- Email: [email protected]
Evoked potentials elicited on the cerebellar cortex by electrical stimulation of the rat spinocerebellar tract
Available online at www.sciencedirect.com
Surgical Neurology 72 (2009) 395 – 400 www.surgicalneurology-online.com
Technique
Evoked potentials elicited on the cerebellar cortex by electrical stimulation of the rat spinocerebellar tract Hiroyuki Muramatsu, MD, PhD⁎, Kyouichi Suzuki, MD, Tatsuya Sasaki, MD, Masato Matsumoto, MD, Jun Sakuma, MD, PhD, Masahiro Oinuma, MD, Takeshi Itakura, BS, Namio Kodama, MD, PhD Department of Neurosurgery, Fukushima Medical University, Fukushima 960-1295, Japan Received 13 March 2009; accepted 8 April 2009
Abstract
Background: In the current study, as a first step to develop a monitoring method of cerebellar functions, we tried to record evoked potentials on the cerebellar cortex by electrical stimulation of the rat SCT, which is located in the Inf-CPed. Methods: The experimental study was performed on rats. Unilateral muscular contractions of quadriceps femoris muscle were elicited by electrical stimulation. The evoked potentials were recorded from the surface of the ipsilateral cerebellum and the contralateral primary sensory cortex. Results: The highly reproducible potentials obtained from the ipsilateral cerebellar hemisphere were named SCEP. The SCEP exhibited one negative peak with a latency of 11.7 ± 0.3 milliseconds (N11). Short-latency somatosensory evoked potential was recorded from the contralateral primary sensory cortex with a latency of 19.1 ± 0.6 milliseconds. Coagulation of the ipsilateral Inf-CPed caused disappearance or marked reduction of the SCEP N11, but it did not change the SSEP. On the other hand, sectioning of the ipsilateral dorsal column resulted in the disappearance of the SSEP, but it did not affect the SCEP N11. Conclusions: Reproducible SCEP was recorded from the rat cerebellar hemisphere by electrical stimulation of the quadriceps femoris muscle. We posit that the SCEP differs from the SSEP, which ascends via the dorsal column, and that it is conducted by the dorsal SCT located in the Inf-CPed. Our results suggest that it may be possible to detect the dysfunction of the Inf-CPed electrophysiologically by using SCEP. © 2009 Elsevier Inc. All rights reserved.
Keywords:
Cerebellum; Evoked potential; Inferior cerebellar peduncle; Spinocerebellar tract
1. Introduction Various kinds of intraoperative monitoring have been used in neurosurgery. We have developed additional intraoperative monitoring techniques and applied them in patients undergoing neurosurgical procedures [11,14,17-22].
Abbreviations: Inf-CPed, inferior cerebellar peduncle; Sup-CPed, superior cerebellar peduncle; SCEP, spinocerebellar evoked potential; SCT, spinocerebellar tract; SSEP, short-latency somatosensory evoked potential. ⁎ Corresponding author. Tel.: +81 24 547 1268; fax: +81 24 548 1803. E-mail address: [email protected] (H. Muramatsu). 0090-3019/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2009.04.015
At present, however, no established methods are available for the intraoperative monitoring of cerebellar function. In the current study, as a first step to develop a monitoring method of cerebellar functions, we tried to record potentials that reflect the Inf-CPed function. We focused on the dorsal SCT, which conducts information from muscle spindles and travels to the cerebellum. The evoked potential elicited from the cerebellar surface by electrical stimulation of rat muscle spindles (SCEP) was recorded. Because the dorsal SCT passes through the Inf-CPed, we also examined the possibility of detecting Inf-CPed dysfunction by monitoring the SCEP.
396
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
2. Materials and methods 2.1. Animals and procedures Experiments were performed on 52 male Wistar rats weighing 280 to 320 g. All experimental procedures complied with the guidelines on animal experiments of Fukushima Medical University, and all experimental protocols were approved by the Institutional Animal Care and Use Committee of Fukushima Medical University. After the rats were anesthetized with 1.5% halothane in a mixture of 75% oxygen and 25% nitrous oxide, anesthesia was maintained by injecting 50 mg/kg per hour of propofol (1% Diprivan, Astra Zeneca, Inc, Osaka, Japan) into the femoral artery under spontaneous respiration. Mean arterial blood pressure was continuously monitored. Blood gases were kept in the reference range (PO2, 90-120 mm Hg; PCO2, 35-45 mm Hg). The body temperature was maintained at 37°C to 38°C with a warming blanket and monitored rectally. During the operation, the rats were in prone position. The ipsilateral hind limb was placed in a natural position to avoid limiting muscle contractions. 2.2. Spinocerebellar evoked potential recordings After exposing the quadriceps femoris muscle unilaterally, 2 stimulation needle electrodes (45386, GE Market Medical System, Tokyo, Japan) were inserted 10 mm apart and along the axis of the muscle fibers. The muscle was electrically stimulated at 2 Hz with a 200millisecond rectangular wave pulse using an electric stimulator (DPS-1100D, Dia Medical System, Tokyo, Japan) and an isolator (5384, NEC SAN-ei Instruments Corp, Tokyo, Japan). After suboccipital craniectomy, both cerebellar hemispheres were exposed widely. A silver ball electrode, approximately 1 mm in diameter (45182, GE Market Medical System), was placed as a recording electrode on the posterior lobe of the cerebellum (Fig. 1). The reference electrode was a silver ball electrode inserted subcutaneously in the ipsilateral auricula. Spinocerebellar evoked potential was recorded through a band-pass filter set for 80 to 3000 Hz, and 150 responses were averaged using a signal processor (Synax1100, NEC Co Ltd, Tokyo, Japan). Spinocerebellar evoked potential was analyzed for the following: (1) specificity of the recording site, (2) the effects of electrical stimulation artifacts, (3) the effect of electrical stimulation intensity increase, and (4) the effects of muscle relaxant administration on SCEP amplitude. 2.3. Short-latency somatosensory evoked potential recordings The same electrical stimulation series and characteristics were used for SSEP as for the SCEP, but the stimulus intensity was 3 mA. After exposing the contralateral parietal lobe, a recording silver ball electrode was placed on the hind
Fig. 1. Schematic representation of SCEP recording (left) and dorsal view of the cerebellum (right). The unilateral quadriceps femoris muscle was electrically stimulated. The recording electrode was fixed 3 mm lateral from the midline of the ipsilateral cerebellar lobule.
