from anesthesia after cardiovascular surgery. Another technique often advocated to limit the postoperative hyperdynamic and metabolic response is delayed recovery in the intensive care unit (ICU). However, the metabolic and hemodynamic benefit of delayed versus early recovery has not been established. In neurosurgical patients, an increase in blood pressure soon after surgery may lead to intracranial bleeding (5). 1 In our institution, neurosurgeons often advocate delayed recovery to attenuate the metabolic and hemodynamic consequences linked to awakening. Ideally, recovery from neuroanesthesia should be smooth and progressive, with no major hemodynamic or metabolic disturbances. It should also be rapid to allow neurological assessment. These two conditions may not be compatible. In the past, a postoperative stabilization period was probably indicated because of perioperative hypothermia, cardiovascular disturbances due to blood loss or the use of controlled hypotension, or cerebral edema due to excessive brain retraction. Progress achieved in surgical and anesthetic techniques may eliminate the indications of delayed awakening. We performed this study to test the hypothesis that delayed recovery would attenuate metabolic and hemodynamic changes due to recovery after intracranial surgery. V̇O2 and catecholamine concentrations were measured as markers of the stress response during emergence from general anesthesia.

Back to Top | Article Outline


After institutional review board approval and written, informed consent, 30 patients, ASA physical status I or II, scheduled for elective intracranial surgery, were included in the study. The exclusion criteria were a preoperative decrease in the level of consciousness, treatment with β-blocking drugs or clonidine, an anticipated duration of surgery >5 h, and complications during surgery. Preoperatively, the patients were randomly divided into two groups. In Group I, the patients were tracheally extubated as soon as possible in the recovery room after surgery. In Group II, the patients were sedated with propofol for 2 h (3–5 mg · kg−1 · h−1 IV) after surgery, and their tracheas were extubated in the ICU. The dose of propofol was adjusted to obtain deep sedation (patient unresponsive to verbal command but showing a motor response to noxious stimulation). No additional hypnotic or analgesic drugs were given during this period. All patients were premedicated with 1 mg of oral lorazepam 1 h before surgery. Anesthesia was maintained with propofol (4–9 mg · kg−1 · h−1 IV), fentanyl (5–10 μg/kg IV), IV atracurium as needed to maintain 0 or 1 response to train-of-four stimulation during surgery, and oxygen in air (inspired oxygen fraction [FIO2] 0.4–0.5). All patients were warmed during anesthesia by using a forced-air warming blanket to obtain a central body temperature >36°C before stopping the administration of propofol. Bladder temperature was recorded at extubation. The criteria for extubation were the ability to obey verbal commands, absence of severe hemodynamic changes, and absence of residual neuromuscular blockade. V̇O2 was measured by indirect calorimetry using the Deltatrac metabolic monitor (Datex, Helsinki, Finland) in the ventilator mode in intubated patients and in the canopy mode after extubation. The Deltatrac has been validated both in vitro and in vivo. In the laboratory, the mean error on V̇O2 was 4% ± 4% (6), and the bias and the standard error of the difference with the reference method (the Douglas bag method) were only −3.5 mL/min and 2.5 mL/min (7). In vivo, the precision of the device was similar to the in vitro measurements except for FIO2 values >0.6 (6,7). All measurements were performed at a FIO2 of 0.5 or 0.4. The FIO2 was not changed for at least 20 min before each measurement period. The patients were ventilated using the Servo 900 C (Siemens) with an additional inspiratory mixing chamber. There were three measurement periods in both groups. The baseline measure was recorded with the patient under anesthesia after the craniotomy. The second measurement period took place from the cessation of propofol to extubation (recovery). The third measure was recorded 30 min after extubation. A pressure calibration was performed before each procedure, and a gas calibration was performed before each measurement period. During anesthesia and 30 min after extubation, the V̇O2 measurements were recorded over 20 min and were averaged after a 10-min stabilization period. The data were recorded every minute between discontinuation of propofol and extubation. The preextubation value for V̇O2 was the average value of the last 5-min measures before disconnection of the ventilator. The mean V̇O2 value from the disconnection of propofol to extubation was noted V̇O2recovery. Venous blood samples were taken during anesthesia, at extubation, and 30 min later for the measurement of plasma catecholamines (adrenaline and noradrenaline). The samples were immediately centrifuged at 4°C and frozen at −80°C. Plasma catecholamines were measured by using high-performance liquid chromatography with reverse-phase ion-pair chromatography and electrochemical detection, according to Hjemdahl (8). The lower limits of detection were 5 pg/mL for each catecholamine. The coefficients of variation for intraassay precision were <7%, and the method was linear over a range of 5–1000 pg/mL. Mean arterial pressure (MAP) and heart rate (HR) were measured at the same time. A MAP >120 mm Hg or a systolic arterial pressure >180 mm Hg was treated with IV nicardipine. The data are presented as mean ± SD. V̇O2, catecholamine blood levels, MAP, and HR data were compared by using repeated-measures analysis of variance. Within-group comparisons used Fisher’s protected least significant difference test, and between-group comparisons for each time period used an unpaired Student’s t-test. The other data were compared by using an unpaired Student’s t-test. A P value <0.05 was considered significant.

