In this post, I link to and excerpt from Electrodiagnostic assessment of the autonomic nervous system: A consensus statement endorsed by the American Autonomic Society, American Academy of Neurology, and the International Federation of Clinical Neurophysiology [PubMed Abstract] [Full-Text HTML] [Full-Text PDF]. Clin Neurophysiol. 2021 Feb;132(2):666-682.
All that follows is from the above resource.
- • Sensitive, validated, noninvasive electrodiagnostic tests of autonomic function have been developed.
- • An international expert panel provides evidence-based recommendations to guide autonomic testing.
- • Recommendations allow for standardized assessment of severity and distribution of autonomic failure.
Evaluation of disorders of the autonomic nervous system is both an art and a science, calling upon the physician’s most astute clinical skills as well as knowledge of autonomic neurology and physiology. Over the last three decades, the development of noninvasive clinical tests that assess the function of autonomic nerves, the validation and standardization of these tests, and the growth of a large body of literature characterizing test results in patients with autonomic disorders have equipped clinical practice further with a valuable set of objective tools to assist diagnosis and prognosis. This review, based on current evidence, outlines an international expert consensus set of recommendations to guide clinical electrodiagnostic autonomic testing. Grading and localization of autonomic deficits incorporates scores from sympathetic cardiovascular adrenergic, parasympathetic cardiovagal, and sudomotor testing, as no single test alone is sufficient to diagnose the degree or distribution of autonomic failure. The composite autonomic severity score (CASS) is a useful score of autonomic failure that is normalized for age and gender. Valid indications for autonomic testing include generalized autonomic failure, regional or selective system syndromes of autonomic impairment, peripheral autonomic neuropathy and ganglionopathy, small fiber neuropathy, orthostatic hypotension, orthostatic intolerance, syncope, neurodegenerative disorders, autonomic hyperactivity, and anhidrosis.
KeywordsAutonomic nervous system diseasesDenervationAutonomicDiabetic autonomic neuropathyValsalva ManeuverTilt table testHypotensionOrthostatic
The discipline of autonomic medicine is concerned with the diagnosis and treatment of diseases and disorders of the autonomic nervous system. Invaluable to accurate diagnosis are autonomic function tests, which supply objective, quantifiable information about the integrity and behavior of autonomic nerves, ganglia, and central nervous system networks.
This review will focus on tests of autonomic function that expert consensus considers to be scientifically valid, reliable, and clinically useful. Tests of autonomic function involve measurement of a physiologic function in response to a specific manipulation of the body under standardized conditions. To be meaningful, the results must be interpreted in reference to normative values and with an understanding of the physiology of autonomic responses in health and in autonomic disorders.
3. Common clinical questions
The aims of autonomic testing are to recognize the presence, distribution, and severity of autonomic dysfunction. Further, autonomic testing may detect characteristic patterns of autonomic failure or hyperfunction that can be related to specific disorders.
When abnormal, the results of autonomic testing can assist with earlier diagnosis, monitor clinical progression, and assess the response to treatment. When normal, the results of autonomic testing can provide objective evidence that a serious autonomic condition is not present.
Table 1. Laboratory grading of autonomic failure by the Composite Autonomic Severity Score.
Sudomotor index Quantitative sudomotor axon reflex test Thermoregulatory sweating test 0 Normal Normal 1 Single site abnormal, or length-dependent pattern (distal sweat volume < 1/3 of proximal values) Anhidrosis present but < 25% 2 Single site < 50% of 5th percentile Anhidrosis 25–50% 3 Two or more sites < 50% of 5th percentile Anhidrosis > 50% Adrenergic index Beat-to-beat measurement of blood pressure in response to the Valsalva maneuver and head- up tilt to 70 degrees 0 Normal 1 Phase II_E decrease of mean blood pressure between 20 and 40 mmHg plus phase II_L or IV absent, or decrease in pulse pressure to ≤ 50% of baseline; PRT 6–10 seconds 2 Phase II_E decrease of < 40 mmHg plus absent phase II_L or IV; PRT 11–20 seconds 3 Phase II decrease of > 40 mmHg plus absent phase II_L and IV; PRT > 20 seconds 4 Criteria met for 3 plus systolic blood pressure decrease of ≥ 30 mmHg (orthostatic hypotension) Cardiovagal index Beat-to-beat measurement of heart rate in response to sinusoidal deep breathing at 5–6 breaths/min and the Valsalva maneuver 0 Normal 1 HRDB or VR mildly decreased (>50% of 5th percentile) 2 HRDB or VR decreased to < 50% of 5th percentile 3 Both HRDB and VR decreased to < 50% of 5th percentile
The table lists the components of autonomic testing results that contribute to the Composite Autonomic Severity Score (CASS). Test methodologies are listed in italics at the top for each category. Results from each category (sudomotor, adrenergic, cardiovagal) are assigned a numerical severity, 0 being normal, which add up to the CASS, which has a maximum score of 10. The sudomotor index can be based on either quantitative sudomotor axon reflex testing (QSART) or the thermoregulatory sweating test (TST). Percentiles are in reference to normative values. HRDB = heart rate response to deep breathing, VR = Valsalva ratio. Phases refer to the components of the Valsalva maneuver: II_E and II_L the early and late portions, respectively, of phase II. PRT = BP recovery time.
