Share this article on:

Anesthetic Requirements and Stress Hormone Responses in Spinal Cord-Injured Patients Undergoing Surgery Below the Level of Injury

Yoo, KyungYeon MD, PhD; Hwang, JaeHa MD, PhD; Jeong, SungTae MD; Kim, SeokJai MD; Bae, HongBeom MD; Choi, JeongIl MD, PhD; Chung, SungSu MD, PhD; Lee, JongUn MD, PhD

doi: 10.1213/01.ane.0000198429.09694.d3
Neurosurgical Anesthesia: Research Report
Chinese Language Editions

Neuraxial anesthesia decreases the minimum alveolar concentration. We determined the effects of spinal cord injury (SCI) on sevoflurane requirements and stress hormone response. Twenty-two chronic SCI patients undergoing surgery below the level of the injury were enrolled in the study, and 15 patients without cord injury served as control patients. Bispectral index score was maintained at 40–50. Measurements included end-tidal sevoflurane concentrations, systolic arterial blood pressure, heart rate, and plasma catecholamine and cortisol concentrations. During surgery, systolic arterial blood pressure, heart rate, and Bispectral index were comparable between SCI and control groups. However, end-tidal sevoflurane concentration was significantly smaller in the SCI (0.81%–1.06%) versus control (1.28%-1.31%) patients. In the control group, plasma norepinephrine and cortisol concentrations were significantly increased during and 1 h after surgery compared with awake baseline values. In the SCI group, the sympathoadrenal and cortisol responses were virtually abolished. We conclude that SCI reduces the anesthetic requirement by 20%–39% during surgery below the level of injury, in association with blunted sympathoadrenal and cortisol responses.

IMPLICATIONS: Sevoflurane requirement is reduced by 20%-39% to maintain the Bispectral index at 40-50 in spinal cord-injured patients during surgery below the level of injury, in association with blunted stress hormone responses.

Department of Anesthesiology, Department of Plastic and Reconstructive Surgery, Department of Physiology, Chonnam National University Medical School, 8 Hak-dong, Gwangju 501-190, South Korea

Accepted for publication November 4, 2005.

Address correspondence and reprint requests to Kyung Yeon Yoo, MD, Department of Anesthesiology and Pain Medicine, Chonnam National University Medical School, 8 Hak-dong, Gwangju 501-190, Korea. Address e-mail to

Neuraxial anesthesia has sedative effects related to the level of sensory block (1) and reduces the amount of sevoflurane required to maintain an adequate depth of anesthesia (2). Afferent impulses initiated from the site of stimulation travel along the spinal cord toward the medulla to activate the hypothalamus. Noxious stimulation transmitted through the spinal cord reaches the cerebral cortex via the brainstem reticular formation and the thalamus, where it evokes an electroencephalographic arousal response.

A complete spinal cord injury or transection not only results in loss of sensory and motor function but also an interruption of sympathetic outflow below the level of injury (3). The anesthetic requirement should thus be reduced in cord-injured patients. Indeed, patients with low, complete lesions may undergo surgery below the level of injury without anesthesia (4). However, few studies have assessed the influence of removal of afferent spinal signaling to the brain (spinal deafferentation) on the anesthetic requirements in cord-injured patients.

The present study was designed to determine the effects of spinal cord injury on anesthetic requirements during a surgery below the injured level. The depth of anesthesia was adjusted to maintain bispectral index (BIS) between 40 and 50 (5,6). Because neurogenic blockade with epidural analgesia inhibits the hormonal response to surgical stress during pelvic and lower limb surgery (7), catecholamine and cortisol concentrations were also measured.

Back to Top | Article Outline


The protocol of the study was approved by the University Hospital Ethics Committee. Informed written patient consent was obtained before starting the study. In patients who were unable to give consent by themselves because of injury, their consent was taken from the next of kin. A power analysis suggested a sample size of 11 patients in each group should be adequate to detect a 30% reduction in the sevoflurane requirement with a power of 0.8 and α of 0.05. In this context, the study enrolled 22 patients, classified as ASA physical status II, with chronic, clinically complete cord injuries scheduled for a surgery for pressure sore below the level of the neurologic lesion under general anesthesia. Fifteen age-matched, nondisabled patients receiving musculocutaneous flap surgery served as controls. None had any cardiovascular, pulmonary, or metabolic diseases. Patients who took medications that would influence autonomic or cardiovascular responses to the surgery were excluded. Patients whose interval from injury to the operation was <6 wk were also excluded, as it takes up to 4–6 wk to recover muscle tone and reflexes in the spinal cord distal to the site of injury (8).