Back to Top | Article Outline


The patients in Group I and Group II were comparable as to age (50 ± 12 vs 52 ± 13 yr), weight (66 ± 12 vs 66 ± 13 kg), body surface area (1.7 ± 0.2 vs 1.7 ± 0.2 m2), length of surgery (264 ± 46 vs 265 ± 59 min), and fentanyl doses (397 ± 111 vs 465 ± 148 μg). Body temperature at extubation was also similar (36.6 ± 0.3 vs 36.8 ± 0.4°C). There was a significant difference between the two groups for V̇O2 and blood noradrenaline concentrations (Figures 1 and 2) and between the time periods for V̇O2, blood noradrenaline concentrations, and MAP (Figures 1 and 2, Table 1). V̇O2 was greater at extubation in Group II than in Group I. V̇O2recovery was 198 ± 38 mL/min in Group I and 268 ± 49 mL/min in Group II (P < 0.01). Blood noradrenaline concentrations were higher in Group II than in Group I at extubation and 30 min after extubation. MAP increased significantly on extubation in both groups, but the differences did not reach statistical significance. Two patients required an infusion of IV nicardipine during recovery in Group II, compared with none in Group I. There was no significant difference between groups or time for adrenaline and HR (Table 1).

Back to Top | Article Outline


Large variations in early postoperative V̇O2 have been reported, mainly due to the variability of body temperature values and shivering in the postoperative period. In a study by Ciofolo et al. (1), the V̇O2 increased from 173 ± 26 mL/min at the end of anesthesia to 457 ± 88 mL/min 20 minutes later, but all patients shivered. Shivering is often associated with a 200%–400% increase in V̇O2 (1,9–11). In one study, the difference in V̇O2 between shivering and nonshivering patients was only 38%. This was explained by the advanced age of the patients (mean 70 years), which may have impaired the thermoregulatory response to hypothermia (12). We did not observe shivering in any patient because no patient was hypothermic at the end of surgery. This may also be related to the use of forced-air warming, which decreases the incidence and the intensity of shivering (13,14). Another reason for the lack of shivering and, hence, low postoperative V̇O2 values, may be the choice of propofol as a hypnotic. It has been demonstrated that patients who received propofol anesthesia had a significantly lower incidence of postanesthetic shivering than those who received isoflurane, despite similar body temperature readings in both groups (15). The low V̇O2 changes we measured, especially in Group I, are in agreement with other data in nonshivering patients (12,16). This suggests that perioperative warming methods and choice of anesthetics have an important influence on metabolic changes during recovery from general anesthesia. There is a close relationship between oxygen demand and delivery during emergence from anesthesia for patients without major cardiovascular disease (17). Patients with limited cardiovascular reserve may not be able to supply the high V̇O2 described in several studies, which would impair tissue oxygenation. This explains why Plunkett et al. (18) investigated the benefit of deep postoperative sedation with propofol after cardiac surgery. It was demonstrated that deep propofol sedation attenuated the postoperative increase in stress hormone compared with a group of patients receiving standard care. However, the concentrations of stress hormones rapidly returned to the control group values when the propofol infusion was stopped. In noncardiac surgery, anesthetic reversal and tracheal extubation induce a major sympathetic stimulation (17,19). It is difficult to distinguish between the response to surgery and to emergence from anesthesia. During surgery, the concentrations of stress hormones are usually low except for intraabdominal or intrathoracic operations (2,20). The increase in stress hormones occurs at the end of the surgical period peaking at emergence (2). Usually, both adrenaline and noradrenaline concentrations increase. In the present study, although adrenaline changes were not significant due to large interindividual variability, the changes followed a pattern similar to that of noradrenaline. The lower metabolic changes measured in Group I versus Group II and the fast decrease in catecholamine concentrations 30 minutes after extubation suggests that recovery from general anesthesia and extubation are major stressful procedures compared with the postoperative effects of surgery. Thus, the stress response after surgery was delayed, but not attenuated, by propofol sedation. In fact, our data suggest that the stress response may be even greater when recovery and extubation are delayed in neurosurgical patients. Modulation of the postoperative stress response with propofol seems limited to the infusion period, which implies that postoperative sedation would be warranted only in patients with physiologic disturbances requiring correction before recovery.

