EPMA Journal (2018) 9:35–45

https://doi.org/10.1007/s13167-017-0124-4

RESEARCH

Heart rate variability: the comparison between high tension and normal
tension glaucoma
Natalia Ivanovna Kurysheva 1,2,3 & Tamara Yakovlevna Ryabova 4 & Vitaliy Nikiforovich Shlapak 4

Received: 4 October 2017 / Accepted: 21 December 2017 / Published online: 22 February 2018
# The Author(s) 2018. This article is an open access publication

Abstract
Relevance Vascular factors may be involved in the development of both high tension glaucoma (HTG) and normal tension
(NTG) glaucoma; however, they may be not exactly the same. Autonomic dysfunction characterized by heart rate variability
(HRV) is one of the possible reasons of decrease in mean ocular perfusion pressure (MOPP).
Purpose To compare the shift of the HRV parameters in NTG and HTG patients after a cold provocation test (CPT).
Methods MOPP, 24-hour blood pressure and HRV were studied in 30 NTG, 30 HTG patients, and 28 healthy subjects. The
cardiovascular fitness assessment was made before and after the CPT. The direction and magnitude of the average group shifts of
the HRV parameters after CPT were assessed using the method of comparing regression lines in order to reveal the difference
between the groups.
Results MOPP and minimum daily diastolic blood pressure were decreased in HTG and NTG patients compared to healthy
subjects. There was no difference in MOPP between HTG and NTG before the CPT. However, all HRV parameters reflected the
predominance of sympathetic innervation in glaucoma patients compared to healthy subjects (P < 0.05).
Before the CPT, the standard deviation of NN intervals (SDNN) of HRV was lower in HTG compared to NTG, 27.2 ± 4.1 ms
and 35.33 ± 2.43 ms (P = 0.02), respectively. After the CPT, SDNN decreased in NTG by 1.7 ms and increased in HTG and
healthy subjects by 5.0 ms and 7.09 ms, respectively (P < 0.05). The analysis of relative shift of other HRV parameters after the
CPT also revealed a significant difference between NTG and HTG in regard to the predominance of sympathetic innervation in
NTG compared to HTG.
Conclusion Patients with NTG have more pronounced disturbance of autonomic nervous system than HTG patients, which is
manifested with the activation of sympathetic nervous system in response to CPT. This finding refers to the NTG pathogenesis
and suggests the use of HRV assessment in glaucoma diagnosis and monitoring.

Keywords High tension glaucoma . Normal tension glaucoma . Ocular blood flow . Heart rate variability . Ocular perfusion
pressure

Introduction
* Natalia Ivanovna Kurysheva
[email protected]
Vascular factors have been recognized to play an important
role in glaucoma pathogenesis. According to the recent re-
1
Consultative-Diagnostic Department of Ophthalmological Center, view, subjects with cardiovascular disease were 2.33 times
Federal Medical and Biological Agency of the Russian Federation,
Moscow, Russian Federation
more likely to develop rapidly progressive glaucoma disease
2
despite significantly lower mean and baseline intraocular pres-
A.I. Burnazyan Federal Medical and Biophysical Center, Federal
Medical and Biological Agency of the Russian Federation,
sures (IOP) [1]. Moreover, the concept that vascular changes
Moscow, Russian Federation in the eye may be an early indicator of heart diseases is also
3
Ophthalmological Department of the Institute of Improvement of
discussed in literature [2].
Professional Skills, Federal Medical and Biological Agency of the It is believed that there are two groups of factors responsi-
Russian Federation, Moscow, Russian Federation ble for the development of glaucomatous optic neuropathy
4
Science Center of Radiation and Chemical Safety and Hygiene, (GON) in case of normal IOP: (1) vascular dysfunction
Medical and Biological Agency, Moscow, Russian Federation (dysregulation) leading to instable ocular blood flow of optic
36 EPMA Journal (2018) 9:35–45

disc [3] and (2) mechanical dysfunction leading to damage of pronounced in NTG patients compared to HTG [16, 21, 22].
scleral membrane and infringement of axons in the optic On the other hand, some authors reported that patients with
nerve. It is generally recognized that high tension glaucoma HTG and NTG exhibit similar alterations in ocular and sys-
(HTG) is most commonly characterized by mechanical dys- temic circulation at the early stages of the disease [23].
function, while vascular dysregulation appears to be at the There is no literature data concerning the influence of CPT
forefront in case of the normal tension glaucoma (NTG) de- on HRV in NTG compared to HTG patients.
velopment [4]. Meanwhile, it is known that instable ocular The objective of this study is to compare the shift of HRV
blood flow plays an important role in the development of both parameters in NTG and HTG patients after CPT.
forms of glaucoma, including with high IOP [5–7]. The ques-
tion that naturally arises: is insufficient blood supply to retina
and optic nerve is more pronounced in NTG? In other words, Materials and methods
are there more hemodynamic-related properties in NTG and
are there any features of the general regulation of blood flow, Study subjects
which are typical for patients with NTG?
Recent studies have shown the role of vascular distur- Eighty-eight eyes of 88 subjects (30 patients with early and
bances and vascular dysregulation in glaucoma progression, moderate NTG, 30 patients with HTG, and 28 age-matched
including NTG [8–13]. This unites glaucoma with such forms healthy subjects) were included in this study.
of pathology as migraine, vasospasm, arterial hypertension, All patients were Caucasian.
and hypotension [9, 14]. HTG was diagnosed on the basis of characteristic chang-
Excessive activity of the sympathetic link of the autonomic es in the optic disc detected by ophthalmoscopy, which was
nervous system (ANS) is one of the possible causes leading performed by one glaucoma specialist (NK) and confirmed
both to disturbance of blood supply to the ONH and to a by two other glaucoma specialists, pathological deviation
decrease in ocular perfusion pressure (OPP) in the vessels of from the normal neuroretinal rim, glaucomatous optic disc
the optic nerve and choroid. Moreover, it has been shown in cupping, peripapillary atrophy, wedge-shaped defects of the
literature that the excessive activity of the sympathetic link of retinal nerve fiber layer (RNFL) adjacent to the edge of
ANS is responsible for glaucoma progression due to an insta- optic disc, hemorrhages at the optic disc boundary,
ble ocular blood flow [10]. glaucomatous visual field (VF) loss on at least two consec-
It is believe that patients with an instable ocular blood flow utive tests, an open angle on gonioscopy (not less than
respond stronger to psychological stress as it has been de- 30°), and ametropia ≤ 0.5 diopter. IOP was higher than
scribed in patients with primary vascular dysregulation 21 mmHg.
(PVD) [4]. It has also been emphasized that any psychological The patients with the same criteria but with IOP of
stress leads to vascular dysfunction [15]. Cold stimulation is a 21 mmHg or lower (without topical treatment), confirmed in
well-established provocation test used for detecting abnormal repeated measurements on different days, were referred to the
vascular reactivity in patients with autonomic failures [16]. NTG group. All patients were followed up at our clinic for at
A significant reduction of retrobulbar blood flow in NTG least 4 years with visits at 3- to 5-month intervals and had no
has been described in literature [17, 18]. In addition, Kaiser ocular pathology other than glaucoma.
et al. demonstrated that ocular blood flow was reduced in both The healthy participants were recruited from the people
NTG patients and in those HTG patients which progressed accompanying the patients and had IOP of less than
despite a normal IOP [17]. 21 mmHg for both eyes, a normal Humphrey Swedish
It has been recently revealed that the cold provocation test Interactive Threshold Algorithm 24-2 standard visual field
(CPT) may increase the ET-1 level in plasma in patients with with mean deviation (MD), and pattern standard deviation
NTG that reflected their vascular dysregulation [1]. This phe- (PSD) within 95% limits of the normal reference. They
nomenon may also indicate the imbalance of ANS that is man- also had a glaucoma hemifield test within 97% limits, a
ifested mostly during provocation tests, including CPT [19]. central corneal thickness ≥ 500 μm, a normal-appearing
Patients with systemic autonomic dysfunction might be at optic nerve head (ONH), a normal RNFL, an open anterior
higher risk for glaucoma progression due to higher suscepti- chamber angle as observed by gonioscopy, and no history
bility of the optic nerve to fluctuations of IOP or OPP. Heart of chronic ocular or systemic corticosteroid use. The age
rate variability (HRV) is a well-known tool that allows study- and race distribution of the healthy subjects matched that
ing the autonomic modulation of the heart sympathovagal of the glaucoma patients.
balance [20]. Exclusion criteria were the following: large refractive er-
According to the literature, vascular risk factors are not rors (outside of ± 6.00 dpt sphere or 2.00 dpt cylinder), pupil
exactly the same between HTG and NTG [2, 9, 21]. It was diameter < 3 mm, systemic administration of beta-blockers
hypothesized that the vascular dysfunction would be more and calcium-channel blockers, concomitant ocular disease
EPMA Journal (2018) 9:35–45 37

