Endothelial dysfunction and vascular stiffness in women with previous pregnancy complicated by early or late pre-eclampsia

Authors

  • R. Orabona,

    Corresponding author
    1. Maternal Fetal Medicine Unit, Department of Obstetrics and Gynecology, University of Brescia, Brescia, Italy
    • Correspondence to: Dr R. Orabona, Piazzale Spedali Civili 1, 25123 Brescia, Italy (e-mail: oraroxy@libero.it)

    Search for more papers by this author
  • E. Sciatti,

    1. Section of Cardiovascular Diseases, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
    Search for more papers by this author
  • E. Vizzardi,

    1. Section of Cardiovascular Diseases, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
    Search for more papers by this author
  • I. Bonadei,

    1. Section of Cardiovascular Diseases, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
    Search for more papers by this author
  • A. Valcamonico,

    1. Maternal Fetal Medicine Unit, Department of Obstetrics and Gynecology, University of Brescia, Brescia, Italy
    Search for more papers by this author
  • M. Metra,

    1. Section of Cardiovascular Diseases, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy
    Search for more papers by this author
  • T. Frusca

    1. Maternal Fetal Medicine Unit, Department of Obstetrics and Gynecology, University of Brescia, Brescia, Italy
    2. Department of Obstetrics and Gynecology, University of Parma, Parma, Italy
    Search for more papers by this author

ABSTRACT

Objectives

Pre-eclampsia (PE) is associated with an increased cardiovascular risk later in life. The persistence of endothelial dysfunction after delivery may represent the link between PE and cardiovascular disease. We aimed to evaluate endothelial function and arterial stiffness after delivery of pregnancy complicated by early-onset (EO) or late-onset (LO) PE and their correlation with gestational age and mean uterine artery pulsatility index at PE diagnosis and birth-weight percentile.

Methods

The study included 30 women with previous EO-PE, 30 with previous LO-PE and 30 controls with no previous PE. Participants were examined at between 6 months and 4 years after delivery. All included women were free from cardiovascular risk factors and drugs. Data on demographic and clinical characteristics during pregnancy were collected retrospectively from obstetrical charts. Endothelial function and arterial stiffness were assessed by peripheral arterial tonometry and pulse-wave analysis.

Results

All vascular parameters were significantly different, indicating circulatory impairment, in women with previous EO-PE. Women with previous LO-PE had higher vascular rigidity than did controls and all had normal values of reactive hyperemia index, although they were significantly lower when compared with those of controls. On multivariate analysis, gestational age and mean uterine artery pulsatility index at the time of PE diagnosis, and birth-weight percentile were all statistically related to the vascular indices studied, after correcting for confounding parameters.

Conclusions

Women with previous pregnancy complicated by PE, in particular those with early-onset disease, showed persistent microcirculatory dysfunction, as suggested by a significant reduction in reactive hyperemia index value, and increased arterial stiffness. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

INTRODUCTION

Evidence in the medical literature suggests that women who have experienced pre-eclampsia (PE) are at increased risk of hypertension, coronary artery disease and fatal stroke later in life[1]. This may be due to maternal systemic endothelial dysfunction resulting from an oxidatively stressed placenta[2]. Although previous investigations have demonstrated the relationship between endothelial dysfunction and PE[3, 4], the majority have been focused on brachial artery (BA) ultrasound, which is considered the gold standard for evaluating endothelial function, measuring the flow-mediated dilatation (FMD). Being highly operator-dependent, it requires specific training. Among non-invasive methods, peripheral arterial tonometry (PAT) is a more reproducible technique for detection of endothelial function, and has been validated and used previously to assess peripheral arterial tone[5-9]. In previously pre-eclamptic women, endothelial dysfunction may present as arterial stiffness[10-12], which describes the reduced capability of the artery to expand and contract according to pressure changes. It is key in the evaluation of cardiovascular (CV) risk profile because of its direct and independent association with CV risk[13].

This study aimed to evaluate endothelial function using PAT and arterial stiffness in women with a history of early-onset (EO) or late-onset (LO) pre-eclampsia (PE), from 6 months to 4 years after delivery, and determine whether vascular status was associated with gestational age and mean uterine artery pulsatility index (UtA-PI) at diagnosis of PE and birth-weight percentile.

