Quantitative Studies of Autonomic Function

Max J. Hilz, MD, PhD; Matthias Dütsch, MD

No one involved in the planning of this CME activity has anyy relevant financial relationships to discllose. Authors/faculty have nothing to disclose.

CME is available 7/15/2008 - 7/15/2011

Copyright: 20006

American Association of Neuromuscular and Electrodiagnostic Medicine

2621 Superior Dr NW Rochester, MN 55901

The ideas and opinions in this Journal Review are solely thhose of the author and do not necessariily represent those of the AANEM

Product: JR07

CME Information Product: JR07 ‐ Quantitative Studies of Autonomic Function

Course Description Dysfunction of the peripheral and central autonomic nervous system is common in many neurological and general medical diseases. The quantitative assessment of sympathetic and parasympathetic function is essential to confirm the diagnosis of autonomic failure, to provide the basis for follow‐up examinations, and potentially to monitor successful treatment. Various procedures have been described as useful tools to quantify autonomic dysfunction. The most important tests evaluate cardiovascular and sudomotor autonomic function. In this review, we therefore focus on standard tests of cardiovascular and sudomotor function such as heart‐rate variability at rest and during deep breathing, active standing, and the Valsalva maneuver, and on the sympathetic skin response. These tests are widely used for routine clinical evvaluation in patients with peripheral neuropathies. Refined methods of studying heart‐rate variaability, baroreflex testing, and detailed measures of sweat output are mostly used for research purposes. In this context, we describe the spectral analysis of slow modulation of heart rate or pressure, reflecting sympathetic and parasympathetic influences, and consider various approaches to baroreflex testing, the thermoregulatory sweat test, and the quantitative sudomotor axon reflex test. Finally, we discuss microneurography as a technique of direct recording of muscle sympathetic activity. Intended Audience This course is intended for Neurologists, Physiatrists, and others who practice neuromuscular, musculoskeletal, and electrodiagnostic medicine with the intent to improve the quality of medical care to patients with muscle and nerve disorders.

Learning Objectives Upon conclusion of this program, participants should be able to: 1. identify the various procedures described as useful tools to quantify autonomic dysfunction. 2. discuss the spectral analysis of slow modulation of heart rate or , reflecting sympathetic and parasympathetic influences, and consider various approaches to baroreflex testing, the thermoregulatory sweat test, and the quantitative sudomotor axon reflex test. 3. recognize microneurography as a technique of direct recording of muscle sympathetic nerve activity.

Release Date: 7/15/2008 Expiration Date: 7/15/2011. Your request to receive AMA PRA Category 1 Credits™ must be submitteed on or before the credit expiration date. Duration/Completion Time: 3 hours

Accreditation and Designation Statements The American Association of Neuromuscular and Electrodiagnostic Medicine (AANEM) is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AANEM designates both sections of this enduring material for a combined maximum of 3 AMA PRA Category 1 creditsTM. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

How to Obtain CME Once you have reviewed the materials, you can obtain CME credit by clicking on the link in the e‐mail received when you purchased this product. Answer the questions and click submit. Once your answers have been submitted, you will be able to view a transcript of your CME by logging into www.aanem.org and clicking View Profile and then My CME. INVITED REVIEW ABSTRACT: Dysfunction of the peripheral and central autonomic nervous system is common in many neurological and general medical diseases. The quantitative assessment of sympathetic and parasympathetic function is essential to confirm the diagnosis of autonomic failure, to provide the basis for follow-up examinations, and potentially to monitor successful treatment. Various procedures have been described as useful tools to quantify auto- nomic dysfunction. The most important tests evaluate cardiovascular and sudomotor autonomic function. In this review, we therefore focus on stan- dard tests of cardiovascular and sudomotor function such as heart-rate variability at rest and during deep breathing, active standing, and the Val- salva maneuver, and on the sympathetic skin response. These tests are widely used for routine clinical evaluation in patients with peripheral neurop- athies. Refined methods of studying heart-rate variability, baroreflex testing, and detailed measures of sweat output are mostly used for research pur- poses. In this context, we describe the spectral analysis of slow modulation of heart rate or blood pressure, reflecting sympathetic and parasympathetic influences, and consider various approaches to baroreflex testing, the ther- moregulatory sweat test, and the quantitative sudomotor axon reflex test. Finally, we discuss microneurography as a technique of direct recording of muscle sympathetic nerve activity. Muscle Nerve 33: 6–20, 2006

QUANTITATIVE STUDIES OF AUTONOMIC FUNCTION

MAX J. HILZ, MD, PhD,1,2 and MATTHIAS DU¨ TSCH, MD2

1 Department of Neurology, New York University Medical Center, 550 First Avenue, Suite NB 7W11, New York, New York 10016, USA 2 Department of Neurology, University of Erlangen-Nuremberg, Erlangen, Germany

Accepted 12 April 2005

Studies of autonomic function are best considered challenge.23,87 There are various parameters of by the systems being tested and then by reference to heart-rate variability.23,84,87 The standard deviation of whether the tests are routine or research techniques. the heart rate, i.e., the square root of its variance, recorded during an interval of, for example, 5 min, CARDIOVASCULAR AUTONOMIC TESTING or the coefficient of variation reflects the influence of the parasympathetic and sympathetic system on Routine Cardiovascular Autonomic Testing: Analysis of heart-rate modulation.84,87 The coefficient of varia- Heart-Rate Variability in the Time-Domain. Biosignals tion, i.e., the standard deviation divided by the mean such as heart rate or blood pressure can be recorded interval between two electrocardiographic R waves as time series and show a continuous variability that (RR interval) during the measuring period, has been is altered with autonomic dysfunction. Similar to the introduced for cardiovascular autonomic evaluation electrocardiogram, the variability of heart rate can as it minimizes the dependence of the standard de- be assessed under resting conditions and during viation on resting heart rate. The standard deviation depends on the intrinsic heart rate of the individual patient and moreover is significantly altered by sin- Abbreviations: ACh, acetylcholine; CFT, cold face test; CPT, cold pressor 29,31 test; E/I ratio, expiratory–inspiratory ratio; E-I difference, expiratory–inspiratory gle extrasystoles or stepwise heart-rate changes. difference; HF, high-frequency; LF, low-frequency; MSNA, muscle sympa- The length of the recorded heart-rate time series thetic nerve activity; pNN50, proportion of differences in consecutive RR 87 intervals Ͼ 50 ms; QSART, quantitative sudomotor axon reflex test; RMSSD, also influences the standard deviation. Usually, root-mean-square of successive differences; SSR, sympathetic skin re- resting heart-rate variability is analyzed from short- sponse; TST, thermoregulatory sweat test; VLF, very-low-frequency 23 Key words: autonomic nervous system; baroreflex testing; heart-rate vari- term segments of 5-min duration. ability; microneurography; sudomotor function The pNN50 is a parameter that indicates the Correspondence to: Max J. Hilz; e-mail: [email protected] proportion of differences in consecutive, so-called © 2005 Wiley Periodicals, Inc. Published online 17 June 2005 in Wiley InterScience (www.interscience.wiley. normal-to-normal RR intervals that are longer than com). DOI 10.1002/mus.20365 50 ms. The parameter is independent of long-term

