New Insights Into the Pathophysiology of Carotid Sinus Syndrome
Background The pathophysiology of carotid sinus syndrome remains poorly understood. Currently, two main hypotheses are provided: a lesion at the level of carotid sinus receptors or a central defect at the level of the nuclei of the autonomic nervous system. The objective of our study was to present arguments in favor of one of these two hypotheses.
Methods and Results Test selection was guided by the following hypothesis: a degenerative central or local lesion could be associated with dysfunctions in the structures surrounding or comprising the baroreflex centers or their pathways. To test this hypothesis, brain stem auditory–evoked potentials; somatosensory–evoked potentials; blink reflexes; sympathetic skin responses; and styloglossus, sternocleidomastoid, and superior trapezius muscle electromyography were systematically performed from the right and left sides in 17 patients with carotid sinus syndrome and in 17 sex- and age-matched control subjects. Similar responses were found in the two groups for the “central” tests. Contrasting with this result, the electromyographic analysis of the sternocleidomastoid muscle differed significantly between the groups: 13 (76%) had pathological responses in the carotid sinus syndrome group compared with only 4 (23.5%) in the control group (P<.01). Furthermore, the abnormality was found on the right and left sides in 9 patients (53%) in the study group and in none of the control group (P<.005).
Conclusions This study strongly suggests that the neuromuscular structures surrounding the carotid mechanoreceptors are involved in the carotid sinus syndrome; however, the exact mechanism remains speculative.
Some sixty years ago, Roskam1 reported the first recognized case of CSS. Three years later, Weiss and Baker2 collected a series of 15 patients and established that CSS was a cause of syncope. However, because treatment was generally unsuccessful, this syndrome was neglected over the next 4 decades. Interest in CSS was rekindled in 1970 when Voss3 published the first case of pacemaker implantation for this condition. In the last 2 decades, many large series have demonstrated the efficacy of pacing,4 5 particularly dual-chamber pacing.6 7 In contrast with this therapeutic progress, the pathophysiology of the syndrome remains poorly understood. In patients with carotid sinus hypersensitivity and tumors of the neck, the exact site of abnormality in the reflex arc seems obvious,8 9 10 11 but these patients represent only a minority of those with CSS, and in most cases, the syndrome remains idiopathic. Some authors have considered CSS a manifestation of sinoatrial disease.12 13 However, patients with CSS usually have normal sinus node function14 but frequently show an exaggerated vasodepressor response.15 16 17 This does not implicate the sinus node and suggests that the abnormality does not lie in the afferent limb of the reflex arc. Some authors have favored an abnormality at the level of the carotid sinus receptors2 18 ; others have considered a central defect.19 Because no hard data substantiate these positions, the exact site (or sites) of hypersensitivity in the reflex arc in patients with CSS remains unknown.
Two groups of patients were selected. Group 1 included patients who met the following criteria: (1) syncope defined as an abrupt loss of consciousness; (2) no known cause of syncope after clinical and laboratory evaluation; (3) abnormal reproducible response to carotid sinus massage (see below) in the absence of drugs known to induce such response, particularly β-blockers; (4) no local cause of compression of the carotid glomus (eg, previous or present neck tumors or adenopathy); and (5) informed consent to undergo noninvasive neurological screening.
Patients were recruited prospectively among those admitted for evaluation of syncope or retrospectively among those attending for pacemaker surveillance. The study group (Table 1⇓) consisted of 17 patients (16 men; mean age, 64.9±11.2 years). All patients had spontaneous syncope (from 1 to >10 episodes); in 5 patients, it resulted in severe injuries. The diagnosis was made after the first episode in 4 patients but after 10 or more episodes in 2 patients. The time between the first episode of syncope and neurological testing varied from 15 days to 22 years (mean, 7±6 years). In 6 patients, a vasodepressor response was evaluated by continuous intra-arterial pressure monitoring during carotid sinus massage while the patient was paced in DDD mode. Of those 6 patients, 3 had a positive response defined as a drop in systolic pressure of ≥50 mm Hg. All patients received a permanent pacemaker (6 VVI pacemakers implanted initially that were upgraded to DDD at the time of replacement and 11 dual-chamber pacemakers).
Group 2 consisted of control patients who (1) never experienced syncope, (2) had a negative result to two carotid massages separated by at least 2 hours, and (3) gave informed consent to undergo neurophysiological screening.
