Acta Physiologica Sinica, August 25, 2003, 55(4): 435£441

Received 2002-10-17Accepted 2003-04-29

The study was supported by Beijing Natural Scientific Foundation (No.7992022).

ª³Corresponding author.

Tel: +86-10-6301-3355 ext.2882;

Fax: +86-10-83152684;

E-mail: zhuangp@sina.com

Research Paper

Characteristics of subthalamic neuronal activities in Parkinson's disease

ZHUANG Ping, LI Yong-Jie*

Beijing Institute of Functional Neurosurgery, Xuanwu Hospital, Capital University of Medical Sciences,

Beijing 100053

Abstract: The relationship between neuronal activity in subthalamic nucleus (STN) and parkinsonian symptoms was investigated. Thirty-five patients with idiopathic Parkinson's disease (PD) received stereotactic surgical treatment. Microelectrode recording in STN and electromyography (EMG) on the limb contralateral to

the surgical side were employed intraoperatively. Single unit firings discriminated from multiple neuronal discharges were recorded, and the correlation between neuronal activity and limb EMG was analyzed. The results showed that there were distinguished characteristics of neuronal discharges in STN and its surrounding areas. Of 346 STN neurons recorded from 36 microrecording trajectories in 35 patients, three patterns of neuronal activities were identified: irregular bursting pattern with a mean frequency of 43.0¡À11.2 Hz (56%, n=244); tonic firing pattern with a mean firing frequency of 41.0¡À12.0 Hz (15%, n=66); and regular bursting pattern with a mean frequency of 47.0¡À11.7 Hz (29%, n=126). The rhythm of regular bursting with the frequency ranging from 3.8 to 6.0 Hz was highly correlated with the frequency of limb tremor measured by EMG (r2=0.66, P<0.01). These cells were therefore called tremor-related neurons or tremor cells. In particular, 80% tremor cells were located in the medio-superior part of STN. In conclusion, our results suggest that microelectrode recording is a critical technique for electrophysiological localization of the target in treating PD. The tremor-related neuronal activity and movement-related neuronal activity recorded from STN are responsible for the clinical parkinsonian symptoms, suggesting that STN plays an important role in the pathophysiology of PD.

