Patients with neurogenic orthostatic hypotension (OH) typically have impaired sympathetic nervous system tone and therefore low levels of upright plasma norepinephrine (NE) (noradrenaline). We report a subset of patients who clinically have typical neurogenic OH but who paradoxically have elevated upright levels of plasma NE. We retrospectively studied 83 OH patients evaluated at the Vanderbilt Autonomic Dysfunction Center between August 2007 and May 2013. Based on standing NE, patients were dichotomized into a hyperadrenergic OH group [hyperOH: upright NE ≥ 3.55 nmol/l (600 pg/ml), n=19] or a non-hyperadrenergic OH group [nOH: upright NE < 3.55 nmol/l (600 pg/ml), n=64]. Medical history and data from autonomic testing, including the Valsalva manoeuvre (VM), were analysed. HyperOH patients had profound orthostatic falls in blood pressure (BP), but less severe than in nOH [change in SBP (systolic blood pressure): −53±31 mmHg compared with −68±33 mmHg, P=0.050; change in DBP (diastolic blood pressure): −18±23 mmHg compared with −30±17 mmHg, P=0.01]. The expected compensatory increase in standing heart rate (HR) was similarly blunted in both hyperOH and nOH groups [84±15 beats per minute (bpm) compared with 82±14 bpm; P=0.6]. HyperOH patients had less severe sympathetic failure as evidenced by smaller falls in DBP during phase 2 of VM and a shorter VM phase 4 BP recovery time (16.5±8.9 s compared with 31.6±16.6 s; P<0.001) than nOH patients. Neurogenic hyperOH patients have severe neurogenic OH, but have less severe adrenergic dysfunction than nOH patients. Further work is required to understand whether hyperOH patients will progress to nOH or whether this represents a different disorder.

CLINICAL PERSPECTIVES

  • We have defined ‘neurogenic hyperOH’ as neurogenic OH in the presence of an upright plasma NE level > 3.55 nmol/l (600 pg/ml).

  • Neurogenic hyperOH occurs in a substantial minority of patients with neurogenic OH and manifests with less severe sympathetic nervous system dysfunction compared to typical neurogenic OH.

  • HyperOH may reflect an earlier stage of classic neurogenic OH, although further studies are required to elucidate the natural history of this disorder.

INTRODUCTION

Orthostatic hypotension (OH) is defined as a reduction in systolic blood pressure (SBP) >20 mmHg or a reduction in diastolic blood pressure (DBP) > 10 mmHg within 3 min of standing [1]. It is an increasingly prevalent clinical problem that is responsible for significant morbidity in the elderly [2,3]. OH is associated with a significant economic burden on the healthcare system with an estimated 160000 hospitalizations annually in the U.S.A. [4]. Whereas ∼50% of these hospitalizations are due to acutely reversible conditions such as dehydration or medication effects, the other half represent chronic and long-standing neurogenic OH related to autonomic nervous system failure [4,5]. Therefore, neurogenic OH is associated with reduced upright plasma norepinephrine (NE) (noradrenaline) levels, a marker of sympathetic nervous system activity [6].

The term ‘hyperadrenergic OH’ (hyperOH) was first coined by Dr David Streeten in 1990 when he described a small group of patients with mild OH and normal or elevated levels of upright plasma NE levels [7]. He did not describe patients with severe neurogenic OH who have elevated upright plasma NE levels (≥3.55 nmol/l or 600 pg/ml) [8].

At our tertiary autonomic disorders referral centre, we have noted that some patients with neurogenic OH have elevated upright plasma NE. This phenomenon has not been previously recognized in neurogenic OH. Furthermore, the clinical features and characteristics of neurogenic OH patients with paradoxically elevated plasma NE have not been described in the literature. We investigated whether symptomatic patients with neurogenic hyperOH could be distinguished from those patients with non-elevated levels of plasma NE [non-hyperadrenergic OH (nOH)] using standard autonomic reflex testing.

MATERIALS AND METHODS

Patient population

Patients evaluated in the Vanderbilt Autonomic Dysfunction Clinic or admitted to the Vanderbilt Clinical Research Center between August 2007 and May 2013 with a diagnosis of chronic autonomic failure were included in the present study if they met the criteria for neurogenic OH [SBP drop ≥ 20 mmHg (or ≥30 mmHg if the supine SBP was ≥160 mmHg) or DBP drop ≥ 10 mmHg; 1] and had undergone autonomic function testing which included a digitally acquired Valsalva manoeuvre (VM) with beat-to-beat blood pressure (BP) recording. The research has been carried out in accordance with the Declaration of Helsinki (2008) of the World Medical Association. Patients with evidence of a square-root wave pattern on VM were excluded as the BP aberrantly stays elevated and constant throughout the strain phases of the VM. This only occurred in three patients. All patients gave their written informed consent and the present study was approved by the Vanderbilt University Institutional Review Board.

Demographics and history of symptoms

We reviewed patient charts and abstracted data for age, height, weight, body mass index (BMI), gender, duration of disease at time of presentation, co-morbidities and medications during presentation. Co-morbid conditions that were noted for analysis included diabetes, obstructive sleep apnoea, multiple system atrophy (MSA) and Parkinson's disease [9]. Medications that were noted for analysis included chronic use of opioid analgesics, benzodiazepines, selective serotonin-reuptake inhibitors (SSRIs), serotonin-NE reuptake inhibitors (SNRIs), fludrocortisone, midodrine, levodopa/carbidopa, β-blockers and α-1 adrenergic antagonists.

Stand test with catecholamines

The ‘stand test’ involved patients lying down for at least 15 min (to allow for a supine steady state), heart rate (HR), SBP and DBP (Dinamap, Critikon Corp) were measured. The patient then stood for 10 min (as tolerated) with HR, SBP and DBP measured at 1, 3, 5 and 10 min standing. Orthostatic changes were calculated as the difference between supine parameters and measurements obtained at the end of standing. Patients unable to stand for 10 min were allowed to sit for the 10 min before blood was sampled. Blood was collected for NE and epinephrine (EPI) (adrenaline) in plastic syringes in both supine and upright positions (after at least 10 min of standing or sitting). Samples were then immediately transferred to chilled vacuum tubes with sodium heparin (BD Biosciences). Plasma was separated by centrifugation at 4°C and stored at −70°C in collection tubes. Clinical samples were assayed within 4 days. Reduced glutathione 6% (Sigma–Aldrich) was added to the Clinical Research Center samples prior to freezing. Concentrations of NE and EPI were measured by HPLC with electrochemical detection following batch alumina extraction [10].

