- Email: [email protected]
Cerebral autoregulation is preserved in multiple sclerosis patients
Journal of the Neurological Sciences 381 (2017) 298–304
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Cerebral autoregulation is preserved in multiple sclerosis patients a,⁎,1
a,b,1
a
c
MARK b
Daniel Ferreira , Pedro Castro , Gonçalo Videira , João Pedro Filipe , Rosa Santos , Maria José Sáb,d, Elsa Azevedoa,b, Pedro Abreua,b a
Department of Clinical Neurosciences and Mental Health, Faculty of Medicine of University of Porto, 4200-319 Porto, Portugal Department of Neurology, São João Hospital Center, 4200-319 Porto, Portugal Department of Neuroradiology, Hospital Center São João, 4200-319 Porto, Portugal d Faculty of Health Sciences, University Fernando Pessoa, 4249-004 Porto, Portugal b c
A R T I C L E I N F O
A B S T R A C T
Keywords: Multiple sclerosis Cerebral autoregulation Autonomic nervous system Heart rate variability Spontaneous baroreflex sensitivity Magnetic resonance imaging
Multiple sclerosis (MS) is an inflammatory disease that may also be associated with vascular dysfunction. One master component of vascular regulation is cerebral autoregulation (CA). We aimed to investigate the integrity of CA in MS patients and study its relationship with autonomic dysfunction (AD), magnetic-resonance-imaging (MRI) lesion load and hemodynamic parameters. We enrolled 20 relapsing-remitting MS and 20 healthy subjects. CA was assessed by transfer function analysis parameters (coherence, gain and phase), as obtained in the very low, low and high-frequency domains (VLF, LF, HF, respectively). We evaluated the autonomic parameters heart rate variability and spontaneous baroreflex sensitivity (BRS). There were no significant differences in CA parameters between MS and controls (p > 0.05). Lesion load was not correlated with any CA parameter. LF gain was positively correlated with BRS in both groups (MS: p = 0.017; controls: p = 0.025). Brainstem lesion load in MS was associated with higher systolic blood pressure (SBP; p = 0.009). Our findings suggest that CA is preserved in our MS cohort. On the other hand, AD in MS patients with brainstem lesions could contribute to the increase of supine SBP. Whether this systemic deregulation could contribute to disease burden remains to be investigated.
1. Introduction Multiple sclerosis (MS) is an autoimmune, demyelinating disease of the central nervous system, where chronic inflammation and axonal injury lead to deficits in motor, sensory, cerebellar and cognitive functions [1]. Cardiovascular (CV) autonomic deregulation is also present in these patients, affecting both the sympathetic and parasympathetic nervous systems, which innervate cerebral vessels [2–4]. Additionally, a higher incidence of cerebrovascular events is present in MS patients in comparison to sex- and age-matched controls, suggesting that vascular dysfunction plays a role throughout the disease course [5,6]. Smoking is a major CV risk factor, as well as an environmental factor for MS development; however, widespread cerebral hypoperfusion seems to be the key factor in predisposing MS patients to ischemic brain lesions [7]. Nevertheless, most of the etiological mechanisms underlying these events remain unclear. Indeed, an adequate blood perfusion is crucial to ensure brain metabolic demand. This mechanism is maintained by three major
⁎
1
components: cerebral autoregulation (CA), cerebrovascular reactivity (VR) and neurovascular coupling [8]. VR was proven to be impaired in MS patients in several studies, even though there are some contradictory results [9,10]. Regarding CA, only one study was conducted, and it showed no significant differences in the autoregulatory mean velocity index between MS patients and controls; however, an autoregulation impairment was described after modeling variation in the cerebral blood flow velocity (CBFV) associated with a 1-mm Hg increase in the mean arterial blood pressure [11]. Thus, it is still unclear whether the CA mechanism remains intact in MS or if it has any relation to vascular and autonomic impairment, and, if CA deregulation plays a role in MS lesion burden. It is worth investigating CA role in MS because we know that it is both impaired by dysautonomia [12,13] and involved in cerebrovascular pathology [14]. Our main objective was to investigate the status of CA in patients with MS. Also, we aimed to study its relationship with the magnitude of autonomic dysfunction (AD) by applying a battery of cardiovascular tests (Ewing battery) and methods to quantify the autonomic system integrity (heart rate variability – HRV – and spontaneous baroreflex
Corresponding author at: Dept. Neurology, Centro Hospitalar de São João, Alameda Professor Hernâni Monteiro, 4200-319 Porto, Portugal. E-mail address: [email protected] (D. Ferreira). Both authors contribute equally to this work.
http://dx.doi.org/10.1016/j.jns.2017.09.009 Received 2 July 2017; Received in revised form 23 August 2017; Accepted 6 September 2017 Available online 08 September 2017 0022-510X/ © 2017 Elsevier B.V. All rights reserved.
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
sensitivity – BRS). MRI brain lesion load and its correlation with CA and hemodynamic parameters were evaluated as well. We hypothesized that an impaired CA in MS patients might play a role in the disease burden.
2.3. Cerebral autoregulation, spontaneous baroreflex sensitivity and heart rate variability measurements Data was visually inspect for artifact removal. Extra-beats were removed by linear interpolation. Beats were marked on R-waves of ECG. For each heartbeat, systolic, mean and diastolic values of arterial BP (SBP, MBP, and DBP, respectively) and CBFV were calculated and visually corrected if needed. CA parameters coherence, gain and phase were calculated using transfer function analysis, as recommended [15]: time-series of spontaneous oscillations in mean CBFV and MBP were interpolated at 10 Hz with a third-order polynomial spline; then an averaged periodogram was calculated by Welch method with Hanning window of window length of 102 s; 50% superposition; and triangular three-point window spectral smoothing filter. Coherence was calculated between input auto-spectra of ABP over cross-spectra of CBFV/ABP and transfer functions of phase and gain were determined by dividing the cross-spectrum by the input autospectrum. Values were reported in three frequency bands: very low (VLF: 0.02–0.07 Hz), low (LF 0.07–0.20 Hz) and high (HF: 0.20–0.50 Hz). The BRS was evaluated in both the time and frequency domains. In the time domain, using the cross-correlation method xBRS, in which the regression slope between SBP and RRi was computed [16]. In resume, xBRS computes the correlation between beat-to-beat systolic blood pressure and normal RRi interval, resampled at 1 Hz, in a sliding 10 s window, with delays of 0–5 s for interval. The delay with the greatest positive correlation is selected and, when significant at p < 0.01, slope and delay are recorded as one xBRS value. The mean value of xBRS is finally calculated. In the frequency domain, the baroreceptor gain (BRG) was calculated based on the cross-correlation gain between the power spectral densities of SBP and RRi. Spectral analysis from timeseries of SBP and RRi were achieved with the same parameters as in described in to access CA. TFA parameter gain was calculated by dividing the cross-spectrum by the input auto-spectrum. BRG was obtained in the band of low frequencies (LF: 0.04–0.15 Hz). xBRS determination of BRS maybe more reliable in patients with low BRS due to autonomic impairment since it is facilitated by the absence of thresholds for pressure and interval variation. Since MS patients might have disautonomia we decided to include to different methods that could strengthen and confirm significant correlations. The frequency domain was used to assess HRV. The fast Fourier transformation was applied to the RRi tachogram, and a spectrum of the HRV signal was obtained. Again, this spectral analysis followed the same parameters described to achieve averaged periodogram described in CA. The following parameters were calculated: total power spectrum of RRi and its low (LF: 0.04–0.15 Hz) and high (HF: 0.15–0.04 Hz) frequency ranges.
2. Materials and methods This matched case-control study was performed in São João Hospital Center in Porto, Portugal. The local institutional ethic committee approved the study and every participant gave written and signed consent. 2.1. Population studied From our MS outpatient clinic, we randomly selected 20 MS patients to participate (12 females). Inclusion criteria were a definite diagnosis of MS, according to the 2010 McDonald criteria, and age above 18 years old. Only relapsing-remitting MS patients were engaged and those who experienced a relapse or had been submitted to corticosteroids treatment within the previous eight weeks were excluded. The control group was selected from the hospital and university staff and consisted of 20 age- and sex-matched healthy subjects. Exclusion criteria were the presence of comorbidities that could affect autonomic function, as known cardiovascular diseases (e.g., myocardial infarction, heart failure, atrial fibrillation) and dysautonomias (e.g., Parkinsonism, multiple system atrophy, diabetes, polyneuropathy). All patients were evaluated by an MS specialist within one month before enrollment, and the following data was recorded: age, gender, neurological examination, Expanded Disability Status Scale (EDSS) score, and duration of disease. 2.2. Experimental design: monitoring and autonomic test protocol Before evaluation, patients and controls were instructed to pause vasoactive drugs in the previous 24 h and abstain from drinking coffee and alcohol or smoke in the previous 12 h. Measurements were performed in a quiet room with a temperature around 22 °C. CBFV was recorded bilaterally from M1 segment of the middle cerebral arteries (MCA), at a depth of 50–55 mm, with 2 MHz monitoring probes held with a headband. (Doppler BoxX, DWL, Singen, Germany). Blood pressure (BP) was continuously monitored in the patient's non-dominant hand using the noninvasive finger cuff Finapres (model 2300; Ohmeda, Englewood, CO, USA). Heart rate and R-R intervals (RRi) were assessed from a 3‑lead ECG. End-tidal carbon dioxide was recorded with a capnograph (Respsen Nonin, Amsterdam, Netherlands) by nasal cannula. All data were synchronized and digitally recorded at 400 Hz with with Powerlab (AD Instruments, Oxford, UK) for offline analysis. Autonomic testing was explained to the volunteers. Maximum contraction strength with the dominant hand was measured using a dynamometer. Patients lied supine for 30 min before evaluation. Assessment of CA, baroreflex function and short-term heart rate variability (HRV) was based on data from a 10-min period of resting in the supine position with uncrossed legs. Afterward, the autonomic tests protocol was performed as described elsewhere [3], based on the Ewing battery. In brief, maximum contraction strength in dominant hand (or non-paretic hand if dominant was affected) was measured with a dynamometer, the subject was positioned supine and rested for 30 min. Next, we calculated: expiratory–to–inspiratory amplitude ratio of HR (E:I ratio) during synchronized deep breathing at 6 cycles/min; the Valsalva ratio, i.e., averaged HR ratio between II and IV phases of three Valsalva maneuvers; the quotient of RRi around 30th beat by that at the 15th after standing; the difference between the systolic BP (SBP) at rest and after standing during 5 min to assess orthostatic hypotension; and diastolic BP (DBP) response to handgrip isometric work at 30% of his maximum strength for 5 min. A final total score (ETS) was obtained by adding 0 if the tests were normal, 1 if borderline and 2 if abnormal [3].