5. Testing of cardiovascular adrenergic function
The Valsalva maneuver and the tilt table test represent the two cornerstones of cardiovascular adrenergic (baroflex-sympathoneural) assessment, which is key to the evaluation of orthostatic hypotension (OH), as discussed in detail in Section 10.1 below. They provide complementary information about autonomic responses, in the case of the Valsalva maneuver, to the transient reduction in cardiac preload induced by increased intrathoracic pressure during straining, and in the case of the tilt table test, to the gravitational redistribution of blood volume in the upright posture. Proper performance requires continuous monitoring of HR and beat-to-beat BP, which is accomplished noninvasively via ECG and photoplethysmography.
5.1. Valsalva maneuver
The Valsalva maneuver is done in a standardized manner with the subject rested and recumbent (or seated, depending on the laboratory) and asked to maintain a column of mercury at 40 mmHg for 15 seconds. A maneuver < 10 seconds or < 20 mm Hg is inadequate (Benarroch et al., 1991). For subjects who have a “flat top” response (where induced BP change mimics expiratory pressure in configuration and early phase II does not develop), the study is repeated with the subject tilted to 20° or if necessary 40° until a significant fall in BP is obtained (Vogel et al., 2008). When BP recordings are done, the subject is asked to repeat the maneuvers until two reproducible beat-to-beat BP recordings are obtained.
Table 2. Summary evaluation of attributes of some autonomic function tests.
Parameters HRV VM QSART TST CASS_Adr3 Sensitivity ≥80% ≥80% 80% 80% 80% Specificity ≥80% ≥80% ≥80% ≥80% ≥80% Reproducibility High1 High High High?2 High Physiologic Basis +++ +++ +++ +++ +++ Clinical Relevance +++ +++ +++ +++ +++ Non-invasive +++ +++ +++ +++ ++ Practical +++ ++ ++ + ++ Availability +++ +++ ++ + ++ Affordability +++ +++ + ++ ++ Known Confounders ++ ++ +++ ++ +++
1Coefficient of variation < 20%; 2Apparently high reproducibility but limited studies done; 3Components are BP/HR to tilt and beat-to-beat BP responses of the Valsalva maneuver as indices of adrenergic failure; +++, easily fulfills criteria; ++, fulfills criteria; +, just fulfills criteria, with some difficulty in some situations. CASS_Adr = Composite Autonomic Severity Score, adrenergic subscale; HRV = heart rate variability; QSART = quantitative sudomotor axon reflex test; TST = thermoregulatory sweating test; VM = Valsalva maneuver.
5.2. Tilt table test
Upon transitioning to the upright posture, the force of gravity causes the blood volume to redistribute away from the cerebral and thoracic vasculature toward the splanchnic and lower extremity vasculature. Within 3 minutes, 500–800 mL blood, or approximately 10% of the total blood volume and 25% of the thoracic blood volume, is displaced downward (Smith et al., 1994). The tilt table creates a controlled environment where the autonomic responses to orthostatic stress can be monitored closely.
In healthy persons, adaptation to the upright posture entails unloading of carotid and cardiopulmonary baroreceptors, which leads to a reflex increase in sympathetic adrenergic outflow, causing increased peripheral vasoconstrictor tone, HR, and inotropic state. In OH and disorders of orthostatic intolerance, the autonomic responses to orthostatic stress are insufficient.
