Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

Contents lists available at ScienceDirect

Seminars in Arthritis and Rheumatism journal homepage: www.elsevier.com/locate/semarthrit

Dysautonomia and its underlying mechanisms in the hypermobility type of Ehlers–Danlos syndrome Inge De Wandele, MSc, PTa,n, Lies Rombaut, PhD, PTa, Luc Leybaert, PhD, MDb, Philippe Van de Borne, PhD, MDc, Tine De Backer, PhD, MDd, Fransiska Malfait, PhD, MDe, Anne De Paepe, PhD, MDe, Patrick Calders, PhDa a

Department of Rehabilitation Sciences and Physiotherapy, Ghent University, Artevelde University College, De Pintelaan 185, 1B3, 9000 Ghent, Belgium Department of Basic Medical Sciences, Physiology Group, Ghent University, Ghent, Belgium Hypertension Clinic, Erasme Hospital, Brussels, Belgium d Heymans Institute of Pharmacology, Ghent University, Ghent, Belgium e Centre for Medical Genetics, Ghent University Hospital, Ghent, Belgium b c

a r t i c l e in fo

Keywords: Ehlers–Danlos syndrome Hypermobility Autonomic dysfunction Orthostatic intolerance

a b s t r a c t Objectives: Many non-musculoskeletal complaints in EDS-HT may be related to dysautonomia. This study therefore aims to investigate whether dysautonomia is present and to explore the underlying mechanisms. Methods: A total of 39 females with EDS-HT and 35 age-matched controls underwent autonomic function testing. Resting autonomic tone was assessed using heart rate variability (frequency domain) and baroreflex sensitivity analysis (cross correlation). Autonomic reactivity was assessed using the Autonomic Reflex Screen test battery. Factors suspected to contribute to dysautonomia, e.g., neuropathy, medication use, decreased physical activity, depression, pain-induced sympathetic arousal, and connective tissue laxity, were quantified using validated questionnaires, the Beighton score, and measurement of skin extensibility. Results: The EDS-HT group showed autonomic deregulation with increased sympathetic activity at rest and reduced sympathetic reactivity to stimuli. Increased resting activity was indicated by a higher LF/HF ratio compared to controls (1.7 7 1.23 vs 0.9 7 0.75, p ¼ 0.002); decreased reactivity by a greater BP fall during valsalva (  19 7 12 vs  8 7 10, p o 0.001), and a smaller initial diastolic BP increase during tilt (7% vs 14%, p ¼ 0.032). Orthostatic intolerance was significantly more prevalent in EDS-HT than controls (74% vs 34%) and was most frequently expressed as postural orthostatic tachycardia. Lowered QSART responses suggest that sympathetic neurogenic dysfunction is common in patients (p o 0.013), which may explain the dysautonomia in EDS-HT. Further, connective tissue laxity and vasoactive medication use were identified as important factors in aggravating dysautonomia (p o 0.035). Conclusion: Dysautonomia consisting of cardiovascular and sudomotor dysfunction is present in EDS-HT. Neuropathy, connective tissue laxity, and vasoactive medication probably play a role in its development. & 2014 Elsevier Inc. All rights reserved.

Introduction Abbreviations: ARS, Autonomic Reflex Screen test battery; BP, blood pressure; BRS, baroreflex sensitivity; CASS, composite autonomic severity score; DIA, diastolic blood pressure; EDS, Ehlers–Danlos syndrome; EDS-HT, Ehlers–Danlos syndrome, hypermobility type; FFT, fast Fourier transform (HRV—frequency analysis technique); FM, fibromyalgia; HADS, Hospital Anxiety and Depression score; HR, heart rate; HRDB, mean heart rate range during deep breathing; HF power, high-frequency power; HRV, heart rate variability; JHS, joint hypermobility syndrome; LF power, low-frequency power; LF/HF ratio, ratio of the low-frequency power to the high-frequency power; n.u., normalized units; OI, orthostatic intolerance; OH, orthostatic hypotension; POTS, postural orthostatic tachycardia syndrome; PP, pulse pressure; PRT, pressure recovery time; QSART, quantitative sudomotor axon reflex testing; OGS, Orthostatic Grading scale; VR, valsalva ratio; VV, vasovagal syncope. n Corresponding author. E-mail address: [email protected] (I. De Wandele). 0049-0172/$ - see front matter & 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.semarthrit.2013.12.006

The Ehlers–Danlos syndrome (EDS) is a heritable connective tissue disorder characterized by generalized joint hypermobility, skin hyperextensibility, and tissue fragility. Patients are classified into 6 subtypes based on the Villefranche Nosology [1]. The hypermobility type (EDS-HT) is considered one of the largest subgroups, with prevalence of 1 in 5000–20,000 births [2]. As a consequence of hypermobility, patients experience recurrent joint dislocations, musculotendinous lesions, and chronic pain [3]. Besides these overt musculoskeletal symptoms, patients with EDS-HT also suffer from a large variety of non-musculoskeletal complaints, among which are palpitations, syncope, diarrhea, constipation, and thermoregulatory complaints [4]. Although

2

I. De Wandele et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

EDS-HT is often perceived as a locomotor disorder, a recent study has shown that the non-musculoskeletal symptoms sometimes dominate the symptom profile and are related to significant impairment in daily life [5]. In fibromyalgia (FM), chronic fatigue syndrome (CFS), and joint hypermobility syndrome (JHS), which all show a large clinical overlap with EDS-HT, many of the nonmusculoskeletal complaints have been attributed to an underlying dysautonomia [6–9]. In EDS-HT, autonomic function has not yet been investigated, although dysautonomia is suspected on a clinical basis and would explain many of the non-musculoskeletal complaints in this population as well. Detection is of importance, because autonomic dysfunction has been related to a poor disease prognosis, decreased quality of life, and increased cardiac morbidity and mortality in other pathologies [10–14]. Individuals with autonomic dysfunction often have an unstable blood pressure, increasing the risk of cardiac complications under general anesthesia [15,16]. If dysautonomia is present in EDS-HT, this risk should be considered, as these patients often undergo surgery in order to treat musculoskeletal and gastrointestinal problems [17]. Although dysautonomia has been demonstrated in JHS and is strongly suspected in EDS-HT, the exact cause is unclear. Several authors have suggested peripheral neuropathy, connective tissue abnormalities, and deconditioning as the possible pathomechanisms in hypermobile patients on a theoretical basis [6,18,19], but evidence supporting these hypotheses is lacking [19]. Identification of the underlying and aggravating factors, however, is necessary to provide an adequate treatment. Therefore, this study aims to evaluate whether dysautonomia is present in EDS-HT, to determine its severity and distribution, and to explore possible underlying mechanisms. Consequently, testing of autonomic function was performed, consisting of cardiac testing (vagal and adrenergic function) and peripheral testing (sudomotor and adrenergic function). In addition, we explored whether neuropathy, medication use, physical activity, depression, pain-induced sympathetic arousal, and parameters of connective tissue laxity were related to autonomic dysfunction.