METHODS

Subjects

This was a prospective single-center case–control study that was carried out in compliance with the Declaration of Helsinki and approved by the local ethics committee. Results are reported following the STROBE guidelines. We searched our electronic database retrospectively for 30 consecutive women with a previous diagnosis of EO-PE and 30 with a previous diagnosis of LO-PE who were treated at the Maternal Fetal Medicine Unit of the Department of Obstetrics and Gynecology, University of Brescia, Italy between January 2009 and December 2013. EO-PE and LO-PE were defined using 34 weeks' gestation as cut-off between the two groups. Thirty healthy women, matched for age, body mass index (BMI) and parity and without CV risk factors, were included as controls. All women were recalled by phone call from 6 months to 4 years after delivery to assess their eligibility. Eligible women received a single appointment at the Cardiology Unit, University of Brescia, Italy, to undergo peripheral blood pressure measurement, endothelial function evaluation and vascular stiffness assessment. All exams were performed in the same temperature-controlled room and in the morning after a night of fasting. All women gave their written informed consent. According to the International Society for the Study of Hypertension in Pregnancy (ISSHP), PE was defined as blood pressure ≥ 140/90 mmHg on two occasions 4–6 h apart, after the 20th week of gestation in previously normotensive women, accompanied by proteinuria ≥ 300 mg/24 h[14].

Small-for-gestational age (SGA) was considered as a birth weight < 10th percentile for gestational age. Intrauterine growth restriction (IUGR) was defined as fetal abdominal circumference < 10th percentile according to local standards[15] and abnormal umbilical artery PI > 95th percentile based on local standards, irrespective of the presence of absent or reversed end-diastolic flow.

None of the women we included had any of the following CV risk factors, confirmed by analyzing the patients' medical charts before and after pregnancy: smoking, dyslipidemia, overweight, diabetes mellitus or chronic hypertension. These CV risk factors were defined according to accepted Italian and European guidelines. The total study population was subjected to other exclusion criteria: multiple pregnancy, chromosomopathies or fetal malformations, maternal cardiopathies or immune disorders, PE superimposed on chronic hypertension and pregestational renal disease. The whole study cohort showed normal blood pressure values and the absence of any pathological proteinuria 6 months after delivery. Data regarding demographic and clinical characteristics during pregnancy were collected from obstetric charts of all studied women.

Blood pressure during pregnancy, 6 months after delivery and at CV evaluation was assessed using a standard, calibrated, electronic sphygmomanometer (OMRON Healthcare, Hoofddorp, The Netherlands). This device inflates the cuff automatically until over-systolic pressure is reached and then deflates gradually, recording the blood pressure by means of oscillometric technology. During blood pressure measurement, the woman was relaxed and sitting at a 45° angle, using an appropriately sized cuff placed at the level of the heart. The mean of three blood pressure measurements was recorded. The arm in which the highest sitting diastolic blood pressure was found was the arm used for all subsequent readings throughout the study. Every effort was made to have the same staff member obtain blood pressure measurements in each individual patient, at the same time of day, using the same equipment. Mean arterial pressure (MAP) was calculated as (SBP + (2 × DBP))/3, where SBP and DBP are systolic and diastolic blood pressures, respectively.

Endothelial function assessment

PAT signals were obtained using the EndoPAT-2000 device (Itamar Medical Ltd, Caesarea, Israel), which has been validated and used previously to assess peripheral arterial tone in other populations[5-9]. Specially designed finger probes were placed on the middle finger of each subject's hands. These probes comprised a system of inflatable latex air cuffs connected by pneumatic tubes to an inflating device controlled through a computer algorithm. A constant counter pressure (predetermined by the baseline DBP) was applied through air cushions. This prevented venous pooling, thus avoiding venoarteriolar reflex vasoconstriction. There was no occlusion of arterial blood flow. Pulsatile volume changes in the distal digit induced pressure alterations in the finger cuff, which were sensed by pressure transducers and transmitted to and recorded by the EndoPAT-2000 device. A decrease in the arterial blood volume in the distal fingertip caused a decrease in pulsatile arterial column changes, reflected as a decrease in the measured PAT signal, and vice versa. Endothelial function was measured via a reactive hyperemia (RH) protocol consisting of a 5-min baseline measurement, after which a blood pressure cuff on the test arm was inflated to 60 mmHg above baseline SBP or to at least 200 mmHg for 5 min. Occlusion of pulsatile arterial flow was confirmed by the reduction of the PAT tracing to zero. After 5 min, the cuff was deflated and the PAT tracing was recorded for a further 5 min. The ratio of the PAT signal after cuff release compared with baseline was calculated through a computer algorithm, automatically normalizing for baseline signal and indexed to the contralateral arm. The calculated ratio is called the RH-PAT index or RH index (RHI). Its normal value is > 2.00 (pure number), while a clear endothelial dysfunction is shown by a value ≤ 1.67. Values between 1.68 and 2.00 are still of unclear significance. At the same time, the software calculated the peripheral augmentation index (AIx), which is a measure of arterial stiffness. In particular, since this parameter is influenced by heart rate, it was standardized automatically per 75 bpm (AIx@75). According to EndoPAT software, peripheral AIx@75 is normal when < 17%. Typical examples of normal and altered EndoPAT traces are shown in Figure 1.