leg area [7,15,16] of the primary sensory cortex. The reference electrode was a silver ball electrode introduced into the ipsilateral auricula. 2.4. Changes in SCEP and SSEP as a result of destruction of the conduction pathways To be able to track the conduction pathways of the SCEP, the ipsilateral dorsal column was sectioned in 4 rats and the ipsilateral Inf-CPed was electrically coagulated in 8 rats. Because of the possible contribution of the contralateral ventral SCT in the conduction of the SCEP via the contralateral Sup-CPed to the ipsilateral cerebellar hemisphere, the contralateral Sup-CPed was coagulated in 4 rats [15,16]. Changes in SCEP and SSEP were examined before and after destruction of these conduction pathways. Sectioning of the ipsilateral dorsal column was done with a sharp knife at the level of the foramen magnum. For electric coagulation of the ipsilateral Inf-CPed or the contralateral Sup-CPed, we applied a bipolar needle coagulator with a 1.5-mm tip attached to an endoscope (HZ-1005B, Machida Corp, Tokyo, Japan). The needle was inserted stereotactically from the cerebellar surface with a micromanipulator (SR-5, Narishige Corp, Tokyo, Japan). To coagulate the ipsilateral Inf-CPed, the needle electrode was inserted from the cerebellar surface (15 mm caudal from the bregma and 3-4 mm lateral from the midline) to the target (9 mm caudal and 7 mm ventral from the bregma and 34 mm lateral from the midline) [15,16]. For electrical coagulation of the contralateral Sup-CPed, the needle was inserted from a point (15 mm caudal from the bregma and 23 mm lateral from the midline) to the target (9 mm caudal and 5 mm ventral from the bregma and 2-4 mm lateral from the midline) [15,16]. Each coagulation was performed with 30 W for 3 seconds. After these experiments, the rats were sacrificed; their brains were fixed in 10% formalin, and the lesions were histologically confirmed for location and size.
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
397
3. Results 3.1. Spinocerebellar evoked potential recordings The specificity of the recording site was examined in 6 rats. A recording electrode was placed at 9 sites on the ipsilateral cerebellar surface and at 1 site on the contralateral cerebellar hemisphere on and around lobule VI (Fig. 2, left). Reproducible potentials were obtained from all 10 sites (Fig. 2, right). A monophasic negative wave with a peak latency of approximately 11 milliseconds was obtained. This peak was defined as N11 in this study. The largest N11 amplitude was obtained 3 mm lateral from the midline on the ipsilateral lobule VI (Fig. 2-3). In 3 of the 6 rats, a monophasic positive wave with a peak latency of 11 milliseconds (P11) was recorded 2 mm lateral from the midline on the ipsilateral lobule VIII (Fig. 2-9). The recording electrode was fixed 3 mm lateral from the midline of ipsilateral lobule VI at the site where the largest N11 amplitude was obtained (identified as 3 on Fig. 2). A highly reproducible negative wave was recorded in all 20 rats studied. The negative peak latency (N11) was 11.7 ± 0.3 milliseconds (mean ± SE) and the amplitude was 16.9 ± 4.9 μV (mean ± SE). To examine the effect of electrical stimulation artifacts, we recorded SCEP and potentials from other sites simultaneously in 3 rats. Electrodes placed on the ipsilateral paravertebral muscle, the posterior cervical muscle, and the ipsilateral forelimb did not record any potentials. We also studied the effect of stimulus intensity in 5 rats. The stimulation threshold ranged from 0.5 to 0.9 mA. The N11 amplitude increased in parallel with stimulation intensity until it reached a plateau at 3 mA (the supramaximal stimulus) (Fig. 3). Lastly, we assessed the effect of a muscle relaxant in 4 rats. Spinocerebellar evoked potentials were recorded before
Fig. 2. Map of the evoked potentials recorded from the cerebellar surface. A recording electrode was placed at 9 sites on the ipsilateral cerebellum on and around lobule VI; one electrode was placed on contralateral lobule VI. N11 and P11, negative and positive peaks, respectively, that appeared 11 milliseconds after stimulation onset.
Fig. 3. Relationship between stimulus intensity and the amplitude of N11. The N11 amplitude increased with stimulus intensity, reaching a plateau at 3 mA.
(control) and 5 and 10 minutes after the intravenous injection of 0.08 mg/kg vecuronium bromide (Musculax, Sankyo Co Ltd, Tokyo, Japan). At 5 minutes postinjection, the N11 amplitude decreased to 13% ± 8% (mean ± SE) and 10 minutes later increased to 80% ± 17% (mean ± SE) of the control level. 3.2. Short-latency somatosensory evoked potential recordings Highly reproducible SSEPs were obtained in all 10 rats. The positive-peak latency was 19.1 ± 0.6 milliseconds (mean ± SE) and the amplitude was 35.5 ± 7.9 μV (mean ± SE). We defined this peak as P19. 3.3. Changes in SCEP and SSEP as a result of the destruction of the conduction pathways In 4 rats we examined the effects of sectioning the ipsilateral dorsal column on the SCEP and SSEP. Sectioning, which was histologically confirmed in all 4 rats (Fig. 4, right), did not produce a change in the amplitude and latency of SCEP N11; however, SSEP P19 disappeared in all 4 rats (Fig. 4 left). The effect of electrical coagulation of the ipsilateral Inf-CPed, which was histologically confirmed, was examined in 8 rats (Fig. 5, right). Introduction of the needle into the target produced no changes in the SCEP and SSEP. After electrical coagulation of the Inf-CPed, N11 disappeared in 4 of the 8 rats. In the other 4 animals, the N11 amplitude decreased to 8% ± 5% (mean ± SE) of the control level (Fig. 5 left). The SSEP did not change in any of these 8 rats. We subjected 4 rats to electrical coagulation of the contralateral Sup-CPed and observed no changes in the SCEP and SSEP (Fig. 6, left). The coagulations in the contralateral Sup-CPed were confirmed histologically (Fig. 6, right).