Data acquisition during autonomic function testing including Valsalva manoeuvre

Autonomic function tests were performed while patients were inpatients on the Vanderbilt University Clinical Research Center or in the outpatient setting during their clinic visit. All non-essential vasoactive medications were held on the day of autonomic testing. SBP and DBP were measured continuously by the finger volume clamp method (Nexfin; BMEYE) and intermittently with an automated oscillometric brachial cuff (Vital-Guard 450C, Ivy Biomedical Systems). HR was determined by continuous ECG monitoring (Vital-Guard 450C, Ivy Biomedical Systems). ECG and BP data were digitalized with 14-bit resolution at a 500 and 1000 Hz sample frequency using a WINDAQ data acquisition system (DI720; DATAQ) and processed off-line using custom software written in PV Wave language (PV Wave; Visual Numerics Inc.) by André Diedrich.

Spectral analysis (Table 1)

Data segments of 300 s recorded at the beginning of each autonomic function test session during quiet respiratory breathing and stable resting conditions were used for the calculation of HR variability parameters, BP variability parameters and cardiovagal baroreflex sensitivity (BRS-v). A QRS detection algorithm, modified from Pan and Tompkins [11], was used to generate beat-to-beat R-R interval and BP values and these values were interpolated, low-pass-filtered (cut-off 2 Hz) and resampled at 4 Hz. Linear trends were removed and power spectral density was estimated with the Fast Fourier transform-based Welch algorithm using segments of 256 data points. The power in the low-frequency range (LF: 0.04–0.15 Hz) and high-frequency range (HF: 0.15–0.40 Hz) were calculated following the North American Society of Pacing and Electrophysiology Task Force Guidelines [12]. The HF component of the R-R intervals (RRI-HF) strongly correlates with parasympathetic HR control, assuming that the respiratory rate is between 9 and 24 breaths/min (0.15–0.4 Hz). Similar methods were employed for the continuous SBP signal. The low-frequency band (SBP-LF) was taken as a measure of sympathetic nervous system tone [13]. In the time-domain, we assessed the percentage of consecutive R-R intervals that were 50 ms different from their neighbour, pNN50 [14], as a marker of parasympathetic tone.

Table 1
Spectral and VM parameters

Spectral and VM manoeuvre parameters with their respective abbreviation, definition, and clinical significance.

AbbreviationDefinitionSignificance
RRI-HF R-R interval variability in the high-frequency band Marker of parasympathetic tone 
RRI-LF R-R interval variability in the low-frequency band Marker of parasympathetic tone 
SBP-LF SBP variability in the low-frequency band Marker of sympathetic tone 
pNN50 Percentage of consecutive R-R intervals that are more than 50 ms different from adjacent R-R interval Marker of parasympathetic tone 
PRT Pressure recovery time Marker of sympathetic tone 
BRS-a Adrenergic baroreflex sensitivity Marker of sympathetic tone 
BRS-v Cardiovagal baroreflex sensitivity Marker of parasympathetic tone 
Phase 2 SBP decrement The decrease in SBP during the VM from the start to the lowest point of phase 2 Marker of sympathetic tone 
Phase 2 DBP decrement The decrease in DBP during the VM from the start to the lowest point of phase 2 Marker of sympathetic tone 
AbbreviationDefinitionSignificance
RRI-HF R-R interval variability in the high-frequency band Marker of parasympathetic tone 
RRI-LF R-R interval variability in the low-frequency band Marker of parasympathetic tone 
SBP-LF SBP variability in the low-frequency band Marker of sympathetic tone 
pNN50 Percentage of consecutive R-R intervals that are more than 50 ms different from adjacent R-R interval Marker of parasympathetic tone 
PRT Pressure recovery time Marker of sympathetic tone 
BRS-a Adrenergic baroreflex sensitivity Marker of sympathetic tone 
BRS-v Cardiovagal baroreflex sensitivity Marker of parasympathetic tone 
Phase 2 SBP decrement The decrease in SBP during the VM from the start to the lowest point of phase 2 Marker of sympathetic tone 
Phase 2 DBP decrement The decrease in DBP during the VM from the start to the lowest point of phase 2 Marker of sympathetic tone 

BRS-v was calculated using cross-spectral analysis of the relationships between R-R interval and SBP. Specifically, it was defined as the mean magnitude value of the transfer function in the SBP-LF with negative phase and squared coherence value greater than 0.5 [15].

Valsalva manoeuvre (Table 1)

Baseline SBP and DBP were obtained just prior to initiation of the VM. Patients were asked to maintain an expiratory pressure of at least 30 mmHg for 15 s. Typical VM tracings are presented in Figure 1 for a healthy volunteer, a hyperOH patient and an nOH patient.

Representative VM Tracings

Figure 1
Representative VM Tracings

Continuous BP tracings of a healthy subject (a), a patient with hyperOH (b) and a patient with nOH (c) are shown. See the text or Figure 2 for details of calculation.

Figure 1
Representative VM Tracings

Continuous BP tracings of a healthy subject (a), a patient with hyperOH (b) and a patient with nOH (c) are shown. See the text or Figure 2 for details of calculation.

Phases 1 and 2 are the ‘strain phases’ of the VM with a Valsalva-induced reduction in cardiac venous return and relative hypotension. Phases 3 and 4 are the ‘recovery phases’ of the VM when cardiac venous return normalizes to baseline levels. Phase 1 includes the period from the onset of the VM until the SBP peak (Figure 2). Phase 2 lasts from the end of phase 1 until release of the VM. The lowest BP in phase 2 is the ‘nadir of phase 2’. If SBP and DBP increased from this nadir during this phase, a ‘late phase 2’ (phase 2L) was present. If no phase 2L was present, the ‘phase 2 BP decrement’ was measured until the end of phase 2. Phase 3 begins after Valsalva release and lasts briefly until BP starts to recover, heralding Phase 4. Phase 4 in normal individuals is notable for a profound recovery in BP, termed ‘phase 4 overshoot’, but usually absent from patients with autonomic failure. The DBP was recorded immediately preceding the relevant SBP peak.

Schematic parameterization of the VM

Figure 2
Schematic parameterization of the VM

VM from one of our OH patients is shown with BP displayed on top and expiratory Valsalva pressure on the bottom. Annotations include the baseline (bsl) SBP prior to the VM, the peak SBP during Valsalva phase 1 (1p), the nadir of Valsalva phase 2 (2n) and the nadir of phase 3 (3n). Arrow a represents the Valsalva phase 2 SBP decrement. Arrow b represents the difference between baseline SBP and the end of Valsalva phase 2n. Arrow c represents the difference between baseline SBP and point 3nPRT is measured as the time from the nadir SBP of Valsalva phase 3 until SBP recovers back to baseline level at point (r). BRS-a is equal to the SBP represented by arrow c divided by the PRT.