Key words: subthalamic neuronal activity; microelectrode recording; stereotaxis

ÅÁ½ðɲ¡ÇðÄԵ׺ËÉñ¾ÔªµÄµç»î¶¯Ìصã

ׯƽ, ÀîÓ½Ü*

Ê׶¼Ò½¿Æ´óѧÐûÎäÒ½Ôº, ±±¾©¹¦ÄÜÉñ¾Íâ¿ÆÑо¿Ëù, ±±¾© 100053

ÕªÒª:±¾Ñо¿Ì½ÌÖÁËÅÁ½ðɲ¡(Parkinson's disease, PD)»¼ÕßÇðÄԵ׺Ë(subthalamic nucleus, STN)Éñ¾Ôªµç»î¶¯µÄÌص㼰ÆäÓëPDÖ¢×´µÄ¹Øϵ¡£ 35ÀýPD»¼ÕßÔÚ½ÓÊÜÊÖÊõÖÎÁƵÄͬʱ, Ó¦ÓÃ΢µç¼«Ï¸°û¼Ç¼ºÍEMG¼Ç¼¼¼Êõ, ¼Ç¼ÊÖÊõ°ÐµãSTN¼°ÆäÖÜΧ½á¹¹Éñ¾ÔªµÄµç»î¶¯ÒÔ¼°ÊÖÊõ¶Ô²àÖ«ÌåµÄEMG¡£ Ó¦Ó÷ÖÎöÈí¼þÕç±ðµ¥Ï¸°ûµç»î¶¯, ·ÖÎöÆäÌص㼰ÆäÓëÖ«ÌåEMGµÄ¹Øϵ¡£ ½á¹û±íÃ÷, STN¼°ÆäÖÜΧ½á¹¹¾ßÓÐÌØÕ÷ÐԷŵç»î¶¯¡£ÔÚ36¸ö¼Ç¼ÕëµÀÖÐ, ¹²·¢ÏÖ436¸öSTNÉñ¾Ôª, ƽ¾ù·ÅµçƵÂÊ44.0¡À20.5 Hz¡£ ÆäÖÐ, 56%µÄÉñ¾Ôª³Ê²»¹æÔò´Ø×´·Åµç; 15%³Ê½ôÕÅÐԷŵç; 29%³Ê¹æÔòµÄ´Ø×´·Åµç, Æä·Åµç½ÚÂÉÓëÖ«ÌåÕð²üµÄEMG¸ß¶ÈÒ»ÖÂ(r2=0.66, P<0.01), ³Æ֮ΪÕð²üϸ°û¡£ ÔÚPDÕð²üÐÍ»¼ÕßµÄSTNÖз¢ÏÖ´óÁ¿Õð²üϸ°û, ÇÒ80%λÓÚSTNÖÐÉϲ¿, ¶øÔÚPD½©Ö±ÐÍ»¼ÕßµÄSTNÖоù·¢ÏÖÓëÔ˶¯Ïà¹ØµÄϸ°ûµç»î¶¯¡£ ±¾Ñо¿Ìáʾ, ͨ¹ý΢µç¼«¼Ç¼¼¼Êõ¿É׼ȷµØÅжÏSTNµÄλÖúͷ¶Î§; ÓëÕð²ü»î¶¯Ïà¹ØµÄϸ°û·ÅµçºÍÓëÔ˶¯Ïà¹Øϸ°ûµÄ·ÅµçÓëPDÖ¢×´ÓÐÄÚÔÚ¹Øϵ; STN²ÎÓëPDÔ˶¯ÕÏ°µÄ²¡ÀíÉúÀí¹ý³Ì¡£

¹Ø¼ü´Ê: Éñ¾ÉúÎïѧ; ÇðÄԵ׺ËÉñ¾Ôªµç»î¶¯; ΢µç¼«¼Ç¼; Á¢Ì嶨Ïò

ÖÐͼ·ÖÀàºÅ: Q25; R33; R338.2

Parkinson's disease (PD) is a progressive neurodegenerative disorder due to the loss of dopaminergic neurons of the substantia nigra pars compacta that results in a cascade of dysfunction involving all components of the basal ganglia circuitry[1]. The clinical manifestations of PD includes tremor, rigidity, bradykinesia, and postural instability. Over the last few years, there has been a growing number of literatures on the clinical benefits of ablative surgery and implantation of long-term stimulat-ing electrode (deep brain stimulation, DBS) in the deep brain nuclei to treat PD[2£6]. A number of reports demonstrated the efficacy of the lesioning and DBS of STN in improving most parkinsonian motor signs, e.g. tremor, rigidity, akinesia and gait freezing[2£6]. In stereotactic neurosurgery, extracellular microelectrode recording has now been widely performed, providing the precision necessary for accurate surgical localization. The most common technique is the single unit recording of the extracellular action potentials using tungsten or platinum-iridium microelectrodes in the STN in humans and monkeys[2]. Microelectrode recording is helpful in the characterization of neuronal activity in STN and its surrounding structure in normal and pathological conditions such as resting tremor and rigidity. The microrecording also provides direct neurophysiological measures of spontaneous and stimulus-evoked cellular activity in human brain. In addition, the analysis of the intraoperative data is important to provide information for the refinement of surgical targeting as well as for the understanding of pathophysiology of basal ganglia, in particular, parkinsonian symptoms in relation to central neuronal activity. In this study, we investigated the neuronal activity of STN and its surrounding structures during the operation of 35 patients with PD and demonstrated the firing characteristics of encountered cells. Our data provide detailed information that can be used to improve the accuracy of targeting the STN and help to understand the pathophysiology of PD.

