Journal of the Neurological Sciences 336 (2014) 220–226

Contents lists available at ScienceDirect

Journal of the Neurological Sciences

j ourna l homepage: www.e lsev ie r .com/ locate / jns

Important casual association of carotid body and glossopharyngeal nerveand lung following experimental subarachnoid hemorrhage in rabbits.First report

Coskun Yolas a, Ayhan Kanat b,⁎, Mehmet Dumlu Aydin c,⁎⁎, Osman Nuri Turkmenoglu d, Cemal Gundogdu e

a Erzurum Regional Research and Education Hospital, Department of Neurosurgery, Erzurum, Turkeyb Recep Tayyip Erdogan University, Medical Faculty, Department of Neurosurgery, Rize, Turkeyc Ataturk University Medical Faculty Department of Neurosurgery, Erzurum, Turkeyd Sisli Etfal Research and Education Hospital, Department of Neurosurgery, Istanbul Turkeye Ataturk University Medical Faculty, Department of Pathology, Erzurum, Turkey

Abbreviations: SAH, subarachnoid hemorrhage; GPN, gtracranial pressure; CB, carotid body.⁎ Correspondence to: A. Kanat, Ekrem Orhon Mah 2. La

Merkez-Rize, Turkey. Tel.: +90 506 5855139; fax: +90 4⁎⁎ Correspondence to: M.D. Aydin, Ataturk UniversityNeurosurgery, Erzurum, Turkey. Tel.: +90 5323228389.

E-mail address: ayhankanat@yahoo.com (A. Kanat).

0022-510X/$ – see front matter. Crown Copyright © 2013http://dx.doi.org/10.1016/j.jns.2013.10.047

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 12 September 2013Received in revised form 27 October 2013Accepted 30 October 2013Available online 8 November 2013

Keywords:Subarachnoid hemorrhageGlossopharyngeal nerveCarotid bodyBrain stem compressionCardiorespiratory arrestDegenerative injury

Object: The glossopharyngeal nerves (GPNs) and carotid bodies (CBs) have an important role in the continuationof the cerebral autoregulation and cardiorespiratory functions. The relationship between degenerative injury ofCB and the GPN in subarachnoid hemorrhage (SAH) was studied.Methods: Twenty rabbits were included in this study. Five of them (n = 5) were used as control group. Theremaining animals (n = 15) were exposed to experimental SAH. In the six animals of the SAH group, severesigns of illness were observed, and these six animals were killed in the first week after SAH. Others animals(n = 9) were followed for 20 days and then sacrificed. GPNs and CBs were examined and, the live anddegenerated GPN axon number, and of CB neuron numbers were stereologically estimated.Results: The mean number of live neurons in CBs was 4206.67 ± 148.35 and live axons of GPNs were1211.66 ± 14.29 in the animals of the control group. The number of degenerated neurons of CBs was2065 ± 110.27 and the number of degenerated axons of GPNs was 530.83 ± 43.48 in early killed animalswith SAH. The number of degenerated neurons of CBs and the number of degenerated axons of GPNs were

found as 1013.89 ± 4184 and 2270.5 ± 134.38 in the living animals with SAH, respectively.Conclusions: High number of degenerated axons of GPN and neurons of CBs of the early killed animals suggestthat themortality in early SAHmight be due to GPNs injury secondary to compression of their axons or supplyingvessels by the probably herniated brainstem, and secondary denervation injury of CBs, and lung.

Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.

1. Introduction

Spontaneous subarachnoid hemorrhage (SAH) is responsible for 5%–7% of strokes [1,2]. It has the highest mortality and morbidity among alltypes of stroke [3–5]. Outcome after SAH depends on several factors,including the severity of the initial event, perioperative medical man-agement, surgical variables, and the incidence of complications [6].Extracerebral organ dysfunction is closely linked to the magnitude ofthe primary neurological insult [7]. Management of patients particularlyfor poor grade subarachnoid hemorrhage inherently carries a significantrisk of complication due to vulnerable brain and cardio-pulmonarydysfunction. It is well established that excessive sympathetic activity,

lossopharyngeal nerve; ICP, in-

leli Sok, Yilmaz Apt 14/3, 5310064 2130364.Medical Faculty Department of

Published by Elsevier B.V. All rights

resulting in the development of cardiopulmonary complications, com-monly occurs in patients who have suffered aneurysmal SAH. Lambertet al. showed that, hearts from animals following acute experimentalSAH exhibit enhanced sensitivity to norepinephrine infusion andsympathetic nerve stimulation, and are more prone to developarrhythmias [8]. Cardiopulmonary complications can hamper successfulpostoperative recovery even in unremarkable surgical situations, sothey are life threatening conditions that have serious implicationswith regard to patient outcome [9]. The vagal nerve injury may alsolead to respiratory problem in SAH [10]. Dysfunction of the carotidbodies (CBs) can result in cerebrovascular and cardiorespiratory failure.These bodies are very sensitive in the arterial pH changes and it isessential for the maintenance of cerebral autoregulation and cardiore-spiratory functions [11]. It receives chiefly sensory nerves from theglossopharyngeal nerve (GPN). So GPN is an information conductor be-tween the CBs and the cardiorespiratory centers. CBs were regarded asbeing derived from cells of the sympathetic nervous system [12], andmay be accepted as the sympathetic ganglia. Sphenopalatine ganglionstimulation activates perivascular vasodilatory nerves in the ipsilateral

reserved.

