Introduction

Multiple sclerosis (MS) is a progressive, autoimmune, demyelinating disease of the central nervous system (CNS) in which chronic inflammation is thought to cause axonal injury and loss, resulting in deficits of motor and cognitive functions [1]. MS affects young adults between ages 20 and 40, with a population prevalence of approximately 400,000 in the USA and one million worldwide [2]. The exact etiology of MS remains unknown; however, it is thought that genetic [3] and epigenetic [4] factors interact with environmental factors, determining the initiation and progression of MS.

MS patients show dysregulation in the neurovisceral integration of cardiovascular modulation, indicated by the dysfunction of both sympathetic and parasympathetic autonomic nervous system (ANS) responses; this dysfunction correlates with MS disease activity and progression to disability [5]. In fact, cardiovascular ANS dysfunction more closely correlates with the level of clinical disability than the extended disability status score (EDSS) and MRI [6]. ANS dysfunction is observed in 45 % of MS patients, although the percentages greatly vary depending on one or a combination of tests. The sympathetic dysfunction is known to be more prevalent than parasympathetic dysfunction (30 vs.18 %), and the latter is often observed in patients with long disease duration and advanced clinical disability [7].

The sympathetic and parasympathetic neurons are the output of the central autonomic network, one of the CNS compartments, comprised of a number of structures, which are reciprocally interconnected. The central autonomic network regulates visceral, endocrine, and behavioral responses, with the information flowing bi-directionally [8]. The post-ganglionic sympathetic output is then mediated by the neurotransmitter norepinephrine, which acts through α- and β-adrenergic receptors, whereas the post-ganglionic parasympathetic nerves use acetylcholine (Ach) as transmitter, which functions through nicotinic and muscarinic receptors.

The output of the sympathetic and parasympathetic neurons at the sinoatrial node of the heart and the interactions between the two branches are manifested by a complex beat-to-beat variability in both frequency and amplitude. This heart rate variability (HRV) connects the central autonomic network output to cardiac autonomic tone [9]. HRV measures the frequency power spectral density analysis of heartbeat interval series, providing information on the distribution of the power as a function of frequency and as a mean to quantify autonomic balance at any given time [10].

The power spectral analysis shows the following two main oscillations; the high-frequency (HF) oscillation components (0.15~0.4 Hz), which correspond to the variation in respiration, also known as respiratory sinus arrhythmia, an index of parasympathetic tone, and the low-frequency (LF) oscillation components (0.04~0.15 Hz), which are the rhythms corresponding to the vasomotor tone, involving mainly the arterial pressure variabilities, an index of sympathetic tone [9, 11]. The ratio of LF to HF is the key indicator for HRV, and assumed to describe the sympathovagal balance, which can be affected by changes in LF, HF, or a combination of both [10, 12].

Brezinova et al. [13] observed reduction in the absolute LF and HF powers in MS patients, as compared to control subjects. The reduction in LF power was more significant (66 %) than reduction in HF power (49 %). Nevertheless, the existence of both sympathetic and parasympathetic dysfunctions in the same MS patient [14] suggests that the parasympathetic dysfunction may not be an independent phenomenon, but may result from an impaired interaction between a dysfunctional MS sympathetic branch, with the parasympathetic branch [9].

Among factors that could lead to MS sympathetic dysfunction is the damage to locus coeruleus [15]. Located in the pons structure of the brain stem, it is the principal site for brain norepinephrine synthesis and is involved in physiological responses to stress. Consistently, MS patients show reduced brain norepinephrine [15] and reduced intracellular catecholamine concentrations in the peripheral blood lymphocytes [16], suggesting the dysfunction of both central and peripheral sympathetic functions.

The results of our recent studies, which aimed to determine the prevalence of cardiovascular risk factors in MS patients, showed significantly lower systolic BP readings and plasma glucose levels, two variables which are determined by the sympathetic ANS function, in MS patients compared to non-MS patients [17, 18]. Among MS patients, 41 and 79 % presented with the systolic BP readings and plasma glucose levels below 120 mmHg and 100 mg/dL, respectively, as compared to 29 and 46 % for age- and gender-matched non-MS patients [17]. The gap between MS and non-MS patients further widened when disease-modifying drug (DMD)-naïve MS patients were compared to non-MS patients [18]. MS patients also exhibit hypothalamus pituitary adrenal (HPA) axis hypoactivity, both at rest and during physiological stress [19]. Sympathetic activation is known to be a prerequisite for the rise in the corticotropin-releasing hormone (CRH) and for the rise in ACTH secretion [20, 21].