1 MATERIALS AND METHODS

1.1 Patients£® The first 35 consecutive patients with idiopathic PD undergoing subthalamotomy at our institute were included in this study. There were 23 men and 12 women. The average age was 56¡À8.7 years and the average disease duration was 6.9¡À5.8 years. A clinical diagnosis of idiopathic PD was made based on the presence of at least two of the cardinal signs: tremor, rigidity and bradykinesia. The inclusion criteria also included the documentation of a positive response to levodopa. Their Hoehn and Yahr stage was at the range of 2.5£4.5 when ¡°off¡± medication[7]. Twenty-two patients had rest tremor and 13 patients presented predominantly rigidity syndrome at limbs.

1.2 Stereotactic surgery and microelectrode recording£® The technique and method of microelectrode-guided surgery have been reported previously[8]. Briefly, CRW stereotactic head frame was tightly fixed on the skull using cranium pin under local anesthesia. Based on the standard stereotactic atlas of Schaltenbrand & Wahren[9], the theoretical coordinates to the target STN were: 4£6 mm below the anterior and the posterior commissures (ACPC) line, 9£12 mm lateral to the midline and 1£2 mm behind the midpoint of ACPC line. MRI stereotactic software provided directly view to determine the anatomical target of three-dimensional axial image. Tungsten microelectrodes (FHC, ME, USA) with a tip diameter of 20£30 ¦Ìm and an impedance of 0.1 to 0.5 Mohm (at 1000 Hz) were used for recording. The electrode was placed in another slightly larger stainless-steel ¡°carrier¡± tube for entering subcortical deep nucleus. The recording procedure started 10 mm above the target in an anterosuperior position advanced in 1 ¦Ìm step with a fine micro-drive. The electrical signals were sampled at 7.5 kHz, amplified (¡Á20000£50000), filtered at bandpass 0.1£5 kHz and recorded by using an acquisition program of PolyView (Astro-Med. Inc., RI, USA). All signals were on-line displayed on both a computer screen and an oscilloscope (HITACHI, V-1560, Japan), and stored in a computer (Pentium ¢ó) for off-line analysis. The electromyography (EMG) was recorded simultaneously using surface electrodes (World Precision Instruments, USA) in the upper limb of extensor carpi radialis (ECR), flexor carpi radialis (FCR) and lower limb of tibialis anterior muscles contralateral to the surgery side.

1.3 Electrical signal analysis£® In the present study, most of the recordings obtained from the trajectory to target were multiple neuronal activities. Single units were therefore identified according to the pattern of action potential and waveforms of neuronal discharges. For data analysis, only spikes (negative upward) that had a signal to noise ratio of greater than two were used. The firing rate and rhythmic burst of neuronal activity were analyzed and a cross-correlation test was carried out to determine the relationship between neuronal activity and simultaneously recorded limb EMG. Gaussian fit test was performed to determine distribution of neurons, and the mean value and standard deviation (SD) were calculated. Significance was set at the level of 0.05. PolyView program (Astro-Med, RI, USA) and Origin 5.0 (Microcal Software, MA, USA) was performed for data analysis.

2 RESULTS

2.1 Localization of STN

In the present study, except for one patient who had two trajectories to achieve the final target, successful localization of the final target was achieved by a single trajectory in 34 patients. Neuronal recordings were therefore obtained from 36 trajectories. It has been established that basal ganglia nuclei have a characteristic spontaneous discharge pattern, which is relatively easy to identify[2]. In the present study, the patterns of neuronal activity in STN were illustrated as follows: (1) reticular thalamus (Rt), which is located above the dorsal area of STN and is the first region that electrode encountered the neuron with bursting discharges at low rate of 7£8 Hz; (2) the Zona incerta (Zi): as the electrode passes Rt, a relatively quiet region was encountered, showing few units and low background noise; (3) substantial nigra (SN), which is located ventral and medial to the STN. The SN was characterized by a regular pattern of firing and higher rate of discharge. As compared to the neuronal activities in the STN, the background noise of SN was lower and the cells in SN were easy to be recognized. We recorded 51 SN neurons that discharged with a mean frequency of 73.2¡À22.0 Hz (from 28 to 120 Hz) and showed distinguishable neuronal activity from its surrounding structures. Figure 1 illustrates typical neuronal activities obtained from the structures described above from one patient with PD.