(except for early cataract), chronic autoimmune diseases, oscillometric method, i.e., by analyzing pulse phenomena
diabetes mellitus, acute circulatory disorders in past medical in a blood pressure cuff.
history, and any concomitant diseases involving the admin- Mean ocular perfusion pressure (MOPP) was calculated on
istration of steroid drugs and antihypertensive medications. the basis of IOP and arterial BP measurements immediately
Patients with any significant cardiovascular, pulmonary, and before the optical coherence tomography (OCT) scanning and
metabolic conditions other than controlled systemic hyper- investigation of retrobulbar blood flow, after a 10-min resting
tension (BP < 140/90 mmHg) were excluded. A history of period in a sitting position. Systemic BP was measured using the
ocular arterial or venous obstruction (branch or central oc- Riva Rocci technique. MOPP was calculated using the formula:
clusion) and systemic conditions associated with venous MOPP = (2/3 diastolic BP + 1/3 systolic BP) × 2/ 3 − IOP.
congestion (e.g., heart failure) were also considered as ex- BP was measured at rest in the sitting position. When mea-
clusion criteria. The patients were instructed to avoid caf- suring BP in the sitting position, the back had a support and
feine intake, smoking, and exercise for 5 h prior to the study the middle shoulder point was at the heart level (the fourth
visit. intercostal space). The measurement was made using Adyutor
If both eyes of a patient were eligible, one eye was random- mechanical tonometer.
ly chosen. Those patients, who previously used antiglaucoma
drops, were asked to discontinue the drug for a period of OCT image acquisition and processing
21 days (drug washout period), while others were newly di-
agnosed glaucoma cases. The medical histories of all patients All subjects also underwent optic disc area measurement at
were carefully obtained with special attention paid to the signs RTVue XR Avanti SD-OCT (Optovue, Inc., Fremont, CA,
of PVD (migraine, vasospasm, and neurocirculatory dystonia) USA) using the traditional ONH scan. All the examinations
and special questions were asked to reveal the symptoms of for a particular subject were performed on the same day. OCT
Flammer syndrome [21]. was performed in the macular area as well. The tracking mode
was used.
The ganglion cell complex (GCC) thickness was deter-
Study examinations mined with the GCC scanning protocol. The characteristics
of GCC (global loss volume, GLV; focal loss volume, FLV)
All participants underwent complete ophthalmologic exam- were also measured.
inations including the best corrected acuity, slit lamp exam-
ination, IOP measurement using analyzer of biomechanical HRV assessment
properties of eyes (Ocular Response Analyzer, ORA,
Reichert Ophthalmic Instruments Inc., Depew, NY), The cardiovascular fitness assessment was made to all pa-
gonioscopy, anterior chamber angle measurement (Visante tients before and after CPT using Rhythm-MET hardware-
OCT, Carl Zeiss, Germany), pachymetry (SP-100, Tomey, software complex developed by the Federal State Unitary
GmbH, Germany), dilated fundus biomicroscopy using 78- Enterprise BScience Center of Radiation and Chemical
diopter lens, stereoscopic optic disc photography, and stan- Safety and Hygiene of the Medical and Biological
dard automated perimetry (SAP) using a Humphrey Field Agency of the Russian Federation^.
Analyzer (HFA, Carl Zeiss Meditec Inc., Dublin, CA) with Rhythm-MET hardware-software complex was used in
SITA. Only reliable SAP results, which were defined as the present study. The Rhythm-MET software and hard-
false-negative and false-positive responses of < 33% and ware system (registration certificate of the Federal
fixation loss of < 20%, were eligible for the study. Supervisory Agency for Healthcare and Social
Glaucomatous VF defects were determined as having a Development No. FSP 2009/04339 dated February 17,
cluster of three or more non-edge points with P < 0.05 2009) was developed to assess the state of health, to iden-
and at least 1 point with P < 0.01 in the pattern deviation tify early arterial hypertension, and to assess functional
probability plot, PSD of less than 5%; or glaucoma state of the central nervous system, psychological, and
hemifield test results outside normal limits. Both glaucoma emotional stability. The method of its work is based on
and normal participants underwent SAP at least twice be- a comprehensive analysis of HRV, systemic hemodynam-
fore this study. ics and autonomic regulation. Photoplethysmograms re-
The study included 24-h blood pressure (BP) monitor- corded from a phalanx with an infrared sensor, located
ing: the automated measurement of BP for 24 h at fixed in the microprocessor module of data input and process-
intervals according to a preset program. Measurements ing, were used as the source of data on HRV and periph-
were made on an outpatient basis in the conditions of nor- eral blood flow [24]. Cardiointervals obtained from
mal patient activity. The device measured the heart rate, photoplethysmograms were processed in accordance with
systolic, and diastolic BP at set intervals using the the recommendations [25] for assessment of HRV
38 EPMA Journal (2018) 9:35–45