Methods Subjects A total of 39 female patients with EDS-HT (age 38.7 7 10.14 years) and 35 age- and sex-matched healthy controls (age 39.5 7 9.43 years) participated. As more than 90% of the EDS-HT patients are female [1], this study included only women. Patients were diagnosed at the Centre for Medical Genetics in Ghent University Hospital using the revised Villefranche Nosology [1]. Exclusion criteria were pregnancy in the current or past year, acute infection, structural heart disease or a disease known to cause dysautonomia (e.g., Parkinson and diabetes), and a history of alcohol or drug abuse. For the control group, 2 exclusion criteria were added: a Beighton score exceeding 3/9 and medication use that might affect the autonomic nervous system (pain medication other than paracetamol, anti-inflammatory drugs, antidepressants, sedatives, nausea treatment, alpha- or beta-blocking agents, diuretics, nitrates, and muscle relaxants). Procedure Participants refrained from heavy meals, coffee, and nicotine for 3 h and from vigorous physical activity for 24 h before testing. If medically permissible, analgesics (e.g., tramadol, morphine, and oxycodone) and medication with (anti)cholinergic activity (e.g., tricyclic antidepressants such as amitriptyline and melitracen) or

(anti)adrenergic activity (e.g., beta-blocking agents, such as bisoprolol; ACE inhibitors, such as lisinopril; and angiotensin II antagonists, such as losartan) were withheld 24 h before testing. Prior to recording the measurements, all subjects underwent an electrocardiogram and were questioned regarding medical history, medication use, and exclusion criteria. The measurements were recorded in a quiet room with constant ambient temperature (21–231C). Heart rate (HR) and beat-to-beat blood pressure (BP) were monitored using a 3-lead ECG (Holter Schiller MT-200) and the Finometers PRO (Finapres medical systems, the Netherlands), respectively. Additional manual and automatic blood pressure measurements (Microlife WatchBP) were recorded to ensure safety and correctness of data. The study was approved by the local Ethics committee of the Ghent University Hospital. Written informed consent was obtained from all participants. Resting autonomic activity Participants were asked to not speak or move during a 15-min supine resting period. Breathing rate was similar between patients and controls (Table 1). Resting heart rate variability analysis Resting HRV provides quantitative information regarding cardiac autonomic tone. The ECG signal quality of the middle 5 min was checked manually and with the Schiller (MT-200 analysis) and Kubios software packages (University of Eastern Finland, Kuopio). If necessary, another appropriate section of 5 min was selected. Detrending was done using the smoothing priors method (λ ¼ 500). Fast Fourier transformation and autoregressive modeling (model order 16) were performed [20,21]. Variability within the HF band (0.15–0.4 Hz) is thought to reflect parasympathetic Table 1 Resting parameters and resting autonomic activity Control

EDS-HT

p

Resting parameters HR (bpm) SYS BP (mmHg) DIA BP (mmHg) Breathing rate (Hz)

62.8 113.3 66.7 0.24

7 7 7 7

8.45 12.27 9.1 0.052

Resting HRV analysis (FFT) LF (ms2) HF (ms2) LF (n.u.) HF (n.u.) LF/HF ratio

242.5 656.5 43.4 56.6 0.9

7 7 7 7 7

341.81 270.8 7 466.94 595.04 502 7 767.81 14.93 57.2 7 15.84 14.93 42.8 7 15.84 0.75 1.7 7 1.23

Resting BRS analysis xBRS (bpm/mmHg) Mean Tau (s) R2 Orthostatic grading scale Frequency Intensity Standing time Other conditions (heath and exercise) Daily life interference Total score

13.4 7 6.03 1.4 7 0.68 0.7 7 0.04

68.4 126.6 75.8 0.23

7 7 7 7

9.58 15.17 10.96 0.070

11.3 7 7.00 1.3 7 0.59 0.7 7 0.04

0.015a,b 0.001a,b 0.001a,c 0.167c 0.364b 0.086c 0.001a,b 0.001a,b 0.002a,c 0.133c 0.840b 0.570b

1.14 1.00 1.12 1.12

o0.001a,c o0.001a,c o0.001a,c o0.001a,c

0.0 7 0.19

1.8 7 0.93

o0.001a,c

1.6 7 1.44

9.0 7 3.63

o0.001a,c

0.6 0.8 0.1 0.1

7 7 7 7

0.69 0.51 0.56 0.23

1.9 2.2 1.6 1.6

7 7 7 7

HR ¼ heart rate; SYS BP ¼ systolic blood pressure; DIA BP ¼ diastolic blood pressure; HRV ¼ heart rate variability; FFT ¼ fast Fourier transform; LF ¼ lowfrequency power; HF ¼ high-frequency power; n.u. ¼ normalized units; xBRS ¼ baroreflex sensitivity; mean Tau ¼ mean baroreflex delay; R2 ¼ regression coefficient of the cross correlation method. a b c

Significantly different between groups (α ¼ 0.05). Results obtained using the Students0 t test. Results obtained using the Mann–Whitney U test.