Fig.1.

Typical neuronal discharges recorded along the depth of microelectrode trajectory in STN and its surrounding areas. The map of 12 mm from mid-sagittal plane is originated from Schaltenbrand & Wahren Atlas. The dashed lines point to the regions where the cells were recorded.

Rt, reticular thalamus; Zi, zona incerta;

IC, internal capsule; SN, substantia nigra;

STN, subthalamic nucleus; GP, globus pallidus;

ACPC, anterior commissure and posterior commissure line.

2.2 Patterns of STN neuronal activity

In the present study, the length where neuronal activity could be detected along the length of trajectory through the STN was 5£8 mm. A total of 436 neurons were identified from 36 trajectories indicating an average of 12.1 neurons in each trajectory (2 neurons/mm). The mean firing rate of all neurons was 44.0¡À20.5 Hz (from 20 to 88 Hz). Three patterns of neuronal activities were identified (Fig.2): (1) 56% (n=244) neurons exhibited irregular bursting with a mean firing frequency of 43.0¡À11.2 Hz; (2) 29% (n=126) neurons showed rhythmic activity synchronized with the contralateral limb EMG activity. These cells were termed as ¡°tremor cells¡±[8,10,11]; and (3) 15% (n=66) neurons showed tonic firing and the amplitude of action potential often co-varied with the heart rhythm. The mean firing rate of these neurons was 41.0¡À12.0 Hz.

Fig.2.

Patterns and frequency analysis of STN neuronal discharges.

A: Irregular neuronal discharges (1) and their frequency analysis (2).

B: Tremor-related neuronal discharges (1) and their frequency analysis (2).

C: Tonic neuronal discharge (1) and their frequency analysis (2).

2.3 Relationship between tremor-related neuronal activity and limb tremor

Tremor-related neuronal activity was observed in STN in all PD patients who had resting tremor (n=22). To explore the tremor-related neuronal activity correlated with limb tremor, 126 neurons were further analyzed. Figure 3A illustrates an example of a typical tremor cell corresponding to limb tremor measured simultaneously in the muscles of ECR and FCR. Frequency analysis showed that the activity of the tremor cells had a rhythm of 3.8£6.0 Hz, which was similar to the frequency of parkinsonian tremor (4£6 Hz). Gaussian curve demonstrated that the average rhythm of the tremor cells was 4.60¡À0.56 Hz and the rhythm of limb tremor was 4.61¡À0.50 Hz (Fig.3B). The coefficient of rhythm burst of neuronal activity with limb tremor was 0.66 (P<0.01), probably indicating the presence of tremor cells in STN and causal relationship between their activities and tremor cells and limb tremor (Fig.3C).

Fig.3.

The association between tremor cell discharges and limb tremor activities.

A: Example of typical rhythmic bursting pattern of a tremor cell in STN. Its discharge was synchronized with limb tremor activities.

B: Frequency analysis of tremor cells and limb tremor measured by EMG.

C: The correlation analysis of the frequency of tremor cells and the frequency of limb tremor (the data

were collected from 35 patients).

2.4 Distribution of tremor cells in STN

During surgery, it was found that tremor cells were easily encountered in the upper portion of STN. Further analysis confirmed that 80% of ¡°tremor cell¡± were located in the dorsomedial region, according to standard human STN atlas[9].Figure 4 demonstrates the pattern of distribution of pooled 38 tremor cells. The data were collected from 5 patients. The anatomical map of STN 12 mm from mid-sagittal plane is originated from Schaltenbrand and Wahren atlas[9].

Fig.4.

Distribution of tremor-related cells in STN.

The STN tremor-related cells whose firing rhythm was synchronized with limb tremor were plotted on a lateral sagittal map 12 mm from the midline of human STN. The pooled cells (n=38) show that most of them were localized to the dorsomedial part of the STN. The cells were discriminated from the data recorded

in 5 PD patients.