parameters and their subsequent generalization, including Statistical analysis

hemodynamics parameters, as well as for assessment of
functional status (FS) and functional reserves (FR) of the The used methods of statistical processing and analysis of
cardiovascular system according to the results of the ex- obtained HRV results were aimed at establishing statistically
amination at rest and after conducting CPT in order to significant differences between the NTG, HTG, and control
form groups that are homogeneous in FS and FB. CPT groups both for the rest period and after CPT. These groups
also analyzed the parameters of time dependence of the were formed on the basis of results of clinical examination, so
amplitude of photoplethysmogram and recovery time after they played the role of learning samples.
the test. In this study, the method of comparing regression lines
A generally accepted CPT was used as a provocation test was applied [27]. The method of comparing regression
in accordance to Weise et al. [26] with our modification. The lines has some advantages, for example, over the widely
testing procedure was the following: a patient’s hand was used method of comparing the mean group values due to
dipped into cold water (+ 4 °C) with small pieces of ice; the fact that it operates immediately with two indicators. In
moreover, the hand was covered with plastic ice bags for addition, it is also suitable for small samples. Based on the
30 s. The registration of RR intervals was made at the end known physiological concepts, selection of parameters of
of the CPT. autonomic regulation was made, for which a reliable sep-
The following characteristics were taken into account in aration of groups of patients from NTG and primary open-
accordance with the international standard [25]: angle glaucoma (POAG) is possible. At the same time, the
first necessary step in the search for these parameters was
& Standard deviation of NN-interval (SDNN): the parameter the assessment of the direction of the average group shifts,
of HRV characterizing the total effect of autonomic blood or the differences in the parameters after the CPT compared
circulation regulation. A reduction in SDNN reflects low to the initial data. Next, the magnitude of the shift was
HRV indicating a high tone of heart sympathetic activity. taken into account (based on how much the index changed
The decrease in SDNN reflects a decrease in HRV, which after the CPT).
indicates an increase in the tone of heart sympathetic Finally, the selected parameters were statistically processed
activity; in order to reveal differences between the groups.
& The parameter of parasympathetic autonomic regulation The conclusion that the response to cold provocation test in
activity (RMSSD); patients with NTG and POAG is statistically reliably different
& Total spectral power (TP): the parameter of the absolute was done on the basis of the negative result of the statistical
activity level of regulatory systems; test for the coincidence of the regression lines.
& Power in the high frequency range (HF): the parameter of The procedure of comparing regression lines includes the
the spectral power of heart rate respiratory undulations comparison of coefficients of slope b (slope ratio), coefficients
reflecting the activity level of respiratory center. The high of shear a, and the comparison of lines in general [28].
frequency band reflects fast changes in beat-to-beat vari- Among all possible regression lines, we used that one,
ability due to parasympathetic activity; which corresponds to the well-known theory of initial value
& Power in the low frequency range (LF): the low fre- in physiology [29]. According to this theory, direction and
quency band is considered to be a fair approximation value of endogenous and exogenous effect depend on the
of sympathetic activity. The very low frequency band initial metabolism and the state of regulatory body systems
reflects mostly sympathetic stimulation. It also reflects before CPT. HRV values reflect the peculiarities of the state
the activity of the baroreflex mechanism ensuring the of regulatory mechanisms of cardiovascular system in NTG,
BP regulation; HTG patients, and control subjects.
& Autonomic regulation index (ARI): the parameter used Taking into account the abovementioned theory, regression
to evaluate the activity of ANS. The increased ARI lines are as follows:
shows the activation of sympathetic regulation, but ΔHRVNTG = aNTG + D bNTG∙HRVNTG,rest, ΔHRVHTG =
the decreased ARI shows the activation of parasympa- a HTG + b HTG ∙HRV HTG,rest , where ΔHRV NTG,CPT and
thetic regulation. ΔHRVHTG,CPT are shifts of the studied HRV parameter after
& The parameter characterizing the degree of tension of reg- CPT for the NTG and HTG groups, respectively; aNTG, bNTG
ulatory systems, in particular, is the activity level of the and aHTG, bHTG are the coefficients of regression lines for
sympathetic regulation mechanisms (S). It may indicate NTG and HTG groups, respectively, whereby a and b are
the inotropic effect of enhancing of sympathetic nerve the coefficients of regression characterizing the degree of con-
influences. This parameter reflects the activation of jugation and the proportionality of shift in characteristics. The
ergotropic regulation mechanisms and increases the inten- coefficient b reflects the strength of connection, i.e., the slope
sity of energy processes. of the line and the intersection of the ordinate axes, the
EPMA Journal (2018) 9:35–45 39

