Insights into cardiac alterations after pre-eclampsia: an echocardiographic study

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. 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
  • 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
  • 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

To investigate cardiovascular (CV) performance status several years after early-onset (EO) or late-onset (LO) pre-eclampsia (PE), using echocardiography to assess myocardial strain and left ventricular (LV) torsional mechanics and ventricular–arterial coupling (VAC).

Methods

Thirty non-pregnant women with a previous singleton pregnancy complicated by EO-PE, 30 who had experienced LO-PE and 30 controls underwent echocardiography with two-dimensional (2D) speckle tracking between 6 months and 4 years after delivery and their findings were compared. All women were free from CV risk factors. VAC was defined as the ratio between aortic elastance (Ea) and LV end-systolic elastance (Ees).

Results

Women in the EO-PE group showed a persistent subclinical impairment in LV systole and a slight alteration in right ventricular function, with reductions in LV 2D strain (circumferential, radial and longitudinal) and right ventricular 2D strain and impairment of LV torsional mechanics, when compared both with women in the LO-PE group and with healthy controls. Although VAC was within the normal range in the whole study cohort, its individual components Ea and Ees were significantly altered more often in the EO-PE group than in both the LO-PE group and controls. All parameters investigated (except right ventricular 2D strain) were associated independently with gestational age at the time of diagnosis of PE.

Conclusions

Women with a history of EO-PE are more likely to have subclinical impairment of systolic biventricular function than are those with a history of LO-PE and controls. The components of VAC (Ea and Ees) show subclinical alterations which are more significant in women with a history of EO-PE than in those with a history of LO-PE and controls, although VAC itself is maintained. Our study supports the use of closer CV monitoring in previously pre-eclamptic women, particularly those in whom PE was preterm. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

INTRODUCTION

Pre-eclampsia (PE) is one of the most common causes of maternal mortality and morbidity worldwide[1, 2]. Although cardiovascular (CV) risk is increased after PE[3], a direct causative relationship has not yet been determined. Previous reports have linked preterm PE with a persistent postpartum impairment in cardiac geometry and systodiastolic function[4-10]. The use of relatively new diagnostic strategies, such as tissue Doppler imaging (TDI) and speckle-tracking echocardiography (STE), may have potential benefits in additional risk stratification by better delineation of cardiac performance status. TDI allows non-invasive measurement of myocardial strain, but is limited by angle dependency, while STE provides an objective quantification of myocardial deformation evaluated in all spatial directions independently from the angle of insonation and from cardiac translational movements. STE promises to reduce inter- and intraobserver variability in assessment of regional left ventricular (LV) function and to improve healthcare cost-effectiveness through the early identification of subclinical disease[11]. Before the introduction of STE, only tagged magnetic resonance imaging had been used to analyze the several deformation components of myocardial dynamics, including LV rotational and torsional dynamics[12]. Understanding LV performance requires not only examination of the properties of the left ventricle itself, but also investigation of the modulating effects of the arterial system on LV performance. Ventricular–arterial coupling (VAC) summarizes the CV performance of a patient, describing the interaction between the left ventricle, as a pump, and the vascular system, as a load. Although myocardial impairment[4-10], aortic stiffening[13] and endothelial dysfunction[14] have been demonstrated in previously pre-eclamptic women, the interaction between LV performance and arterial compliance, globally encompassed in the concept of VAC[15], has been investigated only rarely[16], because of its difficult assessment and interpretation. In this study, we aimed to assess cardiac function using STE and VAC in women with a history of early-onset (EO) or late-onset (LO) PE. Following our previous reports in the same population[13, 14], we tested the hypothesis that cardiac involvement will be more severe in women who have experienced EO-PE compared with those who have experienced LO-PE and healthy controls.

METHODS

This was a cross-sectional single-center case–control study, in compliance with the 1975 Declaration of Helsinki, approved by the local ethics committee and conducted according to STROBE guidelines[17]. We searched retrospectively our electronic database for all women with a diagnosis of PE at the Maternal Fetal Medicine Unit of the Department of Obstetrics and Gynecology, University of Brescia, Italy between January 2009 and December 2013. PE was defined, according to the International Society for the Study of Hypertension in Pregnancy, as blood pressure of at least 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[18]. EO-PE was defined as PE requiring delivery before 34 weeks of gestation.

