Neuroscienee Bulletin September 30，2006，22(5)：26 1—266．http：／／www．neurosci．cn 26l

·Article·

Effects of spinal cord stimulation on cerebrovascular flow：role of sym-

pathetic and parasympathetic innervations

Jun Zhong ，Oren Sagher

Department of Neurosurgery，Xinhua Hospital， Shanghai Jiaotong University School of Medicine，Shanghai 200092，China

Department of Neurosurgery，University of Michigan，Ann Arbor,MI 48 1 09，USA

Abstract：Objective Cervical spinal cord stimulation(SCS)has been found to augment cerebral blood flow(CBF)in a

number of animal models．However，the effective use of SCS is hampered by a lack of understanding of its mechanism(s1

of action．In this paper，we focus on the sympathetic and parasympathetic effects of SCS on CBF． Method Sprague—

Dawley rats were selected f0r the experimental series． The animals were divided into 5 groups to underwent SCS and laser

Doppler flowmeter(LDF)recordings．Control group，the animal underw ent SCS and LDF recordings without any surgery

of the nerve fibers and ganglia．V 1 group．the animal underw ent bilateral resection of the nasociliary and post—ganglionic

parasympathetic nerve fibbers．SCG group，the animal underw ent bilateral resection of supper cervical ganglion． V 1+SCG

group，the animal underw ent both surgeries as V 1一and SCG—group animals did． Sham group，the animal underw ent the

carotid manipulation with blunt-tipped forceps as well as the dissection of nasociliary and post-ganglionic parasympa—

thetic nerve fibers around the ethmoidal foramen，but without cutting any nerves．Results During the SCS，the LDF was

no statistical difference between the V1 or SCG group and the control group．Yet，the effects ofSCS on CBF are completely

abolished in V 1+SCG group．Conclusions Surgical interruption of both the parasympathetic and sympathetic pathways

has the contradict effect on SCS-induced CBF augmentation．

Keywords：spinal cord stimulation；cerebral blood flow；sympathetic；parasympathtic；laser Doppler flowmeter