The control patients were matched to CSS patients for sex and age (age of the study patient±2 years) and were investigated during the same period of time. The control group included 17 patients (16 men; mean age, 64.1±11.8 years). One was treated for diabetes, 2 were drinkers of >80 g/d alcohol, 4 were hypertensive, 10 had documented ischemic heart disease, 2 had valvular heart disease, and 3 were admitted for paroxysmal arrhythmia (atrial fibrillation, junctional tachycardia, and ventricular tachycardia). None had invasive evaluation of the systolic pressure during carotid sinus massage; blood pressure was determined by a cuff sphygmomanometer, and no drop of >20 mm Hg was recorded.
Patients were considered to have unexplained syncope as defined by Kapoor20 if the following series of investigations remained inconclusive: history and clinical examination, routine laboratory tests, orthostatic blood pressure determination, standard 12-lead ECG, 24-hour Holter monitoring, and electrophysiological studies (11 of 17 patients). Carotid sinus massage was performed in supine patients for 5 to 7 seconds with continuous ECG monitoring. For this study, a positive test was defined as ventricular asystole lasting >3 seconds. A second massage was performed at least 30 minutes after the first while the patient was sitting. All patients included in the study group had dizziness during this second procedure.
In the control subjects, carotid sinus massage was performed with patients in the supine position with the same protocol as in the study group. Two patients screened for inclusion in the control group were later excluded because they experienced a pathological response to carotid sinus massage.
The tests performed in this study were guided by the following hypothesis: a central lesion, if responsible for CSS, could be associated with dysfunctions in the structures surrounding the baroreflex pathways. The physiological behavior of the neurological structures around the nucleus tractus solitarii, which could not be directly tested, were investigated systematically and successively from the right and left sides at least twice.
Brain Stem Auditory–Evoked Potentials
The BAEPs were obtained in a quiet, partially soundproof room with subjects sitting comfortably in a reclining chair. Alternating, nonfiltered clicks of 0.2-ms duration, 10-Hz frequency, and 80-dB intensity were delivered through an earphone. The electrophysiological activity of the brain stem was recorded between an active vertex electrode (Cz) and a reference electrode placed over the mastoid process of the side being stimulated. A frontal electrode was used to ground the subject. After amplification (band pass, 100 to 1500 Hz), the 1500 responses were averaged over an analysis time of 10 ms. The BAEPs were elicited by binaural and monoaural stimulation. In the latter case, the ear contralateral to the stimulation was masked by white noise of 60-dBHL intensity.
Jewett and Williston21 defined five peaks (I through V), the presumed origins of which are the distal cochlear nerve (peak I), proximal cochlear nerve (peak II), superior olivary complex (peak III), lateral lemniscus (peak IV), and inferior colliculus (peak V). We compared the interpeak latencies of peaks I-III, III-IV, and I-V in both groups. This examination (Fig 1⇓) explores the auditory pathway from the cochlear nerve to the superior part of the midbrain.
The SSEPs were obtained in subjects comfortably sitting in a reclining chair by stimulation given through surface electrodes placed in relation to the median nerve at the level of the wrist. The receiving electrodes were located in the supraclavicular fossa, at the level of the second cervical vertebra, and on the scalp in the parietal area contralateral to the side of stimulation (2 cm behind and 7 cm lateral to the Cz). The reference electrode was placed frontally (Fz) for reception at the Erb point and at the level of the second cervical vertebra. It was placed on the lobule of the ear, contralateral to the side of stimulation for the encephalic potential. The stimuli were rectangular pulses of 0.3-ms duration delivered at a frequency of 1 Hz. The stimulation intensity was fixed at 1.5 times the motor threshold. After amplification, the 300 evoked responses (band pass, 30 to 1000 Hz) were averaged over an analysis time of 50 ms. The latencies and amplitudes of peaks were measured at the ERB point (N10, negative response recorded 10 ms after stimulation), at the cervical level C2 (N13), and at parietal level (N20). In both groups, we analyzed the interpeak latency of N13 to N20. This examination (Fig 2⇓) explores the proprioceptive and superficial sensory pathways from the periphery to the parietal cortex through the brain stem.