398
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
4. Discussion Iatrogenic injury of the Inf-CPed during neurosurgical procedures is a possibility in cases of tumor infiltration and blood flow affection in the vertebral or posterior-inferior cerebellar arteries [1]. The resulting neurologic deficit may significantly impair the quality of life, especially if gait disturbance appears. Unfortunately, reliable methods of monitoring the Inf-CPed and cerebellar function in general have not been developed. Several electrophysiological techniques to evaluate cerebellar function have been already reported [2,5,6,8,13]. Dow and Anderson [8] recorded potentials from the surface of the rat cerebellum evoked by touching the hair or tapping the tendon of the extremities. Fu et al. [9] recorded the potential of cerebellar Purkinje cells evoked by moving the extremities of a monkey. The experimental protocols, regardless of the common principle of producing an evoked response from cerebellar structures as a result of sensory stimulation, are not applicable for monitoring because their design was intending to prove neuroanatomical and neurophysiologic characteristics of the cerebellum. Lavy [12], who recorded evoked electromyograms of the gastrocnemius muscle and evoked potentials of the peripheral nerve by applying electrical stimulation to the surface of human cerebellum, concluded that the potentials originated from the extrapyramidal system. Hurlbert et al [10] subsequently studied the evoked potentials reported by Lavy by examining their changes after sectioning of the rat dorsal column and lateral funiculus. Although their results suggested the possibility of monitoring the ventral spinal cord, they found it difficult to detect cerebellar disorders
because extrapyramidal cells are distributed widely over the cerebellar surface and because there are many routes from the cerebellum to the spinal cord. We recorded the evoked potential conducted through the dorsal SCT ascends via the Inf-CPed and terminates in the cerebellum and examined the feasibility of using the SCEP for intraoperative Inf-CPed function monitoring. Transient muscle contractions stimulate the muscle spindles and Golgi tendon organs. The impulse originating in the muscle ascends via the ipsilateral dorsal SCT and contralateral ventral SCT, reaching the ipsilateral cerebellum via the ipsilateral Inf-CPed and contralateral SupCPed, respectively [15,16]. The dorsal SCT is primarily involved in carrying sensory inputs derived from peripheral organs such as muscles, joints, and skin to the cerebellum, whereas the ventral SCT carries peripheral inputs and integrates the activity of interneuron groups of the spinal cord [3,4]. As our attempts to record potentials from other sites over the body failed, we concluded that the SCEP was not an artifact attributable to body movement after electric stimulation and that it was elicited by muscle contraction because the N11 amplitude decreased after the administration of a muscle relaxant and recovered after the restoration of muscle contraction. The amplitude of N11 increased with stimulus intensity, reaching a plateau at 3 mA. Because almost all fibers of the quadriceps femoris muscle contract at 3 mA, we used this as the supramaximal stimulus intensity in our experiments. In humans, dorsal SCT fibers project onto lobules VI and VII of the cerebellar hemisphere, onto lobules II to V of the vermis, and onto lobule VIII of the posterior lobe via the InfCPed [16,23]. In rats, as in humans, the dorsal SCT fibers
Fig. 4. Left: Changes in SCEP and SSEP after sectioning of the ipsilateral dorsal column. After sectioning of the dorsal column, the SSEP disappeared. There was no change in the SCEP. Right: Photomicrograph after sectioning of the ipsilateral dorsal column. The diagram shows the rat brainstem at the foramen magnum level. The heavy dashed line indicates the area of sectioning and the filled-in area is the dorsal column (hematoxylin and eosin, original magnification ×10). The diagram is from Paxinous and Watson [15].
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
399
Fig. 5. Left: Changes in SCEP and SSEP produced by electric coagulation of the Inf-CPed. After coagulation, the SCEP disappeared. There was no change in the SSEP. Right: Photomicrograph after coagulation of the ipsilatetal Inf-CPed. The diagram shows a coronal section of the rat brainstem. The dashed line indicates the area of coagulation and the filled-in area is the Inf-CPed (Kluver-Barrera stain, original magnification ×40). The diagram is from Paxinous and Watson [15].
projects widely to the anterior and posterior lobes of the ipsilateral cerebellum via the ipsilateral Inf-CPed. In our examination of the specificity of the SCEP recording site, we studied only the posterior lobe because
the thin cortical bone in this region provides easy access. We obtained highly reproducible potentials at the chosen recording site, which is an area usually exposed in cerebellar surgery. Our findings suggest that SCEP recording may be
Fig. 6. Left: Changes in SCEP and SSEP after coagulation of the Sup-CPed. Coagulation produced no changes in the SSEP and SCEP. Right: Photomicrograph after coagulation of the contralateral Sup-CPed. The diagram shows a coronal section of the rat brainstem. The dashed line indicates the area of coagulation and the filled-in area indicates the Sup-CPed (Kluver-Barrera stain, original magnification ×40). The diagram is from Paxinous and Watson [15].
