Skip main navigation

Maternal Cardiovascular Dysregulation During Early Pregnancy After In Vitro Fertilization Cycles in the Absence of a Corpus Luteum

Originally publishedhttps://doi.org/10.1161/HYPERTENSIONAHA.119.13015Hypertension. 2019;74:705–715

Abstract

Commonly used in vitro fertilization protocols produce pregnancies without a corpus luteum (CL), a major source of reproductive hormones. In vitro fertilization pregnancies without a CL showed deficient gestational increases of central (aortic) arterial compliance during the first trimester and were at increased risk for developing preeclampsia. Here, we investigated whether there was generalized impairment of cardiovascular adaptation in in vitro fertilization pregnancies without a CL compared with pregnancies conceived spontaneously or through ovarian stimulation, which lead to 1 and >1 CL, respectively (n=19–26 participants per cohort). Prototypical maternal cardiovascular adaptations of gestation were serially evaluated noninvasively, initially during the follicular phase before conception, 6× in pregnancy, and then, on average, 1.6 years post-partum. The expected increases of cardiac output, left atrial dimension, peak left ventricular filling velocity in early diastole (E wave velocity), peripheral/central arterial pulse pressure ratio, and global AC, as well as decrease in augmentation index were significantly attenuated or absent during the first trimester in women who conceived without a CL, when compared with the 1 and >1 CL cohorts, which were comparable. Thereafter, these cardiovascular measures showed recovery in the 0 CL group except for E wave velocity, which remained depressed. These results provided strong support for a critical role of CL factor(s) in the transformation of the maternal cardiovascular system in early gestation. Regimens that lead to the development of a CL or replacement of missing CL factor(s) may be indicated to improve cardiovascular function and reduce preeclampsia risk in in vitro fertilization pregnancies.

Introduction

See Editorial, pp 507–508

Accumulating evidence indicates elevated risk for hypertensive disorders of pregnancy after in vitro fertilization (IVF) especially in donor egg recipient and frozen-thawed embryo transfer (FET) cycles. However, the mechanisms responsible for this enhanced risk have been elusive.1 A major contributing factor may be the potential impact of IVF protocols on the maternal hormonal milieu especially in the first trimester when the corpus luteum (CL) is a major source of reproductive hormones before the placenta becomes sufficiently developed to supersede.2 Unassisted (spontaneous), singleton pregnancies develop in the presence of 1 CL. In contrast, with IVF, there can be either formation of multiple CLs after ovarian stimulation in fresh IVF cycles or complete absence of CL development after hypothalamic-pituitary suppression in donor egg and FET cycles. Although estradiol and progesterone are replaced in the first trimester for luteal support when the CL is absent, other factors produced by the CL such as relaxin are not provided. Because relaxin was shown to be critical for maternal circulatory changes at least during early to midterm pregnancy in the conscious gravid rat model3–5 and may contribute to the first trimester increase of GFR in women,6 we reasoned that its deficiency, or perhaps of other unidentified CL factor(s), may compromise circulatory adaptations during early pregnancy in recipients of donor eggs or FET. IVF affords the opportunity to investigate the potential role of the CL in the physiology and outcomes of pregnancy in women.2,7 Of note, perturbed circulatory adaptations during early gestation were previously associated with pathological pregnancy outcomes including preeclampsia.2,8–10

Women conceiving with autologous FET in the absence of a CL experienced increased rates of preeclampsia and preeclampsia with severe features compared with women with one or multiple CL7. In parallel, women conceiving with autologous FET or donor eggs without a CL demonstrated attenuation of the physiological gestational increase in central (aortic) arterial compliance (CAC) during the first trimester, normally a prominent maternal adaptation in early pregnancy facilitating ventricular-arterial coupling.7,11 In contrast, those conceiving by IVF with multiple CL showed gestational changes in CAC and preeclampsia risk that were not different from women conceiving spontaneously without IVF (1 CL).7 Given the increased risk of preeclampsia and reduced central (aortic) compliance in IVF pregnancies without a CL, we hypothesized that absence of a CL is also accompanied by impairment of other maternal circulatory adaptations, again primarily in early gestation. Accordingly, our first aim was to investigate whether abnormal CL number adversely impacted the gestational rise in cardiac output (CO) during the first trimester, another prototypical circulatory adaptation in pregnancy.12 Our second aim was to explore whether gestational changes in wave reflections and global arterial compliance (gAC) were also perturbed by abnormal CL number in the first trimester. Typically, arterial compliance rises during normal pregnancy as reflected by increases in gAC and peripheral/central pulse pressure ratio and decreases in augmentation index and pulse wave velocity.7,13 Of note, association between perturbed arterial compliance and preeclampsia was previously reported.14,15 Identification of additional cardiovascular deficiencies in women conceiving by IVF would further support a potential causal link between aberrant CL number, perturbed maternal circulatory adaptations in early gestation, and adverse pregnancy outcome.2,7

