Transl. Stroke Res. (2015) 6:399–406 DOI 10.1007/s12975-015-0406-x
ORIGINAL ARTICLE
Methodology of Motor Evoked Potentials in a Rabbit Model Stephen D. Waterford 1 & Michelle Rastegar 1 & Erin Goodwin 2 & Paul A. Lapchak 3 & Viviana Juan 1 & Farnaz Haji 1 & René Bombien 1 & Ali Khoynezhad 1
Received: 6 August 2014 / Revised: 22 April 2015 / Accepted: 12 May 2015 / Published online: 20 May 2015 # Springer Science+Business Media New York 2015
Abstract Spinal cord ischemia (SCI) is a devastating complication of aortic operations. Neuromonitoring using motor evoked potentials (MEPs) is a sensitive modality to detect SCI in humans. We describe a leporine SCI model using MEPs to test pharmaceutical therapeutics and other neuroprotective adjuncts. In 80 rabbits, methods to obtain MEPs in normotensive and ischemic rabbits were developed. The effects of isoflurane, propofol, apnea, and hypotension on lower extremity MEPs were studied. Lower extremity MEPs disappear upon SCI induction in 78 of 78 (100 %) rabbits. Prior to SCI induction and during apneic episodes, lower extremity MEPs were lost in all (100 %) and upper extremity MEPs in one (25 %). Isoflurane was used in four experiments, with loss of lower extremity MEPs in all four (100 %) and loss of upper extremity MEPs in zero. With propofol upper extremity, MEPs were obtainable in 80 of 80 rabbits (100 %) and lower extremity MEPs in 78 of 80 rabbits (97.5 %) prior to SCI induction. The presence of these lower extremity MEPs prior to SCI induction was not correlated with systolic or diastolic blood pressure. Disappearance of MEPs occurred in all 45 rabbits with postoperative lower extremity impairment. MEPs in the leporine model correlate closely with paraplegia. MEPs are influenced by inhaled anesthetics and apnea but not by hypotension alone. Propofol anesthesia provides reliable
* Ali Khoynezhad [email protected] 1
Division of Cardiothoracic Surgery, Cedars-Sinai Medical Center, 127 S. San Vicente Blvd., Suite 3306, Los Angeles, CA 90048, USA
2
Comparative Medicine, Cedars-Sinai Medical Center, Los Angeles, CA, USA
3
Department of Neurology and Neurosurgery, Cedars-Sinai Medical Center, Los Angeles, CA, USA
MEPs. This study provides the basis for a reproducible model of SCI to be used for novel therapeutic drug development. Keywords Neuromonitoring . Motor evoked potential . Spinal cord ischemia . Paraplegia . Rabbit . Thoracic aortic
Introduction Spinal cord ischemia (SCI) leading to paraplegia or paraparesis is a devastating complication of aortic surgery. The risk of SCI ranges from 2 to 14 % in endovascular and open surgical treatment of aortic disease [1–3]. To detect SCI, neuromonitoring of transcranial motor evoked potentials (MEPs) is used during aortic surgery [4]. MEP disappearance indicates that ischemia has occurred and prompts corrective action [4]. MEPs record the electrical responses of muscles following stimulation of the motor cortex with scalp electrodes. While MEPs are standard of care in many hospitals, incorporation of MEP in animal models of SCI, while reported, has been scarce [5, 6]. Neuroprotective medications have shown potential to reduce ischemic damage to neuronal tissue, but implementation requires evidence from in vivo models, which can be further improved by incorporating MEP monitoring [7–9]. Reviews of pharmacologic studies of spinal cord protection in the rabbit have noted wide variation in experimental models used to produce SCI, which has limited comparison of the efficacy of various agents [10]. Therefore, establishing a reliable MEP methodology is required, with provision of specific details on anesthetic regimen and physiologic variables required to elicit MEPs in a reproducible fashion. In this study, the effects of cardiopulmonary events and anesthetic medications on MEP findings in a rabbit model are described, and the
400
relationship between MEP findings and postoperative paraplegia is examined. Numerous authors have in fact reported use of MEPs in the rabbit model, but little description of implementation is given. Pokhrel and associates report the effect of reperfusion blood pressure on neurologic outcome in rabbits undergoing spinal cord ischemia [6]. The authors used MEPs, but the process of obtaining baseline MEPs prior to SCI induction was not addressed. Similarly, Murakami and colleagues examined the precise nature of reperfusion MEPs and correlated these results with the motor examination of rabbits after various periods of spinal cord ischemia [11]. While these results provide a method for interpretation of MEPs, they do not focus on establishment of the MEP model itself. We have found that reliable development of neuroprotective medications requires a stable MEP model and an understanding of the factors that influence MEPs. Inability to reliably obtain MEPs can prevent the development of a robust neuroprotective drug development program. Therefore, our objective is to describe the hemodynamic and anesthetic factors, which influence the setup of an MEP model, in order to facilitate the standardization and troubleshooting of such a model.