All women were recalled by phone between 6 months and 4 years after delivery to assess their eligibility. We excluded women with any of the following CV risk factors: smoking habit, dyslipidemia, overweight/obesity, diabetes mellitus, or chronic hypertension, as well as those with multiple pregnancy, chromosomopathy or fetal malformation, maternal cardiopathy, nephropathy or immune disorder, or PE superimposed on chronic hypertension. In addition, we considered only women showing normal blood pressure values and absence of pathological proteinuria 6 months after delivery. Considering the sample-size calculation, cost-effectiveness and available resources, only 60 subjects (30 EO-PE and 30 LO-PE) were selected arbitrarily and requested to attend for postpartum follow-up. Thirty healthy women matched for age, body mass index and parity and without CV risk factors who delivered in our hospital during the same days as the enrolled cases were selected as controls.

Demographic and clinical data during pregnancy were collected from obstetric charts of all included subjects. A baby was considered small-for-gestational age if birth weight was < 10th percentile for gestational age (GA) on the basis of national growth charts[19]. Intrauterine growth restriction was defined as fetal abdominal circumference < 10th percentile, according to local standards[19], with umbilical artery pulsatility index (PI) > 95th percentile. All prenatal ultrasound scans had been performed by experienced sonographers, using an iU22 ultrasound system equipped with a V6-2 curved-array volume transducer (Philips Healthcare, Bothell, WA, USA). Uterine artery (UtA) Doppler measurements were obtained at the apparent crossover of the uterine and external iliac arteries. PIs of both UtAs were measured and their mean calculated.

After providing written informed consent to participate, all women underwent peripheral blood pressure measurement and echocardiography at our Cardiology Unit, in a single, temperature-controlled room. In order to limit intra- and interobserver variability, the study was carried out by a single expert echocardiographer (E.V.) who was blinded to the patients' prior medical history.

Blood pressure measurement

Blood pressure was assessed using a standard, calibrated, electronic sphygmomanometer (OMRON Healthcare, Hoofddorp, The Netherlands), with the woman in a resting state and sitting at a 45° angle. Systolic blood pressure (SBP) was considered to be high if it was > 140 mmHg, while high diastolic blood pressure (DBP) was defined as a value > 90 mmHg. DBP was measured initially in each arm, and the arm with the highest sitting DBP reading was used for a further two measurements and the mean of the three measurements was recorded. Every effort was made to have the same staff member obtain blood pressure measurements in every patient, at the same time of day, using the same equipment. Mean arterial pressure (MAP) was calculated as: SBP + (2 × DBP)/3.

Echocardiography

Echocardiographic examinations were performed using a Vivid 7 ultrasound machine (GE Healthcare, Milwaukee, WI, USA), equipped with a 3.5-MHz transducer. Digital loops were stored on the hard disk of the ultrasound machine and transferred to an EchoPac workstation (GE Healthcare) for offline analysis. Participants were placed in the left lateral decubitus position and images acquired from standard parasternal and apical views. LV dimensions, volumes and mass were obtained according to current guidelines[20] and LV ejection fraction (LVEF) was calculated using Simpson's biplane method[20]. LV mass (in g) was obtained by the equation 0.8 × (1.04 × ((LVEDD + IVST + PWT)3 – LVEDD3)) + 0.6 and relative wall thickness was calculated as 2 × PWT/LVEDD, where LVEDD is LV end-diastolic diameter, IVST is interventricular septal thickness and PWT is posterior wall thickness, at end-diastole. Similarly, right ventricular (RV) systolic function was evaluated according to published guidelines[21], calculating fractional area change (FAC), tricuspid annular plane systolic excursion (TAPSE), basal S′ wave and isovolumic acceleration at TDI. LV and RV diastolic function were defined according to published guidelines, assessing transmitral and transtricuspid Doppler inflows and TDI at basal segments[21, 22]. Myocardial performance index was calculated for both ventricles as (IVCT + IVRT)/ET, where IVCT is isovolumic contraction time, IVRT is isovolumic relaxation time and ET is ejection time at TDI. Valvular alterations were screened according to published guidelines[23, 24]. Systolic pulmonary artery pressure was obtained by adding a right atrial pressure estimate to Bernoulli's simplified equation on tricuspid regurgitation jet velocity by means of continuous-wave Doppler[21]. The stroke work index was calculated as (MAP × SV)/EDV, where SV is stroke volume and EDV is end-diastolic volume.