Methods

The authors declare that all supporting data are available within the article and its online supplementary files (please see http://hyper.ahajournals.org).

Participants

After written informed consent, participants were recruited to this study, which was approved by the University of Florida Institutional Review Board. Women planning to conceive with IVF were identified by Reproductive Endocrinology and Infertility physicians at the University of Florida. Women planning to become pregnant without assisted reproduction were recruited through advertisement. The 3 cohorts included in this investigation were women conceiving (1) naturally without IVF (singleton pregnancies/single CL); (2) by transfer of a frozen-thawed embryo(s) produced using donor or autologous oocytes or of fresh donor oocyte-generated embryos in a programmed, suppressed cycle (absent CL); or (3) after ovarian stimulation, IVF, and fresh embryo transfer (multiple CL).

Baseline measurements before pregnancy were made in the absence of circulating CL products (follicular phase or last 4 days of leuprolide suppression). The participants were then studied 6× during pregnancy and once post-partum. After an overnight fast and abstinence from caffeinated products, alcohol, and pain medications during the preceding 24 hours, participants reported to the Clinical Research Center at ≈8:00 am where they underwent a number of physiological assessments conducted over a period of ≈5 hours.

Standard IVF protocols were used for the fresh IVF cycles. The same programmed cycle protocols were used for donor oocyte recipient fresh embryo transfer, donor oocyte recipient FET, and autologous oocyte FET. The number of CL after ovarian stimulation was estimated by the number of retrieved eggs. In the programmed cycles, absent CL was confirmed in each woman by undetectable concentrations of circulating relaxin at all time points during pregnancy (data not shown). For IVF protocol details, see Methods in online-only Data Supplement.

CO, Structure and Function

The analytical cohort consisted of all participants with at least 4 study visits including a prepregnancy baseline. After the prepregnancy baseline measurement, they were then studied during pregnancy: 5.8±0.1, 8.3±0.1, 11.6±0.1, 15.2±0.1, 24.0±0.1, and 33.6±0.1 weeks of gestation and finally 84.7±1.8 weeks after delivery (mean±SE). As we previously reported, echocardiograms were obtained in the left lateral decubitus position with an iE33 (Philips, The Netherlands) equipped with a broadband S5-1 transducer (frequency transmitted 1.7 MHz, received 3.4 MHz; see Methods in Data Supplement).12

Pulse Wave Analysis

The analytical cohort consisted of all participants with at least 2 study visits including a prepregnancy baseline. After the prepregnancy baseline measurement, they were then studied 6× during pregnancy and once post-partum (see Cardiac Output, Structure and Function above). Wave reflection characteristics and event timing were assessed noninvasively from central (aortic) pressure waveforms, the latter synthesized from radial artery waveforms using a generalized transfer function (SphygmoCor CvMS; AtCor),16,17 as we previously reported (see Figure S1 and Methods in Data Supplement).13

Global Arterial Compliance

gAC was assessed as we previously published.13 See Figure S2 and Methods (in Data Supplement).

Statistics

Characteristics of the study population were described using frequencies and proportions for categorical variables and mean±SEM for continuous variables. χ2 or Fisher exact test was performed to test for differences in proportions of categorical variables between groups. Nonparametric ANOVA (Kruskal-Wallis) test was used to compare continuous variables between 2 or more groups. Wilcoxon signed-rank test with Bonferroni correction was performed for step-down tests for changes in each CL group. Linear mixed models with cubic splines for gestation weeks was fitted to evaluate trends and compare changes over time. Linear mixed models account for the correlation between repeated measurements taken from the same women at several points, include random effects to model the correlation and to allow for missing observations to occur during the study. The area under the curve estimates were computed using the trapezoid method. Significance level was set at the 0.05 level. Also, see Methods (in Data Supplement).