I. De Wandele et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

activity and is mainly caused by respiration [21]. The LF/HF ratio was calculated as a measure of sympathovagal balance. An increased LF/HF ratio is thought to portray sympathetic hyperactivity [21]. Resting baroreflex sensitivity The baroreflex is responsible for maintaining a stable blood pressure [22]. Resting BRS was calculated by the Beatscope software (Finapres medical systems, the Netherlands) based on the systolic BP and interbeat intervals, according to the method by Westerhof et al. [23]. Autonomic reactivity The Autonomic Reflex Screen (ARS) test battery evaluates the peripheral sudomotor and adrenergic function, as well as the cardiac vagal and adrenergic function [24]. The results of the 4 tests are combined into a “composite autonomic severity score” (CASS) ranging from 0 (no autonomic deficit) to 10 (maximal deficit), which corrects for the effects of age and sex and reflects the severity of dysautonomia. The CASS can be split into an adrenergic, vagal, and sudomotor CASS. Quantitative sudomotor axon reflex testing QSART indirectly evaluates the function of the peripheral sympathetic nerve, more specifically the postganglionic cholinergic nerves in the skin. The sweat capsules were attached to the volar wrist, below the fibular head, above the medial malleolus and the dorsum of the foot. A 10% acetylcholine solution was iontophoresed using a constant current of 2 mA for 5 min, evoking a sweat response. The produced sweat volume was measured. Sympathetic nerve dysfunction is characterized by lowered sweat volumes or by a persistent “hung-up” response [24]. Deep breathing The participant breathed maximally at a frequency of 6 breaths per minute, following the lead of an oscillating ball on a computer screen. A total of 8 breathing cycles were recorded and the test was repeated after 2 min of rest. The HR range (HRDB) was calculated as a measure of parasympathetic reactivity [24]. Results were averaged over 2 trials. Valsalva maneuver The participant blew into a mouthpiece between 40 and 50 mmHg for 15 sec [24]. The maneuver was repeated 3 times, or until at least 2 reproducible blood pressure recordings were obtained. Faulty trials (inadequate pressure or duration) were excluded. In between trials 3 min of rest was provided to stabilize HR and BP. The best-performed maneuver was selected for evaluation. The valsalva ratio (VR) was calculated (parasympathetic measure), and the 4 phases in the blood pressure response were quantified (sympathetic reactivity) [24]. In addition, the adrenergic, vagal and globalbaroreflex sensitivity parameterswere calculated using the method by Schrezenmaier et al. [25]. Head-up tilt The participant rested quietly for 5 min. Baseline HR and BP were calculated as the mean from 40 sec to 10 sec before tilting. Next, the table was slowly tilted upright to an angle of 701 for a maximum of 20 min. The patient was asked to report all symptoms and the test was aborted when either orthostatic symptoms or pain became intolerable. Orthostatic hypotension (OH) was defined as a sustained diastolic BP drop of at least 10 mmHg or a systolic BP drop of at least 20 mmHg for normotensive subjects, 30 mmHg for hypertensive subjects, and 15 mmHg for subjects

3

with a resting systolic blood pressure below 120 mmHg [26,27]. Vasovagal syncope (VV) was defined as a sudden BP and HR decline, leading to a loss of consciousness. Postural orthostatic tachycardia (POTS) was defined as a sustained HR rise of at least 30 bpm or a HR of at least 120 bpm in the first 10 min of tilt, without concomitant orthostatic hypotension [28]. Orthostatic hypertension was defined as a systolic blood pressure increase of at least 20 mmHg upon standing [29]. Orthostatic symptoms The Orthostatic Grading Scale (OGS) is a 5-item self-report questionnaire designed to evaluate the frequency and severity of orthostatic symptoms, identify orthostatic stressors, and assess the impact of orthostatic symptoms on activities of daily living and standing time [30]. Respondents rate each item on a scale of 0–4. Adding the scores for the individual items creates a total score, with higher scores reflecting worse orthostatic intolerance. Possible contributing factors to dysautonomia Because the laxity of connective tissue in vascular structures could not be measured directly, we opted to use 2 measures of general connective tissue laxity, namely the current Beighton score and skin extensibility as a substitute [31,32]. Next, information regarding medication use in the past 4 weeks was assembled using a self-report form inquiring about the brand name, dose, frequency, and time of intake of each medicine. Patients were asked to describe the reason for and duration of their medication use, which were later verified during the anamnesis. The Baecke questionnaire was used to assess habitual physical activity [33]. The responses were scored on a 5-point Likert scale (ranging from “never” to “very often”). Anxiety and depression were questioned with the Hospital Anxiety and Depression Scale (HADS). Both subscales contain 7 items that are scored on a Likert scale from 0 to 3 [34]. Further, 2 parts of the painDETECT questionnaire (PDQ) were used [35]. Patients were asked for their current pain intensity and for the strongest and average pain experienced in the past 4 weeks. Next, they filled out the questions regarding the presence of sensory neuropathic symptoms (second part of the PDQ), such as paresthesias, hypoesthesia (numbness), radiating pain, and burning pain. Statistical analysis Data analysis was carried out using the Statistical Package for the Social Sciences 21.0 (SPSS Inc., Chicago, IL, USA). Normality of data was evaluated using the Shapiro–Wilk test and visual assessment of the histograms and QQ plots. Normally, distributed data were compared between groups using independent sample t-tests. If the normality assumption was not fulfilled, the non-parametric Mann–Whitney U test was used. Medication use was compared between groups using the Chi square test. Next, correlation analyses were performed within the EDS-HT group in order to identify factors that possibly contribute to dysautonomia. As such, we evaluated whether autonomic test outcomes were related to vasoactive medication use, deconditioning, depression, and measures of connective tissue laxity because the current literature suggests that these parameters may play a role in the development of dysautonomia. Pearson or Spearman correlation coefficients were used depending on the distribution of the inspected variables. Finally, in order to determine which factors most strongly predict the severity of dysautonomia, a stepwise multiple linear regression analysis was performed in the EDS-HT group (backward selection). The total sudomotor and adrenergic CASS scores were

I. De Wandele et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

QSART Evoked sweat volumes were lower at all testing sites in EDS-HT compared to controls, suggesting peripheral sympathetic nerve dysfunction (Table 2). Only 35% of patients had normal QSART responses. Deep breathing The mean heart rate range during deep breathing (HRDB) did not significantly differ between patients and controls (Table 2). Valsalva The VR was significantly higher in EDS-HT due to a markedly higher HR in phase 2 (Table 2). This demonstrates an adequate parasympathetic activation in patients and controls. By contrast, the BP response was impaired in EDS-HT. Compared to controls, the mean arterial BP showed a significantly larger drop in early phase 2 and less recuperation in late phase 2 (Fig. 1), suggesting insufficient sympathetically induced vasoconstriction. Tilt At baseline, patients showed a significantly higher diastolic BP and mean arterial BP and HR compared to controls (Table 2). In order to correct for this difference in baseline, Figure 2 shows the change in systolic BP, diastolic BP, and HR due to tilt as a percentage of the baseline value. During the tilt, patients initially showed a significantly smaller rise in diastolic and mean arterial BP compared to controls (DIA: p ¼ 0.032 and MAP: p ¼ 0.049 at 30 sec upright). After being tilted for 1 min, these blood pressure differences were no longer statistically significant. From this moment on, however, a significantly larger heart rate rise could be noted in patients, which lasted during the rest of the tilt test (p o 0.04).