Figure 1.

Example of EndoPAT report, showing good endothelial function (a) and endothelial dysfunction (b). Note lower signal amplitude in post-occlusive phase, similar to that in pre-occlusive phase, in case of dysfunction (b), compared with amplitude in case of good endothelial function (a).

Pulse-wave analysis

Central blood pressure was measured by arterial tonometry, using Vascular Explorer (Enverdis GmbH, Jena, Germany). This calculated aortic stiffness parameters from oscillatory recorded pressure waves of the brachial and posterior tibial arteries and from body surface measurements (brachial–jugular, jugular–pubis symphysis, jugular–ankle). Using inflatable upper and lower arm cuffs with high fidelity sensors, pulsatile volume changes (resulting from pulsatile fluctuations of the two arteries) were transduced into pressure curves. Pulse waves were recorded when the two arteries were occluded completely at a cuff pressure of 35–40 mmHg above SBP. A dedicated computer program was used to further analyze the recorded pulse waves. Pulse transit time (PTT) was determined from the decomposition of the general aortic pressure wave using the reflection method. This measurement is based on the fact that the forward traveling pulse wave (generated by the ejection of the left ventricle) is reflected in the periphery, creating a second reflected wave. PTT was determined by the foot–foot difference in time between the forward and the beginning of the reflected pressure wave (reflection method under brachial stop/flow conditions), and aortic pulse-wave velocity (PWV) was calculated automatically from PTT and the travelling distance between the jugulum (sternal notch) and pubis symphysis, according to the manufacturer's recommendations. Moreover, brachial–ankle PWV was determined by means of simultaneous cuff measurements taken on the upper arm and ankle at DBP. Then, carotid–femoral PWV (cfPWV) was calculated from brachial–ankle and aortic measurements. According to the European reference values and techniques, this number was multiplied by 0.8, a correction factor for body surface distance measurements, and values > 9.6 m/s (0.8 × 12 m/s) were considered pathological[16].

Finally, another central hemodynamic parameter measuring arterial stiffness, the aortic AIx@75, was calculated from brachial pressure curves in combination with automated transfer algorithms. The manufacturer's guidelines suggest that aortic AIx@75 is altered when ≥ 35%.

Statistical analysis

All analyses were done using SPSS Statistics v. 20 for Windows (IBM Corp., Armonk, NY, USA). Continuous variables were tested visually for normality using Q–Q plots and represented by the mean ± SD, while categorical variables were represented as frequency (n) and percentage of the sample. For continuous variables, after Levene's test for homoscedasticity, Welch's unequal variances analysis of variance (ANOVA) was performed to analyze the difference between means (independent samples Welch's t-test if two groups), and Dunnett's C test was performed for post-hoc analysis. The chi-square test was used for assessing differences between proportions. Bivariate Spearman's non-parametric correlations were calculated to assess the association between EndoPAT/pulse-wave analysis (PWA) parameters and EO-PE/LO-PE, altered/normal mean UtA-PI, and SGA/non-SGA newborns (as dichotomic variables). Four multivariate linear regression analyses using the ‘enter’ method were run to test the independent ability of gestational age at diagnosis of PE, mean UtA-PI at diagnosis of PE and birth-weight percentile, all considered as continuous variables (independent variables), to predict the vascular parameters, after correcting each variable for the others and for SBP/DBP at diagnosis of PE, Cesarean section and IUGR rates (the only variables which differed between EO-PE and LO-PE on Welch's t-test). Sample-size calculation was adequate for all parameters with 85% power and a 5% Type I risk. For all statistical tests, P-values < 0.05 were considered significant.