Figure 2
Schematic parameterization of the VM

VM from one of our OH patients is shown with BP displayed on top and expiratory Valsalva pressure on the bottom. Annotations include the baseline (bsl) SBP prior to the VM, the peak SBP during Valsalva phase 1 (1p), the nadir of Valsalva phase 2 (2n) and the nadir of phase 3 (3n). Arrow a represents the Valsalva phase 2 SBP decrement. Arrow b represents the difference between baseline SBP and the end of Valsalva phase 2n. Arrow c represents the difference between baseline SBP and point 3nPRT is measured as the time from the nadir SBP of Valsalva phase 3 until SBP recovers back to baseline level at point (r). BRS-a is equal to the SBP represented by arrow c divided by the PRT.

Various Valsalva metrics have been reported to be useful in characterizing the severity of adrenergic dysfunction and were measured in the present study (Figure 2). These include the magnitude of phase 2 SBP and DBP decrement relative to the phase 1 peak [16], difference between baseline SBP and the SBP at the end of phase 2 [17], blood pressure recovery time (PRT) [18] and adrenergic baroreflex sensitivity index (BRS-a) [19]. PRT is defined as the amount of time required for the SBP to recover from the nadir of phase 3 to the baseline SBP level [18]. BRS-a is defined as the SBP drop between baseline and the nadir of phase 3 divided by the PRT [19]. The lowest SBP during phase 3 was used to calculate the starting point for the PRT and BRS-a. Other recorded VM parameters included baseline HR, maximum HR during the VM, HR at VM phase 4 recovery, Valsalva ratio (the maximum HR during phase 2 of VM divided by the lowest HR within 30 s after the maximum HR), presence/absence of a VM phase 2L and presence/absence of a VM phase 4 SBP overshoot.

Statistical analyses

Independent two-tailed Student's t tests were used to compare the continuous variables between the hyperOH and nOH groups and, in the case of PRT, between these two parameters and published control data [18]. Fisher's exact tests were used to analyse categorical data. Spearman's correlation was used to the calculate correlation coefficient between standing plasma NE levels against PRT and BRS-a. Data are presented as means ± S.D. P≤0.05 was considered statistically significant. Statistical analyses were performed using SPSS for Windows version 19 (IBM Corp.) and GraphPad Prism version 5.02. GraphPad Prism was used to create the figures.

RESULTS

Demographics (Table 2)

A total of 83 patients met the inclusion criteria. Upright plasma NE levels ranged between 0.14 and 10.87 nmol/l (23 and 1839 pg/ml; Figure 3). Nineteen patients (55% male) had a plasma upright NE level ≥ 3.55 nmol/l (hyperOH). The nOH group consisted of the other 64 patients (57% male). There was no significant difference in age between hyperOH and nOH patients (69±8 years compared with 65±11 years, P=0.1). The BMI was lower in the hyperOH patients than the nOH patients (24.3±3.9 kg/m2 compared with 27.9±4.7 kg/m2, P=0.03), although height (P=0.6) and weight (P=0.1) were not different between groups. Onset of disease symptoms was non-significantly more recent in the hyperOH patients compared with the nOH patients (5.1±3.5 years compared with 7.1±5.4 years, P=0.1).

Table 2
Demographics and posture study

Data are presented as means ± S.D. Continuous data were analysed using Student's t test comparing hyperOH with nOH groups and categorical data were analysed using a Fisher's exact test. Statistically significant.

HyperOHnOHP-value
Total subjects (n19 64 – 
Male gender (%) 11/20 (52%) 37/65 (56%) 0.879 
Age (years) 69±8 65±11 0.128 
Weight (kg) 74±15 81±18 0.135 
Height (cm) 174±12 173±10 0.590 
BMI (kg/m224±4 27±5 0.026* 
Years of disease at presentation (years) 5.1±3.5 7.1±5.4 0.132 
Supine 
HR (bpm) 67±10 71±13 0.217 
SBP (mmHg) 155±32 158±31 0.776 
DBP (mmHg) 83±14 88±14 0.257 
NE (nmol/l) [pg/ml] 2.81±2.03 [475±344] 0.86±0.76 [145±129] 0.001* 
EPI (nmol/l) [pg/ml] 0.110±0.07 [20±13] 0.09±0.10 [17±19] 0.608 
Standing 
HR (bpm) 84±15 82±14 0.625 
SBP (mmHg) 103±38 90±23 0.166 
DBP (mmHg) 66±21 58±11 0.145 
NE (nmol/l) [pg/ml] 6.12±1.99 [1035±336] 1.34±0.85 [226±144] <0.001* 
EPI (nmol/l) [pg/ml] 0.20±0.25 [37±46] 0.18±0.21[33±39] 0.730 
Change from supine to standing 
HR (bpm) 16±10 10±14 0.073 
SBP (mmHg) −53±31 −68±33 0.050* 
DBP (mmHg) −18±23 −30±17 0.011* 
NE (nmol/l) [pg/ml] 3.30±1.48 [559±251] 0.70±0.48 [81±119] <0.001* 
EPI (nmol/l) [pg/ml] 0.09±0.21 [17±38] 0.08±0.17 [16±32] 0.891 
HyperOHnOHP-value
Total subjects (n19 64 – 
Male gender (%) 11/20 (52%) 37/65 (56%) 0.879 
Age (years) 69±8 65±11 0.128 
Weight (kg) 74±15 81±18 0.135 
Height (cm) 174±12 173±10 0.590 
BMI (kg/m224±4 27±5 0.026* 
Years of disease at presentation (years) 5.1±3.5 7.1±5.4 0.132 
Supine 
HR (bpm) 67±10 71±13 0.217 
SBP (mmHg) 155±32 158±31 0.776 
DBP (mmHg) 83±14 88±14 0.257 
NE (nmol/l) [pg/ml] 2.81±2.03 [475±344] 0.86±0.76 [145±129] 0.001* 
EPI (nmol/l) [pg/ml] 0.110±0.07 [20±13] 0.09±0.10 [17±19] 0.608 
Standing 
HR (bpm) 84±15 82±14 0.625 
SBP (mmHg) 103±38 90±23 0.166 
DBP (mmHg) 66±21 58±11 0.145 
NE (nmol/l) [pg/ml] 6.12±1.99 [1035±336] 1.34±0.85 [226±144] <0.001* 
EPI (nmol/l) [pg/ml] 0.20±0.25 [37±46] 0.18±0.21[33±39] 0.730 
Change from supine to standing 
HR (bpm) 16±10 10±14 0.073 
SBP (mmHg) −53±31 −68±33 0.050* 
DBP (mmHg) −18±23 −30±17 0.011* 
NE (nmol/l) [pg/ml] 3.30±1.48 [559±251] 0.70±0.48 [81±119] <0.001* 
EPI (nmol/l) [pg/ml] 0.09±0.21 [17±38] 0.08±0.17 [16±32] 0.891 

The distribution of upright plasma NE levels

Figure 3
The distribution of upright plasma NE levels

The distribution of upright plasma NE into intervals of 0.30 nmol/l (50 pg/ml) is shown for our entire study population (n=83). A line has been drawn at 600 pg/ml to dichotomize our study population into a hyperOH to the left of the line and an nOH to the right of the line. There is a normal distribution to the left of the line, with outliers to the right of this line.