2.5 Movement-related neurons in STN

Thirteen PD patients who revealed rigidity symptom of limb were tested by passive and active movement during operation. Twenty-two movement-related neurons were obtained along 13 trajectories, showing increased neuronal activity associated with the action of passive or active movements of the limbs. Figure 5 demonstrates a STN neuron in response to passive limb movement.

Fig.5.

The STN neuronal discharges related to upper limb EMG induced by passive movement.

3 DISCUSSION

The knowledge of neurophysiology of the basal ganglia neurons mostly relies on the extensive studies from animal experiments and less information is available from human study. Neurosurgery provides a unique opportunity to directly collect and analyze neuronal activities from human brain using electrophysiological methods during surgical operation. For example, the microelectrode recordings performed on awake PD patients could reflect the actual neuronal activity of the subcortical deep nuclei during disease status[8,10,11].

In the past years, the localization of subcortical nuclei was based on internal landmark of cranium

with reference to the atlas of human brain before surgery. With the development of computer topography and magnetic resonance imaging (MRI) techniques, the localization of human brain structures can be determined by measuring fixed distance from the midpoint of ACPC (landmark). However, there are significant individual variabilities in the spatial coordinates of these targets in ACPC-based coordinates. The brain shift due to cerebrospinal fluid lost after dural opening might result in distortion of the anatomical target. The present study determined the characteristics of the neuronal activities, e.g. firing frequency, patterns of discharges and the degrees of background noise, using microelectrode recording technique along trajectory through nucleus of Rt, Zi, STN and SN orderly during targeting. The recordings revealed a submillimetric precision in the determination of the location and region of lesioning at a cellular level in STN, indicating a significant improvement of the accuracy of target placement.

STN is a small nucleus beneath the thalamus, and has a high density of neurons[10]. In the present study, we recorded about 12 cells in each trajectory and demonstrated that the average neuron density of the STN was 2 cells/mm in STN. Although there was no neuronal activity in STN in normal human subjects studied so far, the firing rate of neuronal activity in STN was significantly increased in PD (44.0¡À20.5 Hz) as compared to the firing rate of normal primates (19.0¡À10.0 Hz)[1]. The results of increased neuronal activity in STN of PD were consistent with those of the recent studies on parkinsonian models of monkey in which the hyperactivity of STN neurons was associated with an increased excitability[12].

Our results on the neurophysiological characteristics of the STN and surrounding regions were consistent with the data obtained from the experiments of PD primate models[1,6,12,13] and patients with PD[10]. In those studies, it has been shown that, in addition to hyperactivity in STN associated with parkinsonian symptoms, there were tremor related neurons[6,10£13]. Furthermore, we have confirmed that most of the tremor cells (80%) localized at the dorsal medial part of STN which are associated with tremor affected distal limbs[10]. In addition, the presence of movement related neurons in STN was also observed in patients with rigidity in our study. STN with movement-related activity has been recorded in normal and 1-methy-4-phenyl-1,2,3,6-tetrathydropyridine (MPTP)-treated PD monkeys[12,14]. The movement-related activity was always excitatory[12,14], which is similar to our finding of predominant excitatory response to limb movements in those PD patients with rigidity syndrome. It may result from an increased glutamatergic input from STN in parkinsonism.

It is known that STN plays an important role in the control of movement by exerting a glutamatergic excitatory influence on the output structures of basal ganglia: the internus globus pallidus (GPi) and the substantial nigra pars reticulat (SNpr)[13,14]. In parkinsonian state, the most significant alterations are the increase in the neuronal firing rate and a change in the firing patterns including a greater tendency to discharge in bursts, a higher degree of synchronization of discharge between neighboring neurons, and a greater proportion of neurons with responses to somatosensory input in the STN and its efferent activity[15]. Such a change in STN, consequently, results in increased inhibition of thalamocortical and brain stem neurons leading to the motor dysfunction in PD[1, 13,15]. In fact, the clinical symptoms of PD were alleviated dramatically with the ablative lesioning or the high frequency stimulation (DBS) in STN[2£5], supporting the concept that the hyperactivity and alteration of activity in STN play a central role in the pathophysiology in PD[15].