Raising the patient on an inclining table is not the same as having the patient stand. Because leg contraction is not required to maintain the upright posture on the tilt table, passive tilting is a stronger orthostatic stimulus. As compared to passive tilt, active standing during the first 30 seconds causes a greater reduction in BP and total peripheral resistance and a larger increase in HR and cardiac output (Tanaka et al., 1996). The optimal conditions for performing tilt table testing have been reviewed in detail elsewhere and depend on the clinical question and context (Cheshire and Goldstein, 2019).
6. Testing of cardiovagal function
The influence of the vagus nerve on heart rate variability is an important aspect of autonomic regulation. This parasympathetic regulation occurs more rapidly than changes in cardiovascular sympathetic adrenergic function.
6.1. Heart rate variability to breathing
Variation in HR in response to breathing has been quantified in a number of ways. In the time domain, those include SD of RR interval, mean circular resultant (MCR), mean successive difference, and mean squared successive difference (typically from using normal, spontaneous breathing), as well as expiratory-inspiratory (E:I) ratio and maximum – minimum HR response to deep breathing (HRDB; usually averaged over 5 cycles) (Weinberg and Pfeifer, 1984). Both the afferent and efferent limbs underlying respiratory sinus arrhythmia are vagal; although there is modulation, under normal conditions, by central and peripheral sympathetic activity (Levy et al., 1966), Bainbridge reflex (Bainbridge, 1920), Hering-Breuer reflex (Hering, 1871), and baroreflexes (Eckberg et al., 1980). There are, in addition, significant effects of medications and end organ (cardiac) disease (Low and Sletten, 2008). That these tests measure cardiovagal function derives from the finding that heart period has a linear relationship to vagal tone, varied under experimental conditions in the dog (Katona and Jih, 1975), and the findings were replicated in humans with pharmacologic blockade (Fouad et al., 1984).
Table 3. Some variables affecting cardiovagal function.
Variable HRDB Effect of Age Age + Decrease with age Respiratory rate Yes Maximal at 6 breaths per minute Hypocapnia Yes Reduces response Sympathetic tone Yes Suppresses HRV Medications Yes Especially anticholinergic Depth of respiration Yes Modest effect Obesity Yes Modest effect
HRDB = heart rate variability with deep breathing; HRV = heart rate variability.
6.2. Valsalva ratio
The Valsalva ratio (VR) is the ratio of the maximum HR that develops in response to BP reduction induced by the Valsalva maneuver (described in Section 5.1), divided by the minimum HR that results from the maneuver-induced BP overshoot. The Valsalva ratio is the ratio of the maximal (following phases II/III) to minimal HR (occurring within 30 s of phase IV peak). Details and mechanisms of the maneuver are described above.
6.3. 30:15 ratio
The 30:15 ratio is measured in the patient who lies quietly and then is asked to stand up unaided. The ratio is obtained by dividing the longest R-R interval at the 30th beat by the shortest R-R interval at the 15th beat. It is one component of Ewing’s battery, which comprises also the Valsalva ratio, the HR response to deep breathing, and the BP response to standing and to sustained handgrip (Ewing et al., 1985). Ewing’s battery is occasionally still used but has largely been supplanted by more precise methods of noninvasive autonomic testing.
7. Testing of sudomotor function
The goal of sudomotor testing is to evaluate the functional integrity of sudomotor neurons. Under controlled conditions a standardized stimulus is administered, and the response is measured.
7.1. Quantitative sudomotor axon reflex test (QSART)
The quantitative sudomotor axon reflex test (QSART) evaluates the functional integrity of the postganglionic sympathetic sudomotor axon.
The QSART has the advantage of assessing the distribution of sudomotor impairment. The test is sensitive and reproducible in healthy controls (Low et al., 1983) and in patients with diabetic neuropathy (Low et al., 1986). Tests repeated on two different days show a decrease in sudomotor volume with a high coefficient of regression. The coefficient of variation was found to be 8% and 14% in two studies (Low et al., 1983, Low and Opfer-Gehrking, 1992). In the diagnosis of distal small fiber neuropathy (DSFN), it has a sensitivity of 80% and a specificity of > 90% (Low et al., 2006, Stewart et al., 1992), using the criteria of absent response or response < 5th percentile at the foot, while sweating is normal in proximal sites (Table 4).