The level and the completeness of spinal cord injury were assessed according to the 1996 American Spinal Injury Association standards (9). Motor function was examined using key muscles for levels C5 through T1 and L2 through S1. Total paralysis of motor strength was regarded as a complete lesion. Sensory level was examined by light touch and pinprick at each dermatome, and anesthesia and analgesia were regarded as a complete lesion. None of the cord-injured patients had touch or pain sensation at the site of surgery.

The patients did not receive any premedication. On arrival in the operating room, Ringer’s lactate solution was initiated at a rate of 10 mL · kg−1 · h−1 IV. Furthermore, a 20-gauge catheter was placed into a radial artery connected to a pressure transducer to measure arterial blood pressure and to collect blood samples. A BIS® monitor strip (BIS® Sensor-Aspect Medical Systems, Natick, MA) was placed on the forehead before induction of anesthesia. BIS values were recorded with a BIS monitor (model A-2000; 3.1 software version; Aspect Medical systems Inc.).

After breathing 100% oxygen, anesthesia was induced with thiopental 5–7 mg/kg IV, followed by vecuronium 0.12 mg/kg. Direct laryngoscopy and endotracheal intubation were performed when neuromuscular block was achieved, and anesthesia was maintained with sevoflurane and 50% nitrous oxide (N2O) in oxygen. All subjects’ lungs were mechanically ventilated using a ventilator to maintain end-tidal carbon dioxide tension at 35–40 mm Hg. Neuromuscular blockade was carefully controlled by train-of-four monitoring, and additional boluses of vecuronium were administered to maintain one twitch response at the orbicularis oculi during the surgical procedure. Routine monitoring included invasive measurement of systemic blood pressure, heart rate (HR) and rhythm by 5-lead electrocardiogram and oxygen saturation by pulse oximetry. Throughout the study, the end-tidal concentrations of carbon dioxide, sevoflurane (ETSEVO), and N2O (ETN2O) were measured using a gas analyzer (Capnomac Ultima; Datex-Ohmeda, Helsinki, Finland).

The concentration of sevoflurane was adjusted to achieve a BIS value of 40-50. Bolus doses of fentanyl 0.5 μg/kg were given if systolic arterial blood pressure (SAP) and/or HR increased by >20% above baseline for >5 min in response to a surgical stimulus. Blood loss, assessed by measuring suctioned blood and by weighing gauzes and drapes, was replaced with Ringer’s lactate solution in a ratio of 1:4 or with packed cells if hemoglobin was <10 g/dL. Hypotension was treated with boluses of ephedrine 5 mg IV if SAP decreased by 30% from the initial SAP. At the completion of surgery, volatile anesthetics were discontinued and residual neuromuscular block was antagonized with atropine 1.0 mg and neostigmine 2.5 mg. Estimated blood loss and amount of fluid or blood administered during the surgery were recorded. All the anesthetic procedures were conducted by an anesthesiologist who changed ETSEVO only on the basis of changes in BIS. Postoperative pain intensity was estimated using the visual analog scale in which 0 = no pain and 10 = worst possible pain. Pain was treated with IV fentanyl and ketorolac to achieve a visual analog scale score <4. Patients were asked whether they recalled any intraoperative events on the first postoperative day.

SAP and HR were recorded before induction of anesthesia, before surgery, and every 5 min during and at 1 h after the surgery. The values during every 1 h of the surgery were averaged to provide a single mean value of corresponding hour periods, i.e., hours 0–1, 1–2 and >2. ETSEVO, ETN2O, and BIS values were similarly averaged except for the time point before induction of anesthesia and at 1 hour after the surgery.

