Hormone Molecular Biology and Clinical Investigation. 2019; 20180060

Hanis Hidayu Kasim1 / MaizatulAzma Masri2 / NorAzila Noh3 / Azlina Mokhtar4 /
Rafidah Hanim Mokhtar1
Clinicalimplicationsof blood pressure variability (BPV)in pregnancies: a review
1Medical Sciences (Physiology), Universiti Sains Islam Malaysia, Kuala Lumpur, Malaysia, Phone: +603-4289 2400, Fax : +603- 4289 2477, E-mail: [email protected]
2Obstetrics and Gynaecology, Universiti Sains Islam Malaysia, Kuala Lumpur, Malaysia
3Medical Sciences (Physiology), Universiti Sains Islam Malaysia, Kuala Lumpur, Malaysia
4Surgical Based Discipline, Universiti Sains Islam Malaysia, Kuala Lumpur, Malaysia Abstract:
Background: Hypertension disorder in pregnancy (HDP) is the second most common contributor to maternal morbidity and mortality worldwide. Blood pressure variability (BPV), with the assistance of ambulatory blood pressure monitoring (ABPM), measures blood pressure readings in pregnant women and has the potential to predict the occurrence of pregnancy-induced hypertension (PIH) or preeclampsia (PE) before any symptoms develop.
Methodology: Studies involving ABPM among pregnant women were identified using electronic databases such as PubMed, Scopus, Google Scholar, ScienceDirect, Medscape, Ovid and ProQuest. These electronic databases were assessed from 1990 to 2018. Keywords used to search for literatures included a combination of BPV matched with pregnancy, pregnant women and HDP, gestational hypertension and/or PE.
Results: Out of 21,526 articles identified, a total of 10 studies met the criteria. Seven articles used the spectral analysis method while another two articles used a combination of spectral analysis, time domain and a non- linear method for BPV analysis. The final article described BPV as vagal baroreflex. Four articles agreed that high frequency (HF) BPV was mainly dominant from the second trimester until 4 days postpartum in HDP patients. This reflects the dominant features of parasympathetic activities among these patients. Two articles that used time domain also agreed that standard deviation (SD) BPV increased in PE patients.
Conclusions: In pregnancy, BPV has a strong impact on the knowledge understanding of the disease in clinical fields, allows a superior ability to predict PIH and PE in mid-pregnancy and offers potential value for address- ing hypertension in pregnancy.
Keywords: ABPM, blood pressure variability, BPV, hypertension disorder in pregnancy, PE, PIH DOI: 10.1515/hmbci-2018-0060
Received: August 2, 2018; Accepted: December 13, 2018

Introduction

Hypertensive disorder in pregnancy (HDP) encompasses chronic hypertension, gestational hypertension, preeclampsia (PE)-eclampsia and PE superimposed with chronic hypertension [1]. They complicate up to 10% of pregnancies worldwide and are associated with severe maternal obstetric and perinatal complications [2]. It is the second major cause that contributes to maternal mortality worldwide [3]. In the fetal and neonatal contexts, it may lead to preterm delivery, fetal intrauterine growth restriction (IUGR), low birth weight (LBW) and perinatal death [4], [5]. Due to these worrying acute and chronic effects of HDP towards mothers and their babies, the need for powerful diagnostic tools for HDP is critical.
Blood pressure (BP) variation is significantly different between nonpregnant healthy women and healthy pregnant women. In general, several studies have concluded that nonpregnant women, when compared to healthy men, have a lower BP, with a more significant decrease among normotensive pregnant women during the 2nd trimester [6], [7]. The differences are immense in pregnancies complicated with HDP or PE compared to normotensive pregnant women [8]. Going through all the research titles using the term variability , the objectives of the study were either related to circadian blood pressure/BP pattern or solely blood pressure variability (BPV). Table 1 explains the differences between circadian BP and BPV studies.
Hanis HidayuKasim is the corresponding author. ©2019 Walter de Gruyter GmbH, Berlin/Boston.

Before the increase of studies in BPV, circadian BP variation was the main topic that was found to be the mirror of the biological rhythm from cardiovascular (CV) system. Circadian BP variation is the average percent- age change in mean blood pressure (MBP) values at night compared with the daytime values. Theoretically, in humans, the circadian BP variation gives a thorough picture of a BP wave for 24 h. Within 24 h, a BP profile known as a morning surge and a dipping pattern may be recorded using ambulatory blood pressure monitor- ing (ABPM). ABPM is known to be superior to clinical BP. In cases of chronic hypertension, cross-sectional and longitudinal studies have shown that 24 h ABPM readings correlate closely with target organ damage and it is independent of mean blood pressure values [19], [20], [21]. However, in pregnancy, the study of circadian BP rhythm carries more significance. With the assistance of ABPM, the BP trends of healthy and complicated preg- nancies become more obvious and are predictable according to gestation age. A circadian BP study has shown that in normal pregnancy, BP decreased till 22 weeks of gestation and then increased up to pre-pregnancy BP until delivery [22], while in a complicated pregnancy, the mean BP readings were stable up to 22 weeks of gestation which then started to rise significantly in relation with gestational age [23]. This is consistent with an- other study that shows that ABPM usage for the study of circadian BP is far more powerful than conventional office measurement in predicting a BP trend and the early identification of pregnancy outcomes in terms of the presence of hypertension disorder in pregnancy.
This article focuses only on BPV studies concerning fluctuations of beat-to-beat BP to observe the clinical implications in pregnant groups. Although many BPV studies have been performed, the conclusions remain controversial. A BPV study can be done in less than 24 h or more based on the aims of a given study [24]. Fluctuations of blood pressure readings may reflect pathological conditions which might be a sign of disease progression [25]. BPV can be estimated by the standard deviation of the blood pressure values over a defined period of the day. Multiple studies that correlate BPV with pregnancy have been done since the 1980s. However, there has been no final conclusion of those studies in the pregnant women group in terms of the clinical benefits and patient care aspects. The most important objective in studying the BPV in pregnancy is to understand the alteration of physiological adaptation of the circulation system in pregnancy involving autonomic nervous system (ANS) and baroreflex system. A BPV study also showed potential in detecting early changes of blood pressure in pregnant women as early as at first trimester that differentiates normotensive and hypertensive pregnant women even before BP begins to rise. In addition to prediction purposes, the study of BPV promises better intervention for hypertensive patients in terms of protecting the target organs [26].