coefficient a determines the position of the line in the coordi- (P) of intragroup differences in parameters after CPT (col-
nate axes. umns 4, 7, and 10) for each group. The right side of Table 2
HRVNTG,rest and HRVHTG,rest are HRV parameters relating gives the information on the results of estimates of statistical
to rest for the NTG and HTG groups, respectively; the shift in significance of intragroup differences of NTG and HTG at rest
HRV parameters is determined by the following ratios: (column 11) and after CPT (column 12).
Δ H RV N T G , C P T = H R V N T G , C P T − H R V N T G , r e s t , The statistical significance of intragroup differences of the
ΔHRVHTG,CPT = HRVHTG,CPT − HRVHTG,rest. HTG, NTG, and control groups at rest and after CPT is given
The conclusion on statistically significant difference of re- in columns 13 and 14. The columns 15 and 16 contain the
gression lines, and, therefore, on the difference of the com- result of estimates of statistical significance of intragroup dif-
pared groups is built on the basis of comparing the calculated ferences of the HTG and control groups.
statistics with the values of Fisher’s test (for coefficient b) and The NTG and HTG groups are significantly different from
the values of Student’s test (for coefficient a and comparison the control group according to the mean group values of HRV
of lines in general). parameters. For example, according to CPT results (column
In this study, the regression lines for the NTG, HTG, and 14, Table 2), the NTG group differs from the HTG and control
control groups were compared using the statistical software groups both at rest and after CPT (columns 15 and 16). At the
package considering the specific character of medical and bi- same time, there are significant differences between the NTG
ological tasks [28]. and HTG groups only at rest (column 11). This was one of the
The following designations were applied to the checks of reasons to use other methods of statistical processing, in par-
statistical significance of difference of regression lines: B1^— ticular, the method of comparing regression lines, to analyze
to the criterion of coincidence of regression lines in general, the intragroup differences.
B2^—to the criterion of comparison of line slopes (b), and To compare the NTG and HTG groups, Fig. 1 shows the
B3^—to the criterion of comparison of shifts (a). regression lines for those HRV parameters, whose regression
Standard methods of descriptive statistics were used as fol- lines are significantly different and the results of the inter-
lows: T tests, methods of non-parametric statistics—Wilcoxon group differences in HRV indicators, revealed by the method
test, Mann-Whitney U test, implemented in the corresponding of comparison of regression lines, are represented in Table 3.
statistics packages (IBM SPSS Statistics v 21, StatPlus). The obtained results relating to HRV parameters confirm
Parameters with Р < 0.05 were considered statistically sig- the earlier assumption on the difference in groups:
nificant. Since a number of parameters (GCC, GLV, systolic, NTG and HTG in each of the pairs of parameters of these
and mean perfusion pressure) depended on the anterior- groups: (SDNN, ΔSDNN), (LF, ΔLF) (ARI, ΔARI), and (TP,
posterior axis and the age of the subjects, we carried out an ΔTP);
adjustment for these parameters on the basis of the linear re-
gression model. & NTG and Control in the pairs of parameters: (SDNN,
ΔSDNN), (ARI, ΔARI).
& HTG and Control in the pairs of parameters: (SDNN,
Results ΔSDNN), (ARI, ΔARI).

The patient characteristics are given in Table 1.

The table shows the mean values and standard deviation (in
parentheses), the exact two-sided Mann-Whitney U test (P*) Discussion
between the healthy eyes (control group) and the NTG pa-
tients, P** between the NTG and HTG groups, and total To the best of our knowledge, this is the first study that has
P*** obtained when comparing these three groups by means revealed the difference in HRV parameters in NTG and HTG
of Kruskal-Wallis test by ranks. All abbreviations are given in patients both at rest and in response to CPT. Our data con-
the list of abbreviations. firmed the predominance of the sympathetic ANS at rest in
No significant differences in SBP, DBP, minimum DBP, NTG compared to HTG and the shift towards sympathetic
and MOPP were identified between NTG and HTG patients activity after CPT, which was more pronounced in NTG com-
across the 24-h measurement period though there were a sig- pared to HTG and control subjects.
nificant difference between both the glaucoma groups and It has been recognized that POAG according to its strict
healthy subjects (Table 1). definition includes HTG and NTG. As far as HTG and NTG
The parameters of HRV in all studied groups are represent- are concerned, they appear to be a continuum of glaucomatous
ed in Table 2. Its left side shows the mean group values of process, in which the underlying mechanisms shifts from pre-
HRV parameters in the NTG, HTG, and control groups. dominantly elevated IOP in HTG to hemodynamic change in
Moreover, the table contains the data on statistical significance NTG. In other words, both HTG and NTG are related to
40 EPMA Journal (2018) 9:35–45

Table 1 Characteristics of the studied groups

Parameter Healthy eyes P NTG eyes P** HTG eyes P***

Age, years 63.8 (9.8) 0.621 63.4 (5.8) 0.26 62.5 (4.3) 0.745
Systolic BP, mm Hg 127.4 (13.4) 0.043 138.2(7.3) 0.055 131.0(5.2) 0.056
Diastolic BP, mm Hg 81.0 (9.3) 0.04 75.5 (8.3) 0.84 74.9 (5.8) 0.03
Minimum daily diastolic BP, mm Hg 38.0 (7.1) 0.02 30.0 (6.2) 0.09 32.0 (8.6) 0.05
Corneal compensated IOP, mm Hg 14.8 (3.6) 0.22 14.1 (3.4) 0.01 24.5 (5.2) 0.02
MOPP, mm Hg 61.1 (8.5) 0.023 53.1 (8.1) 0.051 51.3 (7.3) 0.021
MD, dB − 0.7 (2.12) 0.001 − 5.8 (4.2) 0.31 − 6.5 (3.6) < 0.001
PSD, dB 1.20 (0.58) 0.005 3.79 (1.45) 0.25 4.09 (1.12) 0.05
RNFL, μm 103.9 (7.1) 0.002 85.3 (5.1) 0.34 83.6 (5.1) 0.001
Savg, μm 119.1 (9.9) 0.003 90.9 (6.2) 0.24 92.3 (4.8) 0.03
Iavg, μm 125.1 (10,2) 0.003 97.8 (6.0) 0.09 93.7 (4.0) 0.03
GCC, μm 99.3 (8.2) 0.005 80.3 (6.2) 0.21 76.2 (5.52) 0.004
FLV, % 0.18 (0.09) 0.003 3.57 (1.12) 0.08 5.02 (2.14) 0.002
GLV, % 1.61 (1.86) 0.002 8.35 (1.86) 0.34 7.33 (2.36) 0.001
Axial length, mm 23.1 (1.3) 0.675 23,3 (1.4) 0.45 23,6 (1.8) 0.33
Corneal thickness, μm 532.6 (20.2) 0.923 536 (21.3) 0.38 530 (18.1) 0.423