Speckle-tracking echocardiography (STE)

Two-dimensional (2D) strain represents myocardial deformation from a 2D point of view. Negative strain represents shortening, while positive strain indicates thickening of a given myocardial segment. STE analysis using the commercially available automated function image technique was applied to apical long-axis slices (long-axis and two-chamber and four-chamber views) for assessment of LV global longitudinal strain (GLS)[25]. The endocardial borders were traced in the end-systolic frame of the 2D images from each of the three apical views (each divided into six conventional segments). Speckles were tracked frame-by-frame throughout the LV wall until the software automatically approved the tracking for the six segments. Segments that failed to track were adjusted manually by the operator until the software approved them. GLS was calculated as the average longitudinal strain of all six segments of each of the three views (two-chamber, four-chamber and long-axis, i.e. as the mean strain of all 18 segments) (Figure 1). Moreover, using short-axis views, we calculated LV radial strain and circumferential strain, tracing endocardial borders at the level of the papillary muscles and dividing it into six conventional segments (Figure 1). Reference values of the three LV 2D strains (GLS, radial, circumferential) were according to Kocabay et al.[26]. Finally, using short-axis views at mitral valve and apical levels (six segments each) we derived peak LV torsion (peak LVtor), calculating peak apical and basal rotations and their net difference, as described previously[27]. The software calculated automatically the time domain rotation for each segment at both levels and their net differences (Figure 2). Viewing conventionally from the apex, clockwise rotation was expressed in degrees by a negative value and counterclockwise rotation by a positive one. We used reference values from Kocabay et al.[26]. From the onset of QRS, we obtained time to peak LVtor and time to peak apical and basal rotations, and normalized these intervals as a percentage of systolic duration. Aortic valve closure defined the end of systole. Twisting rate was calculated as (peak LVtor – peak LVtor in early systole)/time between the two, while untwisting rate was calculated as (peak Lvtor – peak LVtor at MVO)/time between the two, where MVO is mitral valve opening. Untwisting rate was also calculated restricted to IVRT. Finally, considering the torsion vs time profile (Figure 2), we obtained peak systolic twisting velocity and peak diastolic untwisting velocity, their times to peak from QRS and their normalization as a percentage of systolic duration.

Figure 1.

Measurement of left ventricular two-dimensional strain using speckle-tracking echocardiography: (a–c) longitudinal strain can be measured on two-chamber view (a), long-axis view (b) or four-chamber view (c) (d,e) circumferential strain (d) and radial strain (e) on parasternal short-axis view.

Figure 2.

Measurement of left ventricular torsion using speckle-tracking echocardiography: (a) basal rotation; (b) apical rotation; (c) left ventricular torsion (LVtor); (d) torsion vs time profile for left ventricle. PDUTV, peak diastolic untwisting velocity; PSTV, peak systolic twisting velocity.

For the right ventricle, 2D strain was assessed only in the four-chamber view, for calculation of RV GLS (Figure 3). In addition, we studied left atrial (LA) 2D strain, calculating its longitudinal peak at the end of LV systole (LAS) and at atrial contraction (LAA) in apical four-chamber view. Moreover, we measured the time to peak longitudinal strain from the R-wave of QRS to LAS (Figure 3). Finally, we derived atrial stiffness, by dividing LV E/E′ by LAS (using average septal and lateral E′), as described previously[28, 29].

Figure 3.