2.4. MRI The brain MRI had been obtained within one year before enrollment. The spinal MRI was the last imaging to be performed, and none of the subjects experienced a relapse after imaging. We assessed the presence of T2-weighted lesions in the following central autonomic network (CAN) structures: insula, anterior cingulate cortex, hypothalamus, amygdala, periaqueductal gray matter, parabrachial nucleus, nucleus of the solitary tract, ventrolateral reticular formation of the medulla and medullary raphe. By adding 1 point for each affected structure, we calculated a CAN score for each patient. The number of T2-weighted lesions in supratentorial areas, brainstem, cerebellum, and spinal cord was calculated, as well as the total number. MRI protocols were performed in 1.5T (Siemens MAGNETOM SymphonyTim syngo) and 3T (Siemens MAGNETOM TrioTim syngo) scanners. These imaging protocols included, at least, coronal and axial T1-weighted images (511750 ms/8.6–8.7 ms [TR/TE]), spin-echo or fast spin-echo axial proton density images (3770–4000 ms/11–22 ms [TR/TE]), T2-weighted images (3770–4000 ms/88–106 ms [TR/TE]) and axial and sagittal fluid-attenuated inversion-recovery (FLAIR) images (8000-9000 ms/ 299
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Table 1 Individual clinical data of MS patients. ID
Gender
Age
Disease duration
EDSS
medication
ARR
Brainstem relapse
Spinal relapse
MSSS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
F F F F M M M F M M M F F F M F F M F F
33 27 23 45 32 48 35 46 32 48 27 48 47 44 35 40 32 35 40 32
4 1 2 25 5 2 5 6 10 19 6 3 7 6 7 6 6 0 0 5
1.0 1.0 0.0 2.0 1.0 2.5 1.0 1.5 0.0 2.5 1.0 1.5 1.5 1.0 1.0 1.0 1.5 1.0 1.0 1.5
Interferon-β1a Interferon-β1a Interferon-β1a Interferon-β1a Interferon-β1a Glatiramer acetate Natalizumab Natalizumab Natalizumab Glatiramer Natalizumab Glatiramer acetate Interferon-β1a Interferon-β1a Fingolimod Dimethyl fumarate Interferon-β1a Interferon-β1a Glatiramer acetate Natalizumab
0.25 0.00 0.00 0.33 0.00 0.00 0.20 0.50 0.00 0.00 0.50 0.43 0.42 0.33 0.42 0.66 0.66 0.00 0.00 0.40
Yes No No Yes No No Yes Yes No Yes Yes No Yes Yes Yes No Yes No No Yes
No No No Yes No No Yes No No Yes Yes Yes No Yes Yes No Yes No No No
1.45 2.44 0.45 0.77 1.28 6.64 1.28 2.33 0.19 1.39 1.13 3.34 2.10 1.13 1.04 1.13 2.33 n.a n.a 2.60
F: female; M: male; EDSS: Expanded Disability Status Scale; ARR: annualised relapse rate; MSSS: Multiple Sclerosis Severity Scale; n.a: not applicable.
3. Results
91–93 ms [TR/TE]), all with 5 mm section thickness.
In the healthy control group, we included 20 subjects with a median age of 36 years. Clinical features of the 20 MS patients are presented in Table 1. Of the 20 patients, 12 were women. MS patients presented a median age of 36 years, EDSS of 1, disease duration of 5.5 years and Multiple Sclerosis Severity Score (MSSS) [17] of 1.3.
2.5. Statistical analyses The normality of the variables was determined by Shapiro-Wilk test. No statistically significant differences (p < 0.05) were detected between the right and left MCA hemodynamic values. Therefore, both values were averaged and used in subsequent analysis. Student t-test or Mann-Whitney test were used to compare hemodynamic characteristics and CA parameters between the two groups. The correlations between two variables were determined by Pearson or Spearman rho coefficients, and a multiple linear regression was performed for posterior multivariate analysis. Post-hoc analysis was performed with ANCOVA analysis to explore the influence of age groups (median as cut-off), sexes and MS vs control over each of the dependent variables. All calculations were performed using the software IMS SPSS Statistics v21™.
3.1. Hemodynamic parameters: comparing the MS patients with the control group Table 2 compares the hemodynamic characteristics between the MS and healthy control groups. No significant differences were found for any of the parameters (p < 0.05). 3.2. Cerebral autoregulation: comparing the MS patients with the control group Transfer function analysis parameters for CA are presented in Table 3. The comparison between the two groups revealed no significant differences (p < 0.05) in any of the CA parameters. A tendency for higher values of phase in the HF band was found in the
Table 2 Comparison of hemodynamic characteristics between MS and control group.
BP, mm Hg BP spectral power, mm Hg2 BP covariance, % Systolic BP, mm Hg CBFV, cm·s− 1 Right Left Average CBFV spectral power, cm2·s− 1 Right Left Average CBFV covariance, % Right Left Average HR, bpm HR covariance, % Mean EtCO2, mm Hg EtCO2 covariance, %
Control group
MS group
p value
75 ± 15 16 ± 9 5 ± 2 115 ± 20
75 ± 13 13 ± 6 4 ± 1 117 ± 16
0.940 0.281 0.257 0.733
67 ± 14 68 ± 14 67 ± 14
66 ± 11 72 ± 17 69 ± 12
0.841 0.445 0.676
16 ± 13 16 ± 15 16 ± 14
15 ± 10 17 ± 12 16 ± 10
0.989 0.640 0.883
5 ± 2 5 ± 1 5 ± 1 68 ± 10 6 ± 3 32 ± 10 8 ± 11
5 ± 1 5 ± 1 5 ± 1 70 ± 8 5 ± 2 37 ± 6 3 ± 2
0.620 0.820 0.718 0.411 0.086 0.264 0.081
Table 3 Comparison of cerebral autoregulation parameters between MS and control group. CA parameters Coherence, a.u VLF LF HF Gaina, %.mm Hg− 1 VLF LF HF Phase, radians VLF LF HF
Control group
MS group
P value
0.38 ± 0.18 0.64 ± 0.15 0.58 ± 0.12
0.32 ± 0.11 0.60 ± 0.16 0.65 ± 0.14
0.151 0.429 0.100
0.91 ± 0.42 1.30 ± 0.47 1.62 ± 0.49
0.82 ± 0.33 1.31 ± 0.36 1.57 ± 0.34
0.221 0.779 0.904
0.90 ± 0.36 0.69 ± 0.23 0.09 ± 0.17
0.96 ± 0.44 0.58 ± 0.14 -0.03 ± 0.16
0.620 0.069 0.059
Values are given in mean ± SD (side deviation); a.u: arbitrary units; VLF: very low frequencies (0.02–0.07 Hz); LF: low frequencies (0.07–0.20 Hz); HF: high frequencies (0.20–0.50 Hz). a None normalized gain (cm·s− 1·mm Hg− 1) also showed no differences between groups (data not shown).
Values are given in mean ± SD (side deviation); BP: blood pressure; CBFV: cerebral blood flow velocity; HR: heart rate; EtCO2: end-tidal carbon dioxide.
300
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Table 4 Correlations between systemic hemodynamics, autonomic tests and cerebral autoregulation (CA) parameters in the control group. CA parameters
Hemodynamics
Gain, %.mm Hg− 1 VLF LF HF Phase, radians VLF LF HF
Ewing battery
BRS
HRV
SBP
DBP
Total score
xBRS
BRG
LF
HF
Total Power
0.046 − 0.391 − 0.482⁎
0.442 −0.263 −0.137
− 0.041 − 0.373 − 0.078
0.211 0.499⁎ 0.409
0.278 0.472⁎ 0.400
0.367 0.200 0.313
0.364 0.132 0.205
0.182 0.146 0.224
0.156 0.379 − 0.153
0.102 0.385 0.034
− 0.045 0.194 − 0.276
0.074 − 0.131 − 0.072
−0.247 −0.026 −0.322
− 0.116 0.203 − 0.367
− 0.051 − 0.027 − 0.226
−0.217 −0.041 −0.241
SBP: systolic blood pressure; DBP: diastolic blood pressure; BRS: baroreflex sensitivity; xBRS: BRS cross correlation in time domain; BRG: transfer function gain of BRS; HRV: heart rate variability in low (LF: 0.04–0.15 Hz) and high (HF: 0.15–0.40 Hz) frequencies; CA parameters in very low (VLF: 0.02–0.07 Hz), low (LF: 0.07–0.20 Hz) and high (HF: 0.20–0.50 Hz) frequencies. ⁎ p < 0.05 significance of Pearson or Spearman's correlation coefficients.
Table 5 Correlations between systemic hemodynamics, autonomic tests and cerebral autoregulation (CA) parameters in MS group. CA parameters
Gain, %.mm Hg VLF LF HF Phase, radians VLF LF HF
Hemodynamics
Ewing battery
BRS
HRV
SBP
DBP
Total score
xBRS
BRG
LF
HF
Total Power
0.424 −0.454⁎ −0.375
− 0.027 − 0.186 − 0.252
− 0.019 − 0.170 − 0.095
0.065 0.526⁎ 0.457⁎
0.156 0.609⁎⁎ 0.459⁎
−0.148 0.045 0.111
0.273 0.677⁎⁎ 0.568⁎⁎
0.073 0.418 0.404
−0.329 0.137 0.356
0.118 0.305 0.061
0.444 0.189 0.246
0.002 0.060 − 0.549⁎
−0.075 −0.071 −0.433
0.167 0.261 −0.256
−0.018 −0.356 −0.200
0.058 −0.148 −0.335
−1
SBP: systolic blood pressure; DBP: diastolic blood pressure; BRS: baroreflex sensitivity; xBRS: BRS cross correlation in time domain; BRG: transfer function gain of BRS; HRV: heart rate variability in low (LF: 0.04–0.15 Hz) and high (HF: 0.15–0.40 Hz) frequencies; CA parameters in very low (VLF: 0.02–0.07 Hz), low (LF: 0.07–0.20 Hz) and high (HF: 0.20–0.50 Hz) frequencies. ⁎ p < 0.05 significance of Pearson or Spearman's correlation coefficients. ⁎⁎ p < 0.01 significance of Pearson or Spearman's correlation coefficients.
control group (0.09 ± 0.17 vs. − 0.03 ± 0.16; p = 0.059). Still, CA is not believed to operate in this frequency range [15].