Locus coeruleus has projections to many brain structures, including hypothalamus, brain stem and spinal cord, cerebral cortex, amygdala, and hippocampus, and it receives input from hypothalamus, prefrontal cortex, and cerebellum, among others. The far-reaching inputs to and outputs from locus coeruleus suggest that the damage to this structure may have widespread adverse effects on physiological processes of many organs. Locus coeruleus is known having roles in cognition, attention and memory, arousal, sleep-wake cycle, posture and balance, blood flow control, mood, and addiction [22–25]. The dysregulation in this system can lead to arrays of clinical symptoms often observed in MS patients, including cognitive dysfunction, sleep disorders, depression, migraine and reduced cerebral blood flow (see paragraphs below for detailed “Discussion”).

In addition, through sensory and immune-related stimuli to the brain, ANS also regulates the function of the immune system. This neurovisceral regulation extends beyond organs such as bone marrow, spleen, and thymus, into the regulation of cells involved in the innate and adaptive immunity [26], and achieved via the expression of adrenergic and cholinergic receptors on immune cells [27, 28]. Signaling through these receptors is known to influence immune cell proliferation and cytokine production, cellular inflammation, adhesion, and cell migration [29]. MS alterations in both adrenergic [30] and cholinergic [31] receptors and their correlation with clinical and MRI indicators of disease activity [32] suggest that ANS dysfunction is likely to play a role in MS immune dysregulation and promotion of inflammatory and neurodegenerative processes.

Thayer et al. [8] proposed dysregulation in the neurovisceral integration of cardiovascular modulation as contributing factor not only to cardiovascular disease but also to the related modifiable and unmodifiable risk factors. In their model, the dysregulation in the neurovisceral integration was related to sympathetic overactivity and parasympathetic underactivity. Predominance of sympathetic over parasympathetic has the potential leading to a reduction in ANS system flexibility and adaptability, which is a prerequisite for quickly responding to environmental demands. This impaired parasympathetic function has been attributed to a reduced prefrontal inhibitory control over sympathoexcitatory circuits in the brain stem [8, 33]. A similar model has been proposed for neurovisceral integration of emotions, suggesting that emotions, such as anxiety, are the by-products of faulty negative feedback, which then gives rise to unregulated and uninhibited positive feedback loops [34].

Both MS patients [13] and patients with cardiovascular disorders [35] present with reduced HRV values when compared to controls. The reduced HRV in patients with cardiovascular disease is due to high levels of sympathetic input, which is known to diminish HRV, whereas in MS patients, the reduced sympathetic outflow is the main contributor to the reduction in the HRV [36], suggesting that both sympathetic hyperactivity and hypoactivity may be associated with pathologies. However, unlike the neurovisceral dysregulation model which is based on the inhibition of vagal activity and dominance of sympathetic activity [8, 33, 34], we propose that lower-than-normal sympathetic output is the main cause for sympathovagal imbalance, leading to MS clinical morbidities.

The role of ANS dysfunction in the pathology of MS and its many clinical presentations is overlooked. Cardiovascular ANS dysfunction is not a benign phenomenon, but it may contribute to the many MS clinical presentations, with a significant negative impact on MS patients’ quality of life. We have recently discussed the role of ANS dysfunction in inflammation, neurodegeneration, and reduced response to immunomodulatory therapies [37], and its role in the high prevalence of osteoporosis often observed in MS patients (under peer review). Our assumption is based on the notion that sympathetic activation, and subsequent norepinephrine release and its binding to β2-adrenergic receptors on peripheral immune cells, has the potential upregulating cAMP, resulting in the inhibition of Th1-derived proinflammatory cytokines while promoting the production of Th2-dervided antiinflammatory cytokines [29, 38]. Similarly, the β2-adrenergic receptors on brain cells have a protective and supportive role [39], and their activation reduces memory impairment [40]. Therefore, the dysregulation in sympathetic activity could contribute to altered Th1/Th2 cytokines’ ratio and to heightened inflammatory state observed in MS patients [41], as well as to the MS cognitive impairments.

