Acute respiratory and metabolic acidosis induced by excessive muscle contraction during spinal evoked stimulation
Y. Tohdoh,
S. Sumita,
T. Kawamata,
K. Omote,
S. Kawana and
A. Namiki
Department of Anesthesiology, Sapporo Medical University School of Medicine, South 1, West 16, Chuo-ku, Sapporo, Hokkaido 060-8543, Japan*Corresponding author
Accepted for publication: December 5, 2000
 |
Abstract
|
---|
Spinal somatosensory evoked potentials (SSEPs) have been used to monitor spinal cord function during corrective scoliosis surgery. We report three cases in which direct epidural stimulation for measurement of SSEPs produced paraspinal muscle contraction, resulting in respiratory and metabolic acidosis. In two of the cases, SSEP-induced acidosis was observed even when only the first twitch of the train-of-four response was detectable after a second dose of muscle relaxant. In one of these two cases, the acidosis was abolished after a sufficient dose of vecuronium to ablate the twitch response. To prevent SSEP-induced respiratory and metabolic acidosis, we recommend that SSEPs should be measured only when profound neuromuscular blockade has been obtained.
Br J Anaesth 2001; 86: 58993
Keywords: monitoring, spinal somatosensory evoked potentials; complications, neuromuscular blockade; complications, acidosis
 |
Introduction
|
---|
Spinal somatosensory evoked potentials (SSEPs) are used to monitor spinal cord function and to reduce the risk of spinal cord injury during spinal surgery. SSEPs have been recorded from the epidural space or from the scalp after epidural or peripheral nerve stimulation during scoliosis surgery.13 SSEPs recorded from the scalp are variable and affected by the effects of anaesthetic agents,4 whereas those recorded from epidural catheter electrodes are almost unaffected by anaesthetic agents.5 Adverse effects after epidural electrical stimulation are rare, so this technique is considered clinically reliable and safe.6 7 We previously reported two cases in which direct epidural stimulation for measurement of SSEPs produced excessive paraspinal muscle contraction, resulting in respiratory and metabolic acidosis.8 We now report the effect of neuromuscular blockade on this adverse effect.
 |
Case reports
|
---|
Case 1
A 16-yr-old, 48 kg girl with idiopathic scoliosis underwent spinal fusion with segmental spinal instrumentation. She received midazolam 2.5 mg and atropine 0.5 mg i.m. 30 min before entering the operating room. Anaesthesia was induced with thiamylal 200 mg; vecuronium 6 mg was administered to facilitate tracheal intubation. Anaesthesia was maintained with 60% nitrous oxide in oxygen and 1.52.5% sevoflurane and fentanyl 12 µg kg1 was given every hour. In order to assess motor function during surgery, a wake-up test was planned. For this test, anaesthesia was adjusted to allow the patient to move their feet according to command. Therefore no other neuromuscular blocking drugs were given during surgery. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide concentration between 4.9 and 5.3 kPa. Arterial blood gas data (FIO2=0.33) were as follows: PaO2, 21.1 kPa; PaCO2, 4.7 kPa; base excess (BE), 0.2 mmol litre1 [HCO3], 23.4 mmol litre1; pH 7.402.
SSEPs were monitored using epidural recordings to prevent spinal cord damage resulting from manipulation of the spinal cord. The electrodes for recording and stimulation were placed directly into the epidural space at T1/T2 and L3/L4, respectively, approximately 2 h after administration of vecuronium. Since paraspinal muscle contraction was observed at a stimulus intensity of 4 mA, SSEPs were recorded using a stimulation intensity of 40 mA, a frequency of 16 Hz and a duration of 0.2 ms, but excessive paraspinal muscle contraction was observed. The systolic blood pressure increased from 13.3 to 20.0 kPa, heart rate from 78 to 110 beats min1 and end-tidal carbon dioxide concentration from 4.4 to 8.0 kPa. Breath sounds and artificial ventilation did not change. There were no changes in rectal temperature or urine colour. Arterial blood gas data (FIO2=0.33) from the radial artery at this time were as follows: PaO2, 13.7 kPa; PaCO2, 8.4 kPa; BE, 7.1 mmol litre1; [HCO3], 22.2 mmol litre1; pH 7.145. Although the patient was hyperventilated to a tidal volume of 500 ml and respiratory rate of 20 b.p.m., end-tidal carbon dioxide concentration remained high 30 min after the stimulation. Figure 1 shows the changes in the continuous intra-arterial blood gas monitoring (Paratrend 7; Biomedical Sensors Ltd, High Wycombe, UK) and end-tidal carbon dioxide concentration, peak inspiratory pressure (PIP) and inspiratory minute volume (MV) recorded by capnography (Ultima; Datex Corp, Helsinki, Finland). The intravascular sensor (Paratrend 7) was placed in the radial artery. Increases in PaCO2 and end-tidal carbon dioxide concentration and decreases in pH and PaO2 were observed immediately after the first measurement of SSEPs. Blood lactate concentrations before stimulation and 10 and 60 min after stimulation were 0.7, 4.2 and 15.2 mmol litre1, respectively. Vecuronium 0.1 mg kg1 was given to produce complete neuromuscular block after the stimulation. Five minutes after administration of vecuronium, a second stimulation (40 mA, 16 Hz, 0.2 ms) was performed (S2 in Figure 1). Paraspinal muscle contraction was not observed and there were no changes in any of the physiological parameters measured (Figure 1).