Results

In the beginning, 21,526 articles were identified using an electronic search engine based on specified keywords. After screening according to inclusion and exclusion criteria, only 10 articles met the requirements. The process is summarized in the flow chart (Figure 1). The study design of the articles chosen are very heterogeneous, where four of them are cross-sectional, five are prospective cohorts and one is animal study.

Figure 1: Flow chart: selection criteria.
To date, BPV studies in pregnancy have typically had two goals. The first is to study the subclinical activ- ities (baroreflex function) in normal and complicated pregnancies with relation to HDP and PE [9], [10], [12], [13], [14], [16], [18], [27] while the second is to determine the superiority of BPV values in predicting HDP and PE in pregnant women [11], [15], [17]. Due to physiological changes in pregnancy, baroreflex functions experi- ence major modification and can be monitored directly using two main CV variability signals; blood pressure and heart rate value. Baroreflex reflexes are predominantly triggered by an ANS component, sympathetic and parasympathetic (vagal). There are three primary methods to determine baroreceptor reflex sensitivity from HR and BP, namely the time domain method, frequency domain method and nonlinear method [24]. Under- standing the baroreflex activities in pregnancy has opened the door to reductions in mortality and morbidity related PE and HDP by creating a definite intervention. The exact cause of PE remains controversial. Targeting physiological changes as a guide for understanding this problem seems promising.
Different methods are reflected in the different findings of each study. Thus, a standard method of measure- ment and analysis should be properly planned for a BPV study. The best established method measurements and proposed mechanism of BPV was built by Parati et al. in 2013 [14]. The BPV method of study is time-based. The nature of variability can be described from beat-to-beat, or minute interval or between visits to clinic or hos- pital. The standard protocol was built by the Task Force of the European Society of Cardiology and the North American Society for Heart Rate Variability (HRV) study. Most studies discussed in this article have combined HRV and BPV assessment so as to study the whole function of autonomic activity and baroreflex function. As baroreflex functions are influenced by environmental circumstances, the protocol was built to see the changes under three main conditions, which are spontaneous and provoked hemodynamic alterations by the Valsalva maneuver and provoked by respiration.
In terms of BPV analysis, there was three main type of methods: the time domain method, frequency domain (spectral analysis) method and nonlinear method. To determine the relationship between BPV and pregnancy outcomes, the time domain method is useful. However, to better determine the subclinical aspects of sympa- thetic and parasympathetic activity, spectral analysis via the frequency domain [the fast Fourier transform (FFT) method] is the method of choice. This is because the frequency domain method consisting of low frequency (LF) and high frequency (HF) is responsible for explaining the dominant modulator behind the BP fluctua- tions. LF arose from respiratory influenced and modulated vagal activity while HF associated with vasomotor

and baroreceptor activity is modulated by sympathetic activity for blood pressure. Using a spectral analysis study, sympathovagal balance activity can be quantified. The most common method of calculation used is FFT. Basically, the blood pressure reading is along the time plane while the standard deviation (SD) value quan- tifies BP in the same plane. Spectral analysis then transforms the time plane into a frequency plane to give a three-dimensional (3D) look into what is behind a reading of blood pressure. Baroreflex is one important reflex that coordinates the heart rhythm by firing the feedback mechanism from activation of ANS (sympathetic and parasympathetic). With relation to pregnancy, there are enormous physiological changes. So, with the study of baroreflex and autonomic activity, a lot of information could be explained for diseases that arise only during the pregnancy period.

