Obstetrics

Heart rate and blood pressure variabilities are increased in

pregnancy-induced hypertension
Eeva M.K. Ekholm, MD, PhD," Kari U.O. Tahvanainen, MSc, b and Tahla Metsfilfi, MD, PhD c
Turku and Tampere, Finland
OBJECTIVE" Our purpose was to study whether cardiovascular changes in pregnancy-induced
hypertension are associated with the increase in sympathetic control of hemodynamics and change in
sympathovagal balance.
STUDY DESIGN; Fourteen women with pregnancy-induced hypertension and 16 women with
uncomplicated pregnancies of similar duration were studied. Electrocardiographic signals and arterial
blood pressure (Finapres monitor, Ohmeda) were continuously measured noninvasively throughout the
study. Heart rate and blood pressure were measured while the subject was breathing (1) with her normal
tidal volume at a frequency of 15 breaths per minute and (2) as deeply as possible at a frequency of six
breaths per minute. Heart rate and systolic blood pressure variability were calculated with use of the
autoregressive model of spectral analysis.
RESULTS: Heart rate and systolic blood pressure variabilities were significantly increased in women with
pregnancy-induced hypertension compared with normotensive pregnant women. This increase was
greatest in the high frequency component of heart rate variability (p = 0.02) while the women were
breathing with a normal tidal volume. Further, the medium frequency (p = 0.03) and high-frequency
variabilities (p = 0.03) of systolic blood pressure were significantly increased in women with preeclampsia
compared with normotensive pregnant subjects.
CONCLUSIONS: Neural control of the heart rate and blood pressure are disturbed in pregnancy-induced
hypertension, as shown by increased heart rate and blood pressure variability. Both the sympathetic and
parasympathetic control of the heart rate and blood pressure appear to be increased. The maladaptation
of the cardiovascular system in women with pregnancy-induced hypertension is manifested as a lack of
the physiologic decline in cardiovascular oscillations. (Am J Obstet Gynecol 1997;177:1208-12.)

Key words: Pregnancy, heart rate variability, blood pressure variability, autonomic nervous
system

Blood pressure and vascular resistance are increased in

women with pregnancy-induced hypertension, the blood
volume is smaller, and blood pressure and heart rate
responses to various provocations are changed compared
with those of normotensive pregnant women. ~4 The
cause of this maladaptation of the cardiovascular system
to pregnancy is unknown. Essential hypertension is associated with hemodynamic changes similar to those of
pregnancy-induced hypertension. In patients with essential hypertension autonomic control of the cardiovascuFrom the Department of Obstetrics and Gynecology, University Central
Hospital, Turku, ~ the Department of Clinical Pkysiolo~, University
Hospital, Tampere,b and the Cardio~'espi~zto~y Researo~ Unit, Univer~
sity of Turku. c
Received for publicatio~ November 22, 1996; revised May 22, 1997,"
accepted June 5, 1997.
Reprint requests: Eeva Ekholm, MD, PhD, Department of Obstetrics and
@necology, University of Turk,u, Kiinamyllynkatu 4-8, FIN-20520
Turk,u, Finland.
Copyright 1997 by Mosby-Year Book, Inc.
0002-9378/97 $5.00 + 0 6/1/83738
1208

lar system shifts toward sympathetic activity, as shown by

greater low-frequency variability and slighter high-frequency variability of the heart rate. 5 The increased
susceptibility to arrhythmias, common in serious preeclampsia, 6 is also ascribed to enhanced sympathetic
activity.
This study was carried out to determine whether
pregnancy-induced hypertension is associated with increased sympathetic control of hemodynamics or with
changed sympathovagal balance. A thorough understanding of hemodynamic control is necessary for appropriate selection of medication and protection of the
delicate control mechanisms that are already disturbed
in pregnancy-induced hypertension.