221C. Yolas et al. / Journal of the Neurological Sciences 336 (2014) 220–226

anterior circle of Willis [13]. Many other studies on ganglionary andneuronal cell degenerations after SAH were published by variousauthors [14–17]. As a ganglion, cell changes CBs following SAH havenot been studied yet. In addition, there are no quantitative studies onthe response of the pulmonary arterial diameter to changes in livingand degenerated neuron number of GPNs and CBs. In this paper, authorsinvestigated the live and degenerated neuron numbers of GPNs and CBsfollowing SAH, and compared with pulmonary arterial diameters. Thissubject was studied for the first-time.

2. Materials and methods

This study was performed on 20 adult male New Zealand rabbits(3.7 ± 0.4 kg). Experiments were carried according to the guidelinesof the experimental research committee of the Ataturk University,Medical Faculty. Five animals (n = 5) were used as control group. Theremaining animals (n = 15) were exposed to experimental SAH.

2.1. Experiment

A balanced injectable anesthesia was used for reducing pain andmortality. After inducing anesthesia with isoflurane by a face mask,0.2 mL/kg of the anesthetic combination (ketamine HCL, 150 mg/1.5 mL; xylazine HCL, 30 mg/1.5 mL; and distilled water, 1 mL) wassubcutaneously injected before surgery. During the operation,0.1 mL/kg anesthetic combination was used when required. Onemilliliter of autologous blood was taken from the auricular arteryand injected into cisterna magna via 22-gauge needle in about1 min. In the six animals of the SAH group, severe signs of illnesswere observed, and these six animals were killed in the first weekafter SAH; the other animals (n = 9) were followed up for 20 dayswithout any medical treatment and then sacrificed.

2.2. Histological procedure

Perfusion fixation was used. Pulmonary arteries (PAs) were ex-amined as described by Dumlu et al. [18]. These arteries were obtain-ed from the longitudinal lungs sections at the levels of 5 mmdistances from the hilus. They were also stained with H&E dyes.For the calculation of vasospasm index, all PA samples were acceptedas a cylinder, and simple geometric formulas were used to estimatetheir surface areas. As a measure of the degree of vasospasm, theuse of PA vasospasm index (VSI) was preferred over the only mea-surement of lumen radius and volume values. PA of all animals wascut 20 segments away from the arising point of the main pulmonaryarteries to the entering points of lung tissue. Then, 20 histopatholog-ical sections, 5 μm apart, were obtained by microtome for each des-ignation and are represented by the lines 1, 2, 3,…20. The meanexternal diameter and internal (luminal) diameters of each sectionwere measured, and external radius was represented as R1 and in-ternal radius was represented as r1. The mean radius value of PAwas calculated as R1 = R1 + R2 + R3 + ….R20/20; and, lumen ra-dius was calculated as r1 = r1 + r2 + r3 + ….r20/20 (Fig. 3). Thewall ring surface values were calculated as the following formula:S1 = πR12 − πr12. The lumen surface area was calculated as thesame method. So, lumen surface value (S2) = πr12. The vasospasmindex was calculated as the proportion of S1/S2. Vasospasm index(VSI) = S1 / S2 = πR12 − πr12 / πr12 = π(R12 − r12) / πr12 =R12 − r12/r12. In summary, VSI = (R2 − r2) / r2. This vasospasmindex (VSI) was used to assess the severity of vasospasm. The valuebetween 1 and 1.5 was accepted as no vasospasm, 1.5–2 was accept-ed as light vasospasm and 2 or greater than 2 was accepted as severevasospasm. Data analysis was done by using two sample t test andcorrelation analyses.

2.3. Tissue processing

All GPNs and CBs were removed bilaterally. The rabbit carotid bodylies medial to the upper end of the common carotid artery and is closelyapplied to the adventitia of this vessel. Carotid bodieswere harvested byremoving the regions of the carotid bifurcations along with a portion ofthe sinus nerves. For GPN transection, the sublingual and submaxillarysalivary glands and the sternohyoid, omohyoid, and posterior belly ofthe digastric muscles were retracted, and the fascia underlying thehypoglossal nerve was carefully dissected to reveal the nerve near theexternal medial wall of the bulla. Sections of CB and GPN were cutinto small blocks, and placed in fresh fixative. They were stored in 10%formaldehyde at 4 °C for 7 days. The fixed tissues were dehydrated ingraded ethyl alcohol series, and embedded in paraffin. The sectionswere cut at a thickness of 5 μm with rotary microtome and collectedon glass-slides for histopathological and stereological analyses. Sectionsstainedwithhematoxylin–eosin (H&E)were examined via lightmicros-copy with attached digital camera and important sections werephotographed.