6 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 trends in heart rate37 and reflects the percentage of creases during orthostatic challenge, mental stress, such intervals in comparison to the total number of and moderate exercise.87 Sympathetic as well as para- analyzed intervals. The pNN50 is a parameter of sympathetic outflow may influence the power of low- parasympathetic activity.23,84,87 Similarly, the root- frequency heart-rate modulation. The heart-rate mean-square of successive differences (RMSSD) re- modulation in the HF range is mainly due to para- flects parasympathetic activity and does not depend sympathetic efferent influences and represents respi- on heart-rate trends.29,84,87 RMSSD is calculated as ratory variability.7,84,87 For heart-rate variability, the the square root of the mean squared differences of power of signal modulation is assessed as the area successive RR intervals.84,87 RMSSD is influenced by under the spectral density curve in the frequency premature ventricular contractions with long com- range considered to reflect sympathetic or parasym- pensatory pauses. It is assumed that this parameter is pathetic activity.37 If heart-rate variability is mea- robust against gradual trends in heart rate over time sured as beats per minute, the power is given in and is independent of mean heart rate. bmp2; if measurements are taken as RR intervals, the We consider the coefficient of variation and the power is expressed in ms2.16 The integral under the RMSSD the most valuable time-domain parameters power spectral density curve is calculated for fre- for routine evaluation at rest as they provide highly quency ranges that have been shown to be influ- reproducible results and are not influenced by mean enced by sympathetic and parasympathetic activity resting heart rate. using sympatholytic medication or atropiniza- tion.16,37 Assessment in Research: Frequency-Domain Analysis Some authors prefer to calculate the relative of Heart-Rate Variability. Using time-domain analy- value of the LF or HF power components in propor- sis, autonomic cardiovascular regulation is often in- tion to the total power in the frequency range from vestigated by measuring mean responses of heart 0.04 to 0.4 Hz.2 Other authors calculate the ratio rate or blood pressure to external stimuli, e.g., tilt- between LF and HF power as a parameter of the ing. However, the calculated net changes of blood sympatho-vagal balance.60,87 During increased sym- pressure or heart rate often disregard information pathetic activity, the LF–HF ratio will show higher on the instantaneous dynamics of autonomic modu- values.59 During predominance of the parasympa- lation.79,87 Analysis of spontaneously occurring fluc- thetic system, the ratio will show smaller values.59 tuations in autonomic tone by means of frequency- Various algorithms have been applied to deter- domain (i.e., spectral) analysis might offer a more mine the frequencies accounting for the modulation subtle insight into the physiological mechanisms of of heart rate and other biosignals by sympathetic or autonomic cardiovascular control.79,87 parasympathetic tone.37,84 Among the algorithms Spectral analysis allows the overall variance of a used for the frequency analysis of biosignals are biosignal to be split into its various underlying fre- parametric and nonparametric methods.37 An im- quency components as most biosignals that vary portant requirement of spectral analysis is the “sta- around a mean value can be reconstructed as a sum tionarity” of the signal, a condition easily fulfilled in of sines and cosines at different frequencies.7 Differ- technical signals but mostly not existing in human ences in the magnitudes of the different frequencies physiology.37,87 If the mechanisms accounting for or oscillations can be attributed to differences in the the signal modulation in a particular frequency are neural influences responsible for cardiovascular reg- changing due to physiological changes (for example, ulation. The magnitudes of slow underlying frequen- during exercise) the significance of a given fre- cies are widely considered indices of sympathetic quency component for the modulation of the bio- and parasympathetic tone.37,84 signal is no longer well-defined.87 Moreover, the Three main components within the slow under- powers of signal oscillations in frequency bands con- lying frequencies have been used in the literature: sidered to reflect sympathetic or parasympathetic the very-low-frequency range (VLF Յ 0.04 Hz), the influence cannot be considered an absolute measure low-frequency range (LF: 0.04–0.15 Hz) and the of sympathetic or parasympathetic outflow.37,58,87 high-frequency range (HF: 0.15–0.4 Hz).37,87 In re- Heart rate during exercise will be high due to a cordings of only 5 min, the interpretation of the VLF predominant sympathetic tone. Yet, the oscillation component is unclear.37,41,87 The fluctuation in of sympathetic influences of heart rate might be heart rate in the LF range is influenced by the lower than under resting conditions because of the baroreceptor system43,84 and reflects sympathetic as continuously high sympathetic outflow.37 Conse- well as parasympathetic influences, particularly un- quently, the power of heart-rate modulation in the der resting conditions.7,60,84 The LF modulation in- frequency range considered to reflect sympathetic

Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 7 activity will be low and might be misinterpreted as a deficiency of sympathetic outflow, although sympa- thetic tone is actually high.19,37 Frequency-domain analysis generally requires rel- atively long, nearly artifact-free recordings. Evalua- tion of the recorded signals is time consuming and requires specific software and computer techniques. Although the parameters of spectral analysis are use- ful measures of cardiovascular autonomic function, this technique is therefore mostly restricted to re- search activities.

Autonomic Challenge Maneuvers. Various maneu- vers are suited to activate the sympathetic or para- sympathetic nervous system.24,50 According to the criteria of the American Diabetes Association, dia- betic autonomic neuropathy—the most common au- tonomic neuropathy in civilized countries—is char- acterized by both, cardiovascular autonomic dysfunction and impaired sudomotor function. The American Diabetes Association and the American Academy of Neurology have recommended a stan- dard battery for the screening of autonomic dysfunc- tion.14 Tests, primarily analyzing parasympathetic heart-rate control are the Valsalva maneuver, metro- nomic breathing, and active standing. Active stand- ing can also be used to monitor sympathetic blood pressure control.14 For routine evaluation of cardiovascular auto- FIGURE 1. Normal heart-rate variability during a Valsalva ma- neuver in a 36-year-old subject, showing four defined phases nomic function, we consider the Valsalva maneuver, (I–IV) and baroreflex-mediated bradycardia (a). Absent modula- active standing, and especially metronomic breath- tion of heart rate in a 37-year-old diabetic patient. During strain, ing the most valuable tests. The many other tests there is no increase of heart rate; after release of the strain, there activating sympathetic or parasympathetic tone in- is no reflex bradycardia (b). (bpm, beats per minute; s, seconds.) clude the cold pressor test, the cold face test, squat- ting, coughing, and mental arithmetic.24,50 Valsalva Maneuver. The Valsalva maneuver is duces cardiac stroke volume, which again results in a known to influence heart rate, but actually evaluates baroreflex-mediated and peripheral va- the baroreflex arc and its sympathetic and parasym- soconstriction. Phase 3 describes the first 1–2 s after pathetic responses. During the maneuver, the per- release of the expiratory strain. There is a passive son breathes into a special mouthpiece that is con- decline in blood pressure and increase in heart rate. nected to a manometer and maintains an expiratory Finally, phase 4 shows a blood pressure overshoot pressure of 40 mmHg for 15–20 s.5 The maneuver that is due to the persistent increase in peripheral consists of four phases5 (Fig. 1). Phase 1 occurs resistance, and a normalization of venous return and during the first 2–3 s of the forced expiration and stroke volume.37 Mean blood pressure can increase shows a brief decrease in heart rate and increase in by more than 10 mmHg, but an increase is absent in blood pressure due to mechanical compression of patients with sympathetic dysfunction.63 The blood the aorta. During phase 2, blood pressure first de- pressure increase mediates a baroreflex-induced bra- creases and then increases in the late portion of this dycardia. This bradycardia is a measure of baroreflex phase. The initial decrease in blood pressure acti- buffer capacity and vagal cardiac innervation. Blood vates the baroreflex and results in an increase of pressure changes during phases 2 and 4 reflect sym- sympathetic activity and a consequent increase of pathetic responses. peripheral resistance and blood pressure during the The Valsalva ratio is used as an index of the late stage of phase 2. Venous return is lowered be- baroreflex-mediated bradycardia and is calculated as cause of the continuous expiratory strain and re- the ratio of the highest heart rate during expiration