Subjects
Fourteen women with pregnancy-induced hypertension and 16 women with uncomplicated pregnancies
were studied. Table I shows patient characteristics. Preg-

Ekholm, Tahvanainen, and Mets&l& 1209

Volume 177, Number 5

Am J Obstet Gy~ecol

Table I. Patient characteristics

Auscultatory blood pressure

Gestational week

PIH
Normotensive

Age (yr)

Nulliparous

At test

At delivery

Systolic

Diastolic

Birth weight (gin)

28 (18-36)
28 (24-34)

11/14
10/16

37 (31-40)
37 (31-40)

37 (31-41)
39 (37-41)

153 (130-190)*
120 (106-130)

105 (86-111)*
75 (60-88)

2895 (1220-4340)*
3405 (2890-4240)

Blood pressure measured in clinical setting. Data are expressed as median and range. PIH, Pregnancy-induced hypertension.
*p < 0.05 between groups by Student t test.

n a n c y - i n d u c e d hypertension was defined as a blood

pressure > 1 4 0 / 9 0 m m Hg o n two occasions d u r i n g the
last trimester of pregnancy. T h i r t e e n of the patients also
had noninfective proteinuria (>-0.3 gm/1). Diurnal proteinuria measured in six of the patients with pregnancyi n d u c e d hypertension r a n g e d from 1.1 to 16.6 gm per 24
hours. The patients were recruited from the perinatal ward
of Turku University Central Hospital, where they were
admitted for pregnancy-induced hypertension. The patients were studied o n the second day (median, range 1 to
17 days) of the hospital stay. None of them were receiving
any antihypertensive medication at the time of investigation. The control subjects were outpatients with gestational
ages similar to those of the patients with pregnancy-induced
hypertension. They remained normotensive throughout
their pregnancies. The study was approved by the Ethical
Committee of the Turku University Hospital, and all participants gave informed consent.
Methods
The electrocardiographic signal from lead V5 was
recorded t h r o u g h o u t the study. Arterial blood pressure
was measured noninvasively on a beat-to-beat basis from
the middle p h a l a n x of the left middle finger with use of
the finger-cuffmethod (Finapres 2300 BP monitor, Ohmeda). The R-R interval data a n d blood pressure values
were stored on-line on a microcomputer. Data acquisition a n d analysis were carried out with use of m e n u driven software (CAFTS, Medikro Oy, Finland). 7 To
avoid any errors potentially caused by hydrostatic pressure in blood pressure recordings, the left arm was fixed
to the heart level with a b a n d a g e on the front of the
chest. The subject was lying in a left lateral r e c u m b e n t
position to avoid compression of the inferior vena cava.
Because the cuffwas above the heart level in all subjects,
the recorded blood pressure values were consistently
lower than true values.
First, heart rate a n d blood pressure were recorded
while the subject was breathing with her n o r m a l tidal
volume for 7 minutes. The m i c r o c o m p u t e r s o u n d generator was used to pace the b r e a t h i n g rhythm to 0.25 Hz,
15 breaths per minute. Second, a deep-breathing test was
carried out to obtain maximal respiratory variability in
heart rate a n d blood pressure, s In this test the subject