Table 4. Comparison of studies of sudomotor testing in distal small fiber neuropathy.
Autonomic Test Reference Number of patients Diagnostic Criteria Sensitivity TST Stewart et al. 1992 25 Distal anhidrosis 72% Low et al. 2006 125 Distal anhidrosis 74% QSART Stewart et al. 1992 40 Distal abnormality 43% Any abnormality 80% Tobin et al. 1999 15 Any abnormality 80% Novak et al. 2001 92 Distal abnormality 71% Any abnormality 73% Singer et al. 2004 11 Any abnormality 64% Low et al. 2006 125 Distal abnormality 62% Any abnormality 77% Thaisetthawatkul et al. 2013 121 Any abnormality 52% Abnormal and at least 1 additional abnormality in pin sense, quantitative sensory testing, or skin biopsy 82%
QSART = quantitative sudomotor axon reflex test; TST = thermoregulatory sweating test.
7.2. Thermoregulatory sweat test
The thermoregulatory sweat test (TST) evaluates the integrity of central and peripheral sympathetic sudomotor pathways. The pathway is complex. From central nervous system thermoregulatory centers including the hypothalamus, bulbospinal tracts descend to the intermediolateral cell columns, which send forth preganglionic neurons that emerge from the spinal cord and travel through the sympathetic chain ganglia. Preganglionic neurons synapse onto postganglionic sudomotor axons, which innervate eccrine sweat glands (Fealey et al., 1989, Low and Sletten, 2008).
7.3. Sympathetic skin response
The sympathetic skin response (SSR) evaluates the momentary change in the electrical potential associated with induced sweating in the palms and soles. The response can be evoked by psychological or local nociceptive stimuli (Vetrugno et al., 2003). As the response is emotionally activated rather than thermoregulatory, it is highly variable and has limited sensitivity and specificity in the diagnosis of sudomotor nerve impairment (Arunodaya and Taly, 1995, Gutrecht, 1994, Maselli et al., 1989, Niakan and Harati, 1988).
10. Autonomic testing in specific clinical situations
The range of clinical presentations of autonomic disorders is vast. Whereas a complete differential diagnosis of autonomic disorders is beyond the scope of this paper, a number of common indications may be identified in which autonomic testing has been shown to be useful in the clinical evaluation. In the following sections we review the evidence that supports the use of autonomic testing for each indication.
10.1. Orthostatic hypotension
Often the most disabling of autonomic symptoms, orthostatic hypotension (OH) is defined as a sustained reduction of systolic BP of ≥ 20 mmHg or diastolic BP of ≥ 10 mmHg within 3 minutes of standing or head-up tilt to at least 60° (Freeman et al., 2011). The change in systolic BP correlates more closely than diastolic BP with orthostatic symptoms (Fedorowski et al., 2017). Tilt table testing for 3–5 minutes is usually adequate to detect OH (Cheshire and Goldstein, 2019), but in some cases the development of OH can be delayed (Gibbons and Freeman, 2015).
An important clinical distinction is that of neurogenic OH. Whereas OH from inadequate fluid intake, excessive fluid loss, deconditioning, venous pooling, or medications is quite common, a minority of patients with OH will have neurogenic OH, which is a cardinal manifestation of cardiovascular sympathetic adrenergic failure (Goldstein and Sharabi, 2009).
A decreased HR response to orthostatic hypotension has been shown to be a reasonable surrogate marker of neurogenic OH. In a prospective study of 423 patients, 378 of whom had α-synucleinopathies, neurogenic OH was reliably distinguished from other causes of OH when the ΔHR/ΔSBP ratio at 3 minutes of tilt was < 0.5 beats/min per mmHg (Norcliffe-Kaufmann et al., 2018). However, this criterion may not be adequate in patients with partial or early autonomic failure, patients taking beta blockers, or those with a cardiac pacemaker (Cheshire and Goldstein, 2019).
10.2. Orthostatic intolerance
Many patients present with symptoms that occur when standing and are relieved by lying down and yet do not have OH. When disabling postural symptoms result from the physiologic response to the orthostatic stress in which gravity causes blood to pool upon standing, these patients are said to have orthostatic intolerance. The most recognized variety of chronic orthostatic intolerance is the postural tachycardia syndrome (POTS), which is a heterogeneous clinical condition defined as a sustained HR increment of ≥ 30 beats/min (≥40 beats/min for patients < 20 years of age) within 10 minutes of standing or head-up tilt in the absence of orthostatic hypotension (Benarroch, 2012, Freeman et al., 2011, Singer et al., 2012). The diagnosis is based on averaged, rather than momentary peak, heart rates.
