Maternal cardiac adaptation and fetal growth





Background


Pregnancies with small-for-gestational-age fetuses are at increased risk of adverse maternal-fetal outcomes. Previous studies examining the relationship between maternal hemodynamics and fetal growth were mainly focused on high-risk pregnancies and those with fetuses with extreme birthweights, such as less than the 3rd or 10th percentile and assumed a similar growth pattern in fetuses above the 10th percentile throughout gestation.


Objective


This study aimed to evaluate the trends in maternal cardiac function, fetal growth, and oxygenation with advancing gestational age in a routine obstetrical population and all ranges of birthweight percentiles.


Study Design


This was a prospective, longitudinal study assessing maternal cardiac output and peripheral vascular resistance by bioreactance at 11 +0 to 13 +6 , 19 +0 to 24 +0 , 30 +0 to 34 +0 , and 35 +0 to 37 +0 weeks’ gestation, sonographic estimated fetal weight in the last 3 visits and the ratio of the middle cerebral artery by umbilical artery pulsatility indices or cerebroplacental ratio in the last 2 visits. Women were divided into the following 5 groups according to birthweight percentile: group 1, <10th percentile (n=261); group 2, 10 to 19.9 percentile (n=180); group 3, 20 to 29.9 percentile (n=189); group 4, 30 to 69.9 percentile (n=651); and group 5, ≥70th percentile (n=508). The multilevel linear mixed-effects model was performed to compare the repeated measures of hemodynamic variables and z scores of the estimated fetal weight and cerebroplacental ratio.


Results


In visit 2, compared with visit 1, in all groups, cardiac output increased, and peripheral vascular resistance decreased. At visit 3, groups 1, 2, and 3, compared with 4 and 5, demonstrated an abrupt decrease in cardiac output and increase in peripheral vascular resistance. From visit 2, group 1 had a constant decline in estimated fetal weight, coinciding with the steepest decline in maternal cardiac output and rise in peripheral vascular resistance. In contrast, in groups 4 and 5, the estimated fetal weight had a stable or accelerative pattern, coinciding with the greatest increase in cardiac output and lowest peripheral vascular resistance. Groups 2 and 3 showed a stable growth pattern with intermediate cardiac output and peripheral vascular resistance. Increasing birthweight was associated with higher cerebroplacental ratio. Groups 3, 4, and 5 had stable cerebroplacental ratio across visits 3 and 4, whereas groups 1 and 2 demonstrated a significant decline ( P <.001).


Conclusion


In a general obstetrical population, maternal cardiac adaptation at 32 weeks’ gestation parallels the pattern of fetal growth and oxygenation; babies with birthweight<20th percentile have progressive decline in fetal cerebroplacental ratio, decline in maternal cardiac output, and increase in peripheral vascular resistance.


Introduction


Newborn size is related to maternal anthropometric factors, such as height, prepregnancy weight status, and gestational weight gain, and the intrauterine environment immediately before delivery, which could be affected by the degree of successful placentation and maternal health throughout pregnancy. Traditionally, fetuses with birthweight (BW) of <10th percentile are classified as small for gestational age (SGA) and they are at higher risk of adverse outcomes. However, adverse outcomes are not confined to this group. Indeed, there is evidence that there is a proportionate rise in perinatal mortality, adverse neonatal outcomes, and intrapartum compromise as the BW falls <50th percentile. Even fetuses with a BW between the 50th and 75th percentile still represent an “at-risk” group. Although there is no consensus on the optimal BW percentile, the perinatal and intrapartum risks seem to plateau when the BW is >75th percentile.



AJOG at a Glance


Why was this study conducted?


This study aimed to investigate longitudinally the trends in maternal cardiac function, fetal growth, and oxygenation, with advancing gestation.


Key findings


There were similar trends in fetal growth and maternal cardiac function with advancing gestational age. At 32 weeks’ gestation, fetuses with lower estimated fetal weight continued with a growth deceleration pattern and reduced cerebroplacental ratio, whereas fetuses with greater estimated fetal weight had an accelerative growth pattern with increased cerebroplacental ratio. The deviation in fetal growth pattern occurred at the same time when women with smaller fetuses experienced a decline in cardiac output and increase in peripheral vascular resistance, whereas women with bigger fetuses continued to adapt favorably with greater cardiac output and lower peripheral vascular resistance.