b r e a t h e d slowly a n d deeply with b r e a t h i n g cycles of 10

seconds (0.1 Hz) for 4 to 5 minutes.
Analysis of heart rate and systolic blood pressure
variability. To quantify the periodic c o m p o n e n t s of
heart rate a n d systolic blood pressure variability, we used
spectral analysis. Low-frequency variability is associated
with sympathetic vasomotor regulation. M e d i u m frequency variability relates to baroreflex activity a n d is
m o d u l a t e d by both sympathetic a n d parasympathetic
control. High-frequency variability is vagally mediated
and reflects respiratory sinus arrhythmia. 9
A stationary region, free from ectopic beats, was defined from heart rate a n d blood pressure recordings.
The m e a n R-R interval, heart rate, systolic, a n d m e a n a n d
diastolic blood pressure values were calculated.
Power spectral densities were calculated on the basis of
R-R intervals a n d systolic blood pressures after linear
d e t r e n d i n g of the signals. Modified covariance autoregressive m o d e l i n g was used for spectral analysis. A model
order of 14 was chosen as the most suitable for the data.
Power spectra were quantified in four frequency bands:
total power from 0.00 Hz to 0.40 Hz, low-frequency
power from 0.00 to 0.07 Hz, medium-frequency power
from 0.07 to 0.15 Hz, a n d high-frequency power from
0.15 to 0.40 Hz. Sympathovagal balance was estimated by
dividing (1) the heart rate variability in the m e d i u m frequency b a n d by that in the high-frequency b a n d a n d
(2) systolic blood pressure variability in the m e d i u m frequency b a n d by that in the high-frequency band. 1
Statistical analysis. The Student t test was used for
statistical evaluation of the data. The data were analyzed
for n o r m a l distribution a n d were f o u n d to be suitable.
The data are expressed as m e a n (SD). T h e n u m b e r of
subjects was defined with power calculations that were
based on our previous studies with healthy p r e g n a n t
subjects.
Results
The clinical blood pressure measurements show that
blood pressure was significantly higher in women with
pregnancy-induced hypertension compared with normotensive women (p < 0.0001). The babies of the patients with
pregnaucy-induced hypertension were smaller than the
babies of the normotensive mothers (p = 0.04) (see Table

1210 Ekholm, Tahvanainen, and MetsN~.

November 1997
Am J Obstet Gynecol

Table II. Heart rate and blood pressure variability (mean and SD) during controlled breathing at frequency of 0.25 Hz
P I H women (n = 14)

Heart rate (beats/min)

Systolic BP (mm Hg)
Mean BP (mm Hg)
Diastolic BP (mm Hg)
Heart rate variability
Total power (ms 2)
Low-frequency power (ms 2)
Medium-frequency power (ms 2)
High-frequency power (ms 2)
Medium-to-high-frequency ratio
Blood pressure variability
Total power (ram Hg 2)
Low-frequency power (mm Hg 2)
Medium-frequency power (mm Hg 2)
High-frequency power (ram Hg 2)
Medium-to-high-frequency ratio

75
116
88
71
1397
562
200
596
47

Normotensive women (n = 16)

(13)
(19)
(11)
(10)
(I092)
(416)
(229)
(538)
(38)

16.0 (11.0)
11.2 (9.6)
2.6 (1.9)
2.1 (1.0)
153 (171)

Significance

77
97
70
55

(8)
(13)
(10)
(9)

p
p
p
p

=
=
=
=

0.60
0.004*
0.0001'
0.0001"

880
497
108
247
68

(688)
(592)
(79)
(165)
(59)

p
p
p
p
p

=
=
=
=
=

0.13
0.73
0.17
0.03*
0.27

p=
pp=
p=
p=

0.20
0.40
0.02*
0.03*
0.82

103.6 (11.4)
8.0 (11.1)
1.2 (0.8)
1.3 (0.8)
140 (140)

Blood pressure values shown were recorded with Finapres monitor and were consistently lower than true values. PIH, Pregnanc~
induced hypertension; BP, blood pressure.
*p < 0.05 between groups by Student t test.

Table IlL Heart rate and blood pressure variability (mean and SD) during controlled breathing at frequency of 0.1 Hz
PIH

Heart rate variability

Total power (ms ~)
Low-frequency power (ms 2)
Medium-frequency power (ms 2)
High-frequency power (ms 2)
Medium-to-high-frequency ratio
Blood pressure variability
Total power (ram Hg 2)
Low-frequency power (mm Hg 2)
Medium-frequency power (ram
Hg 2)
High-frequency power (mm Hg 2)
Medium-to-high-frequency ratio

3996
599
2830
502
1107

Normotensive

(2923)
(905)
(1829)
(540)
(774)

3767
452
2292
300
1298

35.2 (21.9)
11.6 (7.4)
21.4 (17.6)

(3188)
(354)
(1562)
(248)
(1020)

24.1 (15.0)
8.3 (5.8)
14.9 (14.9)

2.0 (2.6)
1808 (2078)

0.9 (0.7)
2308 (2630)

Significance

p
p
p
p
p

=
=
=
=
=

0.84
0.59
0.40
0.22
0.58

p = 0.12
p = 0.20
p = 0.29
p = 0.12
p = 0.58

PIH, Pregnancy-induced hypertension.