MS patients show alterations in adrenergic influences of the immune cells, indicated by lower cellular norepinephrine levels [42, 43] and higher number and density of dysfunctional [30] β2-adrenergic receptors on immune cells [32, 43] These receptors are absent from the brain of chronic progressive MS patients [44], and their absence is associated with MS clinical and MRI indicators of disease activity [32, 45]. The low norepinephrine levels and the dysfunction of the β2-adrenergic receptors have the potential of activating the α2-adrenergic receptors on macrophages, upregulating inflammatory processes [46]. The activation of the same receptors in the brain is associated with neurodegenerative processes [47].

The association between low sympathetic output and inflammation is consistent with the reduction in the rate of clinical relapse in female MS patients during pregnancy, when a progressive rise in the production of norepinephrine plasma levels is observed [48], whereas the post-partum clinical relapse is characterized by an approximately 40 % reduction in the norepinephrine plasma levels and reduction in the number and the affinity of the β2-adrenergic receptors on peripheral blood mononuclear cells [49]. Furthermore, the reduction in the α2-adrenergic receptors during pregnancy and their upregulation during post-partum [50] suggest the involvement of these receptors in inflammation and their contribution to MS clinical relapse.

The neuroprotective effects of increasing CNS norepinephrine have been shown in clinical trials of MS patients treated with antidepressants [51], and in its animal model, the experimental autoimmune encephalomyelitis (EAE) [52]. These results are consistent with studies showing a correlation between LF domain of the HRV and verbal learning and memory [53].

Similarly, the activation of the cholinergic axis has the potential exerting antiinflammatory and neuroprotective effects [54, 55]. PET studies show reduced nAChRs in the frontal, cingulate, temporal, parietal, and occipital cortices, and cerebellum of MS patients, and the correlation between reduction in these receptors and impaired memory and attention [56]. MS patients also show alterations in muscarinic cholinergic receptors on peripheral immune cells [31], suggesting both central and peripheral parasympathetic dysfunctions.

ANS dysfunction could also reduce patients’ clinical response to immunomodulatory therapies. It is known that chronic progressive MS patients, who present with a more severe ANS dysfunction [5], are often the ones who have poor response to immunomodulatory therapies [57]. Monocytes derived from these patients show low intracellular norepinephrine levels and do not adequately respond to the IL-2-induced rise in β2-adrenergic receptors density [43].

The reduced and or loss of response to long-term immunomodulatory therapy may stem from the potential of immunomodulators to downregulate the sympathetic ANS function, either directly or by shifting the profile of Th1 to Th2 cells [58]. Th2 cells do not possess β2-adrenergic receptors and hence are unable to stimulate the sympathetic ANS function [59]. Analogously, chronic progressive MS patients, who have predominantly Th2 cytokines [59], do not respond favorably to treatment with agents that increase Th2 cytokines [60]. The adverse effects of immunomodulators on sympathetic ANS function may especially be apparent in older MS patients and those patients with longer disease duration [61, 62]. These patients are likely to have a more severe ANS dysfunction, resulting from both age-induced impairment in ANS function and from the MS-related disease burden [63, 64].

The reduced response to immunomodulatory therapies depends on the duration, and often occurs within 2 years of use, correlating closely with patients’ clinical disability [65]. For example, an acute interferon beta (IFN-β) administration enhances lymphocytes’ production of norepinephrine in IFN-β-naïve MS patients [66]. But, lymphocytes derived from long-term IFN-β users show reduced cytokine response to IFN-β treatment [67], as well as reduced baseline and post-IFN-β-induced rise in intracellular norepinephrine [16]. The association between ANS dysfunction and reduced response to immunomodulatory therapy is supported by the result of a clinical study showing improved clinical efficacy of glatiramer acetate when combined with the β-adrenergic agonist, albuterol [68].