400
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
clinically applicable. We recorded P11 from lobule VIII in the caudal portion of the cerebellum. This suggests that the SCEP ascends from deep portions of the brain to the cerebellar surface of lobule VIII. Because we did not examine the SCEP origin, we cannot draw definitive conclusions. Because there are anatomical fiber tracts other than the ipsilateral dorsal SCT, we cannot exclude the possibility that the SCEP we recorded was derived from the ventral SCT. However, based on our findings after coagulation of the contralateral Sup-CPed, it appears unlikely that the SupCPed contributed to any significant degree. Our results suggest that the SCEP elicited by stimulation of the rat quadriceps femoris muscle and recorded from the cerebellar surface originated from stimulation of muscle spindles and the Golgi tendon organ, ascended the dorsal SCT, and reached the cerebellum via the Inf-CPed. 5. Conclusions Reproducible SCEPs were recorded from the rat cerebellar hemisphere by electric muscle stimulation. The SCEP did not change after sectioning the ipsilateral dorsal column and coagulation of the contralateral Sup-CPed, but it disappeared after coagulation of the ipsilateral Inf-CPed. We conclude that the ascent was via the dorsal SCT and that the SCEP reached the cerebellum via the Inf-CPed. Our results suggest that the SCEP monitoring might be intraoperatively applicable to evaluate the functional integrity of the Inf-CPed of patients undergoing posterior fossa surgery. References [1] Amarenco P. The spectrum of cerebellar infarctions. Neurology 1991; 41:973-9. [2] Armstrong DM, Drew T. Response in the posterior lobe of the rat cerebellum to electrical stimulation of cutaneous afferents to the snout. J Physiol 1980;309:357-74. [3] Arshawsky YI, Gerkinbilt MB, Fukson OI, et al. Recordings of neurons of the dorsal spinocerebellar tract during evoked locomotion. Brain Res 1972;43:272-5. [4] Arshawsky YI, Gerkinbilt MB, Fukson OI, et al. Origin of modulation in neurons of the ventral spinocerebellar tract during locomotion. Brain Res 1972;43:276-9. [5] Bremar F. Cerebral and cerebellar potentials. Physiol Rev 1958;38: 357-88. [6] Brookhart JM, Moruzzi G, Snider RS. Origin of cerebellar waves. J Neurophysiol 1951;14:181-90. [7] Donoghue JP, Kerman KL, Ebner FF. Evidence for two organizational plans within the somatic sensory-motor cortex of the rat. J Comp Neurol 1979;183:647-63. [8] Dow RS, Anderson R. Cerebellar action potentials in response to stimulation of proprioceptors and exteroceptors in the rat. J Neurophysiol 1942;5:121-36. [9] Fu QG, Flament D, Coltz JD, et al. Relationship of cerebellar Purkinje cell simple spike discharge to movement kinematics in the monkey. J Neurophysiol 1997;78:478-91.
[10] Hurlbert RJ, Tator CH, Fehlings MG. Evoked potentials from direct cerebellar stimulation for monitoring of the rodent spinal cord. J Neurosurg 1992;76:280-91. [11] Kikuchi Y, Sasaki T, Matsumoto M, et al. Optic nerve evoked potentials elicited by electrical stimulation. Neurol Med Chir (Tokyo) 2005;45:349-55. [12] Lavy WJ. Clinical experience with motor and cerebellar evoked potential monitoring. Neurosurgery 1987;20:169-82. [13] Mares P. Development of cerebellar somatosensory evoked potentials in the rat. Acta Biol Med Germ 1976;35:55-62. [14] Oikawa T, Matsumoto M, Sasaki T, et al. Experimental study of medullary trigeminal evoked potentials: Development of a new method of intraoperative monitoring of the medulla oblongata. J Neurosurg 2000;93:68-76. [15] Paxinous G, Watson C. The rat brain in stereotaxic coordinates, ed 4. San Diego: Academic press; 1998. [16] Paxinous G. The rat nervous system, ed 4. Sydney: Academic press; 1995. [17] Sakuma J, Matsumoto M, Ohta M, et al. Glossopharyngeal evoked potentials after stimulation of the posterior part of the tongue in dogs. Neurosurgery 2002;51:1026-33. [18] Sakuma J, Suzuki K, Sasaki T, et al. Monitoring and preventing blood flow insufficiency due to clip rotation after the treatment of internal carotid artery aneurysms. J Neurosurg 2004;100:960-2. [19] Sasaki T, Kodama N, Matsumoto M, et al. Blood flow disturbance in perforating arteries attributable to aneurysm surgery. J Neurosurg 2007;107:60-7. [20] Sasaki T, Suzuki K, Matsumoto M, et al. Origins of surface potentials evoked by electrical stimulation of oculomotor nerves: Are they related to electrooculographic or electromyographic events? J Neurosurg 2002;97:941-4. [21] Sato M, Kodama N, Sasaki T, et al. Olfactory evoked potentials: experimental and clinical studies. J Neurosurg 1996;85:1122-6. [22] Suzuki K, Kodama N, Sasaki T, et al. Intraoperative monitoring of blood flow insufficiency in the anterior choroidal artery during aneurysm surgery. J Neurosurg 2003;98:507-14. [23] Yamada J, Shirao K, Kitamura T. Trajectory of spinocerebellar fibers passing through the inferior and superior cerebellar peduncles in the rat spinal cord: A study using horseradish peroxidase with pedunculotomy. J Comp Neurol 1991;304:147-60.