BRs were obtained in supine, relaxed subjects after stimulation delivered by surface electrodes; the cathode was applied at the emergence point of the supraorbital nerve (a collateral of the ophthalmic branch of the trigeminal nerve). Shocks of 0.5-ms duration, the intensity of which was irregularly varied to avoid habituation, ranging from 30 to 80 mA were delivered. The detection of signals was obtained by two pairs of cup-shaped surface electrodes: the active electrode was applied below the inferior eyelid in the inferolateral quadrant of the orbit and the indifferent electrode on the external orbital margin. The grounding electrode was placed around the arm (band pass, 20 to 2000 Hz).
Several trials were made with increasing current intensity to obtain the best response: when the latency was the shortest and the amplitude the largest. Each electric stimulus of the supraorbital nerve evoked on the ipsilateral side an early reflex response and a late bilateral response.22 The early reflex response (oligosynaptic) results from direct connections at the level of the pons between the principal sensory nucleus of the trigeminal nerve and the motor nucleus of the facial nerve. The late reflex responses are polysynaptic: in the afferent limbs, the sensory fibers of the trigeminal nerve reach the principal sensory nucleus, particularly the inferior trigeminal segment that descends just to the level of the second cervical segment of the spinal cord. This nucleus distributes by direct or crossed pathways to the facial motor nucleus (Fig 3⇓). Early and late reflex response latencies were studied.
Sympathetic Skin Responses
SSRs were obtained in supine, relaxed subjects after electric stimulation delivered by surface electrodes placed in relation to the median nerve at the wrist. The receiving electrodes were placed on the palmar and dorsal surfaces of the hands. The stimuli were rectangular 0.2-ms-long pulses delivered at irregular intervals and at increasing intensities. After amplification (band pass, 0.16 to 3200 Hz), the five evoked responses were averaged over an analysis time of 10 seconds. The cutaneous nervous response is a slow electric wave evoked by activation of postganglionic sympathetic fibers that, in turn, activate the sweat glands in a synchronous manner. The studied parameter was the upper limb SSR latency.
Styloglossus, Sternocleidomastoid, and Superior Trapezius Muscle Electromyography
Electromyographic analysis of the muscles was performed on the right and left sides. The styloglossus muscle is innervated by the hypoglossal nerve, the nucleus of which is composed of large multipolar cells arranged in an extended column along the whole length of the medulla oblongata to the floor of the fourth ventricle. The sternocleidomastoid and superior trapezius muscles are innervated by the spinal nerves, the nuclei of which are formed by lateral cell groups of the anterior horns of the spinal cord extending from the sixth cervical segment to the base of the brain stem (Fig 4⇓). The electrical activities of these muscles were detected by a Bronck’s coaxial needle electrode. In the normal resting muscle, there is no electrical activity. In the case of increasing voluntary muscular contraction, a single potential pulsating at a frequency of <10 Hz is recorded for a weak contraction; for a contraction of moderate force, two or three potentials are recorded; for a maximal contraction against resistance, there are sustained potentials interfering with one another. Each potential is the sum of the action potentials of all the muscle fibers of the motor unit, which are set in activity together because they are innervated by the same motor neurone. In the case of neurological affection, these potentials have an abnormal firing rate with effort (Fig 5⇓).
All data are expressed as mean±SEM. Comparisons between categorical data were made with the χ2 with Yates’ correction for electromyographic analysis and the unpaired Student’s t test for the other parameters. A significance level of P<.05 was assumed.
Neurophysiological investigations were completed without any marked side effects. As noted in Table 2⇓, the results for BAEPs, SSEPs, BR, and SSR were not statistically different between the groups. The electromyographic analysis of styloglossus and upper trapezius was normal in all the patients and control subjects (Table 3⇓). Contrasting with this result, the electromyographic analysis of the sternocleidomastoids differs markedly between the groups: 13 of the 17 patients (76.5%) in the study group had a pathological result compared with only 4 (23.5%) in the control group (P<.01). Furthermore, the abnormality was found on both the right and left sides (Table 4⇓) in 9 patients (53%) of the study group and in none of the control group (P<.01).
In 4 of the 17 patients in the study group, electromyographic analysis was normal. The time elapsed between the first syncope and analysis was ≤1 month in 2 of these patients (for the other 2 patients, it was 5 and 6 years); in the remaining 13 patients of the study group, only 1 had a delay between symptoms and analysis ≤1 month (P=NS).