Tilt table testing with beat-to-beat assessment of BP and HR represents the standard of care in diagnosing POTS and is recommended by the Heart Rhythm Society for this purpose (Cheshire and Goldstein, 2019, Sheldon et al., 2015). Diagnostic evaluation should also include estimation of intravascular volume status. Additional autonomic testing may be warranted in selected patients to assess for peripheral denervation and hyperadrenergic state, as these patients have been found to have lower resting muscle sympathetic nerve activity, impaired cardiovagal responses, exaggerated BP drops in response to the Valsalva maneuver, and delayed cardiovascular adrenergic responses to hypotensive challenge (Arnold et al., 2018, Jacob et al., 2019, Low et al., 2009).
Whereas isolated instances of neurally mediated syncope are quite common and often can be diagnosed by a careful history, recurrent or unexplained syncope are more serious matters that can be diagnostically challenging (Cheshire, 2017).
[See this section for details.]
10.4.1. Distal small fiber neuropathy
The “burning feet” syndrome is perhaps the most common presentation of distal small-fiber neuropathy (DSFN) in clinical practice. These patients complain of distal burning, prickling, and some stabbing discomfort, with variable allodynia. They have completely normal motor function, intact tendon reflexes, and nerve conduction studies. About 10% of such cases are due to diabetes, and the cause remains unknown in as many as half of cases (Terkelsen et al., 2017). As both somatic and autonomic C fibers are involved, but large myelinated fibers are usually spared, this condition can be diagnosed by sudomotor testing when nerve conduction studies are normal (Illigens and Gibbons, 2009, Low et al., 2006, Thaisetthawatkul et al., 2013). Test results should be correlated with the patient’s symptoms and small fiber sensory findings on physical examination. Table 5 categorizes the degree of diagnostic confidence associated with specific QSART profiles.
Table 5. Degrees of confidence in diagnosing small fiber neuropathy.
QSART Profile Probability of SFN Normal at all sites No evidence Abnormal but with potential medication effect Indeterminate Sudomotor volume < 5th percentile at 1 site without neuropathic symptoms or sensory exam abnormalities Low probability Sudomotor volume < 5th percentile at 1 or more sites with uncertain correlation anatomically with symptoms or exam findings Intermediate probability Sudomotor volume < 5th percentile at the foot and a length-dependent decrease, defined as a sudomotor volume at the foot less than one-third that at proximal sites, or Sudomotor volume < 5th percentile at the foot and either a history of distal dysesthesia or a distal deficit to small fiber sensory modalities on physical exam High probability
QSART = quantitative sudomotor axon reflex test; SFN = small fiber neuropathy.
10.4.2. Regional neuropathy
The TST is particularly sensitive in detecting the neuroanatomical contours of regional anhidrosis, which are useful in identifying localizable deficits. The boundaries of segmental anhidrosis can define thoracic radiculopathies, cervical sympathetic deficits associated with Horner or Harlequin syndromes, or map the extent of surgical sympathectomy or subsequent reinnervation (Cheshire and Freeman, 2003, Fealey et al., 1989).
10.4.3. Inflammatory demyelinating neuropathy
Acute inflammatory demyelinating neuropathy, also known as Guillain-Barré syndrome, is an immune-mediated disorder of nerves and nerve roots that presents acutely. Autonomic involvement occurs in at least two-thirds of patients and can include tachycardia, bradycardia, hypertension, OH, urinary sphincter disturbances, and anhidrosis (Singh et al., 1987, Zochodne, 1994). In a series of 100 Guillain-Barré patients, low R-R interval variation on deep breathing was associated with increased incidence of serious cardiac rhythm disturbances (Winer and Hughes, 1988).