Commentary This report provides a basis for testing the spinocerebellar pathways. This may have a human intraoperative correlate in the future. The authors have presented several control studies to show that they indeed did record from the spinocerebellar pathway, not just a far field recording of a lemniscal sensory pathway. I expect others to explore how this may be extended and applied to monitoring the inferior cerebellar peduncle or spinal cord during surgeries that place those structures at risk. Such a human monitoring technique may require recording from subdural strip electrodes placed over the lateral cerebellum. Marc Nuwer, MD, PhD David Geffen School of Medicine at UCLA Los Angeles, CA 90095, USA E-mail address: [email protected]
Surgical Neurology 72 (2009) 395 – 400 www.surgicalneurology-online.com
Technique
Evoked potentials elicited on the cerebellar cortex by electrical stimulation of the rat spinocerebellar tract Hiroyuki Muramatsu, MD, PhD⁎, Kyouichi Suzuki, MD, Tatsuya Sasaki, MD, Masato Matsumoto, MD, Jun Sakuma, MD, PhD, Masahiro Oinuma, MD, Takeshi Itakura, BS, Namio Kodama, MD, PhD Department of Neurosurgery, Fukushima Medical University, Fukushima 960-1295, Japan Received 13 March 2009; accepted 8 April 2009
Abstract
Background: In the current study, as a first step to develop a monitoring method of cerebellar functions, we tried to record evoked potentials on the cerebellar cortex by electrical stimulation of the rat SCT, which is located in the Inf-CPed. Methods: The experimental study was performed on rats. Unilateral muscular contractions of quadriceps femoris muscle were elicited by electrical stimulation. The evoked potentials were recorded from the surface of the ipsilateral cerebellum and the contralateral primary sensory cortex. Results: The highly reproducible potentials obtained from the ipsilateral cerebellar hemisphere were named SCEP. The SCEP exhibited one negative peak with a latency of 11.7 ± 0.3 milliseconds (N11). Short-latency somatosensory evoked potential was recorded from the contralateral primary sensory cortex with a latency of 19.1 ± 0.6 milliseconds. Coagulation of the ipsilateral Inf-CPed caused disappearance or marked reduction of the SCEP N11, but it did not change the SSEP. On the other hand, sectioning of the ipsilateral dorsal column resulted in the disappearance of the SSEP, but it did not affect the SCEP N11. Conclusions: Reproducible SCEP was recorded from the rat cerebellar hemisphere by electrical stimulation of the quadriceps femoris muscle. We posit that the SCEP differs from the SSEP, which ascends via the dorsal column, and that it is conducted by the dorsal SCT located in the Inf-CPed. Our results suggest that it may be possible to detect the dysfunction of the Inf-CPed electrophysiologically by using SCEP. © 2009 Elsevier Inc. All rights reserved.
Keywords:
Cerebellum; Evoked potential; Inferior cerebellar peduncle; Spinocerebellar tract
1. Introduction Various kinds of intraoperative monitoring have been used in neurosurgery. We have developed additional intraoperative monitoring techniques and applied them in patients undergoing neurosurgical procedures [11,14,17-22].
Abbreviations: Inf-CPed, inferior cerebellar peduncle; Sup-CPed, superior cerebellar peduncle; SCEP, spinocerebellar evoked potential; SCT, spinocerebellar tract; SSEP, short-latency somatosensory evoked potential. ⁎ Corresponding author. Tel.: +81 24 547 1268; fax: +81 24 548 1803. E-mail address: [email protected] (H. Muramatsu). 0090-3019/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2009.04.015
At present, however, no established methods are available for the intraoperative monitoring of cerebellar function. In the current study, as a first step to develop a monitoring method of cerebellar functions, we tried to record potentials that reflect the Inf-CPed function. We focused on the dorsal SCT, which conducts information from muscle spindles and travels to the cerebellum. The evoked potential elicited from the cerebellar surface by electrical stimulation of rat muscle spindles (SCEP) was recorded. Because the dorsal SCT passes through the Inf-CPed, we also examined the possibility of detecting Inf-CPed dysfunction by monitoring the SCEP.
396
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
2. Materials and methods 2.1. Animals and procedures Experiments were performed on 52 male Wistar rats weighing 280 to 320 g. All experimental procedures complied with the guidelines on animal experiments of Fukushima Medical University, and all experimental protocols were approved by the Institutional Animal Care and Use Committee of Fukushima Medical University. After the rats were anesthetized with 1.5% halothane in a mixture of 75% oxygen and 25% nitrous oxide, anesthesia was maintained by injecting 50 mg/kg per hour of propofol (1% Diprivan, Astra Zeneca, Inc, Osaka, Japan) into the femoral artery under spontaneous respiration. Mean arterial blood pressure was continuously monitored. Blood gases were kept in the reference range (PO2, 90-120 mm Hg; PCO2, 35-45 mm Hg). The body temperature was maintained at 37°C to 38°C with a warming blanket and monitored rectally. During the operation, the rats were in prone position. The ipsilateral hind limb was placed in a natural position to avoid limiting muscle contractions. 2.2. Spinocerebellar evoked potential recordings After exposing the quadriceps femoris muscle unilaterally, 2 stimulation needle electrodes (45386, GE Market Medical System, Tokyo, Japan) were inserted 10 mm apart and along the axis of the muscle fibers. The muscle was electrically stimulated at 2 Hz with a 200millisecond rectangular wave pulse using an electric stimulator (DPS-1100D, Dia Medical System, Tokyo, Japan) and an isolator (5384, NEC SAN-ei Instruments Corp, Tokyo, Japan). After suboccipital craniectomy, both cerebellar hemispheres were exposed widely. A silver ball electrode, approximately 1 mm in diameter (45182, GE Market Medical System), was placed as a recording electrode on the posterior lobe of the cerebellum (Fig. 1). The reference electrode was a silver ball electrode inserted subcutaneously in the ipsilateral auricula. Spinocerebellar evoked potential was recorded through a band-pass filter set for 80 to 3000 Hz, and 150 responses were averaged using a signal processor (Synax1100, NEC Co Ltd, Tokyo, Japan). Spinocerebellar evoked potential was analyzed for the following: (1) specificity of the recording site, (2) the effects of electrical stimulation artifacts, (3) the effect of electrical stimulation intensity increase, and (4) the effects of muscle relaxant administration on SCEP amplitude. 2.3. Short-latency somatosensory evoked potential recordings The same electrical stimulation series and characteristics were used for SSEP as for the SCEP, but the stimulus intensity was 3 mA. After exposing the contralateral parietal lobe, a recording silver ball electrode was placed on the hind
Fig. 1. Schematic representation of SCEP recording (left) and dorsal view of the cerebellum (right). The unilateral quadriceps femoris muscle was electrically stimulated. The recording electrode was fixed 3 mm lateral from the midline of the ipsilateral cerebellar lobule.