However, it is clear that the magnitude of the individual parkinsonian signs such as tremor, rigidity and akinesia are highly variable between patients. Conceivably, these signs may be the result either of independent pathophysiologic mechanisms, or of the same pathophysiologic mechanism affecting different motor circuits.

***

Acknowledgment: We would like to thank Prof. Chen Biao for the reviewing and editing of the manuscript and Mr. Liu Han for the computer assistance during the study.

REFERENCES

[1]DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990;13(7): 281£285.

[2]Starr PA, Vitek JL, Bakay RAE. Ablative and deep brain stimulation for Parkinson's disease. Neurosurg 1998;43:989£1015.

[3]Alvarez L, Macias R, Guridi J, Lopez G, Alvarez E, Maragoto C, Teijeiro J, Torres A, Pavon N, Rodriguez-Oroz MC, Ochoa L, Hetherington H, Juncos J, DeLong MR, Obeso JA. Dorsal subthalamotomy for Parkinson's disease. Mov Disord 2001;16(1):72£78.

[4]Krack P, Pollak P, Limousin P, Benazzouz A, Benabid AL. Stimulation of subthalamic nucleus alleviate tremor in Parkinson's disease. Lancet 1997;350:1675.

[5]Li YJ (ÀîÓ½Ü). Surgical treatment for movement disorders: a review of 1135 cases. Chin J Neurosurg (ÖлªÉñ¾Íâ¿ÆÔÓÖ¾) 2001;17(6):350£353 (Chinese, English abstract).

[6]Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesion of the subthalamic nucleus. Science 1990;249:1436£1438.

[7]Fahn S, Elton RL, and Members of the UPDRS Development Committee. The Unified Parkinson's Disease Rating Scale. In Fahn S, Marsden CD, Calne DB, Goldstein M, eds. Recent developments in Parkinson's disease. Florham Park: Macmillan Health care Information, 1987, 153£163.

[8]Li YJ (ÀîÓ½Ü), Zhuang P, Zhao GG, Shi CQ, Zhang YQ, Li JY. Tremor-related neuronal activity in globus pallidus internus in parkinsonian patients. Chin J Neurosurg (ÖлªÉñ¾Íâ¿ÆÔÓÖ¾) 2002;18(1):18£21 (Chinese, English abstract).

[9]Schaltenbrand G. Wahren W. Atlas for stereotaxy of the human brain. 2nd Stuttgart: Thieme, 1977.

[10]Rodriguez-Oroz MC, Rodriguez M, Guridi J, Mewes K, Chockkman V, Vitek J, DeLong MR, Obeso JA. The subthalamic nucleus in Parkinson's disease: somatotopic organization and physiological characteristics. Brain 2001;124(9):1777£1790.

[11]Lenz FA, Tasker RR, Kwan HC, Schnider S, Kwong R, Murayama Y. Single unit analysis of the human ventral thalamus nuclear group: correlation of thalamic ¡°tremor cells¡± with the 3£6 Hz component of parkinsonian tremor. J Neurosci 1988;8:754£764.

[12]Bergman H, Wichmann T, Karmon B, DeLong MR. The primate subthalamic nucleus. ¢ò: Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 1994;72:507£520.

[13] Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders.

Trends Neurosci 1989;12(10):366£375.

[14]Wichmann T, Bergman H, DeLong MR. The primate subthalamic nucleus. I Functional properties in intact animals. J Neurophysiol 1994;72:494£506.

[15]Wichmann T, DeLong MR, Vitek JL. Pathophysiological consideration in basal ganglia surgery: role of the basal ganglia in hypokinetic and hyperkinetic movement disorders. In: Lozano AM eds. Movement disorders surgery. Prog Neurol Surg. Basel, Karger 2000;15:31£57.