Materials Eighty New Zealand White rabbits, Oryctolagus cuniculus, between 3 to 4 kg, were obtained from a United States Department of Agriculture class A vendor. Equipment used is as follows: 24-gauge intravenous catheters, 3.0 mm inner diameter endotracheal tubes, and gauze. Three French Fogarty embolectomy catheters were used (Edwards Lifesciences, Irvine, CA, USA). MEP measurements were performed using an intraoperative neuromonitoring machine (16 channels; Cascade, Cadwell, Kennewick, WA, USA) and commercially available electrodes (Rochester Electro-Medical, Lutz, FL, USA). Medications used were ketamine, xylazine, isoflurane, propofol, carprofen, and buprenorphine, obtained from our institutional pharmacy. Pentobarbital and phenytoin euthanasia solution (Euthasol) was obtained from Virbac Animal Health, Fort Worth, TX, USA.
Methods Rabbits received care according to the Guide for the Care and Use of Laboratory Animals [12] and with approval by the Institutional Animal Care and Use Committee. A 24-gauge intravenous catheter was placed in the lateral ear vein, and general anesthesia was induced intravenously with ketamine (10 mg/mL) and xylazine (2 mg/mL) in saline. Rabbits were intubated with direct laryngoscopy with a 3.0 mm inner diameter cuffed endotracheal tube and the cuff inflated. For four
Transl. Stroke Res. (2015) 6:399–406
experiments, anesthesia was maintained with inhaled isoflurane (0.5–3 %) for the duration of the operation. For other operations, anesthesia was maintained with a continuous infusion of propofol, without inhaled anesthetic. For analgesia, rabbits received carprofen 5 mg/kg subcutaneously after induction and 0.01 mg/kg buprenorphine subcutaneously upon awakening. One liter per minute of oxygen was delivered via endotracheal tube. Vitals signs including non-invasive blood pressure were monitored. After induction, the right groin was clipped. After administration of 1 % lidocaine, a 2 cm longitudinal incision was performed 1 cm distal to the palpable common femoral artery pulse, and dissection was carried down to the superficial femoral artery (SFA). After proximal and distal control of SFA, an arteriotomy was made with fine scissors, and a three French Fogarty embolectomy catheter was advanced into the aorta to 18 cm, established on necropsy to be just below the renal arteries. The rabbit spinal column consists of seven lumbar vertebrae, and the spinal cord travels down to the lumbosacral junction. The lumbar, infrarenal portion therefore supplies the lower extremities, analogous to the human thoracic aorta. The balloon was subsequently inflated to initiate aortic occlusion and distal SCI for a period of 6 to 40 min. These occlusion times will produce varying degrees of motor impairment, and investigators using this protocol will need to calibrate the degree of paraplegia obtained with various occlusion times, with their individual source of rabbits and choice of anesthetic regimen. MEPs were measured immediately following SCI induction, and subsequently, at 30-s intervals until loss of lower extremity, MEPs were obtained. Following reperfusion, MEPs were measured for 5 min prior to awakening the animal from general anesthesia. At the conclusion of experiments, euthanasia was performed with 2 mL intravenous pentobarbital and phenytoin (Euthasol). Technical Requirements For measurement of MEPs, we use an intraoperative neuromonitoring machine with a Cascade 16 channel amplifier (Cadwell, Kennewick, WA, USA) equipped with Cascade software version 2.2.45 and commercially