Measurement using speckle-tracking echocardiography of two-dimensional (2D) strain in right ventricle (a) and left atrium (b) in apical four-chamber view. LAA, left atrial peak longitudinal strain at atrial contraction; LAS, left atrial peak longitudinal strain at end of left ventricular systole; TPLS, time to peak longitudinal strain LAS.

Images were considered to be of good quality if at least four segments out of six did not require manual interpolation. No patient was excluded from STE analyses.

Ventricular–arterial coupling (VAC) (Figure 4)

Aortic elastance (Ea) was estimated as end-systolic pressure/SV[30]. End-systolic pressure was calculated as SBP × 0.9[15, 31]. LV outflow tract diameter, from which cross-sectional area was determined, was measured at the base of the aortic leaflets in a parasternal long-axis view. SV was measured from the proximal aorta pulse-wave Doppler flow (apical five-chamber view) and aortic cross-sectional area[31]. LV end-systolic elastance (Ees) was defined as end-systolic pressure/end-systolic volume. The modified single-beat method was used to obtain Ees according to Chen et al.[15]: Ees = (DBP – (ENd(est) × SBP × 0.9))/(SV × ENd(est)). ENd(est) was calculated as 0.0275 – 0.165 × LVEF + 0.3656 × (DBP/(SBP × 0.9)) + 0.515 × ENd(avg), with LVEF expressed in decimals and calculated as above[20]. ENd(avg) was given by a seven-term polynomial function: 0.35695 – 7.2266 × tNd + 74.249 × tNd2 – 307.39 × tNd3 + 684.54 × tNd4 – 856.92 × tNd5 + 571.95 × tNd6 – 159.1 × tNd7, in which tNd is the ratio of pre-ejection time (pre-ET) to total systolic period (pre-ET + ET), with the time at onset and termination of flow determined by aortic Doppler analysis. VAC was defined as the ratio Ea/Ees and a value of VAC > 1.3 was considered pathological. Ea and Ees were considered altered if they were > 95th percentile of values reported previously according to sex[32]. LV volume at a theoretical end-systolic pressure of 0 mmHg (V0) derives from the end-systolic pressure–volume relationship and was calculated as ESV – (SBP × 0.9)/Ees[15].

Figure 4.

Modified single-beat method to calculate ventricular–arterial coupling (VAC). (a) Left ventricular pressure–volume relationship: Ees is slope of end-systolic pressure–volume relationship; V0 is intercept of this relation at end-systolic pressure of 0 mmHg; Ea is negative slope of line joining point at end-systolic pressure and volume to point at end-diastolic volume and 0 mmHg. (b) Tissue Doppler imaging at lateral mitral annulus to calculate pre-ejection time (from onset of QRS complex to onset of S′ wave) and ejection time (from onset to end of S′ wave). DBP, diastolic blood pressure; Ea, aortic elastance; EDV, end-diastolic volume; Ees, left ventricular end-systolic elastance; ESP, end-systolic pressure; ESV, end-systolic volume; ET, ejection time; LVEF, left ventricular ejection fraction (in decimals); SBP, systolic blood pressure; SV, stroke volume; V0, left ventricular volume at a theoretical end-systolic pressure of 0 mmHg.

Statistical analysis

Continuous variables were tested visually for normality using Q-Q plots and were expressed as mean and SD, while categorical variables were expressed as frequency (n) and percentage of the sample. After Levene's test for homoscedasticity, Welch's unequal variances analysis of variance (ANOVA) was performed to analyze the difference between means for continuous variables (independent samples Welch's t-test if two groups), and Dunnett C test was used for post-hoc analysis. The χ2 test was used for assessing differences between proportions. Finally, multivariate regression analysis using the ‘enter’ method was performed to assess the association between 2D strain parameters 6 months to 4 years after delivery (as the dependent variables) and pregnancy data that differed significantly between EO-PE and LO-PE as the independent variables (GA, SBP, DBP, UtA-PI at diagnosis of PE and birth weight < 10th percentile). The post-hoc sample-size calculation showed that the size was adequate with 85% power and a 0.05 Type-I risk for all parameters. Statistical analysis was performed using IBM SPSS Statistics 20 for Windows, version 20.0 (IBM Corp., Armonk, NY, USA). All values were two-tailed; statistical significance was set at P < 0.05.