2.4. Stereological analyses

Stereological analyses of histopathological data were evaluatedaccording to the principles described previously [19–21]. To obtain anestimation of the total axon number, we used the two-dimensional dis-sector technique. A counting frame was placed on a monitor, and thesampled area was chosen by a systematic uniform random manner viathe dial indicator controlled specimen stage. This ensures that all loca-tions within a nerve cross section were equally represented [15]. Thephysical dissectormethodwas used to evaluate the numbers of neuronsof CBs and GPNs. This method can easily estimate the particle number,be readily performed, is intuitively simple, is free from assumptionsabout particle shape, size and orientation, and is unaffected by overpro-tection and truncation. Two consecutive sections (dissector pairs)obtained from tissue samples with named reference were mounted oneach slide. Reference and look-up sections were reversed in order todouble the number of dissector pairs without taking new sections. Themean numerical density of neurons of CBs (NvGN) per mm3 was esti-mated using the following formula:

NvGN ¼ ΣQ−N=t � A

where ΣQ − N is the total number of counted neurons appearing onlyin the reference sections, t is the section thickness, and A is the area ofthe counting frame. The Cavalieri volume estimation method was usedto obtain the total number of neurons in each specimen. This was calcu-lated by multiplication of the volume (mm3) and numerical density ofneurons in each CB and GPN. The number of living and degeneratedneurons of GPNs and CBs was counted for both right and left sides andmeans were calculated.

3. Results

As noted earlier, six of the animals of the SAH group (n = 6) werekilled within the first week and the remaining animals (n = 9) werefollowed 20 days. Clinically, meningeal irritation signs, consciousness,convulsive attacks, fever, apnea, cardiac arrhythmia and breath distur-bances were observed frequently in premortal periods of all of killedsix animals. In histopathological examination of GPNs, axonal andperi-axonal thinning, axonal loss and large inter-axonal space wereaccepted as axonal degeneration criteria. Fig. 1A,B shows stereologicalcell counting of the carotid body in a rabbit. The mean number of liveaxons in the GPNs of in rabbits was found as 1211.66 ± 14.29,1013.89 ± 41.84 and 558.66 ± 46 in the control group, the late killedand the early killed animals, respectively. The number was higher inthe late killed animals than that in early killed (p b 0.001). The mean

Fig. 1. Macroscopic appearance of brain with SAH (yellow arrow) and pulmonary contu-sion (black arrow).

222 C. Yolas et al. / Journal of the Neurological Sciences 336 (2014) 220–226

number of degenerated axons of GPNs was 5 ± 1, 159.33 ± 10.36 and530.83 ± 43.48 in controls, the late killed and the early killed animalsafter SAH, respectively (p b 0.001). All stereological results weresummarized in Table 1. In histopathological examination of CBs, cytoplas-mic condensation, nuclear shrinking, cellular angulations and peri-cytoplasmic halo formation secondary to cytoplasmic regression wereaccepted as the neuronal degeneration criteria for CBs neurons. Stereolog-ically, themean total volume of the CBswas calculated as 0.8 ± 0.2 mm3,and the mean neuron number of CBs was 4206.67 ± 148.35. The liveneuron numbers of CBs were 3997.77 ± 96.86 and 2270.5 ± 134.38 inthe late and early killed animals, respectively, which were higher in thelate killed animals than in the early killed one (p b 0.001). Also, themean numbers of degenerated neurons of CBs were 11.33 ± 3.21,255.78 ± 41.92, and 2065 ± 110.27 in controls, early killed and latekilled animals, respectively. The numbers were significantly higherin the early killed animals than those in the late killed animals(p b 0.001). Table 2 shows the mean number of live and degeneratedneurons in CBs by groups. There was a significant correlation betweenthe number of live axons of GPNs and the number of live CB neurons(r = 0.87) and the number of degenerated axons of GPNs and the num-ber of degenerated CB neurons (r = 0.92). There is significant linear cor-relation. This correlation shows a direct cause and effect relationshipbetween live and degenerated axons of GPN and CB neurons.

Fig. 2 shows histological appearance of a normal pulmonary artery.Mean VSI of pulmonary arteries and live and degenerated neuron den-sity of CB was 0.767 ± 0.041,7580 ± 500/mm3, 40 ± 11/mm3;1.218 ± 0.190, 6420 ± 450/mm3; 1260 ± 110/mm3; 1.745 ± 0.220,3840 ± 370/mm3; and 3250 ± 318/mm3 in the control group, and inanimals with slight and severe pulmonary artery vasospasm, respec-tively (see Table 1). Histopathological appearance of pulmonary arteryof an animal with light vasospasm was shown in Fig. 3. A less ischemiccarotid body is seen at the left upper corner. Fig. 4 shows histopatholog-ical appearance of severe vasospasm pulmonary artery of an animal

Table 1Mean numbers of live and degenerated axons of glossopharyngeal nerves by groups.