8 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 brainstem nuclei and thus modulate the sympathetic and parasympathetic outflow to the heart.78 Heart- rate modulation during deep breathing depends on parasympathetic cardiac control and is largely re- duced with atropinization.28,44 A respiratory fre- quency of six cycles per minute induces maximal heart-rate variability in healthy persons.1,69,90 Low et al. suggest that the average of the five largest con- secutive responses in eight respiratory cycles is ana- lyzed.53 Others prefer to use shorter segments in order to avoid hypocarbia6 with prolonged hyper- ventilation.47 The expiratory–inspiratory difference (E-I difference) or the expiratory–inspiratory ratio (E/I ratio) can be determined from the maximum and minimum heart rate during respiration at six cycles per minute.31 The values depend not only on age and sex but also on body position, body weight, drugs, and respiratory effort.50,68,90 Single reference values are therefore misleading.29 E-I differences ex- ceeding 15 heart beats per minute are considered to FIGURE 2. Heart-rate variability (HRV) during 6/min breathing: be normal.57 E-I differences below 10 beats per normal HRV in a 39-year-old control (a); absent HRV in a 38- minute are abnormal in persons below the age of 40 year-old patient with diabetic autonomic neuropathy (b). (bpm, 67 beats per minute.) years. Differences below 5 beats per minute are abnormal in persons older than 50 years.90 The E/I ratio should exceed values of 1.23 in persons and the lowest heart rate during the first 20 s after younger than 20 years.82 The E/I ratio should be the expiratory strain.37 The ratio depends not only higher than 1.06 in patients between 76 and 80 on age and gender of the tested person but also on years.82 body position, and the duration and intensity of the Sustained Handgrip Test. Ewing and Clarke24 in- expiratory strain.50 The maneuver should be re- troduced the sustained handgrip test as a standard peated several times to assure reproducibility and part of a battery used to diagnose diabetic cardiac reliability of results.50 When performed carefully and autonomic neuropathy. The test results are, how- evaluated correctly, the Valsalva maneuver is a help- ever, quite variable.37 First, the subject is asked to ful tool in the assessment of cardiovascular auto- press a handgrip dynamometer with full strength. nomic function. Low and coworkers established age- Then, the handgrip should be maintained for 3–5 and gender-specific normative data.50 A Valsalva ra- min at one-third of the maximum contraction tio below 1.10 is abnormal; according to some au- strength.37 The consequent increase of sympathetic thors, even a ratio below 1.2 is abnormal.63 activity and vasoconstriction induces a blood pres- Deep (Metronomic) Breathing. Deep metronomic sure rise that is considered to reflect sympathetic breathing at a rate of six cycles per minute is prob- autonomic activity.37 The early acceleration of heart ably the most common and reliable test to assess rate during the maneuver is due to a withdrawal of respiratory sinus , with acceleration of vagal activity, whereas the late heart-rate acceleration heart rate during inspiration and deceleration dur- results from sympathetic activation.11,50 Normally, di- ing expiration under optimized conditions (Fig. astolic blood pressure at the end of the effort is at 2).1,69,90 There is a decrease of sinus arrhythmia with least 16 mmHg higher than before the maneuver.37 increasing age.53,64 A diastolic blood pressure increase by only 10 mmHg The central neuronal circuitry involved in inspi- or less is abnormal. Many patients perform a Valsalva ration likely inhibits cardiovagal neurons.49,72 maneuver during the handgrip test and conse- Changes of the central venous blood volume during quently bias the test result.65 respiration also influence heart rate by activation of Cold Pressor Test. The cold pressor test (CPT) the Bainbridge reflex while the mechanical modula- consists of immersion of one hand and arm in ice tion of blood pressure activates or inhibits the cold (0–4°C) water for 40–180 s. The cold stimulus baroreflex and thus affects heart rate.78 Moreover, activates afferent pain and temperature fibers from stretch receptors in the lungs and chest walls activate the skin.32,42 The impulses pass via the spinothalamic

Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 9 tract to several brain areas and result in sympathet- tients unable to cooperate with other challenge ma- ically mediated heart-rate acceleration, peripheral neuvers.38 During the CFT, the bradycardia is in- vasoconstriction, and an increase in blood pressure duced by cold, wet, or noxious stimulation of the (Fig. 3).32,39 In the first 30 s, blood pressure seems to face via reflex centers located in the brainstem re- increase due to a rise of heart rate and cardiac gion.46 Efferent sympathetic pathways mediate pe- output. Robertson reports an average increase of ripheral vasoconstriction and consequent blood systolic blood pressure by 20 mmHg and of heart pressure increase.34 Efferent cardiac parasympa- rate by 10 beats per minute (bpm).74 Vasoconstric- thetic pathways mediate bradycardia,46 which can be tion can be recorded by monitoring superficial skin abolished by atropinization or .34,46 A dis- blood flow, e.g., at the digit pulps, using laser Dopp- turbance in integrity of the trigeminal-brainstem- ler flowmetry.54 According to Yamamoto et al., pe- vagal reflex arc at any level yields a pathological CFT ripheral vasoconstriction contributes to blood pres- response with absent or diminished bradycardia.46 sure elevation at a later stage of CPT.93 With hand Khurana et al. observed an attenuated cardiovagal immersion into 4°C water, muscle sympathetic nerve response or even minimal tachycardia in response to activity increased after 30 s and peaked after 60–90 s cold face stimulation in patients with diabetes melli- of cold stimulation.93 In contrast, Low and cowork- tus, brainstem stroke, or Shy–Drager syndrome.46 ers observed a 51% reduction in skin blood flow at the index finger pulp after only 20 s of hand immer- Orthostatic Challenge by Head-Up Tilting or Active sion in 0–2°C cold water.54 These results suggest that Standing. Orthostatic challenge causes an early car- responses differ significantly depending on the stim- diovascular response that occurs within the first ulus duration, water temperature, and whether the 30 s.91,92 During assumption of the upright position, hand or arm is immersed in the ice water. 300–900 ml of blood are redistributed from central In patients with afferent small-fiber neuropathy blood vessels to the lower extremity.10,12,75 The early or spinothalamic tract dysfunction or with central or circulatory stabilization occurs after 1–2 min of or- efferent sympathetic lesions, such as occur in dia- thostasis. Finally, there is a response to prolonged betic or alcoholic autonomic neuropathies, the re- orthostasis lasting for more than 5 min.91,92 sponses to CPT are diminished or even absent.54,61 The early phase of stabilization and the adjust- Reproducibility of CPT results should be assured, ments during prolonged orthostasis are similar for although subjects often consider the test unpleasant active standing and passive tilting.91,92 Passive tilt- and are reluctant to participate in a second test ing is better suited to assess neurocardiogenic syn- run.74 Therefore, age-matched control data are re- copes, i.e., to evaluate neuronal control during quired for a meaningful interpretation of the re- long-duration orthostatic challenge.91,92 More- sults.29 over, passive tilting is more suited than active Cold Face Test. The cold face test (CFT) consists standing if patients have lower-limb weakness or if of application of cold compresses (1°C to 2°C) to the there is a need to return patients rapidly to the forehead and maxillary region of the subject’s face supine position.91,92 for a period of 60–180 s.38 Cold stimulation of the Active standing (the so-called Ewing maneuver) forehead and maxillary region activates the periph- is more suited to assess responses during the initial eral sympathetic and the cardiac parasympathetic phase of orthostatic challenge.91,92 During active nervous system.38 The combined activation induces a standing, there is a contraction of abdominal and leg decrease of heart rate as well as a peripheral vaso- muscles and a subsequent compression of the resis- constriction with subsequent blood pressure in- tance and capacitance vessels with increase of the crease (Fig. 4).27,34,46 Bradycardia is the best docu- intra-abdominal pressure.10,86,91,92 Consequently, ve- mented and most easily quantified of the circulatory nous return and cardiac output increase. Yet, this responses following CFT.46 Baroreflex activation is increase does not fully compensate for the decline in probably not essential to induce bradycardia, as the total peripheral resistance occurring upon active heart rate decrease has been recorded even in the standing.91,92 Therefore, there is a transient reduc- absence of blood pressure changes.27,34 tion in blood pressure during active standing.91,92 The CFT is a modification of the so-called diving Muscle contraction activates an exercise reflex that reflex that occurs with immersion of the face in results in a rapid withdrawal of parasympathetic ac- water. The CFT is more comfortable than the diving tivity and a consecutive heart-rate acceleration reflex, but has similar sensitivity towards inducing a within the first 3 s after standing up. The initial cardiovagal response. CFT is noninvasive and can be reduction in peripheral resistance accounts for a used to evaluate parasympathetic responses in pa- brief drop in blood pressure. The resulting inhibi-

10 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 FIGURE 3. Heart rate (HR), blood pressure (BP), and skin blood flow (SBF) during cold pressor test in a control and in a patient with familial (FD). (BPdia, diastolic blood pressure; BPmean, mean blood pressure; BPsys, systolic blood pressure.) In the patient, HR and SBF remain stable, BP increase is reduced and delayed. In the control, SBF decreases initially and shows secondary hyperperfusion (Hunting phenomenon).

Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 11 FIGURE 4. Heart rate (HR) (a), systolic blood pressure (BPsys) (b), and skin blood flow (SBF) (c) during cold face test (CFT). In the control, HR and SBF decrease while BP increases. In the patient with sympathetic and parasympathetic failure, responses are attenuated. tion of the baroreflex induces an ongoing inhibition creased baroreflex activity and reduced stimulation of cardiac parasympathetic activity as well as an aug- of cardiopulmonary receptors assures blood pres- mentation of sympathetic outflow, both resulting in sure recovery after approximately 7 s. Healthy per- a secondary, gradual rise in heart rate.91,92 The de- sons might even show a blood pressure overshoot.37

12 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 The initial decline and secondary overshoot of systolic and diastolic blood pressure within the first seconds of orthostatic challenge can be recorded by continuous noninvasive blood pressure measure- ments. According to Wieling et al., an initial drop in systolic blood pressure by more than 40 mmHg or in diastolic blood pressure by more than 25 mmHg is abnormal.92 After standing for 30 s, heart rate and blood pressure are normalized.91,92 Ewing suggested that the ratio is calculated between the highest heart rate, i.e., the shortest RR interval, after approxi- mately 15 heart beats from standing up, and the slowest heart rate, i.e., the longest RR interval, after approximately 30 heart beats from beginning the challenge, as a parameter of orthostatic cardiac re- sponse.95 This 30/15 ratio should be above 1.04, yet values are age-dependent.20,22,95 Responses can also be quantified as the difference between the highest heart rate within 15 s after standing and the baseline heart rate.91,92 According to Wieling et al., heart rate should initially increase by more than 20 bpm in young persons (10–14 years old); the increase should not be below 11 bpm in persons aged 75–80 years.91,92 After the initial response to active standing (or passive head-up tilt), there is the early phase of stabilization during 1–2 min of standing. This phase shows an increase in diastolic blood pressure by approximately 10 mmHg and a sympathetic heart- rate increase by 10 bpm.91,92 During 5–10 min of active standing or passive FIGURE 5. Blood pressure (BP) and heart rate (HR) during head-up tilt in a control and in a patient with familial dysautono- head-up tilt, the sympathetic outflow is rather con- mia (FD). (BPdia, diastolic blood pressure; BPmean, mean blood stant and heart rate and blood pressure remain sta- pressure; BPsys, systolic blood pressure.) The patient shows ble.91,92 orthostatic and absent reflex tachycardia. The initial cardiovascular responses to ortho- static challenge as well as the responses during the first 1–2 min of standing are under neurocardiovas- mmHg within 3 min of active or passive orthostatic cular control.91,92 During a prolonged standing challenge as criteria of .14 phase of more than 5 min, humoral mechanisms contribute to maintaining the blood pressure. Due Head-Up Tilting. Cardiovascular adaptation to pro- to an activation of the sympathetic system and the longed orthostatic challenge can be tested by means renin-angiotensin-aldosterone system, plasma cate- of passive head-up tilting.91,92 The cardiovascular cholamine levels increase.91,92 changes during head-up tilt are more gradual than According to Ewing and Clarke, a decline by during active standing.91,92 There is no biphasic re- more than 30 mmHg in systolic blood pressure dur- sponse of heart rate and blood pressure, but instead ing active standing is abnormal.23 Wieling et al. con- a gradual increase in diastolic pressure and heart sider a persistent decrease in systolic blood pressure rate and no major change in systolic pressure.91,92 by more than 20 mmHg after 1–2 min of standing or Consequently, passive orthostatic challenge results a decrease in diastolic blood pressure by more than in a 5–10 mm increase of mean blood pressure (Fig. 5–10 mmHg as abnormal responses.91,92 The consen- 5). The different effects seem to result from the sus committee of the American Autonomic Society abdominal and lower-limb muscle contractions dur- and the American Academy of Neurology defined a ing active standing and the absence of such contrac- decrease in systolic blood pressure by at least 20 tions during passive tilting.91,92 However, muscle ac- mmHg or in diastolic blood pressure by at least 10 tivation can only be avoided if the patient is not

Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 13 brought to the fully upright position. The gravita- rate and blood pressure upon changes of body posi- tional stress or hydrostatic effect during orthostasis tion. corresponds to the tilt angle.91,92 If the patient is The so-called sequence method analyzes the re- tilted to a 70° angle, the orthostatic challenge is lation between fluctuations in heart rate and blood almost identical with the challenge in the upright pressure under resting conditions.30,66 The barore- position (sin 70° ϭ 0.94).91,92 To assure that there is flex sensitivity is calculated as the slope of the regres- no muscle contraction, we only tilt to a 60° angle (sin sion between spontaneous sequences of blood pres- 60° ϭ 0.87).45,62 sure increases (or decreases) and simultaneously It should be noted that the 60° or 70° tilt angle occurring increases or decreases in the electrocar- cannot be used to assess mild cardiovagal dysfunc- diographic RR intervals.30,66 Using this approach, tion during the initial phase of orthostatic respons- Parati and coworkers found a baroreflex sensitivity es.91,92 These tilt angles are used to evaluate neuro- of 7.6 Ϯ 2.0 ms/mmHg for blood pressure increases, cardiovascular control during the phase of i.e., positive sequences, and a sensitivity of 6.4 Ϯ 1.5 stabilization, i.e., after 1–2 min of challenge and ms/mmHg for blood pressure decreases, i.e., nega- during prolonged 5–10 min, orthostasis.91,92 During tive sequences.66 In contrast to the standard method orthostatic stress, there is an increase of venous pres- of phenylephrine injection,70 the sequence method sure in the feet from 5–10 mmHg to approximately is noninvasive and does not require patient cooper- 90 mmHg.88 The higher hydrostatic pressure ac- ation. However, the sequence method only evaluates counts for progressive fluid transudation from circu- the baroreflex sensitivity without challenge and un- lating blood into the lower-extremity tissues. The der resting conditions and therefore only shows a volume shift is better demonstrated during pro- small portion of the baroreflex curve.37 longed passive tilting than during active standing Spectral analysis of heart-rate and blood pressure vari- with increased skeletal muscle tone that counteracts ability can also provide a measure of spontaneous blood pooling.81,92 The reduction of venous return baroreflex sensitivity.37 Instead of comparing the di- and diastolic cardiac filling upon orthostasis63,92 in- rect changes of blood pressure and consecutive duces a decrease of cardiac output.91,92 During pro- changes in heart rate, the changes in sympathetically longed orthostasis, stroke volume can be reduced by mediated spectral components of blood pressure 30–40%.91,92 Heart rate accelerates by approxi- modulation and the consecutive changes in the fre- mately 20% due to baroreflex activation. This heart- quencies reflecting sympathetic heart-rate modula- rate increase counteracts the cardiac output reduc- tion are used to describe baroreflex sensitivity.37 tion to some extent.91,92 Consequently, the decrease With this method, baroreflex sensitivity is assessed by of cardiac output is approximately 20%.91,92 analyzing the relationship between spontaneous, The increase in hydrostatic pressure in the lower- sympathetically mediated fluctuations of the blood extremity veins also activates the venoarteriolar axon pressure, reflecting the input activity of the barore- reflex, which increases vascular resistance by up to flex, and corresponding heart-rate fluctuations, re- 40% during standing.35,36,92 The venoarteriolar axon flecting the reflex output activity. The amplification reflex further reduces the decrease of cardiac output between input and output of the reflex is an index of and slows the loss of fluid into the surrounding baroreflex sensitivity. Mathematically, this amplifica- tissue.35,36 In response to orthostatic challenge, there tion equals the so-called gain of the transfer function is also a constriction of splanchnic resistance vessels between the oscillations of blood pressure and heart with expulsion of the splanchnic blood reservoir and rate in the LF range, provided there is sufficient a consecutive increase of venous return and cardiac coherence, i.e., a stable and reliable relationship filling.91,92 This splanchnic vasoconstriction is essen- between both biosignals. tial for orthostatic tolerance.92 Robbe and coworkers demonstrated that this The decrease in blood pressure and cardiac fill- baroreflex gain provides an index of baroreflex sen- ing upon orthostatic challenge activates the barore- sitivity that is closely correlated with the results of the flex which, in turn, inhibits cardiovagal outflow, en- phenylephrine method.73 The values assessed by cal- hances sympathetic activity, and thus increases heart culating the gain of the transfer function between rate, peripheral vascular resistance, and blood pres- low-frequency oscillations of blood pressure and low- sure.11,19,92 frequency oscillations of the RR interval (18.1 Ϯ 8.9 ms/mmHg) are similar to the baroreflex sensitivity Baroreflex Sensitivity Testing. Various approaches assessed by calculating heart-rate deceleration in re- can be used to test the sensitivity of the most impor- sponse to phenylephrine-induced blood pressure tant reflex arc assuring short-term stability of heart increases (16.2 Ϯ 7.3 ms/mmHg).73 Both the se-