The difference in catecholamine concentrations between groups should lead to differences in hemodynamic variables. We did not find a significant difference between groups for arterial blood pressure, although the increase seemed larger in Group II on extubation. This may be due to insufficient statistical power of the study, because the aim of the study was not to search for differences in blood pressure between groups. The lack of difference in blood pressure between groups may also be due to the need to treat the patients who develop hypertension during emergence because two patients required nicardipine treatment in Group II, compared with none in Group I.

The cardiovascular and metabolic changes may seem modest in this study compared with others. Such changes are usually inconsequential, but neurosurgical patients may experience unfavorable sequelae. Many authors have reported that endotracheal suctioning increases intracranial pressure (ICP) (21–23). Although the effect of recovery and extubation on ICP have not been investigated, it is well demonstrated that tracheal suctioning, especially when associated with coughing or bucking, is linked to increases in ICP. Arterial hypertension is associated with postoperative intracranial hemorrhage (5).1 Thus, it seems justified to limit the metabolic and hemodynamic changes during emergence after intracranial procedures. It is also important to consider the maximal increase in each group because the hemodynamic changes are often treated when a value exceeds a given threshold. In Group I, the maximal noradrenaline blood concentration and V̇O2, respectively, were 354 and 291 mL/min; in Group II, the respective values were 833 and 434 mL/min. Thus, early extubation after surgery may limit the number of patients experiencing a metabolic stress at risk of neurological or cardiovascular complications.

A limitation of our study is that we did not evaluate postoperative pain, but there is a link between pain and catecholamine release. Epidural anesthesia blunts the increase in plasma catecholamines both during and after surgery, and there is a relationship between pain score and noradrenaline concentrations (2,24). Although intracranial surgery is not very painful postoperatively, a mean fentanyl dose of 397 ± 111 μg/kg in Group I should have provided analgesia during recovery. In Group II, after two hours of propofol anesthesia, the patients might have experienced more pain than those in Group I because of a progressive decrease in the fentanyl blood concentration. One study supported the importance of perioperative analgesia on V̇O2 in the immediate postoperative period (25). Twelve patients scheduled for maxillary surgery were randomized into two groups receiving either 3 or 1 μg · kg−1 · h−1 fentanyl during surgery with enflurane in both groups. No patient was hypothermic, and body temperature was similar in both groups after surgery. During the first 30 minutes after anesthesia, the V̇O2 in the small-dose fentanyl group was >60% greater than the V̇O2 of the other group. Another study has shown that the increase in arterial blood pressure was attenuated by a dose of fentanyl as small as 1 μg/kg given three to five minutes before extubation (26). This emphasizes the importance of analgesia during the recovery period and on extubation.

It can be argued that the differences between the groups were due to differences in the management of recovery. In fact, the tracheas of patients in Group I were extubated in the recovery room, whereas those of patients in Group II were extubated in the neurosurgical ICU. All patients were cared for by the same anesthesiologists, who attempted to extubate the trachea in an identical manner in both groups. In our hospital, the environment of the ICU is much less noisy than that in the recovery room, which would probably limit the number of stressful stimuli in the ICU. Other undetected factors may explain the results, but we tried to perform the study under realistic clinical, rather than experimental, conditions. Thus, the results may truly reflect the metabolic changes of recovery from neurosurgical anesthesia.

In conclusion, we confirmed that V̇O2 is low during emergence from anesthesia in normothermic neurosurgical patients. Because no specific intervention was undertaken after discontinuing propofol anesthesia, it is likely that the anesthetic technique influenced metabolic changes during recovery. The metabolic variables were recorded in nearly ideal conditions for recovery in this study and may not reflect the changes after more invasive surgical procedures. However, in this study, the stress response to emergence was greater with prolonged versus immediate extubation. Considering the development of new short-acting anesthetics, especially opioids, the metabolic consequences of the anesthetic technique on recovery in various surgical procedures deserve further study. In neurosurgical patients, delayed recovery cannot be recommended after intracranial surgery to attenuate the metabolic or hemodynamic changes in normothermic patients.

Back to Top | Article Outline


1 Basali A, Schubert A, Kalfas I. Perioperative hypertension and postcraniotomy intracranial hemorrhage [abstract]. Anesthesiology 1994;81(Suppl):A203. Cited Here...