hemodynamics, but it was hypothesized that the evidence of of both glaucoma groups exhibited similar alterations in ocu-
vascular dysfunction would be more pronounced in NTG pa- lar and systemic circulation compared to healthy subjects, no
tients. One of the possible reasons for this is autonomic dys- significant differences were found in retinal arterial or venous
function that may contribute to unstable or fluctuating blood flicker response, nocturnal blood pressure, systemic arterial
pressure and thereby may induce the dysfunction of autoreg- stiffness, and intima-media thickness between HTG and
ulation leading to glaucoma development and progression [3]. NTG patients [23].
The autonomic dysfunction in patients with NTG using According to Bossuyt et al., OPP was significantly reduced
short-term and a 24-h heart rate variability analysis have been in HTG and NTG patients compared with controls, suggesting
reported by different authors [10, 30–33]. However, the liter- that perfusion-related vascular alterations are likely to be
ature data on autonomic dysfunction in HTG and NTG are playing a part in the pathogenesis of both conditions [34].
controversial. According to Riccadonna M. et al., HRV and These findings allow us to propose that it is unreasonable to
nocturnal diastolic BP variability were reduced in NTG com- differentiate NTG and HTG.
pared to HTG [31]. Furthermore, these differences were more On the other hand, there are some important differences
prominent in more severe clinical forms of NTG. The authors between HTG and NTG. For example, the VF progression
suggested a correlation between the extent of autonomic dis- pattern in NTG is reported to be different when compared with
order and severity of glaucoma. other types of glaucoma [35]. Notably, NTG eyes progressed
Brown et al. evaluated baroreflex control of the heart and more frequently in the central region of the VF, and this re-
blood vessels in HTG and NTG patients using the sinusoidal sponse was related to unstable or large fluctuations of 24-h
neck suction. They revealed that the response of ANS in the mean ocular perfusion pressure, and excessive nocturnal dips
healthy subjects was significantly greater than that in glauco- of systemic blood pressure (BP) [10, 36]. Hence, a separation
ma patients. However, they did not detect any difference be- between NTG and HTG is still usual in clinical practice [37].
tween NTG and HTG. According to their data, the decreased We have recently revealed that ocular blood flow was sig-
sympathetic and parasympathetic modulation during barore- nificantly reduced both in NTG and HTG compared to healthy
ceptor stimulation in the patients with HTG and NTG sug- subjects. The reduction of arterial ocular blood flow was more
gested that autonomic dysfunction that may contribute to the significant in HTG than in NTG while the lower venous blood
pathogenesis of both diseases [32]. flow was detected in NTG patients [18]. It was emphasized in
Mroczkowska et al. compared HTG and NTG patients with literature that the reduction of blood flow velocities in the
early stage of the disease using 24-h ambulatory blood pres- central retinal vein and central retinal artery was significantly
sure monitoring and measuring peripheral pulse-wave analy- associated with glaucoma progression both in NTG and HTG
sis, and carotid intima-media thickness. Retinal vascular reac- patients with well controlled IOP (21 mmHg or less) [17]. It
tivity to flicker light was assessed as well. Though the patients means that circulatory disorders may occur both in NTG and
 EPMA Journal (2018) 9:35–45

Table 2 The mean group values of HRV before and after cold provocation test in the normal tension glaucoma (NTG), high tension glaucoma (HTG), and control groups

HRV parameters NTG group HTG group Control group Statistical significance of differences between groups (Р)

NTGrest NTGCPT NTGrest NTGCPT HTGrest HTGCPT

⇕ ⇕ ⇕ ⇕ ⇕ ⇕
HTGrest HTGCPT Controlrest ControlCPT Controlrest ControlCPT
Rest CPT P Rest CPT P Rest CPT P P P P P P P

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
SDNN (ms) 35.3 ± 12.4 33.6 ± 11.3 0.08 27.2 ± 10.0 32.2 ± 13.2 0.014 42.03 ± 5.12 49.12 ± 9.02 0.004 0.016 0.43 0.20 0.004 0.001 0.001
RMSSD (ms) 34.2 ± 11.3 35.5 ± 16.4 0.71 28.1 ± 12.2 32.3 ± 16.6 0.29 35.6 ± 5.1 35.7 ± 16.2 0.98 0.046 0.39 0.17 0.23 0.014 0.038
HF (ms2) 242.4 ± 169.3 250.2 ± 10.1 0.78 216.4 ± 87.2 253.33 ± 10.12 0.78 307.12 ± 304 400.1 ± 149.3 0.085 0.045 0.37 0.58 0.003 0.35 0.024
LF (ms2) 233.2 ± 132.1 272.4 ± 180.2 0.82 236.1 ± 117 310.1 ± 191.3 0.024 743.5 ± 292.1 716.6 ± 360.3 0.90 0.03 0.81 0.96 0.54 0.23 0.18
S 1.61 ± 0.28 1.75 ± 0.37 0.95 2.40 ± 0.61 2.21 ± 0.77 0.65 1.04 ± 0.43 1.29 ± 0.42 0.02 0.14 0.80 0.09 0.001 0.001 0.001
ARI 315.3 ± 86.12 357.42 ± 129.1 0.91 453.2 ± 111 405.7 ± 104. 0.26 169.1 ± 35.2 189.38 ± 59.63 0.09 0.16 0.62 0.14 0.001 0.001 0.001
TP 1082 ± 924 964 ± 571 0.65 655 ± 528 918 ± 688 0.14 1773 ± 1542 2339 ± 2063 0.27 0.018 0.29 0.10 0.009 0.001 0.001

Mann-Whitney U test was used to assess the statistical significance of the groups. Italic value demonstrate significant difference of the compared values (for visual clarity)
CPT, cold provocation test; NTGrest, HTGrest, collective designation of belonging of values of NTG and HTG groups at rest, respectively; NTGCPT, HTGCPT, collective designation of belonging of values of
NTG and HTG groups after CPT, respectively; ⇕, symbol denoting the comparison of HRV values of two groups
41
42 EPMA Journal (2018) 9:35–45

SDNN

LF
SDNN LF
- NTG, - HTG, - Control - NTG, - HTG, - Control
A SDNNNTG − SDNNHTG (F=3.26, P=0.037), B LF NTG − LFHTG (t = 45, P=0.041).
(t = 45, P=0.001; SDNNNTG − SDNNControl
(F=8.74, P<0.001) (t=42.6, P<0.001);
SDNNHTG − SDNNControl (t=3.42, P=0.001).
ARI
TP

TP ARI
- NTG, - HTG - Control - NTG, - HTG - Control
C TP NTG − TPHTG (F=2.86, P=0.039); D ARI NTG − ARIHTG (F=7.89, P<0.001);
TPNTG − TPControl (F=8.56, P<0.001) ARINTG − ARIControl (F=8.74, P<0.001).
Fig. 1 Comparison of regression lines for HRV parameters in the NTG, (HRVrest). A positive slope of the line indicates an increase in ΔHRV at
HTG, and control subject groups. Note: the slope of regression lines HRVrest increase; a negative slope reflects a decrease in ΔHRVat HRVrest
characterizes the relation between two variables: shift in HRV increase.
parameter (ΔHRV = HRVCPT − HRVrest) after CPT and its value at rest