14 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 quence method and the calculation of the low-fre- quency transfer function gain provide noninvasive measures of baroreflex sensitivity, but only evaluate the reflex under resting conditions; they do not assess blood pressure modulation in response to baroreflex activation.37 Moreover, the gain of the transfer function between systolic blood pressure and heart rate may depend on variables such as mechanical, respiratory, or movement-induced changes in blood pressure or heart rate.19,37,79 Pharmacological baroreflex testing is the standard approach for measuring baroreflex sensitivity that was introduced by Smyth et al.83 This so-called Ox- ford-method assesses heart-rate changes in response to pharmacological blood pressure stimulation. Rapid infusion of phenylephrine increases blood FIGURE 6. Activation of carotid baroreceptors by means of “neck suction” using a moldable lead collar. Sympathetic and parasym- pressure and thereby activates the baroreflex and pathetic baroreflex responses are induced by applying different generates a bradycardia.83 The baroreflex sensitivity levels of positive and negative pressures to the neck and the is calculated as the slope of the linear regression baroreceptors. between beat-to-beat systolic blood pressure values and values of the RR interval.83 Smyth and col- leagues calculated the baroreflex sensitivity between intervals.17 The sigmoid baroreflex response curve 2 and 15.5 ms/mmHg in awake subjects and between can be assessed by applying different levels of posi- 4.5 and 28.9 ms/mmHg in sleeping persons.83 tive and negative pressures. Eckberg and Fritsch18 The use of cardiovascular depressor or pressor used positive pressures of approximately 40 mmHg drugs better characterizes the baroreflex sensitivity and then applied stepwise decreases of pressure to 16,76 over a wide range of arterial pressures. The ap- Ϫ65 mmHg to determine various parameters of RR plication of ramps of increasing or decreasing arte- interval changes, such as the maximum, minimum, 18 rial pulses allows the entire sigmoid baroreflex and range of responses, the maximum slope of the 18 curve to be described. However, the method must reflex curve, or the operational point of the barore- be avoided in patients with arterial hypertension. ceptors.18 In contrast to pharmacological activation Moreover, it defines heart-rate responses but not of the baroreflex, neck chamber stimulation not blood pressure responses to baroreflex activation. only determines heart-rate responses but also blood In 1957, Ernsting and Parry described baroreflex pressure responses to baroreflex stimulation.18,19 20 stimulation by means of a neck chamber. The However, some patients do not tolerate the neck method was further modified by Eckberg and co- collar. Moreover, it is somewhat uncertain to what 17–19 7–9 workers and Bernardi et al. The carotid sinus extent the pressure within the collar is transmitted to and baroreceptors can be mechanically deformed by the baroreceptors.18,19 applying a positive or negative pressure to the neck. Nowadays, a moulded lead collar is used to cover the TESTING OF SUDOMOTOR FUNCTION anterior part of the neck (Fig. 6), and a positive or negative atmospheric pressure is applied to the res- Thermoregulatory Sweat Test. Central and periph- ervoir between the neck and the collar. A modified eral sudomotor function can be tested by means of vacuum cleaner is normally used to induce a nega- the thermoregulatory sweat test (TST).25,26 In order tive pressure within the neck collar. To avoid inter- to identify skin areas with sweat production, an indi- ference with the respiratory sinus arrhythmia, step- cator powder is required.25,26 The humidity of the wise pressure changes should be applied while the sweat changes the color of this powder. Conse- patient holds his or her breath at the end of normal quently, the patient’s entire skin surface has to be expiration.18,20 The negative pressure distends the dusted with the powder, and the patient remains in neck tissue including the baroreceptors and carotid a sweat chamber that is used to heat the body core artery, similar to the carotid artery and baroreceptor temperature, usually with infrared heaters.25,26 The distension occurring with a blood pressure increase. sweat chamber must be temperature- and humidity- The distending pressure is compared to baroreflex- controlled. Air temperature should be at 45–50°C, mediated changes in the electrocardiographic RR and humidity at 35%–40%.25,26 The heating time is

Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 15 approximately 45–65 min.25,26 There must be an increase in oral temperature by at least 1.0°C or to 38°C,3,13,25,26 but not to above 38.5°C.25,26 Anatomi- cal drawings or photographic recordings can be used to document areas of absent sweating, reduced sweating, or hyperhidrosis. Areas of abnormal sweat- ing are calculated as a percentage of the anterior body surface.13,25,26 The test can be used to assess sudomotor dysfunction in various diseases such as primary autonomic failure or central and peripheral secondary autonomic dysfunction, for example in neuropathies, myelopathy, or skin or sweat gland 25,26 disorders. Global anhidrosis can be seen in pri- FIGURE 7. Quantitative sudomotor axon reflex test: (1) baseline mary autonomic failure.3,13 The TST is abnormal in start, (2) iontophoresis start, (3) iontophoresis end, (4) baseline, patients with central sudomotor dysfunction whereas (A) maximum of humidity, (B) baseline humidity. The area under postganglionic sweating—as assessed with the quan- the curve reflects the sweat volume. titative sudomotor axon reflex test—usually shows preserved function of the postganglionic sudomotor 34.5°C to optimize the sweat response.52 The ionto- fibers and sweat glands in these patients.85 However, phoresis of acetylcholine is induced by a constant, the TST can also be used to monitor peripheral 2-mA anodal current from the stimulus compart- sudomotor function and assess severity and progres- ment to the skin. The cathode is placed on the skin sion of anhidrosis in an autonomic neuropathy, e.g., near to the stimulation site. in diabetic patients.25,26 Sweat responses are usually recorded at baseline, during 5 min of iontophoresis, and 5–10 min after Quantitative Sudomotor Axon Reflex Test. The quan- iontophoresis to observe the return of sweat output titative sudomotor axon reflex test (QSART) assesses to baseline. Normally, there is a 1–2 min latency the postganglionic sudomotor nerve fibers and the between the onset of iontophoresis and reflex-medi- sweat glands in localized areas of the skin. Acetylcho- ated sweating. The area under the curve describing line (ACh) is brought into the skin by means of the sweat output is a parameter of the axon reflex– iontophoresis, and then binds to the muscarinic re- mediated sweat response (Fig. 7). The tests provides ceptors on the eccrine sweat glands. This results in a highly reproducible results and detects postgangli- first, direct sweat response. ACh, however, also binds onic sudomotor dysfunction.50 to nicotinic receptors of the postganglionic sudomo- However, QSART testing can only evaluate the tor axons and induces antidromic C-fiber activation. sweat output at the site where the multicompartmen- The impulse travels along the C-fibers and reaches tal sweat chamber covers the skin. The technique branching points from where it activates sweat does not allow a general statement on postgangli- glands by orthodromic activation of the branching onic sudomotor function in an individual patient. C-fiber nerve terminals. The activation of C-fiber Due to the dying-back pathology of many peripheral branches finally evokes the indirect, axon reflex– neuropathies, QSART testing may be useful to map mediated sweat gland response next to the site of the stocking- or glove-like distribution of postgangli- initial iontophoretic stimulation.50–52,55,56 The sweat onic sudomotor dysfunction. The proximal dorsal response is terminated by the cleavage of ACh into foot, the lateral proximal calf, the distal medial leg, acetate and choline by acetylcholinesterase.50 A mul- and the middle forearm can be used as standard ticompartmental sweat cell is used to measure the recording sites.50 Low reported various types of sweat production. The cell contains a compart- QSART responses.50,51 In painful neuropathies or in ment—usually the inner compartment—that is patients with reflex sympathetic dystrophy, there are filled with 10% acetylcholine. There is also a second persistent QSART responses that might result from compartment—usually the outer compartment— enhanced sympatho-sympathetic reflexes.50,51 The that takes up the humidity from the axon-induced QSART shows abnormal responses in patients with sweat production. The sweat in this compartment is postganglionic sudomotor fiber dysfunction, e.g., in evaporated and carried to a hygrometer by means of various neuropathies (Fig. 8). In preganglionic, cen- a nitrogen gas flow. The hygrometer continuously tral disorders such as multisystem atrophy or spinal monitors the relative humidity. During the test, the cord injuries, QSART responses are usually nor- skin temperature should be kept at approximately mal.50,51