View larger version (35K):
[in this window]
[in a new window]
|
Fig 1 (A) Original trend display of arterial blood pH, PaCO2 and PaO2 recorded by continuous intra-arterial blood gas monitoring (Paratrend 7) and (B) airway pressure, end-expiratory minute volume and end-tidal carbon dioxide concentration recorded by capnography (Ultima) in case 1. Large arrows indicate direct epidural stimuli (S1 and S2). The small arrow indicates tracheal suctioning. Muscle relaxant was not given before the first stimulus (S1) but was given before the second (S2). PaCO2 increased and pH and PaO2 decreased at S1 but not at S2.
|
|
Case 2
A 14-yr-old girl weighing 37 kg with no history of neuromuscular disease underwent corrective surgery for idiopathic scoliosis. Premedication and induction were the same as in case 1. Anaesthesia was maintained with 65% nitrous oxide in oxygen and 1.52.0% sevoflurane and fentanyl 12 µg kg1 was given every hour. To evaluate the degree of neuromuscular blockade, we observed the response of the thumb to train-of-four (TOF) stimulation of the right ulnar nerve using mechanomyography. The electrodes used for measuring SSEPs were placed directly in the epidural space at T1/T2 and L3/L4. Paraspinal muscle contraction appeared at a stimulus intensity of 0.8 mA. Direct epidural stimulation was performed (9 mA, 16 Hz, 0.2 ms) 10 min after administration of vecuronium 0.05 mg kg1. At this time, only the first twitch of the TOF response was detectable. Slight paraspinal muscle contraction was observed. The Paratrend 7, placed in the radial artery, showed an increase in PaCO2 and decreases in pH and PaO2, and the Ultima showed an increase in end-tidal carbon dioxide concentration (Figure 2, S1). Further measurements of SSEPs were carried out after additional doses of vecuronium (Figure 2, S2 and S3). Only slight changes in the PaCO2, pH, PaO2 and end-tidal carbon dioxide concentration were observed. When the TOF ratio was 100%, measurement of SSEPs was performed again. This measurement was performed without warning the anaesthetists. Marked changes in pH, end-tidal carbon dioxide concentration, PaCO2 and PaO2 were observed (Figure 2, S4). The blood lactate concentration also increased, from 0.9 mmol litre1 before epidural stimulation to 8.0 and 12.4 mmol litre1 10 and 60 min after epidural stimulation, respectively.

View larger version (38K):
[in this window]
[in a new window]
|
Fig 2 (A) Original trend display of arterial blood gas analyses recorded by Paratrend 7 and (B) capnogram recorded by Ultima in case 2. Large arrows indicate direct epidural stimuli (S1S4). Vecuronium was given before S1, S2 and S3, but slight paraspinal muscle contractions were observed. Only slight changes in pH, PaCO2 and end-tidal carbon dioxide concentration were observed. When SSEPs were unexpectedly applied, however, significant changes in all these parameters were observed (S4).