In this evidence-based review, we propose that the dysregulation of the neurovisceral integration of cardiovascular modulation (Fig. 1) cannot only promote inflammatory and neurodegenerative processes, but it can also lead to many other MS-related morbidities, including depression, fatigue and sleep disorders, migraine, and cerebral hemodynamic impairments; the latter can lead to cerebral venous remodeling and subsequent stenosis. We demonstrate that these clinical presentations are intertwined with MS inflammatory and neurodegenerative processes.

Fig. 1
figure 1

The Clinical effects of impaired neurovisceral integration of cardiovascular modulation. The diagram shows that an impaired neurovisceral integration of cardiovascular modulation has widespread adverse clinical effects. It can promote inflammation/neurodegeneration, reduce the response to immunomodulatory therapies, and contribute to the clinical presentations of fatigue/sleep disorders, depression, migraine, osteoporosis, and HPA axis hypo-responsiveness; It can lead to cerebral hemodynamic impairments, indicated by reduced cerebral blood flow, diminished cerebral autoregulation, and altered cerebrospinal fluid dynamics; It can lead to hypoxia-induced-TCR response diminution, with further adverse effects on cerebral blood flow; It can influence cerebral venous microvascular and macrovascular structures, indicated by venous remodeling and compression/stenosis respectively. HPA hypothalamus pituitary adrenal; TCR trigemino-cardiac reflex

Full size image

Discussion

ANS Dysfunction in MS Clinical Morbidities

Depression

Depression is a common clinical presentation in MS, affecting up to 50 % MS patients during their lifetime [69]. MS patients are twice as likely to be on antidepressants as compared to non-MS patients (35 vs.18 %) [17]. The pathophysiology of depression is complex, but it is known to partly involve reduced brain norepinephrine and dopamine (the latter is metabolized to norepinephrine by the enzyme dopamine β-hydroxylase) [70]. Lymphocytes of individuals with depression show lower β-adrenergic receptor binding [71] and sensitivity [72] compared to those in healthy controls. Furthermore, a significant reduction in the parasympathetic-related HRV parameters have been observed in patients with clinical symptoms of depression [73], suggesting that both adrenergic and cholinergic dysfunctions may play roles in the pathology of MS depression.

Consistent with this conclusion, individual treatment with the antihypertensive drug reserpine, which depletes monoamine stores, induces depression [74], whereas monoamine oxidase inhibitors (MAOIs) and tricyclic antidepressants (TCAs), which are known to increase synaptic levels of CNS norepinephrine, reduce depressive symptoms. In addition, vagal afferent nerve fiber stimulation is known to benefit patients with depression [75]. However, this FDA-approved strategy is thought to exert beneficial effects, in part, by transmitting impulses to the locus coeruleus [75] and increasing CNS norepinephrine concentrations [76]. Similar beneficial effects of vagal stimulation have been shown in MS patients [77], who are known to have locus coeruleus damage [15].

In addition, lower brain-derived neurotrophic factor (BDNF) serum levels are known to be associated with depressive disorders [78]. This observation agrees with the reported low BDNF serum levels in MS patients. The levels correlate with neuroimaging indicators of disease activity and are increased after antidepressant therapy [79, 80]. BDNF is produced by both immune cells and neurons with a potential for exerting immunomodulatory and neuroprotective effects [81]. The increase in neuronal norepinephrine [82] and vagus nerve stimulation [76] are known to increase BDNF levels, suggesting a relationship between ANS function, BDNF levels, depression, and MS inflammatory and neurodegenerative processes.

Gender differences in cardiovascular ANS regulation have been reported, showing lower sympathetic function in females than males [64, 83]. Our own study showed a significantly higher percentage of DMD-naïve female MS patients with systolic BP below 120 mmHg, as compared to male patients (55 vs. 17 %). This trend was reversed for systolic BP above 130 mmHg (27 % of females vs. 50 % of males) [17].

These differences may contribute to a higher prevalence of depression observed in MS females, as compared to males [84]. In addition, MS patients with longer disease duration, and chronic progressive MS patients, show more severe clinical depression compared to patients with a shorter disease duration and patients with other form of MS [84, 85]. These observations are consistent with the greater severity of ANS dysfunction in chronic progressive MS patients and those with longer disease duration [5, 63].