Figure 3
The distribution of upright plasma NE levels

The distribution of upright plasma NE into intervals of 0.30 nmol/l (50 pg/ml) is shown for our entire study population (n=83). A line has been drawn at 600 pg/ml to dichotomize our study population into a hyperOH to the left of the line and an nOH to the right of the line. There is a normal distribution to the left of the line, with outliers to the right of this line.

Stand test with catecholamines (Table 2)

The supine HR (P=0.2), SBP (P=0.8) and DBP (P=0.3) were not different between groups. Maximum standing times were not different between the hyperOH and nOH groups (875±585 s compared with 728±639 s, P=0.375). Standing HR was similar for hyperOH patients and nOH patients [84±15 beats per minute (bpm) compared with 82±14 bpm, P=0.6], as were the standing SBP (P=0.2) and DBP (P=0.1). The orthostatic increase in HR was similar between hyperOH and nOH patients (16±10 bpm compared with 10±14 bpm, P=0.07). Both groups experienced large drops in BP with standing, but the drop in SBP (−53±31 mmHg compared with −68±33 mmHg, P=0.05) and DBP (−18±23 mmHg compared with −30±17 mmHg, P=0.01) were smaller in hyperOH patients than nOH patients.

Supine plasma NE was much greater in the hyperOH group than the nOH group [92.81±2.03 nmol/l (475±344 pg/ml) compared with 0.86±0.76 nmol/l (145±129 pg/ml), P=0.001]. On standing, the plasma NE was considerably higher in the hyperOH group [6.12±1.99 nmol/l (1035±336 pg/ml) compared with 1.34±0.85 nmol/l (226±144 pg/ml), P<0.001].

Co-morbidities and medication use (Table 3)

The prevalence of Parkinson's disease (P=0.4), diabetes mellitus (P=0.3) and obstructive sleep apnoea (P=0.5) were not different between the groups. MSA was exclusively seen in the nOH group with a prevalence of 20%.

Table 3
Comorbidities and medication use

Data are presented as incidence of disease or medication use in each study group. Reported P-values are for Chi-square tests using Fisher's exact method. Statistically significant.

HyperOHnOHP-value
Total subjects (n19 64 – 
Co-morbidities 
Parkinson's disease 3 (16%) 5 (8%) 0.376 
Diabetes 2 (11%) 3 (5%) 0.322 
MSA 0 (0%) 13 (20%) 0.033* 
Obstructive sleep apnoea 2 (11%) 12 (19%) 0.506 
Medication usage 
Opioid analgesic use 3 (16%) 4 (6%) 0.193 
Benzodiazepine use 9 (47%) 5 (8%) <0.001* 
Tricyclic antidepressant use 0 (0%) 1 (2%) 1.000 
Stimulant use 0 (0%) 2 (3%) 1.000 
Levodopa use 3 (16%) 5 (8%) 0.376 
Midodrine use 5 (26%) 31 (48%) 0.116 
Fludrocortisone use 1 (5%) 19 (30%) 0.033* 
SSRI use 7 (37%) 14 (22%) 0.232 
SNRI use 2 (11%) 3 (5%) 0.322 
α-Blocker use 0 (0%) 3 (5%) 1.000 
β-Blocker use 6 (32%) 6 (9%) 0.026* 
HyperOHnOHP-value
Total subjects (n19 64 – 
Co-morbidities 
Parkinson's disease 3 (16%) 5 (8%) 0.376 
Diabetes 2 (11%) 3 (5%) 0.322 
MSA 0 (0%) 13 (20%) 0.033* 
Obstructive sleep apnoea 2 (11%) 12 (19%) 0.506 
Medication usage 
Opioid analgesic use 3 (16%) 4 (6%) 0.193 
Benzodiazepine use 9 (47%) 5 (8%) <0.001* 
Tricyclic antidepressant use 0 (0%) 1 (2%) 1.000 
Stimulant use 0 (0%) 2 (3%) 1.000 
Levodopa use 3 (16%) 5 (8%) 0.376 
Midodrine use 5 (26%) 31 (48%) 0.116 
Fludrocortisone use 1 (5%) 19 (30%) 0.033* 
SSRI use 7 (37%) 14 (22%) 0.232 
SNRI use 2 (11%) 3 (5%) 0.322 
α-Blocker use 0 (0%) 3 (5%) 1.000 
β-Blocker use 6 (32%) 6 (9%) 0.026* 

The use of SNRI medications (P=0.3), SSRI medications (P=0.2), opioid analgesics (P=0.2), levodopa/carbidopa (P=0.4) and α-adrenergic antagonists (P=1.000) were similar between the groups. β-Blockers were used by more hyperOH patients than nOH patients (32% compared with 7%, P=0.03), as were benzodiazepines (47% compared with 9%, P<0.001). In contrast, fludrocortisone use was less prevalent in the hyperOH patients than the nOH patients (5% compared with 30%, P=0.03), whereas midodrine use was similar between both groups (26% compared with 48%, P=0.1).

Spectral analyses (Table 4)

In the time-domain, pNN50 was lower in both hyperOH patients (10.5±24.5%) and nOH patients (6.5±16.3%) than we have previously found in healthy control subjects (36.5±25.5%) [20]. pNN50 did not differ between hyperOH and nOH patients (P=0.4).

Table 4
BRS-v, HR variability and BP variability

Data are presented as means ± S.D. Reported P-values are for Student's t tests comparing hyperOH with nOH. Abbreviations: HF-RRI, RRI power in the high-frequency range; LF-RRI, RRI power in the low-frequency range; pNN50, percentage of adjacent R-R intervals (RRI) greater than 50 ms. The data for healthy subjects that are presented for comparison are from [20].

HyperOHnOHP-valueHealthy subjects
HR variability 
pNN50 (%) 10.5±24.5 6.5±16.3 0.403 36.5±25.5 
LF-RRI (ms2364±719 220±610 0.398 1144±1093 
HF-RRI (ms2297±678 165±517 0.375 2147±3688 
BP variability 
LF-SYS (mmHg29.0±13.0 5.0±6.0 0.229 6.7±5.4 
BRS-v (ms/mmHg) 6.0±6.2 4.1±3.5 0.113 12.5±7.6 
HyperOHnOHP-valueHealthy subjects
HR variability 
pNN50 (%) 10.5±24.5 6.5±16.3 0.403 36.5±25.5 
LF-RRI (ms2364±719 220±610 0.398 1144±1093 
HF-RRI (ms2297±678 165±517 0.375 2147±3688 
BP variability 
LF-SYS (mmHg29.0±13.0 5.0±6.0 0.229 6.7±5.4 
BRS-v (ms/mmHg) 6.0±6.2 4.1±3.5 0.113 12.5±7.6 

RRI-HF was not different between hyperOH patients (297±678 ms2) and nOH patients (164±517 ms2; P=0.4).