16 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 MICRONEUROGRAPHY Whereas measurements of cardiovascular modula- tion and sudomotor function are indirect means to monitor sympathetic and parasympathetic activity, microneurography is used for direct recordings of bursts of efferent muscle sympathetic nerve activity (MSNA).15,89 However, due to its invasiveness and the time-consuming nature of the procedure, MSNA may be more relevant for research purposes than for routine evaluations of autonomic function. In addi- tion, abnormal findings at one site do not necessarily allow extrapolation to suggest generalized auto- nomic dysfunction. Still, the technique is probably the best and most direct measure of sympathetic activity. Since MSNA reflects vasoconstrictor activity FIGURE 8. Normal quantitative sudomotor axon reflex test to intramuscular vessels, it is closely related to intra- (QSART) in a healthy control (A) and abnormal QSART in a 89 patient with dysfunction of the postganglionic sympathetic sudo- muscular vasoconstriction and vascular resistance. motor axon due to small-fiber neuropathy (B). In contrast to the Moreover, a close correlation exists between plasma patient, sweating increases during mental arithmetic (1), and norepinephrine levels and MSNA.71 MSNA can be acoustic stimulation (2), in the control prior to acetylcholine ion- recorded from peripheral such as the me- tophoresis (3). dian, tibial, or peroneal nerve.89 Most often, the peroneal nerve is examined. A recording electrode is inserted into a single nerve fascicle while the refer- ence electrode is inserted subcutaneously.63,89 Filters Sympathetic Skin Response. Various authors recom- are set at 700 Hz and2kHz, the signal is preampli- mend measurements of the sympathetic skin re- fied and rectified, and the mean voltage or voltage sponse (SSR) to evaluate sudomotor function. How- area is assessed.89 The number of bursts per minute ever, the results of sympathetic skin response testing is multiplied by the mean burst amplitude or mean 80 should be interpreted carefully. voltage area to measure sympathetic nerve activity.63 Any arousal stimulus can be used to activate The MSNA bursts occur with an individually stable sympathetic sudomotor nerve fibers and thereby latency from the electrocardiographic R-wave, indi- change skin resistance.2,4 The central pathways cating that the MSNA bursts are related to an inhib- mediating the arousal response and sudomotor itory effect of systole on the arterial barorecep- activation are not known completely.94 The SSR is tors.77,89 MSNA activity increases after baroreceptor usually recorded from the palms of the hands and denervation and the relation to the cardiac rhythm soles of the feet, with the reference electrodes on disappears.15 MSNA is inhibited during baroreflex the dorsum of the hands and feet. The low-fre- activation.15 In contrast, MSNA increases with un- quency filter should be as low as possible, e.g., at loading of baroreceptors, for example during ortho- 0.1 Hz, and the high-frequency filter should be at static challenge by passive tilting, or during the 30 Hz.48 SSR can be evoked by various types of blood pressure decrease in phase 2 of the Valsalva stimulation, such as electrical, acoustic, or inspira- maneuver.63,89 tory gasp stimuli. As the response habituates, stim- uli must be delivered at randomized intervals and CONCLUSION 80 with an increasing stimulus intensity. The re- Many tests have been used to quantify autonomic sponse is rather variable and there are no clear dysfunction and are described in standard text- 80 criteria regarding abnormal responses. Yokota books. All of the tests show age-dependent results and coworkers consider the SSR abnormal if am- and often significant differences between women plitudes between the left and right side differ by at and men.50 There are concerns that the variability of least 50% or if one of four limb responses is ab- procedures used for autonomic testing is excessive, sent.94 Although the SSR is easy to perform,33,40 but study designs often fail to take into account the the technique is not very sensitive in the early need to standardize procedures to assure reproduc- detection of small-fiber neuropathy and in detect- ibility of the results. The results of a Valsalva maneu- ing autonomic sudomotor dysfunction.80 ver will differ greatly, not only if the level or duration

Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 17 of expiratory strain varies inter- or intra-individually, 12. Borst C, Wieling W, Brederode J, De Rijk L, Hond A, Dunning A. Mechanisms of initial heart rate response to postural but also if the test is carried out in supine, sitting, or change. Am J Physiol 1982;243:H676–H681. standing position.5,37,50,63 Similarly, results vary if one 13. Cohen J, Low P, Fealey R, Sheps S, Jiang N-S. Somatic and test is performed early in the morning and another autonomic function in progressive autonomic failure and 37,63 multiple system atrophy. Ann Neurol 1987;22:692–699. during the late afternoon. Room temperature 14. Consensus Committee of the American Autonomic Society and humidity, and disturbances in background noise and the American Academy of Neurology. Consensus state- will affect the results of autonomic testing.37,63 Spec- ment on the definition of orthostatic hypotension, pure au- tonomic failure, and multiple system atrophy. Neurology tral analysis yields results that depend on the algo- 1996;46:1470. rithm, the length of the analyzed signal epoch, and 15. Delius W, Hagbarth K, Hongell A, Wallin B. General charac- the stationarity of the analyzed biosignals.37,63 Fre- teristics of sympathetic activity in human muscle nerves. Acta Physiol Scand 1972;84:65–81. quently, a compromise has to be found between the 16. Ebert TJ, Cowley AW Jr. Baroreflex modulation of sympa- need for long enough signal epochs and the require- thetic outflow during physiological increases of vasopressin in ment of stationarity. Individual conditions such as humans. Am J Physiol 1992;262:H1372–H1378. 17. Eckberg DL, Cavanaugh MS, Mark AL, Abboud FM. A simpli- the level of stress and exhaustion or relaxation might fied neck suction device for activation of carotid barorecep- also vary from one examination to the next and thus tors. J Lab Clin Med 1975;85:167–173. influence the test results. However, the quality and 18. Eckberg DL, Fritsch JM. How should human baroreflexes be tested? News Physiol Sci 1993;8:7–12. reproducibility of autonomic test results will increase 19. Eckberg DL, Sleight P. Human baroreflexes in health and significantly and fulfill the requirements needed for disease. Monographs of the Physiological Society, Vol. 43. serial examinations if the many different variables New York: Oxford University Press; 1992. 588 p. 20. Ernsting J, Parry DJ. Some observations on the effects of influencing autonomic responses are first identified, stimulating the stretch receptors in the carotid artery of man. taken into consideration, and then kept constant, J Physiol (Lond) 1957;136:45–46. and if testing is highly standardized. 21. Ewing DJ. Which battery of cardiovascular autonomic func- tion tests? Diabetologia 1990;33:180–181. 22. Ewing DJ, Bellavere F, Espi F, McKibben BM, Buchanan KD, Riemersma RA, et al. Correlation of cardiovascular and neu- roendocrine tests of autonomic function in diabetes. Metab- REFERENCES olism 1986;35:349–353. 23. Ewing DJ, Clarke BF. Autonomic neuropathy: its diagnosis 1. Angelone A, Coulter N. Respiratory sinus arrhythmia: a fre- and prognosis. Clin Endocrinol Metab 1986;15:855–888. quency dependent phenomenon. J Appl Physiol 1964;19:479– 24. Ewing DJ, Clarke BF. Diagnosis and management of diabetic 482. autonomic neuropathy. Br Med J 1982;285:916–918. 2. Baba M, Watahiki Y, Matsunaga M, Takebe K. Sympathetic 25. Fealey RD. Thermoregulatory sweat test. In: Low PA, editor. skin response in healthy man. Electromyogr Clin Neuro- Clinical autonomic disorders. Philadelphia: Lippincott– physiol 1988;28:277–283. Raven; 1997. p 245–257. 3. Bannister R, Ardill L, Fentem P. Defective autonomic control 26. Fealey RD, Low PA, Thomas JE. Thermoregulatory sweating of blood vessel in idiopathic hypotension. Brain 1967;90:725– abnormalities in diabetes mellitus. Mayo Clin Proc 1989;64: 746. 617–628. 4. Baser S, Meer J, Polinsky R, Hallett M. Sudomotor function in 27. Finley JP, Bonet JF, Waxman MB. Autonomic pathways re- autonomic failure. Neurology 1991;41:1564–1566. sponsible for bradycardia on facial immersion. J Appl Physiol 5. Benarroch EE. Central autonomic network: functional orga- 1979;47:1218–1222. nization and clinical correlations. Armonk: Futura; 1997. 697 28. Fouad F, Tarazi R, Ferrario C, Fighaly S, Alicandri C. Assess- p. ment of parasympathetic control of heart rate by a noninva- 6. Bennett T, Farquhar I, Hosking D, Hampton J. Assessment of sive method. Am J Physiol 1984;246:H838–H842. methods for estimating autonomic nervous control of the 29. Freeman R, Weiss ST, Roberts M, Zbikowski SM, Sparrow D. heart in patients with diabetes mellitus. Diabetes 1978;27: The relationship between heart rate variability and measures 1167–1174. of body habitus. Clin Auton Res 1995;5:261–266. 7. Bernardi L, Bianchini B, Spadacini G, Leuzzi S, Valle F, 30. Fritsch JM, Eckberg DL, Graves LD, Wallin BG. Arterial pres- Marchesi E, et al. Demonstrable cardiac reinnervation after sure ramps provoke linear increases of heart period in hu- human heart transplantation by carotid baroreflex modula- mans. Am J Physiol 1986;251:R1086–R1090. tion of RR interval. Circulation 1995;92:2895–2903. 31. Genovely H, Pfeifer MA. RR-Variation: the autonomic test of choice in diabetes. Diabetes Metab Rev 1988;4:255–271. 8. Bernardi L, Leuzzi S, Radaelli A, Passino C, Johnston JA, 32. Godden J, Roth G, Hines E. The changes in intraarterial Sleight P. Low-frequency spontaneous fluctuations of R-R pressure during emersion of the hand in ice cold water. interval and blood pressure in conscious humans: a barore- Circulation 1965;12:963–973. ceptor or central phenomenon? Clin Sci 1994;87:649–654. 33. Gutrecht JA. Sympathetic skin response. J Clin Neurophysiol 9. Bernardi L, Passino C, Robergs R, Appenzeller O. Acute and 1994;11:519–524. persistent effects of a 46-kilometre wilderness trail run at 34. Heistad DH, Abboud FM, Eckstein JW. Vasoconstrictor re- altitude: cardiovascular autonomic modulation and barore- sponse to simulated diving in man. J Appl Physiol 1968;25: flexes. Cardiovasc Res 1997;34:273–280. 542–549. 10. Blomqvist CG, Nixon JV, Johnson RL Jr, Mitchell JH. Early 35. Henriksen O. Sympathetic control of arteriole tone in periph- cardiovascular adaptation to zero gravity simulated by head- eral tissue during orthostatic stress. Physiological and clinical down tilt. Acta Astronaut 1980;7:543–553. aspects. Ugeskr Laeger 1985;147:1922–1928. 11. Borst C, Hollander AP, Bouman LM. Cardiac acceleration 36. Henriksen O, Skagen K. Local and central sympathetic vaso- elicited by voluntary muscle contractions of minimal dura- constrictor reflexes in human limbs during orthostatic stress. tion. J Appl Physiol 1972;32:70–77. In: Christensen N, Henriksen O, Lassen N, editors. The sym-