available electrodes (Rochester Electro-Medical, Lutz, FL, USA). For the extremities, twisted pair 12 mm subdermal electrodes are used. For the ground electrode, a single 12 mm electrode is used. Electrodes One pair of electrodes is placed in each limb, using the long head of the triceps brachii muscle for the upper extremity and the gracilis muscle for the lower extremity. Upper extremities provide a positive control, because their MEPs are not affected by lumbar SCI [13]. Upper extremities are not shaved; thus, muscles are palpable but not visible. The muscle therefore
Transl. Stroke Res. (2015) 6:399–406
must be firmly grasped, and electrodes placed directly into muscle, avoiding subcutaneous placement or injury of neurovascular structures (Fig. 1). The 12 mm electrode is inserted entirely. For lower extremity electrodes, the midportion of the SFA lies deep to the gracilis muscle, and electrodes are placed in the medial gracilis to avoid the neurovascular bundle (Fig. 2). Extremity electrodes are placed apart from each other by approximately 3 mm, to allow for measurement of a potential between them. Next cranial electrodes are placed subcutaneously tangential to the scalp overlying the primary motor cortex (M1). In the rabbit, M1 is best stimulated with an electrode rostral to the ear, overlying the fronto-scutular muscle [14], in a palpable ridge of tissue between the ear and the eye. The tissue is elevated, and the electrode is inserted parallel to the scalp (Fig. 3). This method uses visible anatomic landmarks, because electrodes cannot be placed according to the 10–20 system in a rabbit [15], as the brain is too small for application of these coordinates. The ground electrode is placed into the subcutaneous tissue of either shoulder. Electrode impedance is checked, and must be below 5000 Ω. High impedance indicates that the electrode has fallen out of the muscle partially or completely. Next, the voltage of the cortical stimulus is set and should range from 60 to 140 V in the rabbit. Given the fact that SCI can increase the voltage threshold for MEPs, we used a voltage of 100 V, high enough to confirm that lower extremity motor potentials had truly disappeared when ischemia was induced. Next, a gauze bite block is placed between the teeth to prevent jaw muscles from contracting and causing selfinjury or damage to the endotracheal tube. Transcranial stimuli are then triggered from each cranial electrode, and MEP tracings recorded.
Fig. 1 Ventral view of rabbit thoracic limb. The long head of the triceps brachii muscle of the extremity is grasped and elevated, and electrodes placed into the muscle (illustration: VJ)
401
Fig. 2 Ventral view of rabbit pelvic limbs. The femoral pulse is palpated, and electrodes placed 2 cm medial to this in the palpable gracilis muscle (illustration: VJ)
Postoperatively, motor function is rated from 0 to 4 using the Tarlov scale [5]. Animals were examined 2 h postoperatively. This scale has been adapted for use in rabbits and has five scores: 0, paraplegic with no lower extremity function; 1, poor lower extremity function, weak antigravity movement only; 2, some lower extremity motor function with good antigravity strength but inability to draw legs under body or hop;
Fig. 3 Scalp electrodes are placed in a ridge of tissue between the ear and the eye, by grasping and elevating this tissue and placing the electrode parallel to the skull, avoiding bone (illustration: VJ)
402
3, ability to draw legs under body and hop but not normally; 4, normal motor function.
