- Top of page
Objective To assess maternal cardiac function in nulliparous women in the first trimester of pregnancy and evaluate its potential role for predicting pre-eclampsia and small for gestational age (SGA).
Design Prospective, observational, cross-sectional study.
Setting Maternity unit of a teaching hospital.
Population Nulliparous women with singleton pregnancies presenting consecutively for routine antenatal care (n= 534).
Methods Two-dimensional and M-mode echocardiography and uterine artery Dopplers were carried out at 11-14 weeks.
Main outcome measures Cardiac output (CO), stroke volume (SV), mean arterial pressure (MAP), total vascular resistance and uterine artery pulsatility index (UAPI) were compared in four outcome groups according to the development of pre-eclampsia and/or SGA.
Results Compared with the normal outcome group (n= 457), in those with pre-eclampsia but not SGA (n = 8), CO and MAP were increased; in the group with pre-eclampsia and SGA (n= 19) MAP, TRP and UAPI were increased and in the group with SGA but no pre-eclampsia (n= 50) total peripheral resistance and UAPI were increased. Independent predictors of pre-eclampsia were MAP, SV and UAPI and of SGA SV and UAPI.
Conclusions Alterations in maternal cardiac function and UAPI are observed in the first trimester of pregnancy in nulliparous women that subsequently develop pre-eclampsia and/or SGA.
- Top of page
Pre-eclampsia is not a uniformly high-resistance, volume-contracted state as previously thought.1 Earlier studies, using direct haemodynamic measurements in women with severe pre-eclampsia requiring invasive monitoring, have described variations in haemodynamic characteristics ranging from a hyperdynamic state with a higher-than-normal cardiac output (CO) and increased left ventricular function to a vasoconstrictive state with decreased CO and diminished left ventricular function. These studies also document variable systemic resistance profiles, ranging from normal- to high-resistance states.2–4 However, since the introduction and evaluation of the Doppler technique for maternal cardiac function in pregnancy,5,6 a series of echocardiographic studies have confirmed the existence of apparent haemodynamic subgroups in women with overt or preclinical pre-eclampsia.7–9 These studies have advocated a maternal hyperdynamic circulation7,9 that is preceding the clinically overt disease9 and may also be present to a variable degree during the more severe stages.7 Furthermore, in one study the effect of maternal haemodynamics on fetal growth in hypertensive pregnancies have been investigated.10 Easterling et al. observed that high-resistance hypertension was associated with lower percentile weights for gestational age, while high CO and low-resistance hypertension were associated with normal fetal growth.10 However, the cardiac investigations in their study were performed in women who were clinically hypertensive.
In previous studies on pre-eclampsia, the influence of the variable characteristics in the maternal central and peripheral haemodynamics on the sensitivity of screening tests or the efficacy of treatment has been masked by the analysis of all pre-eclampsia cases as a single disease entity. However, there is emerging epidemiological evidence to suggest that preterm pre-eclampsia (before 37 weeks) and term pre-eclampsia associated with low-birthweight infants (between 37 and 42 weeks) are likely to share similar disease origins, while term pre-eclampsia may also be associated with large or normal for gestational age fetuses and may thus represent a different subgroup of women.11 Additionally, there is evidence of disparity in screening for pre-eclampsia with uterine artery Dopplers either in the first12 or in the second trimester13 since it has been shown that there is a remarkably higher sensitivity in women with pre-eclampsia complicated by small-for-gestational-age (SGA) babies compared with uncomplicated pre-eclampsia or SGA alone. Furthermore, nondistinction of these varied haemodynamic states may, in part, be why studies utilising volume-loading in severe pre-eclampsia have shown no clear maternal nor fetal benefit.14,15 However, although maternal cardiac function might play an important role in screening and treatment for pre-eclampsia, there is no information to date regarding the degree of changes in the first trimester of pregnancy, presumably because previous investigators were concentrating at the later stages of the disease when it would be more applicable for screening and treatment. Furthermore, in most of the studies so far, hypertensive women have been examined collectively without addressing separately the issue of fetal growth and its interaction on maternal cardiovascular physiology.
The objective of this study was to prospectively investigate maternal cardiac function and peripheral haemodynamics in the first trimester of pregnancy to study the cohort according to eventual pregnancy outcome and assess its impact on screening for pre-eclampsia and SGA.