leg area [7,15,16] of the primary sensory cortex. The reference electrode was a silver ball electrode introduced into the ipsilateral auricula. 2.4. Changes in SCEP and SSEP as a result of destruction of the conduction pathways To be able to track the conduction pathways of the SCEP, the ipsilateral dorsal column was sectioned in 4 rats and the ipsilateral Inf-CPed was electrically coagulated in 8 rats. Because of the possible contribution of the contralateral ventral SCT in the conduction of the SCEP via the contralateral Sup-CPed to the ipsilateral cerebellar hemisphere, the contralateral Sup-CPed was coagulated in 4 rats [15,16]. Changes in SCEP and SSEP were examined before and after destruction of these conduction pathways. Sectioning of the ipsilateral dorsal column was done with a sharp knife at the level of the foramen magnum. For electric coagulation of the ipsilateral Inf-CPed or the contralateral Sup-CPed, we applied a bipolar needle coagulator with a 1.5-mm tip attached to an endoscope (HZ-1005B, Machida Corp, Tokyo, Japan). The needle was inserted stereotactically from the cerebellar surface with a micromanipulator (SR-5, Narishige Corp, Tokyo, Japan). To coagulate the ipsilateral Inf-CPed, the needle electrode was inserted from the cerebellar surface (15 mm caudal from the bregma and 3-4 mm lateral from the midline) to the target (9 mm caudal and 7 mm ventral from the bregma and 34 mm lateral from the midline) [15,16]. For electrical coagulation of the contralateral Sup-CPed, the needle was inserted from a point (15 mm caudal from the bregma and 23 mm lateral from the midline) to the target (9 mm caudal and 5 mm ventral from the bregma and 2-4 mm lateral from the midline) [15,16]. Each coagulation was performed with 30 W for 3 seconds. After these experiments, the rats were sacrificed; their brains were fixed in 10% formalin, and the lesions were histologically confirmed for location and size.
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
397
3. Results 3.1. Spinocerebellar evoked potential recordings The specificity of the recording site was examined in 6 rats. A recording electrode was placed at 9 sites on the ipsilateral cerebellar surface and at 1 site on the contralateral cerebellar hemisphere on and around lobule VI (Fig. 2, left). Reproducible potentials were obtained from all 10 sites (Fig. 2, right). A monophasic negative wave with a peak latency of approximately 11 milliseconds was obtained. This peak was defined as N11 in this study. The largest N11 amplitude was obtained 3 mm lateral from the midline on the ipsilateral lobule VI (Fig. 2-3). In 3 of the 6 rats, a monophasic positive wave with a peak latency of 11 milliseconds (P11) was recorded 2 mm lateral from the midline on the ipsilateral lobule VIII (Fig. 2-9). The recording electrode was fixed 3 mm lateral from the midline of ipsilateral lobule VI at the site where the largest N11 amplitude was obtained (identified as 3 on Fig. 2). A highly reproducible negative wave was recorded in all 20 rats studied. The negative peak latency (N11) was 11.7 ± 0.3 milliseconds (mean ± SE) and the amplitude was 16.9 ± 4.9 μV (mean ± SE). To examine the effect of electrical stimulation artifacts, we recorded SCEP and potentials from other sites simultaneously in 3 rats. Electrodes placed on the ipsilateral paravertebral muscle, the posterior cervical muscle, and the ipsilateral forelimb did not record any potentials. We also studied the effect of stimulus intensity in 5 rats. The stimulation threshold ranged from 0.5 to 0.9 mA. The N11 amplitude increased in parallel with stimulation intensity until it reached a plateau at 3 mA (the supramaximal stimulus) (Fig. 3). Lastly, we assessed the effect of a muscle relaxant in 4 rats. Spinocerebellar evoked potentials were recorded before
Fig. 2. Map of the evoked potentials recorded from the cerebellar surface. A recording electrode was placed at 9 sites on the ipsilateral cerebellum on and around lobule VI; one electrode was placed on contralateral lobule VI. N11 and P11, negative and positive peaks, respectively, that appeared 11 milliseconds after stimulation onset.
Fig. 3. Relationship between stimulus intensity and the amplitude of N11. The N11 amplitude increased with stimulus intensity, reaching a plateau at 3 mA.
(control) and 5 and 10 minutes after the intravenous injection of 0.08 mg/kg vecuronium bromide (Musculax, Sankyo Co Ltd, Tokyo, Japan). At 5 minutes postinjection, the N11 amplitude decreased to 13% ± 8% (mean ± SE) and 10 minutes later increased to 80% ± 17% (mean ± SE) of the control level. 3.2. Short-latency somatosensory evoked potential recordings Highly reproducible SSEPs were obtained in all 10 rats. The positive-peak latency was 19.1 ± 0.6 milliseconds (mean ± SE) and the amplitude was 35.5 ± 7.9 μV (mean ± SE). We defined this peak as P19. 3.3. Changes in SCEP and SSEP as a result of the destruction of the conduction pathways In 4 rats we examined the effects of sectioning the ipsilateral dorsal column on the SCEP and SSEP. Sectioning, which was histologically confirmed in all 4 rats (Fig. 4, right), did not produce a change in the amplitude and latency of SCEP N11; however, SSEP P19 disappeared in all 4 rats (Fig. 4 left). The effect of electrical coagulation of the ipsilateral Inf-CPed, which was histologically confirmed, was examined in 8 rats (Fig. 5, right). Introduction of the needle into the target produced no changes in the SCEP and SSEP. After electrical coagulation of the Inf-CPed, N11 disappeared in 4 of the 8 rats. In the other 4 animals, the N11 amplitude decreased to 8% ± 5% (mean ± SE) of the control level (Fig. 5 left). The SSEP did not change in any of these 8 rats. We subjected 4 rats to electrical coagulation of the contralateral Sup-CPed and observed no changes in the SCEP and SSEP (Fig. 6, left). The coagulations in the contralateral Sup-CPed were confirmed histologically (Fig. 6, right).