Results MEP tracings were obtained prior to SCI induction with use of propofol total intravenous anesthesia (Fig. 4). Normal MEP tracings, obtained with a 100 V transcranial stimulus to one side of the cortex, are shown in Fig. 4. Firing one transcranial electrode generated muscle contraction in all four extremities, owing to spread of current to the contralateral motor cortex. Upper extremity MEPs were of greater magnitude than lower extremity MEPs in 78 of 78 rabbits (100 %), expected from the shorter distance from the cortex to the upper extremity. Lower extremity MEPs were present at the start of the experiment in 78 of 80 rabbits (97.5 %), while upper extremity MEPs were present in 80 of 80 rabbits (100 %). Following SCI induction, within 60 s, lower extremity MEPs disappeared completely in 78 of 78 rabbits (100 %), while upper extremity MEPs were lost in 0 of 80 rabbits (0 %) (Fig. 5, Table 1), establishing this as a highly sensitive model for SCI detection. Absent MEPs are defined as zero amplitude. As hypotension is common under anesthesia and can alter perfusion of the cerebral cortex and descending pathways, we
Fig. 4 Normal pattern of extremity MEPs following stimulation by one cranial electrode at 100 V. The cranial electrode delivers four stimuli (arrows), and MEPs are indicated. Upper extremity MEPs appear with shorter latency than lower extremity MEPs, owing to shorter neuronal distance
Transl. Stroke Res. (2015) 6:399–406
investigated whether hypotension abolishes MEPs. Using a cutoff for systolic blood pressure of 85 mmHg, we examined 30 animals whose lowest systolic blood pressure was at or above 85 mmHg and 47 animals whose lowest systolic blood pressure was below 85 mmHg. Thirty of 30 animals with a nadir systolic blood pressure at or above 85 mmHg had baseline lower extremity MEPs (100 %), while 45 of 47 animals with a nadir systolic blood pressure below 85 mmHg had baseline lower extremity MEPs (95.7 %) (p=0.52, Table 1). Using a cutoff for diastolic blood pressure of 40 mmHg, we examined 33 animals whose lowest diastolic blood pressure was at or above 40 mmHg and 43 animals whose lowest systolic blood pressure was below 40 mmHg. Thirty-three of 33 animals with a nadir diastolic blood pressure at or above 40 mmHg had baseline lower extremity MEPs (100 %), while 41 of 43 animals with a nadir diastolic blood pressure below 40 mmHg had baseline lower extremity MEPs (95.3 %) (p= 0.50, Table 1). While these results are not statistically significant, the two rabbits without MEPs had a systolic pressure below 85 mmHg and a diastolic pressure below 40 mmHg. In particular, the blood pressure of these two animals was 70/20 and 83/37. When the period of SCI was terminated, allowing for spinal cord reperfusion, we continued to monitor MEPs. We have found three reperfusion responses. Lower extremity MEPs can be absent, can return in a normal pattern, or can return
Fig. 5 Pattern of MEPs following SCI induction, showing absence of lower extremity MEPs and preservation of upper extremity MEPs. This pattern may also be seen prior to SCI induction with inadequate stimulatory voltage or with episodes of apnea and isoflurane use
Transl. Stroke Res. (2015) 6:399–406
403
Table 1 Effect of clinical parameters on loss of upper and lower extremity MEPs Clinical Scenario
Loss of upper extremity MEPs
Loss of lower extremity MEPs
Induction of spinal cord ischemia Nadir systolic blood pressure at or above 85 mmHg Nadir systolic blood pressure below 85 mmHg Nadir diastolic blood pressure at or above 40 mmHg Nadir diastolic blood pressure below 40 mmHg Apnea Isoflurane use Propofol total intravenous anesthesia Postoperative lower extremity impairment
0 of 80 (0 %)
78 of 78 (100 %)
0 of 30 (0 %)
0 of 30 (0 %)a
0 of 47 (0 %)
2 of 47 (4.3 %)a
0 of 33 (0 %)
0 of 33 (0 %)b
0 of 43 (0 %)
2 of 43 (4.7 %)b
1 of 4 (25 %) 0 of 4 (0 %) 0 of 74 (0 %)
4 of 4 (100 %)c 4 of 4 (100 %)c 2 of 74 (3.8 %)c
0 of 46 (0 %)
45 of 45 (100 %)
Fisher exact test p=0.52 for nadir systolic blood pressure≥85 mmHg vs. nadir systolic blood pressure
ORIGINAL ARTICLE
Methodology of Motor Evoked Potentials in a Rabbit Model Stephen D. Waterford 1 & Michelle Rastegar 1 & Erin Goodwin 2 & Paul A. Lapchak 3 & Viviana Juan 1 & Farnaz Haji 1 & René Bombien 1 & Ali Khoynezhad 1
Received: 6 August 2014 / Revised: 22 April 2015 / Accepted: 12 May 2015 / Published online: 20 May 2015 # Springer Science+Business Media New York 2015