Chronic inflammatory demyelinating neuropathy (CIDP) is considered to be a chronic form of Guillain-Barré syndrome. Autonomic involvement has been described in 21% to 76% of cases of both the demyelinating and axonal subtypes (Stamboulis et al., 2006). Autonomic testing has detected subclinical signs of involvement in as many as 80% of patients without overt autonomic symptoms (Ingall et al., 1990, Lyu et al., 2002). Autonomic manifestations, though frequent, tend to be mild (Stamboulis et al., 2006). In a retrospective study of 47 CIDP patients, CASS scores were abnormal in 47%, but mild, with a mean ± SD of 0.8 ± 0.9, being ≤ 3 in all cases (Figueroa et al., 2012). This leads to the conclusion that, if autonomic testing in suspected CIDP were to show extensive or severe autonomic failure, then an alternative diagnosis should be sought.
10.4.4. Autoimmune autonomic ganglionopathy
Impaired cholinergic ganglionic synaptic transmission is one of the causes of severe autonomic failure. A subset of autonomic ganglionopathies is autoimmune and may be positive for ganglionic α3-AChR antibody (Vernino et al., 2000). These patients have severe dysautonomia with postural hypotension, gastrointestinal dysmotility, anhidrosis, sicca symptoms, and pupillary and erectile dysfunction (Vernino et al., 2009).
Autonomic testing has proven useful in in characterizing the various autonomic presentations of autoimmune autonomic ganglionopathy (Sandroni and Low, 2009). In a retrospective study of 289 patients with positive ganglionic α3-AChR antibodies who had undergone autonomic testing, CASS scores correlated closely with antibodies, with levels above 0.40 nmol/L predicting CASS scores of ≥ 7 (Cutsforth-Gregory et al., 2018).
10.4.5. Sensory neuronopathy
Autonomic dysfunction frequently accompanies sensory neuronopathies, which are characterized by non-length-dependent sensory deficits. Cardiovascular adrenergic, cardiovagal, and postganglionic sudomotor deficits have been described in nearly all patients. The autonomic involvement can be disabling, with orthostatic hypotension in 60% of patients adding to the risk of falling from sensory ataxia (Damasceno et al., 2011, Martinez et al., 2019).
10.4.6. Acute intermittent porphyria
Acute intermittent porphyria is an autosomal dominant inborn error of metabolism in which a block in the enzymatic biosynthesis of heme leads to excessive secretion of porphyrins and porphyrin precursors. Although quite rare, it is of importance for this discussion because symptomatic attacks are typically heralded by an autonomic neuropathy. The most common presenting symptom is abdominal pain caused by splanchnic autonomic dysfunction. Parasympathetic dysfunction may be an early feature. Other autonomic manifestations can include tachycardia, labile hypertension, orthostatic hypotension, hyperhidrosis, vomiting, bladder dysfunction, and constipation or diarrhea (Laiwah et al., 1985). Abnormalities on autonomic testing during symptomatic attacks have been shown to be reversible (Laiwah et al., 1985).
10.5. Diabetic neuropathies
Particular attention is given here to diabetic neuropathy, as diabetes mellitus is the most common cause of autonomic neuropathy in the developed world, and prevalence is increasing in the developing world (Freeman, 2005, Freeman, 2014).
In summary, there is strong evidence that autonomic function is impaired in diabetes and progresses over time. The presence of DAN worsens the prognosis by mechanisms that are likely a combination of dysautonomia and vascular disease (Agashe and Petak, 2018). As tight diabetic control is known significantly to prevent the development of diabetic autonomic neuropathy (Nathan et al., 1993), there has been considerable interest in the development and utilization of autonomic function tests in diabetes (Bernardi et al., 2011, Bissinger, 2017). The Toronto Consensus Panel on Diabetic Neuropathy recommends that all diabetic patients should be screened for autonomic neuropathy (Bernardi et al., 2011).
The term pandysautonomia is often used when the patient has severe generalized autonomic failure, which can occur on the basis of central or peripheral autonomic nervous system disease. Typically the patient has orthostatic hypotension and involvement of at least two other systems such as neurogenic bladder or bowel or thermoregulatory failure.
Pandysautonomia occurs in a number of settings. One is end-stage diabetic neuropathy, where the patient has severe retinopathy, nephropathy, and neuropathy with progressive autonomic failure over a number of years (Low, 1996, Low and Hilz, 2008). Another is the autonomic neuropathy associated with primary and familial amyloidosis. Pandysautonomia frequently accompanies the parkinsonism or cerebellar ataxia that together are the clinical hallmarks of multiple system atrophy.
10.7. Familial dysautonomia