Arterial blood samples were drawn before anesthesia induction (baseline), immediately before the start of surgery, 1 and 2.5 h into the surgery (during flap incision and elevation, and closure), and 1 h after the end of surgery. The samples were collected into prechilled tubes containing EDTA-Na and immediately centrifuged at 3000 rpm for 10 min at 4°C. The plasma was stored at −70°C until assayed. Plasma concentrations of norepinephrine and epinephrine were measured in duplicate by using high-pressure liquid chromatography (10). The assay sensitivity was 10 pg/mL, and within-run precision coefficients of variation were 13.5% and 14.2% for norepinephrine and epinephrine, respectively. Plasma concentrations of cortisol were determined by radioimmunoassay (DSL-2000 SP Aktive® Cortisol; Diagnostic System Laboratories, Sinsheim, Germany).

All data are presented as mean ± sd. Demographic and surgical data were compared using Student’s unpaired, t-tests. Differences between the groups of variables (SAP, HR, BIS values, ETSEVO, and stress hormones) were analyzed by two-way analysis of variance with repeated measures. A Scheffé test was used for multiple pairwise comparisons when a significant difference was indicated with analysis of variance. Linear regression was performed to determine the relation between measured ETSEVO and the level of injury (i.e., the most cephalic level of complete sensory lesions) in cord-injured patients. A P value <0.05 was considered statistically significant.

Back to Top | Article Outline


Demographic and surgical data were similar between the two groups (Table 1). All surgeries were musculocutaneous flaps done for management of pressure sores in cord-injured patients and reconstructions in the controls. Total operative time, intraoperative fluid requirement, and blood loss did not differ between groups. Figure 1 shows the distribution of lesions of cord-injured patients. Of the 22 patients, the level of lesion was between T2 and L1 with above T7 in 5 patients and below T7 in 17. All control patients received IV analgesics (fentanyl, 100 ± 0 μg; ketorolac, 46 ± 11 mg) during the 1-h period after the end of surgery. Analgesics were not required by the cord-injured patients during this interval. The day after the surgery, no patients recalled awareness during anesthesia and surgery.

Table 1

Table 1

Figure 1

Figure 1

During the surgery, ETSEVO was significantly smaller in the cord-injured versus control patients (0.81%–1.06% versus 1.26%–1.31%; P < 0.01) (Table 2). In addition, the cord-injured patients who had small ETSEVO (<0.8%, n = 14) during the first hour period manifested significant increases in anesthetic concentrations (61% ± 40%) during flap surgery after débridement of necrotic tissue, whereas those who had large ETSEVO (>0.8%, n = 8) showed little change (2% ± 11%). Although two patients demonstrated unexpectedly large ETSEVO, there was no significant correlation between the sensory level and ETSEVO (Fig. 2). BIS, SAP, and HR were similar between the two groups.

Table 2

Table 2

Figure 2

Figure 2

Two control patients received fentanyl (0.5 μg/kg each) at the start of surgery, whereas 2 cord-injured patients received ephedrine (one patient 5 mg and another 10 mg) during the late part of surgery. ETN2O concentration was 46% ± 1% throughout the study, which did not differ between the groups. One hour after surgery, SAP and HR did not differ from the baseline value in the cord-injured group, but increased significantly in the control group (P < 0.05).

Figure 3 shows the plasma concentrations of norepinephrine, epinephrine, and cortisol. Preoperative baseline hormonal concentrations did not differ between the groups. Anesthetic induction did not affect norepinephrine or cortisol but significantly reduced epinephrine in both groups. Norepinephrine and cortisol significantly increased during the surgery and 1 h after the end of surgery when compared with baseline values in the control group. In contrast, in the cord-injured group, norepinephrine plasma concentrations remained unchanged throughout the study. Epinephrine and cortisol decreased to less than baseline values during surgery and recovered to baseline values at 1 h after the end of surgery (P < 0.05). During surgery and 1 h after the end of surgery, all stress hormones measured in the cord-injured group were significantly less than in the control.