Abstract Spinal cord ischemia (SCI) is a devastating complication of aortic operations. Neuromonitoring using motor evoked potentials (MEPs) is a sensitive modality to detect SCI in humans. We describe a leporine SCI model using MEPs to test pharmaceutical therapeutics and other neuroprotective adjuncts. In 80 rabbits, methods to obtain MEPs in normotensive and ischemic rabbits were developed. The effects of isoflurane, propofol, apnea, and hypotension on lower extremity MEPs were studied. Lower extremity MEPs disappear upon SCI induction in 78 of 78 (100 %) rabbits. Prior to SCI induction and during apneic episodes, lower extremity MEPs were lost in all (100 %) and upper extremity MEPs in one (25 %). Isoflurane was used in four experiments, with loss of lower extremity MEPs in all four (100 %) and loss of upper extremity MEPs in zero. With propofol upper extremity, MEPs were obtainable in 80 of 80 rabbits (100 %) and lower extremity MEPs in 78 of 80 rabbits (97.5 %) prior to SCI induction. The presence of these lower extremity MEPs prior to SCI induction was not correlated with systolic or diastolic blood pressure. Disappearance of MEPs occurred in all 45 rabbits with postoperative lower extremity impairment. MEPs in the leporine model correlate closely with paraplegia. MEPs are influenced by inhaled anesthetics and apnea but not by hypotension alone. Propofol anesthesia provides reliable
* Ali Khoynezhad [email protected] 1
Division of Cardiothoracic Surgery, Cedars-Sinai Medical Center, 127 S. San Vicente Blvd., Suite 3306, Los Angeles, CA 90048, USA
2
Comparative Medicine, Cedars-Sinai Medical Center, Los Angeles, CA, USA
3
Department of Neurology and Neurosurgery, Cedars-Sinai Medical Center, Los Angeles, CA, USA
MEPs. This study provides the basis for a reproducible model of SCI to be used for novel therapeutic drug development. Keywords Neuromonitoring . Motor evoked potential . Spinal cord ischemia . Paraplegia . Rabbit . Thoracic aortic
Introduction Spinal cord ischemia (SCI) leading to paraplegia or paraparesis is a devastating complication of aortic surgery. The risk of SCI ranges from 2 to 14 % in endovascular and open surgical treatment of aortic disease [1–3]. To detect SCI, neuromonitoring of transcranial motor evoked potentials (MEPs) is used during aortic surgery [4]. MEP disappearance indicates that ischemia has occurred and prompts corrective action [4]. MEPs record the electrical responses of muscles following stimulation of the motor cortex with scalp electrodes. While MEPs are standard of care in many hospitals, incorporation of MEP in animal models of SCI, while reported, has been scarce [5, 6]. Neuroprotective medications have shown potential to reduce ischemic damage to neuronal tissue, but implementation requires evidence from in vivo models, which can be further improved by incorporating MEP monitoring [7–9]. Reviews of pharmacologic studies of spinal cord protection in the rabbit have noted wide variation in experimental models used to produce SCI, which has limited comparison of the efficacy of various agents [10]. Therefore, establishing a reliable MEP methodology is required, with provision of specific details on anesthetic regimen and physiologic variables required to elicit MEPs in a reproducible fashion. In this study, the effects of cardiopulmonary events and anesthetic medications on MEP findings in a rabbit model are described, and the
400
relationship between MEP findings and postoperative paraplegia is examined. Numerous authors have in fact reported use of MEPs in the rabbit model, but little description of implementation is given. Pokhrel and associates report the effect of reperfusion blood pressure on neurologic outcome in rabbits undergoing spinal cord ischemia [6]. The authors used MEPs, but the process of obtaining baseline MEPs prior to SCI induction was not addressed. Similarly, Murakami and colleagues examined the precise nature of reperfusion MEPs and correlated these results with the motor examination of rabbits after various periods of spinal cord ischemia [11]. While these results provide a method for interpretation of MEPs, they do not focus on establishment of the MEP model itself. We have found that reliable development of neuroprotective medications requires a stable MEP model and an understanding of the factors that influence MEPs. Inability to reliably obtain MEPs can prevent the development of a robust neuroprotective drug development program. Therefore, our objective is to describe the hemodynamic and anesthetic factors, which influence the setup of an MEP model, in order to facilitate the standardization and troubleshooting of such a model.