What does this add to what is known?


Maternal cardiac adaptation at 32 weeks’ gestation parallels fetal growth and oxygenation.



Women who deliver SGA babies have suppressed cardiac output (CO) and raised peripheral vascular resistance (PVR). Previous studies demonstrate the existence of a relationship between fetal size and fetoplacental Doppler findings with maternal cardiovascular response to pregnancy. , However, they were mainly cross-sectional, focused on high-risk populations and the extremes of BW, assuming a similar fetal growth pattern in non-SGA groups. Cerebroplacental ratio (CPR), a marker of fetal oxygenation, is commonly used to differentiate fetal growth restriction (FGR) from SGA. In the third trimester of pregnancy, CPR has been associated with intrapartum fetal compromise, increased rate of cesarean delivery, and admission to the neonatal intensive care unit (NICU). We have previously demonstrated that in cases of SGA fetuses, maternal hemodynamics deviate most markedly in those with reduced CPR.


This study aimed to investigate longitudinally, in a routine obstetrical population stratified according to BW percentile, the trends in maternal cardiac function, fetal growth, and oxygenation, assessed by the CPR, with advancing gestational age.


Materials and Methods


Study population


Women with singleton pregnancies attending routine pregnancy care at 11 +0 to 13 +6 weeks’ gestation in 6 maternity hospitals in the United Kingdom between November 2015 and May 2016 were invited to participate in the study for longitudinal maternal hemodynamic and fetal assessment, and 1918 women (99% of those approached) agreed to participate in the study. Gestational age was confirmed from the measurement of fetal crown-rump length. After the first visit, 13 pregnancies were diagnosed with fetal anomalies, and 16 resulted in subsequent miscarriage or termination because of fetal abnormalities. After the exclusion of those with poor signals, those with missing pregnancy outcomes, and those who withdrew consent from the study, a total of 1789 women were followed up at 19 +0 to 24 +0 , 30 +0 to 34 +0 , and 35 +0 to 37 +0 weeks’ gestation ( Figure 1 ). During each of these visits, we measured maternal weight and blood pressure and assessed cardiovascular function noninvasively. In visits 2, 3, and 4, we recorded the ultrasonographic measurements of fetal head circumference (HC), abdominal circumference (AC), and femur length (FL) and calculated the estimated fetal weight (EFW). In addition, in visits 3 and 4, we conducted fetal Doppler studies for measuring the umbilical artery pulsatility index (UA-PI) and middle cerebral artery pulsatility index (MCA-PI). The study was approved by the National Health Service Research Ethics Committee (REC reference number 13/LO/1479).




Figure 1


Flowchart of study recruitment

Ling et al. Maternal cardiac function and fetal growth. Am J Obstet Gynecol 2021 .


Maternal factors and pregnancy outcomes


Maternal factors recorded included age, height, weight, and body surface area (BSA), racial origin (White, Black, South Asian, East Asian, and mixed), methods of conception (spontaneous or artificial reproductive techniques [ART]), cigarette smoking, chronic hypertension, diabetes mellitus, and parity (nulliparous, parous with and without previous preeclampsia [PE]). Pregnancy outcomes included PE, pregnancy-induced hypertension (PIH), gestational diabetes mellitus (GDM), gestational age at delivery, preterm birth, induction of labor, operative delivery for fetal distress, BW, and perinatal mortality.


Maternal cardiovascular function and fetal biometry and Dopplers


Maternal cardiac function was assessed using a noninvasive, bioreactance technology (Nirma International Conference on Management, Cheetah Medical Ltd, Maidenhead, Berkshire, United Kingdom), which we have previously validated for use in pregnancy. The bioreactance technology uses the simultaneous relative phase shifts to calculate sensitivity value (SV) when an alternating electrical current traverse the thoracic cavity. After 15 minutes of rest, 4 electrodes were applied across the maternal back and cardiac variables (CO, SV, hazard ratio [HR], PVR, and mean arterial pressure [MAP]) were recorded in a sitting position for 10 minutes at 30-second intervals (20 cycles). The averages of the final 10 cycles of hemodynamic variables were included in the analysis.