1 Introduction

Spinal cord stimulation(SCS)was firstly accepted to

improve microcirculatory parameters in patients with criti-

callimb ischemia] 1．Later Hosobuchi]21 noted that global

cerebral blood flow(CBF)increased in a small cohort of

patients treated with SCS for chronic pain．Recently，cervi-

cal SCS has been found to augment CBF in a number of

animal models[ ’ ．Regardless the promise of clinical ben．

efit of SCS in the treatment of cerebralischemia．the effec—

tive use of SCS is ham pered by a lack of understanding of

its mechanism(s)of action．

Cerebral blood flow is controlled by a complex array of

mechanisms that include metabolic coupling，vascular auto-

regulation and control of brainstem vasomotor centers．as

Corresponding author：Jun ZHONG

Tel：86—21．65790000—7705

Fax：86—21．65l53984

E—mal：ZhongM DPhD@ l26．com

Article ID：l673．7067(2006)05．0261．06

CLC number：O425

Document code：A

Received date：2006．07．05

well as peripheral lnnervatlons．Whereas the local meta—

boric requirements of neurons and hemodynamics are likely

to dictate regional CBF，global changes in CBF are likeiy to

be subject to central and peripheral vasomotor innervations．

Evidence from studies of the peripheral vasculature

indicates that a decrease in sympathetic nervous system

activity during stimulation may lead to the observed aug—

mentation in blood flow ．We have previously shown sig—

nificant attenuation in the SCS—induced CBF response af-

ter administration of the ganglionic blocker hexametho—

nium Sympathetic tone plays a role in SCS—induced CBF

augmentation，mediated primarily by 仅1一adrenergic

receptors【81．Studies indicated that cerebral arteries．both

extracerebral and intraparenchymal，are diffusedly inner—

vated by adrenergic nerves arising from the superior cervi—

cal ganglion(SCG)．There is evidence that neurons from

the medial aspects of laminae I，II and V at high cervical

levels project directly to the SCG ”．

Other studies have also demonstrated the presence of

acetyl choline(Ach)and high levels of choline acetyl trans-

ferase rCHAT) as well as functional receptors of the pow—

维普资讯 http://www.cqvip.com

262 Neuroscience Bulletin September 30，2006，22(5)：26 1—266

erful vasodilator vasoactive intestinal polypeptide(VIP)in

fibers innervating pial and arteriorlar vessels．Using the

immunofluorescence technique，the origins and pathways

of CHAT-and VIP—positive fibers to pial arteries are

delineated．In the rat，the principal origins of those para-

sympathetic fibers are the sphenopalatine ganglion，otic

ganglion and internal carotid ganglionf1 3_ ．The chiefsource

of parasympathetic pial innervation is the sphenopalatine

ganglion．Fibers from the ganglion pass up the medial wall

of the orbit，enter the skull through the ethmoidal foramen

and follow the internal ethmoidal artery back to the circle of

Willis．

One potential upstream target of SCS is the para—

sympathetic outfow from the brainstem．Parasympathetic

innervation of pial vasculature，involving Ach and calcito—

nin gene—related peptide(CGRP)，is known to emanate from

the brainstem primarily via the trigeminal nerve and enter

the cranial circulation through the nasociliary nerve．A

change in parasympathetic tone therefore is a possible link

between SCS and cerebrovascular tone．

In the present study，we sought to determine the role

of the parasympathetic pial innervation as well as the sym—

pathetic pathway in the SCS response．

2 M aterials and methods

2．1 Animal preparation Adult male Sprague-Dawley rats．

each weighing 250 g to 350 g，were selected for the experi—

mental series．Th e animals were housed in standard condi．

tions in a laboratory environment with free access to food

and water．Care was taken to minimize stress and nocicep—

tive input to the animals before and during surgery．Gen—

eral anesthesia was induced with 5％ isoflurane fAerrane)．

After intubation and initiation of mechanical ventilation

with a rodent ventilator(model 683；Harvard Apparatus，

Inc．．S．Natick．USA)．isoflurane was titrated between 1．5％

and 2-25％ to maintain a mean arterial pressure between 80

mmHgand 120mmHgandanorm alPaC02levelbetween35

mmHgand45mmHgduringtheexperiment．The ratswere

paralyzed with gallamine(10 mg&g，i．v．；Sigma Chemical

Co．，St．Louis，USA)as a muscle relaxant．The animal’s body

temperature was maintained at(37~0．5)oC with an automatic

heating device(model 73A—YSI；Yellow Springs Instruments，

Yellow Springs，USA)．The femoral artery was cannulated

for continuous monitoring of arterial blood pressure and

arterial blood gas levels．

2．2 Cerebral blood flow measurement A mi dline incision

was made to expose the skull and cervical spine．A mi cro—

surgical drill was then used to create a 2．5一mm—diameter

burrhole6l-nlnlateraland 1 l-nlnposteriortotheBregma，as

previously described[31．To ensure reproducibility of

recording．this location was held constant．Great care was

taken to maintain dural integrity．Real—time cortical blood

flow was then recorded continuously by using a laser Dop—

pier flowmeter(Laserflow BMP2；Vasamedics，Inc．，Little

Canada，USA)．All data were collected and recorded on a

Macintosh computer using standard charting software

(Chart V4．1．1：ADInstruments；Mountain View,Canada)．

2．3 Spinal cord stimulation Th e animal was placed in a

sterotactic frame(model 900；Kopf Instruments，T~unga，

Canada)．A midline incision was made from the occiput to

the cervical spine．Th e first and second lami na were identi—

fled and cleared of paraspinous muscles．A lami nectomy

between C 1 and C2 was perform ed，sparing the dura．Me—

ticulous hemostasis was maintained to keep the dorsal sur-

faceof spinalduradryforthedurationoftheexpe riment．A

standard stimulation setup was used，which consisted of a

stimulator(model S48；Grass Instruments，Quincy，USA)，a

constant—current unit(model CCU1一A：Grass Instruments)