No significant relationship was found for the CAN score or for the total number of lesions and involvement of brain, cerebellum and spinal cord. The exception was the number of brainstem lesions, which was strongly associated with a higher SBP (r = 0.628; p = 0.005) and with the VLF gain (r = 0.474; p = 0.047). Subsequent multivariate analysis using logistic regression analysis was conducted to study this correlation. It showed no significant relation between brainstem lesion burden and the VLF gain function, but a higher number of lesions was still associated with a higher systolic BP (β = 0.552; p = 0.009), as presented by a linear regression function in Fig. 1. Different averaged values of SBP regarding the presence or absence of brainstem lesions are plotted in Fig. 2.
3.3. Cerebral autoregulation correlation with hemodynamics and autonomic tests The association between systemic hemodynamics, autonomic tests (ETS, BRS and HRV) and CA parameters was calculated for the control and MS groups and is presented in Tables 4 and 5, respectively. In the control group, CA parameters did not correlate with ETS and HRV. Regarding BRS, a positive correlation was found with LF gain for xBRS (r = 0.499; p = 0.025) and BRG (r = 0.472; p = 0.036). Also, we found a negative correlation between HF gain and SBP (r = −0.482; p = 0.031). In the MS group, we found no correlation between ETS and CA parameters as well. Concerning BRS, xBRS was positively correlated with the gain parameter in the LF (r = 0.526; p = 0.017) and HF (r = 0.457; p = 0.043) bands and negatively with the HF Phase (r = − 0.549; p = 0.012). BRG was also found to be positively correlated with gain in the LF (r = 0.609; p = 0.004) and HF (r = 0.459; p = 0.042) bands. Additionally, SBP was negatively correlated with LF gain (r = − 0.454; p = 0.044) and HRV was found to be associated with gain in the LF (r = 0.677; p = 0.001) and HF (r = 0.568; p = 0.009) bands.
3.5. Age and gender effects Post-hoc analysis was performed with ANCOVA analysis to explore each of the dependent variables (CA measures, xBRS and RR total power) adjusting for gender, age strata (< 37 vs ≥37 years-old; median) and group (MS vs control). We found that males (6.7 ± 2.2 vs 10.7 ± 6.0 mm Hg/ms, p = 0.021) and older individuals (7.7 ± 1.9 vs 15 ± 10.1 mm Hg/ms, p = 0.002) tend to have lower xBRS similarly between groups. HRV was also reduced in older individuals (RR total power 635 ± 326 vs 2323 + 2240 ms2, p = 0.014) irrespective of gender or group. CA measures did not differ in gender and age (p > 0.05), which is in accordance with the results of a larger healthy cohort we published [8]. In what concerns the relationship between number of brain stem lesions and SBP we found that was not disturbed by gender or age (p > 0.05)
3.4. MS group: correlation between MRI lesion burden, hemodynamics and CA measurements Table 6 presents the correlation coefficient between lesion burden and cerebrovascular and systemic hemodynamics in the MS patients. 301
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Table 6 Correlations between MRI lesions in the CNS of MS patients and measurements of hemodynamics and CA parameters. CAN scorea
Hemodynamics SBP, mm Hg DBP, mm Hg CBFV, cm·sec− 1 CA parameters Gain, %.mm Hg− 1 VLF LF HF Phase, radians VLF LF HF
Number of lesions Brain
Brainstem
Cerebellum
Spinal cord
Total
0.306 −0.147 0.112
− 0.189 0.027 − 0.044
0.628⁎⁎ 0.099 0.217
0.032 0.287 0.061
0.253 0.130 0.121
−0.115 −0.083 −0.034
0.227 0.146 0.015
0.254 0.422 0.247
0.474⁎ 0.165 − 0.055
0.184 0.343 0.255
0.394 0.022 −0.145
0.065 0.450 0.255
−0.413 −0.326 0.295
0.078 − 0.189 0.328
− 0.152 0.114 0.420
0.172 0.138 0.067
−0.382 −0.270 0.298
0.078 −0.287 0.275
CAN: central autonomic network; SBP: systolic blood pressure; DBP: diastolic blood pressure; CBFV: cerebral blood flow velocity; VLF: very low frequencies (0.02–0.07 Hz); LF: low frequencies (0.07–0.20 Hz); HF: high frequencies (0.20–0.50 Hz). ⁎ p < 0.05 significance of Pearson's correlation coefficient. ⁎⁎ p < 0.01 significance of Pearson's correlation coefficient. a CAN score was calculated as described elsewhere [3] by adding 1 point for each MRI T2-weighted lesion in the white matter adjacent to CAN structures (insula, anterior cingulate cortex, hypothalamus, amygdala, periaqueductal gray matter, parabrachial nucleus, nucleus of the solitary tract, ventrolateral reticular formation of the medulla and medullary raphe).
4. Discussion
relationship between HRV in the HF band and a worst CA response, reflected by a higher LF gain. To our knowledge, this is the first time this association is documented in MS patients. The presence of AD in this MS cohort it detailed in a previous publication [3], in which we documented lower HRV indexes and BRS in MS compared to healthy subjects. CAN lesions does not seem to justify this impairment. Our study also revealed a positive correlation between supine systolic BP and lower LF gain, indicating that CA adjusted itself adequately to higher levels of BP in our patients. In fact, Serrador et al. reported that hypertensive patients demonstrated a better attenuation of cerebral blood flow fluctuations in response to BP changes in LF gain, possibly revealing a permanent remodeling of the cerebrovasculature [27]. Also, in patients with small vessel disease, a positive linear relationship between mean 24-h BP and the autoregulatory index (ARI) was found [28]. Altogether, this data suggest that MS patients may also present a counteractive CA response to increased systemic BP variability. After multivariate analysis adjustment for a possible confounding factor (VLF gain), we found a significant positive correlation between systolic BP in the supine position and brainstem lesion load. In agreement with these findings, patients presenting at least one MRI
Our study main conclusion is that CA is preserved in the mild stages of MS. We found no differences in the CA parameters coherence, gain and phase between the two groups. Our results are in agreement with other studies suggesting a preserved CA mechanism in subjects with AD [18,19] or other neurological disorders with AD, such as Alzheimer disease [20], familial dysautonomia [21] and Shy-Drager Syndrome [22]. However, some studies in patients with autonomic failure [13,23] or with AD and autoimmune diseases [24,25] revealed an impaired CA. These contrasting results can be a consequence of a distinct underlying disease pathophysiology that may ultimately lead to different cerebrovascular responses to AD. Concerning the relationship between AD parameters and CA, we showed that there was an inverse relationship between the efficacy of CA and baroreflex mechanisms. This was shown by the fact that a more tuned baroreflex, i.e., higher values of xBRS or BRG, correlated positively with higher values of CA LF gain, meaning a worse capability of the brain to dampen blood pressure oscillations. This association had already been described [26] in healthy subjects and was also shown in our MS group, suggesting the integrity of a possible compensatory mechanism in these patients. Additionally, we described a similar
Fig. 2. MS patients' averaged systolic blood pressure (BP) in the supine position, regarding the presence or absence of MRI brainstem lesions. *p < 0.05, significant difference in paired t-test within group.
Fig. 1. Linear correlation between the number of MRI brainstem lesions and systolic blood pressure (BP) in the supine position, in MS patients.
302
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Conflict of interest
brainstem lesion had higher systolic BP comparing to those with no lesions. Higher values of supine BP in these patients might indicate some level of cardiovascular autonomic impairment, which was already known to be correlated with lesions in the brainstem, according to other studies [29,30]. An elevated systolic BP in the supine position, conjugated with the CV dysfunction in these patients might point toward a potential manifestation of supine hypertension (SH) and/or orthostatic hypotension (OH). Some studies in patients with OH have demonstrated a self-adjustment of the autoregulatory range of CA, maintaining a relatively constant cerebral perfusion [31,32]. For instance, patients with Parkinson's disease manifesting SH revealed a higher prevalence of white matter lesions (WML) [33], which were demonstrated to be an important prognostic factor for stroke, cognitive impairment, dementia and death [34]. Even though ARI might not be directly correlated with WML load [28], there seems to be a complex interplay between systemic BP, cerebrovascular hemodynamics and brain lesion load in diseases with AD. Altogether, these factors may contribute to the disease burden in MS. Age and gender are important factors that influence baroreflex function but not CA [8]. We did not find any significant influence of these demographic variables in the alterations of MS patients. MS patients have shown widespread cerebral hypoperfusion which seems not to be secondary to axonal degeneration. Our results are contrary to the theory that this is due to a primary arterial vascular dysfunction namely in terms of autoregulation. Our findings adds more evidence to the claims of some authors that hypo perfusion might be a result of reduced axonal/astrocyte activity and energy metabolism. CA assessment is focused in arterial vessels. Our negative results might be cannot exclude some deregulation in venous level. Some authors suggest slower cerebral venous flow in patients with MS can contribute to disease. Nevertheless, some limitations concerning this study deserve mentioning, and considerations should be made regarding the interpretation of these results. Firstly, AD with progressive impairment of cardiac autonomic balance in MS correlates with higher disease severity and longer duration [4], with a study showing that changes in CBFV were more prominent in patients with EDSS > 2 [35]. Our MS sample had a low median EDSS and disease duration, which did not allow us to study the possibility of CA impairment becoming apparent over the course of the disease and, thus, limits the generalization of these results to the broad spectrum of MS manifestations. Additionally, we cannot rule out the present of more asymptomatic lesions at time of the evaluation since MRI scan was allowed to be within one year from patient assessment. Secondly, even though preliminary studies have not detected differences in the CBFV response to different maneuvers [36], this possibility should be kept in mind when comparing different approaches to measure the influence of autonomic nervous system in cerebral blood flow. In a steady-state situation, we found no significant differences between the two groups. However, stronger challenges to the control of cerebral blood flow, such as squat-stand maneuvers, could be necessary for changes in CA to become apparent. Lastly, it is important to emphasize that merely CBFV was measured by transcranial Doppler, hence only reflecting the changes in cerebral blood flow if MCA's diameter remains constant [15,36], which is expected in the study protocol conditions. Overall, this data indicates that cerebral autoregulation is preserved in our MS cohort. Autonomic dysfunction in multiple sclerosis patients with brainstem lesions could contribute to their increased supine systolic blood pressure. However, these patients seem to present an adaptive mechanism with a decrease in CA gain that dampens possible deleterious effects of higher levels of blood pressure. Whether this systemic deregulation could contribute to disease burden remains to be investigated. Nevertheless, our data suggest that MS physicians should make an effort to carefully monitor cardiovascular risks, especially through the evaluation of blood pressure, in MS patients.
None. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Acknowledgments This work was partially supported by the “Associação para o Estudo da Doença Vascular Cerebral do Hospital de S. João”. The authors acknowledge the work of Isabel Melo for helping with English editing. References [1] A. Compston, A. Coles, Multiple sclerosis, Lancet 372 (9648) (2008) 1502–1517. [2] C.-A. Haensch, J. Jörg, Autonomic dysfunction in multiple sclerosis, J. Neurol. 253 (1) (2006) i3–i9. [3] G. Videira, P. Castro, B. Vieira, J.P. Filipe, R. Santos, E. Azevedo, M.J. Sá, P. Abreu, Autonomic dysfunction in multiple sclerosis is better detected by heart rate variability and is not correlated with central autonomic network damage, J. Neurol. Sci. 367 (2016) 133–137. [4] K. Nasseri, B. TenVoorde, H. Ader, B. Uitdehaag, C. Polman, Longitudinal follow-up of cardiovascular reflex tests in multiple sclerosis, J. Neurol. Sci. 155 (1) (1998) 50–54. [5] N.B. Allen, J.H. Lichtman, H.W. Cohen, J. Fang, L.M. Brass, M.H. Alderman, Vascular disease among hospitalized multiple sclerosis patients, Neuroepidemiology 30 (4) (2008) 234–238. [6] C.F. Christiansen, S. Christensen, D.K. Farkas, M. Miret, H.T. Sorensen, L. Pedersen, Risk of arterial cardiovascular diseases in patients with multiple sclerosis: a population-based cohort study, Neuroepidemiology 35 (4) (2010) 267–274. [7] M. D'Haeseleer, M. Cambron, L. Vanopdenbosch, J. De Keyser, Vascular aspects of multiple sclerosis, Lancet Neurol. 10 (7) (2011) 657–666. [8] J. Madureira, P. Castro, E. Azevedo, Demographic and systemic hemodynamic influences in mechanisms of cerebrovascular regulation in healthy adults, J. Stroke Cerebrovasc. Dis. 26 (3) (2017) 500–508. [9] O. Marshall, H. Lu, J.-C. Brisset, F. Xu, P. Liu, J. Herbert, R.I. Grossman, Y. Ge, Impaired cerebrovascular reactivity in multiple sclerosis, JAMA Neurology 71 (10) (2014) 1275–1281. [10] N. Uzuner, S. Ozkan, N. Cinar, Cerebrovascular reactivity in multiple sclerosis patients, Mult. Scler. 13 (6) (2007) 737–741. [11] Z. Mezei, L. Olah, L. Kardos, R.K. Kovacs, L. Csiba, T. Csepany, Cerebrovascular hemodynamic changes in multiple sclerosis patients during head-up tilt table test: effect of high-dose intravenous steroid treatment, J. Neurol. 260 (9) (2013) 2335–2342. [12] E. Azevedo, P. Castro, R. Santos, J. Freitas, T. Coelho, B. Rosengarten, R. Panerai, Autonomic dysfunction affects cerebral neurovascular coupling, Clin. Auton. Res. 21 (6) (2011) 395–403. [13] P.M. Castro, R. Santos, J. Freitas, R.B. Panerai, E. Azevedo, Autonomic dysfunction affects dynamic cerebral autoregulation during Valsalva maneuver: comparison between healthy and autonomic dysfunction subjects, J. Appl. Physiol. 117 (3) (2014) 205–213. [14] P. Castro, J.M. Serrador, I. Rocha, F. Sorond, E. Azevedo, Efficacy of cerebral autoregulation in early ischemic stroke predicts smaller infarcts and better outcome, Front. Neurol. 8 (2017) 113. [15] J.A. Claassen, A.S. Meel-van den Abeelen, D.M. Simpson, R.B. Panerai, N. International Cerebral Autoregulation Research, Transfer function analysis of dynamic cerebral autoregulation: A white paper from the International Cerebral Autoregulation Research Network, J. Cereb. Blood Flow. Metab. 36 (4) (2016) 665–680. [16] B.E. Westerhof, J. Gisolf, W.J. Stok, K.H. Wesseling, J.M. Karemaker, Time-domain cross-correlation baroreflex sensitivity: performance on the EUROBAVAR data set, J. Hypertens. 22 (7) (2004) 1371–1380. [17] R.H. Roxburgh, S.R. Seaman, T. Masterman, A.E. Hensiek, S.J. Sawcer, S. Vukusic, I. Achiti, C. Confavreux, M. Coustans, E. le Page, G. Edan, G.V. McDonnell, S. Hawkins, M. Trojano, M. Liguori, E. Cocco, M.G. Marrosu, F. Tesser, M.A. Leone, A. Weber, F. Zipp, B. Miterski, J.T. Epplen, A. Oturai, P.S. Sorensen, E.G. Celius, N.T. Lara, X. Montalban, P. Villoslada, A.M. Silva, M. Marta, I. Leite, B. Dubois, J. Rubio, H. Butzkueven, T. Kilpatrick, M.P. Mycko, K.W. Selmaj, M.E. Rio, M. Sa, G. Salemi, G. Savettieri, J. Hillert, D.A. Compston, Multiple Sclerosis Severity Score: using disability and disease duration to rate disease severity, Neurology 64 (7) (2005) 1144–1151. [18] A.P. Blaber, R.L. Bondar, F. Stein, P.T. Dunphy, P. Moradshahi, M.S. Kassam, R. Freeman, Transfer function analysis of cerebral autoregulation dynamics in autonomic failure patients, Stroke 28 (9) (1997) 1686–1692. [19] A. Hetzel, M. Reinhard, B. Guschlbauer, S. Braune, Challenging cerebral autoregulation in patients with preganglionic autonomic failure, Clin. Auton. Res. 13 (1) (2003) 27–35.
303
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
[28] J. Birns, J. Jarosz, H.S. Markus, L. Kalra, Cerebrovascular reactivity and dynamic autoregulation in ischaemic subcortical white matter disease, J. Neurol. Neurosurg. Psychiatry 80 (10) (2009) 1093–1098. [29] A. Acevedo, C. Nava, N. Arriada, A. Violante, T. Corona, Cardiovascular dysfunction inmultiple sclerosis, Acta Neurol. Scand. 101 (2) (2000) 85–88. [30] G. Vita, M.C. Fazio, S. Milone, A. Blandino, L. Salvi, C. Messina, Cardiovascular autonomic dysfunction in multiple sclerosis is likely related to brainstem lesions, J. Neurol. Sci. 120 (1) (1993) 82–86. [31] D.J. Brooks, S. Redmond, C.J. Mathias, R. Bannister, L. Symon, The effect of orthostatic hypotension on cerebral blood flow and middle cerebral artery velocity in autonomic failure, with observations on the action of ephedrine, J. Neurol. Neurosurg. Psychiatry 52 (8) (1989) 962–966. [32] V. Novak, P. Novak, J.M. Spies, P.A. Low, Autoregulation of cerebral blood flow in orthostatic hypotension, Stroke 29 (1) (1998) 104–111. [33] Y.S. Oh, J.S. Kim, K.S. Lee, Orthostatic and supine blood pressures are associated with white matter hyperintensities in Parkinson disease, J. Mov. Disord. 6 (2) (2013) 23–27. [34] S. Debette, H.S. Markus, The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis, BMJ 341 (2010) c3666. [35] M. Gonul, T. Asil, K. Balci, Y. Celik, N. Turgut, I. Uzunca, Changing cerebral blood flow velocity detected by transcranial Doppler ultrasound during head up tilt in patients with multiple sclerosis, Eur. J. Neurol. 15 (7) (2008) 725–729. [36] R.B. Panerai, Transcranial Doppler for evaluation of cerebral autoregulation, Clin. Auton. Res. 19 (4) (2009) 197–211.
[20] E.D. Gommer, E.G. Martens, P. Aalten, E. Shijaku, F.R. Verhey, W.H. Mess, I.H. Ramakers, J.P. Reulen, Dynamic cerebral autoregulation in subjects with Alzheimer's disease, mild cognitive impairment, and controls: evidence for increased peripheral vascular resistance with possible predictive value, J. Alzheimers Dis. 30 (4) (2012) 805–813. [21] M. Hilz, F. Axelrod, U. Haertl, C. Brown, B. Stemper, Transcranial Doppler sonography during head up tilt suggests preserved central sympathetic activation in familial dysautonomia, J. Neurol. Neurosurg. Psychiatry 72 (5) (2002) 657–660. [22] T. Briebach, J. Laubenberger, P.-A. Fischer, Transcranial Doppler sonographic studies of cerebral autoregulation in Shy-Drager syndrome, J. Neurol. 236 (6) (1989) 349–350. [23] A.C. Chao, H.H. Hu, K.K. Liao, C.M. Chern, Dynamic cerebrovascular regulation in patients with autonomic dysfunction: a transcranial Doppler study, J. Chin. Med. Assoc. 66 (6) (2003) 339–345. [24] K.G. Hollingsworth, D.E. Jones, R. Taylor, J. Frith, A.M. Blamire, J.L. Newton, Impaired cerebral autoregulation in primary biliary cirrhosis: implications for the pathogenesis of cognitive decline, Liver Int. 30 (6) (2010) 878–885. [25] N. Nasr, M. Czosnyka, F. Arevalo, H. Hanaire, B. Guidolin, V. Larrue, Autonomic neuropathy is associated with impairment of dynamic cerebral autoregulation in type 1 diabetes, Auton Neurosci-Basic 160 (1–2) (2011) 59–63. [26] Y.C. Tzeng, S.J. Lucas, G. Atkinson, C.K. Willie, P.N. Ainslie, Fundamental relationships between arterial baroreflex sensitivity and dynamic cerebral autoregulation in humans, J. Appl. Physiol. 108 (5) (2010) 1162–1168. [27] J.M. Serrador, F.A. Sorond, M. Vyas, M. Gagnon, I.D. Iloputaife, L.A. Lipsitz, Cerebral pressure-flow relations in hypertensive elderly humans: transfer gain in different frequency domains, J. Appl. Physiol. 98 (1) (2005) 151–159.