Groups Glossopharyngeal nerve

Live axon Degenerated

Mean ± SD Mean ± SD

Control (n = 5) 1211.67 ± 14.29 5 ± 1LK animal (n = 9) 1013.89 ± 41.84 159.33 ± 10EK animal (n = 6) 558.66 ± 46.38 530.83 ± 43

Degenerated axons were lower in the LK (late killed) animals (group III) than in the early kille

with massive SAH. Table 2 shows the relationship between vasospasmindex of pulmonary artery and live and degenerated neuron density ofCBs of groups. The VSI values were higher in early killed animals thanthose of the late killed animals (p b 0.05).

4. Discussion

At present, neurosurgical practice is confronted by an explosion oftechnology [22,23]; however, spontaneous SAH still results in signifi-cant morbidity and mortality, even among patients who reach medicalattention in good neurological condition. Many patients have neurolog-ical decline in the peri or post-operative period, which may lead tochange the outcome. The pathophysiology that immediately follows an-eurysm rupture has been recognized as the greatest contributor tomor-tality after SAH for over a decade [24]. It is well known that the patientswith SAH suffer from global or regional cerebral ischemia [25,26]. In pa-tients suffering from SAH, increasing intracranial pressure and decreas-ing cerebral perfusion pressure are common complications caused bymultiple pathological mechanisms [27]. In our study, six of the animalswith SAH (n = 6)were killed within the first week as a result of severesigns and symptoms of SAH. The major likely mechanism leading to se-vere clinical condition following SAH in these animals is GPN and CB in-jury secondary to increased intracranial pressure (ICP). Another mostimmediate event following the rupture of an intracranial aneurysm isthe acute rise in ICP [24]. It was recently reported that, at the earlyphase of subarachnoid hemorrhage, hydrocephalus may be stimulatedthe irritant receptors glossopharyngeal and vagal nerve endings whichinnervate the healthy choroid plexuses epithelium and arteries [28].Raised ICP following SAH predicts poor outcome and is due to hemor-rhage volume and possibly, brain edema, hydrocephalus and increasedvolume of circulating intracranial blood. Interventions that reduceedema may therefore reduce ICP and improve outcome. Authors de-scribe here, a mechanism for morbidity or mortality in SAH that in-volves communication between glossopharyngeal nerve and thecardiopulmonary system.

4.1. A neglected cranial nerve

Compared with other lower cranial nerves, the GPN is well hiddenwithin the jugular foramen, at the infratemporal fossa, and in the deeplayers of the neck [29].

Among the lower six cranial nerves, the glossopharyngeal nerve isthe smallest in terms of nerve diameter. It may be a neglected cranialnerve. The important branch of GPN is the carotid sinus nerve (nerveof Hering). This nerve supplies the carotid body and carotid sinus. It con-veys chemoreceptor and stretch baroreceptor information centrally forrespiratory and circulatory reflex function. It branches from theglossopharyngeal nerve at the level of the carotid bifurcation [30], sothis nerve is an information conductor between the CBs and the cardio-respiratory centers receive chemoreceptor andbaroreceptor afferent in-puts from the carotid body and carotid sinuses respectively. The cellbodies of these sensory cells lie in the petrous ganglion. The afferent fi-bers for chemoreceptors and baroreceptors pass on to the dorsal

Carotid bodies

axon Live neuron Degenerated neuron

Mean ± SD Mean ± SD

4206.67 ± 148.35 11.33 ± 3.21.36 3997.77 ± 6.86 255.78 ± 41.92.48 2270.5 ± 134.38 2065 ± 110.27

d (EK) animals (group II) (p b 0.001).

Table 2The relationship between vasospasm index of pulmonary artery and neuron density of CBs of groups.

Groups Carotid Bodies Pulmonary arteries

Live neuron density/mm3 Degenerated neuron density/mm3 Vasospasm index

Mean ± SD Mean ± SD Mean ± SD Severity

Control (n = 5) 7580 ± 500 40 ± 11 0.767 ± 0.041 NormalLK animals (n = 9) 6420 ± 450 1260 ± 110 1.218 ± 0.190 NormalEK animals (n = 6) 3840 ± 370 3250 ± 318 1.745 ± 0.220 Slight