- Top of page
The pre-eclampsia rate for nulliparous women was 5%. From the total of 534 nulliparous women recruited in the study, 457 had normal outcome, 8 developed pre-eclampsia without SGA, 19 developed pre-eclampsia with SGA and 50 had SGA without pre-eclampsia.
The demographic characteristics of the four groups of women are summarised in Table 1. There were no differences between the groups in terms of maternal age, height and gestation at entry into the study. Women with pre-eclampsia and SGA compared with the other groups had a higher proportion of women of black race and an earlier gestation at delivery compared with all other groups. Birthweight was significantly different between all four groups but birthweight centile was significantly lower in the two groups (pre-eclamptic and normotensive) that delivered babies with SGA.
Table 1. Demographic characteristics of the study populations
| ||Normal (n= 457)||Pre-eclampsia (n= 8)||Pre-eclampsia and Bwt < tenth centile (n= 19)||Bwt < tenth centile (n= 50)||P-value|
|Maternal age (years)||30.2 (25.4–33.3)||31.2 (21.6–33.2)||30.6 (24.2–34.5)||29.0 (21.6–32.3)||0.34|
|Height (cm)||165.0 (160.0–168.0)||166.5 (157.0–174.0)||168.0 (160.7–168.7)||163.0 (155.0–165.0)||0.60|
|Weight (kg)||62.0 (57.0–70)||72.5 (61.7–91.7)*||71.0 (57.0–81.0)*||60.0 (53.0–72.2)||0.03|
|Ethnicity, n (%)|
|White||278 (61)||6 (75)||8 (42)||21 (42)||0.02|
|Black**||128 (28)||2 (25)||10 (53)||18 (36)|
|Other||51 (11)|| ||1 (5)||11 (22)|
|Gestation at entry (days)||88.0 (87.0–90.0)||87.5 (87.0–91.5)||89.0 (87.0–93.5)||88.0 (87.0–90.0)||0.68|
|Gestation at delivery (days)||282.0 (274.0–286.0)||275.0 (262.0–280.0)||244.0 (231.0–258.7)***||278.0 (270.0–286.0)||<0.0001|
|Birthweight (kg)****||3.4 (0.4)||3.1 (0.4)||1.7 (0.5)||2.5 (0.3)||<0.001|
|Birthweight centile||47.2 (28.8–69.8)*****||28.5 (15.0–72.4)*****||2.8 (0.8–6.7)||4.5 (1.5–7.9)||<0.0001|
Tables 2 and 3 summarise the differences in cardiac function and UAPI between the four groups.
Table 2. Comparison between the groups (normally distributed data: mean and standard deviation; non-normally distributed data: median and interquartile range)
|Variable||Normal (n= 457)||Pre-eclampsia (n= 8)||Pre-eclampsia and Bwt < tenth centile (n= 19)||Bwt < tenth centile (n= 50)||P-value|
|Cardiac output (l/min)||4.9 (4.3–5.5)||6.2 (5.4–7.1)*||4.9 (4.1–5.6)||4.6 (3.9–5.3)||<0.001|
|Cardiac index (l/min/m2)||2.9 (2.6–3.3)||3.3 (3.0–4.0)*||2.8 (2.4–3.0)||2.7 (2.4–3.3)||0.004|
|SV (ml)||67.0 (11.7)||87.9 (15.2)*||66.3 (10.9)||61.9 (11.5)**||<0.001|
|Heart rate (bpm)||75.2 (10.4)||73.6 (11.5)||74.4 (11.2)||76.5 (8.7)||0.7|
|MAP (mmHg)||77.3 (72.7–83.3)***||90.0 (82.0–91.0)||92.0 (81.3–95.3)||78.0 (68.0–86.0) ****||<0.0001|
|SBP pressure (mmHg)||104 (100–118.5)***||116 (111–127.5)||114 (110–130)||104 (100–110)||<0.0001|
|DBP pressure (mmHg)||60 (54–70)***||72 (66–77.5)||76 (66–80)||64 (60–70)||<0.0001|
|Total vascular resistance (dynes/sec/cm5)||1260 (1110–1460)||1105 (920–1280)*****||1410 (1300–1530)******||1345 (1120–1590)||<0.001|
|Mitral valve annulus shortening (mm)||15.5 (2.0)||17.8 (1.7)*||14.8 (1.7)||15.0 (1.9)||0.001|
|Transmitral E-wave velocity (mm/sec)||82.4 (13.6)||95.7 (14.6)*||79.9 (18.6)||82.6 (13.4)||0.04|
|Transmitral A-wave velocity (mm/sec)||48.9 (8.4)||51.5 (13.9)||51.8 (13.9)||51.7 (10.8)||0.1|
|UAPI||1.72 (0.5)||1.87 (0.5)||2.43 (0.6)*******||2.07 (0.6)*******||<0.001|
Table 3. Differences in maternal central and peripheral haemodynamics between the uncomplicated pregnancies and those complicated by pre-eclampsia without SGA, pre-eclampsia with SGA and SGA alone