RESULTS

In total, 388 cases of PE were found in our electronic database during the period considered, of which 41 EO-PE and 57 LO-PE were eligible for the study (Figure 2). Considering the sample-size calculation, cost effectiveness and available resources, 60 cases (30 EO-PE and 30 LO-PE) were chosen randomly and enrolled. Women with a previous uncomplicated pregnancy who delivered at our hospital in the same period were selected according to the exclusion criteria and included randomly as controls. All included women were free from any medication at the time of assessment, including oral contraceptives.

Figure 2.

Flowchart of women with previous pre-eclampsia (PE) considered for the study. *58 subjects had more than one cardiovascular (CV) risk factor. EO-PE, early-onset PE; LO-PE, late-onset PE.

Demographic and clinical characteristics of the study cohort, during pregnancy and at cardiovascular assessment after delivery, are shown in Tables 1 and 2. Women in the EO-PE group had worse obstetric and fetal outcomes than those in the LO-PE group or controls: higher DBP (but lower SBP), greater UtA-PI at diagnosis of PE, more Cesarean sections and higher IUGR rates. In addition, at 6 months to 4 years after delivery, women in the EO-PE group had significantly higher SBP (P = 0.007), DBP (P = 0.003) and MAP (P = 0.001).

Table 1. Clinical data of study cohort of women with previous early-onset (EO) or late-onset (LO) pre-eclampsia (PE) and normal controls at cardiovascular evaluation 6 months to 4 years after delivery
VariableEO-PE (n = 30)LO-PE (n = 30)Controls (n = 30)Intergroup ANOVA P
  • Data are given as mean ± SD. Post-hoc two-sample comparison of groups:
  • *P < 0.05, EO-PE vs controls;
  • P < 0.05, EO-PE vs LO-PE.
Time from delivery (years)2.3 ± 0.72.5 ± 0.82.2 ± 0.60.1
Maternal age at assessment (years)38 ± 436 ± 637 ± 40.08
Body mass index (kg/m2)23.2 ± 2.322.3 ± 2.423.1 ± 2.50.3
Body surface area (m2)1.68 ± 0.141.67 ± 0.121.62 ± 0.080.1
Systolic blood pressure (mmHg)125 ± 13116 ± 11119 ± 80.007
Diastolic blood pressure (mmHg)80 ± 9* 73 ± 974 ± 60.003
Mean arterial pressure (mmHg)95 ± 10* 87 ± 989 ± 40.001
Heart rate (bpm)78 ± 977 ± 1079 ± 70.6
Table 2. Demographic and clinical data of study cohort of women with previous early-onset (EO) or late-onset (LO) pre-eclampsia (PE) and normal controls, obtained retrospectively
VariableEO-PE (n = 30)LO-PE (n = 30)Controls (n = 30)Intergroup ANOVA P
  • Data are given as mean ± SD or n (%). Post-hoc two-sample comparison of groups:
  • *P < 0.05, EO-PE vs controls;
  • P < 0.05, LO-PE vs controls;
  • P < 0.05, EO-PE vs LO-PE.
  • DBP, diastolic blood pressure; DIC, disseminated intravascular coagulation; GA, gestational age; HELLP, hemolysis, elevated liver enzymes, low platelets; IUGR, intrauterine growth restriction; SBP, systolic blood pressure; UtA-PI, uterine artery pulsatility index.
Maternal age at delivery (years)36 ± 434 ± 635 ± 40.06
Parity   0.1
Nulliparous18 (60.0)24 (80.0)21 (70.0) 
Primiparous9 (30.0)3 (10.0)6 (20.0) 
Multiparous3 (10.0)3 (10.0)3 (10.0) 
GA at diagnosis of PE (weeks)27 + 5 ± 2 + 436 + 4 ± 1 + 2< 0.001
Mean UtA-PI at diagnosis of PE1.56 ± 0.391.13 ± 0.430.001
SBP at diagnosis of PE (mmHg)161 ± 27163 ± 130.007
DBP at diagnosis of PE (mmHg)117 ± 33104 ± 180.003
Proteinuria (mg/24 h)3258 ± 9263012 ± 26180.8
GA at delivery (weeks)30 + 6 ± 3 + 6*37 + 1 ± 1 + 239 + 1 ± 1 + 00.03
Cesarean section30 (100)*17 (56.7)5 (16.7)< 0.001
IUGR23 (76.7)13 (43.3)0 (0)< 0.001
Male neonate15 (50.0)10 (33.3)17 (56.7)0.4
Birth weight (g)928 ± 539*2483 ± 5613315 ± 485< 0.001
Birth-weight percentile14.1 ± 20.7*20.7 ± 22.348.0 ± 21.90.02
Maternal complications   
HELLP syndrome0 (0)0 (0)0 (0) 
Eclampsia0 (0)0 (0)0 (0) 
Placental abruption1 (3.3)0 (0)0 (0) 
DIC1 (3.3)0 (0)0 (0) 