1.8 1.1 3.0 3.4

Deep breathing HRDB (bpm) Valsalva VR Highest HR during valsalva (bpm) BRSa (mmHg/sec) BRSv (sec/mmHg) BRSg PRT (sec) PP drop (mmHg)

7 7 7 7

1.84 1.00 2.41 2.72

0.7 0.6 1.1 1.3

7 7 7 7

1.00 0.82 1.40 1.62

17.7 7 7.69

18.6 7 7.43

1.9 7 0.33 98.6 7 11.53

2.2 7 0.52 108.8 7 17.58

27.3 6.8 164.6 2.6 26.3

Tilt Baseline SYS (mmHg) Baseline DIA (mmHg) Baseline MAP (mmHg) Baseline HR (bpm) Standing time First manifestation of OI Orthostatic intolerance Vasovagal syncope Orthostatic hypotension POTS Orthostatic hypertension Early stop of tilt test Due to vasovagal syncope Due to orthostatic hypotension Due to POTS Due to pain

7 7 7 7 7

0.001a,b 0.013a,b o0.001a,b 0.001a,b 0.387b 0.007a,c 0.007a,b

23.40 41.9 7 33.49 0.034a,b 2.92 5.4 7 3.08 0.057b 124.96 218.7 7 229.80 0.586b 4.19 3.2 7 5.49 0.340b 11.43 35.8 7 9.51 o0.001a,b

126.2 7 13.60 71.8 7 11.19 92.5 7 12.07 63.6 7 7.73 19 min 32 sec 8 min 22 sec 12 (34.3%) 3 (8.6%) 6 (17.1%) 4 (11.4%) 0 (0.0%) 5 (14.3%) 3 (8.6%) 2 (5.7%)

132.9 7 15.97 77.1 7 9.33 99.5 7 11.60 70.7 7 10.08 15 min 39 sec 3 min 38 sec 29 (74.4%) 5 (12.8%) 10 (25.6%) 16 (41.0%) 2 (5.1%) 17 (43.6%) 5 (12.8%) 6 (15.4%)

0 (0.0%) 0 (0.0%)

3 (7.7%) 3 (7.7%)

0.065c 0.026a,b 0.021a,b 0.002a,c 0.001a,c 0.017a,c 0.001a,d 0.712 0.379 0.003a,d 0.494 0.010a,d

QSART ¼ quantitative sudomotor axon reflex test; HRDB ¼ mean heart rate range during deep breathing; VR ¼ valsalva ratio; BRSa ¼ adrenergic baroreflex sensitivity; BRSv ¼ vagal baroreflex sensitivity; BRSg ¼ global baroreflex sensitivity; PRT ¼ pressure recovery time in phase 3 of the valsalva maneuver; PP drop ¼ pulse pressure fall from baseline during valsalva; POTS ¼ Postural Orthostatic Tachycardia Syndrome. a

Significantly different between groups (α ¼ 0.05). Results obtained using the Students0 t test. Results obtained using the Mann–Whitney U test. d Results obtained using the Chi square test. b c

orthostatic symptoms. In both groups, vasovagal syncope and orthostatic hypotension were the main reason for an early stop (Table 3). Remarkably, in 3 patients POTS also caused symptoms severe enough to prematurely end the test, although systemic blood pressure remained at an adequate level.

25.0 20.0 15.0 10.0 5.0 0.0 -5.0 -10.0

Orthostatic intolerance Orthostatic intolerance (OI) was present in the majority of patients (74%), with POTS as the most typical form (Table 3). If patients showed objective signs of OI, this usually manifested earlier during the tilt than controls. In addition, more patients than controls prematurely ended the tilt test due to the severity of

wrist (μl) foot (μl) fibula (μl) malleolus (μl)

p

-15.0

Phase 4

Autonomic reactivity

volume volume volume volume

Phase 3

The resting systolic and diastolic BP was significantly higher in the patient group. Further, the higher HR in patients compared to controls, in combination with a higher LF/HF ratio and a higher normalized LF power, strongly suggests that resting cardiac sympathetic activity is increased in patients. Concerning resting cardiac parasympathetic activity, the results are inconclusive. Patients had a significantly lower normalized HF power, but the absolute HF power, which is a better marker of parasympathetic activity, is not significantly different between groups. Also, resting baroreflex sensitivity (xBRS), mainly a cardiac parasympathetic marker, was not significantly different between groups.

QSART Sweat Sweat Sweat Sweat

EDS-HT

Phase 2L

Resting autonomic activity (Table 1)

Control

Phase 2E

Results

Table 2 Autonomic reactivity

Phase 1

used as independent variables. The vasoactive medication use, Baecke score, HADS depression score, Beighton score, skin extensibility, strongest pain experienced during the past 4 weeks, and age were used as the dependent variables because these variables constitute the theoretically plausible factors that also significantly correlated with the autonomic test outcomes.

Baseline

4

-20.0 -25.0 Control

EDS-HT

Fig. 1. Mean arterial pressure response to the valsalva maneuver.

% of paents upright

I. De Wandele et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

120% 100% 100%

100%

100% 80%

100%

100% 100%

100%

92%

100%

100%

97%

5

89%

90% 77%

60%

69% 56%

40% 20% 0% 60%

HR change (relave to baseline)

50% 40% 30% 20% 10% 0% -10%

SYS BP change (relave to baseline)

10% 5% 0% -5%

Controls

3 min supine

1 min supine

20 min upright

15 min upright

10 min upright

5 min upright

3 min upright

1 min upright

30 sec upright

16% 14% 12% 10% 8% 6% 4% 2% 0% Baseline

DIA BP change (relave to baseline)

-10%

EDS-HT

Fig. 2. Response to tilt.