EFW and fetal biometry HC, AC, and FL z scores were calculated on the basis of the formulas derived from normal ranges in 1040 singleton pregnancies. The CPR was calculated as the ratio between the MCA-PI and UA-PI, measured as described by Vyas et al.


Definitions


BW percentile for gestational age was derived from the Fetal Medicine Foundation reference range. We classified the study population into the following 5 groups based on BW percentiles: group 1, <10th percentile; group 2, 10.0th to 19.9th percentile; group 3, 20.0th to 29.9th percentile; group 4, 30.0th to 69.9th percentile; and group 5, ≥70th centile. The definitions of PE and PIH were those of the International Society for the Study of Hypertension in Pregnancy.


Statistical analysis


We examined the longitudinal changes of maternal cardiovascular variables and z scores of fetal biometry, EFW, and CPR stratified according to the BW percentile groups as described above. The Kolmogorov-Smirnov test was used to assess the normality of the distribution of numeric data. For comparison of continuous data, the Kruskal-Wallis test or the 1-way analysis of variance (ANOVA) test was used for not-normally and normally distributed data, respectively. For categorical data, the chi-square test or the Fisher exact test was used, where appropriate. Data were presented as median (interquartile range) and mean (standard deviation) for not-normally and normally distributed continuous variables, respectively and as number (percentage) for categorical variables. The distribution of maternal weight, CO, SV, MAP, and PVR were performed using the Gaussian after Log 10 transformation.


We performed a multilevel linear mixed-effects model for the repeated measures analysis of the maternal hemodynamic variables and ultrasound assessment of fetal biometry z scores (EFW, HC, AC, FL) and CPR controlling for maternal age; maternal BSA; racial origin; smoking, previous PE or FGR; parity; maternal chronic hypertension and diabetes; pregnancy-related complications, such as PE, PIH, and GDM; BW percentile group; time (the 4 visits); and the interaction between the BW group and time. The likelihood radio test was used to define the best multilevel model comparing the base model with either the random intercept or random intercept and slope. The estimated marginal means of each hemodynamic variable, fetal biometry, and CPR at each BW percentile group and time combination are presented.


The software program International Business Machines (IBM) Statistical Package for the Social Sciences (SPSS) was used for the statistical analysis (SPSS Statistics for Windows 2015, version 25.0; IBM Corporation, Armonk, NY).


Results


In total, 1789 women were included in the final analysis of the longitudinal changes in maternal hemodynamics and fetal biometry z scores. The 5 groups included 261 women in group 1, 180 in group 2, 189 in group 3, 651 in group 4, and 508 in group 5.


Maternal demographics and pregnancy outcomes


The maternal demographic characteristics and pregnancy outcomes are presented in Table 1 . There was no significant difference in maternal age or prevalence of ART among the 5 BW groups. However, there was a significant difference in the BSA, smoking, racial origin, parity, previous PE, or FGR and chronic hypertension, gestational age of delivery, proportion of women who underwent induction of labor, operative birth for fetal distress, and perinatal mortality between the 5 BW groups. There was no difference in the prevalence of PE, PIH, GDM, rates of emergency cesarean delivery, and NICU admission.