andan oscilloscope (model511l-A；Tektronix，Gaithersburg，

USA)．A ground electrode was placed in the left proximal

hindquaaers of the rat．A 1 mm platinum—ball electrode

mounted on a sterotactic electrode holder was then low—

ered onto the exposed dorsal dural surface in the mi dline．

After a stable laser Doppler flowmetry value(LDF)baseline

was established，monopolar electrical simulation was

perform ed．According to our previous study，we chose the

optimal simulation set up，i．e．1．5 mA of current am plitude，

250 ms ofpulse width an d 50 Hz offrequency．Th e stimu-

lation length was 2 mi n[41．

2．4 Parasympathetic pathway resection A midline inci—

sion was made between the orbits，and the skin and

priosteum were reflected past the superior orbital margin．

Intraorbital structures were retracted laterally and the struc—

tures in the ethmoidal foram en were exposed．After coagu—

lating the anterior ethmoidal artery and vein，we cut the

nasociliary and the post-ganglionic parasympathetic nerve

fibres just outside the ethmoidal foram en．

2．5 Superior cervical gangli0n removal A mi dline super

cervical incision was made．The cervical muscles were re—

tracted to expose the carotid arteries．Th e sympathetic chain

was then visible dorsal to the arteries on the lateral verte—

bral border in the cervical region．Th e SCG was identified

维普资讯 http://www.cqvip.com

Jun Zhong，el al Mechanisms of SCS on CBF 263

around the bification of the carotid and was removed

bilaterally．

2．6 Animal classificatiOn The rat was divided into 5

groups．Control group，the animal underwent SCS and LDF

recordings without any surgery of the nerve fibers and

ganglia．V1 group，the animal underw ent bilateral resection

of the nasociliary and post—ganglionic parasympathetic

nerve fibbers．SCG group，the animal underw ent bilateral

resection of supper cervical ganglion．V I+SCG group， the

animal underw ent both surgeries as V 1·and SCG-．group

animalsdid．Sham group，theanimalunderw entthe carotid

manipulation with blunttipped forceps as well as the dis—

section of nasociliary and post·ganglionic parasympathetic

nerve fibers around the ethmoidal foramen，but without

cutting any nerves．

2．7 Statistical analysis The LDF measurement of CBF

was converted to percentage change from the baseline

leve1．It was presented as mean ±SEM．Comparisons of

values were made by student’S t-test and single factor

ANOVA with post—hoc analysis using Bonferroni／Dunn test．

Statistical significance was accepted at P<0．05．

3 Results

3．1 Parasympathetic innervation effect on CBF during

SCS In order to evaluate the role of parasympathetic out．

flowonthe cerebrovascular effectsofSCS．we cutthenaso—

ciliary nerve and the post·ganglionic never fibers from the

sphenopalatine gangalion．SCS induced significant LDF

increase in both parasympathetic transaction and control

animals．There was no statistical difference between the

twogroups(58．5％±3-2％ VS67．6％±2．7％，P>0．05，Tlab．1)．

However．the peak LDF changes seemed to occur later in

100

90

80

70

60

一

50

40 —

30

20

10

0

一 lO

the V I group compared with that in the control group仃1ab．

2)．Table 1 depicted the LDF changes(％)during SCS am ong

different groups．Figure 1 demonstrates the time course of

LDF changes before．during and after SCSinVl and con—

trol groups．Table 2 shows the every 10 S LDF changes in

V1 and control groups．

3．2 Sympathetic innervation effect on CBF during SCS

When the SCG was cut bilaterally．SCS stillinduced a ro．

bust LDF increasing of 58．5％ ±3．2％ fP<0．05 VS Control

Tab．1 The LDF changes during SCS among different groups

(％1

尸 <0．001 VS othergroups

Tab．2 The LDF changes in V1 group compared with Control

group (％ 1

— 6 __4 —2 0 2 4 6

Fig．1 The time course of LDF changes before，du~ng and after SCS in VI and control groups

维普资讯 http://www.cqvip.com

264 Neuroscience Bulletin September 30， 2006，22(5)：261—266

group)．However，the peak increase did not occur until 1．5

min after the SCS began．Table 3 shows the LDF changes in

every 1 0 S in SCG group compared with control group．Fig—

ure 2 delineates the time course of LDF changes during the

whole experiment in SCG and control groups．

3．3 Interruption ofboth parasympathetic and sympathetic

nerve abolish SCS-induced LDF augmentation Since ou,r

previous study had observed that SCS continued to exhibit

a robust effect on CBF following bilateral SCG removalt ，

our next series ofthe experiment focused on both the para—

sympathetic and sympathetic pathways．In this experiment，

the effects of SCS on CBF are completely abolished by

interruption of both the parasympathetic and sympathetic

pathways．TheLDFchangeinV1+SCGgroupwas 3．5±0．8

(P<0．05 VS other groups，Tab．1)．Figure 3 illustrates the

LDF changes in V 1+SCG and sham groups during the

Tab．3 The LDF changes in SCG group compared with Con-

trol group(％)

Time(min)

Fig·2 The time course of LDF changes during the whole experiment in SCG and control groups

Time(min)