BPV study among normal pregnancy

Studies comparing nonpregnant women and healthy pregnant women basically aim to understand the au- tonomic activity of the CV system caused by physiological changes in normal pregnancy. Out of 10 studies discussed here, six involved a healthy pregnant women group matched with nonpregnant women. A cross- sectional study by Voss et al. which was done once between 28 and 40 weeks of gestation showed no significant different of BPV between healthy pregnant and nonpregnant women [17]. Another study took BPV measure- ments in three main subgroups of pregnant women, 20 normal pregnant women, 40 complicated pregnancy [20 pregnancy induced hypotension (PIH) and 20 PE patients] and another 20 nonpregnant women [16] once in the 3rd trimester. This study reports the variability as vagal baroreflex gain and measured BPV by the spectral analysis method. At the end of the study, it found that vagal baroreflex gain in normal pregnancy is signifi- cantly decreased compared to the nonpregnant group (p = 0.001). A limitation of this study was that it used vagal baroreflex gain in the replacement of LF and HF terminology, as at that time the research team disagreed with the assumption of LF reflects sympathetic activity. The strength of this study was that they used baroreflex measurement as per protocol that involved three main methods spontaneous heart rate and BPV, the deep breathing test and the Valsalva maneuver. From spontaneous HR and BPV results, they calculated integrated baroreflex gain through frequency bands. By definition, the baroreflex gain studied in that particular article is the ratio of the R-R interval of heart rate and systolic arterial blood pressure in spectra (α = ¼ p{RRI power/SBP power}) for LF or HF, which theoretically explained the relationship of heart period response to arterial BP changes. Even though it did not mention BP variability directly, the calculation results consistently describe the function of baroreceptor from systolic BP changes related to endothelial changes. Vagal baroreflex gain is literally compared HF bands between these two determinants. When vagal baroreflex gain is reduced, the re- sponse of the heart towards BP changes is delayed. This study thus showed that the heart response is delayed in normal pregnancy compared to the nonpregnant group. Indirectly, this is consistent with lower HF of HR and higher HF systolic BPV in normal pregnancy.
Another study that followed-up patients to term and did multiple BPV measurements throughout the gesta- tional age revealed that BPV levels were unchanged during normal pregnancy throughout pregnancy [18]. The objective was to evaluate BPV and other autonomic determinants values during different gestational periods. In one study, the team followed-up 82 women from before they were pregnant and started the BPV assessment at the 1st half and the 2nd half of the menstrual cycle [9]. Half of them got pregnant within a year and the test continued at gestational ages of 6, 8, 12, 16, 20 and 32 weeks. The results reported unchanged BPV in normal pregnancy compared to the pre-pregnancy state. The same conclusion held for the singular experimental study which found that BPV had no effect in healthy pregnant rats [13].
Five study results show consistently that BPV did not change throughout pregnancy in a healthy pregnancy, while one study described BPV as one of the components in their calculations of vagal baroreflex gain. This may indicate that there are statistically significant changes between nonpregnant and normotensive pregnant women.

BPV in complicated pregnancy

Studies that involved complicated pregnancies or high-risk pregnancies were aimed to determine the effect of autonomic function on the pathological condition. Rang et al. enrolled only high-risk patients with a previous history of PE or primigravida. The study reported eight cases of PE among 42 high-risk pregnant women. However, there were no changes reported in the BPV values of PE patients compared to normotensive pregnant and nonpregnant women. Another longitudinal study by Flood et al. followed-up with normal pregnant women until delivery (at recruitment) and found that in women that develop PIH, LF-BPV is increased at 28 weeks gestation [11]. However, no changes were seen in PE patients. The study by Flood et al. has the largest sample

size (n = 385). Both studies were consistent in showing that BPV is not influenced by PE state and thus concluded that PE is unpredictable with BPV assessment.
All the studies that enrolled patients with complicated pregnancies during the 3rd trimester were consistent with findings that HF-BPV is increased in PIH. All studies that enrolled patients with complicated pregnancies at the 3rd trimester were consistent with the finding that HF-BPV was increased in PIH. An early study by Ekholm et al. from 1997 enrolled 14 women who were diagnosed with PIH and matched with 16 normoten- sive pregnant women between 31 and 40 weeks of gestational age. This cross-sectional study assessed BPV once during the 3rd trimester at three frequencies: LF, MF and HF. As discussed LF and HF were markers for sympathetic and parasympathetic activities, respectively. Medium frequency (MF) contains the overlap of sympathetic and vagal activity [10]. The study reported an increase of both MF and HF of systolic BPV in PIH patients compared to normotensive pregnant women. HF is known to be related to respiration and is controlled parasympathetically. The assumption is that increment of systolic BPV in PIH patients is related to the changes of intrathoracic pressure and a rise of blood volume during inhalation and expiration. The MF increment in this study was explained as a sympathetic response for fluctuations of vascular tone and systemic vascular resistance. However, the ratio of MF/HF was the same in PIH and normal pregnant women, showing that sympathovagal activity remains balanced in PIH patients. A study by Silver et al. recruited 20 PE patients and 20 PIH patients matched with another 20 nonpregnant and 20 normotensive pregnant women between 34 and 38 weeks of gestation. Vagal baroreflex gain was reported to be further decreased in PIH and PE patient in comparison to normotensive pregnant women at the 3rd trimester.
A later study by Walther et al. in 2008 [12] observed the sensitivity and specificity of BPV with a combination of Doppler ultrasound (US) examination to predict PE. In normotensive patients who had abnormal findings based on Doppler US findings described as a perfusion disorder of uterine arteries at the 2nd trimester, BPV was measured between 18th and 26th weeks gestation. This method is unique because, at that time, the Doppler examination was one of the gold standards to predict PE in a clinical setting with a positive predictive accuracy (PPA) of 30%. Interestingly, all patients enrolled were normotensive and showed no signs or symptoms of PE at that time except Doppler US. They also measured BPV in three types of analysis, including time domain, frequency domain and the nonlinear method. All patients were followed-up until delivery and reported HF of both systolic and diastolic BPV were increased in patients who develop PE after 30 weeks gestational age with HF diastolic BPV led to the most significant changes in PE patients besides very low frequency (VLF) HRV. Baroreflex function measurement consisting of HRV, BPV and baroreflex sensitivity (BRS) was concluded to be the best prediction parameter of PE, with sensitivity and specificity of 87.5%, positive prediction accuracy (PPA) of 70%, and negative prediction accuracy (NPA) of 95%. The author suggested a combination method of Doppler US and baroreflex function measurements by ABPM as the best tools to predict PE.
In 2014 another research group attempted to determine the autonomic control from delivery to 4 days post- partum [14]. They compared the BPV, HRV and BRS in time, frequency domain and nonlinear method between normal pregnant, PE pregnant women matched with nonpregnant women. Of the 10 articles discussed here, this is the only study done during the postpartum period. The findings are supportive of physiological changes of blood pressure in pregnancy where LF BPV and HF BPV is high as well as the SD of BPV (time domain) in PE patients just before the delivery compared to normal pregnancy. Postpartum, due to sympathetic neural ac- tivation, BPV was also higher in normal pregnancy and comparable to PE patients. At the same time, BPV in PE gradually decreased towards normal pregnancy levels. This was assumed to be because in normal pregnancy, blood volume increment after delivery increased stroke volume and subsequently the systolic blood pressure. However, due to the pathological dysfunction of endothelium in PE, these physiological changes were often overlooked. It can be concluded that PE related changes started to disappear within 4 days postpartum.
From the gestational age point of view, only one study measured BPV at pre-pregnancy and in the 1st trimester [9] then continued until 32 weeks. The result showed unchanged baroreflex function throughout nor- mal pregnancy compared to the pre-pregnancy state. Two studies assessed BPV during the 2nd trimester. One study found an increase LF BPV in PIH, but not in PE [11], while the other study was inconsistent as within the 2nd trimester, there were statistically significant increments of HF BPV in the PE group compared to nor- mal pregnancy [12]. In the 3rd trimester, all authors agreed that HF BPV is higher in PIH [10] and in PE until delivery [14] and in both PE and PIH patients [16]. For the postpartum period, only one study was done. The results showed an increased LF and HF BPV, and the SD BPV among PE and PIH patients compared to normal pregnancy and also statistically significant higher HF and LF BPV when compared to the normal pregnancy and nonpregnant groups.