Table 1

Demographic characteristics and pregnancy outcomes of the study groups




























































































































































































































Variables Group 1 (n=261) Group 2 (n=180) Group 3 (n=189) Group 4 (n=651) Group 5 (n=508) P value
Age (y) 30.7 (±5.7) 30.7 (±5.4) 30.8 (±5.3) 31.3 (±5.2) 31.6 (±5.2) .064
Body surface area 1.7 (1.6–1.8) 1.7 (1.6–1.8) 1.7 (1.6–1.9) 1.7 (1.6–1.9) 1.8 (1.7–1.9) .000
Smoking 24 (9.2) 10 (5.6) 12 (6.3) 29 (4.5) 21 (4.1) .033
Racial origin .000
White 174 (66.7) 106 (58.9) 129 (68.3) 500 (76.8) 413 (81.3)
Black 49 (18.8) 40 (22.2) 37 (19.6) 94 (14.4) 57 (11.2)
South Asian 26 (10.0) 21 (11.7) 13 (6.9) 24 (3.7) 14 (2.8)
East Asian 6 (2.3) 3 (1.7) 8 (4.2) 14 (2.2) 9 (1.8)
Mixed 6 (2.3) 10 (5.6) 2 (1.1) 19 (2.9) 15 (3.0)
Nulliparous 165 (62.8) 104 (57.8) 119 (63.0) 327 (50.2) 219 (43.1) .000
Multiparous .000
Previous PE or FGR 33 (12.6) 17 (9.4) 8 (4.2) 36 (5.5) 24 (4.7)
No previous PE or FGR 64 (24.5) 59 (32.8) 62 (32.8) 288 (44.2) 265 (52.2)
Artificial reproductive techniques 6 (2.3) 7 (3.9) 7 (3.7) 12 (1.8) 11 (2.2) .398
Chronic hypertension 9 (3.4) 8 (4.4) 5 (2.6) 10 (1.5) 5 (1.0) .020
Preexisting diabetes 2 (0.8) 1 (0.6) 3 (1.6) 4 (0.6) 4 (0.8) .748
Pregnancy outcomes
Gestational age at birth (wk) 39.0 (37.7 – 40.1) 39.6 (38.7 – 40.6) 39.7 (38.7 – 40.6) 40.0 (39.0 – 40.9) 40.1 (39.1 – 41.0) .000
Preeclampsia 14 (5.4) 5 (2.8) 7 (3.7) 14 (2.2) 13 (2.6) .116
Pregnancy-induced hypertension 12 (4.6) 9 (5.0) 2 (1.1) 22 (3.4) 16 (3.4) .216
Gestational diabetes 14 (5.4) 5 (2.8) 8 (4.2) 26 (4.0) 27 (5.3) .572
Induction of labor 99 (37.9) 54 (30.0) 58 (30.7) 163 (25.0) 156 (30.7) .004
Emergency cesarean 54 (20.7) 26 (14.4) 33 (17.5) 96 (14.7) 98 (19.3) .111
Operative birth (fetal distress) 46 (17.6) 18 (10.0) 23 (12.2) 71 (10.9) 51 (10.0) .025
Neonatal intensive care unit admission 21 (8.0) 11 (6.1) 8 (4.2) 30 (4.6) 30 (5.9) .285
Perinatal mortality 4 (1.8) 0 (0.0) 0 (0.0) 1 (0.2) 0 (0.0) .001

Data are presented mean (±standard deviation), median (interquartile range), or number (percentage). The 5 groups were compared using the chi-square test or Fisher exact test for categorical variables. The Kruskal-Wallis test or the 1-way ANOVA tests with post hoc analysis was used for not-normally and normally distributed data, respectively.

ANOVA, analysis of variance; FGR , fetal growth restriction; PE , preeclampsia.

Ling et al. Maternal cardiac function and fetal growth. Am J Obstet Gynecol 2021 .


Maternal hemodynamic changes in different birthweight groups


The fixed effects of the best multilevel models and the pairwise comparison of estimated marginal means with 95% confidence intervals (CIs) are shown in Table 2 , Table 3 , Table 4 , and Figure 2 for Log 10 CO, Log 10 PVR, EFW, and CPR. For maternal HR, Log 10 SV and Log 10 MAP and fetal HC, AC, and FL z score results are shown in the Supplemental Results section, Supplemental Table 1 and Supplemental Table 2 , and Supplemental Figure 1 and Supplemental Figure 2 .