Materials Eighty New Zealand White rabbits, Oryctolagus cuniculus, between 3 to 4 kg, were obtained from a United States Department of Agriculture class A vendor. Equipment used is as follows: 24-gauge intravenous catheters, 3.0 mm inner diameter endotracheal tubes, and gauze. Three French Fogarty embolectomy catheters were used (Edwards Lifesciences, Irvine, CA, USA). MEP measurements were performed using an intraoperative neuromonitoring machine (16 channels; Cascade, Cadwell, Kennewick, WA, USA) and commercially available electrodes (Rochester Electro-Medical, Lutz, FL, USA). Medications used were ketamine, xylazine, isoflurane, propofol, carprofen, and buprenorphine, obtained from our institutional pharmacy. Pentobarbital and phenytoin euthanasia solution (Euthasol) was obtained from Virbac Animal Health, Fort Worth, TX, USA.
Methods Rabbits received care according to the Guide for the Care and Use of Laboratory Animals [12] and with approval by the Institutional Animal Care and Use Committee. A 24-gauge intravenous catheter was placed in the lateral ear vein, and general anesthesia was induced intravenously with ketamine (10 mg/mL) and xylazine (2 mg/mL) in saline. Rabbits were intubated with direct laryngoscopy with a 3.0 mm inner diameter cuffed endotracheal tube and the cuff inflated. For four
Transl. Stroke Res. (2015) 6:399–406
experiments, anesthesia was maintained with inhaled isoflurane (0.5–3 %) for the duration of the operation. For other operations, anesthesia was maintained with a continuous infusion of propofol, without inhaled anesthetic. For analgesia, rabbits received carprofen 5 mg/kg subcutaneously after induction and 0.01 mg/kg buprenorphine subcutaneously upon awakening. One liter per minute of oxygen was delivered via endotracheal tube. Vitals signs including non-invasive blood pressure were monitored. After induction, the right groin was clipped. After administration of 1 % lidocaine, a 2 cm longitudinal incision was performed 1 cm distal to the palpable common femoral artery pulse, and dissection was carried down to the superficial femoral artery (SFA). After proximal and distal control of SFA, an arteriotomy was made with fine scissors, and a three French Fogarty embolectomy catheter was advanced into the aorta to 18 cm, established on necropsy to be just below the renal arteries. The rabbit spinal column consists of seven lumbar vertebrae, and the spinal cord travels down to the lumbosacral junction. The lumbar, infrarenal portion therefore supplies the lower extremities, analogous to the human thoracic aorta. The balloon was subsequently inflated to initiate aortic occlusion and distal SCI for a period of 6 to 40 min. These occlusion times will produce varying degrees of motor impairment, and investigators using this protocol will need to calibrate the degree of paraplegia obtained with various occlusion times, with their individual source of rabbits and choice of anesthetic regimen. MEPs were measured immediately following SCI induction, and subsequently, at 30-s intervals until loss of lower extremity, MEPs were obtained. Following reperfusion, MEPs were measured for 5 min prior to awakening the animal from general anesthesia. At the conclusion of experiments, euthanasia was performed with 2 mL intravenous pentobarbital and phenytoin (Euthasol). Technical Requirements For measurement of MEPs, we use an intraoperative neuromonitoring machine with a Cascade 16 channel amplifier (Cadwell, Kennewick, WA, USA) equipped with Cascade software version 2.2.45 and commercially