Short-termBPV study
All studies discussed in this article were assessment of short-term and beat-to-beat BPV values. The neural CV activity is modulated by the ANS. This dynamic activity is an important mechanism to act in any delay of

baroreflex response from seconds to minutes. Such a response in terms of sympathetic activation will increase heart rate, while parasympathetic activation will decrease the heart rate as well as the stroke volume and cardiac output. In pregnancy, physiological changes are a major issue to be understood. The effects of baroreflex and autonomic function may be affected or not in pregnancy; however, with the existence of PIH and PE events which are highly risky to the mother and fetus, this BPV and HRV study will give information on the real pathological and subclinical ongoing process. Short-term BPV study is the best method of choice to describe the changes in CV baroreflex activity. All the article descriptions are simplified in Table 2.

Table 2: Comparison between circadian BP study and blood pressure variability (BPV) study.
Research/Clinical Implications Circadian BP Blood pressure variability

Definitions
Average percentage change in mean blood pressure (MBP) values at night compared with the daytime values throughout 24 h or 48 h
Study of beat to beat BP fluctuation in a period of time either short-term (<24 h) or long-term (visit-to-visit)

Results Report as
Day and night trend morning surge
dipping,
According to method: Time domain, e.g.: SD
Spectral analysis, e.g.: LF, MF, HF non linear

Theory of disease
Non sinusidal waveform of BP in pregnant woman as physiological adaptation in pregnancy state
Baroreflex activity differentiate PIH and PE pathophysiology
PIH was known to be the result of maladaptation of the ANS
PE started by inflammatory process and dysfunctional endothelia which later interrupting the ANS
HF BPV (modulator by the parasympathetic nervous system) higher since the 2nd trimester till day 4 postpartum in PIH patients

Predictions profile
Different BP pattern between normotensive and complicated pregnancy:
in normotensive pregnancy: BP was decreased till 22 weeks of gestation and then increased up to pre-pregnancy BP till delivery
In hypertensive pregnancy: Stable mean BP up to 22 weeks of gestation which then started to raise significantly in relation with gestational age
Specific changes of BPV differentiate normotensive and complicated pregnancy:
variability parameters (BPV and HRV) started to change from 2nd trimester onwards in PIH/PE pregnancy group, where diagnosis of HDP defined after 20 weeks of gestational age

Potential clinical benefits

Early screening for high risk pregnant women

1.Early screening for high risk pregnant women
2.Potential for pharmacological study to tackle the ongoing pathophysiology

Limitations

1.Duration of study: 24 h or 48 h
2.ABPM: patient wear at home may cause error in reading

1. Short term: no BP reading recorded during sleep

Strengths
1.Gives detailed result of BP reading in 24 h
2.Real BP reading during individual routine activity
1.Duration: can be done as short as 5 min
2.Specific clinical information of the ANS

ANS, autonomous nervous system; HF, high frequency; HRV, heart rate variability; LF, low frequency; MF, medium frequency; PE, preeclampsia; PIH, pregnancy-induced hypertension; SD, standard deviation.