Table 2

Fixed effects of multilevel linear mixed-effects models for Log 10 cardiac output and Log 10 peripheral vascular resistance, estimated fetal weight, and cerebroplacental ratio z score














































































































































































































































































Parameter Log 10 cardiac output Log 10 peripheral vascular resistance Estimated fetal weight (z score) Cerebroplacental ratio (z score)
Fixed part Estimate(SE) P value Estimate(SE) P value Estimate(SE) P value Estimate(SE) P value
Intercept 0.599 (0.022) <.0001 3.120 (0.024) <.0001 0.109 (0.231) <.001 0.190 (0.046) .010
Age (y) −0.002 (0.000) <.0001 0.002 (0.0004) <.0001
Body surface area 0.141 (0.009) <.0001 −0.076 (0.010) <.0001 0.564 (0.114) <.001
Race (reference White) <.0001 <.0001 .020
Black −0.011 (0.004) .016 0.008 (0.005) .147 0.152 (0.055) .006
South Asian −0.035 (0.007) <.0001 0.031 (0.008) <.0001 0.160 (0.089) .072
East Asian −0.047 (0.011) <.0001 0.044 (0.013) <.0001 0.163 (0.131) .215
Mixed −0.015 (0.010) .118 0.006 (0.011) .564 0.146 (0.114) .201
Smoking (reference nonsmokers)
Preexisting diabetes
Parity (reference nulliparous) .009 <.0001 <.001
Multiparous, previous PE or SGA 0.011 (0.007) .092 −0.011 (0.007) .165 −0.123 (0.081) .131
Multiparous, no previous PE/SGA 0.010 (0.003) .004 −0.017 (0.004) <.0001 −0.193 (0.043) <.001
Birthweight group (reference group 5) .019 <.0001 <.001 <.001
Group 1 (<10th percentile) −0.035 (0.008) <.0001 0.040 (0.009) <.0001 −2.337 (0.101) <.001 −0.726 (0.081) <.001
Group 2 (10.0th–19.9th percentile) −0.026 (0.009) .006 0.032 (0.010) .002 −1.482 (0.108) <.001 −0.657 (0.087) <.001
Group 3 (20.0th–29.9th percentile) −0.022 (0.009) .018 0.020 (0.010) .050 −1.309 (0.108) <.001 −0.296 (0.087) .001
Group 4 (30.0th–69.9th percentile) −0.003 (0.006) .630 0.003 (0.007) .672 −0.732 (0.075) <.001 −0.229 (0.060) <.001
Hypertensive disorders (reference number) <.0001 <.0001 .001 .008
Preeclampsia −0.021 (0.010) .037 0.048 (0.011) <.0001 0.334 (0.118) .005 −0.151 (0.120) .210
Pregnancy-induced hypertension −0.033 (0.009) <.0001 0.066 (0.010) <.0001 0.177 (0.107) .099 0.046 (0.101) .643
Chronic hypertension −0.022 (0.013) 0.089 0.064 (0.015) <.0001 0.397 (0.157) .012 −1.029 (0.372) .006
Gestational diabetes −0.025 (0.008) .002 0.032 (0.009) <.0001
Time (4 visits) <.0001 <.0001 <.001 <.001
Interaction birthweight groups with time <.0001 <.0001 <.001 .002

Estimated fetal weight and cerebroplacental ratio z scores were assessed in 3 and 2 visits, respectively. There was no significant contribution from smoking, preexisting diabetes, and chronic hypertension on Log 10 cardiac output. There was no significant contribution from smoking and preexisting diabetes on Log 10 peripheral vascular resistance. There was no significant contribution from maternal age, race, smoking, preexisting diabetes, and gestational diabetes on estimated fetal weight z score. There was no significant contribution from maternal age, body surface area, smoking, parity, and gestational diabetes on cerebroplacental ratio z score.

BW , birthweight; PE , preeclampsia; SE , standard error; SGA , small for gestational age.

Ling et al. Maternal cardiac function and fetal growth. Am J Obstet Gynecol 2021.