|
|
Case 3
A 17-yr-old boy weighing 57 kg with no history of neuromuscular disease underwent corrective surgery for idiopathic scoliosis. Premedication, induction and maintenance of anaesthesia were performed as in case 1. To evaluate the degree of neuromuscular blockade, we observed the response of the thumb to TOF stimulation of the right ulnar nerve using mechanomyography. The electrodes used for measuring SSEPs were located directly into the epidural space at T1/T2 and L3/L4. After confirming the occurrence of paraspinal muscle contraction at a stimulus intensity of 0.8 mA, vecuronium 0.1 mg kg1 was given to obtain complete neuromuscular block. Ten minutes after vecuronium administration, only the first twitch of the TOF response was detectable. Slight paraspinal muscle contraction with epidural stimulation (8 mA, 16 Hz, 0.2 ms) was observed. There were slight increases in arterial pressure and heart rate. The Paratrend 7 showed an increase in PaCO2 and decreases in pH and PaO2, and the Ultima showed an increase in end-tidal carbon dioxide concentration (Figure 3, S1). Twenty minutes after stimulation, vecuronium 0.08 mg kg1 was given. Epidural stimulation was performed again when there was no response to TOF stimulation. Paraspinal muscle contraction was not observed. Arterial pressure, heart rate and end-tidal carbon dioxide concentration remained stable. There were no changes in pH, blood gas or end-tidal carbon dioxide concentration (Figure 3, S2).

View larger version (39K):
[in this window]
[in a new window]
|
Fig 3 (A) Trend display of arterial blood gas analyses recorded by Paratrend 7 and (B) capnogram recorded by Ultima in case 3. A slight increase in PaCO2 and decreases in pH and PaO2 were observed at S1, but not at S2.
|
|
In all cases, blood samples directly obtained from the radial artery were also analysed using an ABL 300 blood gas analyser. The results were similar to those measured by the Paratrend 7.
 |
Discussion
|
---|
The aim of scoliosis surgery is to obtain extensive correction of the spinal curvature without damaging the spinal cord. Measurement of SSEPs has often been used for intra-operative assessment of spinal cord function.13 An alternative method is the wake-up test, which provides only a qualitative assessment of motor function.9 For the latter test, anaesthesia should be adjusted, allowing the patient to move his or her feet according to command. SSEPs reflect somatosensory but not motor function of the spinal cord. To measure SSEPs, the stimulus is gradually increased until minimal twitching of the paraspinal muscle is observed.10 Usually, summation of the SSEPs is repeated 20100 times by recurrent stimuli of 0.10.2 ms in duration, each at a rate of 1030 Hz and at five to 10 times the intensity that induces paraspinal muscle contraction or SSEPs.11 In our patients, after observation of paraspinal muscle contraction at a given stimulus intensity, we applied epidural stimulation at a 10-fold higher intensity, at a rate of 16 Hz and a duration of 0.2 ms and the action potential was recorded.
There have been a few reports of complications arising from measurement of SSEPs using epidural stimulation. Accidental epidural perforation or nerve root injury can occur during insertion of the electrodes into the epidural space. Epidural stimulation can also induce such excessive paraspinal muscle contraction that surgery is prevented.11 However, to the best of our knowledge, our previous report8 was the first of epidural stimulation inducing excessive paraspinal muscle contraction and changing the patients haemodynamic state and acidbase balance.
Intense muscle contraction-induced systemic acidosis with the accumulation of muscle metabolites has been reported after grand mal convulsions12 or after succinylcholine medication.13 14 Erbguth and colleagues13 reported an increase in carbon dioxide production 1 and 5 min after succinylcholine-induced muscle fasciculation. They concluded that significantly increased metabolic activity was associated with the muscle fasciculation. Christensen and colleagues14 also reported that administration of succinylcholine 1 mg kg1 induced systemic muscle fasciculation and increased carbon dioxide production, but these phenomena were abolished when pancuronium 0.01 mg kg1 was injected before succinylcholine and the non-depolarizing agent prevented fasciculation. These results suggest that the muscle fasciculation generated by succinylcholine caused increases in carbon dioxide production.
In our patients, we observed an increase in PaCO2 and decreases in pH and PaO2 immediately after muscle contraction. In case 1 and 2, lactate concentrations rose 15- to 20-fold 1 h after stimulation. These changes after the measurement of SSEPs were prevented by administration of sufficient vecuronium to produce complete neuromuscular block. These results indicate that excessive muscle contraction induced by direct epidural stimulation increases oxygen demand to a level exceeding oxygen supply and consequently causes lactic acidosis.
Orriger and colleagues12 studied the time course of lactic acidosis after a single grand mal seizure. Their results indicated that metabolic acidosis was a primary result of the seizure, not a secondary result of respiratory or renal acidosis.