Data on EndoPAT and PWA in the three groups are shown in Table 3. All PAT and PWA indices were significantly different in the EO-PE group compared with the LO-PE group and controls. Women in the LO-PE group showed higher vascular rigidity than the controls, but all had normal RHI values, although they were significantly lower when compared with the controls.

Table 3. EndoPAT and pulse-wave analysis parameters for measurement of endothelial function and arterial stiffness in study cohort of women with previous early-onset (EO) or late-onset (LO) pre-eclampsia (PE) and normal controls, measured 6 months to 4 years after delivery
VariableEO-PE (n = 30)LO-PE (n = 30)Controls (n = 30)Intergroup ANOVA P
  • Data are given as mean ± SD or n (%). Post-hoc two-sample comparison of groups:
  • *P < 0.05, EO-PE vs controls;
  • P < 0.05, LO-PE vs controls;
  • P < 0.05, EO-PE vs LO-PE.
  • AIx@75, augmentation index corrected for 75 bpm; cfPWV, carotid–femoral pulse-wave velocity; RHI, reactive hyperemia index.
RHI1.70 ± 0.42*2.51 ± 0.492.89 ± 0.35< 0.001
RHI ≤ 1.6711 (36.7)*0 (0)0 (0)< 0.001
RHI ≤ 2.0023 (76.7)*2 (6.7)0 (0)< 0.001
Peripheral AIx@75 (%)17 ± 19*6 ± 13−2 ± 6< 0.001
Peripheral AIx@75 ≥ 17%12 (40.0)*6 (20.0)0 (0)< 0.001
Aortic AIx@75 (%)44.0 ± 12.5*19.1 ± 10.114.3 ± 6.0< 0.001
Aortic AIx@75 ≥ 35%7 (23.3)*5 (16.7)0 (0)< 0.001
cfPWV × 0.8 (m/s)8.42 ± 1.92*6.16 ± 1.115.86 ± 0.50< 0.001
cfPWV × 0.8 > 9.6 m/s3 (10.0)*0 (0)0 (0)< 0.001

At diagnosis of PE, we found an altered UtA-PI in 37 of 49 (75.5%) patients in whom it was measured. Considering PAT and PWA data in relation to UtA-PI, the proportion of women with RHI ≤ 2.00 (P = 0.04) and cfPWV × 0.8 (P = 0.03) were the only parameters that were significantly higher in patients with an altered UtA Doppler velocimetry compared with those with normal UtA Doppler.

Thirty-two (53.3%) women with previous PE had an SGA newborn. AIx@75 was the only parameter that was significantly higher in the SGA group (P = 0.048).

The correlations between EndoPAT/PWA parameters and gestational age and mean UtA-PI at diagnosis of PE and birth-weight percentile are reported in Table 4. Gestational age at diagnosis of PE was directly correlated with all parameters except for RHI ≤ 1.67 and ≤ 2.00. An altered UtA-PI correlated positively with an RHI ≤ 2.00 (P = 0.02), peripheral AIx@75 (P = 0.02) and cfPWV × 0.8 (P = 0.03), while an SGA newborn correlated positively with an RHI ≤ 2.00 (P = 0.04) and cfPVW × 0.8 > 9.6 m/s (P = 0.03). On multivariate analysis after correcting for possible confounders, gestational age and mean UtA-PI at diagnosis of PE and birth-weight percentile were all significantly associated with PAT and PWA indices (Table 5). These findings remained statistically significant even after excluding women with a time interval since delivery of < 1 year (n = 6).