Different style ofstudydesign

Among the10 articles discussed here, four used a cross-sectional designated study to observe the BPV in preg- nancy [10], [11], [16], [17]. Three out of four studies were done among normal pregnancies and only one study done among normal and complicated pregnancies based on this method [16]. Surprisingly, the results in all these studies are very contradictory to one another. Voss et al. found no significant different of BPV between normal pregnancy and nonpregnant women. Both studies by Ekholm et al. and Flood et al. found there is an increment of BPV in pregnant women that develop PIH later in pregnancy, however, the first study suggests an increment of MF and HF BPV, while the later one observed only LF BPV increment. Hence these two studies gave different interpretations and different suggestions at the end of their study. While the study by Silver et al. presented the BPV as vagal baroreflex interpreted as higher HF BPV in PIH and PE [16]. However, the list of cross-sectional studies discussed here still differs on the exact time of the BPVs were measured either in the 1st, 2nd or 3rd trimester or at several times during the pregnancy.
On the other hand, five articles were based on a prospective cohort study [9], [12], [14], [15], [18]. Four of the studies were performed among complicated pregnancies or high-risk pregnancies while only one study was done among normal pregnant women. The results reported were more homogenous whereby three out of four studies among complicated pregnancies revealed an increment of HF BPV and another two studies showed the unchanged BPV throughout the pregnancies in normal and complicated pregnancies. Furthermore, there were also two studies in this category that interpreted BPV in the time domain (SD) and both of the studies reported similar findings of increments of BPV in the SD among high-risk pregnant women who later developed PE [14], [15]. Overall, this makes a total of five out of 10 studies that recorded changes of HF BPV which were similar either using a cross-sectional or cohort study design.
In order to discuss the effect of different study designs towards the interpreted BPV results, more study is needed. Human BP may differ beat by beat. However, in pregnancy, we are still unsure of the physiological significance of BP fluctuations throughout a normal pregnancy, unlike the specific trend observed in BP and PR in every trimester of normal pregnancy which primarily reflects the normal physiological CV changes. Is the BPV going to change as the pregnancy grows or is it going to adapt to the physiological changes of the CV system which may be present with unchanged BPV? The various study designs are useful to answer the puzzles where if there are any changes recorded, what is the specific trend observed either in a normal or a complicated pregnancy.

Discussion

As discussed earlier, complications associated with HDP are critical for both the mother and the fetal/neonatal status. Diagnosing HDP using clinical BP seems to be at a dead end, failing to catch single abnormal readings during antenatal visits. Statistically, factors related to HDP or PE are among the top five leading causes of maternal deaths. Current science has provided evidence that ABPM is able to catch abnormal readings which are very significant in predicting the occurrence of HDP. Circadian BP rhythm has shown differences between normotensive and hypertensive pregnant women from before 20 weeks gestational age. BPV study has a high impact on the knowledge and understanding of the disease, and has a superior ability to predict PIH and PE in mid-pregnancy and has a potential value for intervention purposes in hypertension in pregnancy. In the future, we suggest looking for the relationship of BPV with the severity of HDP and also the evaluation of BPV changes in pregnant women on antihypertensive medications to determine the superiority of BPV control instead of mean BP control with maternal and fetal outcomes.

Theory of disease

The most important implication from the study of BPV is the contribution to the general body of knowledge. The pathophysiologies of PE and PIH are not the same. PIH was known to be the result of maladaptation of the ANS, while the pathophysiology of PE remains debatable. One assumption by Flood et al. is that PE is a systemic inflammatory condition which causes damage to all vascular wall. This is consistent with the findings of abnormal perfusion in uterine arteries as seen by Doppler US and may support the pathology of HELLP syndrome in PE patients.
Looking at the frequency domain results of BPV analysis, HF BPV was mainly dominant from the 2nd trimester till 4 days postpartum. This reflects the dominant features of parasympathetic activity in PIH patients. Other studies also observed the increment of LF BPV and MF BPV in the 2nd and 3rd trimester. Comparison

with normal pregnancy showed that there were distinct disturbances in the neural control of variability pa- rameters relevant to PIH and PE cases. These neural control changes were assumed to be the response from the increment of BP in PIH cases that triggered the parasympathetic activity to reduce the BP. This is where the difference of PIH and PE can be seen as in PIH, the BP changes are de novo triggered parasympathetic activity, while in PE, the systemic endothelial dysfunction is responsible for causing alterations in the neural CV control. Dysfunctional endothelial in PE was explained in a previous study [28] with evidence that released inflammatory molecules damaged the endothelial wall predominantly at night-time in PE.
The American College of Obstetricians and Gynecologists (ACOG) has recommended that in PIH, BP should return to normal in the postpartum period [2]. Walther et al. showed that, just before delivery in normal preg- nancy, systolic BP remains stable, but diastolic BP was high. In the postpartum period, due to changes in blood volume, systolic BP started to elevate, and diastolic BP remained at a high level and that caused nearing a normal pulse pressure value. The BP increment is higher compared to the nonpregnant group postpartum. In PE, due to the pathological background of the disease, the SBP increment is not related to the blood volume changes. BP remains high but comparable to the BP in the normal pregnant group. This is in contrast to the theory that PIH and PE are transient conditions in which pathologic changes are reversed directly after pla- cental removal [25]. The author only assessed the BP for 4 days postpartum and the previous theory is actually exclusive to the 12-week period postpartum. They concluded that the autonomic function is still modulated 4 days postpartum.
In other spectra, the first 48 h are known to be the most critical period of PE and even PIH patients to developing eclampsia (28%). In severe PE, the production of circulating antagonists to vascular endothelial growth factor-receptors is responsible for causing endothelial abnormality to the extent of causing cerebral ischemic and edema as the possible cause to developing seizures in eclampsia [29]. A recent study by Faber et al. suggested that the increase of VLF BPV followed by HRV in epileptic pregnant women may lead to the development of grand mal seizure at the 24th week of gestation [30]. VLF is mainly a reflecting sympathetic activity. However, no study was done to assess BPV changes before or during a seizure in the eclamptic patient.