CASS score in EDS-HT Analysis of the CASS score demonstrated that 47% of patients had mild dysautonomia, 33% had moderate dysautonomia, and 3.3% had severe dysautonomia. Autonomic dysfunction consisted of sympathetic dysfunction (mean adrenergic CASS: 1.9 7 0.94) and sudomotor dysfunction (mean sudomotor CASS: 1.5 7 1.29), with a normal parasympathetic reactivity (mean vagal CASS 0.2 7 0.43; mean t score below 1). Orthostatic symptoms Patients had significantly higher scores on the OGS compared to controls, both for frequency and intensity of orthostatic complaints (Table 1).

Possible contributing factors to dysautonomia Parameters of connective tissue laxity showed significant correlations with autonomic outcomes. Higher Beighton scores were generally related to lower blood pressure. As such, the Beighton score was inversely correlated with the systolic BP at rest (r ¼  0.402, p ¼ 0.034) and with the pulse pressure (r ¼  0.828, p ¼ 0.003), systolic BP (r ¼  0.880, p ¼ 0.001), diastolic BP (r ¼  0.825, p ¼ 0.003), and mean arterial BP at the end of the tilt (r ¼  0.755, p ¼ 0.012). Besides this association with lower BP, the Beighton score was also related to a larger HR increase during the tilt (at 1 min: r ¼ 0.484, p ¼ 0.019; at 5 min: r ¼ 0.513, p ¼ 0.021) and a lower TPR at the end of the tilt (r ¼  0.413, p ¼ 0.049). Furthermore, having a more extensible skin was associated

I. De Wandele et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

6

Table 3 Possible factors contributing to dysautonomia Control

EDS-HT

p

Connective tissue laxity Beighton Skin extensibility

1.2 7 1.31 5.8 7 2.08 o 0.001a,b 0.9 7 0.45 1.4 7 0.66 0.033a,b

Vasoactive medication use Opiates Tricyclic antidepressants Antihypertensives Beta-blocking agents Ace inhibitors Diuretics Angiotensin II antagonists

0 0 0 0 0 0 0 0

Habitual physical activity Baecke total score

5.8 7 0.95 4.5 7 1.15

Hospital Anxiety and Depression Scale Anxiety Depression

4.6 7 2.73 7.8 7 3.83 o 0.001a,d 1.1 7 1.34 6.1 7 3.81 o 0.001a,b

(0%) (0%) (0%) (0%) (0%) (0%) (0%) (0%)

22 (56%) 15 (38%) 5 (13%) 12 (31%) 7 (18%) 3 (8%) 1 (3%) 3 (8%)

PainDETECT questionnaire Current pain intensity 0.4 7 1.10 5.3 7 2.31 Intensity of strongest pain in 4 weeks 2.3 7 2.76 8.4 7 1.12 Sensory neuropathic symptoms 4 (11.4%) 30 (76.9%)

o 0.001a,c o 0.001a,c 0.055c o 0.001a,c 0.012c 0.24c 1c 0.24c o 0.001a,d

o 0.001a,b o 0.001a,b o 0.001a,c

a

Significantly different between groups (α ¼ 0.05). Results obtained using the Mann–Whitney U test. c Results obtained using the Chi square test. d Results obtained using the Students0 t test. b

with a higher total CASS score (r ¼ 0.522, p ¼ 0.009) and a larger drop in TPR during the tilt (at 3 min upright: r ¼  0.525, p ¼ 0.030; at the end of the tilt: r ¼  0.603, p ¼ 0.006). Of 39 patients, 22 (56%) used vasoactive medication, mostly blood pressure-lowering agents, antidepressants, and opiates (Table 3). Overall, 8 patients used only 1 vasoactive medicine, 6 patients used 2, 5 patients used 3, and 3 patients used 4 vasoactive medicines. Using more vasoactive medicines was related to more severely impaired autonomic test results; for instance, a larger BP drop during valsalva (r ¼  0.396, p ¼ 0.021), a lower TPR during the first part of the tilt (r between  0.433 and  0.488 from the start of the tilt until 3 min upright; p between 0.018 and 0.030), and a larger HR decrease at the end of the tilt (r ¼  0.384, p ¼ 0.027). The HADS depression score was associated with a lower TPR during the tilt and during supine recuperation after the tilt (at 3 min upright: r ¼  0.488, p ¼ 0.034; at 15 min upright: r ¼  0.663, p ¼ 0.014; at 1 min supine: r ¼  0.534, p ¼ 0.033; at 3 min supine: r ¼  0.746, p ¼ 0.001). In addition, patients with more intense pain in the past 4 weeks generally showed a higher HR. The pain intensity was correlated with the HR during rest (r ¼ 0.384, p ¼ 0.030), during valsalva (r ¼ 0.399, p ¼ 0.024), and during the tilt (30 s upright: r ¼ 0.556, p ¼ 0.007; 1 min upright: r ¼ 0.426, p ¼ 0.048; 3 min upright: r ¼ 0.0489, p ¼ 0.034). By contrast, habitual physical activity (Baecke score) was not significantly correlated with any autonomic outcomes. Lastly, a positive correlation existed between the score on the sensory profile of the PDQ and the OGS (r ¼ 0.401, p ¼ 0.023), which indicates that the presence of more and more intense neuropathic symptoms is related to more and more severe orthostatic symptoms. Determinants for dysautonomia severity Skin extensibility was identified as a significant predictor, lowering the total CASS (standardized Beta: β ¼ 0.737, p o 0.001), adrenergic (β ¼ 0.446, p ¼ 0.014), and sudomotor CASS

(β ¼ 0.507, p ¼ 0.037). Vasoactive medication use was an additional predictor lowering the adrenergic CASS (β ¼ 0.363, p ¼ 0.035).