available electrodes (Rochester Electro-Medical, Lutz, FL, USA). For the extremities, twisted pair 12 mm subdermal electrodes are used. For the ground electrode, a single 12 mm electrode is used. Electrodes One pair of electrodes is placed in each limb, using the long head of the triceps brachii muscle for the upper extremity and the gracilis muscle for the lower extremity. Upper extremities provide a positive control, because their MEPs are not affected by lumbar SCI [13]. Upper extremities are not shaved; thus, muscles are palpable but not visible. The muscle therefore
Transl. Stroke Res. (2015) 6:399–406
must be firmly grasped, and electrodes placed directly into muscle, avoiding subcutaneous placement or injury of neurovascular structures (Fig. 1). The 12 mm electrode is inserted entirely. For lower extremity electrodes, the midportion of the SFA lies deep to the gracilis muscle, and electrodes are placed in the medial gracilis to avoid the neurovascular bundle (Fig. 2). Extremity electrodes are placed apart from each other by approximately 3 mm, to allow for measurement of a potential between them. Next cranial electrodes are placed subcutaneously tangential to the scalp overlying the primary motor cortex (M1). In the rabbit, M1 is best stimulated with an electrode rostral to the ear, overlying the fronto-scutular muscle [14], in a palpable ridge of tissue between the ear and the eye. The tissue is elevated, and the electrode is inserted parallel to the scalp (Fig. 3). This method uses visible anatomic landmarks, because electrodes cannot be placed according to the 10–20 system in a rabbit [15], as the brain is too small for application of these coordinates. The ground electrode is placed into the subcutaneous tissue of either shoulder. Electrode impedance is checked, and must be below 5000 Ω. High impedance indicates that the electrode has fallen out of the muscle partially or completely. Next, the voltage of the cortical stimulus is set and should range from 60 to 140 V in the rabbit. Given the fact that SCI can increase the voltage threshold for MEPs, we used a voltage of 100 V, high enough to confirm that lower extremity motor potentials had truly disappeared when ischemia was induced. Next, a gauze bite block is placed between the teeth to prevent jaw muscles from contracting and causing selfinjury or damage to the endotracheal tube. Transcranial stimuli are then triggered from each cranial electrode, and MEP tracings recorded.
Fig. 1 Ventral view of rabbit thoracic limb. The long head of the triceps brachii muscle of the extremity is grasped and elevated, and electrodes placed into the muscle (illustration: VJ)
401
Fig. 2 Ventral view of rabbit pelvic limbs. The femoral pulse is palpated, and electrodes placed 2 cm medial to this in the palpable gracilis muscle (illustration: VJ)
Postoperatively, motor function is rated from 0 to 4 using the Tarlov scale [5]. Animals were examined 2 h postoperatively. This scale has been adapted for use in rabbits and has five scores: 0, paraplegic with no lower extremity function; 1, poor lower extremity function, weak antigravity movement only; 2, some lower extremity motor function with good antigravity strength but inability to draw legs under body or hop;
Fig. 3 Scalp electrodes are placed in a ridge of tissue between the ear and the eye, by grasping and elevating this tissue and placing the electrode parallel to the skull, avoiding bone (illustration: VJ)
402
3, ability to draw legs under body and hop but not normally; 4, normal motor function.