304
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Cerebral autoregulation is preserved in multiple sclerosis patients a,⁎,1
a,b,1
a
c
MARK b
Daniel Ferreira , Pedro Castro , Gonçalo Videira , João Pedro Filipe , Rosa Santos , Maria José Sáb,d, Elsa Azevedoa,b, Pedro Abreua,b a
Department of Clinical Neurosciences and Mental Health, Faculty of Medicine of University of Porto, 4200-319 Porto, Portugal Department of Neurology, São João Hospital Center, 4200-319 Porto, Portugal Department of Neuroradiology, Hospital Center São João, 4200-319 Porto, Portugal d Faculty of Health Sciences, University Fernando Pessoa, 4249-004 Porto, Portugal b c
A R T I C L E I N F O
A B S T R A C T
Keywords: Multiple sclerosis Cerebral autoregulation Autonomic nervous system Heart rate variability Spontaneous baroreflex sensitivity Magnetic resonance imaging
Multiple sclerosis (MS) is an inflammatory disease that may also be associated with vascular dysfunction. One master component of vascular regulation is cerebral autoregulation (CA). We aimed to investigate the integrity of CA in MS patients and study its relationship with autonomic dysfunction (AD), magnetic-resonance-imaging (MRI) lesion load and hemodynamic parameters. We enrolled 20 relapsing-remitting MS and 20 healthy subjects. CA was assessed by transfer function analysis parameters (coherence, gain and phase), as obtained in the very low, low and high-frequency domains (VLF, LF, HF, respectively). We evaluated the autonomic parameters heart rate variability and spontaneous baroreflex sensitivity (BRS). There were no significant differences in CA parameters between MS and controls (p > 0.05). Lesion load was not correlated with any CA parameter. LF gain was positively correlated with BRS in both groups (MS: p = 0.017; controls: p = 0.025). Brainstem lesion load in MS was associated with higher systolic blood pressure (SBP; p = 0.009). Our findings suggest that CA is preserved in our MS cohort. On the other hand, AD in MS patients with brainstem lesions could contribute to the increase of supine SBP. Whether this systemic deregulation could contribute to disease burden remains to be investigated.
1. Introduction Multiple sclerosis (MS) is an autoimmune, demyelinating disease of the central nervous system, where chronic inflammation and axonal injury lead to deficits in motor, sensory, cerebellar and cognitive functions [1]. Cardiovascular (CV) autonomic deregulation is also present in these patients, affecting both the sympathetic and parasympathetic nervous systems, which innervate cerebral vessels [2–4]. Additionally, a higher incidence of cerebrovascular events is present in MS patients in comparison to sex- and age-matched controls, suggesting that vascular dysfunction plays a role throughout the disease course [5,6]. Smoking is a major CV risk factor, as well as an environmental factor for MS development; however, widespread cerebral hypoperfusion seems to be the key factor in predisposing MS patients to ischemic brain lesions [7]. Nevertheless, most of the etiological mechanisms underlying these events remain unclear. Indeed, an adequate blood perfusion is crucial to ensure brain metabolic demand. This mechanism is maintained by three major
⁎
1
components: cerebral autoregulation (CA), cerebrovascular reactivity (VR) and neurovascular coupling [8]. VR was proven to be impaired in MS patients in several studies, even though there are some contradictory results [9,10]. Regarding CA, only one study was conducted, and it showed no significant differences in the autoregulatory mean velocity index between MS patients and controls; however, an autoregulation impairment was described after modeling variation in the cerebral blood flow velocity (CBFV) associated with a 1-mm Hg increase in the mean arterial blood pressure [11]. Thus, it is still unclear whether the CA mechanism remains intact in MS or if it has any relation to vascular and autonomic impairment, and, if CA deregulation plays a role in MS lesion burden. It is worth investigating CA role in MS because we know that it is both impaired by dysautonomia [12,13] and involved in cerebrovascular pathology [14]. Our main objective was to investigate the status of CA in patients with MS. Also, we aimed to study its relationship with the magnitude of autonomic dysfunction (AD) by applying a battery of cardiovascular tests (Ewing battery) and methods to quantify the autonomic system integrity (heart rate variability – HRV – and spontaneous baroreflex
Corresponding author at: Dept. Neurology, Centro Hospitalar de São João, Alameda Professor Hernâni Monteiro, 4200-319 Porto, Portugal. E-mail address: [email protected] (D. Ferreira). Both authors contribute equally to this work.
http://dx.doi.org/10.1016/j.jns.2017.09.009 Received 2 July 2017; Received in revised form 23 August 2017; Accepted 6 September 2017 Available online 08 September 2017 0022-510X/ © 2017 Elsevier B.V. All rights reserved.
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
sensitivity – BRS). MRI brain lesion load and its correlation with CA and hemodynamic parameters were evaluated as well. We hypothesized that an impaired CA in MS patients might play a role in the disease burden.
2.3. Cerebral autoregulation, spontaneous baroreflex sensitivity and heart rate variability measurements Data was visually inspect for artifact removal. Extra-beats were removed by linear interpolation. Beats were marked on R-waves of ECG. For each heartbeat, systolic, mean and diastolic values of arterial BP (SBP, MBP, and DBP, respectively) and CBFV were calculated and visually corrected if needed. CA parameters coherence, gain and phase were calculated using transfer function analysis, as recommended [15]: time-series of spontaneous oscillations in mean CBFV and MBP were interpolated at 10 Hz with a third-order polynomial spline; then an averaged periodogram was calculated by Welch method with Hanning window of window length of 102 s; 50% superposition; and triangular three-point window spectral smoothing filter. Coherence was calculated between input auto-spectra of ABP over cross-spectra of CBFV/ABP and transfer functions of phase and gain were determined by dividing the cross-spectrum by the input autospectrum. Values were reported in three frequency bands: very low (VLF: 0.02–0.07 Hz), low (LF 0.07–0.20 Hz) and high (HF: 0.20–0.50 Hz). The BRS was evaluated in both the time and frequency domains. In the time domain, using the cross-correlation method xBRS, in which the regression slope between SBP and RRi was computed [16]. In resume, xBRS computes the correlation between beat-to-beat systolic blood pressure and normal RRi interval, resampled at 1 Hz, in a sliding 10 s window, with delays of 0–5 s for interval. The delay with the greatest positive correlation is selected and, when significant at p < 0.01, slope and delay are recorded as one xBRS value. The mean value of xBRS is finally calculated. In the frequency domain, the baroreceptor gain (BRG) was calculated based on the cross-correlation gain between the power spectral densities of SBP and RRi. Spectral analysis from timeseries of SBP and RRi were achieved with the same parameters as in described in to access CA. TFA parameter gain was calculated by dividing the cross-spectrum by the input auto-spectrum. BRG was obtained in the band of low frequencies (LF: 0.04–0.15 Hz). xBRS determination of BRS maybe more reliable in patients with low BRS due to autonomic impairment since it is facilitated by the absence of thresholds for pressure and interval variation. Since MS patients might have disautonomia we decided to include to different methods that could strengthen and confirm significant correlations. The frequency domain was used to assess HRV. The fast Fourier transformation was applied to the RRi tachogram, and a spectrum of the HRV signal was obtained. Again, this spectral analysis followed the same parameters described to achieve averaged periodogram described in CA. The following parameters were calculated: total power spectrum of RRi and its low (LF: 0.04–0.15 Hz) and high (HF: 0.15–0.04 Hz) frequency ranges.
2. Materials and methods This matched case-control study was performed in São João Hospital Center in Porto, Portugal. The local institutional ethic committee approved the study and every participant gave written and signed consent. 2.1. Population studied From our MS outpatient clinic, we randomly selected 20 MS patients to participate (12 females). Inclusion criteria were a definite diagnosis of MS, according to the 2010 McDonald criteria, and age above 18 years old. Only relapsing-remitting MS patients were engaged and those who experienced a relapse or had been submitted to corticosteroids treatment within the previous eight weeks were excluded. The control group was selected from the hospital and university staff and consisted of 20 age- and sex-matched healthy subjects. Exclusion criteria were the presence of comorbidities that could affect autonomic function, as known cardiovascular diseases (e.g., myocardial infarction, heart failure, atrial fibrillation) and dysautonomias (e.g., Parkinsonism, multiple system atrophy, diabetes, polyneuropathy). All patients were evaluated by an MS specialist within one month before enrollment, and the following data was recorded: age, gender, neurological examination, Expanded Disability Status Scale (EDSS) score, and duration of disease. 2.2. Experimental design: monitoring and autonomic test protocol Before evaluation, patients and controls were instructed to pause vasoactive drugs in the previous 24 h and abstain from drinking coffee and alcohol or smoke in the previous 12 h. Measurements were performed in a quiet room with a temperature around 22 °C. CBFV was recorded bilaterally from M1 segment of the middle cerebral arteries (MCA), at a depth of 50–55 mm, with 2 MHz monitoring probes held with a headband. (Doppler BoxX, DWL, Singen, Germany). Blood pressure (BP) was continuously monitored in the patient's non-dominant hand using the noninvasive finger cuff Finapres (model 2300; Ohmeda, Englewood, CO, USA). Heart rate and R-R intervals (RRi) were assessed from a 3‑lead ECG. End-tidal carbon dioxide was recorded with a capnograph (Respsen Nonin, Amsterdam, Netherlands) by nasal cannula. All data were synchronized and digitally recorded at 400 Hz with with Powerlab (AD Instruments, Oxford, UK) for offline analysis. Autonomic testing was explained to the volunteers. Maximum contraction strength with the dominant hand was measured using a dynamometer. Patients lied supine for 30 min before evaluation. Assessment of CA, baroreflex function and short-term heart rate variability (HRV) was based on data from a 10-min period of resting in the supine position with uncrossed legs. Afterward, the autonomic tests protocol was performed as described elsewhere [3], based on the Ewing battery. In brief, maximum contraction strength in dominant hand (or non-paretic hand if dominant was affected) was measured with a dynamometer, the subject was positioned supine and rested for 30 min. Next, we calculated: expiratory–to–inspiratory amplitude ratio of HR (E:I ratio) during synchronized deep breathing at 6 cycles/min; the Valsalva ratio, i.e., averaged HR ratio between II and IV phases of three Valsalva maneuvers; the quotient of RRi around 30th beat by that at the 15th after standing; the difference between the systolic BP (SBP) at rest and after standing during 5 min to assess orthostatic hypotension; and diastolic BP (DBP) response to handgrip isometric work at 30% of his maximum strength for 5 min. A final total score (ETS) was obtained by adding 0 if the tests were normal, 1 if borderline and 2 if abnormal [3].