Discussion This study aimed to evaluate whether dysautonomia is present in EDS-HT, to determine its severity and distribution, and to explore the underlying mechanisms. Until today, autonomic function has only been tested in JHS and has been limited to testing the reactivity of the cardiovascular system [6]. Information regarding resting cardiovascular activity and the sudomotor system was added by this study. The results demonstrate that dysautonomia is indeed present in the EDS-HT population, that it is not limited to the cardiovascular system, and that it is mainly characterized by sympathetic deregulation. Dysautonomia in EDS-HT The sympathetic deregulation was apparent in both the cardiovascular and sudomotor tests. The HRV measurements suggested that resting sympathetic activity is increased in EDS-HT compared to controls, as indicated by the higher HR in combination with the higher LF/HF ratio. By contrast, sympathetic reactivity to acute cardiovascular challenges seems to be decreased, as shown by the responses to the valsalva maneuver and the tilt test. While performing the valsalva maneuver, patients showed a greater decrease in blood pressure (phase 2E) compared to controls, followed by an insufficient recovery (phase 2L). These results are similar to the study of Gazit et al. [6] in JHS and suggest that the adrenergic sympathetic nerves are unable to induce an adequate level of vasoconstriction [36]. Our patient group also showed a smaller diastolic blood pressure rise compared to controls during the initial response to tilt. Because diastolic blood pressure is mainly determined by peripheral resistance, this again implies insufficient sympathetically induced vasoconstriction. Further, sympathetic impairment was also reflected in the sudomotor function test. The patient group showed a decreased sweat response to QSART at all sites, suggesting dysfunction of the sympathetic nerves in the skin. Taken together, the EDS-HT group shows reduced sympathetic reactivity despite a resting overactivity. Similar findings have been reported in FM and severe heart failure [37,38]. In contrast to the substantial sympathetic deregulation, the results showed a fairly normal parasympathetic regulation. In accordance with the study by Gazit et al. in JHS, we found that parasympathetic reactivity to acute stressors, namely valsalva and deep breathing, was normal in EDS-HT. Only 1 parameter, the normalized HF power, suggested that parasympathetic activity at rest may be lowered compared to controls, which has also been found in FM [39]. One of the most debilitating consequences of dysautonomia in daily life is orthostatic intolerance (OI). Its prevalence in individuals with hypermobility is high, ranging between 74% in our study sample with EDS-HT and 78% in JHS [6]. In agreement with Rowe et al. [18] we found that POTS was the most typical response to tilt in EDS-HT (41% of patients) and that gradual orthostatic hypotension occurs in about 20% of patients (26% in our study and 22% in the study by Gazit et al. [6]). Both forms of orthostatic intolerance are accompanied by symptoms of dizziness, lightheadedness, vision disturbances, nausea, sweating, and sometimes chest tightness. Patients reported that, besides standing upright, these symptoms can also be provoked by physical exercise, heavy meals, a warm environment, hot showers, and baths. Although the upright blood pressure remains within adequate limits in POTS,

I. De Wandele et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

symptoms of cerebral hypoperfusion can become severe enough to limit standing time. Our patients reported a standing time limited to 5–15 min in daily life, moderately interfering with activities, such as remaining in the upright posture at a reception, during shopping, or while walking. In addition to this direct discomfort, previous research has further indicated that orthostatic intolerance is related to fatigue [9,18,40], which is the second most reported complaint in EDS-HT after pain, and to a lowered quality of life in general [13]. Underlying mechanism of dysautonomia Although dysautonomia is prevalent in EDS-HT, its cause is currently unknown. The second aim of this study was therefore to explore several theoretically plausible pathomechanisms. The results indicate that the autonomic impairment in EDS-HT is probably of multifactorial origin. First, our study seems to point to peripheral neuropathy as one of the common underlying mechanisms. Neurogenic abnormalities have been previously described in EDS [41], and the lowered QSART sweat volumes strongly suggest that this neuropathy is also affecting sympathetic fibers. The neuropathy hypothesis is further supported by the high prevalence of sensory neuropathic symptoms in our patient group and the insufficient sympathetically mediated vasoconstriction during valsalva and tilt. If the neuropathy hypothesis is true, it may also explain why POTS occurs so frequently in EDS-HT. In case of peripheral neuropathy, the sympathetic nerves are unable to create sufficient vasoconstriction during the upright posture. Consequently, the cardiac sympathetic nerves then cause a compensatory tachycardia in order to maintain sufficient cardiac output [42,43], leading to tachycardia. Because not all patients showed QSART abnormalities (35% showed normal responses), we can be ascertain that other mechanisms besides neuropathy are responsible for dysautonomia in EDSHT. Although this has never been investigated, multiple authors have speculated that the blood vessels of patients with EDS-HT have an increased distensibility, allowing for more venous pooling during the upright posture [6,18,42,44]. Our results provide some support for this idea, as we found that parameters reflecting collagen laxity, such as skin extensibility and the Beighton score, were related to parameters of vasodilatation (lowered peripheral vascular resistance, lowered supine, and standing BP) and to increased HR. Moreover, in the linear regression analysis, skin extensibility was identified as the most important predictor for the severity of sympathetic dysfunction, stronger than vasoactive medication use. Unfortunately, we were unable to measure vascular distensibility directly. Theoretically, the fact that both skin and vascular tissue contain a lot of collagen type 1 may explain how a parameter such as skin extensibility could —to some extent—reflect vascular distensibility. More substantial research is necessary to accept or refute this hypothesis. Besides neurogenic and connective tissue abnormalities, medication used in EDS-HT may profoundly influence vasomotor regulation [45,46]. Vasoactive medicines were used by 54% of patients and consisted of opiates, trazodone, blood pressurelowering agents, and tricyclic antidepressant. Opiates may cause vascular dilatation due to histamine release and depression of the vasomotor center; trazodone and blood pressure-lowering agents are known to inhibit sympathetic activation and tricyclic antidepressants block adrenergic receptors, all contributing to sympathetic failure [46]. Although vasoactive medication use was identified as the second strongest predictor for the severity of sympathetic dysfunction, it is definitely not the only causative mechanism behind dysautonomia in hypermobile individuals. In fact, dysautonomia had a similar prevalence in the study by Gazit et al. [6] in patients who did not use any medication with autonomic influence.