18 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 pathoadrenal system: physiology and pathophysiology. sudden cardiac death. J Cardiovasc Electrophysiol 1994;5: Copenhagen: Minksgaard; 1988. p 83–94. 274–286. 37. Hilz MJ. Quantitative autonomic functional testing in clinical 60. Malliani A, Pagani M, Lombardi F, Cerutti S. Cardiovascular trials. In: Brown WF, Bolton CF, Aminoff MJ, editors. Neuro- neural regulation explored in the frequency domain. Circu- muscular function and disease. Basic, clinical and electrodi- lation 1991;84:482–492. agnostic aspects. Philadelphia: WB Saunders; 2002. p 1899– 61. Mathias CJ. Autonomic diseases: clinical features and labora- 1929. tory evaluation. J Neurol Neurosurg Psychiatry 2003;74:31– 38. Hilz MJ, Stemper B, Sauer P, Haertl U, Singer W, Axelrod FB. 41. Cold face test demonstrates parasympathetic cardiac dysfunc- 62. Mathias CJ. Role of autonomic evaluation in the diagnosis tion in familial dysautonomia. Am J Physiol 1999;276:R1833– and management of . Clin Auton Res 2004;14:45–54. R1839. 63. Mosqueda-Garcia R. Evaluation of autonomic failure. In: Rob- 39. Hines E, Brown G. The cold pressor test for measuring the ertson D, Biaggioni I, editors. Disorders of the autonomic reactibilityof the blood pressure: data concerning 571 normal nervous system. Luxembourg: Harwood Academic Publishers; and hypertensive subjects. Am Heart J 1936;11:1–9. 1995. p 25–60. 40. Hoeldtke R, Davis KM, Hshieh PB, Gaspar SR, Dworkin GE. 64. O’Brien I, O’Hare P, Corrall R. Heart rate variability in Autonomic surface potential analysis: assessment of reproduc- healthy subjects: effect of age and the derivation of normal ibility and sensitivity. Muscle Nerve 1992;15:926–931. ranges for tests of autonomic function. Br Heart J 1986;55: 41. Hojgaard MV, Holstein-Rathlou NH, Agner E, Kanters JK. 348–354. Dynamics of spectral components of heart rate variability 65. Oribe E, Caro S, Perera R, Winters SL, Gomes JA, Kaufmann during changes in autonomic balance. Am J Physiol 1998;275: H. Syncope: the diagnostic value of head-up tilt testing. Pac- H213–H219. ing Clin Electrophysiol 1997;20:874–879. 42. Jamieson G, Ludbrook J, Wilson A. Cold hypersenstivity in 66. Parati G, Di Rienzo M, Bertinieri G, Pomidossi G, Casadei R, Raynaud’s phenomenon. Circulation 1971;44:254–264. Groppelli A, et al. Evaluation of the baroreceptor-heart rate 43. Kamath MV, Fallen EL. Power spectral analysis of heart rate reflex by 24-hour intra-arterial blood pressure monitoring in variability: a noninvasive signature of cardiac autonomic func- humans. Hypertension 1988;12:214–222. tion. Crit Rev Biomed Eng 1993;21:245–311. 67. Persson A, Solders G. R-R variations, a test of autonomic 44. Katona P, Jih F. Respiratory sinus arrhythmia: noninvasive dysfunction. Acta Neurol Scand 1983;67:285–293. measure of parasympathetic cardiac control. J Appl Physiol 68. Pfeifer M, Cook D, Brodsky J, Tice D, Reenan A, Swedine S, et 1975;39:801–805. al. Quantitative evaluation of cardiac parasympathetic activity 45. Khurana RK, Nicholas EM. Head-up tilt table test: how far and in normal and diabetic man. Diabetes 1982;31:339–345. how long? Clin Auton Res 1996;6:335–341. 69. Pfeifer MA, Jung S, Crain G, Schumer M. Autonomic neurop- 46. Khurana RK, Watabiki S, Hebel JR, Toro R, Nelson E. Cold athy. Diabetes Med 1993;10:70S–73S. face test in the assessment of trigeminal-brainstem-vagal func- 70. Pitzalis MV, Mastropasqua F, Passantino A, Massari F, Ligurgo tion in humans. Ann Neurol 1980;7:144–149. L, Forleo C, et al. Comparison between noninvasive indices of 47. Kronenberg R, Hamilton F, Gabel R, Hickey R, Read D, baroreceptor sensitivity and the phenylephrine method in Severinghaus J. Comparison of three methods for quantitat- post-myocardial infarction patients. Circulation ing respiratory response to hypoxia in man. Respir Physiol 1998;97:1362–1367. 1972;16:109–125. 48. Levy DM, Reid G, Rowley DA. Quantitative measures of sym- 71. Rea R, Eckberg D, Fritsch J, Goldstein D. Relation of plasma pathetic skin response in diabetes: relation to sudomotor and norepinephrine and sympathetic traffic during hypotension neurological function. J Neurol Neurosurg Psychiatry 1992; in humans. Am J Physiol 1990;58:R982–R986. 10:902–908. 72. Richter DW, Ballanyi K, Schwarzacher S. Mechanisms of re- 49. Levy MN, Degeest H, Zieske H. Effects of respiratory center spiratory rhythm generation. Curr Opin Neurobiol 1992;2: activity on the heart. Circ Res 1966;18:67–78. 788–793. 50. Low PA. Testing the autonomic nervous system. Semin Neu- 73. Robbe HW, Mulder LJ, Ruddel H, Langewitz WA, Veldman rol 2003;23:407–421. JB, Mulder G. Assessment of baroreceptor reflex sensitivity by 51. Low PA. Evaluation of sudomotor function. Clin Neuro- means of spectral analysis. Hypertension 1987;10:538–543. physiol 2004;115:1506–1513. 74. Robertson D. Clinical pharmacology: assessment of auto- 52. Low PA, Caskey PE, Tuck RR, Fealey RD, Dyck PJ. Quantita- nomic function. In: Baughman K, Greene B, editors. Clinical tive sudomotor axon reflex test in normal and neuropathic diagnostic manual for the house officer. Baltimore: Williams subjects. Ann Neurol 1983;14:573–580. & Wilkins; 1981. p 86–101. 53. Low PA, Denq JC, Opfer-Gehrking TL, Dyck PJ, O’Brien PC, 75. Rowell LB. Reflex control of regional circulations in humans. Slezak JM. Effect of age and gender on sudomotor and car- J Auton Nerv Syst 1984;11:101–114. diovagal function and blood pressure response to tilt in nor- 76. Rudas L, Crossman AA, Morillo CA, Halliwill JR, Tahvanainen mal subjects. Muscle Nerve 1997;20:1561–1568. KU, Kuusela TA, et al. Human sympathetic and vagal barore- 54. Low PA, Neumann C, Dyck PJ, Fealey RD, Tuck RR. Evalua- flex responses to sequential nitroprusside and phenylephrine. tion of skin vasomotor reflexes by using laser Doppler veloci- Am J Physiol 1999;276:H1691–H1698. metry. Mayo Clin Proc 1983;58:583–592. 77. Sanders J, Ferguson D, Mark A. Arterial baroreflex control of 55. Low PA, Opfer-Gehrking TL, Kihara M. In vivo studies on sympathetic nerve activity during elevation of blood pressure receptor pharmacology of the human eccrine sweat gland. in normal man: dominance of aortic baroreceptors. Circula- Clin Auton Res 1992;2:29–34. tion 1988;77:279–288. 56. Low PA, Opfer-Gehrking TL, Proper CJ, Zimmerman I. The 78. Saul J, Cohen R. Respiratory sinus arrhythmia. In: Levy M, effect of aging on cardiac autonomic and postganglionic Schwartz P, editors. Vagal control of the heart: experimental sudomotor function. Muscle Nerve 1990;13:152–157. basis and clinical implications. Armonk, NY: Futura; 1994. p 57. Mackay JD, Page MM, Cambridge J, Watkins PJ. Diabetic 511–536. autonomic neuropathy. The diagnostic value of heart rate 79. Saul JP, Berger RD, Albrecht P, Stein SP, Chen MH, Cohen monitoring. Diabetologia 1980;18:471–478. RJ. Transfer function analysis of the circulation: unique in- 58. Malik M, Camm AJ. Components of heart rate variability— sights into cardiovascular regulation. Am J Physiol 1991;261: what they really mean and what we really measure. Am J H1231–H1245. Cardiol 1993;72:821–822. 80. Schondorf R. Skin potentials: normal and abnormal. In: Low 59. Malliani A, Lombardi F, Pagani M, Cerutti S. Power spectral PA, editor. Clinical autonomic disorders. Philadelphia: Lip- analysis of cardiovascular variability in patients at risk for pincott–Raven; 1997. p 221–231.

Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006 19 81. Smith JJ, Porth CM, Erickson M. Hemodynamic response to 89. Wallin BG, Fagius J. Peripheral sympathetic neural activity in the upright posture. J Clin Pharmacol 1994;34:375–386. conscious humans. Annu Rev Physiol 1988;50:565–576. 82. Smith S. Reduced sinus arrhythmia in diabetic autonomic 90. Wheeler T, Watkins PJ. Cardiac denervation in diabetes. Br neuropathy: diagnostic value of an age-related normal range. Med J 1973;4:584–586. Br Med J 1982;285:1599–1601. 91. Wieling W, van Lieshout JJ. Investigation and treatment of 83. Smyth HS, Sleight P, Pickering GW. Reflex regulation of autonomic circulatory failure. Curr Opin Neurol Neurosurg arterial pressure during sleep in man. A quantitative method 1993;6:537–543. of assessing baroreflex sensitivity. Circ Res 1969;24:109–121. 92. Wieling W, Smit AA, de Jong-de Vos van Steenwijk CC, van 84. Stein PK, Bosner MS, Kleiger RE, Conger BM. Heart rate Lieshout JJ, Karemaker JM. Pathophysiological mechanisms variability: a measure of cardiac autonomic tone. Am Heart J underlying vasovagal syncope in young subjects. Pacing Clin 1994;127:1376–1381. Electrophysiol 1997;20:2034–2038. 85. Stewart JD, Low PA, Fealey RD. Distal small fiber neuropathy: 93. Yamamoto K, Iwase S, Mano T. Responses of muscle sympa- results of tests of sweating and autonomic cardiovascular re- thetic nerve activity and cardiac output to the cold pressor flexes. Muscle Nerve 1992;15:661–665. test. Jpn J Physiol 1992;42:239–252. 86. Tanaka H. Cardiac output and blood pressure during active 94. Yokota T, Matsunaga T, Okiyama R, Hirose K, Tanabe H, and passive standing. Clin Physiol 1996;16:157–170. Furukawa T, et al. Sympathetic skin response in patients with 87. Task Force of the European Society of Cardiology and the multiple sclerosis compared with patients with spinal cord North American Society of Pacing and Electrophysiology. transsection and normal controls. Brain 1991;114:1381–1394. Heart rate variability: standards of measurement, physiologi- 95. Ziegler D, Dannehl K, Muhlen H, Spuler M, Gries FA. Prev- cal interpretation, and clinical use. Circulation 1996;93:1043– alence of cardiovascular autonomic dysfunction assessed by 1065. spectral analysis, vector analysis, and standard tests of heart 88. Walker AJ, Longland CJ. Venous pressure measurement in rate variation and blood pressure responses at various stages foot as aid to investigate disease in leg. Clin Sci 1950;9:101. of diabetic neuropathy. Diabetes Med 1992;9:806–814.

20 Quantitative Studies of Autonomic Function MUSCLE & NERVE January 2006