Fig·3 The LDF changes in VI+SCG and sham groups during the experiment

6 8 10

维普资讯 http://www.cqvip.com

Jun Zhong，et al Mechanisms of SCS on CBF 265

experiment．

4 Discussion

There is an abundant supply of sympathetic，para-

sympathetic and sensory nerves to the adventitia of blood

vessels of the brain surface and their parenchymal branches．

It is well established that in most species the sympathetic

fibers originate in the SCG，but only recently have the ori-

gins and pathways of the cerebrovascular parasympathetic

fibers been clarifled．Various methods including immuno．

histoc hemical mapping demonstrated the presence of para-

sympathetic fibers in cerebral vessels，cranial ganglia and

pathways．Goadsby et a1．“ showed that if the trigeminal

ganglion is stimulated unilaterally in the cat，there is a bi—

lateral increase in regional CBF in the frontal and parietal

cortex，butffthefacialnerveisthencutbilaterallythefrontal

and the ipsilateral parietal increases in CBF are abolished．

This indicates a neurally mediated vasodilatation via the

~geminalnerve，spinalIrigeminalnucleus，andthefacialnerve

pathway containing the parasympathetic ganglia and fibers．

Th e role of SCG and the cervical sympathetic chain

was examined by observing the effects of surgical sympa-

thectomy on SCS-induced CBF changes in our previous

study【8】．At baseline．surgical sympathectomy did not sig-

nificantly alter CBF．This is consisted with the findings of

other studies which have similarly shown that sympathetic

tone does not significantly impact resting CBF ．W hen

SCS was applied following surgical ablation of the SCG the

resulting CBF response remained robust，indicating that

the CBF response to SCS is less even if dependent on cer-

vical sympathetic outflow only．

Th e sympatheticfibers arisingfrom theSCG mayhave

some role in extending the upper limi t of autoregulation

and，in addition，have trophic effects on blood vessels『l91．

The parasympathetic nerves arising in the superior

salivatory nucleus traverse the sphen0palatine ganglion

to the cerebral vessels．The sensory nerves arise from the

trigemi nal ganglion to innervate the cerebral circulation【 。

and project to second organ neurons in the trigeminal

nucleus caudalis and its caudal extension into the C 1 and

C2 cervical spinalcord．Ithas been shownthat stimulationof

the trigeminal ganglion increases CBF in cat[15,211 and rat『221．

Nevertheless，when this parasympathic pathway was

blocked，SCS still induced a significant CBF increase in the

present study．

Th e phenomenon，blocking either sympathetic or para—

sympathetic pathway failed to counteract the cerebra—

vascular effects of SCS while blocking both sympathetic

and parasympathetic pathway did，implied that the

mechanism(s)underlying SCS-induced augmentation of

CBF seem to involve both sympathetic and parasympa-

thetic neura】jnnervatjon of the cerebral b】ood vesse】s．

Aknowledgements：This work was supported by a

grant form Xinhua Hospital affiliated to Shanghai Jiaotong

University School of Medicine．

References：

[1】 Augustinsson LE，Carlsson CA，Holm J，Jivegard L．Epidural

electrical stimulation in severe limb ischemia．Pain relief,in—

creased blood flow,and a possible limb—saving effect．Ann Surg

l985．202：l04一ll0．

[2】 Hosobuchi Y．Electrical stimulation of the cervical spinal cord

increases cerebral blood flow in humans．Appl Neurophysiol

l 985．48：372—376．

[3】 Sagher O，Huang DL．Effects of cervical spinal cord stimula—

tion on cerebral blood flow in the rat．J Neurosurg 2000，93(1

Supp1)：71—76．

[4】 Zhong J，Huang DL，Sagher O．Parameters influencing aug—

mentation of cerebral blood flow by cervical spinal cord

stimulation．Acta Neurochir(Wien)2004．146：1227—1234．

[5】 Visocchi M，Cioni B，Ve唱 S，Marano G，Pentimalli L，Meglio

M．Spinal cord stimulation and cerebral blood flow：an experi—