RESULTS

We identified in our electronic database 388 women with previous pregnancy complicated by PE during the period between January 2009 and December 2013; of these women, 41 who had had EO-PE and 57 who had had LO-PE were eligible for the study and 30 from each group were included (Figure 5). All women were free from any medication at the time of assessment, including oral contraceptive. Pregnancy data are reported in Table 1, and data for baseline patient characteristics at cardiological assessment are given in Table 2. Blood pressure measurements, although within normal range (< 140/90 mmHg), were significantly higher in the EO-PE group than in the LO-PE group and DBP was also higher than in controls.

Figure 5.

Flowchart of pre-eclamptic pregnancies considered for inclusion in the study. *Fifty-eight subjects had more than one cardiovascular (CV) risk factor. BP, blood pressure; EO-PE, early-onset pre-eclampsia; LO-PE, late-onset pre-eclampsia; PE, pre-eclampsia.

Table 1. Demographic and clinical characteristics of study population of women with previous singleton pregnancy complicated by early-onset (EO) or late-onset (LO) pre-eclampsia (PE) and controls, obtained retrospectively
 Cases  
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, EO-PE vs LO-PE.
  • DBP, diastolic blood pressure; GA, gestational age; HELLP, hemolysis, elevated liver enzymes, low platelets; IUGR, intrauterine growth restriction; MA, maternal age; NA, not applicable; SBP, systolic blood pressure; UtA-PI, uterine artery pulsatility index.
MA at delivery (years)36 ± 434 ± 635 ± 40.061
Parity   0.136
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 + 2NA< 0.001
Mean UtA-PI at diagnosis of PE1.56 ± 0.391.13 ± 0.43NA0.001
SBP at diagnosis of PE (mmHg)161 ± 27163 ± 13NA0.007
DBP at diagnosis of PE (mmHg)117 ± 33104 ± 18NA0.003
Proteinuria (mg/24 h)3258 ± 9263012 ± 26180.841
GA at delivery (weeks)(30 + 6) ± (3 + 6)*(37 + 1) ± (1 + 2)*(39 + 1) ± (1 + 0)0.033
Cesarean delivery30 (100.0)* 17 (56.7)*5 (16.7)< 0.001
IUGR23 (76.7)13 (43.3)0 (0)< 0.001
Male sex15 (50.0)10 (33.3)17 (56.7)0.405
Birth weight (g)928 ± 5392483 ± 561*3315 ± 485< 0.001
Birth-weight percentile14.1 ± 20.7*20.7 ± 22.3*48.0 ± 21.90.022
Maternal complications    
HELLP syndrome000 
Eclampsia000 
Placental abruption1 (3.3)00 
Disseminated intravascular coagulation1 (3.3)00 
Table 2. Clinical data of study cohort of non-pregnant women with previous singleton pregnancy complicated by early-onset (EO) or late-onset (LO) pre-eclampsia (PE) and controls with previous normal pregnancy, at cardiovascular evaluation 6 months to 4 years after delivery
 Cases  
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.
  • BMI, body mass index; BSA, body surface area; DBP, diastolic blood pressure; HR, heart rate; MAP, mean arterial pressure; SBP, systolic blood pressure.
Time from delivery (years)2.3 ± 0.72.5 ± 0.82.2 ± 0.60.115
Age at assessment (years)38 ± 436 ± 637 ± 40.084
BMI (kg/m2)23.2 ± 2.322.3 ± 2.423.1 ± 2.50.329
BSA (m2)1.68 ± 0.141.67 ± 0.121.62 ± 0.080.112
SBP (mmHg)125 ± 13116 ± 11119 ± 80.007
DBP (mmHg)80 ± 9* 73 ± 974 ± 60.003
MAP (mmHg)95 ± 10* 87 ± 989 ± 40.001
HR (bpm)78 ± 977 ± 1079 ± 70.604