Back to Top | Article Outline


1. Ciofolo MJ, Clergue F, Devillier C, et al. Changes in ventilation, oxygen uptake, and carbon dioxide output during recovery from isoflurane anesthesia. Anesthesiology 1989; 70: 737–41.
2. Breslow MJ, Parker SD, Franck SM, et al. Determinants of catecholamine and cortisol responses to lower extremity revascularization. Anesthesiology 1993; 79: 1202–9.
3. Udelsman R, Norton JA, Jelenich SE, et al. Responses of the hypothalamic-pituitary-adrenal and renin-angiotensin axes and the sympathetic system during controlled surgical and anesthetic stress. J Clin Endocrinol Metab 1987; 64: 986–94.
4. McSPI-EUROPE Research group. Perioperative sympatholysis: beneficial effects of the α2-adrenoreceptor agonist mivazerol on hemodynamic stability and myocardial ischemia. Anesthesiology 1997; 86: 346–63.
5. Kalfas IH, Little JR. Postoperative hemorrhage: a survey of 4992 intracranial procedures. Neurosurgery 1988; 23: 343–7.
6. Takkala J, Keinänen O, Väisänen P, Kari A. Measurement of gas exchange in intensive care: Laboratory and clinical validation of a new device. Crit Care Med 1989; 17: 1041–7.
7. Tissot S, Delafosse B, Bertrand O, et al. Clinical validation of the Deltatrac monitoring system in mechanically ventilated patients. Intensive Care Med 1995; 21: 149–53.
8. Hjemdahl P. Catecholamine measurements by high performance liquid chromatography. Am J Physiol 1984; 247: E13–20.
9. Bay J, Nunn JF, Prys-Roberts C. Factors influencing arterial P O2 during recovery from anesthesia. Br J Anaesth 1968; 40: 398–407.
10. MacIntyre PE, Pavlin EG, Dwersteg JF. Effect of meperidine on oxygen consumption, carbon dioxide production, and respiratory gas exchange in postanesthetic shivering. Anesth Analg 1987; 66: 751–5.
11. Ralley FE, Wynands E, Ramsay JG, et al. The effects of shivering on oxygen consumption and carbon dioxide production in patients rewarming from hypothermic cardio-pulmonary bypass. Can J Anaesth 1988; 35: 332–7.
12. Franck SM, Fleischer LA, Olson KF, et al. Multivariate determinants of early postoperative oxygen consumption in elderly patients. Anesthesiology 1995; 83: 241–9.
13. Sharkey A, Lipton JM, Murphy MT, Giesecke AH. Inhibition of postanesthetic shivering with radiant heat. Anesthesiology 1987; 66: 249–52.
14. Lennon RL, Hosking MP, Conover MA, Perkins WJ. Evaluation of a forced-air system for warming hypothermic postoperative patients. Anesth Analg 1990; 70: 424–7.
15. Cheong KF, Low TC. Propofol and postanaesthetic shivering. Anaesthesia 1995; 50: 550–2.
16. Rodriguez JL, Weissman C, Damask MC, et al. Morphine and postoperative rewarming in critically ill patients. Circulation 1983; 68: 1238–46.
17. Hosoda R, Hattori M, Shimada Y. Favorable effects of epidural analgesia on hemodynamics, oxygenation and metabolic variables in the immediate post-anesthetic period. Acta Anaesthesiol Scand 1993; 37: 469–74.
18. Plunkett JJ, Reeves JD, Ngo L, et al. Urine and plasma catecholamine and cortisol concentrations after myocardial revascularization. Anesthesiology 1997; 86: 785–96.
19. Lowrie A, Johnston PL, Fell D, Robinson SL. Cardiovascular and plasma catecholamine responses at tracheal extubation. Br J Anaesth 1992; 68: 261–3.
20. Naito Y, Tami S, Slinger K, et al. Responses of plasma adrenocorticotropic hormone, cortisol and cytokines during and after upper abdominal surgery. Anesthesiology 1992; 77: 426–31.
21. Donegan MF, Bedford RF. Intravenously administered lidocaine prevents intracranial hypertension during endotracheal suctioning. Anesthesiology 1980; 52: 516–8.
22. White PF, Schlobohm RM, Pitts LH, Lindauer JM. A randomized study of drugs for preventing increases in intracranial pressure during endotracheal suctioning. Anesthesiology 1982; 57: 242–4.
23. Brucia J, Rudy E. The effects of suction catheter insertion and tracheal stimulation in adults with severe brain injury. Heart Lung 1996; 25: 295–303.
24. Rutberg H, Hakanson E, Anderberg B, et al. Effects of the extradural administration of morphine, or bupivacaine, on the endocrine response to upper abdominal surgery. Br J Anaesth 1984; 56: 233–7.
25. Combes P, Lavagne P. Influence de l’analgésie peropératoire sur la consommation d’oxygène postopératoire immédiate. Ann Fr Anesth Reanim 1991; 10: 343–7.
26. Fuhrman TM, Ewel CL, Pippin WD, Weaver JM. Comparison of the efficacy of esmolol and alfentanil to attenuate the hemodynamic responses to emergence and extubation. J Clin Anesth 1992; 4: 444–7.
© 1999 International Anesthesia Research Society