223C. Yolas et al. / Journal of the Neurological Sciences 336 (2014) 220–226

nucleus of the vagus, which is the main autonomic nucleus, and fromthere to the respiratory and vasomotor centers [30]. In cranium, theGPN emerges from the post-olivary sulcus and course ventral to theflocculus and choroid plexus of the lateral recess of the fourth ventricle.The nerve then enters the jugular foramen through the uppermostporus [29]. GPNs injury may result in CBs denervation atrophy andcardiorespiratory reflexes can be disrupted. The hypothalamus mayalso have an important role in the modulation of cardiorespiratory re-flexes. In this study, intracranial changes following SAH probably ledto dysfunction of the efferent pathway or chemoreceptor afferents.Baroreceptor fibers were likely affected; we did not study the bloodpressure of animals; however, it is well known that blood pressure,the result of neural innervation from several systems, contributes tothe autonomic control of blood flow in the body [31]. Cholinergic inputsof hypothalamus and CBs received with GPNs and it is activated by hy-potension. Baroreceptor nerve endings that innervate the aortic archand carotid sinus and detect acute fluctuations in the arterial pressurenerve terminals of the baroreceptor reflexes are regulated by GPN[31]. In this study, dysfunction of these baroreceptor reflexes regulatedby GPN was likely the cause of bad outcome following SAH. The com-pression or irritability of the GPN nerve is the most likely pathophysio-logic mechanism that can account for severe illness. Degeneration maylead to a heightened vagal response. This may result in cardiac dys-rhythmia, bradycardia, hypotension, and even asystole and subsequentsyncope and death. This effect is similar to that seen in carotid sinusmassage for the treatment of supraventricular tachycardias. Massagingthe carotid sinus causes a hyperstimulation of the glossopharyngeal af-ferent pathway, resulting in an exaggerated parasympathetic vagal ef-ferent response [30]. In some instances, as in our killed animals, anirritated GPN following SAH can also be potentially deadly and life-threatening by producing pulmonary arterial vasospasm secondary tocarotid body degeneration

Fig. 2. CB and its afferent nerve GPN (LM, H&E, ×40).

4.2. Degeneration of the carotid body

The human carotid body is located at the bifurcation of the commoncarotid artery into the internal and external carotid arteries. It is themost vascularized and chemosensitive structures in the body [32].On the basis of animal models, it has been hypothesized that thechemosensitive elements in the CB are the afferent nerve terminalswhose cell bodies are located in the sensory ganglia of theglossopharyngeal nerve, with central projections to respiratory centersin the brain stem. Stimuli such as hypoxia, hypercapnia, and acidosisevoke action potentials along the afferent terminals and through secre-tion of a neurotransmitter cause release of dopamine from chief cells[33]. A pronounced activation of the sympathetic nervous system andthe inflammatory response may occur in SAH. Fig. 1A,B shows stereo-logical cell counting of the carotid body in a rabbit. Mean number ofdegenerated neurons was significantly higher in the killed animalsthan in the living animals (p b 0.001). Table 1 shows the mean numberof live and degenerated neurons in CBs in groups. These results likelyoccurred because of the injury of the CB. Aydin et al. previously reportedthat ischemic neuronal damage of the CBs may be recovered via earlyrevascularization of the CBs via posterior cerebral circulation [32].

4.3. Pulmonary arterial vasospasm

CBs are assumed to play a similar homeostatic role in ventilatoryregulation in humans [11], and they may be very important in mediat-ing rapid ventilatory responses to hypoxia, metabolic acidosis, ordynamic states ofmuscular exercise [34]. The compression or irritabilityof the GPN may affect the CBs reflexes. These reflexes arising from thecarotid bodies may play an important role in respiratory changes inSAH. Some studies have stressed the importance of extracerebral

Fig. 3. Histological view of GPN with degenerated axons (red arrow) in a rabbit withsevere SAH signs (LM, H&E, ×20).

Fig. 4. Histological appearance of lung and pulmonary artery of a rabbit without SAH (LMH&E, ×20). Normal CB neuron (NN) at the right upper corner (LM, H&E, ×100).

Fig. 6. Histological view of a rabbit with severe pulmonary artery vasospasm after severeCB degeneration at the right upper corner (LM, H&E, ×20; LM, H&E, ×100).

224 C. Yolas et al. / Journal of the Neurological Sciences 336 (2014) 220–226

organ dysfunction on the development of delayed ischemic deficitsmore than the intrinsic cerebral pathophysiologic process [35]. Pulmo-nary hypertension is the most troubling complication of subarachnoidhemorrhages [18]. When cardiopulmonary complications occur, suc-cessful microsurgical clipping might result in an unfavorable outcome.Pulmonary arterial narrowing, unstable cardiac performance resultingin hypoperfusion, and low ventilation are major factors hinderinggood outcome. The VSI values were higher in killed animals with severedegeneration CB andGPN than those of the living animals (Table 2). His-tological appearance of a normal pulmonary artery in Fig. 2, light vaso-spasm in Fig. 3, and severe vasospasm in Fig. 4 were shown. In thisstudy, we showed that these major factors were augmented with highnumber degeneratedneuron of GPN and CB. It is of considerable interestto see that the degeneration of the carotid body led to pulmonary arte-rial narrowing. This narrowing is likely the cause of deaths of this animal(Figs. 5 and 6).