7

Further, depression [47], deconditioning [6,44], and paininduced sympathetic arousal have been suggested to contribute to dysautonomia. Although deconditioning often leads to orthostatic intolerance, we did not find any association between the Baecke questionnaire and orthostatic intolerance in EDS-HT. By contrast, depression was related to a lower TPR and blood pressure during the tilt, but this is probably merely attributable to the vasodilatory effect of the antidepressants used by these patients. Lastly, we found that the strongest pain experienced over the past 4 weeks was associated with increased HR at rest and during the tilt. The autonomic nervous system is known to wind up very well but may calm down quite poorly [48]. Excessive or repeated sympathetic arousal, for instance caused by intense musculoskeletal pain, may lead to a chronically heightened arousal over time, called a “hyperadrenergic state.” Further research is needed to clarify the relationship between pain and dysautonomia in EDS-HT. Limitations The present results must be viewed within the limitations of the study. First, we used indirect tests to evaluate autonomic function. Although this kind of testing is a valuable tool in determining the dysautonomia severity and distribution [24], it also reflects end-organ function rather than pure neural activity in the sympathetic and parasympathetic nerves [49]. The QSART, valsalva, and tilt results together strongly suggest peripheral sympathetic nerve dysfunction, but are all confounded by skin structure or vascular structure, which might be altered in EDS-HT. Tests directly assessing sympathetic nerve function, e.g., microneurography, are necessary to confirm neurogenic damage. A second limitation is the fact that the effect of medication could not be excluded. Many patients were medically not allowed to stop their medicines in time to wash out all effects. However, as vasoactive medication use is highly prevalent in EDS-HT, this study realistically reflects the average patient in daily life. Third, we used correlation analysis to identify factors possibly contributing to dysautonomia. As this is a cross-sectional study, inferences about cause and consequence are not fully justified. However, because of the evidence regarding the causative nature of these factors in other pathologies, we named neuropathy, deconditioning, depression, and hyperadrenergic state as a possible “causes” of dysautonomia. The results should be interpreted as a first screening towards underlying mechanisms. Conclusion In conclusion, patients with EDS-HT suffer from dysautonomia in the cardiovascular and sudomotor domain. This dysautonomia seems to consist of resting sympathetic overactivity but decreased sympathetic reactivity to stimuli. The autonomic impairment is probably of multifactorial origin. Peripheral sympathetic neuropathy and medication use with autonomic side effects are the most probable causative mechanisms. Furthermore, we postulate that patients with EDS-HT may be genetically prone to autonomic complaints due to their increased collagen laxity. Further research on this topic is warranted, however, as vascular distensibility could not be measured directly. Factors such as depression and deconditioning were found to be of less importance, but they should be addressed in individual patients if present because in combination with other factors they may contribute to the clinical expression of symptoms. Role of the funding source This work was supported by the Methusalem Grant BOF08/ 01M01108 from the Flemish Government and Ghent University to

8

I. De Wandele et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

Anne De Paepe. This did not influence the study design, collection, analysis, or the interpretation of data.

Acknowledgments We would like to express our gratitude towards all patients participating in the study. Further, we want to thank Dr. Wim Peersman for his advice on statistical analysis, Dr. Johan Ryckaert for the technical help in QSART testing, and Elke Derynck for her work in the data analysis. References [1] Beighton P, De Paepe A, Steinmann B, Tsipouras P, Wenstrup RJ. Ehlers–Danlos syndromes: revised nosology, Villefranche, 1997. Ehlers–Danlos National Foundation (USA) and Ehlers–Danlos Support Group (UK). Am J Med Genet 1998;77:31–7. [2] Levy HP. Ehlers–Danlos syndrome, hypermobility type. In: Pagon RA, Adam MP, Bird TD, Dolan CR, Fong CT, Stephens K, editors. Gene Reviews. University of Washington, Seattle, Seattle, WA; 1993. [3] Rombaut L, Malfait F, Cools A, De Paepe A, Calders P. Musculoskeletal complaints, physical activity and health-related quality of life among patients with the Ehlers–Danlos syndrome hypermobility type. Disabil Rehabil 2010;32:1339–45. [4] Castori M, Camerota F, Celletti C, Danese C, Santilli V, Saraceni VM, et al. Natural history and manifestations of the hypermobility type Ehlers–Danlos syndrome: a pilot study on 21 patients. Am J Med Genet A 2010;152A:556–64. [5] De Wandele I, Rombaut L, Malfait F, De Backer T, De Paepe A, Calders P. Clinical heterogeneity in patients with the hypermobility type of Ehlers–Danlos syndrome. Res Dev Disabil 2013;34:873–81. [6] Gazit Y, Nahir AM, Grahame R, Jacob G. Dysautonomia in the joint hypermobility syndrome. Am J Med 2003;115:33–40. [7] Martinez-Lavin M, Hermosillo AG. Autonomic nervous system dysfunction may explain the multisystem features of fibromyalgia. Semin Arthritis Rheum 2000;29:197–9. [8] Freeman R. The chronic fatigue syndrome is a disease of the autonomic nervous system. Sometimes. Clin Auton Res 2002;12:231–3. [9] van Dijk N, Boer MC, Mulder BJ, van Montfrans GA, Wieling W. Is fatigue in Marfan syndrome related to orthostatic intolerance? Clin Auton Res 2008;18:187–93. [10] Singer DH, Martin GJ, Magid N, Weiss JS, Schaad JW, Kehoe R, et al. Low heart rate variability and sudden cardiac death. J Electrocardiol 1988;21(Suppl): S46–S55. [11] Fedorowski A, Melander O. Syndromes of orthostatic intolerance: a hidden danger. J Intern Med 2013;273:322–35. [12] Schwartz PJ, La Rovere MT, Vanoli E. Autonomic nervous system and sudden cardiac death. Experimental basis and clinical observations for postmyocardial infarction risk stratification. Circulation 1992;85(Suppl. 1):I77–91. [13] Benrud-Larson LM, Dewar MS, Sandroni P, Rummans TA, Haythornthwaite JA, Low PA. Quality of life in patients with postural tachycardia syndrome. Mayo Clin Proc 2002;77:531–7. [14] Low PAaB. Quality of life in dysautonomia. In: Low PA, editor. Clinical Autonomic Disorders. 3rd ed. Lippincott Williams & Wiklins Rochester; 2010; 179–84. [15] Mustafa HI, Fessel JP, Barwise J, Shannon JR, Raj SR, Diedrich A, et al. Dysautonomia: perioperative implications. Anesthesiology 2012;116:205–15. [16] Kernan S, Tobias JD. Perioperative care of an adolescent with postural orthostatic tachycardia syndrome. Saudi J Anaesth 2010;4:23–7. [17] Rombaut L, Malfait F, De Wandele I, Cools A, Thijs Y, De Paepe A, et al. Medication, surgery, and physiotherapy among patients with the hypermobility type of Ehlers–Danlos syndrome. Arch Phys Med Rehabil 2011;92:1106–12. [18] Rowe PC, Barron DF, Calkins H, Maumenee IH, Tong PY, Geraghty MT. Orthostatic intolerance and chronic fatigue syndrome associated with Ehlers–Danlos syndrome. J Pediatr 1999;135:494–9. [19] Benarroch EE. Postural tachycardia syndrome: a heterogeneous and multifactorial disorder. Mayo Clin Proc 2012;87:1214–25. [20] Boardman A, Schlindwein FS, Rocha AP, Leite A. A study on the optimum order of autoregressive models for heart rate variability. Physiol Meas 2002;23:325–36. [21] Heart rate variability: standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Eur Heart J 1996; 17:354–81.