|Variable||Pre-eclampsia no-SGA||Pre-eclampsia with SGA||SGA|
|Total vascular resistance (dynes/sec/cm5)|
|Left ventricular systolic function|
|Stroke volume (ml)|
|Cardiac output (L/min)|
|Cardiac index (L/min/m2)|
|Mitral valve annulus shortening (mm)|
|Left ventricular diastolic function|
|Transmitral E-wave velocity (mm/sec)|
|Transmitral A-wave velocity (mm/sec)|
Systolic function of the left ventricle
Women with eventual pre-eclampsia without SGA compared with all other groups had a similar heart rate but a greater SV, CO and cardiac index. This group of women also showed greater mitral annulus displacement compared with all other groups.
Women with eventual pre-eclampsia and SGA showed no differences in heart rate, SV and CO when compared with those with normal outcome. They were also similar to the normal outcome group in terms of mitral valve annulus displacement.
The group of women that subsequently delivered babies with SGA without pre-eclampsia had the lowest SV of all the groups.
Diastolic function of the left ventricle
Women with eventual pre-eclampsia without SGA had higher transmitral diastolic E-wave velocities than all other groups. However, there were no differences in the A-wave maximum velocities between the four groups.
MAP and total vascular resistance
While none of the women was clinically hypertensive when undergoing echocardiographic assessment at the end of first trimester, those who would ultimately go on to develop pre-eclampsia without SGA or pre-eclampsia with SGA already demonstrated significantly higher MAPs compared with those with eventual normal outcome. The group with eventual pre-eclampsia with SGA also showed higher MAPs when contrasted with those with SGA without pre-eclampsia. SBP and DBP demonstrated similar differences between groups as those observed for MAP.
In women with eventual pre-eclampsia without SGA total vascular resistance was maintained at levels similar to the normal outcome group, despite higher MAPs, but lower than the two groups that subsequently delivered babies with SGA. The group of women that subsequently developed pre-eclampsia with SGA had the highest total vascular resistance compared with any other group.
Uterine artery Dopplers
Women that delivered babies with SGA, with and without pre-eclampsia, had higher mean UAPI compared with the normal and the uncomplicated pre-eclamptic groups.
Multiple logistic regression
Multiple logistic regression demonstrated that significant independent contribution in the prediction of pre-eclampsia was provided by MAP, UAPI and SV (Table 4). The ROC curve for the final model is presented in Figure 1. The best point on the ROC curve for prediction of pre-eclampsia was at the probability of more than 5.4% with a sensitivity of 77.8%, specificity of 79.1% and a screen-positive rate of 20%.
Table 4. Logistic regression models for prediction of pre-eclampsia and birthweight below the tenth centile
|Variable||Coefficient||Odds ratio (95% CI)||P-value|
|Stroke volume||0.03||1.03 (1.0–1.07)||0.04|
|Birthweight below tenth centile|
|Stroke volume||−0.04||0.95 (0.92–0.98)||0.001|
Similarly, multiple logistic regression demonstrated that significant independent contribution in the prediction of SGA was provided by UAPI and SV (Table 4). The ROC curve for the final model is presented on Figure 1. The best point on the ROC curve for prediction of SGA was at the probability of more than 11.8% with a sensitivity of 66%, specificity of 77.5% and a screen-positive rate of 22% (Figure 1).