Figure 3

Figure 3

Back to Top | Article Outline


Our results demonstrate that the amount of sevoflurane required to maintain BIS within the range of 40–50 is reduced by 20%–39% in cord-injured patients. This was not related to the age of the patients. Concomitantly, sympathoadrenal and cortisol responses were significantly attenuated. These findings are consistent with those reported by Antognini et al. (11), in which isoflurane’s action in the spinal cord indirectly inhibited brain cortical activity induced by electrical stimulation of the reticular formation. It is suggested that blockade of ascending somatosensory transmission either by neuraxial blockade or spinal cord injury increase susceptibility to anesthetics. Although two patients showed unexpectedly large ETSEVO, there was no specific correlation between the sensory level and ETSEVO (Fig. 2).

The stress response to surgery is characterized by sympathetic activation and enhanced secretion of pituitary hormones (12). The adrenergic response is related to the extent of surgical trauma and is determined by afferent signals (nociceptive pathways and humoral mediators) originated from the site of injury (13). Therefore, it is not surprising to observe different neuroendocrine stress responses between the two groups. During recovery, control patients showed significant catecholamine and pressor responses associated with consumption of analgesics. In contrast, in the cord-injured group, catecholamine and hemodynamic responses were markedly attenuated and no patients experienced any pain at the operative site. Autonomic hyperreflexia with a clinically significant increase of catecholamines generally occurs in patients with a sensory level above T6 but not in those with lesions below T10 (14). Nevertheless, the norepinephrine responses at hour 2.5 into surgery did not differ between the cord-injured and the control patients, although 16 of 22 (73%) of the cord-injured patients had lesions below T10. The surgical stress may have activated postganglionic sympathetic neurons through a spinal reflex, resulting in release of norepinephrine regardless of the injured level. In fact, Levin et al. (15) observed that resting sympathoadrenal tone, as indicated by plasma catecholamine levels, was either normal or increased in supine, chronic, quadriplegic male humans, speculating that the activity of these systems is under the control of local spinal reflexes to maintain basal autonomic functions.

Cortisol rapidly increased during surgery in the control group, whereas it decreased in the cord-injured group. This finding is consistent with a previous report that neurogenic blockade at the spinal level by epidural anesthesia inhibits anterior pituitary responses including cortisol in female patients who underwent pelvic or lower limb surgery (7). Huang et al. (16) also found that the hypothalamus-pituitary-adrenal axis was impaired in patients with spinal cord injuries.

However, plasma cortisol concentrations increased significantly over baseline values at 1 hour after the end of surgery in the cord-injured patients, although none suffered from pain during recovery. The peak cortisol response was also observed in the control group at this time. This finding is in line with a previous report that the strongest stress response occurs during emergence from anesthesia (17). Nociceptive pathways may only be partially responsible for the activation of the stress response (18). Some substances directly released at the site of surgery and cytokines themselves may stimulate the release of both adrenocorticotropic hormone and cortisol (19). The exact trigger of such adrenocortical responses is not completely understood.

This study has a few limitations. First, an inhaled anesthetic requirement is generally quantified by either a lack of response in terms of movement (minimum alveolar concentration) or a lack of hemodynamic responses (minimum alveolar concentration needed to blockade adrenergic response). In patients with spinal cord injuries both responses are altered, i.e., the paralyzed patient does not move even in the presence of inadequate anesthesia and hemodynamic responses are sometimes altered by abnormal reflexes. Therefore, we determined the sevoflurane requirements between these two groups when titrated to a BIS value of 40 to 50 irrespective of whether this was adequate or inadequate anesthesia. Second, although SAP and HR did not differ between groups during surgery, two control patients showed pressor and tachycardiac responses and needed fentanyl at skin incision, indicating the depth of anesthesia was inadequate for surgery. It is thus likely that anesthetic requirements in the cord-injured patients would have been further reduced if the anesthetic dose was based on cardiovascular responses. Third, we examined the degree of cord injury only by American Spinal Injury Association standards (9). Spinal cord injury is not an ‘all or nothing‘ phenomenon. Therefore, it may be difficult to guarantee that all patients had a complete transection with no neural function below the level of injury in this study. It may have been better to stratify the patients by the level of autonomic dysfunction or to evaluate autonomic cardiovascular control in relation to the intensity of baroreceptor stimulation or to vagolysis. Finally, to prevent possible occurrence of bucking and thus awakening during surgery with light anesthesia, patients were paralyzed with neuromuscular blocking drugs. This may have led to loss of spasmodic reflex that might have occurred with light anesthesia during surgery.

