Changes in Central Hemodynamics in Women With Hypertensive Pregnancy Between Before and After Delivery

Abstract

The authors tested the hypothesis that central hemodynamic parameters in women with hypertensive disorders of pregnancy (HDP) change between before and after delivery. A total of 137 pregnant women were studied: 72 with HDP, 42 with chronic hypertension (CH), and 23 with white‐coat hypertension (WCH; control group). Aortic augmentation index adjusted by heart rate 75 beats per minute (AIx@75), central pulse pressure (PP), total peripheral resistance (TPR), and cardiac output (CO) before and after delivery were recorded. AIx@75 and central PP were higher in the HDP group than in the control group, but both parameters declined after delivery until they were similar to the controls. AIx@75 and central PP, but not TPR or CO, were significantly decreased after delivery in the HDP group, but no such effects were seen in the other groups. These findings suggest that increased wave reflection caused by the stiffened aorta could be a key factor in the pathophysiology of HDP.

Hypertensive disorders of pregnancy (HDP), which include preeclampsia (PE), gestational hypertension (GH), and superimposed PE, but not chronic hypertension (CH), occur in approximately 5% of all pregnancies.1 In this paper, we defined HDP as either PE, GH, or superimposed PE, but did not include CH in this category. Even in developed countries, HDP accounts for approximately 16% of all of the causes of maternal death.2 The early diagnosis and management of HDP are thus very important. HDP can cause pregnancy‐associated complications such as preterm birth, a small‐for‐gestational‐age (SGA) infant, intrauterine growth restriction, and placenta abruption. Although various causal factors of HDP have been examined over many years, it is still not known which factor best predicts HDP.

In nonpregnant hypertensive patients, parameters of arterial stiffness such as carotid intima‐media thickness and pulse wave velocity are established predictors of cardiovascular diseases.3, 4 The augmentation index (AIx), a marker of wave reflection, is also an emerging marker of cardiovascular risk.5 It was reported that central blood pressure (BP) calculated by radial artery pulse wave analysis was a better predictor of cardiovascular prognosis than conventional brachial BP measurements.6, 7, 8 Khalil and colleagues reported that the AIx at 11 to 13 weeks of gestation was associated with better predictive values of PE compared with conventional brachial BP.9

A few studies have investigated AIx and central BP in pregnant women throughout the course of pregnancy, with conflicting results.10, 11 In addition, several studies have specifically examined AIx and central BP in patients with HDP9, 12, 13, 14, 15 and found that both parameters were higher in pregnant women with HDP than in healthy pregnant women.

Only limited data are available regarding the changes of the AIx and central BP between before and after delivery, and they show that the AIx in patients with HDP decreased after delivery.15, 16 However, these studies enrolled small numbers of patients and lacked sufficient power to draw definitive conclusions. Moreover, there have been no studies examining cardiac output (CO) and total peripheral resistance (TPR) using the same timing as the measurements of central hemodynamics in patients with HDP. In the present study, therefore, we considered how the indices of central hemodynamics in HDP change over the course of pregnancy. Specifically, we tested the hypothesis that in patients with HDP, the central hemodynamics increase during pregnancy and decrease after delivery, and that increased CO or TPR may be the mechanism underlying these changes. Our results add several important facts to the contributions made by previous studies, as detailed in Table 1.

Table 1.

Summary

What is already known on this subject

The central hemodynamic parameters such as the AIx and central BP are higher in patients with HDP during pregnancy than in normal pregnant women

The AIx in patients with HDP is increased during pregnancy but decreases after delivery

The TPR value in patients with HDP is higher than in normal pregnant women

The unique contributions of this study

The AIx@75 and central PP in the patients with HDP are higher than in controls during pregnancy but decline until they are identical to the control levels after delivery

The reductions of AIx@75 and central PP in patients with HDP after the delivery are not explained by the changes in TPR or CO

Other factors, such as angiogenic factors or sex hormones, might influence these central hemodynamic parameters

Abbreviations: AIx, augmentation index; AIx@75, augmentation index adjusted by heart rate 75 beats per minute; BP, blood pressure; CO, cardiac output; HDP, hypertensive disorders of pregnancy; PP, pulse pressure; TPR, total peripheral resistance.