Predictions profile
In our setting, we still depend on isolated BP measurements in diagnosing hypertension in pregnancy, which poorly predicts PE [31]. BPV together with Heart Rate Variability (HRV) has been suggested as a non-invasive method for predicting hypertensive disease in pregnancy [11] as well as PE [23].
Most of the papers (six) reviewed in this article show distinct alterations of BPV and HRV values as the modulator of ANS pathology at the certain gestational ages between normal pregnancy and PIH/PE. Based on these findings, the predictive profile of BPV has been confirmed. However, when the exact time the BPV should be checked in pregnant women has to be more precise. From the observation analysis among all the articles discussed here, variability parameters started to change from the 2nd trimester onwards. Combining the information from the circadian BP study, BP started to rise at around 22 weeks of gestation onwards [23]
and gestational hypertension was defined by an increase of BP of 140 mm Hg or more and 90 mm Hg or more in both systolic and diastolic BP, respectively. The 2nd trimester, around 20 weeks, may be the best time to assess any alterations of variability parameters in order to predict PIH and PE in pregnant women. To support this, the prediction of PE variability parameters (combination of BPV, HRV and BRS) alone has sensitivity and specificity of 87.5%, PPA 70% and NPA 95% [12]. It is thus suggested that combining the Doppler examination of the uterine arteries and measurement of variability parameters would the best predictor in mid-pregnancy for the occurrence of PE.
The next question is who should be examined for BPV changes. Following the risk factors listed in both the ACOG and the National Institute for Health and Care Excellence (NICE) hypertension in pregnancy guidelines, this simple and feasible test is suggested for this group because according to the method applied from all the articles discussed here, it only takes a minimum of 5 min or as long as 30 min to get the variability parameters. With advanced computerized devices, BPV measurement seems easier and more doable in an outpatient center and enhance the quality of pregnancy. However, a larger sample size is needed to decide on the frequency of parameters assessment by comparing and following up intraindividual BPV changes within different gesta- tional ages. This information helps us to stratify more specific and smaller circles of subsets among the listed high-risk pregnant women in order to predict PIH/PE and later the need for prophylaxis to prevent PE.

Prognostic profiles
The severity of PIH or PE in relation to BPV has not been reported in any of the 10 papers reviewed. To evaluate the power of BPV in giving a prognostic value among PIH and PE patients, we recommend studies to follow

up a set of pregnant women who have been already diagnosed with PIH and PE and serial intraindividual variability parameters are necessary. The foreseen outcomes include severe PE, eclampsia or death and adverse neonatal outcomes.
As in chronic hypertension in the nonpregnant group, BPV information in terms of ABPM gives beneficial input towards prognostics in terms of CV morbidity and mortality. A longitudinal study of PE patients should continue after pregnancy to observe the effects of BP fluctuations from pregnancy until after pregnancy.

Intervention benefit

In medical practice, all pregnant women need to be seen at regular follow-ups in the antenatal clinics. HDP were diagnosed based on BP findings of >140 mm Hg for systolic and/or >80 mm Hg for diastolic BP. The presence of proteinuria or other signs of organ dysfunction (renal insufficiency, liver, neurological or hematological com- plications, uteroplacental insufficiency) marked the diagnosis of PE. Managing HDP in all pregnant women can be challenging but crucial at the same time as the background of understanding the disease is still superfi- cial and interventions should be discussed and studied. Basically, the main management steps in dealing with HDP can be simplified into three parts. The first part is by doing prevention from the initial state to the more advanced hypertensive status or any mother and fetal complications. Secondly, it is important to maintain the BP at the optimal level by the use of medications as well as to monitor the fetal growth profiles and the lastly, postpartum care and follow up are important. To achieve the first aim, a prophylaxis that was widely used is an antiplatelet prophylaxis (aspirin 60 80 mg/day) and at least of 1 g/day of calcium intake prophylaxis and for treating high BP in pregnancy, first line antihypertensives used in pregnancy are methyldopa and labetalol.
One of the expected contributions from the study of BPV is an understanding of the subclinical aspects of PIH and PE and then intervening directly with the causal factors. The current treatment of hypertension in pregnancy aimed for the reduction of the mean of office BP, approaching target BP. Another important potential contribution is to evaluate the efficacy of antihypertensive medications or prophylaxis pharmacology in the control of BP for 24 h rather than the mean clinical BP during antenatal visits. White coat hypertension in pregnant women was also underestimated and the antihypertensive prescription may be unnecessary [32]. In the nonpregnant group, the BPV value gives additional and independent prognostic information on CV and non-CV risk mortality and morbidity in comparison to mean BP [33], [34] and is a good predictor of total organ damage (TOD) [35], [36]. It has also been suggested that antihypertensive treatment targeting the reduction of systolic BPV may be able to reduce stroke risk instead of just targeting reduction of mean BP [37], [38]. Unfortunately, discussions of this topic were not found among the 10 articles observed here.
The major limitation from this article is that the 10 articles discussed had a heterogeneous style of study design either cross-sectional, prospective cohort or animal study. This limitation is due to the minimum num- ber of previous studies that fulfilled our inclusion criteria. Overall, the study of BPV among pregnant women is still very minimal compared to other groups of subjects like the elderly and those with chronic diseases. Considering the potential deleterious harm to mothers and their offspring in relation to HDP, more BPV study among pregnant women is expected.