The TOF ratio is often used to evaluate the degree of neuromuscular blockade. When only the first twitch of the TOF response is detected, the degree of neuromuscular block is approximately 9095% and surgery can normally be performed.15 In these circumstances, epidural stimulation in cases 2 and 3 of the present study caused sufficient muscle contraction to result in slight changes in haemodynamics and acidbase balance. On the other hand, after no response to TOF stimulation in case 3, paraspinal muscle contraction, acidosis and haemodynamic changes were not observed. These observations indicated that complete neuromuscular block is necessary to abolish the adverse effects of epidural stimulation-induced paraspinal muscle contraction.
We used the Paratrend 7 in addition to arterial blood gas analysis using an ABL 300 analyser. The values obtained from the Paratrend 7 have been reported to correlate with arterial values.16 The Paratrend 7 was therefore thought to be effective in the detection of a change in arterial blood gases in acute acidosis.
In conclusion, direct epidural stimulation to measure SSEPs induces excessive paraspinal muscle contraction and severe metabolic and respiratory acidosis. However, these changes can be prevented by administration of sufficient vecuronium to prevent a response to TOF stimulation. We recommend that SSEPs should be measured after confirmation of complete neuromuscular blockade.
 |
References
|
---|
1 Shimoji K, Kano T, Higashi H, Morioka T, Henschel EO. Evoked electroencephalogram record from epidural space in man. J Appl Physiol 1972; 33: 46871[Free Full Text]
2 Engler GL, Spielhotz NI, Bernhard WN, Danziger F, Merkin H, Wolff T. Somatosensory evoked potentials during Harrington instrumentation for scoliosis. J Bone Joint Surg Am 1978; 60: 52832[Abstract]
3 Grundy BL, Nash CI Jr, Lorig RA, Brown RH. Arterial manipulation alters spinal cord function during correction of scoliosis. Anesthesiology 1981; 54: 24953[ISI][Medline]
4 McPherson RW, Mahla M, Johnson R, Traystman RJ. Effects of enflurane, isoflurane, and nitrous oxide on somatosensory evoked potentials during fentanyl anesthesia. Anesthesiology 1985; 62: 62633[ISI][Medline]
5 Tamaki T, Noguchi T, Takano H, et al. Spinal cord monitoring as a clinical utilization of the spinal evoked potentials. Clin Orthopaed 1984; 184: 5864[Medline]
6 Fujioka H, Shimoji M, Tomita S, et al. Spinal cord potential recordings from the extradural space during scoliosis surgery. Br J Anaesth 1994; 73: 3506[Abstract]
7 Jones SJ, Edgar MA, Ransford AO, Thomas NP. A system for the electrophysiological monitoring of the spinal cord during operations for scoliosis. J Bone Joint Surg Br 1983; 65: 34139
8 Tohdoh Y, Sumita S, Iwasaki H, Kawamata T, Tsuchida H, Namik A. Two cases of severe metabolic acidosis after monitoring of evoked spinal potential. Masui 1997; 46: 147982[Medline]
9 Lonstein JE. Adolescent idiopathic scoliosis. Lancet 1994; 344: 140712[ISI][Medline]
10 Machida M, Weinstain SL, Yamada T, Kimura J. Electrophysiological measure of sensory and motor function during spinal surgery. Spine 1985; 10: 40713[ISI][Medline]
11 Kurokawa T. Evoked electrospinogram (I,III). Jpn J Clin Electroencephalogr 1975; 17: 5766, 186195
12 Orringer CE, Eustace JC, Wunsch CD, Gardner LB. Natural history of lactic acidosis after grand-mal seizures. New Engl J Med 1977; 13: 7969
13 Erbguth PH, Bergman NA. The effect of a single dose of succinylcholine on oxygen consumption. Anesthesiology 1973; 39: 4479[ISI][Medline]
14 Christensen KJS, Olesen AS, Jorgensen S. Effects of suxamethonium on CO2-production. Acta Anaesthesiol Scand 1985; 29: 4246[ISI][Medline]
15 Viby-Mogensen J. Neuromuscular monitoring. In: Miller RD, ed. Anesthesia, 3rd edn. New York: Churchill Livingstone, 1990: 12202
16 Venkatesh B, Clutton Brock TH, Hendry SP. A multiparameter sensor for continuous intraarterial blood gas monitoring: a prospective evaluation. Crit Care Med 1994; 22: 58894[ISI][Medline]