4.4. Importance of the present study

Any contribution to our knowledge of the cause of themorbidity andmortality of SAH is always welcome. Present study has four major

Fig. 5. Histological section of a rabbit with slight pulmonary artery vasospasm after littleCB degeneration at the right upper corner (LM, H&E, ×20; LM, H&E, ×100).

findings for unfavorable outcome following SAH. (a) Degenerative injuryof glossopharyngeal nerve occurred in killed animalswith severe signs ofSAH. (b) This injury led to degeneration of carotid body. (c) Degenerationof carotid body resulted in vasospasm of pulmonary arteries. All of thesefinding may be the cause of mortality in subarachnoid hemorrhage. Insummary, we demonstrated degeneration of GPN axons and CB neuronsin an experimentalmodel of SAH. This is a novel findingwhichmay havean important role in respiratory and cardiopulmonary complications fol-lowing SAH. Such changes were reported for the first time, and a mech-anism for morbidity or mortality that involves communication betweenneural networks and the cardiopulmonary system was proposed. Wethought that this injury was probably produced by pulmonary arterialvasospasm secondary to GPN and carotid body cell degeneration. Thereare also other mechanisms that may induce cardiorespiratory dysfunc-tion as a consequence of SAH. Increased ICP and herniation may affectthe solitarius nucleus as well as other cardiorespiratory centers in themedulla by local compression. In addition cranial vessels vasospasmmay also produce ischemic injury to the medulla. Our theory in thisstudy is very interesting. The cause of worse outcome following SAHsuch as decreased airway diameter and hypoxemia may be preventedby early surgical clipping and evacuation of SAH before degenerative in-jury of GPN and CB cells occurs. Airways responses may be due to thestimulation of the carotid body chemoreceptors. Functional outcomehas become a key parameter for the determination of the efficacy of ther-apeutic interventions [36]. It is accepted now that the best treatment ofaneurysmal SAH is early surgical intervention [37]. However, today, thetraditional notion of aneurysmal obliterationwithmicrosurgical clippingis currently being challenged, with an evolving body of knowledge dem-onstrating that endovascular treatment may have lower mortalities andmorbidities. It has been suggested that coiling should be considered asfirst-line treatment in aneurysmal SAH [38]. Howcanendovascular treat-ment of an aneurysm prevent the glossopharyngeal nerve injury or brainherniation? SAH clot removal which can be only made by early surgerymay be effective in preventing GPN cell injury. It seems to be imperativeto understand this process to better target treatments. Although thisstudy is experimental and deals with rabbit SAH model, it may havesome relevance to the human SAH model. Authors suggest thatglossopharyngeal is injured due to SAH and aggravates themortal effectsof SAH.We copiously stress the fact thatwe are thefirst ones to report onGPN injury following SAHand that the recognition of this fact is of impor-tance. If indeed one is the first to report something, that something is ofvalue. The opening of new horizons of this kind of knowledge will helpunderstand the complex challenge of SAH. More data are needed onthis subject.

225C. Yolas et al. / Journal of the Neurological Sciences 336 (2014) 220–226

4.5. Study limitations

This is a study of rabbit SAH showing an association of decreased ca-rotid body neurons and glossopharyngeal nerve axons, and several limi-tations of this study deserve mention. Stereological methods were usedfor the determination of degenerative changes of the GPNs and CBs,and an estimate of the number of live or degenerated neurons in eachspecimen was the basis of our results. It was concluded that “the causeof cardiorespiratory arrest in SAH may be due to secondary injury tothese nerve structures due to herniation”, but we have not proven herni-ation in this experiment; however the cause and effect relationship wasrecognized at the end of the experiment.Whenwe obtained our final re-sults, control examination of other nerves and vessels and brainstemher-niation had not been performed. This is a great disadvantage of the study.In the future, this subject should be studied; however, theremay be othermechanisms that may induce cardiorespiratory dysfunction as a conse-quence of SAH. Increased ICP and herniationmay affect the solitarius nu-cleus as well as other cardiorespiratory centers in the medulla by localcompression or direct compression to brainstem, blood flow impairmentto brainstem, catecholamine surge, and so on. Vagal and aortic body re-flexes perturbations might be accepted as another possible mechanismimpairing cardiorespiratory functions and should be further investigated.In addition cranial vessels vasospasm may also produce ischemic injuryto themedulla. Another control group (SHAM group) of saline injectionsto the cisternamagnamight have been a better control group to evaluatethe mechanisms of these consequences. However, narrowing of pulmo-nary arteries and injuries of GPN and CB have been reported for thefirst time in this study. The pathologic changes which were found inthe present study may occur not only due to the injury of GPNs andCB. Perhaps the most important limitation of the study is that it is diffi-cult to evaluate these changes in vivo or through autopsy, particularly inman with SAH. Another limitation is that the experimental animalmodel of SAH may not accurately mimic the human disease process[39].