I). The resting heart rate in the study patients was similar to
that in the normotensive subjects. Systolic (p = 0.004),
diastolic (p = 0.0001), and mean blood pressures (p =
0.0001) were higher in women with pregnancy-induced
hypertension than in the control subjects (Table II).

Variability of heart rate and systolic blood pressure

during controlled breathing at 0.25 Hz. H e a r t rate and
systolic blood pressure variability were increased in pregnancy-induced hypertension. T h e increase was greatest
in the high-frequency c o m p o n e n t of heart rate variability
(p = 0.03). Further, m e d i u m - and high-frequency components of systolic b l o o d pressure variability were increased. T h e ratios between m e d i u m - and high-frequency variability were u n c h a n g e d (see Table II).
Variability of heart rate and systolic b l o o d pressure
during controlled deep breathing at 0.1 Hz. Power
spectral estimates of heart rate and systolic b l o o d pressure variability were greater in w o m e n with pregnancy-

i n d u c e d hypertension, although the difference between

the groups was n o t statistically significant. T h e ratio
between m e d i u m - and high-frequency variability did n o t
change (Table III).

Comment
We f o u n d that heart rate variability and systolic blood
pressure variability, as assessed with power spectral analysis, were increased in w o m e n with pregnancy-induced
hypertension c o m p a r e d with normotensive p r e g n a n t
women. T h e increase was greatest in the high-frequency
range.
T h e increase seen in high-frequency oscillation of the
heart rate implies that cardiac vagal m o d u l a t i o n is significantly c h a n g e d and vagal activity may be increased in
w o m e n with pregnancy-induced hypertension c o m p a r e d
with w o m e n with normotensive pregnancies. T h e highfrequency c o m p o n e n t of heart rate variability is vagally

177, Number5
AmJ Obstet Gynecol
Volume

mediated and u n d e r central and peripheral control. ~

The effect of central mechanisms is considered to be the
most important. 12 The peripheral factors that affect
high-frequency oscillation of the heart rate consist of
changes in the stimulation of baroreceptors, stretch
receptors in the lungs, and receptors in the right atrium
due to venous return. 1~ Central factors probably explain,
at least partially, the current results. Peripheral factors
may, however, play a role, because the differences between the patients with pregnancy-induced hypertension
and the normotensive pregnant controls showed a decrease while the women were breathing with maximal
tidal volume and at a slower frequency of 0.1 Hz. During
deep breathing the stimulation of the stretch receptors
of the lungs and the low-pressure receptors of the heart
is different from that occurring during breathing with a
normal tidal volume and at a higher frequency, affecting
afferent neural inputs. Changes in respiration are known
to influence heart rate spectra] ~
The high-frequency oscillation of systolic blood pressure was increased in women with pregnancy-induced
hypertension. This increase may be the result of the
observed changes in heart rate variability,9 but it may also
be a result of alterations in intrathoracic pressure and
blood volume during the respiratory cycle. 14 Previous
studies do not suggest that pregnancy-induced hypertension changes respiratory mechanics. 15
Sympathetic nerve activity to skeletal muscle has been
reported to be increased in preeclampsia, a6 Our finding
on the increase in the medium-frequency variation of
blood pressure implies that the sympathetic control of
blood pressure is also enhanced in pregnancy-induced
hypertension. It has been suggested that the mediumfrequency variation of blood pressure arises from sympathetically controlled ~ fluctuations in vascular tone and
peripheral resistance a7 that are partly of central origin. 9
The medium-frequency oscillations of blood pressure
and heart rate also reflect baroreflex-mediated vasomotor control. The increase observed in the mediumfrequency component of blood pressure variability may
indicate a change in baroreflex function in women with
pregnancy-induced hypertension compared with those
with normotensive pregnancy, in agreement with the
findings of previous studies. ~s
The medium-frequency to high-frequency ratios of
heart rate and blood pressure variability were similar in
women with pregnancy-induced hypertension and those
with normotensive pregnancies. These ratios are considered to be useful index values of sympathovagal balance. 1 Our results suggest that sympathovagal balance is
not affected by pregnancy-induced hypertension. Subjects with normotensive pregnancy showed a significant
decrease in vagally and sympathetically mediated heart
rate and blood pressure variabilities compared with
subjects with pregnancy-induced hypertension. This find-