[22] Rudas L, Crossman AA, Morillo CA, Halliwill JR, Tahvanainen KU, Kuusela TA, et al. Human sympathetic and vagal baroreflex responses to sequential nitroprusside and phenylephrine. Am J Physiol 1999;276(5 Pt 2):H1691–8. [23] Westerhof BE, Gisolf J, Stok WJ, Wesseling KH, Karemaker JM. Time-domain cross-correlation baroreflex sensitivity: performance on the EUROBAVAR data set. J Hypertens 2004;22:1371–80. [24] Novak P. Quantitative autonomic testing. J Vis Exp 2011;53 [pii: 2502. doi: 10.3791/2502]. [25] Schrezenmaier C, Singer W, Swift NM, Sletten D, Tanabe J, Low PA. Adrenergic and vagal baroreflex sensitivity in autonomic failure. Arch Neurol 2007; 64:381–6. [26] Freeman R, Wieling W, Axelrod FB, Benditt DG, Benarroch E, Biaggioni I, et al. Consensus statement on the definition of orthostatic hypotension, neurally mediated syncope and the postural tachycardia syndrome. Auton Neurosci 2011;161:46–8. [27] Wieling W, Schatz IJ. The consensus statement on the definition of orthostatic hypotension: a revisit after 13 years. J Hypertens 2009;27:935–8. [28] Low PAS, Sandroni P, Joyner MJ, Shen WK. Postural tachycardia syndrome. In: Low PA, editor. Clinical Autonomic Disorders. 3rd ed. Lippincott Williams & Wiklins Rochester; 2010;515–33. [29] Fessel J, Robertson D. Orthostatic hypertension: when pressor reflexes overcompensate. Nat Clin Pract Nephrol 2006;2:424–31. [30] Schrezenmaier C, Gehrking JA, Hines SM, Low PA, Benrud-Larson LM, Sandroni P. Evaluation of orthostatic hypotension: relationship of a new self-report instrument to laboratory-based measures. Mayo Clin Proc 2005;80:330–4. [31] Juul-Kristensen B, Rogind H, Jensen DV, Remvig L. Inter-examiner reproducibility of tests and criteria for generalized joint hypermobility and benign joint hypermobility syndrome. Rheumatology 2007;46:1835–41. [32] Remvig L, Duhn P, Ullman S, Arokoski J, Jurvelin J, Safi A, et al. Skin signs in Ehlers–Danlos syndrome: clinical tests and para-clinical methods. Scand J Rheumatol 2010;39:511–7. [33] Baecke JA, Burema J, Frijters JE. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr 1982; 36:936–42. [34] Snaith RP. The Hospital Anxiety and Depression Scale. Health Qual Life Outcomes 2003;1:29. [35] Freynhagen R, Baron R, Gockel U, Tolle TR. painDETECT: a new screening questionnaire to identify neuropathic components in patients with back pain. Curr Med Res Opin 2006;22:1911–20. [36] Low PAaS DM. Laboratory evaluation of the autonomic nervous system. In: Low PA, editor. Clinical Autonomic Disorders. 3rd ed. Rochester: Lippincott Williams & Wiklins; 2010;143. [37] Martinez-Lavin M. Biology and therapy of fibromyalgia. Stress, the stress response system, and fibromyalgia. Arthritis Res Ther 2007;9:216. [38] van de, Borne P, Montano N, Pagani M, Oren R, Somers VK. Absence of lowfrequency variability of sympathetic nerve activity in severe heart failure. Circulation 1997;95:1449–54. [39] Cohen H, Neumann L, Shore M, Amir M, Cassuto Y, Buskila D. Autonomic dysfunction in patients with fibromyalgia: application of power spectral analysis of heart rate variability. Semin Arthritis Rheum 2000;29:217–27. [40] Grubb BP, Kanjwal MY, Kosinski DJ. Review: the postural orthostatic tachycardia syndrome: current concepts in pathophysiology diagnosis and management. J Interv Card Electrophysiol 2001;5:9–16. [41] Voermans NC, Drost G, van Kampen A, Gabreels-Festen AA, Lammens M, Hamel BC, et al. Recurrent neuropathy associated with Ehlers–Danlos syndrome. J Neurol 2006;253:670–1. [42] Kanjwal Y, Kosinski D, Grubb BP. The postural orthostatic tachycardia syndrome: definitions, diagnosis, and management. Pacing Clin Electrophysiol 2003;26:1747–57. [43] Raj SR. The postural tachycardia syndrome (POTS): pathophysiology, diagnosis & management. Indian Pacing Electrophysiol J 2006;6:84–99. [44] Mathias CJ, Low DA, Iodice V, Owens AP, Kirbis M, Grahame R. Postural tachycardia syndrome—current experience and concepts. Nat Rev Neurol 2012;8:22–34. [45] Bradley JG, Davis KA. Orthostatic hypotension. Am Fam Physician 2003;68:2393–8. [46] Milazzo V, Di Stefano C, Servo S, Crudo V, Fulcheri C, Maule S, et al. Drugs and Orthostatic hypotension: evidence from the literature. J Hypertens 2012;1:104. [47] Hausberg M, Hillebrand U, Kisters K. Addressing sympathetic overactivity in major depressive disorder. J Hypertens 2007;25:2004–5. [48] Rahman A, Holman JA. Fibromyalgia and hypermobility. In: Hakim AJ, Keer R, Grahame R, editors. Hypermobility, Fibromyalgia and Chronic Pain. 1st ed. Elsevier; 2010;61–8. [49] Nahm F, Freeman R. Autonomic testing. In: Blum AS, editor. The Clinical Neurophysiology Primer. 1st ed. Humana Press Totowa, New Jersey; 2007; 447–60.