5. Conclusion

Systemic organ involvement following SAH is associated with in-creased mortality and neurological impairment, even after adjustmentfor other outcome predictors such as the severity of the initial neurolog-ical injury. This may be a reflection of secondary brain injury precipitat-ed by hypoxemia, circulatory failure, fever, or hyperglycemia, all ofwhich have been linked to adverse clinical outcomes. In addition,brain-stem herniation is known to be one of the most important factorsin both the etiology of respiratory disturbances and sudden death inneurosurgical practices [40]. Interventions to avert or reverse theseand other perturbations need to be tested in clinical trials as they repre-sent opportunities to improve survival and neurological recovery in pa-tients with SAH [7]. In the present study, the number of degeneratedaxons of GPN and neurons of CBswas higher in the dead than in the liv-ing animals after SAH. Also, there was correlation between the numberof live axons of GPNs and the number of live CBs neurons. The same cor-relation was seen between the number of degenerated axons of GPNsand the number of degenerated CBs neurons. It can be stated that SAHconcomitantly caused neurodegeneration in bothGPNs and CBs. This in-vestigation has revealed several new features relevant to the role of thecarotid body following SAH. Possibly, an important cause of worse out-come in SAH may be the axonal degeneration of GPNs and neuronaldeath in CBs by the destruction of chemoreceptor and/or baroreceptorreflex arcs directed by GPN and CB.

Conflict of interest

There is no conflict of interest.

References

[1] Szydełko M, Kwolek A, Druzbicki M. Results of rehabilitation in patients after sub-arachnoid haemorrhage from ruptured intracranial aneurysm and after surgicaltreatment. Neurol Neurochir Pol 2008;42:116–22.

[2] Morga R, Czepko R, Dembińska-Kieć A, Danilewicz B. Assessment of the haemostaticsystem in patients surgically treated for ruptured cerebral aneurysm. NeurolNeurochir Pol 2007;41:296–305.

[3] Wong GK, Wong R, Mok V, Wong A, Poon WS. Natural history and medical treat-ment of cognitive dysfunction after spontaneous subarachnoid haemorrhage: re-view of current literature with respect to aneurysm treatment. J Neurol Sci 152010;299(1-2):5–8.

[4] Cagiran E, Derbent A, Cagiran I, Yuksel UG, Dalbasti T. Admission blood glucose andmorbidity–mortality in subarachnoid hemorrhage patients. J Neurol Sci (Turk)2013;30:168–77.

[5] Weaver JP, Fisher M. Subarachnoid hemorrhage: an update of pathogenesis, diagno-sis and management. J Neurol Sci 1994;125:119–31.

[6] Komotar RJ, Zacharia BE, Valhora R, Mocco J, Connolly Jr ES. Advances in vasospasmtreatment and prevention. J Neurol Sci 15 2007;261(1-2):134–42.

[7] Stevens RD, Nyquist PA. The systemic implications of aneurysmal subarachnoidhemorrhage. J Neurol Sci 15 2007;261(1–2):143–56.

[8] Lambert E, Du XJ, Percy E, Lambert G. Cardiac response to norepinephrine and sym-pathetic nerve stimulation following experimental subarachnoid hemorrhage. JNeurol Sci 15 2002;198(1-2):43–50.

[9] Warnell P. The cardiopulmonary complications of aneurysmal subarachnoid hemor-rhage: current trends in management. Axone 1992;14:24–8.

[10] Cakir M, Sengul G, Aydın MD, Gursan N. The role of vagal nerve root injury on respi-ration disorders in subarachnoid hemorrhage. Turk Neurosurg 2013. http://dx.doi.org/10.5137/1019-5149.JTN.7062-12.1 [accepted for publication].

[11] Prabhakar NR, Jacono FJ. Cellular andmolecular mechanisms associated with carotidbody adaptations to chronic hypoxia. High Alt Med Biol 2005;6:112–20.

[12] Lecompte PM. Tumors of the carotid body. Am J Pathol 1948;24:305–37.[13] Takahashi M, Zhang ZD, Macdonald RL. Sphenopalatine ganglion stimulation

for vasospasm after experimental subarachnoid hemorrhage. J Neurosurg 2011;114:1104–9.

[14] Aydin MD, Akyol SI, Sahin O. Histopathological changes in ciliary ganglion of rabbitswith subarachnoid hemorrhage. Int J Neurosci 2005;115:1595–602.

[15] Aydin MD, Serarslan Y, Gundogdu C, Aydin N, Aygul R, Kotan D, et al. The relation-ship between the neuron density of the trigeminal ganglion and the posterior com-municating artery vasospasm in subarachnoid hemorrhage: an experimental study.Neurosurg Q 2012;22:1–6.

[16] Kanat A, Yilmaz A, Aydin MD, Musluman M, Altas S, Gursan N. Role of degeneratedneuron density of dorsal root ganglion on anterior spinal artery vasospasm insubarachnoid hemorrhage: experimental study. Acta Neurochir (Wien) 2010;152:2167–72.

[17] Kayaci S, Kanat A, Aydin MD, Musluman AM, Eseoglu M, Karalar M, et al. Role ofneuron density of the stellate ganglion on regulation of the basilar artery volumein subarachnoid hemorrhage: an experimental study. Auton Neurosci 72011;165:163–7.

[18] Aydin MD, Ungoren MK, Izci Y, Acikel M, Gundogdu C, Gursan N. Vagal nerve degen-eration and pulmonary artery vasospasm after subarachnoid hemorrhage. J NeurolSci (Turk) 2013;30:369–76.