Methods

Our study population consisted of 137 women who visited Jichi Medical University Hospital, Tochigi, Japan, from August 2009 to July 2013 and were initially diagnosed as having HDP. This study was approved by the institutional review board of our hospital and participants gave informed consent. Gestational age was defined by ultrasound for the first trimester screening by obstetricians. All patients were referred to the hypertension clinic in the same hospital for the parallel follow‐up. For referred patients with HDP, the medical history, ambulatory BP monitoring, and home BP measurements were re‐examined in order to confirm the diagnosis of HDP. All patients were classified as having white‐coat hypertension (WCH), CH, or transient hypertension during the first trimester based on the 2014 Japanese Society of Hypertension guidelines.17 Next, the patients were evaluated for possible subtypes of HDP, such as PE, GH, and superimposed PE, when the appearance of proteinuria and/or hypertension occurred after 20 weeks of gestation. The BPs in the first trimester were unavailable in some women, but we confirmed that the BP levels were normal in the first trimester in the HDP group by referring to maternity health records and referral letters.

Following the current guidelines, we defined GH as BP ≥140/90 mm Hg after 20 weeks of amenorrhea, with BP falling to normal within 12 weeks after delivery. PE was defined as proteinuria ≥300 mg/24 h in addition to the diagnostic criteria of GH.18 The diagnosis of superimposed PE (on CH) was based on the following criteria: new‐onset proteinuria ≥300 mg/24 h in hypertensive women without proteinuria before 20 weeks of gestation and a sudden increase in proteinuria or BP in hypertensive women with proteinuria before 20 weeks of gestation.

CH was defined as hypertension that was present and observable before pregnancy or was diagnosed before the 20th week of gestation, as well as hypertension that was diagnosed for the first time during pregnancy and did not resolve after delivery.19 WCH was defined as elevated clinic BP ≥140/90 mm Hg, but awake ambulatory BP <135/85 mm Hg and/or home BP <135/85 mm Hg. When there was a history of hypertension, the patient was classified as having CH. Patients with PE, GH, or superimposed PE were classified into the HDP group. Thus, patients were classified into three groups: the HDP, CH, and WCH groups. In theory, the control group would consist of entirely normotensive pregnant women, but these women were often reluctant to commit time to visiting our hospital after delivery. Because WCH has been reported to be a relatively benign state during pregnancy compared with PE and CH,20 patients with WCH were set as controls. Of note, when women with an initial diagnosis of WCH progressed to PE or GH, the final diagnoses were made as PE or GH. Therefore, the WCH group in our study consisted of patients with true WCH, and could reasonably be treated as a control group.

At a gestational age of approximately 37 weeks, we measured the indices of hemodynamics using both applanation tonometry and impedance cardiography. We repeated the same test within 6 months after delivery. The follow‐up rate was 93 of 137 (68%) patients. The reasons for the low follow‐up rate are described in the study limitations section.

Measurement of Hemodynamics and Wave Reflection

Applanation tonometry (SphygmoCor; AtCor Medical, Sydney, New South Wales, Australia) was performed to record radial pulse waves. These waves were calibrated with brachial cuff pressure and generated to aortic waveforms using a generalized transfer function. The patients' central BP, central pulse pressure (PP), and aortic AIx were calculated from the estimated aortic waveforms, which were previously validated.21 The AIx adjusted by heart rate (HR) (75 beats per min; AIx@75) was used in this study based on previous pregnancy studies.9, 13, 16 A single doctor (K.E.) performed the applanation tonometry three times for each patient. In difficult cases, at least one good‐quality waveform was adopted. The values with the highest operator index, a measure of good quality data by the SphygmoCor device, were used for analyses. In our previous investigation and another related study, the intraobserver reproducibility of this method was within the acceptable range.22, 23

Impedance cardiography (Task Force Monitor; CNSystems Medizintechnik, Graz, Austria) was performed to record CO and TPR. This impedance cardiographic device allows the noninvasive evaluation of CO, and constantly records the changes in impedance to current aortic flow through the chest between strategically placed electrodes. The flow volume of the aorta influences and changes the transthoracic impedance. By analyzing the beat‐to‐beat changes in transthoracic impedance, it is possible to calculate the CO and TPR. The CO and TPR values calculated by the Task Force Monitor were previously validated.24