Standard echocardiographic parameters are shown in Table S1, while STE data are reported in Tables 3 and S2. Focusing on LV function, the EO-PE group had a slightly reduced LVEF (56 ± 7% vs 61 ± 5% and 63 ± 4%) with higher peripheral resistance (2.1 ± 0.57 vs 1.5 ± 0.25 and 1.53 ± 0.23) than had the LO-PE group and controls. STE confirmed and extended these findings, showing an impairment of circumferential strain and GLS in the EO-PE group relative to the LO-PE group and to matched controls, and a significant reduction of radial strain with respect to controls. Data on LV torsional mechanics and VAC are summarized in Tables 3 and S2. Peak LVtor was impaired significantly in the EO-PE group, relative to both the LO-PE group and controls, as was peak apical rotation, which was altered in 10.0% of EO-PE cases. Considering VAC components, Ea and Ees were more frequently abnormal in the EO-PE group than in the LO-PE group or controls. Nevertheless, VAC showed normal values in the entire study cohort, despite being significantly higher in the EO-PE group compared with the others.

Table 3. Rate of alteration in two-dimensional (2D) left ventricular strain and torsion measured by speckle-tracking echocardiography, and in ventricular–arterial coupling (VAC) parameters in non-pregnant women with previous singleton pregnancy complicated by early-onset (EO) or late-onset (LO) pre-eclampsia (PE) and controls with previous normal pregnancy, at cardiovascular evaluation 6 months to 4 years after delivery
 Cases  
ParameterEO-PE (n = 30)LO-PE (n = 30)Controls (n = 30)Intergroup ANOVA P
  • Data are given as n (%). Post-hoc two-sample comparison of groups:
  • *P < 0.05, EO-PE vs controls;
  • P < 0.05, EO-PE vs LO-PE.
  • circumf, circumferential; Ea, aortic elastance; Ees, end-systolic left ventricular elastance; GLS, global longitudinal strain; LA, left atrial; LAA, left atrial peak longitudinal strain at atrial contraction; LAS, left atrial peak longitudinal strain at end of left ventricular systole; LV, left ventricular; LVtor, left ventricular torsion; RV, right ventricular.
2D strain    
LV radial strain7 (23.3)*2 (6.7)0 (0.0)0.008
LV circumf strain 10 (33.3)*4 (13.3)0 (0.0)0.002
LV-GLS16 (53.3)* 2 (6.7)0 (0.0)< 0.001
RV strain7 (23.3)*2 (6.7)0 (0.0)0.008
LAA strain0 (0.0)0 (0.0)0 (0.0)1.000
LAS strain3 (10.0)0 (0.0)0 (0.0)0.132
LA stiffness0 (0.0)0 (0.0)0 (0.0)1.000
LV torsion    
Peak LVtor6 (20.0)* 0 (0.0)0 (0.0)0.002
Peak apical rotation3 (10.0)* 0 (0.0)0 (0.0)0.041
Peak basal rotation 0 (0.0)0 (0.0)0 (0.0)1.000
VAC    
Ea20 (66.7)* 2 (6.7)0 (0.0)< 0.001
Ees16 (53.3)* 0 (0.0)0 (0.0)< 0.001
VAC0 (0.0)0 (0.0)0 (0.0)1.000

RV function was compromised subclinically in the EO-PE group alone, as evidenced by a lower TAPSE (22.8 ± 2.6 vs 25.8 ± 2.6 and 25.1 ± 1.6 mm) and a prolonged E-wave deceleration time (260 ± 49 vs 210 ± 34 and 182 ± 9 ms) when compared with the other two groups (Table S1). RV systolic involvement was, again, confirmed by STE; RV 2D strain was reduced significantly more in EO-PE than in controls (Table 3).

Regarding the left atrium, EO-PE cases had a greater anteroposterior diameter than did controls (27.6 ± 4.1 mm vs 24.9 ± 2.2 mm), but normal volume (24.6 ± 10.2 mL vs 25.9 ± 2.2 mL) (Table S1). LA function was preserved in most subjects, LAA strain and atrial stiffness being within normal range; LAS was altered in EO-PE cases only, but not statistically so (Table 3).