Results

CO, Structure, and Function

Participant Characteristics, Infertility Diagnoses, and Obstetric and Neonatal Outcomes

Participant race, parity, maternal smoking, and history of hypertensive disease of pregnancy were not different among the 3 cohorts. On average, maternal age and BMI were 4 years and 3 to 4 kg/m2 greater, respectively, in the 0 CL compared with the single and multiple CL cohorts (both P<0.05; Table). Infertility diagnoses were comparable between women conceiving with IVF involving multiple or 0 CL with the major exception of diminished ovarian reserve and male factor infertility, which were more frequent in the 0 CL and multiple CL cohorts, respectively (Table S1). Although the study was not powered for detecting differences in adverse obstetric outcomes, numerically more occurred in women conceiving by IVF (Table S2). The cesarean section rate (Table 1) and twin pregnancies were significantly greater, and gestational age at delivery and newborn weight were significantly less in pregnancies conceived by IVF compared with those conceived spontaneously (Table S3).

Table. Participant Demographic and Clinical Characteristics

Characteristic0 CL1 CLMultiple CLP Value
N242219
Maternal age, y, at 5–6 wks36±1 (29–46)32±1 (23–45)32±1 (26–41)0.03
BMI, kg/m2, prepregnancy27±1 (18–36)24±1 (18–43)23±1 (18–38)0.021
Body weight, kg, prepregnancy74±3 (47–101)66±3 (44–115)65±3 (48–100)0.156
Height, cm165±2 (154–183)165±1 (152–172)167±2 (157–182)0.836
Participant race
White2020160.54
American Indian or Alaska Native000
Asian000
Black211
Other212
Nulliparity, No. (%)11(46)11 (50)12 (63)0.51
Maternal smoking, No. (%)2 (10)1 (4)00.77
Hypertensive disease in prior pregnancy, No.000
Cesarean section, No. (%)10(42)2 (9)3(16)0.029

Mean±SEM (range). BMI indicates body mass index; and CL, corpus luteum.

Cardiac Output

CO was determined by multiplying the LV outflow tract velocity time integral of each beat by the estimated average LV outflow cross-sectional area determined by averaging the measured radii of all beats (aVaD_CO; Table S4; Figure 1). Because we hypothesized that the potential impact of the CL would be the greatest in the first trimester, we analyzed CO across the 3 cohorts for the first 4 visits. In the full mixed model, there was a significant effect of time (P<0.001) and group (P=0.050), and a borderline significant group by time interaction (P=0.092). After grouping, the 1 and >1 CL cohorts together, and then comparing this combination with the 0 CL cohort, the group by time interaction remained of borderline significance (P=0.065). Wilcoxon signed-rank test with Bonferroni correction was used to compare the 3 time points in the first trimester with the prepregnancy baseline for each of the 3 cohorts. CO was increased at all 3 time points in the first trimester for the 1 and >1 CL cohorts (P<0.017) but only at the 10 to 12 gestational week time point for 0 CL group (Figure 1A). Moreover, the increase in CO from the prepregnant baseline among the cohorts was significantly different during gestational weeks 5 to 6: 0 CL 186±148 mL/minute compared with cohorts 1 and >1 CL combined 669±129 mL/minute (P=0.024); and gestational weeks 7 to 9: 0 CL 124±154 mL/minute compared with cohorts 1 and >1 CL combined 740±133 mL/minute (P=0.018; Figure 1B). Finally, the area under the curve of the overall increase in CO during the first trimester was also significantly greater for the 1 and >1 CL cohorts combined compared with the 0 CL cohort (P=0.039; Figure 1B). In summary, the gestational rise in CO during the first trimester was attenuated in the 0 CL cohort.

Figure 1.

Figure 1. Cardiac output in pregnancies conceived with or without IVF. A, Absolute values of cardiac output (CO) before, during, and after pregnancy, and (B) changes in cardiac output from prepregnancy baseline in women who conceived with or without assisted reproduction. aVaD_CO, CO was determined by multiplying the LV outflow tract velocity time integral of each beat by (aV) the estimated average LV outflow cross-sectional area determined by averaging the measured radii of all beats (aD). Wilcoxon signed-rank test with Bonferroni correction was used to compare the 3 time points in the first trimester with the prepregnancy baseline for each cohort. BP indicates before pregnancy; CL, corpus luteum; and PP, postpartum. +P<0.017 vs BP. *P<0.05 for 0 CL compared with 1 and >1 CL cohorts combined during 5 to 6 and 7 to 9 gestational wks by nonparametric ANOVA.