mental study．Stereotact Funct Neurosurg 1 994，62：1 86—1 90．

[6】 Linderoth B，Herregodts P，Meyerson BA．Sympathetic me—

diation of peripheral vasodilation induced by spinal cord

stimulation：animal studies of the role of cholinergic and adr—

energic receptor subtypes．Neurosurgery 1994，35：71 1—719．

[7】 Myklebust JB，Cusick JF,Boerboom LE，Prieto TE，Khan TA．

Vascular effects of spinal cord stimulation in the monkey．

Stereotac Funct Neurosurg 1 995，64：32—39．

[8】 Patel S，Huang DL，Sagher O．Sympathetic mechanisms in

cerebral blood flow alteration s induced by spinal cord

stimulation．J Neurosurg 2003．99：754—761．

[9】 Nicholas MK，Sagher O，Hartley JP,Stefansson K，Arnason

BG．A phenotypic analysis of T lymphocytes isolated from

the brains of mice with allogeneic neural transplants．Prog

Brain Res．78：249—259．

[1 0】Tamamaki N，Nojyo Y．Intracranial trajectories of sympa—

thetic nerve fibers originating in the superior cervical gan—

glion in the rat：WGA—HRP anterograde labeling study．Brain

Res 1987，437：387—392．

[1 1】Jansen AS，Loewy AD．Neurons lying in the white matter of

the upper cervical spinal cord project to the intermediolateral

cell column．Neuroscience 1997．77：889—898．

[1 2】Florence VM，Bevan JA．Biochemical determinations of cho—

linergic innervation in cerebral arteries．Circ Res 1 979．45：

维普资讯 http://www.cqvip.com

266 Neuroscience Bulletin September 30，2006，22(5)：26 1—266

【1 3】

[14】

[1 5】

[16】

[17】

[18】

2l2—2l8．

Suzuki Y，McMaster D，Huang M ，Lederis K，Rorstad OP．

Characte rjzation of functional receptors for vasoactive intes—

tinal peptide in bovine cerebral arteries．J Neurochem 1 985，

45：890—899．

Hara H，Weir B．Pathway of acetylcholjnesterase containing

nerves tO the major cerebral arteries in rats．J Comp Neurol

1986，250：245·252．

Goadsby PJ，Duckworth JW．Effect of stimulation of trigemi-

nal ganglion on regional cerebral blood flow in cats．Am J

Physiol 1 987，253(2 Pt 2)：R270—274．

Eklof B，Ingvar DH，Kagstrom E，Olin T．Persistence of cere-

bral blood flow autoregulation following chronic bilateral cer-

vical sympathectomy in the monkey．Acta Physiologica

Scandinavica 1971，82：172．176．

Heistad DD，M arcus ML，Gross PM ．Effects of sympathetic

nerves on cerebral vessels in dog，cat，and monkey．Am J

Physiol 1978，235：H544-552．

Edvinsson L，Owman C，Siesjo B．Physiological role of cerebro一

颈髓 电刺激增加脑 血流的作用机制

vascular sympathetic nerves in the autoregulation of cerebral

blood flow．Brain Res 1976，1 17：5 19·523．

[1 9】Dimitriadou V'Aubineau P-Taxi J，Seylaz J．Ultrastructural

changes in the cerebral artery wall induced by long··term sym。。

pathetic denervation．Blood Vessels 1 988．25：1 22一l43．

[20】Liu—Chen LY,Gillespie SA，Norregaard T V，Moskowitz MA．

Co—localization of retrogradely transported wheat germ agglu—

tinin and the putative neurotransmitter substance P within

trigeminal ganglion cells projecting tO cat middle cerebral

artery．J Comp Neurol 1984．225：187-192．

[2 1】Kaube H，Keay KA，Hoskin KL，Bandler R，Goadsby PJ．Ex-

pression of c-Fos—like immunoreactivity in the caudal medulla

and upper cervical spinal cord following stimulation of the

superior sagittal sinus in the cat．Brain Res 1993，629：95—102．

[22】Seylaz J，Hara H，Pinard E，Mraovitch S，MacKenzie ET，

Edvinsson L．Effect of stimulation of the sphenopalatine gan—

glion on cortical blood flow in the rat．J Cereb Blood Flow

Metab l 988．8：875．878．

研究

仲骏 ，Oren Sagher

上海交通大学医学院附属新华医院神经外科，上海 200092

密歇根大学神经外科，安娜堡，密歇根州 48109，美国

摘要：目的 尽管颈髓电刺激(scs)在许多动物实验中已被证实能增加脑血流，但其作用机理还不甚明了。本

文将从脑血管的交感和副交感通路方面探讨SCS对脑血流的影响机理。方法 对SD大鼠给予SCS并用激光多普

勒(LDF)实时记录脑血流。动物分成 5组，对照组：不对神经纤维和神经节作任何手术处理；v1组：切断双

侧鼻睫状神经及其副交感节后纤维；SCS组：切断双侧颈上神经节；v1+SCG：切断双侧鼻睫状神经及其副

交感节后纤维及双侧颈上神经节；假手术组：手术暴露双侧鼻睫状神经和其副交感节后纤维及双侧颈上神经节，

但不切断神经纤维或神经节。结果 SCS时的LDF变化在对照组、V1~IJSCG组问没有显著性差异。但在VI+SCG

组中，SCS的升脑血流效应却被明显抑制了。结论 SCS的升脑血流效应可能是通过交感和副交感双重通路实

现的。

关键词：颈髓电刺激；脑血流；交感神经；副交感神经；激光多普勒(LDF)

维普资讯 http://www.cqvip.com