Statistical Analysis

All statistical analyses were performed using SPSS (version 19.0; IBM‐SPSS Inc, Armonk, NY). The normality of the distribution of the data was examined with the Kolmogorov‐Smirnov test. Data are expressed as the mean±standard deviation or median (interquartile range) for normally and non‐normally distributed data, respectively. The hemodynamic values and the changes of the hemodynamic values before and after delivery among the three patient groups were compared using one‐way analysis of variance with Tukey post hoc tests and the Mann‐Whitney U‐test with Bonferroni post hoc tests for pair‐wise comparisons. A P value <.05 was considered significant.

Results

Characteristics of the Patients

The characteristics of the patients are shown in Table 2. Age and history of HDP were similar among the three groups, but body mass index (BMI) was lower in the HDP group than in the CH group. The number of antihypertensive agents used in the HDP and CH groups was significantly greater than that in the control group, but was similar between the HDP and CH groups.

Table 2.

Characteristics of the Patients

	HDP (n=72)	CH (n=42)	Controls (n=23)	P Value	Significant Pairs
Age, y	32.8±5.8	34.5±5.2	33.8±5.0	.3
BMI, kg/m2	26.2±5.3	30.4±6.7	27.9±6.5	.002	HDP vs CH
History of HDP, No. (%)	9 (13)	8 (19)	2 (9)	.5
Nulliparous, No. (%)	47 (65)	16 (38)	7 (30)	.002	HDP vs CH, HDP vs controls
Gestational age, wk	36.2±1.5	36.4±1.7	37.2±1.6	<.001	HDP vs CH, HDP vs controls
BP medications before delivery, No.	0.6±0.8	0.7±0.7	0.1±0.3	.03	HDP vs controls, CH vs controls
BP medications after delivery, No.	0.5±0.7	0.9±0.9	0.1±0.4	.005	CH vs controls
Heart rate, beats per min	84±15	87±11	88±11	.4
Brachial SBP, mm Hg	140±13	137±11	126±16	<.001	HDP vs controls, CH vs controls
Brachial DBP, mm Hg	87±10	84±11	74±11	<.001	HDP vs controls, CH vs controls
Brachial MAP, mm Hg	107±11	103±11	92±13	<.001	HDP vs controls, CH vs controls
Brachial PP, mm Hg	54±9	52±10	51±11	.6
Central SBP, mm Hg	127±14	122±12	109±15	<.001	HDP vs controls, CH vs controls
Central DBP, mm Hg	89±11	86±11	76±12	<.001	HDP vs controls, CH vs controls
Central MAP, mm Hg	107±11	103±11	92±13	<.001	HDP vs controls, CH vs controls
Central PP, mm Hg	39±9	36±9	33±8	.02	HDP vs controls
Aortic AIx, %	22.6±11.3	17.9±11.0	11.6±12.1	<.001	HDP vs controls
Aortic AIx@75, %a	26.7±8.6	24.0±8.9	18.1±11.1	<.001	HDP vs controls
Stroke volume, mL	66 (49–76)	54 (52–75)	67 (57–82)	.7
Cardiac output, L/min	4.9 (4.0–6.2)	5.2 (4.5–5.6)	5.5 (4.7–6.4)	.2
TPR, dynes/s1/cm5	1614 (1190–2080)	1442 (1266–1836)	1283 (917–1527)	.01	HDP vs controls

Abbreviations: AIx, augmentation index; AIx@75, augmentation index adjusted by heart rate 75 beats per minute; BMI, body mass index; CH, chronic hypertension; DBP, diastolic blood pressure; HDP, hypertensive disorders of pregnancy; MAP, mean arterial pressure; PP, pulse pressure; SBP, systolic blood pressure; TPR, total peripheral resistance. Data are expressed as mean±standard deviation or median (interquartile range). ^aThe number of patients analyzed was 133 because the AIx adjusted by HR 75 beats per minute was calculated only if the patient's HR was 40 to 110 beats per minute. P values using one‐way analysis of variance for overall significance.