The first trimester rise in CO determined by multiplying the single best LV outflow tract velocity time integral by the single best LV outflow cross-sectional area (bVbD_CO) was also significantly attenuated in the 0 CL group, as was the gestational increase of cardiac index calculated using aVaD_CO (Table S5; Figures S3 and S4; and Results in the Data Supplement).

For analysis of mean arterial pressure and systemic vascular resistance (SVR), see Results in the Data Supplement. The potential contributions of heart rate (HR), stroke volume, and SVR to the attenuated gestational rise of CO in the women who conceived by IVF in the absence of a CL are also presented in Results and Discussion in the Data Supplement.

Left Atrial Dimension

Left atrial dimension (LAD) derived from m-mode echocardiography is depicted in Table S6 and Figure 2. We analyzed LAD across the 3 cohorts for the first 4 visits. In the full mixed model, there was a significant effect of group (P=0.014) but not of time or group by time. LAD was significantly increased at 7 to 9 (P=0.014) and 10 to 12 (P=0.004) gestational weeks compared with prepregnancy baseline in the 1 CL group but at none of the time points in the first trimester for the 0 or >1 CL groups (Figure 2A). The area under the curves for the overall first trimester change in LAD was also significantly different among the 3 cohorts (P=0.012; Figure 2B). In particular, the increase in LAD from the prepregnancy baseline between the 0 and 1 CL cohorts was significantly different especially during gestational weeks 7 to 9: 0 CL −0.10±0.06 and 1 CL +0.14±0.06 cm (P=0.004); and gestational weeks 10 to 12: 0 CL −0.02±0.05 and 1 CL +0.20±0.06 cm (P=0.012; Figure 2B). In summary, the gestational rise in LAD during the first trimester was greatest in the 1 CL cohort, and the 0 CL group showed little change from prepregnancy baseline (which, inexplicably, was also 0.3 cm greater than the other 2 cohorts; P=0.034). The increase of LAD in the >1 CL group was intermediate between the 0 and 1 CL cohorts.

Figure 2.

Figure 2. Left atrial dimension in pregnancies conceived with or without IVF. A, Absolute values of left atrial dimension before, during, and after pregnancy and (B) changes in left atrial dimension from prepregnancy baseline in women who conceived with or without assisted reproduction. BP indicates before pregnancy; CL, corpus luteum; and PP, postpartum. +P<0.017 vs BP by Wilcoxon signed-rank test with Bonferroni correction. *P<0.05 for 0 CL compared with 1 CL cohort during 7 to 9 and 10 to 12 gestational wks by nonparametric ANOVA.

E Wave Velocity

In general, peak LV filling velocity in early diastole (E wave velocity) increased significantly during pregnancy in the 1 and >1 CL, but not in the 0 CL cohort (Table S6; Figure 3). We analyzed E wave velocity across the 3 cohorts for the first trimester and for all 7 visits (excluding the postpartum visit). In the full mixed model, there were significant group by time interactions for both analyses (P=0.046 and 0.032, respectively; Figure 3A). After combining E wave velocity for the 1 and >1 CL cohorts, which were not different from each other, and comparing this combined cohort to the 0 CL group using the full mixed model, the group by time interaction became highly significant during the first trimester (P=0.009). E wave velocity was significantly increased during gestational weeks 5 to 6 and 7 to 9 relative to prepregnancy baseline in the 1 CL cohort (P<0.017), and during 7 to 9 gestational weeks in the >1 CL group, but this increase was of borderline significance (P=0.050). Notably, E wave velocity was not significantly different from prepregnancy baseline at any time point during the first trimester in the 0 CL cohort (Figure 3A). Changes in E wave velocity from the prepregnancy baseline among the 3 cohorts were significantly different at all first trimester visits, that is, gestational weeks 5 to 6: 0 CL −6.2±3.3, 1 CL +8.3±2.4 and >1 CL +3.5±3.1 cm/s (P=0.007); gestational weeks 7 to 9: 0 CL −3.9±2.8, 1 CL +8.9±3.0, and >1 CL +9.8±3.8 cm/s (P=0.012); and gestational weeks 10 to 12: 0 CL −7.2±3.4, 1 CL +4.2±2.6, and >1 CL +2.9±4.1 cm/s (P=0.044; Figure 3B). Finally, the area under the curve for the overall change in E wave velocity during pregnancy was also significantly greater for the 1 and >1 CL cohorts, which were similar, compared with the 0 CL group (P<0.001 first trimester; P=0.009 all of pregnancy; Figure 3B).