Ekholm, Tahvanainen, and Mets~lfi 1211

ing is in agreement with those of previous studies showing that heart rate variability diminishes during normal
pregnancy compared with the n o n p r e g n a n t state in the
second trimester of pregnancy. ~9' ~0
The current study shows that the physiologic decrease
in heart rate variability is absent in patients with pregnancy-induced hypertension. The etiology of the decrease in heart rate variability or of the changes in heart
rate and blood pressure responses to various provocations in normotensive pregnancy is not known. Sympathicotonia2I and decreased vagal outflow resulting from
central or peripheral factors have been suggested as
causes of these pregnancy-associated changes. 9' 92 Animal studies support the concept of changed neural reflex
mechanisms underlying the altered pressor responsiveness. 23 Further, changes in baroreflex sensitivity may also
reduce hemodynamic oscillations,z4 The extensive
changes that occur in maternal hemodynamics during
pregnancy probably also play a role: the increased blood
volume alters the preload of the heart and thus may
account for the reduction in heart rate variability and the
heart rate and pressor responses to various provocations.25, 26 The maladaptation of the cardiovascular control system may be an appropriate response to an inadequate increase in plasma volume in pregnancy-induced
hypertension.
In conclusion, the neural control of the heart rate and
blood pressure is disturbed in pregnancy-induced hypertension. Both the sympathetic and parasympathetic control of heart rate and blood pressure seem to increase,
and the baroreflex function may be decreased in hypertensive pregnancy compared with normotensive pregnancy. The maladaptation of the cardiovascular system to
pregnancy in pregnancy-induced hypertension is manifested as a lack of the physiologic decline in cardiovascular oscillations.
REFERENCES

1. Airaksinen KEJ, Kirkinen P, Takkunen JT. Autonomic nervous dysfunction in severe pre-eclampsia. Eur J Obstet
Gynecol Reprod Biol 1985;19:269-76.
2. Castr6n O, Saarikoski S, Siimes A. Vascular reactivity in
normal pregnancy and in pregnancies complicated by hypertension. IntJ Gynaecol Obstet 1973;11:236-46.
3, Nisell H, Hjemdahl P, Linde B, Lunell N-O. Sympathoadrenal and cardiovascular reactivity in pregnancy-induced
hypertension. II. Responses to tilting. Am J Obstet Gynecol
1985;152:554-60.
4. Ekholm E, Piha S, Tahvanainen K, Antila K, Erkkola R.
Autonomic hemodynamic control in pregnancy-induced
hypertension. Hypertens Pregnancy 1994;13:253-64.
5. Guzzetti S, Piccaluga E, Casati R, Cerutti S, Lombardi F,
Pagani M, et al. Sympathetic predominance in essential
hypertension: a study employing spectral analysis Of heart
rate variability.J Hypertens 1988;6:711-7.
6. Naidoo D, Bhorat IE, Moodley J, Naidoo JK, Mitha AS.
Continuous electrocardiographic monitoring in hypertensive crises in pregnancy. Am J Obstet Gynecol 1991;164:
530-3.
7. Tahvanainen K, Lfinsimies E, Tikkanen P, Hartikainen J,

1212

8.
9.
10.

11.
12.
13.
14.

15.

16.