In conclusion, chronic spinal cord transection reduced the sevoflurane concentration by 20%–39% required to maintain BIS between 40 and 50 and blunted the sympathoadrenal and cortisol responses when the surgery was performed below the level of the lesion.

Back to Top | Article Outline


1. Gentili M, Huu PC, Enel D, et al. Sedation depends on the level of sensory block induced by spinal anaesthesia. Br J Anaesth 1998;81:970–1.
2. Hodgson PS, Liu SS. Epidural lidocaine decreases sevoflurane requirements for adequate depth of anesthesia as measured by the bispectral index monitor. Anesthesiology 2001;94:799–803.
3. Mathias CJ, Frankel HL. The cardiovascular system in tetraplegia and paraplegia. In: Frankel HL, editor. Spinal cord trauma, handbook of clinical neurology. New York: Elsevier Science Publishers; 1992;435–56.
4. Hambly PR, Martin B. Anaesthesia for chronic spinal cord lesions. Anaesthesia 1998;53:273–89.
5. Glass PSA, Bloom M, Kearse LAJ, et al. Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 1997;86:836–47.
6. Katoh T, Suzuki A, Ikeda K. Electroencephalograhic derivatives as a tool for predicting the depth of sedation and anesthesia induced by sevoflurane. Anesthesiology 1998;88:642–50.
7. Hagen C, Brandt MR, Kehlet H. Prolactin, LH, FSH, GH and cortisol response to surgery and the effect of epidural analgesia. Acta Endocrinol 1980;94:151–4.
8. Teasell RW, Arnold MO, Krassioukov A, Delaney GA. Cardiovascular consequences of loss of supraspinal control of the sympathetic nervous system after spinal cord injury. Arch Phys Med Rehabil 2000;81:506–16.
9. American Spinal Injury Association. International standards for neurological and functional classification of spinal cord injury, 5th ed. Chicago: American Spinal Injury Association, 1996:1–24.
10. Holly JMP, Makin HLJ. The estimation of catecholamines in human plasma: a review. Anal Biochem 1983;128:257–74.
11. Antognini JF, Atherley R, Carstens E. Isoflurane action in spinal cord indirectly depresses cortical activity associated with electrical stimulation of the reticular formation. Anesth Analg 2003;96:999–1003.
12. Desborough JP, Hall GM. Endocrine response to surgery. In: Kaufman L. Anesthesia Review, Vol. 10. Edinburgh: Churchill Livingstone, 1993;131–48.
13. Shintani F, Kanba S, Nakaki T, et al. Interleukin-1beta augments release of norepinephrine, dopamine and serotonin in rat anterior hypothalamus. J Neurosci 1993;13:3574–81.
14. Colachis SC. Autonomic hyperreflexia with spinal cord injury. J Am Paraplegia Soc 1992;15:171–86.
15. Levin BE, Martin BF, Natelson BH. Basal sympatho-adrenal function in quadriplegic man. J Auton Nerv Syst 1980;2:327–36.
16. Huang TS, Wang YH, Lee SH, Lai JS. Impaired hypothalamus-pituitary-adrenal axis in men with spinal cord injuries. Am J Phys Med Rehabil 1998;77:108–12.
17. Furuya K, Shimizu R, Hirabayashi Y, et al. Stress hormone responses to major intra-abdominal surgery during and immediately after sevoflurane-nitrous oxide anaesthesia in elderly patients. Can J Anaesth 1993;40:435–9.
18. Schricker T, Carli F, Schreiber M, et al. Propofol/sufentanil anesthesia suppresses the metabolic and endocrine response during, not after, lower abdominal surgery. Anesth Analg 2000;90:450–5.
19. Scarborough DE. Cytokine modulation of pituitary hormone secretion. Ann N Y Acad Sci 1990;594:169–87.
© 2006 International Anesthesia Research Society