The association between depression and ANS dysfunction in MS patients is mostly unknown. However, MS depression closely correlates with inflammation and neurodegeneration. For example, MS patients with depression show higher hyperintense and hypointense lesion volumes in the left medial inferior frontal region than patients without depression [86]. A diffusion tensor imaging of MS patients with depression shows reduced normal appearing white matter and elevated mean diffusivity in normal appearing grey matter in the temporal lobes [87]. Furthermore, an association between MS depressive symptoms and a reduced hippocampal volume has been observed [88]. These studies are in line with the reported negative influences of depression on MS cognitive function [89] and the role of ANS dysfunction in promoting neurodegeneration.

Fatigue-Sleep Disorders

Fatigue is one of the most common clinical symptoms affecting 80 % of MS patients, 50 % of whom are severely affected [90, 91], with a significant impact on the quality of life [92]. Fatigue is closely associated with depression [92, 93], with a higher prevalence in females than males [94]. MS patients can experience peripheral and or central fatigue. The peripheral fatigue is known as muscular fatigue, correlating with EDSS, whereas the pathophysiology of central fatigue involves impaired cognitive processing, memory, concentration, and deficits in attention and arousal [95, 96].

A number of clinical scales are used to measure fatigue in MS patients. These scales include Fatigue Severity Scale (FSS), Fatigue Impact Scale (FIS), Modified Fatigue Impact Scale (MFIS), and Neurological Fatigue Index (NFI). However, these clinical scales are based on self-reported questionnaires and do not always correlate with each other [97]. Fatigued MS patients are unable to adequately increase BP during handgrip exercise [98] and during stressful physical tests [99], suggesting the involvement of sympathetic dysfunction in MS fatigue. In addition, MS fatigue, as measured by the FSS and the MFIS, correlates with HRV [100]. The reduced HRV is especially apparent during the short-term memory test [99], suggesting a relationship between ANS dysfunction, cognitive impairment, and MS fatigue.

The perception of fatigue can be increased by comorbid conditions such as depression and sleeping disturbances [101]. In fact, similar to fatigue and depression, sleeping disturbances are three times more likely in MS patients than non-MS patients, and reduced sleep quality is twice as likely in MS patients than non-MS patient [102]. In addition, MS patients with fatigue are more likely to experience disrupted sleep, indicated by frequent awakening, than non-fatigued MS patients and healthy controls [103]. Cardiovascular ANS dysfunction, especially a reduced parasympathetic function, is known to contribute to MS sleep disturbance [104].

Strober et al. [101] proposed a model showing that fatigue, depression, and sleeping disorders are interrelated and each independently correlate with MS disease severity. This model agrees with the reported relationship between MS fatigue and higher white and grey matter atrophies, and between MS fatigue and higher T1 and T2 lesion loads [105, 106]. The relationship between fatigue and MS neurodegenerative processes is consistent with ANS dysfunction contributing to MS physical and cognitive disability [107]. The coexistence of fatigue and ANS dysfunction has been also demonstrated in other autoimmune [108] and neurodegenerative [109] diseases.

Migraine

Headache is another common MS clinical symptom, with a 50 % higher prevalence in MS patients than in control subjects [110]. MS patients present with primary headache, including migraine (with and without aura) and tension type headache, although many MS patients have unclassified headache [111]. Migraine is more prevalent in females and in young relapsing-remitting MS patients, correlating with disease exacerbations [112], whereas tension-type headache is more often observed in secondary progressive male MS patients [113].

Studies show that the relative risk of developing MS is significantly higher in migraineurs than in non-migraineurs, and the odds of being diagnosed with migraine is higher in women with MS than those without MS [114], suggesting that MS and migraine may be related at some level. Factors such as demyelinating brain stem lesions [115] and or treatment with DMDs [116] have been thought to contribute to the high prevalence of migraine in MS patients, but the role of ANS dysfunction in MS migraine has not been investigated.

The cranial circulation has both sympathetic and parasympathetic components [117], suggesting a role for ANS dysfunction in migraine. A wide variety of diagnostic tests indicate the presence of a sympathetic ANS dysfunction in migraineurs. Pupillary sympathetic hypofunction, orthostatic hypotension, a poor response to cold pressor test, a decreased overshoot in the Valsalva maneuver, and low levels of plasma norepinephrine have all been documented in patients with migraine and tension headache [118–121]. The reduced plasma norepinephrine level is also apparent in migraineurs during the headache-free period [121].