398
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
4. Discussion Iatrogenic injury of the Inf-CPed during neurosurgical procedures is a possibility in cases of tumor infiltration and blood flow affection in the vertebral or posterior-inferior cerebellar arteries [1]. The resulting neurologic deficit may significantly impair the quality of life, especially if gait disturbance appears. Unfortunately, reliable methods of monitoring the Inf-CPed and cerebellar function in general have not been developed. Several electrophysiological techniques to evaluate cerebellar function have been already reported [2,5,6,8,13]. Dow and Anderson [8] recorded potentials from the surface of the rat cerebellum evoked by touching the hair or tapping the tendon of the extremities. Fu et al. [9] recorded the potential of cerebellar Purkinje cells evoked by moving the extremities of a monkey. The experimental protocols, regardless of the common principle of producing an evoked response from cerebellar structures as a result of sensory stimulation, are not applicable for monitoring because their design was intending to prove neuroanatomical and neurophysiologic characteristics of the cerebellum. Lavy [12], who recorded evoked electromyograms of the gastrocnemius muscle and evoked potentials of the peripheral nerve by applying electrical stimulation to the surface of human cerebellum, concluded that the potentials originated from the extrapyramidal system. Hurlbert et al [10] subsequently studied the evoked potentials reported by Lavy by examining their changes after sectioning of the rat dorsal column and lateral funiculus. Although their results suggested the possibility of monitoring the ventral spinal cord, they found it difficult to detect cerebellar disorders
because extrapyramidal cells are distributed widely over the cerebellar surface and because there are many routes from the cerebellum to the spinal cord. We recorded the evoked potential conducted through the dorsal SCT ascends via the Inf-CPed and terminates in the cerebellum and examined the feasibility of using the SCEP for intraoperative Inf-CPed function monitoring. Transient muscle contractions stimulate the muscle spindles and Golgi tendon organs. The impulse originating in the muscle ascends via the ipsilateral dorsal SCT and contralateral ventral SCT, reaching the ipsilateral cerebellum via the ipsilateral Inf-CPed and contralateral SupCPed, respectively [15,16]. The dorsal SCT is primarily involved in carrying sensory inputs derived from peripheral organs such as muscles, joints, and skin to the cerebellum, whereas the ventral SCT carries peripheral inputs and integrates the activity of interneuron groups of the spinal cord [3,4]. As our attempts to record potentials from other sites over the body failed, we concluded that the SCEP was not an artifact attributable to body movement after electric stimulation and that it was elicited by muscle contraction because the N11 amplitude decreased after the administration of a muscle relaxant and recovered after the restoration of muscle contraction. The amplitude of N11 increased with stimulus intensity, reaching a plateau at 3 mA. Because almost all fibers of the quadriceps femoris muscle contract at 3 mA, we used this as the supramaximal stimulus intensity in our experiments. In humans, dorsal SCT fibers project onto lobules VI and VII of the cerebellar hemisphere, onto lobules II to V of the vermis, and onto lobule VIII of the posterior lobe via the InfCPed [16,23]. In rats, as in humans, the dorsal SCT fibers
Fig. 4. Left: Changes in SCEP and SSEP after sectioning of the ipsilateral dorsal column. After sectioning of the dorsal column, the SSEP disappeared. There was no change in the SCEP. Right: Photomicrograph after sectioning of the ipsilateral dorsal column. The diagram shows the rat brainstem at the foramen magnum level. The heavy dashed line indicates the area of sectioning and the filled-in area is the dorsal column (hematoxylin and eosin, original magnification ×10). The diagram is from Paxinous and Watson [15].
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
399
Fig. 5. Left: Changes in SCEP and SSEP produced by electric coagulation of the Inf-CPed. After coagulation, the SCEP disappeared. There was no change in the SSEP. Right: Photomicrograph after coagulation of the ipsilatetal Inf-CPed. The diagram shows a coronal section of the rat brainstem. The dashed line indicates the area of coagulation and the filled-in area is the Inf-CPed (Kluver-Barrera stain, original magnification ×40). The diagram is from Paxinous and Watson [15].
projects widely to the anterior and posterior lobes of the ipsilateral cerebellum via the ipsilateral Inf-CPed. In our examination of the specificity of the SCEP recording site, we studied only the posterior lobe because
the thin cortical bone in this region provides easy access. We obtained highly reproducible potentials at the chosen recording site, which is an area usually exposed in cerebellar surgery. Our findings suggest that SCEP recording may be
Fig. 6. Left: Changes in SCEP and SSEP after coagulation of the Sup-CPed. Coagulation produced no changes in the SSEP and SCEP. Right: Photomicrograph after coagulation of the contralateral Sup-CPed. The diagram shows a coronal section of the rat brainstem. The dashed line indicates the area of coagulation and the filled-in area indicates the Sup-CPed (Kluver-Barrera stain, original magnification ×40). The diagram is from Paxinous and Watson [15].