There were no significant differences in HR or CO among the three groups. Central systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP) values were significantly higher in the HDP and CH groups compared with the control group. Central PP, AIx@75, and TPR values were significantly higher in the HDP group compared with the control group, whereas these values in the CH group were similar to those of the controls. By contrast, brachial PPs were similar among the three groups. The HDP group consisted of 72 women; 45 had early‐onset HDP and 27 had late‐onset HDP. Central PP, AIx@75, CO, and TPR were similar between these groups both before and after delivery (Table S1). Therefore, we combined the early‐onset HDP and the late‐onset HDP groups and defined them as the HDP group.

With regard to the correlations between hemodynamic parameters and perinatal outcomes, central PP and AIx@75 showed weak but statistically significant inverse correlations with gestational weeks and Apgar scores (Table S2). In contrast, CO and TPR were not correlated with gestational week and Apgar score. Before delivery, central PP alone showed weak correlations with the number of antihypertensive agents. After delivery, CO and TPR values showed weak correlations with the number of antihypertensive agents.

Hemodynamic Parameters

Figure 1 shows the central PP and AIx@75 values before and after delivery. In the HDP group, the central PP and AIx@75 values were significantly lower after delivery, whereas those changes were not observed in the other groups. The changes in other central BP values are shown in Table 3. The number of antihypertensive agents did not change between before and after delivery in any of the three groups. In the HDP group, each of the BP values—ie, brachial and central SBP, DBP, MAP, and PP—was significantly lower after delivery, in parallel with the change in AIx@75. In the controls, the AIx@75 was significantly higher after delivery, but the central PP did not change between before and after delivery. Thus, the direction of the change in the AIx@75 between before and after delivery in the HDP group was the opposite of the direction in the controls (HDP vs controls: −3.2±1.4 vs 5.5±1.8, P=.004). In the control group, none of the BP values changed after delivery, in contrast to AIx@75. In the CH group, central PP, AIx@75, and all BP values were unchanged after delivery.

Figure 1.

Central pulse pressure (PP) and augmentation index adjusted by heart rate 75 beats per minute (AIx@75) between before and after delivery. HDP indicates hypertensive disorders of pregnancy; CH, chronic hypertension. Error bars=standard errors.

Table 3.

Changes in Hemodynamic Parameters Between Before and After Delivery

	HDP (n=49)			CH (n=30)			Controls (n=14)
	Before	After	P Value	Before	After	P Value	Before	After	P Value
Heart rate, beats per min	82±15	75±11	<.001	88±12	75±9	<.001	87±9	78±8	<.001
Brachial SBP, mm Hg	141±13	127±15	<.001	137±12	137±14	1.0	125±18	124±12	.7
Brachial DBP, mm Hg	87±11	80±11	<.001	86±11	89±10	.2	76±12	80±10	.1
Brachial MAP, mm Hg	108±11	98±12	<.001	104±11	108±11	.3	94±14	97±11	.2
Brachial PP, mm Hg	54±9	47±9	<.001	52±10	48±12	.1	49±12	43±8	.1
Central SBP, mm Hg	129±14	117±16	<.001	123±13	127±14	.2	109±15	114±11	.2
Central DBP, mm Hg	89±11	82±11	<.001	88±12	90±10	.3	78±12	81±11	.1
Central MAP, mm Hg	108±11	98±12	<.001	104±11	108±11	.3	94±14	97±11	.2
Central PP, mm Hg	40±9	35±10	.02	36±9	37±11	.5	31±7	32±5	.7
Aortic AIx	24.2±10.8	25.2±10.8	.5	16.6±11.9	24.4±10.8	.001	13.5±8.7	23.6±6.7	<.001
Aortic AIx@75a	27.7±8.3	24.6±8.7	.03	23.1±9.7	24.5±9.1	.4	19.4±10.6	24.9±8.8	.01
Stroke volume, mL	62 (49–75)	64 (53–76)	.6	64 (53–79)	66 (56–72)	.7	67 (53–81)	62 (55–85)	.4
Cardiac output, L/min	5.1 (4.1–5.9)	4.7 (3.8–5.2)	.3	5.4 (4.7–6.4)	4.6 (4.0–5.5)	<.001	5.6 (4.7–6.6)	4.9 (4.3–6.0)	.3
TPR, dynes/s1/cm5	1570 (1257–2022)	1583 (1270–2029)	.3	1413 (1163–1617)	1497 (1319–1905)	.1	1242 (935–1528)	1383 (1026–1816)	.6