HTG despite the IOP level. One of the reasons is the increased In the present study, we observe a significant dipping of
sympathetic neural activity (SNA). It leads to an increase in diastolic BP both in NTG and HTG patients compared to
vascular resistance and especially under circumstances of the healthy subjects. This may be a consequence of the activation
endothelial dysfunction may have circulatory implications rel- of the sympathetic innervation. Chronic increased SNA can
evant to glaucoma pathogenesis. SNA causes an increase of lead to arterial and cardiac remodeling, endothelial dysfunc-
heart rate, stroke volume, and vasoconstriction. It regulates the tion, increased tissue oxygen demand, and subsequent de-
circadian variation of BP and is closely linked to nocturnal creasing of the ischaemia threshold in all organs, including
dipping. the eye. There is the evidence of the presence of the choroidal
EPMA Journal (2018) 9:35–45 43

Table 3 The intergroup

differences in HRV indicators, No. Compared indicators Values of the difference criteria Significance level
revealed by the method of
comparison of regression lines 1 SDNNNTG and SDNNHTG FFisher = 3.26 P = 0.037
tStudent = 45 P = 0.001
2 SDNNNTG and SDNNcontrol FFisher = 8.74 P < 0.001
tStudent = 42.6 P < 0.001
3 SDNNHTG and SDNNcontrol tStudent = 3.42 P = 0.001
4 LFNTG and LFHTG FFisher = 3.21 P = 0.05
5 TPHTG and TPHTG FFisher = 2.86 P = 0.039
6 TPHTG and TPcontrol FFisher = 8.56 P < 0.001
7 ARINTG and ARIHTG FFisher = 7.89 P < 0.001
8 ARINTG and ARIcontrol FFisher = 8.74 P < 0.001

neuroplexus represented by numerous internal autonomic As the result of this fact, we revealed the most important
ganglia, forming the autonomous perivascular network data of this study: a considerable increase in the activity of the
around the choroidal vessels [6]. It is believed that it plays sympathetic ANS in NTG patients in response to the CPT. The
vasodilatory function aimed at enhancing the ocular blood changes of the basic HRV parameters (SDNN, HF, LF, S, and
flow. Apparently, the vascular mechanisms of failure of optic ARI) after the CPT emphasize a significant difference between
nerve and retinal trophism and their autonomic regulation play HTG and NTG patients. It is known that an increase of the
a significant role in ocular physiology and pathophysiology in sympathetic ANS in response to the provocation tests is typ-
general and, particularly, in glaucoma. Vasoconstriction oc- ical for people with PVD. The development of NTG is asso-
curs in the setting of the predominance of sympathoadrenal ciated with possible PVD [2, 21]. However, currently this fact
influences on arterioles and capillaries, as well as due to the is not absolutely certain, and therefore, NTG is considered to
decreased activity of parasympathetic influences on retinal be a form of an open-angle glaucoma.
vessels. Although the role of PVD in the pathogenesis of GON has
Altered ocular blood flow or reduced visual field sensitivity been discussed for many years, only recent studies due to the
during sympathetic provocation tests has been demonstrated use of modern technologies could prove that patients with
in POAG patients [33, 38, 39]. NTG, but not healthy individuals, suffer from the retinal blood
The present study has identified the evidence of altered flow autoregulation failure in the conditions of provocation
MOPP both in HTG and NTG patients compared to healthy tests [5]. From this point of view, the dysfunction of the auto-
subjects. However, there were no difference in the MOPP nomic blood flow regulation seems to be of high importance
between HTG and NTG that is consistent with previous re- and its study attracts attention of the researches. Wierzbowska
search [23, 31, 38]. Such findings suggest that a considerable et al. studied HRV in NTG and revealed the sympathovagal
overlap may exist in the development of HTG and NTG, es- balance of ANS in NTG patients that shifted towards sympa-
pecially at the early stage of the disease [23, 40]. From this thetic activity with no change of 24-h pattern of BP variability
point of view, it has been assumed that provocation tests may as compared to the healthy subjects [33]. Na et al. also ob-
be needed to reveal alterations in cardiovascular function in served significantly decreased SDNN values in patients with
NTG patients [34]. In the present study, we applied CPT to NTG [30].
reveal the difference between NTG and HTG patients. Park and co-authors studied the NTG patients with differ-
Before CPT, a significant difference was revealed for all ent types of HRV and reported that VF progression in patients
HRV parameters at rest between both the glaucoma groups with sympathetic predominance is faster than that in patients
and between HTG and control subjects. The CPT confirmed with higher HRV. The authors concluded that the autonomic
a significant difference between glaucoma patients and control dysfunction, especially the decrease of SDNNs, was a predic-
subjects. tor of central VF progression in NTG [10].
This was the reason to compare regression lines for analyz- The new data, which testify to the influence of vascular
ing the intergroup differences. The mentioned method using factors on the development of NTG, are especially relevant.
the example of comparing the NTG and HNG groups by It can be assumed that violations of autonomic innervation
SDNN shows that significance of differences in groups can underlying PVD are an important cause of NTG development,
be established immediately by two characteristics: initial but not its specific feature. Being present in POAG patients,
values of the parameter at rest and its changes as a result of including HTG, an imbalance of the ANS can be also consid-
the reaction to CPT. ered as a risk factor for unfavorable course of GON. In any
44 EPMA Journal (2018) 9:35–45