Stepwise multivariate regression analysis (Table 4) showed that LV circumferential/radial/longitudinal strains, LVtor, LAA and LAS strains, LA stiffness, Ea, Ees, VAC and V0 were associated independently with GA at diagnosis of PE, after correcting for SBP/DBP, mean UtA-PI and birth weight < 10th percentile, while RV 2D strain was not associated with GA.

Table 4. Stepwise multivariate regression analysis, run on the 60 cases with previous pre-eclamptic pregnancy, to assess linear association between two-dimensional (2D) strain and ventricular–arterial coupling (VAC) parameters after delivery (as dependent variables) and obstetric data (as independent variables)
 LV radial 2D strainLV circumferential 2D strainLV-GLSRV 2D strainLAA strainLAS strainLA stiffnessLVtorEaEesVACV0
Obstetric dataβPβPβPβPβPβPβPβPβPβPβPβP
  1. BW, birth weight; DBP, diastolic blood pressure at diagnosis; Ea, aortic elastance; Ees, end-systolic left ventricular elastance; GA, gestational age at diagnosis; GLS, global longitudinal strain; LA, left atrial; LAA, left atrial peak longitudinal strain at atrial contraction; LAS, left atrial peak longitudinal strain at the end of left ventricular systole; LV, left ventricular; LVtor, left ventricular torsion; p, percentile; RV, right ventricular; SBP, systolic blood pressure at diagnosis; UtA-PI, mean uterine artery pulsatility index at diagnosis; V0, left ventricular volume at end-systolic pressure of 0 mmHg.
GA0.7150.002−0.2680.041−0.315< 0.001NS−0.456< 0.001−0.640< 0.001−0.456< 0.001−0.640< 0.001−0.456< 0.001−0.640< 0.001−0.5990.008−0.237< 0.001
SBPNSNS−0.3270.001−2.364< 0.0010.4820.0020.8180.0050.4820.0020.8180.0050.4820.0020.8180.005NS0.890< 0.001
DBPNSNS−0.465< 0.0011.354< 0.001−1.156< 0.001−1.412< 0.001−1.156< 0.001−1.412< 0.001−1.156< 0.001−1.412< 0.0010.8480.024−1.662< 0.001
BW < 10th pNS−0.7120.001−0.605< 0.001−0.869< 0.0010.714< 0.0010.5780.0030.714< 0.0010.5780.0030.714< 0.0010.5780.003NS0.575< 0.001
UtA-PI0.9090.001−0.625< 0.0010.579< 0.0010.290< 0.001NSNSNSNSNSNSNS0.377< 0.001

DISCUSSION

The main findings of this study can be summarized as follows: (1) most EO-PE cases, together with a smaller proportion of LO-PE cases, showed subclinical impairment of systolic biventricular function, as evidenced by reductions in LV 2D strain (circumferential, radial and longitudinal) and RV 2D strain, and impairment of LV torsional mechanics, when compared with matched controls; (2) atrial function was preserved in the entire study cohort, although LAS strain was impaired in the EO-PE group, thought not statistically so; (3) VAC was maintained within normal range, albeit its components (Ea and Ees) were altered in approximately half of cases with previous EO-PE; (4) STE and VAC parameters (except RV 2D strain) were associated independently with GA at diagnosis of PE, after correcting for possible confounders (i.e. SBP/DBP, mean UtA-PI and birth weight < 10th percentile).