Abbreviations: AIx, augmentation index; AIx@75, augmentation index adjusted by heart rate 75 beats per minute; BMI, body mass index; CH, chronic hypertension; DBP, diastolic blood pressure; HDP, hypertensive disorders of pregnancy; MAP, mean arterial pressure; PP, pulse pressure; SBP, systolic blood pressure; TPR, total peripheral resistance. Data are expressed as mean±standard deviation or median (interquartile range). ^aThe number of patients analyzed was 89 because the AIx adjusted by HR 75 beats per minute was calculated only if the patient's HR was 40 to 110 beats per minute.

When the AIx was not adjusted by HR 75 beats per min, it was not significantly changed in the HDP group, but it was significantly increased in the CH and control groups after delivery (Figure S1). The extent of the change in the AIx in the HDP group was significantly less than in those of the CH and control groups (HDP vs CH: 1.1±1.7 vs 7.9±2.0, P=.03; HDP vs controls: 1.1±1.7 vs 10.1±2.0, P=.03).

Figure 2 shows the changes in TPR and CO between before and after delivery. There was a significant reduction in CO in the CH group, but not in the other groups. In all three groups, there were no significant changes in TPR after delivery. The extent of the changes was similar between each pair of the three groups.

Figure 2.

Changes in total peripheral resistance and cardiac output between before and after delivery. HDP indicates hypertensive disorders of pregnancy; CH, chronic hypertension. Error bars=standard errors.

When the HDP group was subdivided into the PE, GH, and superimposed PE groups, the central PP and AIx@75 values decreased after delivery only in the PE group (Table S3). The CO and TPR values were similar between before and after delivery in all groups.

The hemodynamic parameters after delivery are shown in Table S4. There were no significant differences in HR, brachial PP, central PP, AIx@75, CO, or TPR among the three groups.

Table S5 shows details of the antihypertensive medications before and after delivery. In both the HDP and the CH group, the prescriptions most often prescribed shifted from α‐methyldopa and hydralazine before delivery to once‐daily calcium channel blockers after delivery. The use of once‐daily angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers was also increased in these groups, but few drugs were prescribed in the control group throughout the study.

Finally, we checked the relationships between the severity of hypertension and perinatal outcomes (Table S6). The birth weights were lower in the HDP group than in the CH and control groups. The rates of SGA and severe HT (>160/110 mm Hg) were higher in the HDP group than in the CH and control groups.

Discussion

The main findings of this study were that the AIx@75 and the central PP were higher in the patients with HDP than in the controls during pregnancy, but were similar between the two groups after delivery (Table S4). Generally, arterial stiffness is highly dependent on BP levels25 and vasoconstrictions. Therefore, it is possible that functional stiffness was increased in a relatively short period without presenting structural changes. These central hemodynamic changes specific to pregnancy deserve further discussion when considering the pathophysiology of HDP.

Changes in AIx@75 Between Before and After Delivery

In the HDP patients, the AIx@75 was high during pregnancy but significantly decreased after delivery, which suggests that the central hemodynamic changes of the HDP patients between before and after delivery were clearly different from those of the controls and the patients with CH. By contrast, in the control group, the AIx@75 was lower than in the HDP group during pregnancy but similar to that in the HDP group after delivery. This is in agreement with previous studies of normal pregnant women, which revealed that the AIx@75 decreased from the first trimester to the second trimester, reached its nadir in the second trimester, and increased gradually during the postpartum period.10, 11 Before delivery, only central PP was higher in the HDP group compared with the control group, while brachial PP was similar between the two groups. In the HDP group, central PP was significantly decreased after pregnancy, but not in the other groups. Based on these findings, we speculate that a large degree of central hemodynamic change could occur between before and after delivery in patients with HDP. In other words, reflection waves from the arterial tree may have been augmented during pregnancy in patients with HDP, but ameliorated by delivery.