case, the results obtained convincingly demonstrate the role of range; GON, glaucomatous optic neuropathy; HTG, high tension glauco-
ma; HRV, heart rate variability; ILM, internal limiting membrane; IOP,
PVD in the NTG pathogenesis. Our results that demonstrate intraocular pressure; LF, low frequency range; MD, mean deviation;
the predominance of SNA in NTG patients may be used for MOPP, mean ocular perfusion pressure; NTG, normal tension glaucoma;
distinguishing NTG from HTG. OCT, optical coherence tomography; ONH, optic nerve head; OPP, ocular
This conclusion has an important practical implication for perfusion pressure; POAG, primary open-angle glaucoma; PSD, pattern
standard deviation; PVD, primary vascular dysregulation; RMSSD, root
detecting NTG (or if it is suspected), determining the progno- mean square of the successive differences; RNFL, retinal nerve fiber
sis and choosing more appropriate therapy, as well as making layer; RPE, retinal pigment epithelium; SAP, standard automated
recommendations to patients concerning the proper lifestyle. perimetry; SBP, systolic blood pressure; SDNN, standard deviation of
Further studies are needed to verify our findings as well as the NN intervals; SD-OCT, spectral-domain optical coherence tomogra-
phy; SNA, sympathetic neural activity; TP, total spectral power; VF,
studies on any therapies that favorably influence ANS activity visual field
in patients with glaucoma.
Our study has several limitations that must be acknowl-
Open Access This article is distributed under the terms of the Creative
edged. First, we did not study the progression pattern of the
Commons Attribution 4.0 International License (http://
disease and we did not analyze the relation between the shift creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
of HRV parameters after CPT and functional loss in HTG and distribution, and reproduction in any medium, provided you give appro-
NTG, although it might reveal the influence of HRV on glau- priate credit to the original author(s) and the source, provide a link to the
coma progression and its difference between the two studied Creative Commons license, and indicate if changes were made.
glaucoma groups.
Second, we did not evaluate the circulatory parameters and
their relation to the HRV parameters. References
Third, we did not assess test-retest variability, though it is
known that HRV is a moderately to fairly good reliable mea- 1. Chan T, Bala C, Siu A, Wan F, White A. Risk factors for rapid
glaucoma disease progression. Am J Ophthalmol. 2017;180:151–
surement [41]. 7. https://doi.org/10.1016/j.ajo.2017.06.003 .
Fourth, MOPP was calculated using the formula: MOPP = 2. Flammer J, Konieczka K, Flammer AJ. The primary vascular dys-
(2/3 diastolic BP + 1/3 systolic BP) × 2/3 − IOP, though it has regulation syndrome: implications for eye diseases. EPMA J.
been emphasized in literature, that retinal venous pressure is 2013;4(1):14. https://doi.org/10.1186/1878-5085-4-14.
3. Flammer J, Konieczka K, Bruno RM, Virdis A, Flammer AJ,
very often markedly higher than IOP, that means that our Taddei S. The eye and the heart. Eur Heart J. 2013;34(17):1270–
calculation may not be adequate enough in some patients [42]. 8. https://doi.org/10.1093/eurheartj/eht023.
However, our study has an advantage over other studies on 4. Nicolela MT, Ferrier SN, Morrison CA, Archibald ML, LeVatte TL,
HRV in NTG and POAG as we did not include patients on Wallace K, et al. Effects of cold-induced vasospasm in glaucoma:
the role of endothelin-1. Invest Ophthalmol Vis Sci. 2003;44(6):
antihypertensive medication or topical medications that could 2565–72. https://doi.org/10.1167/iovs.02-0913.
influence BP and heart rate values. 5. Drance CM, Douglas DV, Schilzer M. Response of blood flow to
In summary, we found out that patients with NTG have warm and cold in normal and low tension glaucoma patients. Am J
more pronounced disturbance of the autonomic nervous sys- Ophthalmol. 1988;105(1):35–9. https://doi.org/10.1016/0002-
9394(88)90118-3.
tem than HTG patients that is manifested by the activation of 6. Grunwald J, Piltz J, Hariprasad S, DuPont J. Optic nerve and cho-
sympathetic nervous system in response to the CPT. This roidal circulation in glaucoma. Invest Ophthalmol Vis Sci.
finding refers to the NTG pathogenesis and suggests the use 1998;39(12):2329–36.
of HRV assessment in glaucoma diagnosis and monitoring. 7. Findl O, Rainer G, Dallinger S, et al. Assessment of optic disc blood
flow in patients with open-angle glaucoma. Am J Ophthalmol.
2000;130(5):589–96.
8. Galassi F, Sodi A, Ucci F, Renieri G, Pieri B, Baccini M. Ocular
Compliance with ethical standards The study was approved by the
hemodynamics and glaucoma prognosis: a color Doppler imaging
Ethical Committee (Institutional Review Board) at the Institution of the
study. Arch Ophthalmol. 2003;121(12):1711–5. https://doi.org/10.
Federal Medical and Biological Agency of the Russian Federation and
1001/archopht.121.12.1711.
was conducted in accordance with Good Clinical Practice within the
9. Flammer J, Orgül S, Costa V, Orzalesi N, Krieglstein G, et al. The
tenets of the Declaration of Helsinki. Each patient/healthy subject was
impact of ocular blood flow in glaucoma. Prog Retin Eye Res.
required to sign an informed consent form before being enrolled in the
2002;21(3):359–93. https://doi.org/10.1016/S1350-9462(02)
study and prior to any measurements being taken.
00008-3.
10. Park HYL, Park SH, Park CK. Central visual field progression in
Conflict of interest The authors declare that they have no conflicts of normal tension glaucoma patients with autonomic dysfunction.
interest. Invest Ophthalmol Vis Sci. 2014;55(4):2557–63. https://doi.org/
10.1167/iovs.13-13742.
Abbreviations ANS, autonomic nervous system; ARI, autonomic reg- 11. Drance S, Anderson DR, Schulzer M. Collaborative normal-tension
ulation index; BP, blood pressure; CPT, cold provocation test; DBP, dia- glaucoma study group. Risk factors for progression of visual field
stolic blood pressure; FLV, focal loss volume; FR, functional reserves; abnormalities in normal-tension glaucoma. Am J Ophthalmol.
FS, functional status; GCC, ganglion cell complex; GLV, global loss 2001;131(6):699–708. https://doi.org/10.1016/S0002-9394(01)
volume; GON, glaucomatous optic neuropathy; HF, high frequency 00964-3.
EPMA Journal (2018) 9:35–45 45