Table 4. Spearman's correlation between EndoPAT and pulse-wave analysis parameters and early-onset (EO)/late-onset (LO) pre-eclampsia (PE), altered/normal mean uterine artery pulsatility index (UtA-PI) and small-for-gestational-age (SGA)/appropriate-for-gestational-age (AGA) newborn in women with previous pre-eclampsia
 EO-PE/LO-PE (n = 60)Altered/normal UtA-PI (n = 49)SGA/AGA (n = 60)
VariableρPρPρP
  1. AIx@75, augmentation index corrected for 75 bpm; cfPWV, carotid–femoral pulse-wave velocity; RHI, reactive hyperemia index.
RHI−0.2720.009−0.1180.4−0.1590.2
RHI ≤ 1.670.0610.60.2520.080.1450.3
RHI ≤ 2.00< 0.00110.3230.020.2650.04
Peripheral AIx@75 (%)0.2430.020.3530.020.2380.08
Peripheral AIx@75 ≥ 17%0.2790.040.00010.2280.09
Aortic AIx@75 (%)0.677< 0.0010.2210.20.0440.8
Aortic AIx@75 ≥ 35%0.5040.0010.1410.40.0220.9
cfPWV × 0.8 (m/s)0.542< 0.0010.3680.03−0.1420.4
cfPWV × 0.8 > 9.6 m/s0.4930.0010.2210.20.3490.03
Table 5. Multivariate linear regression between EndoPAT and pulse-wave analysis parameters and gestational age at diagnosis of pre-eclampsia (PE), mean uterine artery pulsatility index (UtA-PI) at diagnosis of PE and birth-weight percentile (considered as independent variables) in women with previous PE, after correcting for possible confounders*
 RHIPeripheral AIx@75Aortic AIx@75cfPWV × 0.8
VariablePβPβPβPβ
  • *Possible confounders are systolic (SBP) and diastolic (DBP) blood pressure, Cesarean section rate and intrauterine growth restriction (IUGR).
  • AIx@75, augmentation index corrected for 75 bpm; cfPWV, carotid–femoral pulse-wave velocity; RHI, reactive hyperemia index.
Gestational age< 0.0010.966< 0.001−0.198< 0.001−0.996< 0.001−1.281
Mean UtA-PI0.041−0.282< 0.0010.276< 0.0010.737< 0.0010.894
Birth-weight percentile0.0360.1620.001−0.262< 0.001−0.406< 0.001−0.818
SBP0.001−1.110< 0.0010.867< 0.0010.121< 0.0010.219
DBP0.014−0.618< 0.0011.339< 0.0010.307< 0.0010.517
Cesarean section rate0.9980.0000.0050.107< 0.0010.072< 0.0010.384
IUGR0.6380.0570.422−0.030< 0.0010.812< 0.0010.656
Adjusted R20.7240.8620.5050.497

Time between delivery and the cardiovascular evaluation after delivery was not statistically significantly different among the three groups, nor between EO-PE and LO-PE. Nevertheless, we evaluated the changes in PAT and PWA indices according to this time interval (Figure 3).

Figure 3.

Changes in vascular parameters: (a) reactive hyperemia index (RHI; P = 0.09); (b) peripheral augmentation index corrected for 75 bpm (AIx@75; P = 0.9); (c) aortic AIx@75 (P = 0.9); and (d) carotid–femoral pulse-wave velocity (cfPWV × 0.8; P = 0.3), according to time of assessment after delivery.