Table 3

Multilevel linear mixed-effects models: estimated marginal means with 95% confidence interval for maternal Log 10 cardiac output, Log 10 peripheral vascular resistance, and estimated fetal weight and cerebroplacental ratio z scores


























































































































































Variable Visit 1 Visit 2 Visit 3 Visit 4
Log 10 cardiac output
Group 1 0.716 (0.699–0.734) 0.735 (0.717–0.753) 0.725 (0.707–0.743) 0.704 (0.685–0.722)
Group 2 0.714 (0.694–0.733) 0.732 (0.712–0.752) 0.733 (0.713–0.753) 0.712 (0.693–0.732) e , g
Group 3 0.704 (0.684–0.723) 0.728 (0.708–0.748) 0.733 (0.713–0.753) 0.716 (0.696–0.736) d , g
Group 4 0.701 (0.686–0.716) a , d 0.730 (0.715–0.745) 0.749 (0.734–0.764) b 0.735 (0.720–0.751) c , d , j
Group 5 0.714 (0.698–0.730) g 0.742 (0.726–0.758) 0.751 (0.735–0.767) b 0.739 (0.722–0.755) c , e , j
Log 10 peripheral vascular resistance
Group 1 3.158 (3.138–3.177) 3.134 (3.114–3.154) 3.148 (3.128–3.168) 3.178 (3.157–3.198)
Group 2 3.163 (3.141–3.185) 3.141 (3.119–3.163) 3.139 (3.117–3.161) d , g 3.170 (3.148–3.192) e , h
Group 3 3.180 (3.158–3.201) 3.139 (3.117–3.162) 3.133 (3.110–3.155) 3.158 (3.136–3.181)
Group 4 3.179 (3.162–3.196) a 3.138 (3.121–3.155) 3.118 (3.101–3.135) b 3.141 (3.123–3.158) c
Group 5 3.170 (3.153–3.188) 3.129 (3.111–3.147) 3.117 (3.099–3.135) b 3.138 (3.119–3.156) c
Estimated fetal weight z score
Group 1 −0.701 (−0.884 to −0.519) −0.830 (−1.015 to −0.644) f −1.057 (−1.246 to −0.868) f
Group 2 −0.383 (−0.588 to −0.178) a −0.138 (−0.344 to 0.067) c , f , i −0.202 (−0.408 to 0.002) c , f , i
Group 3 −0.277 (−0.487 to −0.067) b −0.017 (−0.227 to 0.192) c , f , i −0.029 (−0.240 to 0.181) c , f , i
Group 4 −0.201 (−0.347 to −0.056) c 0.492 (0.346–0.638) c , f 0.546 (0.399–0.694) c , f
Group 5 0.205 (0.050–0.361) c 1.133 (0.975–1.291) c 1.279 (1.120–1.438) c
Cerebroplacental ratio z score
Group 1 −0.428 (−0.663 to −0.194) −0.688 (−0.925 to −0.451) f
Group 2 −0.278 (−0.523 to −0.032) e , g −0.618 (−0.863 to −0.373) f , i , k
Group 3 −0.170 (−0.422 to 0.081) b −0.258 (−0.510 to −0.006) c , e
Group 4 −0.100 (−0.320 to 0.118) c −0.191 (−0.410 to 0.028) c , f
Group 5 −0.001 (−0.225 to 0.223) c 0.038 (−0.185 to 0.262) c

Estimated fetal weight and cerebroplacental ratio z scores were assessed in 3 and 2 visits, respectively.

Ling et al. Maternal cardiac function and fetal growth. Am J Obstet Gynecol 2021.

Compared with group 1: a P< .05, b P <.01, c P <.001; compared with group 5: d P <.05, e P <.01, f P <.0001; compared with group 4: g P <.05, h P <.01, i P <.001; compared with group 3: j P <.05, k P <.01, l P <.0001.


Table 4

Change between visits: comparison of the Log 10 cardiac output and Log 10 peripheral vascular resistance, estimated fetal weight, and cerebroplacental ratio z scores ( P values)

Jun 12, 2021 | Posted by in GYNECOLOGY | Comments Off on Maternal cardiac adaptation and fetal growth

Full access? Get Clinical Tree

Get Clinical Tree app for offline access