400
H. Muramatsu et al. / Surgical Neurology 72 (2009) 395–400
clinically applicable. We recorded P11 from lobule VIII in the caudal portion of the cerebellum. This suggests that the SCEP ascends from deep portions of the brain to the cerebellar surface of lobule VIII. Because we did not examine the SCEP origin, we cannot draw definitive conclusions. Because there are anatomical fiber tracts other than the ipsilateral dorsal SCT, we cannot exclude the possibility that the SCEP we recorded was derived from the ventral SCT. However, based on our findings after coagulation of the contralateral Sup-CPed, it appears unlikely that the SupCPed contributed to any significant degree. Our results suggest that the SCEP elicited by stimulation of the rat quadriceps femoris muscle and recorded from the cerebellar surface originated from stimulation of muscle spindles and the Golgi tendon organ, ascended the dorsal SCT, and reached the cerebellum via the Inf-CPed. 5. Conclusions Reproducible SCEPs were recorded from the rat cerebellar hemisphere by electric muscle stimulation. The SCEP did not change after sectioning the ipsilateral dorsal column and coagulation of the contralateral Sup-CPed, but it disappeared after coagulation of the ipsilateral Inf-CPed. We conclude that the ascent was via the dorsal SCT and that the SCEP reached the cerebellum via the Inf-CPed. Our results suggest that the SCEP monitoring might be intraoperatively applicable to evaluate the functional integrity of the Inf-CPed of patients undergoing posterior fossa surgery. References [1] Amarenco P. The spectrum of cerebellar infarctions. Neurology 1991; 41:973-9. [2] Armstrong DM, Drew T. Response in the posterior lobe of the rat cerebellum to electrical stimulation of cutaneous afferents to the snout. J Physiol 1980;309:357-74. [3] Arshawsky YI, Gerkinbilt MB, Fukson OI, et al. Recordings of neurons of the dorsal spinocerebellar tract during evoked locomotion. Brain Res 1972;43:272-5. [4] Arshawsky YI, Gerkinbilt MB, Fukson OI, et al. Origin of modulation in neurons of the ventral spinocerebellar tract during locomotion. Brain Res 1972;43:276-9. [5] Bremar F. Cerebral and cerebellar potentials. Physiol Rev 1958;38: 357-88. [6] Brookhart JM, Moruzzi G, Snider RS. Origin of cerebellar waves. J Neurophysiol 1951;14:181-90. [7] Donoghue JP, Kerman KL, Ebner FF. Evidence for two organizational plans within the somatic sensory-motor cortex of the rat. J Comp Neurol 1979;183:647-63. [8] Dow RS, Anderson R. Cerebellar action potentials in response to stimulation of proprioceptors and exteroceptors in the rat. J Neurophysiol 1942;5:121-36. [9] Fu QG, Flament D, Coltz JD, et al. Relationship of cerebellar Purkinje cell simple spike discharge to movement kinematics in the monkey. J Neurophysiol 1997;78:478-91.
[10] Hurlbert RJ, Tator CH, Fehlings MG. Evoked potentials from direct cerebellar stimulation for monitoring of the rodent spinal cord. J Neurosurg 1992;76:280-91. [11] Kikuchi Y, Sasaki T, Matsumoto M, et al. Optic nerve evoked potentials elicited by electrical stimulation. Neurol Med Chir (Tokyo) 2005;45:349-55. [12] Lavy WJ. Clinical experience with motor and cerebellar evoked potential monitoring. Neurosurgery 1987;20:169-82. [13] Mares P. Development of cerebellar somatosensory evoked potentials in the rat. Acta Biol Med Germ 1976;35:55-62. [14] Oikawa T, Matsumoto M, Sasaki T, et al. Experimental study of medullary trigeminal evoked potentials: Development of a new method of intraoperative monitoring of the medulla oblongata. J Neurosurg 2000;93:68-76. [15] Paxinous G, Watson C. The rat brain in stereotaxic coordinates, ed 4. San Diego: Academic press; 1998. [16] Paxinous G. The rat nervous system, ed 4. Sydney: Academic press; 1995. [17] Sakuma J, Matsumoto M, Ohta M, et al. Glossopharyngeal evoked potentials after stimulation of the posterior part of the tongue in dogs. Neurosurgery 2002;51:1026-33. [18] Sakuma J, Suzuki K, Sasaki T, et al. Monitoring and preventing blood flow insufficiency due to clip rotation after the treatment of internal carotid artery aneurysms. J Neurosurg 2004;100:960-2. [19] Sasaki T, Kodama N, Matsumoto M, et al. Blood flow disturbance in perforating arteries attributable to aneurysm surgery. J Neurosurg 2007;107:60-7. [20] Sasaki T, Suzuki K, Matsumoto M, et al. Origins of surface potentials evoked by electrical stimulation of oculomotor nerves: Are they related to electrooculographic or electromyographic events? J Neurosurg 2002;97:941-4. [21] Sato M, Kodama N, Sasaki T, et al. Olfactory evoked potentials: experimental and clinical studies. J Neurosurg 1996;85:1122-6. [22] Suzuki K, Kodama N, Sasaki T, et al. Intraoperative monitoring of blood flow insufficiency in the anterior choroidal artery during aneurysm surgery. J Neurosurg 2003;98:507-14. [23] Yamada J, Shirao K, Kitamura T. Trajectory of spinocerebellar fibers passing through the inferior and superior cerebellar peduncles in the rat spinal cord: A study using horseradish peroxidase with pedunculotomy. J Comp Neurol 1991;304:147-60.
Commentary This report provides a basis for testing the spinocerebellar pathways. This may have a human intraoperative correlate in the future. The authors have presented several control studies to show that they indeed did record from the spinocerebellar pathway, not just a far field recording of a lemniscal sensory pathway. I expect others to explore how this may be extended and applied to monitoring the inferior cerebellar peduncle or spinal cord during surgeries that place those structures at risk. Such a human monitoring technique may require recording from subdural strip electrodes placed over the lateral cerebellum. Marc Nuwer, MD, PhD David Geffen School of Medicine at UCLA Los Angeles, CA 90095, USA E-mail address: [email protected]