2.4. MRI The brain MRI had been obtained within one year before enrollment. The spinal MRI was the last imaging to be performed, and none of the subjects experienced a relapse after imaging. We assessed the presence of T2-weighted lesions in the following central autonomic network (CAN) structures: insula, anterior cingulate cortex, hypothalamus, amygdala, periaqueductal gray matter, parabrachial nucleus, nucleus of the solitary tract, ventrolateral reticular formation of the medulla and medullary raphe. By adding 1 point for each affected structure, we calculated a CAN score for each patient. The number of T2-weighted lesions in supratentorial areas, brainstem, cerebellum, and spinal cord was calculated, as well as the total number. MRI protocols were performed in 1.5T (Siemens MAGNETOM SymphonyTim syngo) and 3T (Siemens MAGNETOM TrioTim syngo) scanners. These imaging protocols included, at least, coronal and axial T1-weighted images (511750 ms/8.6–8.7 ms [TR/TE]), spin-echo or fast spin-echo axial proton density images (3770–4000 ms/11–22 ms [TR/TE]), T2-weighted images (3770–4000 ms/88–106 ms [TR/TE]) and axial and sagittal fluid-attenuated inversion-recovery (FLAIR) images (8000-9000 ms/ 299
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Table 1 Individual clinical data of MS patients. ID
Gender
Age
Disease duration
EDSS
medication
ARR
Brainstem relapse
Spinal relapse
MSSS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
F F F F M M M F M M M F F F M F F M F F
33 27 23 45 32 48 35 46 32 48 27 48 47 44 35 40 32 35 40 32
4 1 2 25 5 2 5 6 10 19 6 3 7 6 7 6 6 0 0 5
1.0 1.0 0.0 2.0 1.0 2.5 1.0 1.5 0.0 2.5 1.0 1.5 1.5 1.0 1.0 1.0 1.5 1.0 1.0 1.5
Interferon-β1a Interferon-β1a Interferon-β1a Interferon-β1a Interferon-β1a Glatiramer acetate Natalizumab Natalizumab Natalizumab Glatiramer Natalizumab Glatiramer acetate Interferon-β1a Interferon-β1a Fingolimod Dimethyl fumarate Interferon-β1a Interferon-β1a Glatiramer acetate Natalizumab
0.25 0.00 0.00 0.33 0.00 0.00 0.20 0.50 0.00 0.00 0.50 0.43 0.42 0.33 0.42 0.66 0.66 0.00 0.00 0.40
Yes No No Yes No No Yes Yes No Yes Yes No Yes Yes Yes No Yes No No Yes
No No No Yes No No Yes No No Yes Yes Yes No Yes Yes No Yes No No No
1.45 2.44 0.45 0.77 1.28 6.64 1.28 2.33 0.19 1.39 1.13 3.34 2.10 1.13 1.04 1.13 2.33 n.a n.a 2.60
F: female; M: male; EDSS: Expanded Disability Status Scale; ARR: annualised relapse rate; MSSS: Multiple Sclerosis Severity Scale; n.a: not applicable.
3. Results
91–93 ms [TR/TE]), all with 5 mm section thickness.
In the healthy control group, we included 20 subjects with a median age of 36 years. Clinical features of the 20 MS patients are presented in Table 1. Of the 20 patients, 12 were women. MS patients presented a median age of 36 years, EDSS of 1, disease duration of 5.5 years and Multiple Sclerosis Severity Score (MSSS) [17] of 1.3.
2.5. Statistical analyses The normality of the variables was determined by Shapiro-Wilk test. No statistically significant differences (p < 0.05) were detected between the right and left MCA hemodynamic values. Therefore, both values were averaged and used in subsequent analysis. Student t-test or Mann-Whitney test were used to compare hemodynamic characteristics and CA parameters between the two groups. The correlations between two variables were determined by Pearson or Spearman rho coefficients, and a multiple linear regression was performed for posterior multivariate analysis. Post-hoc analysis was performed with ANCOVA analysis to explore the influence of age groups (median as cut-off), sexes and MS vs control over each of the dependent variables. All calculations were performed using the software IMS SPSS Statistics v21™.
3.1. Hemodynamic parameters: comparing the MS patients with the control group Table 2 compares the hemodynamic characteristics between the MS and healthy control groups. No significant differences were found for any of the parameters (p < 0.05). 3.2. Cerebral autoregulation: comparing the MS patients with the control group Transfer function analysis parameters for CA are presented in Table 3. The comparison between the two groups revealed no significant differences (p < 0.05) in any of the CA parameters. A tendency for higher values of phase in the HF band was found in the
Table 2 Comparison of hemodynamic characteristics between MS and control group.
BP, mm Hg BP spectral power, mm Hg2 BP covariance, % Systolic BP, mm Hg CBFV, cm·s− 1 Right Left Average CBFV spectral power, cm2·s− 1 Right Left Average CBFV covariance, % Right Left Average HR, bpm HR covariance, % Mean EtCO2, mm Hg EtCO2 covariance, %
Control group
MS group
p value
75 ± 15 16 ± 9 5 ± 2 115 ± 20
75 ± 13 13 ± 6 4 ± 1 117 ± 16
0.940 0.281 0.257 0.733
67 ± 14 68 ± 14 67 ± 14
66 ± 11 72 ± 17 69 ± 12
0.841 0.445 0.676
16 ± 13 16 ± 15 16 ± 14
15 ± 10 17 ± 12 16 ± 10
0.989 0.640 0.883
5 ± 2 5 ± 1 5 ± 1 68 ± 10 6 ± 3 32 ± 10 8 ± 11
5 ± 1 5 ± 1 5 ± 1 70 ± 8 5 ± 2 37 ± 6 3 ± 2
0.620 0.820 0.718 0.411 0.086 0.264 0.081
Table 3 Comparison of cerebral autoregulation parameters between MS and control group. CA parameters Coherence, a.u VLF LF HF Gaina, %.mm Hg− 1 VLF LF HF Phase, radians VLF LF HF
Control group
MS group
P value
0.38 ± 0.18 0.64 ± 0.15 0.58 ± 0.12
0.32 ± 0.11 0.60 ± 0.16 0.65 ± 0.14
0.151 0.429 0.100
0.91 ± 0.42 1.30 ± 0.47 1.62 ± 0.49
0.82 ± 0.33 1.31 ± 0.36 1.57 ± 0.34
0.221 0.779 0.904
0.90 ± 0.36 0.69 ± 0.23 0.09 ± 0.17
0.96 ± 0.44 0.58 ± 0.14 -0.03 ± 0.16
0.620 0.069 0.059
Values are given in mean ± SD (side deviation); a.u: arbitrary units; VLF: very low frequencies (0.02–0.07 Hz); LF: low frequencies (0.07–0.20 Hz); HF: high frequencies (0.20–0.50 Hz). a None normalized gain (cm·s− 1·mm Hg− 1) also showed no differences between groups (data not shown).
Values are given in mean ± SD (side deviation); BP: blood pressure; CBFV: cerebral blood flow velocity; HR: heart rate; EtCO2: end-tidal carbon dioxide.
300
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Table 4 Correlations between systemic hemodynamics, autonomic tests and cerebral autoregulation (CA) parameters in the control group. CA parameters
Hemodynamics
Gain, %.mm Hg− 1 VLF LF HF Phase, radians VLF LF HF
Ewing battery
BRS
HRV
SBP
DBP
Total score
xBRS
BRG
LF
HF
Total Power
0.046 − 0.391 − 0.482⁎
0.442 −0.263 −0.137
− 0.041 − 0.373 − 0.078
0.211 0.499⁎ 0.409
0.278 0.472⁎ 0.400
0.367 0.200 0.313
0.364 0.132 0.205
0.182 0.146 0.224
0.156 0.379 − 0.153
0.102 0.385 0.034
− 0.045 0.194 − 0.276
0.074 − 0.131 − 0.072
−0.247 −0.026 −0.322
− 0.116 0.203 − 0.367
− 0.051 − 0.027 − 0.226
−0.217 −0.041 −0.241
SBP: systolic blood pressure; DBP: diastolic blood pressure; BRS: baroreflex sensitivity; xBRS: BRS cross correlation in time domain; BRG: transfer function gain of BRS; HRV: heart rate variability in low (LF: 0.04–0.15 Hz) and high (HF: 0.15–0.40 Hz) frequencies; CA parameters in very low (VLF: 0.02–0.07 Hz), low (LF: 0.07–0.20 Hz) and high (HF: 0.20–0.50 Hz) frequencies. ⁎ p < 0.05 significance of Pearson or Spearman's correlation coefficients.
Table 5 Correlations between systemic hemodynamics, autonomic tests and cerebral autoregulation (CA) parameters in MS group. CA parameters
Gain, %.mm Hg VLF LF HF Phase, radians VLF LF HF
Hemodynamics
Ewing battery
BRS
HRV
SBP
DBP
Total score
xBRS
BRG
LF
HF
Total Power
0.424 −0.454⁎ −0.375
− 0.027 − 0.186 − 0.252
− 0.019 − 0.170 − 0.095
0.065 0.526⁎ 0.457⁎
0.156 0.609⁎⁎ 0.459⁎
−0.148 0.045 0.111
0.273 0.677⁎⁎ 0.568⁎⁎
0.073 0.418 0.404
−0.329 0.137 0.356
0.118 0.305 0.061
0.444 0.189 0.246
0.002 0.060 − 0.549⁎
−0.075 −0.071 −0.433
0.167 0.261 −0.256
−0.018 −0.356 −0.200
0.058 −0.148 −0.335
−1
SBP: systolic blood pressure; DBP: diastolic blood pressure; BRS: baroreflex sensitivity; xBRS: BRS cross correlation in time domain; BRG: transfer function gain of BRS; HRV: heart rate variability in low (LF: 0.04–0.15 Hz) and high (HF: 0.15–0.40 Hz) frequencies; CA parameters in very low (VLF: 0.02–0.07 Hz), low (LF: 0.07–0.20 Hz) and high (HF: 0.20–0.50 Hz) frequencies. ⁎ p < 0.05 significance of Pearson or Spearman's correlation coefficients. ⁎⁎ p < 0.01 significance of Pearson or Spearman's correlation coefficients.
control group (0.09 ± 0.17 vs. − 0.03 ± 0.16; p = 0.059). Still, CA is not believed to operate in this frequency range [15].