The objective of the present study was to compare the results of neurophysiological testing in two groups of age- and sex-matched patients who differed by the existence of overt CSS in one group. The principal findings were the documentation of very similar results of the central tests in the two groups and the unexpected observation of a very high proportion of abnormal sternocleidomastoid electromyographic results in the CSS group.
Previous studies of the pathophysiology of CSS have presented conflicting views.2 18 19 In fact, very few articles have been devoted to this subject, and our knowledge in the field is the result of intellectual speculation rather than experimental data. Two theories are proposed as the cause of CSS: a lesion of the nucleus tractus solitarii and a peripheral disease at the level of the carotid sinus baroreceptors. An argument for the latter hypothesis was the demonstration of a high proportion of CSS in patients with treated or untreated neoplastic neck tumors.8 9 10 11 However, in idiopathic CSS, the few histological reports in the literature have failed to document abnormality in the carotid baroreceptors.23 Furthermore, the different responses obtained during carotid massage (cardioinhibitory and/or vasodepressive) were considered an argument against an abnormality in the afferent part of the reflex pathway, as was the finding of asystole exceeding the duration of carotid sinus massage. The central structures are the most difficult part of the carotid sinus reflex pathway to study in humans, which explains why the hypothesis that this site is diseased in CSS is based on the exclusion of the responsibility of other sites.
CSS usually is diagnosed in patients in their 60s or 70s.4 5 6 At this age, a degenerative process seems a likely explanation of CSS, and if it occurs in the central nervous system, many structures besides the nucleus tractus solitarii also may be damaged at random. In our study, the values of all central tests are very similar between the two groups. This result seems to rule out the hypothesis of an organic central defect in the reflex pathway.
Proposed Explanations for Sternocleidomastoid Involvement
The finding of a neurogenic degenerative process of the sternocleidomastoid muscles in patients with CSS has not been previously reported and raises many questions. Because vasomotor centers are far from the motor nuclei of the sternocleidomastoid, a central impairment is unlikely; a pathological process involving these structures simultaneously would be expected to cause many other clinical disturbances. Chronic loss of innervation of the sternocleidomastoid is a strong argument for a peripheral origin of CSS, the result being an increased sensitivity of the baroreflex arc.
Theoretically, three explanations of such a pathological association have to be considered. The first explanation is that a common process leads to both CSS and denervation of the sternocleidomastoid. The sternocleidomastoid and the upper trapezius receive double innervation by the spinal nerve and the cervical plexus, occurring mostly from the second, third, and fourth cervical roots through anastomoses. Numerous interindividual variations of these anatomoses have been described.24 No anatomic relationship can be found between the baroreflex peripheral arc in its classic conception and the motor innervation tracts of the sternocleidomastoid.
The second explanation is that CSS causes sternocleidomastoid denervation. Because the neurological process seems to be limited to the sternocleidomastoid, no argument substantiates this hypothesis.
The final reasoning is that the neurological involvement of sternocleidomastoid leads to CSS. No evident explanation can be offered to confirm this relationship. However, an original hypothesis could be developed. The neurogenic abnormalities in the sternocleidomastoid are consistent with a lesion of the terminal motor pathways of this muscle, with a normal upper trapezius. They also can be considered a strong indirect argument for involvement of the proprioceptive sensory pathways. Proprioceptive sensitivity provides the medulla and higher centers with information about the contraction state (neuromuscular spindles) and the state of stretch (Golgi apparatus) of the sternocleidomastoid, particularly during active or passive movement. Animal experiments have clearly established the functional implication of the proprioceptive pathways in baroreflex regulation.25 On the other hand, the baroreceptors in the carotid sinus are considered to be stretch receptors that respond to deformation of the vessel wall in any direction.25 During neck movements, mild stretches of both carotid receptors and sternocleidomastoid tension receptors occur. Our hypothesis is that proprioceptive sensitivity of the sternocleidomastoids could provide important information on the mechanical origin of the baroreceptor stretching during neck movement and could modulate information coming from the carotid baroreceptors. Thus, the sternocleidomastoid proprioceptive information could produce a strong inhibitory effect on the baroreflex arc, according to the concept of a central gating of the baroreflex as developed for cardiorespiratory interactions.26
Finally, an impairment of these proprioceptive sensory pathways, whatever the anatomic level or pathological mechanism, would liberate the baroreflex arc, which could then respond inappropriately to mechanical extravascular stimulation of the carotid sinus but is perceived and integrated as a pressure elevation in the vessels. The close anatomic relationship between the sternocleidomastoid and the carotid bifurcation, the spatial orientation of the muscle led to consideration of the sternocleidomastoid as a privileged muscular structure for such a baroreflex regulation. The pathological process leading to chronic terminal denervation of the sternocleidomastoid remains unknown (vascular?); further studies are needed.