SBP-LF was not different between hyperOH patients (9.0±13.0 mmHg2) and nOH patients (5.0±6.0 mmHg2; P=0.2).

Mean BRS-v gain did not differ between hyperOH patients and nOH patients, although there was a non-significant trend towards higher BRS-v in hyperOH patients (6.0±6.2 ms/mmHg/ms compared with 4.1±3.5 ms/mmHg/ms; P=0.1).

Valsalva manoeuvre (Table 5)

During the VM, baseline SBP (143±24 mmHg compared with 146±28 mmHg, P=0.6), baseline HR (73±14 compared with 73±11, P=0.8), maximum HR (85±15 compared with 84±13, P=0.8), HR during phase 4 recovery (77±13 compared with 74±12, P=0.278) and the Valsalva ratio (1.18±0.14 compared with 1.12±0.12, P=0.08) did not differ between hyperOH and nOH groups.

Table 5
VM metrics

Data are presented as means ± S.D. Reported P-values are for Student's t test comparing hyperOH with nOH unless otherwise noted for continuous data and Fisher's exact test for categorical data. Statistically significant.

HyperOH (n=19)nOH (n=64)P-value
PRT (s) 16.5±8.9 31.6±16.6 <0.001* 
BRS-a (ms/mmHg/s) 5.8±4.0 3.9±4.3 0.093 
Phase 2 SBP decrement (mmHg) 77±38 78±25 0.895 
Phase 2 DBP decrement (mmHg) 16±10 25±13 0.008* 
Difference between baseline SBP and end of phase 2 SBP (mmHg) 52±32 63±27 0.140 
Difference between baseline DBP and end of phase 2 DBP (mmHg) 9±10 13±14 0.310 
Valsalva ratio 1.18±0.15 1.12±0.12 0.081 
Baseline HR (bpm) 73±14 73±11 0.840 
Maximum HR (bpm) 85±15 84±13 0.825 
HR at conclusion of VM (bpm) 78±13 74±12 0.278 
Presence of phase 2 L (%) 4 (21) 8 (13) 0.464 
Presence of phase 4 overshoot (%) 2 (11) 2 (3) 0.223 
HyperOH (n=19)nOH (n=64)P-value
PRT (s) 16.5±8.9 31.6±16.6 <0.001* 
BRS-a (ms/mmHg/s) 5.8±4.0 3.9±4.3 0.093 
Phase 2 SBP decrement (mmHg) 77±38 78±25 0.895 
Phase 2 DBP decrement (mmHg) 16±10 25±13 0.008* 
Difference between baseline SBP and end of phase 2 SBP (mmHg) 52±32 63±27 0.140 
Difference between baseline DBP and end of phase 2 DBP (mmHg) 9±10 13±14 0.310 
Valsalva ratio 1.18±0.15 1.12±0.12 0.081 
Baseline HR (bpm) 73±14 73±11 0.840 
Maximum HR (bpm) 85±15 84±13 0.825 
HR at conclusion of VM (bpm) 78±13 74±12 0.278 
Presence of phase 2 L (%) 4 (21) 8 (13) 0.464 
Presence of phase 4 overshoot (%) 2 (11) 2 (3) 0.223 

Sympathetically-mediated vasoconstriction

Valsalva phase 2L was present in four out of 19 hyperOH patients (21%) compared with eight out of 64 (13%) nOH patients (P=0.5). Valsalva phase 4 overshoot occurred in two out of 19 hyperOH patients (10%) compared with two of 64 (5%) nOH patients (P=0.2). No individuals from either group had both a phase 2L and a phase 4 overshoot during their VM.

Valsalva markers of adrenergic function

PRT was shorter in the hyperOH than nOH patients (16.5±8.9 s compared with 31.6±16.6 s, P<0.001; Figure 4a). PRTs for both hyperOH (P<0.001) and nOH (P<0.001) were longer than published data for PRT in healthy control subjects (2.00±1.98 s) [18]. There was a non-significant trend towards greater BRS-a in hyperOH patients than in nOH patients (5.8±4.0 ms/mmHg compared with 3.9±4.3 ms/mmHg, P=0.093; Figure 4b). Phase 2 SBP decrement from phase 1 was not different between the hyperOH and the nOH patients (77±38 mmHg compared with 78±25 mmHg; P=0.9; Figure 4c), whereas phase 2 DBP decrement was considerably smaller in hyperOH patients than in nOH patients (16±10 mmHg compared with 25±13, P=0.008; Figure 4d). PRT was negatively and significantly correlated with standing plasma NE (ρ=−0.459, P<0.001) and BRS-a was positively and significantly correlated with standing plasma NE (ρ=0.364, P=0.001)

Summary Valsalva metrics between the patient groups
Figure 4
Summary Valsalva metrics between the patient groups

Summary Valsalva data are presented for hyperOH and nOH patients. The panels show PRT (a), BRS-a (b), the fall in SBP (c) and the fall in DBP (d) during phase 2 of the VM. Data are presented as means ± S.E.M. Student's t tests were used to generate P-values.

Figure 4
Summary Valsalva metrics between the patient groups

Summary Valsalva data are presented for hyperOH and nOH patients. The panels show PRT (a), BRS-a (b), the fall in SBP (c) and the fall in DBP (d) during phase 2 of the VM. Data are presented as means ± S.E.M. Student's t tests were used to generate P-values.

There were non-significant trends to smaller drops in BP with Valsalva in hyperOH patients than nOH patients. These include the difference between baseline SBP and the SBP at the end of phase 2 (52±31 mmHg compared with 63±27 mmHg, P=0.1), difference between baseline DBP and the DBP at the end of phase 2 (9±10 mmHg compared with 13±14 mmHg, P=0.3) and the difference between baseline SBP and the SBP at the nadir of phase 3 (73±33 mmHg compared with 87±25 mmHg, P=0.1).

DISCUSSION

We define ‘hyperOH’ as the presence of paradoxically high upright plasma NE level ≥ 3.55 nmol/l (600 pg/ml) in neurogenic OH patients and extensively describe this novel group of patients with severe neurogenic OH for the first time. Compared with classical OH patients with low plasma NE, these patients have less severe adrenergic impairment.