Patients with migraine and tension-type headache also show reduction in the HRV-related parameters of parasympathetic ANS activity, as well as reduced HR responses during sustained handgrip, and RR-interval oscillatory response to neck suction [122–124]. These results collectively show the involvement of both sympathetic and parasympathetic dysfunctions in the pathology of migraine. Similar to MS patients, migraineurs suffer from greater fatigue as well as depressive tendencies [122, 125], observations which are in line with the reported role of sympathetic impairment as the cause of migraine [126]. Consistent with the role of ANS dysfunction in neurodegeneration, migraineurs also exhibit increased serum markers of glial activation and neuronal damage, apparent even in migraine-free periods [127].

Cerebral Hemodynamic Impairments

The newer theory of vascular etiology of MS, which has been set forward by Zamboni, describes impairments in the principal pathways of extracranial venous drainage, affecting mostly the internal jugular veins (IJVs), the vertebral, and the azygos veins [128, 129]. However, the normal cerebrospinal fluid (CSF) ferritin levels in MS patients [130] exclude the role of venous reflux as the factor contributing to MS inflammatory and neurodegenerative processes. We have recently proposed that dysregulation in neurovisceral integration of cardiovascular modulation can lead to cerebral hemodynamic impairments, associated with the chronic cerebrospinal venous insufficiency (CCSVI). In this review, we provide additional evidence supporting our original proposal, showing a close association between venous stenosis and other clinical manifestations of ANS dysfunction.

ANS-dysfunction-induced arterial BP reduction [131] is among factors contributing to CCSVI occurrence. Our recent study shows that CCSVI-positive MS patients show deviations from normal ranges in arterial BP. Approximately, 50 % of patients present with systolic BP below 120 mmHg [132]. Furthermore, 91 % of MS patients present with three or more symptoms of ANS dysfunction, including cognitive impairment, fatigue, sleep disorders, and headache, and a significant statistical relationship between various clinical symptoms has been observed [132]. The coexistence of CCSVI and cardiovascular ANS dysfunction in MS, as well as in other autoimmune and neurological diseases [133, 134], further supports the notion that the two phenomena may be related.

Reduced arterial BP could also to a lower than normal cerebral perfusion pressure, which is determined by the differences in the pressure gradient between mean arterial BP and the intracranial pressure [135]. Although intracranial venous pressure has been shown to be normal [136], PET or single-photon emission CT techniques show hypoperfusion of MS brain parenchyma, involving both cerebral white and grey matters [137, 138]. These hemodynamic impairments correlate with CCSVI severity [139], with MS disease progression, with neuropsychological dysfunction, and with fatigue scores [137, 138, 140].

The Starling resistor hypothesis dictates that an increase in intracranial pressure and cerebral venous pressure would accompany a reduction in cerebral blood flow and cerebral perfusion pressure [141]. However, despite reduced cerebral blood flow [142, 143], MS patients show normal intracranial pressure, as measured by venous pressure [136]. This discrepancy may be due to differences in the degree of compliance of various brain regions, leading to an intracranial pressure gradient. Therefore, the level of the intracranial pressure may depend on the area of measurement [144].

The hypoperfusion is assumed to be the result of deficiency in astrocytic β2-adrenergic receptors and a reduced formation of cAMP [138], rather than the presence [138] or absence [137, 138] of CCSVI. These results confirm that sympathetic dysfunction, rather than venous obstruction, may contribute to the observed cerebral hypoperfusion. The improvement in cerebral blood flow post-balloon angioplasty [145] could be attributed to sympathetic activation [146] due to venous distention [147], rather than correction of the venous obstruction. The reported transient increase in cerebral blood flow, before lesion development [148], indicates sympathetic ANS activation in a setting of acute inflammation.