[19] Gundersen HJ, Bendtsen TF, Korbo L, Marcussen N, Møller A, Nielsen K, et al. Somenew, simple and efficient stereological methods and their use in pathological re-search and diagnosis. APMIS 1988;96:379–94.

[20] Gundersen HJ. Stereology of arbitrary particles: a review of unbiased number andsize estimators and the presentation of some new ones, in memory of William RThompson. J Microsc 1986;143:3–45.

[21] Sterio DC. The unbiased estimation of number and sizes of arbitrary particles usingthe dissector. J Microsc 1981;34:127–36.

[22] Kanat A, Epstein CR. Challenges to neurosurgical professionalism. Clin NeurolNeurosurg 2010;112:839–43.

[23] Kanat A, Yazar U, Kazdal H, Yilmaz A, Musluman M. Neurosurgery is a profession.Neurol Neurochir Pol 2009;43:286–8.

[24] Ayer R, Zhang J. Connecting the early brain injury of aneurysmal subarachnoid hem-orrhage to clinical practice. Turk Neurosurg 2010;20:159–66.

[25] Durmaz R, Ozkara E, Kanbak G, Arslan OC, Dokumacioğlu A, Kartkaya K, et al. Nitricoxide level and adenosine deaminase activity in cerebrospinal fluid of patients withsubarachnoid hemorrhage. Turk Neurosurg 2008;18:157–64.

[26] Erdi MF, Guney O, Kiyici A, Esen H. The effects of alpha lipoic acid on cerebral vaso-spasm following experimental subarachnoid hemorrhage in the rabbit. TurkNeurosurg 2011;21:527–33.

[27] von Gottberg P, Psychogios M, Schuetze G, Cohnen J, Zwaka P, Knauth M, et al. Fea-sibility of flat panel detector computed tomography for position assessment of ex-ternal ventricular drainage. Neurol Neurochir Pol 2013;47:32–42.

[28] Kanat A, Turkmenoglu O, AydinMD, Yolas C, Aydin N, Gursan N, et al. Toward changingof the pathophysiologic basis of acute hydrocephalus after subarachnoid hemorrhage: apreliminary experimental study. World Neurosurg 2013;80(3-4):390–5.

[29] OzverenMF, TureU,OzekMM,PamirMN.Anatomic landmarks of theglossopharyngealnerve: a microsurgical anatomic study. Neurosurgery 2003;52:1400–10.

[30] Soh KBK. The glossopharyngeal nerve, glossopharyngeal neuralgia and the Eagle'ssyndrome—current concepts and management. Singapore Med J 1999;40.

[31] Musluman AM, Aydin MD, Yilmaz A, Canseven T, Kanat A, Gundogdu C. The effect ofdegenerated neuron density of petrosal ganglion on the development of blood pres-sure variabilities after subarachnoid hemorrhage in a rabbit model: an experimentalstudy. Turk Neurosurg 2011;21:1–8.

226 C. Yolas et al. / Journal of the Neurological Sciences 336 (2014) 220–226

[32] Aydin MD, Ozkan U, Gundogdu C, Onder A. Protective effect of posterior cerebral cir-culation on carotid body ischemia. Acta Neurochir 2002;144:369–72.

[33] McDonald DM. Regulation of chemoreceptor sensitivity in the carotid body: the roleof presynaptic sensory nerves. Fed Proc 1980;39:2627–35.

[34] Whipp BJ, Wasserman K. Carotid bodies and ventilatory control dynamics in man.Fed Proc 1980;39:2668–73.

[35] Solenski NJ, Haley Jr EC, Kassell NF, Kongable G, Germanson T, Truskowski L, et al.Medical complications of aneurysmal subarachnoid hemorrhage: a report of themulticenter, cooperative aneurysm study. Crit Care Med 1995;23:1007–17.

[36] Thal SC, Mebmer K, Schmid-Elsaesser R, Zausinger S. Neurological impairment inrats after subarachnoid hemorrhage—a comparison of functional tests. J Neurol Sci15 2008;268(1–2):150–9.

[37] Tuzgen S, Kucukyuruk B, Aydin S, Ozlen F, Kizilkilic O, Abuzayed B. Decompressivecraniectomy in patients with cerebral infarction due to malignant vasospasm afteraneurysmal subarachnoid hemorrhage. J Neurosci Rural Pract 2012;3:251–5.

[38] Kanat A, Kayaci S, Yazar U, Sahin Y, Yaman O, Guvercin AR. One of the giants of neu-rological surgery left usmore than a decade ago, and neurosurgical literature did notshow much interest. Neurol Neurochir Pol 2011;45:63–7.

[39] Whitmore RG, Grant RA, Leroux P, El-Falaki O, Stein SC. How large is the typical sub-arachnoid hemorrhage? A review of current neurosurgical knowledge. WorldNeurosurg 2012;77:686–97.

[40] Aydin MD, Eroglu A, Turkyilmaz A, Erdem AF, Alıcı HA, Aydin N, et al. The contribu-tion of chemoreceptor-network injury to the development of respiratory arrest fol-lowing subarachnoid hemorrhage. EAJM 2010;42:47–52.