Nomenclature: hyperadrenergic orthostatic hypotension

Dr David Streeten first coined the term ‘hyperOH’ in 1990 to describe mild OH in the presence of normal to elevated levels of upright plasma NE in eight young patients with large orthostatic increases in HR [7]. In contrast with the patients reported in that study, the patients in our hyperOH group had large orthostatic falls in BP with only a modest increase in HR (Table 1). This haemodynamic pattern is typical of neurogenic OH and not consistent with a transient cause of OH such as hypovolsaemia [21,22]. Previously, ‘hyperadrenergic’ has been used to describe patients with upright plasma NE > 3.55 nmol/l (600 pg/ml), a threshold met by only two of the eight patients in Streeten's study [8]. Moreover, those two patients were younger adults (age 49 and 37 years) [7]. The hyperOH patients in the present study represent older adults with OH in the presence of truly elevated upright plasma NE ≥ 3.55 nmol/l (600 pg/ml) and represent a different population than those patients described by Dr Streeten.

Hyperadrenergic OH patients have less severe adrenergic dysfunction

Despite fairly significant OH in both groups, patients with hyperOH had less severe sympathetic noradrenergic nerve dysfunction than nOH patients.

The PRT is a function of sympathetic nervous system mediated vasoconstriction. BP will recover quickly with normal vasoconstrictive function. With worsening adrenergic failure, the BP recovery time (as measured by PRT) will be longer. The PRT is a validated, reliable and convenient metric that can be used even in cases when phase 2L is absent, as is common in patients with neurogenic OH [1719]. In addition, PRT correlates well with muscle sympathetic nerve activity (MSNA), the gold standard for evaluating sympathetic nerve function [19,23].

Whereas a normal PRT in healthy individuals is 2.00±1.98 s [18], the PRTs for the hyperOH and the nOH patients were significantly longer (Figure 4a). A previous study has reported a PRT of 30 s in a group with OH, which is almost identical with the PRT in our nOH patients [18]. Vogel et al. [18] reported that a ’borderline OH’ group (defined as orthostatic SBP drop of >10 mmHg but <30 mmHg) had a PRT of 6.6 s. Our hyperOH patients had a PRT of 16 s, indicating that the severity of their adrenergic dysfunction lies somewhere in the continuum between moderate and severe adrenergic failure.

The BRS-a correlates well with adrenergic function and this has been validated with MSNA in a previous study [19]. The BRS-a was severely blunted in both OH groups compared with healthy individuals (24.5±19.3 ms/mmHg) [19]. The BRS-a has been shown to be less sensitive than PRT in discriminating those individuals with moderate OH from those with severe OH, which is consistent with the patients in our study [17].

Hyperadrenergic OH: clinical characteristics

Underlying diagnoses

MSA is a rapidly progressive neurodegenerative disorder associated with severe OH and has a mean survival of less than 9 years [24]. The absence of MSA in the hyperOH group suggests that a hyperOH state may not be a clinical feature of MSA.

Chronic OH associated with many conditions, such as diabetes, Parkinson's and MSA, is progressive [5] and increases with severity over time [25,26]. The nOH group in our study also had evidence of more severe disease compared with the hyperOH group. This suggests that the hyperOH phenotype may be a manifestation of a milder and less-well-developed stage of neurogenic OH. The hyperOH group presented with a trend towards a shorter duration of disease, which is consistent with this theory.

Concomitant medications

There were several medications that were disproportionately used more often in one group than the other. Overall, this discrepant pattern of medication use is most consistent with less severe disease in the hyperOH group. Midodrine and fludrocortisone are commonly prescribed in the treatment of severe neurogenic OH [21]. Significantly fewer patients with hyperOH were on fludrocortisone than nOH patients and there was a similar trend to less frequent midodrine use in hyperOH patient. Benzodiazepine use was also considerably greater in the hyperOH population. Physicians may be discouraged from prescribing benzodiazepines to nOH patients due to their greater symptoms and higher potential for falls [27]. Both acutely and chronically administered benzodiazepines decrease levels of NE at baseline and during times of stress [28,29], making it unlikely to be a cause of the higher NE seen in the hyperOH group.

Finally, the greater use of β-blockers in the hyperOH population reflects less severe disease in this group. The use of β-blockers has acutely been shown to increase levels of NE due to impaired clearance [30,31], but this effect on the study is probably minimal given that only a minority of patients in the hyperOH group were taking a β-blocker (six of 19). The more prevalent use of β-blockers in this population is better explained by a pattern of pharmaceutical prescribing more appropriate for a less severe disease phenotype.

Possible mechanism

The underlying mechanism behind hyperOH may be a partial autonomic neuropathy of the sympathetic nerves innervating the lower extremities, as proposed by Dr Streeten in his seminal paper [7]. In his study, the few patients with an elevated upright plasma NE level > 3.55 nmol/l (600 pg/ml) exhibited signs of greater responsiveness in the lower compared with upper extremity to NE infusion, indicating autonomic sympathetic denervation preferentially in the lower limbs [7]. Selective impairment of these nerves results in excessive pooling of blood and causes OH. It is unclear why these nerves in the lower extremity are affected preferentially but it may be related to their greater length, which are affected first in certain types of neuropathy, such as diabetic length-dependent distal symmetric polyneuropathy [32]. HyperOH may then be an earlier form of typical nOH when the upper extremities, which have not yet been affected, perceive this OH and via the baroreflex, compensate for autonomic sympathetic insufficiency in the lower extremities by producing excessive amounts of NE.

Another possibility could be decreased clearance of NE coupled with impaired NE sensitivity resulting in OH. NE has been shown to be elevated in some elderly individuals due to decreased renal clearance [33]. Although NE sensitivity was not measured, endocrinopathies such as Type 2 diabetes have a similar pathogenesis. Type 2 diabetics are well-documented to have excessive insulin secretion in an effort to stem growing insulin resistance to maintain normal blood glucose, followed by frank diabetes when insulin secretion drops below even normal levels [3436]. Neurogenic hyperOH patients share, perhaps, a similar fate when they arrive at this ‘burn-out phase’, akin to the old cliché of a candle burning brightest just before it goes out.

Limitations

The main limitation to the present study is the relatively modest sample size for the hyperOH group, although this is currently the largest study examining this hyperOH phenomenon. It is possible that a larger study sample size would allow more complete clinical characterization of this population, with further discriminating features achieving statistical significance. Another limitation is that we were unable to control for baseline hypovolaemia. Hypovolaemia has previously been shown to significantly impact the SBP changes during VM phase 2 [37].

Conclusion

In conclusion, neurogenic hyperOH describes the condition of neurogenic OH in the presence of elevated levels of plasma NE. Our data suggest that these individuals have severe OH, but less severe adrenergic impairment compared with typical neurogenic OH patients. These patients comprise a substantial minority of patients with neurogenic OH. From this limited study population, we postulate that hyperOH may be an earlier manifestation of typical neurogenic OH and these patients may develop typical nOH over time. Further studies will be required to elucidate the natural history of this disorder.