Despite alteration in arterial BP, little or no changes in cerebral blood flow is expected to occur as long as the perfusion pressure is greater than cerebral autoregulation, which is known to maintain constant cerebral blood flow despite fluctuations in the cerebral perfusion pressure. Nevertheless, in pathological states, cerebral blood flow could become pressure dependent due to impairment in cerebral autoregulation. Indeed, subsequent to the publication of our recent review, where we postulated the likelihood of the impairment of MS cerebral autoregulation, Mezei et al. [149] reported such impairment, indicated by changes in MS cerebral blood flow velocity as a function of changes in mean arterial BP. The impairment in cerebral autoregulation is likely to be the result of ANS dysfunction, since cerebral autoregulation is under autonomic neural control [150, 151].

The combination of low arterial BP and impairment in cerebral autoregulation, and high levels of nitric oxide, produced due to the chronic MS brain inflammatory processes [152], will ultimately reduce transmural pressure, promoting the critical closure pressure (CrCP) [153, 154]. Alterations in the CrCP contribute to the compression or the collapse of the cerebral venous system in general, and IJVs in particular, owing to their cytoarchitectionial construction. The IJVs collapsibility also explains their common tendency for restenosis after venous angioplasty [155]. Furthermore, IJVs reopening after their collapse due to orthostatic challenge may be hindered due to a suboptimal central venous pressure, the latter resulting from a low sympathetic outflow [156].

Phase contrast CSF flow imaging studies show reduced net CSF flow, correlating with CCSVI severity [157, 158]. It is known that CSF dynamic is coupled to cerebral arterial and venous systems through windkessel effect [159]. The intracranial venous flow is normal in MS patients, showing no sign of reflux, but the arterial flow is reduced [158], suggesting that the impairment in CSF flow may be the by-product of reduced arterial pressure rather than venous obstruction.

A dynamic secretion and resorption of CSF is required for maintaining intracranial pressure equilibria. Circulatory movement due to hydrostatic pressure gradients regulates both CSF secretion at the choroid plexuses and CSF resorption into the venous system at the arachnoid granulations. However, a dynamic CSF flow depends also on the vasomotor waves or Traube-Herring-Mayer (THM) waves which are cyclic changes in arterial BP brought about by oscillations in arterial baroreceptors [160]. The frequency and the peak occurrence of THM waves depend on the functions of both sympathetic and parasympathetic ANS [160].

THM waves along with increased HRV are considered to be markers of autonomic balance. MS patients present with baroreflex dysfunction, indicated by reduced systolic BP oscillations and reduced RR-interval oscillations in response to baroreceptor stimulation [14]. These alterations have the potential for reducing the THM wave. Whether the observed MS improvement in CSF dynamics [161] is the result of an increase in THM wave due to balloon angioplasty-induced sympathetic activation [146] is unknown and warrants further investigations.

The hypoperfusion of MS brain parenchyma [137, 138] could lead to hypoxia, and an increase in the expression of hypoxia inducible factor-1α (HIF-1α), promoting myelin-associated glycoprotein degradation and CNS damage [162], as well as compromise the blood–brain barrier integrity [163]. Oligodendrocytes are known to be particularly sensitive to ischemic insults [164], and a mild chronic hypoperfusion in the range of 15 % is able to lead to the disruption of axon-myelin integrity [165].

Contrast-enhanced MRI and the use of echo planar technique, which measured parameters of brain perfusion, such as cerebral blood flow, cerebral blood volume, and vascular mean transit time, observed a 50 % reduction in cerebral blood flow and more than a twofold prolonged mean transit time and reduction in cerebral blood volume in normal appearing white matter (NAWM) in relapsing-remitting MS compared with controls [142, 143]. The chronic focal lesions in MS white matter showed a more decreased perfusion when compared with contralateral NAWM [142]. Similar results were observed also in primary progressive MS patients [143]. Such significant reduction in cerebral blood flow has the potential causing microglial activation, white matter damage, and axon sparing [166].