Results MEP tracings were obtained prior to SCI induction with use of propofol total intravenous anesthesia (Fig. 4). Normal MEP tracings, obtained with a 100 V transcranial stimulus to one side of the cortex, are shown in Fig. 4. Firing one transcranial electrode generated muscle contraction in all four extremities, owing to spread of current to the contralateral motor cortex. Upper extremity MEPs were of greater magnitude than lower extremity MEPs in 78 of 78 rabbits (100 %), expected from the shorter distance from the cortex to the upper extremity. Lower extremity MEPs were present at the start of the experiment in 78 of 80 rabbits (97.5 %), while upper extremity MEPs were present in 80 of 80 rabbits (100 %). Following SCI induction, within 60 s, lower extremity MEPs disappeared completely in 78 of 78 rabbits (100 %), while upper extremity MEPs were lost in 0 of 80 rabbits (0 %) (Fig. 5, Table 1), establishing this as a highly sensitive model for SCI detection. Absent MEPs are defined as zero amplitude. As hypotension is common under anesthesia and can alter perfusion of the cerebral cortex and descending pathways, we
Fig. 4 Normal pattern of extremity MEPs following stimulation by one cranial electrode at 100 V. The cranial electrode delivers four stimuli (arrows), and MEPs are indicated. Upper extremity MEPs appear with shorter latency than lower extremity MEPs, owing to shorter neuronal distance
Transl. Stroke Res. (2015) 6:399–406
investigated whether hypotension abolishes MEPs. Using a cutoff for systolic blood pressure of 85 mmHg, we examined 30 animals whose lowest systolic blood pressure was at or above 85 mmHg and 47 animals whose lowest systolic blood pressure was below 85 mmHg. Thirty of 30 animals with a nadir systolic blood pressure at or above 85 mmHg had baseline lower extremity MEPs (100 %), while 45 of 47 animals with a nadir systolic blood pressure below 85 mmHg had baseline lower extremity MEPs (95.7 %) (p=0.52, Table 1). Using a cutoff for diastolic blood pressure of 40 mmHg, we examined 33 animals whose lowest diastolic blood pressure was at or above 40 mmHg and 43 animals whose lowest systolic blood pressure was below 40 mmHg. Thirty-three of 33 animals with a nadir diastolic blood pressure at or above 40 mmHg had baseline lower extremity MEPs (100 %), while 41 of 43 animals with a nadir diastolic blood pressure below 40 mmHg had baseline lower extremity MEPs (95.3 %) (p= 0.50, Table 1). While these results are not statistically significant, the two rabbits without MEPs had a systolic pressure below 85 mmHg and a diastolic pressure below 40 mmHg. In particular, the blood pressure of these two animals was 70/20 and 83/37. When the period of SCI was terminated, allowing for spinal cord reperfusion, we continued to monitor MEPs. We have found three reperfusion responses. Lower extremity MEPs can be absent, can return in a normal pattern, or can return
Fig. 5 Pattern of MEPs following SCI induction, showing absence of lower extremity MEPs and preservation of upper extremity MEPs. This pattern may also be seen prior to SCI induction with inadequate stimulatory voltage or with episodes of apnea and isoflurane use
Transl. Stroke Res. (2015) 6:399–406
403
Table 1 Effect of clinical parameters on loss of upper and lower extremity MEPs Clinical Scenario
Loss of upper extremity MEPs
Loss of lower extremity MEPs
Induction of spinal cord ischemia Nadir systolic blood pressure at or above 85 mmHg Nadir systolic blood pressure below 85 mmHg Nadir diastolic blood pressure at or above 40 mmHg Nadir diastolic blood pressure below 40 mmHg Apnea Isoflurane use Propofol total intravenous anesthesia Postoperative lower extremity impairment
0 of 80 (0 %)
78 of 78 (100 %)
0 of 30 (0 %)
0 of 30 (0 %)a
0 of 47 (0 %)
2 of 47 (4.3 %)a
0 of 33 (0 %)
0 of 33 (0 %)b
0 of 43 (0 %)
2 of 43 (4.7 %)b
1 of 4 (25 %) 0 of 4 (0 %) 0 of 74 (0 %)
4 of 4 (100 %)c 4 of 4 (100 %)c 2 of 74 (3.8 %)c
0 of 46 (0 %)
45 of 45 (100 %)
Fisher exact test p=0.52 for nadir systolic blood pressure≥85 mmHg vs. nadir systolic blood pressure