Ekholm, Tahvanainen, a n d MetsNfi

Kfirki T, Lyyra T, et al. Microcomputer-based monitoring of

cardiovascular functions in simulated microgravity. Adv
Space Res 1992;12:227-36.
Hirsch JA, Bishop B. Respiratory sinus arrhythmia in humans, how breathing pattern modulates heart rate. Am J
Physiol 1981;241:H620-9.
Akselrod S, Gordon D, MadwedJB, Snidman NC, Shannon
DC, Cohen RJ. Hemodynamic regulation: investigation by
spectral analysis. Am J Physiol 1985;249:H867-75.
Pagani M, Lombardi F, Guzzetti S, Rimoldi O, Furlan R,
Pizzinelli P, et al. Power spectral analysis of heart rate and
arterial pressure variabilities as a marker of sympatho-vagal
interaction in man and conscious dog. Circ Res 1986;59:
178-93.
SaulJP, Berger RD, Chen MH, Cohen RJ. Transfer function
analysis of autonomic regulation II. Respiratory sinus arrhythmia. Am J Physiol 1989;256:H153-61.
Melcher A. Respiratory sinus arrhythmia in man: a study in
heart rate regulating mechanisms. Acta Physiol Scand
1976(Suppl 435):7-31.
Brown TE, Beightol LA, Koh J, Eckberg DL. Important
influence of respiration on human R-R interval power
spectra is largely ignored. J Appl Physiol 1993;75:2310-7.
Saul JP, Berger RD, Albrecht P, Stein SP, Chen MH,
Cohen RJ. Transfer function of the circulation: unique
insight into cardiovascular regulation. Am J Physiol 1991;
261:H1231-45.
Rees GB, Broughton Pipkin F, Symonds EM, PatrickJM. A
longitudinal study of respiratory changes in normal human
pregnancy with cross-sectional data on subjects with pregnancy-induced hypertension. AmJ Obstet Gyneco11990;162:
826-30.
Schobel HP, Fischer T, Heuszer K, Geiger H, Schmieder RE.
Preeclampsia--a state of sympathetic overactivity. N Engl
J Med 1996;455:1480-5.

November 1997
Am J Obstet Gynecol

17. Parati G, SaulJP, Di Rienzo M, Mancia G. Spectral analysis

of blood pressure and heart rate variability in evaluating
cardiovascular regulation: a critical appraisal. Hypertension
1995;25:1276-86.
18. Wasserstrum N, Kirshon B, Rossavik IK, Willis RS, Moise KJ,
Cotton DB. Implications of sino-aortic baroreceptor reflex
dysfunction in severe preeclampsia. Obstet Gynecol 1989;
74:34-9.
19. Ekholm EMK, Piha SJ, Antila KJ, Erkkola RU. Cardiovascular autonomic reflexes in mid-pregnancy. BrJ Obstet Gynaecol 1993;100:177-82.
20. Ekholm E, Antila K, PihaJ, JalonenJ, Metsfil~iT, Erkkola R.
Changes in autonomic cardiovascular control in mid-pregnancy. Clin Physiol 1992;12:527-36.
21. Assali NS, Prystowsky H. Studies on autonomic blockade. II.
Observations on the nature of blood pressure fall with high
selective spinal anesthesia in pregnant women. J Clin Invest
1950;29:1369-75.
22. Ekholm E, Piha sJ, Antila KJ, Erkkola RU. Autonomic
cardiovascular responses in pregnancy: a longitudinal study.
Clin Auton Res 1994;4:161- 5.
23. Hines T, Lindheimer MD, Barron W. Total autonomic
blockade eliminates the attenuated pressor response to
angiotensin II in pregnant rats. Am J Physiol 1993;265:
R1270-5.
24. Ramsay MM, Broughton Pipkin F, Rubin PC. Pressor, heart
rate and plasma catecholamine responses to noradrenaline
in pregnant and nonpregnant women. BrJ Obstet Gynaecol
1993;100:170-6.
25. Py6rfilg T. Cardiovascular response to the upright position
during pregnancy. Acta Obstet Gynecol Scand 1966;
45 (Suppl 5):1-116.
26. Scott All, Tattersall RB, McPherson M. Improvement of
postural hypotension and severe diabetic autonomic neuropathy during pregnancy. Diabetes Care 1988;11:369-70.