Cerebral hypoxic ischemia has the potential for fast and reversible activation of the oxygen-conserving reflexes, similar to their activation during hypoxic cerebral ischemia of stroke. Also termed trigeminocardiac reflexes (TCRs), these protective sympathetically mediated vasomotor responses originate from reticulospinal neurons of the rostroventrolateral medulla, projecting to the upper brain stem and/or thalamus, culminating in the activation of a population of neurons in the cortex. The activation of these cortical centers can in turn modulate cardiovascular centers, increasing cerebral vasodilation and blood flow, without changing cerebral oxygen or glucose metabolism, hence minimizing hypoxia-induced ischemic brain damage [167, 168]. However, the suboptimal MS sympathetic output may compromise TCRs activation and their downstream neuroprotective effects, leading to reduction in MS cerebral blood flow [142, 143], in the absence of optimal TCR corrective measures.

In addition, hemodynamic impairments due to ANS dysfunction can also lead to morphological and metabolic changes in the venous system [169], with the potential for promoting venous compression and or closure. A series of studies by Zamboni and colleagues [170–172] report morphological changes including IJV calcification patterns [170], absence of endothelial cells in the intraluminal obstacles and their replacement with fibrous lamina [171], and an increase in the proportions of type III to type I collagen within the adventitia [172]. These pathological processes in the venous system mirror those in the arterial system, where hemodynamic impairments could lead to arterial remodeling, arterial hardening, and stenosis, a process known as atherosclerosis.

One such example is the hemodynamic alterations in the surgically created arteriovenous (AV) fistula that often leads to vessel remodeling and AV graft stenosis [173]. In this model, immune cell infiltration into the AV fistula occurs early and transiently in the stenosis process, and the levels correlate with the vessel wall shear stress [174]. These results suggest that the absence of immune cells observed in the diseased IJVs [172] may be due to early and transient infiltration of immune cells into the diseased IJVs and or due to a very low physiological shear stress in these vessels compared to the shear stress in the arteries (1 dyne/cm2 in the venous system vs. 15 dyne/cm2 in arteries) [175]. The efficacy of 25-hydroxy vitamin D in MS could be related to its ability to beneficially influence ANS function [176], preventing the subsequent venous remodeling [177].

Compared to cerebral arteries, the cerebral venous system presents with a complex three-dimensional structure and extensive anatomic variabity; these could in turn influence the biophysics and hemodynamics of the cerebral venous system, differentiating them from those parameters, known to govern in the arterial system [178]. Although the cerebral venous occlusion may be partially compensated by collateral circulation, this occlusion often leads to venous infarction and neurological deterioration, with an underlying pathology that differs from cerebral arterial infarction [179]. Assuming that ANS dysfuntion plays a role in venous occlusion, the improvement in the function of this system has the potential for reducing the high rate of MS cerebral venous thromboembolic incidences [180, 181]. In contrast, MS low sympathetic output may contribute to reduced risk of ischemic stroke [182], ischemic heart disease, and myocardial infarction [183].

Concluding Remarks

We have shown through our studies and those of others that impaired neurovisceral integration of cardiovascular modulation plays a role in MS pathology, with the potential for promoting not only inflammation and neurodegeneration but also many other MS-related morbidities, including depression, fatigue and sleep disorders, migraine, and cerebral hemodynamic impairments; the latter could result in venous compression and or stenosis, in the absence of corrective measures. We have also recently describe the contribution of ANS dysfunction to the high prevalene of osteoporosis often observed in these patients (under peer review).

The close association between ANS dysfunction and MS pathological processes suggests that the routine ANS function testing should be conducted and used as a tool for improving the assessment of MS disease stage and the rate of its progression. Such data could be also used for evaluating the clinical efficacy of DMDs.

In a sequel to this review, we have described the pharmacological, non-pharmacological, and surgical strategies that could be adopted to improve neurovisceral integration of cardiovascular modulation, promoting a state of sympathovagal balance [184], which would likely have a positive impact on MS patients’ quality of life. Nevertheless, the corrective strategies should take into account the individual differences in the nature, whether sympathetic or parasympathetic, and the degree of the autonomic imbalance. In addition, genetic, epigenetic, and factors can also modulate ANS functional activity. These factors interact with environmental factors in a complex manner. This complexity is further enhanced in MS patients due to MS disease processes, disease duration, and the chronic use of many MS-related and unrelated drugs. Therefore, an optimal corrective strategy(s) in MS patients should take into consideration these additional factors and the interactions among them [185].

These strategies could be also applied to other autoimmune and neurodegenerative diseases where autonomic dysfunction plays a role.