AUTHOR CONTRIBUTION

Philip Mar, Cyndya Shibao and Satish Raj participated in the development of study design and drafting of the manuscript for content. Philip Mar, Satish Raj, Italo Biaggioni, André Diedrich and David Robertson participated in the analysis and interpretation of data. Emily Garland and Bonnie Black participated in the conduction of the study and collection of data. All authors were involved with subsequent revisions of the manuscript prior to its final form.

We thank our patients who participated in this project and recognize the highly professional care provided by the staff of the Elliot V. Newman Clinical Research Center.

FUNDING

This work was supported by the National Institutes of Health [grant numbers R01 HL102387, R01 HL071784, P01 HL56693 and U54 NS065736] and a Clinical and Translational Science Award [grant number UL1 TR000445].

Abbreviations

     
  • BMI

    body mass index

  •  
  • BP

    blood pressure

  •  
  • bpm

    beats per minute

  •  
  • BRS-a

    adrenergic baroreflex sensitivity

  •  
  • BRS-v

    cardiovagal baroreflex sensitivity

  •  
  • DBP

    diastolic blood pressure

  •  
  • EPI

    epinephrine

  •  
  • HF

    high-frequency range

  •  
  • HR

    heart rate

  •  
  • hyperOH

    hyperadrenergic OH

  •  
  • LF

    low-frequency range

  •  
  • MSA

    multiple system atrophy

  •  
  • MSNA

    muscle sympathetic nerve activity

  •  
  • NE

    norepinephrine

  •  
  • nOH

    non-hyperadrenergic OH

  •  
  • OH

    orthostatic hypotension

  •  
  • phase 2L

    late phase 2

  •  
  • PRT

    pressure recovery time

  •  
  • RRI-HF

    HF component of the R-R intervals

  •  
  • SBP

    systolic blood pressure

  •  
  • SBP-LF

    SBP variability in the low-frequency band

  •  
  • SNRI

    serotonin–norepinephrine-reuptake inhibitor

  •  
  • SSRI

    selective serotonin-reuptake inhibitor

  •  
  • VM

    Valsalva manoeuvre

References

References
1
Freeman
 
R.
Wieling
 
W.
Axelrod
 
F.B.
Benditt
 
DG.
Benarroch
 
E.
Biaggioni
 
I.
Cheshire
 
W.P.
Chelimsky
 
T.
Cortelli
 
P.
Gibbons
 
C.H.
, et al 
Consensus statement on the definition of orthostatic hypotension, neurally mediated syncope and the postural tachycardia syndrome
Clin. Auton. Res.
2011
, vol. 
21
 (pg. 
69
-
72
)
[PubMed]
2
Low
 
P.A.
 
Prevalence of orthostatic hypotension
Clin. Auton. Res.
2008
, vol. 
18
 
Suppl. 1
(pg. 
8
-
13
)
[PubMed]
3
Poon
 
I.O.
Braun
 
U.
 
High prevalence of orthostatic hypotension and its correlation with potentially causative medications among elderly veterans
J. Clin. Pharm. Ther.
2005
, vol. 
30
 (pg. 
173
-
178
)
[PubMed]
4
Shibao
 
C.
Grijalva
 
C.G.
Raj
 
S.R.
Biaggioni
 
I.
Griffin
 
M.R.
 
Orthostatic hypotension-related hospitalizations in the United States
Am. J. Med.
2007
, vol. 
120
 (pg. 
975
-
980
)
[PubMed]
5
Robertson
 
D.
Robertson
 
R.M.
 
Causes of chronic orthostatic hypotension
Arch. Intern. Med.
1994
, vol. 
154
 (pg. 
1620
-
1624
)
[PubMed]
6
Ziegler
 
M.G.
Lake
 
C.R.
Kopin
 
I.J.
 
The sympathetic-nervous-system defect in primary orthostatic hypotension
N. Engl. J. Med.
1977
, vol. 
296
 (pg. 
293
-
297
)
[PubMed]
7
Streeten
 
D.H.
 
Pathogenesis of hyperadrenergic orthostatic hypotension. Evidence of disordered venous innervation exclusively in the lower limbs
J. Clin. Invest
1990
, vol. 
86
 (pg. 
1582
-
1588
)
[PubMed]
8
Jacob
 
G.
Ertl
 
A.C.
Shannon
 
J.R.
Furlan
 
R.
Robertson
 
R.M.
Robertson
 
D.
 
Effect of standing on neurohumoral responses and plasma volume in healthy subjects
J. Appl. Physiol.
1998
, vol. 
84
 (pg. 
914
-
921
)
[PubMed]
9
Velseboer
 
D.C.
de Haan
 
R.J.
Wieling
 
W.
Goldstein
 
D.S.
de Bie
 
R.M.
 
Prevalence of orthostatic hypotension in Parkinson's disease: a systematic review and meta-analysis
Parkinsonism Relat. Disord.
2011
, vol. 
17
 (pg. 
724
-
729
)
[PubMed]
10
Jacob
 
G.
Shannon
 
J.R.
Black
 
B.
Biaggioni
 
I.
Mosqueda-Garcia
 
R.
Robertson
 
R.M.
Robertson
 
D.
 
Effects of volume loading and pressor agents in idiopathic orthostatic tachycardia
Circulation
1997
, vol. 
96
 (pg. 
575
-
580
)
[PubMed]
11
Pan
 
J.
Tompkins
 
W.J.
 
A real-time QRS detection algorithm
IEEE Trans. Biomed. Eng.
1985
, vol. 
32
 (pg. 
230
-
236
)
[PubMed]
12
Malik
 
M.
Bigger
 
T.
Malliani
 
A.
Moss
 
A.
Schwartz
 
P.
 
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
Circulation
1996
, vol. 
93
 (pg. 
1043
-
1065
)
[PubMed]
13
Diedrich
 
A.
Jordan
 
J.
Tank
 
J.
Shannon
 
J.R.
Robertson
 
R.
Luft
 
F.C.
Robertson
 
D.
Biaggioni
 
I.
 
The sympathetic nervous system in hypertension: assessment by blood pressure variability and ganglionic blockade
J. Hypertens.
2003
, vol. 
21
 (pg. 
1677
-
1686
)
[PubMed]
14
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
Circulation
1996
, vol. 
93
 (pg. 
1043
-
1065
)
[PubMed]
15
Persson
 
P.B.
DiRienzo
 
M.
Castiglioni
 
P.
Cerutti
 
C.
Pagani
 
M.
Honzikova
 
N.
Akselrod
 
S.
Parati
 
G.
 
Time versus frequency domain techniques for assessing baroreflex sensitivity
J. Hypertens.
2001
, vol. 
19
 (pg. 
1699
-
1705
)
[PubMed]
16
Sandroni
 
P.
Benarroch
 
E.E.
Low
 
P.A.
 