12. Satilmis M, Orgul S, Doubler B, Flammer J. Rate of progression of 27. Wilcox RR. Comparing the slopes of two independent regression
glaucoma correlates with retrobulbar circulation and intraocular lines when there is complete heteroscedasticity. Br J Math Stat
pressure. Am J Ophthalmol. 2003;135(5):664–9. Psychol. 1997;50(2):309–17. https://doi.org/10.1111/j.2044-8317.
13. Zink J, Grunwald J, Piltz-Seymour J, Staii A, Dupont J. Association 1997.tb01147.x.
between lower optic nerve laser Doppler blood volume measure- 28. Stanton AG. Primer of biostatistics. New York Toronto. 1994:459.
ments and glaucomatous visual field progression. Br J Ophthalmol. 29. Berntson GG, Uchino BN, Cacioppo JT. Origins of baseline vari-
2003;135(12):1487–91. ance and the law of initial values. Psychophysiology. 1994;31(2):
14. Phelps CD, Corbett JJ. Migraine and low tension glaucoma. A case- 204–10. https://doi.org/10.1111/j.1469-8986.1994.tb01042.x.
control study. Invest Ophthalmol Vis Sci. 1985;25(8):1105–8. 30. Na KS, Lee NY, Park SH, Park CK. Autonomic dysfunction in
15. Neves VJ, Moura MJ, Almeida BS, Costa R, Sanches A, Ferreira R, normal tension glaucoma: the short-term heart rate variability anal-
et al. Chronic stress, but not hypercaloric diet, impairs vascular ysis. J Glaucoma. 2010;19(6):377–81. https://doi.org/10.1097/IJG.
function in rats. Stress. 2012;15:138–48. 0b013e3181c4ae58.
16. Gherghel D, Hosking SL, Cunliffe IA. Abnormal systemic and 31. Riccadonna M, Covi G, Pancera P, et al. Autonomic system activity
ocular vascular response to temperature provocation in primary and 24-hour blood pressure variations in subjects with normal and
open-angle glaucoma patients: a case for autonomic failure? high tension glaucoma. J Glaucoma. 2003;12:156–63.
Invest Ophthalmol Vis Sci. 2004;45(10):3546–54. https://doi.org/ 32. Brown CM, Dutsch M, Michelson G, Neundorfer B, Hilz M.
10.1167/iovs.04-0290. Impaired cardiovascular responses to baroreflex stimulation in open
17. Kaiser HJ, Schoetzau A, Stümpfig D, Flammer J. Blood-flow ve- angle and normal-pressure glaucoma. Clin Sci. 2002;102(6):623–
locities of the extraocular vessels in patients with high-tension and 30. https://doi.org/10.1042/cs1020623.
normal-tension primary open-angle glaucoma. Am J Ophthalmol. 33. Wierzbowska J, Wierzbowski R, Stankiewicz A, Siesky B, Harris
1997 Mar;123(3):320–7. https://doi.org/10.1016/S0002-9394(14) A. Cardiac autonomic dysfunction in patients with normal tension
70127-8. glaucoma: 24-h heart rate and blood pressure variability analysis.
Br J Ophthalmol. 2012;96(5):624–8. https://doi.org/10.1136/
18. Kurysheva N, Kiseleva T, Irtegova E, Parshunina O. Venous ocular
bjophthalmol-2011-300945.
blood flow in primary open angle glaucoma. Int J Ophthalmol Eye
34. Bossuyt J, Vandekerckhove G, Backer T, Velde S, et al. Vascular
Res. 2015;3:1):1–7.
dysregulation in normal-tension glaucoma is not affected by struc-
19. Feke GT, Pasquale LR. Retinal blood flow response to posture
ture and function of the microcirculation or macrocirculation at rest.
change in glaucoma patients compared with healthy subjects.
Medicine. 2015;94(2):425.
Ophthalmology. 2008;115(2):246–52. https://doi.org/10.1016/j.
35. Ahrlich KG, De Moraes CG, Teng CC, et al. Visual field progres-
ophtha.2007.04.055.
sion differences between normal-tension and exfoliative high-
20. Narita K, Murata T, Hamada T, Takahashi T, Omori M, Suganuma
tension glaucoma. Invest Ophthalmol Vis Sci. 2010;51(3):1458–
N, et al. Interactions among higher trait anxiety, sympathetic activ-
63. https://doi.org/10.1167/iovs.09-3806.
ity, and endothelial function in the elderly. J Psychiatr Res.
36. Choi J, Jeong J, Cho HS, Kook MS. Effect of nocturnal blood
2007;41(5):418–27. https://doi.org/10.1016/j.jpsychires.2006.01.
pressure reduction on circadian fluctuation of mean ocular perfu-
003.
sion pressure: a risk factor for normal tension glaucoma. Invest
21. Konieczka RR, Traverso C, Myung D, Kida T, Kurysheva N, et al. Ophthalmol Vis Sci. 2006;47(3):831–6. https://doi.org/10.1167/
Flammer syndrome. EPMA J. 2014;5(1):11. https://doi.org/10. iovs.05-1053.
1186/1878-5085-5-11. 37. Thonginnetra O, Greenstein VC, Chu D, Liebmann JM, Ritch R,
22. Gasser P, Flammer J. Blood-cell velocity in the nailfold capillaries Hood DC. Normal versus high tension glaucoma: a comparison of
of patients with normal-tension and high-tension glaucoma. Am J functional and structural defects. J Glaucoma. 2010;19(3):151–7.
Ophthalmol. 1991;111(5):585–8. https://doi.org/10.1016/S0002- https://doi.org/10.1097/IJG.0b013e318193c45c.
9394(14)73703-1. 38. Gherghel D, Hosking SL, Cunliffe IA, Heitmar R. Transient cardiac
23. Mroczkowska S, Benavente-Perez A, Negi A, Sung V, Patel SR, ischaemia and abnormal variations in systemic blood pressure in
Gherghel D. Primary open-angle glaucoma vs normal-tension glau- unselected primary open angle glaucoma patients. Ophthalmic
coma: The vascular perspective. JAMA Ophthalmol. 2013;131(1): Physiol Opt. 2010;30(2):175–81. https://doi.org/10.1111/j.1475-
36–43. https://doi.org/10.1001/2013.jamaophthalmol.1. 1313.2009.00704.x.
24. Ryabova Т, Shlapak V, Dvorkovich A. Device for automatic re- 39. Chai E, Goldberg I, Chia A, Chen J. Visual field responses to a hand
cording and processing the pulse curve. Finnish Patent FI100939 vibration stimulus. Surv Ophthalmol. 1999;43(1):79–86.
В, Class А 61 B5/024 dated 30.03.1998. 40. Koch E, Staab J, Fuest M, Witt K, et al. Blood pressure and heart
25. Task force of the European Society of Cardiology and the North rate variability to detect vascular dysregulation in glaucoma. J
American Society of Pacing and Electrophysiology. Heart rate var- Ophthalmol. 2015;2015:1–9. https://doi.org/10.1155/2015/798958.
iability. Standards of measurement, physiological interpretation and 41. Piepoli M, Radaelli A, Ponikowski P, et al. Reproducibility of heart
clinical use. Institute of cardiology equipment. 2000:64. rate variability indices during exercise stress testing and inotrope
26. Weise F, Laude D, Girard A, Zitoun P, Siché JP, Elghozi JL. Effects infusion in chronic heart failure patients. Clin Sci. 1996;91:87–8.
of the cold pressor test on short-term fluctuations of finger arterial 42. Flammer J, Konieczka K. Retinal venous pressure: the role of
blood pressure and heart rate in normal subjects. Clin Auton Res. endothelin. EPMA J. 2015;26(6):21. https://doi.org/10.1186/
1993;3:303-10. s13167-015-0043-1.