Outlook

Due to the high-quality information on BPV that could provide us, in terms of understanding the process of disease and prediction ability, in the next 5 10 years in combination with bio-engineering, ABPM machines are expected to be smaller, cheaper and more user-friendly. They may not only be used by clinicians but also patients themselves to monitor their own bodies. In line with the expansion of knowledge among non-health personnel, this could be a breakthrough allowing the preventive management of the disease. There may be other potential BPV and other variability parameters that remain poorly understood. Research efforts should focus on discovering these parameters.

Highlights

– BPV studies potentially provide a background of ongoing pathology of BP fluctuations in pregnant women. BPV studies have the ability to predict the occurrence of PE or PIH among pregnant women if monitored in
a group of risky patients.

Understanding information concerning the relationships among BPV with ANS activity and baroreflex ac- – tivity may have potential in a pharmacological context.
Further study on the evaluation of the power of BPV in terms of prognostic value among PIH and PE patients – is needed.
A standardized protocol on BPV and other variability parameters study is needed for more homogenous
conclusions.

Author statement

Research funding: Authors state no funding involved. Conflict of interest: Authors state no conflict of interest. Informed consent: Not applicable.
Ethical approval: Not applicable.

References

[1]Cunningham FG, Leveno KJ, Bloom FL, Spong CY, Dashe JS, Hoffman BL, Casey BM, Sheffield JS. Obstetrical Complications. WilliamsOb-
stetrics, 2014 ed. Vol. 24th. McGraw-Hill Education, 2014:728 79.51 pp. 978-0-07-179894-5 .
[2]Roberts JM, Druzin M, August PA, Gaiser RR, Bakris G, Granger JP, et al. Hypertension in Pregnancy. AmericanCollegeofObstetriciansand Gynecologists. 2013;(5):31 46. DOI: 10.1097/01.AOG.0000437382.03963.88 .
[3]Say L, Chou D, Gemmill A, Tunçalp Ö, Moller AB, Daniels J, et al. Global causes of maternal death: A WHO systematic analysis. Lancet Glob Heal. 2014;2:1 11.
[4]Sibai BM, Barton JR. Expectant management of severe preeclampsia remote from term: patient selection, treatment, and delivery indica-
tions. Am J Obstet Gynecol. 2007;196:514.e1 9.
[5]Sibai BM. Evaluation and management of severe preeclampsia before 34 weeks gestation. Am J Obstet Gynecol. 2011;205:191 8.
[6]Hermida RC, Ayala DE, Mojón A, Fontao MJ, Chayán L, Fernández JR. Differences between men and women in ambulatory blood pressure
thresholds for diagnosis of hypertension based on cardiovascular outcomes. Chronobiol Int. 2013;30:221 32.
[7]Manfredini R, Fabbian F, Pala M, Tiseo R, De Giorgi A, Manfredini F, et al. Seasonal and weekly patterns of occurrence of acute cardiovas-
cular diseases: does a gender difference exist? J Women s Heal. 2011;20:1663 8.
[8]Hermida RC, Ayala DE, Mojon A, Fernandez JR, Alonso I, Aguilar MF, Ucieda R, Iglesias M. Differences in circadian blood pressure variabil-
ity during gestation between healthy and complicated pregnancies. AmJHypertens. 2003;16:200 8.
[9]Rang S. Serial assessment of cardiovascular control shows early signs of developing pre-eclampsia. J Hypertens 2004;22:369 76.
[10]Ekholm EM, Tahvanainen KU, Metsälä T. Heart rate and blood pressure variabilities are increased in pregnancy-induced hypertension. Am J Obstet Gynecol. 1997;177:1208 12.
[11]Flood P, McKinley P, Monk C, Muntner P, Colantonio LD, Goetzl L, et al. Beat-to-beat heart rate and blood pressure variability and hyper- tensive disease in pregnancy. Am J Perinatol. 2015;32:1050 8.
[12]Walther T, Faber R, Stepan H. Analysis of heart rate and blood pressure variability in pregnancy new method for the prediction of preeclampsia. Proc First Int Conf Bio-inspired Syst Signal Process. 2008;24:47 55.
[13]Slangen B, Out I. Blood pressure and heart rate variability in early pregnancy in rats. Am J Physiol Hear Circ Physiol. 1997;273:H1794 9.
[14]Walther T, Voss A, Baumert M, Truebner S, Till H, Stepan H, et al. Cardiovascular variability before and after delivery: recovery from arte- rial stiffness in women with preeclampsia 4 days post partum. Hypertens Pregnancy. 2014;33:1 14.
[15]Malberg H, Bauernschmitt R, Voss A, Walther T, Faber R, Stepan H, et al. Analysis of cardiovascular oscillations: A new approach to the early prediction of pre-eclampsia. Chaos. 2007;17:015113.
[16]Silver HM, Tahvanainen KU, Kuusela TA, Eckberg DL. Comparison of vagal baroreflex function in nonpregnant women and in women with normal pregnancy, preeclampsia, or gestational hypertension. Am J Obstet Gynecol. 2001;184:1189 95.
[17]Voss A, Malberg H, Schumann AY, Wessel N, Walther T, Stepan H, et al. Baroreflex sensitivity, heart rate, and blood pressure variability in normal pregnancy. Am J Hypertens. 2000;13:1218 25.
[18]Moertl MG, Ulrich D, Pickel KI, Klaritsch P, Schaffer M, Flotzinger D, et al. Changes in haemodynamic and autonomous nervous system parameters measured non-invasively throughout normal pregnancy. Eur J Obstet Gynecol Reprod Biol. 2009;144(Suppl 1): S179 83.
[19]Frattola A, Parati G, Cuspidi C, Albini F, Mancia G. Prognostic value of 24-hour blood pressure variability. J hypertens. 1993;11:1133 7.
[20]Parati G, Ulian L, Santucciu C, Omboni S, Mancia G. Blood pressure variability, cardiovascular risk and antihypertensive treatment. J Hypertens Suppl. 1995;13:S27 34.
[21]Parati G, Lantelme P. Blood pressure variability, target organ damage and cardiovascular events. J Hypertens. 2002;20:1725 9.
[22]Ayala DE, Hermida RC. Predictable blood pressure variability in a healthy pregnancy assessed by longitudinal monitoring. Proc Annu Conf Eng Med Biol. 1990;12(pt 2):687 8.
[23]Hermida RC, Ayala DE, Iglesias M. Predictable blood pressure variability in healthy and complicated pregnancies. Am Hear Assoc [Inter- net]. 2001;38:736 41.