DISCUSSION

Pregnancy leads to maternal CV adaptation characterized by high cardiac output and low systemic vascular resistance in order to respond to the increased metabolic demands[17, 18]. In cases complicated by PE, arterial compliance is reduced and vascular stiffness increased[19-21]. Playing a key role in the pathogenesis of PE[2], endothelial dysfunction may represent a link between placentation defects and CV risk later in life[22]. The acute atherosis that characterizes the spiral arteries during PE[23, 24] is similar to the early stages of atherosclerosis (Type I and II lesions, according to the American Heart Association[25]). Endothelial dysfunction seems to be related to poor placentation with smooth muscle cells retained in the vascular walls of the placental bed, vasoconstriction, hypoxic–ischemic damage and oxidative stress[2]. Using PAT, we found a significant microcirculatory impairment in the EO-PE group compared with the LO-PE group and matched controls. Women with previous LO-PE showed a normal RHI value, although it was significantly lower than that of controls. Several studies have hypothesized the persistence of endothelial dysfunction in previously pre-eclamptic women. In particular, they consistently showed an endothelium-dependent reduction in FMD[4, 26-30], which is considered the gold standard for assessment of endothelial function. Only two studies have reported using the PAT technique[31, 32]. Carty et al.[31] did not observe any difference in RHI value between women who had hypertensive and normotensive pregnancies. However, they did not differentiate between PE and gestational hypertension (27 women in total) and their mean gestational age at delivery was 39.9 weeks, which is consistent with LO-PE. Our results are consistent with those of Yinon et al.[32] who showed a reduced RHI in women with PE compared to controls, indicating that endothelial dysfunction characterizes pre-eclamptic pregnancies, though their study lacked postpartum assessment. Our study is the first to apply clearly the PAT technique at a short to medium interval after a pre-eclamptic pregnancy. Its strength is the complete absence of CV risk factors in the total study cohort. EndoPAT is a newer technique than BA ultrasound. Its most important advantage is its operator independence. In addition, it is easy to use and does not need specific training[33]. On the other hand, it is more expensive than BA ultrasound because the finger probes need to be disposed of after every use. According to several authors, EndoPAT and BA ultrasound do not fully correlate with each other[34] because they focus on different vascular beds: microcirculation vs muscular arteries. EndoPAT is associated with Framingham CV risk factors[34, 35] and it predicts future CV events[34, 36-38] better than the Reynolds Risk Score (which is Framingham risk score plus high-sensitivity C-reactive protein)[39].

Arterial stiffness plays a major role in hypertensive disorders, being both the cause and effect of high blood pressure. In the context of pregnancy-induced hypertension, there is significant evidence that arterial stiffness precedes the clinical phase of the disease both in early pregnancy and even prior to pregnancy in women at risk[19-21]. According to the findings of some[28, 40], our data suggest that aortic stiffness, as quantified by cfPWV, is significantly greater in women with previous EO-PE when compared with the LO-PE group and controls, with the exception of aortic AIx@75 ≥ 35%, which was also significantly greater in the LO-PE group. Aortic stiffness is largely predictive of future CV events[41-43]. Recently, its additional value in defining the CV profile has been consolidated[41, 42, 44, 45] and has been included in the European Society of Cardiology/European Society of Hypertension guidelines[46].

PAT and PWA indices were related to gestational age at diagnosis of PE, mean UtA-PI assessed at diagnosis of PE and birth-weight percentile suggesting that these factors predicted not only an adverse pregnancy outcome but also subsequent development of maternal CV impairment. These findings agree and extend what is already known about this topic, particularly regarding arterial stiffening[47-49].

It is already known that pregnancy transiently improves vascular compliance. This phenomenon seems to persist for a few years after delivery[50-52]. However, these findings are based on a slight reduction in MAP in a subsequent pregnancy. Albeit that our study was not designed for this issue, we evaluated directly microcirculation and arterial stiffness in a wide time period after delivery and did not find any difference between short- and medium-term assessments. We acknowledge that the small number of patients involved in our study may hamper our findings. Nevertheless, the persistence of an inflammatory response many years after PE has been demonstrated through blood samples[53]; our data seem to support this.

The present study has some limitations. We lack data on preconceptional endothelial function in order to clarify whether the alterations we found are consequences of PE or the first clinical expression of a maternal predisposition to endothelial dysfunction already present before pregnancy. The limitations of EndoPAT have been discussed previously. Furthermore, the relatively small number of patients involved prevented us from performing more complete multivariate regression analysis, although the study was powered enough for its aim. The time range of the study (6 months to 4 years after delivery) may appear too wide; we chose 6 months as an inferior cut-off because we wanted to be sure that both proteinuria and hormonal alterations typical of pregnancy had normalized. We found the same cardiovascular alterations at both 6 months and 4 years after delivery. Lastly, the method we used to evaluate arterial stiffness has not been validated fully; nevertheless, a system operating on the same mechanism, namely Vicorder, has been compared with the other most used technologies and no difference has been found among them in quantifying arterial stiffness[54].

In conclusion, in previously pre-eclamptic women, in particular those with early-onset disease, microcirculatory dysfunction seems to persist after delivery, as suggested by a significant reduction in RHI value and increased arterial stiffness. These findings are associated with a higher CV risk later in life, because of the independent relationship between RHI and arterial stiffness and multiple traditional CV risk factors[35] and their predictive ability of future events[36-39, 41-45].

Ancillary