Although it is not still understood whether PE causes permanent CV damage or whether pre-eclamptic women have pre-existing alterations, the development of PE may lead to identification of women at high risk before other CV risk factors become apparent. Previous reports on cardiac performance in previously pre-eclamptic women reported a permanent subclinical LV dysfunction[4-10], namely, asymptomatic heart failure Stage B[33]. The early detection of functional alterations is challenging using traditional ejection phase indices (e.g. LVEF), which depend on loading condition, heart rate and LV geometry. Recently, STE has been applied to overcome these limitations. It is angle-independent, not greatly influenced by preload/afterload and not affected by heart movements. Studying longitudinal, circumferential and radial deformations, 2D strain gives a more comprehensive evaluation of LV systolic function, from both regional and global points of view, focusing on subendocardial fibers, which are the first to be damaged in CV disorders. GLS has better prognostic value for predicting major adverse CV events than does LVEF[34], and it is highly reproducible[35]; it could even provide additional information when LVEF is normal or almost normal[36, 37]. Our findings extend those of Shahul et al.[38] to the postpartum, showing persistence of myocardial dysfunction, particularly in cases affected by preterm compared with term PE. This discrepancy between EO-PE and LO-PE may be based in part on an impairment of apical rotation which occurs in EO-PE alone. Viewing the left ventricle from the apex, STE is able to assess the complex torsional heart movement focusing on apical rotation, which has been demonstrated recently as a useful marker in routine clinical practice to detect preclinical LV deterioration preceding heart failure, being age-independent and correlated with LVEF[27, 39, 40].

In previously pre-eclamptic women, LV function is further worsened by arterial stiffness[13, 14]. Aortic stiffening contributes to pathological changes in the left ventricle, increasing pulse pressure[10, 39-41]. The complex interaction between the heart, as a pump, and the arterial tree, as a load, is globally encompassed in the concept of VAC[15], which is considered to be a central determinant of net CV performance[42-44] and so an independent prognostic factor in overt cardiac diseases[45, 46]. In pathological conditions, when heart and vessels lose their favorable interaction, resulting in energy dissipation and less effective stroke work, VAC rises above 1.3[15]. The echocardiographic single-beat method was validated recently for VAC assessment[11], replacing the invasive method (i.e. heart catheterization) used previously. In the context of PE, ventricular–arterial interaction has been investigated very poorly. In agreement with Estensen et al.[47], our findings suggest that VAC remains within normal range in previously pre-eclamptic women, as expected in Stage-B heart failure. However, its components (Ea and Ees) were impaired significantly more often in the EO-PE group, confirming the arterial inability to react to myocardial contraction/relaxation in these women. Our VAC data highlight the vicious circle that arises from the alteration in both arterial elasticity and LV systolic function, probably preceding overt LV impairment.

With regard to the right ventricle, we confirmed the results of Melchiorre et al.[4], improving on them with our use of STE. Assessing 2D strain, we detected a subclinical involvement of RV function in 23% of EO-PE cases, despite their having normal values of conventional RV systolic indices (i.e. FAC, TAPSE and basal S′-wave). Mild RV alteration might be a consequence of increased LV and LA filling pressures causing higher pulmonary resistance, or a phenomenon reflecting interventricular septal impairment.

While the left atrium has been studied during PE[4, 41, 48], no postpartum data are available. Applying STE, which has been suggested recently as a method able to detect subclinical LA dysfunction, we found normal but greater LA strains in EO-PE cases than in controls, although morphology and stiffness were not altered. These findings seem to reflect a tendency towards damage of the LA, perhaps secondary to LV impairment[49]. These findings are consistent with the theory that in women who have experienced EO-PE, LV dysfunction has already affected LA performance and even RV activity.

Limitations of the present study include the small number of patients involved and the lack of data from preconceptional CV evaluation, which prevented us from demonstrating a cause–effect relationship between PE and myocardial/arterial alterations. Preconceptional studies are needed to clarify whether subclinical myocardial impairment is related directly to PE or whether it represents an expression of a maternal genetic predisposition already present before pregnancy.

Since CV risk factors (e.g. glucose intolerance, smoking habit, dyslipidemia, overweight/obesity) have a key role to play in cardiac damage, our consideration of a pre-existing maternal CV risk profile as a possible confounder and so as an exclusion criterion was a major strength of this study.

To conclude, women with a history of EO-PE are more likely to have subclinical impairment of systolic biventricular function than are those with a history of LO-PE and controls. The components of VAC (Ea and Ees) show subclinical alterations which are more significant in women with a history of EO-PE than in those with a history of LO-PE and controls, although VAC itself is maintained. Our study supports the use of closer CV monitoring in previously pre-eclamptic women, particularly those in whom PE was preterm.

Ancillary