[24]Parati G, Ochoa JE, Lombardi C, Bilo G. Assessment and management of blood-pressure variability. Nat Rev Cardiol 2013;10:143 55.
[25]Mammaro A, Carrara S, Cavaliere A, Ermito S, Dinatale A, Pappalardo EM, et al. Hypertensive disorders of pregnancy. J Prenat Med 2009;3:1 5.
[26]Xie H-H, Zhang X-F, Chen Y-Y, Shen F-M, Su D-F. Synergism of hydrochlorothiazide and nifedipine on blood pressure variability reduction and organ protection in spontaneously hypertensive rats. Hypertens Res. 2008;31:685 91.
[27]Tejera E, Areias MJ, Rodrigues AI, Nieto-Villar JM, Rebelo I. Blood pressure and heart rate variability complexity analysis in pregnant women with hypertension. Hypertens Pregnancy. 2012;31:91 106.
[28]Bouchlariotou S, Liakopoulos V, Dovas S, Giannopoulou M, Kiropoulos T, Zarogiannis S, et al. Nocturnal hypertension is associated with an exacerbation of the endothelial damage in preeclampsia. Am J Nephrol. 2008;28:424 30.
[29]Mustafa R, Ahmed S, Gupta A. A comprehensive review of hypertension in pregnancy. J Pregnancy 2012;2012:105918.
[30]Faber R, Stepan H, Baumert M, Voss A, Walther T. Changes of blood pressure and heart rate variability precede a grand mal seizure in a pregnant woman. J Perinat Med. 2004;32:538 40.
[31]Hermida RC, Ayala DE, Mojón A, Iglesias M. High sensitivity test for the early diagnosis of gestational hypertension and preeclampsia. II. Circadian blood pressure variability in healthy and hypertensive pregnant women. J Perinat Med. 1997;25:153 67.
[32]Brown MA, Robinson A, Bowyer L, Buddle ML, Martin A, Hargood JL, Cario GM, et al. Ambulatory blood pressure monitoring in preg- nancy: what is normal? AmJObstetGynecol. 1998;178:836 42.
[33]Tatasciore A, Renda G, Zimarino M, Soccio M, Bilo G, Parati G, et al. Awake systolic blood pressure variability correlates with target-organ damage in hypertensive subjects. Hypertension. 2007;50:325 32.
[34]Zakopoulos NA, Tsivgoulis G, Barlas G, Spengos K, Manios E, Ikonomidis I, et al. Impact of the time rate of blood pressure variation on left ventricular mass. J Hypertens. 2006;24:2071 7.
[35]Matsui Y, Ishikawa J, Eguchi K, Shibasaki S, Shimada K, Kario K. Maximum value of home blood pressure: A novel indicator of target or- gan damage in hypertension. Hypertension. 2011;57:1087 93.
[36]Parati G, Pomidossi G, Albini F, Malaspina D, Mancia G. Relationship of 24-hour blood pressure mean and variability to serverity of target-organ damage in hypertension. J Hypertens. 1987;5:93 98.bpV
[37]Rothwell PM. Does blood pressure variability modulate cardiovascular risk? Curr Hypertens Rep. 2011;13:177 86.
[38]Rothwell PM, Howard SC, Dolan E, O Brien E, Dobson JE, Dahlöf B, et al. Effects of beta blockers and calcium-channel blockers on within- individual variability in blood pressure and risk of stroke. Lancet Neurol. 2010;9:469 80.