Pharmacological dissection of components of the Valsalva maneuver in adrenergic failure
J. Appl. Physiol.
1991
, vol. 
71
 (pg. 
1563
-
1567
)
[PubMed]
17
Novak
 
P.
 
Assessment of sympathetic index from the Valsalva maneuver
Neurology
2011
, vol. 
76
 (pg. 
2010
-
2016
)
[PubMed]
18
Vogel
 
E.R.
Sandroni
 
P.
Low
 
P.A.
 
Blood pressure recovery from Valsalva maneuver in patients with autonomic failure
Neurology
2005
, vol. 
65
 (pg. 
1533
-
1537
)
[PubMed]
19
Schrezenmaier
 
C.
Singer
 
W.
Swift
 
N.M.
Sletten
 
D.
Tanabe
 
J.
Low
 
P.A.
 
Adrenergic and vagal baroreflex sensitivity in autonomic failure
Arch. Neurol.
2007
, vol. 
64
 (pg. 
381
-
386
)
[PubMed]
20
Nwazue
 
V.C.
Paranjape
 
S.Y.
Black
 
B.K.
Biaggioni
 
I.
Diedrich
 
A.
Dupont
 
W.D.
Robertson
 
D.
Raj
 
S.R.
 
Postural tachycardia syndrome and inappropriate sinus tachycardia: role of autonomic modulation and sinus node automaticity
J. Am. Heart Assoc.
2014
, vol. 
3
 pg. 
e000700
 
[PubMed]
21
Freeman
 
R.
 
Clinical practice. Neurogenic orthostatic hypotension
N. Engl. J. Med.
2008
, vol. 
358
 (pg. 
615
-
624
)
[PubMed]
22
Tuboly
 
G.
Rudas
 
L.
Csillik
 
A.
Nagy
 
A.
Benedek
 
K.
Benedek
 
G.
Braunitzer
 
G.
 
Haemodynamic parameters and cognitive function during modeled acute volume loss
Acta Physiol. Hung.
2012
, vol. 
99
 (pg. 
118
-
125
)
[PubMed]
23
Rudas
 
L.
Crossman
 
A.A.
Morillo
 
C.A.
Halliwill
 
J.R.
Tahvanainen
 
K.U.
Kuusela
 
T.A.
Eckberg
 
D.L.
 
Human sympathetic and vagal baroreflex responses to sequential nitroprusside and phenylephrine
Am. J. Physiol.
1999
, vol. 
276
 (pg. 
H1691
-
H1698
)
[PubMed]
24
Schrag
 
A.
Wenning
 
G.K.
Quinn
 
N.
Ben-Shlomo
 
Y.
 
Survival in multiple system atrophy
Mov. Disord.
2008
, vol. 
23
 (pg. 
294
-
296
)
[PubMed]
25
Stefanova
 
N.
Bucke
 
P.
Duerr
 
S.
Wenning
 
G.K.
 
Multiple system atrophy: an update
Lancet Neurol.
2009
, vol. 
8
 (pg. 
1172
-
1178
)
[PubMed]
26
Mabuchi
 
N.
Hirayama
 
M.
Koike
 
Y.
Watanabe
 
H.
Ito
 
H.
Kobayashi
 
R.
Hamada
 
K.
Sobue
 
G.
 
Progression and prognosis in pure autonomic failure (PAF): comparison with multiple system atrophy
J. Neurol. Neurosurg. Psychiatry
2005
, vol. 
76
 (pg. 
947
-
952
)
[PubMed]
27
Leipzig
 
R.M.
Cumming
 
R.G.
Tinetti
 
M.E.
 
Drugs and falls in older people: a systematic review and meta-analysis: I. Psychotropic drugs
J. Am. Geriatr. Soc.
1999
, vol. 
47
 (pg. 
30
-
39
)
[PubMed]
28
Stratton
 
J.R.
Halter
 
J.B.
 
Effect of a benzodiazepine (alprazolam) on plasma epinephrine and norepinephrine levels during exercise stress
Am. J. Cardiol.
1985
, vol. 
56
 (pg. 
136
-
139
)
[PubMed]
29
Hossmann
 
V.
Maling
 
T.J.
Hamilton
 
C.A.
Reid
 
J.L.
Dollery
 
C.T.
 
Sedative and cardiovascular effects of clonidine and nitrazepam
Clin. Pharmacol. Ther.
1980
, vol. 
28
 (pg. 
167
-
176
)
[PubMed]
30
Best
 
J.D.
Halter
 
J.B.
 
Blood pressure and norepinephrine spillover during propranolol infusion in humans
Am. J. Physiol.
1985
, vol. 
248
 (pg. 
R400
-
R406
)
[PubMed]
31
Rosen
 
S.G.
Supiano
 
M.A.
Perry
 
T.J.
Linares
 
O.A.
Hogikyan
 
R.V.
Smith
 
M.J.
Halter
 
J.B.
 
Beta-adrenergic blockade decreases norepinephrine release in humans
Am. J. Physiol.
1990
, vol. 
258
 (pg. 
E999
-
E1005
)
[PubMed]
32
Albers
 
J.W.
Pop-Busui
 
R.
 
Diabetic neuropathy: mechanisms, emerging treatments, and subtypes
Curr. Neurol. Neurosci. Rep.
2014
, vol. 
14
 pg. 
473
 
[PubMed]
33
Esler
 
M.
Skews
 
H.
Leonard
 
P.
Jackman
 
G.
Bobik
 
A.
Korner
 
P.
 
Age-dependence of noradrenaline kinetics in normal subjects
Clin. Sci.
1981
, vol. 
60
 (pg. 
217
-
219
)
[PubMed]
34
Leahy
 
J.L.
 
Pathogenesis of type 2 diabetes mellitus
Arch. Med. Res.
2005
, vol. 
36
 (pg. 
197
-
209
)
[PubMed]
35
Weyer
 
C.
Bogardus
 
C.
Mott
 
D.M.
Pratley
 
R.E.
 
The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus
J. Clin. Invest.
1999
, vol. 
104
 (pg. 
787
-
794
)
[PubMed]
36
Fukushima
 
M.
Usami
 
M.
Ikeda
 
M.
Nakai
 
Y.
Taniguchi
 
A.
Matsuura
 
T.
Suzuki
 
H.
Kurose
 
T.
Yamada
 
Y.
Seino
 
Y.
 
Insulin secretion and insulin sensitivity at different stages of glucose tolerance: a cross-sectional study of Japanese type 2 diabetes
Metabolism
2004
, vol. 
53
 (pg. 
831
-
835
)
[PubMed]
37
van Kraaij
 
D.J.
Jansen
 
R.W.
Hoefnagels
 
W.H.
 
Monitoring hypovolemia in healthy elderly subjects by measuring blood pressure response to Valsalva's maneuver
Geriatr. Nephrol. Urol.
1999
, vol. 
9
 (pg. 
73
-
79
)
[PubMed]