No significant relationship was found for the CAN score or for the total number of lesions and involvement of brain, cerebellum and spinal cord. The exception was the number of brainstem lesions, which was strongly associated with a higher SBP (r = 0.628; p = 0.005) and with the VLF gain (r = 0.474; p = 0.047). Subsequent multivariate analysis using logistic regression analysis was conducted to study this correlation. It showed no significant relation between brainstem lesion burden and the VLF gain function, but a higher number of lesions was still associated with a higher systolic BP (β = 0.552; p = 0.009), as presented by a linear regression function in Fig. 1. Different averaged values of SBP regarding the presence or absence of brainstem lesions are plotted in Fig. 2.
3.3. Cerebral autoregulation correlation with hemodynamics and autonomic tests The association between systemic hemodynamics, autonomic tests (ETS, BRS and HRV) and CA parameters was calculated for the control and MS groups and is presented in Tables 4 and 5, respectively. In the control group, CA parameters did not correlate with ETS and HRV. Regarding BRS, a positive correlation was found with LF gain for xBRS (r = 0.499; p = 0.025) and BRG (r = 0.472; p = 0.036). Also, we found a negative correlation between HF gain and SBP (r = −0.482; p = 0.031). In the MS group, we found no correlation between ETS and CA parameters as well. Concerning BRS, xBRS was positively correlated with the gain parameter in the LF (r = 0.526; p = 0.017) and HF (r = 0.457; p = 0.043) bands and negatively with the HF Phase (r = − 0.549; p = 0.012). BRG was also found to be positively correlated with gain in the LF (r = 0.609; p = 0.004) and HF (r = 0.459; p = 0.042) bands. Additionally, SBP was negatively correlated with LF gain (r = − 0.454; p = 0.044) and HRV was found to be associated with gain in the LF (r = 0.677; p = 0.001) and HF (r = 0.568; p = 0.009) bands.
3.5. Age and gender effects Post-hoc analysis was performed with ANCOVA analysis to explore each of the dependent variables (CA measures, xBRS and RR total power) adjusting for gender, age strata (< 37 vs ≥37 years-old; median) and group (MS vs control). We found that males (6.7 ± 2.2 vs 10.7 ± 6.0 mm Hg/ms, p = 0.021) and older individuals (7.7 ± 1.9 vs 15 ± 10.1 mm Hg/ms, p = 0.002) tend to have lower xBRS similarly between groups. HRV was also reduced in older individuals (RR total power 635 ± 326 vs 2323 + 2240 ms2, p = 0.014) irrespective of gender or group. CA measures did not differ in gender and age (p > 0.05), which is in accordance with the results of a larger healthy cohort we published [8]. In what concerns the relationship between number of brain stem lesions and SBP we found that was not disturbed by gender or age (p > 0.05)
3.4. MS group: correlation between MRI lesion burden, hemodynamics and CA measurements Table 6 presents the correlation coefficient between lesion burden and cerebrovascular and systemic hemodynamics in the MS patients. 301
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Table 6 Correlations between MRI lesions in the CNS of MS patients and measurements of hemodynamics and CA parameters. CAN scorea
Hemodynamics SBP, mm Hg DBP, mm Hg CBFV, cm·sec− 1 CA parameters Gain, %.mm Hg− 1 VLF LF HF Phase, radians VLF LF HF
Number of lesions Brain
Brainstem
Cerebellum
Spinal cord
Total
0.306 −0.147 0.112
− 0.189 0.027 − 0.044
0.628⁎⁎ 0.099 0.217
0.032 0.287 0.061
0.253 0.130 0.121
−0.115 −0.083 −0.034
0.227 0.146 0.015
0.254 0.422 0.247
0.474⁎ 0.165 − 0.055
0.184 0.343 0.255
0.394 0.022 −0.145
0.065 0.450 0.255
−0.413 −0.326 0.295
0.078 − 0.189 0.328
− 0.152 0.114 0.420
0.172 0.138 0.067
−0.382 −0.270 0.298
0.078 −0.287 0.275
CAN: central autonomic network; SBP: systolic blood pressure; DBP: diastolic blood pressure; CBFV: cerebral blood flow velocity; VLF: very low frequencies (0.02–0.07 Hz); LF: low frequencies (0.07–0.20 Hz); HF: high frequencies (0.20–0.50 Hz). ⁎ p < 0.05 significance of Pearson's correlation coefficient. ⁎⁎ p < 0.01 significance of Pearson's correlation coefficient. a CAN score was calculated as described elsewhere [3] by adding 1 point for each MRI T2-weighted lesion in the white matter adjacent to CAN structures (insula, anterior cingulate cortex, hypothalamus, amygdala, periaqueductal gray matter, parabrachial nucleus, nucleus of the solitary tract, ventrolateral reticular formation of the medulla and medullary raphe).
4. Discussion
relationship between HRV in the HF band and a worst CA response, reflected by a higher LF gain. To our knowledge, this is the first time this association is documented in MS patients. The presence of AD in this MS cohort it detailed in a previous publication [3], in which we documented lower HRV indexes and BRS in MS compared to healthy subjects. CAN lesions does not seem to justify this impairment. Our study also revealed a positive correlation between supine systolic BP and lower LF gain, indicating that CA adjusted itself adequately to higher levels of BP in our patients. In fact, Serrador et al. reported that hypertensive patients demonstrated a better attenuation of cerebral blood flow fluctuations in response to BP changes in LF gain, possibly revealing a permanent remodeling of the cerebrovasculature [27]. Also, in patients with small vessel disease, a positive linear relationship between mean 24-h BP and the autoregulatory index (ARI) was found [28]. Altogether, this data suggest that MS patients may also present a counteractive CA response to increased systemic BP variability. After multivariate analysis adjustment for a possible confounding factor (VLF gain), we found a significant positive correlation between systolic BP in the supine position and brainstem lesion load. In agreement with these findings, patients presenting at least one MRI
Our study main conclusion is that CA is preserved in the mild stages of MS. We found no differences in the CA parameters coherence, gain and phase between the two groups. Our results are in agreement with other studies suggesting a preserved CA mechanism in subjects with AD [18,19] or other neurological disorders with AD, such as Alzheimer disease [20], familial dysautonomia [21] and Shy-Drager Syndrome [22]. However, some studies in patients with autonomic failure [13,23] or with AD and autoimmune diseases [24,25] revealed an impaired CA. These contrasting results can be a consequence of a distinct underlying disease pathophysiology that may ultimately lead to different cerebrovascular responses to AD. Concerning the relationship between AD parameters and CA, we showed that there was an inverse relationship between the efficacy of CA and baroreflex mechanisms. This was shown by the fact that a more tuned baroreflex, i.e., higher values of xBRS or BRG, correlated positively with higher values of CA LF gain, meaning a worse capability of the brain to dampen blood pressure oscillations. This association had already been described [26] in healthy subjects and was also shown in our MS group, suggesting the integrity of a possible compensatory mechanism in these patients. Additionally, we described a similar
Fig. 2. MS patients' averaged systolic blood pressure (BP) in the supine position, regarding the presence or absence of MRI brainstem lesions. *p < 0.05, significant difference in paired t-test within group.
Fig. 1. Linear correlation between the number of MRI brainstem lesions and systolic blood pressure (BP) in the supine position, in MS patients.
302
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
Conflict of interest
brainstem lesion had higher systolic BP comparing to those with no lesions. Higher values of supine BP in these patients might indicate some level of cardiovascular autonomic impairment, which was already known to be correlated with lesions in the brainstem, according to other studies [29,30]. An elevated systolic BP in the supine position, conjugated with the CV dysfunction in these patients might point toward a potential manifestation of supine hypertension (SH) and/or orthostatic hypotension (OH). Some studies in patients with OH have demonstrated a self-adjustment of the autoregulatory range of CA, maintaining a relatively constant cerebral perfusion [31,32]. For instance, patients with Parkinson's disease manifesting SH revealed a higher prevalence of white matter lesions (WML) [33], which were demonstrated to be an important prognostic factor for stroke, cognitive impairment, dementia and death [34]. Even though ARI might not be directly correlated with WML load [28], there seems to be a complex interplay between systemic BP, cerebrovascular hemodynamics and brain lesion load in diseases with AD. Altogether, these factors may contribute to the disease burden in MS. Age and gender are important factors that influence baroreflex function but not CA [8]. We did not find any significant influence of these demographic variables in the alterations of MS patients. MS patients have shown widespread cerebral hypoperfusion which seems not to be secondary to axonal degeneration. Our results are contrary to the theory that this is due to a primary arterial vascular dysfunction namely in terms of autoregulation. Our findings adds more evidence to the claims of some authors that hypo perfusion might be a result of reduced axonal/astrocyte activity and energy metabolism. CA assessment is focused in arterial vessels. Our negative results might be cannot exclude some deregulation in venous level. Some authors suggest slower cerebral venous flow in patients with MS can contribute to disease. Nevertheless, some limitations concerning this study deserve mentioning, and considerations should be made regarding the interpretation of these results. Firstly, AD with progressive impairment of cardiac autonomic balance in MS correlates with higher disease severity and longer duration [4], with a study showing that changes in CBFV were more prominent in patients with EDSS > 2 [35]. Our MS sample had a low median EDSS and disease duration, which did not allow us to study the possibility of CA impairment becoming apparent over the course of the disease and, thus, limits the generalization of these results to the broad spectrum of MS manifestations. Additionally, we cannot rule out the present of more asymptomatic lesions at time of the evaluation since MRI scan was allowed to be within one year from patient assessment. Secondly, even though preliminary studies have not detected differences in the CBFV response to different maneuvers [36], this possibility should be kept in mind when comparing different approaches to measure the influence of autonomic nervous system in cerebral blood flow. In a steady-state situation, we found no significant differences between the two groups. However, stronger challenges to the control of cerebral blood flow, such as squat-stand maneuvers, could be necessary for changes in CA to become apparent. Lastly, it is important to emphasize that merely CBFV was measured by transcranial Doppler, hence only reflecting the changes in cerebral blood flow if MCA's diameter remains constant [15,36], which is expected in the study protocol conditions. Overall, this data indicates that cerebral autoregulation is preserved in our MS cohort. Autonomic dysfunction in multiple sclerosis patients with brainstem lesions could contribute to their increased supine systolic blood pressure. However, these patients seem to present an adaptive mechanism with a decrease in CA gain that dampens possible deleterious effects of higher levels of blood pressure. Whether this systemic deregulation could contribute to disease burden remains to be investigated. Nevertheless, our data suggest that MS physicians should make an effort to carefully monitor cardiovascular risks, especially through the evaluation of blood pressure, in MS patients.
None. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Acknowledgments This work was partially supported by the “Associação para o Estudo da Doença Vascular Cerebral do Hospital de S. João”. The authors acknowledge the work of Isabel Melo for helping with English editing. References [1] A. Compston, A. Coles, Multiple sclerosis, Lancet 372 (9648) (2008) 1502–1517. [2] C.-A. Haensch, J. Jörg, Autonomic dysfunction in multiple sclerosis, J. Neurol. 253 (1) (2006) i3–i9. [3] G. Videira, P. Castro, B. Vieira, J.P. Filipe, R. Santos, E. Azevedo, M.J. Sá, P. Abreu, Autonomic dysfunction in multiple sclerosis is better detected by heart rate variability and is not correlated with central autonomic network damage, J. Neurol. Sci. 367 (2016) 133–137. [4] K. Nasseri, B. TenVoorde, H. Ader, B. Uitdehaag, C. Polman, Longitudinal follow-up of cardiovascular reflex tests in multiple sclerosis, J. Neurol. Sci. 155 (1) (1998) 50–54. [5] N.B. Allen, J.H. Lichtman, H.W. Cohen, J. Fang, L.M. Brass, M.H. Alderman, Vascular disease among hospitalized multiple sclerosis patients, Neuroepidemiology 30 (4) (2008) 234–238. [6] C.F. Christiansen, S. Christensen, D.K. Farkas, M. Miret, H.T. Sorensen, L. Pedersen, Risk of arterial cardiovascular diseases in patients with multiple sclerosis: a population-based cohort study, Neuroepidemiology 35 (4) (2010) 267–274. [7] M. D'Haeseleer, M. Cambron, L. Vanopdenbosch, J. De Keyser, Vascular aspects of multiple sclerosis, Lancet Neurol. 10 (7) (2011) 657–666. [8] J. Madureira, P. Castro, E. Azevedo, Demographic and systemic hemodynamic influences in mechanisms of cerebrovascular regulation in healthy adults, J. Stroke Cerebrovasc. Dis. 26 (3) (2017) 500–508. [9] O. Marshall, H. Lu, J.-C. Brisset, F. Xu, P. Liu, J. Herbert, R.I. Grossman, Y. Ge, Impaired cerebrovascular reactivity in multiple sclerosis, JAMA Neurology 71 (10) (2014) 1275–1281. [10] N. Uzuner, S. Ozkan, N. Cinar, Cerebrovascular reactivity in multiple sclerosis patients, Mult. Scler. 13 (6) (2007) 737–741. [11] Z. Mezei, L. Olah, L. Kardos, R.K. Kovacs, L. Csiba, T. Csepany, Cerebrovascular hemodynamic changes in multiple sclerosis patients during head-up tilt table test: effect of high-dose intravenous steroid treatment, J. Neurol. 260 (9) (2013) 2335–2342. [12] E. Azevedo, P. Castro, R. Santos, J. Freitas, T. Coelho, B. Rosengarten, R. Panerai, Autonomic dysfunction affects cerebral neurovascular coupling, Clin. Auton. Res. 21 (6) (2011) 395–403. [13] P.M. Castro, R. Santos, J. Freitas, R.B. Panerai, E. Azevedo, Autonomic dysfunction affects dynamic cerebral autoregulation during Valsalva maneuver: comparison between healthy and autonomic dysfunction subjects, J. Appl. Physiol. 117 (3) (2014) 205–213. [14] P. Castro, J.M. Serrador, I. Rocha, F. Sorond, E. Azevedo, Efficacy of cerebral autoregulation in early ischemic stroke predicts smaller infarcts and better outcome, Front. Neurol. 8 (2017) 113. [15] J.A. Claassen, A.S. Meel-van den Abeelen, D.M. Simpson, R.B. Panerai, N. International Cerebral Autoregulation Research, Transfer function analysis of dynamic cerebral autoregulation: A white paper from the International Cerebral Autoregulation Research Network, J. Cereb. Blood Flow. Metab. 36 (4) (2016) 665–680. [16] B.E. Westerhof, J. Gisolf, W.J. Stok, K.H. Wesseling, J.M. Karemaker, Time-domain cross-correlation baroreflex sensitivity: performance on the EUROBAVAR data set, J. Hypertens. 22 (7) (2004) 1371–1380. [17] R.H. Roxburgh, S.R. Seaman, T. Masterman, A.E. Hensiek, S.J. Sawcer, S. Vukusic, I. Achiti, C. Confavreux, M. Coustans, E. le Page, G. Edan, G.V. McDonnell, S. Hawkins, M. Trojano, M. Liguori, E. Cocco, M.G. Marrosu, F. Tesser, M.A. Leone, A. Weber, F. Zipp, B. Miterski, J.T. Epplen, A. Oturai, P.S. Sorensen, E.G. Celius, N.T. Lara, X. Montalban, P. Villoslada, A.M. Silva, M. Marta, I. Leite, B. Dubois, J. Rubio, H. Butzkueven, T. Kilpatrick, M.P. Mycko, K.W. Selmaj, M.E. Rio, M. Sa, G. Salemi, G. Savettieri, J. Hillert, D.A. Compston, Multiple Sclerosis Severity Score: using disability and disease duration to rate disease severity, Neurology 64 (7) (2005) 1144–1151. [18] A.P. Blaber, R.L. Bondar, F. Stein, P.T. Dunphy, P. Moradshahi, M.S. Kassam, R. Freeman, Transfer function analysis of cerebral autoregulation dynamics in autonomic failure patients, Stroke 28 (9) (1997) 1686–1692. [19] A. Hetzel, M. Reinhard, B. Guschlbauer, S. Braune, Challenging cerebral autoregulation in patients with preganglionic autonomic failure, Clin. Auton. Res. 13 (1) (2003) 27–35.
303
Journal of the Neurological Sciences 381 (2017) 298–304
D. Ferreira et al.
[28] J. Birns, J. Jarosz, H.S. Markus, L. Kalra, Cerebrovascular reactivity and dynamic autoregulation in ischaemic subcortical white matter disease, J. Neurol. Neurosurg. Psychiatry 80 (10) (2009) 1093–1098. [29] A. Acevedo, C. Nava, N. Arriada, A. Violante, T. Corona, Cardiovascular dysfunction inmultiple sclerosis, Acta Neurol. Scand. 101 (2) (2000) 85–88. [30] G. Vita, M.C. Fazio, S. Milone, A. Blandino, L. Salvi, C. Messina, Cardiovascular autonomic dysfunction in multiple sclerosis is likely related to brainstem lesions, J. Neurol. Sci. 120 (1) (1993) 82–86. [31] D.J. Brooks, S. Redmond, C.J. Mathias, R. Bannister, L. Symon, The effect of orthostatic hypotension on cerebral blood flow and middle cerebral artery velocity in autonomic failure, with observations on the action of ephedrine, J. Neurol. Neurosurg. Psychiatry 52 (8) (1989) 962–966. [32] V. Novak, P. Novak, J.M. Spies, P.A. Low, Autoregulation of cerebral blood flow in orthostatic hypotension, Stroke 29 (1) (1998) 104–111. [33] Y.S. Oh, J.S. Kim, K.S. Lee, Orthostatic and supine blood pressures are associated with white matter hyperintensities in Parkinson disease, J. Mov. Disord. 6 (2) (2013) 23–27. [34] S. Debette, H.S. Markus, The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis, BMJ 341 (2010) c3666. [35] M. Gonul, T. Asil, K. Balci, Y. Celik, N. Turgut, I. Uzunca, Changing cerebral blood flow velocity detected by transcranial Doppler ultrasound during head up tilt in patients with multiple sclerosis, Eur. J. Neurol. 15 (7) (2008) 725–729. [36] R.B. Panerai, Transcranial Doppler for evaluation of cerebral autoregulation, Clin. Auton. Res. 19 (4) (2009) 197–211.
[20] E.D. Gommer, E.G. Martens, P. Aalten, E. Shijaku, F.R. Verhey, W.H. Mess, I.H. Ramakers, J.P. Reulen, Dynamic cerebral autoregulation in subjects with Alzheimer's disease, mild cognitive impairment, and controls: evidence for increased peripheral vascular resistance with possible predictive value, J. Alzheimers Dis. 30 (4) (2012) 805–813. [21] M. Hilz, F. Axelrod, U. Haertl, C. Brown, B. Stemper, Transcranial Doppler sonography during head up tilt suggests preserved central sympathetic activation in familial dysautonomia, J. Neurol. Neurosurg. Psychiatry 72 (5) (2002) 657–660. [22] T. Briebach, J. Laubenberger, P.-A. Fischer, Transcranial Doppler sonographic studies of cerebral autoregulation in Shy-Drager syndrome, J. Neurol. 236 (6) (1989) 349–350. [23] A.C. Chao, H.H. Hu, K.K. Liao, C.M. Chern, Dynamic cerebrovascular regulation in patients with autonomic dysfunction: a transcranial Doppler study, J. Chin. Med. Assoc. 66 (6) (2003) 339–345. [24] K.G. Hollingsworth, D.E. Jones, R. Taylor, J. Frith, A.M. Blamire, J.L. Newton, Impaired cerebral autoregulation in primary biliary cirrhosis: implications for the pathogenesis of cognitive decline, Liver Int. 30 (6) (2010) 878–885. [25] N. Nasr, M. Czosnyka, F. Arevalo, H. Hanaire, B. Guidolin, V. Larrue, Autonomic neuropathy is associated with impairment of dynamic cerebral autoregulation in type 1 diabetes, Auton Neurosci-Basic 160 (1–2) (2011) 59–63. [26] Y.C. Tzeng, S.J. Lucas, G. Atkinson, C.K. Willie, P.N. Ainslie, Fundamental relationships between arterial baroreflex sensitivity and dynamic cerebral autoregulation in humans, J. Appl. Physiol. 108 (5) (2010) 1162–1168. [27] J.M. Serrador, F.A. Sorond, M. Vyas, M. Gagnon, I.D. Iloputaife, L.A. Lipsitz, Cerebral pressure-flow relations in hypertensive elderly humans: transfer gain in different frequency domains, J. Appl. Physiol. 98 (1) (2005) 151–159.
304