Some patients with CSS have normal sternocleidomastoids. This situation was encountered in only 4 patients, and no definite explanation was found. Thus, the only hypotheses that could be formulated were that (1) the association is only fortuitous, but this contradicts the statistical analysis; (2) the abnormality of the sternocleidomastoid is too mild to be detected; (3) the disease is too recent and the lesions are yet dissociated (an argument for the third explanation is that CSS symptoms were very recent in only 2 of the 4 patients); and (4) the peripheral pathological process involves only the sensory pathways or receptors of the sternocleidomastoid (in accordance with our hypothesis).
Some patients without CSS have abnormal sternocleidomastoids. This occurred in 4 patients in the control group. Here, too, only hypotheses can be given: CSS has not been diagnosed, even by two carotid sinus massages (this is not totally unlikely because the response to carotid sinus massage is known to vary, particularly when the disease is still in its early phase), and the response is still dissociated—the sternocleidomastoid is diseased but a large proportion of proprioceptive afferents are, according to our postulate, still intact. An argument for this explanation is that the 4 patients had unilateral lesions, whereas bilateral lesions were very predominant in the study group.
In these two aberrant groups of patients, only careful follow-up could favor one of these hypotheses; eg, in control patients with abnormal electromyographic results, regular carotid sinus massage could disclose whether some of these patients later develop CSS.
Finally, this study raises strong arguments for the participation of structures, particularly the sternocleidomastoids, surrounding carotid receptors in the CSS. However, the exact mechanism that induces centers to react by severe general sympathetic inhibition remains speculative, and our hypothesis has to be tested by further studies.
Selected Abbreviations and Acronyms
|BAEP||=||brain stem auditory–evoked potential|
|CSS||=||carotid sinus syndrome|
|SSR||=||sympathetic skin responses|
- Received August 16, 1995.
- Revision received October 30, 1995.
- Accepted October 30, 1995.
- Copyright © 1996 by American Heart Association
Roskam J. Un syndrome nouveau: syncopes cardiaques graves et syncopes répétées par hypereflexivité sinocarotidienne. Presse Med. 1930;38:590-591.
Morley CA, Perrins EJ, Grant P, Chan SL, Mc Brien DJ, Sutton R. Carotid sinus syncope treated by pacing: analysis of persistent symptoms and role of atrioventricular sequential pacing. Br Heart J. 1982;47:411-418.
Fourquet N, Genet L, Davy-Chedaute F, Jezequel J, Blanc JJ. Hypersensibilité sinocarotidienne associée à un cancer oto-rhino-laryngologique traité: etude chez 103 malades. Presse Med. 1991;20:1713-1716.
Cicogna R, Bonomi FG, Curnis A, Mascioli G, Bollati A, Visioli O, Rossi L. Parapharyngeal space lesions syncope-syndrome: a newly proposed reflexogenic cardiovascular syndrome. Eur Heart J. 1993;14:1476-1483.
Leatham A. Carotid sinus syncope. Br Heart J. 1982;47:409-410.
Morley CA, Hudson WM, Perrins EJ, Sutton R. Is there a difference between sick sinus syndrome and carotid sinus syndrome? Br Heart J. 1983;49:620-621.
Almquist A, Gornick C, Benson W, Dunningan A, Benditt DG. Carotid sinus hypersensitivity: evaluation of the vasodepressor component. Circulation. 1985;71:927-936.
Morley CA, Perrins EJ. The pathophysiology of carotid sinus syndrome. In: Perez Gomez F, ed. Cardiac Pacing Electrophysiology Tachyarrhythmias. Madrid, Spain: Grouz Publishers; 1985:200-213.
Jewett DL, Williston JS. Auditory evoked far fields averaged from the scalp of human. Brain. 1971;94:681-696.
Smith HL, Moerrsch FD. Further study on the hypersensitive carotid reflex. Mayo Clin Proc. 1935;11:380-383.
Kirchheim HR. Systemic arterial baroreceptor reflexes. Physiol Rev. 1976;56:100-176.