When the AIx was not adjusted by HR 75 beats per minute, it was low in the CH and control groups during pregnancy, but returned to the preconception state after delivery. On the other hand, in the HDP group, the AIx was similar between before and after delivery (Figure S1). The changes in wave reflection in the HDP group after delivery were concealed by the lesser reduction of HR relative to the CH and control groups.

The TPR values during pregnancy were significantly higher in patients with HDP than in controls. This is in agreement with previous study findings, which showed that TPR in patients with HDP was higher than in normal pregnant women.26, 27 The TPR is a parameter of peripheral vasoconstriction. Namely, vasoconstriction characterizes the pathophysiology of HDP whether its mechanism is primary or not. However, in the present study, the TPR did not significantly decrease between before and after delivery in any of the three groups. It could be speculated that increased TPR in the HDP group did not decrease early after delivery, but TPR in the CH and control groups did not change even during pregnancy.

Generally, the increase in the AIx@75 has a strong relationship with vasoconstriction. Indeed, only vasodilating antihypertensive agents, not nonvasodilating antihypertensive agents, improve AIx.28, 29 In the present study, the AIx@75 decreased significantly after delivery only in the HDP group, while the TPR did not significantly change. The decrease in the AIx@75 after delivery in the HDP group indicates that vasodilatation might not be a principal factor, but functional stiffness of mid to large conduit arteries could be the main mechanism. Factors associated with the pathogenesis of HDP, such as sex hormones, angiogenic factors, and endothelial dysfunction, might have key roles in this phenomenon.

Differences Between Patients With HDP and CH

Interestingly, the AIx@75 and central PP values during pregnancy were similar between the CH and control groups, although the patients with CH showed significantly higher SBP, DBP, and MAP values compared with controls. Our finding that the central hemodynamics in the CH group were similar to those of the controls is consistent with a previous study.30 We suspect that the reason for this is that there was no significant difference in the TPR between the CH and control groups, although the HDP group had higher TPR values than the CH and control groups (Table 2). This implies that if the BP is controlled, the hemodynamic load in the CH patients is low enough to be similar to that of controls, but in the HDP patients, even when BP is controlled with drugs, the hemodynamic load would not be sufficiently reduced.

Sex Hormones, Angiogenic Factors, and Central Hemodynamics

Several studies have shown that the levels of soluble fms‐like tyrosine kinase 1 (sFlt1), a circulating antiangiogenic protein released from placenta, are increased in the serum of patients with PE.31 sFlt1 prevents angiogenic factors such as placental growth factor and vascular endothelial growth factor from interaction with endothelial receptors.32 This causes endothelial dysfunction, which plays a central role in the pathogenesis of HDP.

Delivery results in major changes in the serum concentration of sex hormones such as estrogen, progesterone, and human chorionic gonadotropin (hCG). Serum hCG concentration is increased in patients with HDP.33 It has also been reported that serum hCG level has a positive correlation with severity of HDP.33 Thus, it is possible that the reductions of hCG and aFlt1 are additional factors that contribute to the reduction of AIx@75 in patients with HDP.

Study Strengths and Limitations

The strength of our study is that we were able to repeat the physiological examinations in pregnant women before and after delivery using exactly the same methods. In addition, the patients included both women with high‐risk and those with low‐risk hypertensive pregnancies.

There were also several limitations in this study. First, the sample size was small, and thus the results will need to be confirmed in larger studies. Second, the follow‐up rate was relatively low. The main reason for this is that our hospital is a tertiary hospital that receives referrals from many other clinics and hospitals. Therefore, follow‐up could not be performed in a large number of patients because of geographical constraints. Another reason was that some patients were unwilling to continue with follow‐up because of the time commitment required. That is, the AIx and impedance cardiography examinations required patients to spend time resting quietly before the measurements, and some patients found themselves too busy taking care of their children to commit to this. However, there were no differences in clinical characteristics between the follow‐up and dropout groups. Namely, the high dropout rate did not result in selection bias (Table S7). Third, the population of the patient group was not well balanced. A bias was present in that most of the patients referred to our hospital were complex or serious cases. Thus, there were more patients in the HDP group, followed by the CH and control groups. Fourth, we used patients with WCH as the control group in this study. In regard to this limitation, we note that WCH has been reported to be a benign state compared with PE and CH during pregnancy.20 We thus believe that patients with WCH were suitable as controls. Fifth, we cannot exclude the possibility that the use of antihypertensive agents or the differences of antihypertensive medication class influenced the collected data. However, the number of antihypertensive drugs per person was essentially the same among the three patient groups, with only very minor differences, and the number of antihypertensive agents was similar between before and after delivery in all three groups. In addition, in order to perform the hemodynamic measurements under the same conditions both before and after delivery, we performed the measurements at postpartum when the patients were still taking antihypertensive agents. In both the HDP and CH groups, α‐methyldopa and hydralazine were frequently used, but the use of calcium channel blockers and angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers was significantly increased after delivery (Table S5). Because calcium channel blockers could ameliorate the extent of arterial stiffness,28, 29 there is a concern that the improvement of central hemodynamics in the HDP group before and after delivery could have been caused by the change in antihypertensive medications. However, these significant hemodynamic changes were observed only in the HDP group, and not in the CH group. Therefore, we believe that the changes in central PP and AIx@75 observed in the HDP group were not caused by the changes of antihypertensive classes. There were a few patients taking β‐blockers, but the number was small enough to ignore this in the analyses. Sixth, the methods used to examine the central hemodynamic parameters CO and TPR have not been validated in pregnant women. However, these tests were repeated in the same individuals using exactly the same methods. Seventh, the AIx unadjusted by HR 75 beats per minute did not show consistent results in the HDP group; however, most of the previous studies9, 13, 16 adopted the AIx@75 in pregnant women. We therefore used the AIx@75 in the present study. Eighth, when the HDP group was subdivided into the PE, GH, and superimposed PE groups, the central PP and AIx@75 values decreased after delivery only in the PE group. However, this result was based on a very small sample size in the GH and superimposed PE groups. Finally, we did not measure pulse wave velocity.

Clinical Implications and Future Perspectives

AIx and central PP might be useful in separating patients with HDP from other hypertensive pregnancies such as those with CH or WCH. As mentioned above, because the pathophysiology of HDP involves enhanced central hemodynamic parameters, the measurement of AIx and/or central PP can play an important role in the early detection and prediction of HDP. In addition, drugs that reduce the stiffness of mid to large arteries might be effective in preventing HDP‐associated maternal complications. However, one important barrier for the development of new therapies is the exposure of not only the mother but also the fetus to the treatment.

It was recently reported that women with a history of PE have higher cardiovascular risk than those with normal pregnancies.34 However, it is not clear whether PE itself or the concomitant conditions is the risk factor. Further research is needed on this issue.

Conclusions

Central hemodynamic parameters were elevated in patients with HDP but returned to normal after delivery. The opposite findings were observed in the controls. Because the CO and TPR values did not significantly change between before and after delivery, increased body fluid or vasoconstriction might not be a main factor. Other factors that increased the stiffness of mid to large arteries, such as angiogenic factors, sex hormones, or NO sensitivity, could also have influenced the central hemodynamics in patients with HDP.

Source of Funding

None.

Disclosure

The authors report no specific funding in relation to this research and have no conflicts of interest to disclose. The authors alone are responsible for the content and writing of the paper.

Supporting information

Figure S1. Augmentation index values before and after delivery Error bars=standard errors. HDP, hypertensive disorders of pregnancy; CH, chronic hypertension.

Table S1. Characteristics of hemodynamic parameters in the early‐ and the late‐onset HDP groups.

Table S2. Correlations between hemodynamic parameters and perinatal outcomes.

Table S3. Changes in central hemodynamic parameters between before and after delivery.

Table S4. Hemodynamic parameters after delivery among the three groups of pregnant patients.

Table S5. Antihypertensive medications before and after the delivery.

Table S6. Perinatal outcomes and severity of hypertension.

Table S7. Comparisons between the follow‐up and drop‐out groups.

J Clin Hypertens (Greenwich). 2016;18:329–336. DOI: 10.1111/jch.12670 © 2015 Wiley Periodicals, Inc.