Figure 3.

Figure 3. Mitral E wave velocity in pregnancies conceived with or without IVF. A, Absolute values of peak left ventricular filling velocity in early diastole (mitral E wave velocity) before, during, and after pregnancy and (B) changes in Mitral E wave velocity from prepregnancy baseline in women who conceived with or without assisted reproduction. BP indicates before pregnancy; CL, corpus luteum; and PP, postpartum. +P<0.017 vs BP for 1 CL; +P=0.05 vs BP for >1 CL (borderline significance) by Wilcoxon signed-rank test with Bonferroni correction. *P<0.05 among the 3 cohorts at all 3 visits in the first trimester by nonparametric ANOVA.

For A wave velocity, E/A ratio, and left ventricular mass and wall dimensions, see Table S6 and Results in the Data Supplement.

Pulse Wave Analysis

Participant Characteristics, Infertility Diagnoses, and Obstetric and Neonatal Outcomes

These variables were similar to those listed above for CO, Structure and Function. However, because the participants were selected on the basis of completing at least 2 rather than 4 visits, they were not identical as we previously published.7

Peripheral and aortic pressures, aortic augmentation pressures, durations, and cardiac parameters for the 3 cohorts before, during, and after pregnancy are presented in Tables S7 through S11.

Pulse Pressure Amplification Ratio (PPAmp)

PPAmp is the peripheral/central pulse pressure ratio. PPAmp rose in the first trimester but significantly less so in the 0 CL compared with the 1 and >1 CL cohorts (group by time interaction P=0.015; Table S8; Figure 4A). By applying the full mixed model to analyzing 2 groups—0 CL versus cohorts 1 and >1 CL combined—the group by time interaction became highly significant (P<0.001). Wilcoxon signed-rank test with Bonferroni correction was used to compare the 3 time points in the first trimester with the prepregnancy baseline for each of the 3 cohorts. PPAmp was significantly increased at all first trimester time points in each of the 3 cohorts (P<0.017; Figure 4A). However, the increase in PPAmp from the prepregnant baseline among the cohorts was significantly different during gestational weeks 5 to 6: 0 CL 0.06±0.02, 1 CL 0.15±0.02, and >1 CL 0.18±0.03 (P=0.006); and gestational weeks 7 to 9: 0 CL 0.06±0.02, 1 CL 0.13±0.02, and >1 CL 0.17±0.03 (P=0.034; Figure 4B). Finally, the area under the curve of the overall increase in PPAmp during the first trimester was significantly different among the 3 cohorts (P=0.024); and for combined cohorts 1 and >1 CL versus 0 CL (P=0.007; Figure 4B). In summary, the gestational rise in PPAmp during the first trimester was attenuated in the 0 CL cohort.

Figure 4.

Figure 4. Peripheral/central arterial pulse pressure ratio in pregnancies conceived with or without IVF. A, Absolute values of peripheral/central arterial pulse pressure ratio (PPAmpRatio) before, during, and after pregnancy and (B) changes in PPAmpRatio from prepregnancy baseline in women who conceived with or without assisted reproduction. BP indicates before pregnancy; CL, corpus luteum; and PP, postpartum. +P<0.017 vs BP by Wilcoxon signed-rank test with Bonferroni correction. *P<0.05 among the 3 cohorts during 5 to 6 and 7 to 9 gestational wks by nonparametric ANOVA.

Augmentation Index %

Consistent with the findings for PPAmp, the gestational decline in Augmentation Index % (AIx%) during the first trimester was significantly subdued in the 0 CL cohort when compared with the combined cohorts 1 and >1 CL, which were comparable (group by time interaction P=0.003; Table S9; Figure 5A). Alx% was significantly decreased at all first trimester time points compared with prepregnancy baseline in each of the 3 cohorts (P<0.017; Figure 5A). However, the decline in AIx% from the prepregnant baseline among the 3 cohorts was significantly different during gestational weeks 5 to 6: 0 CL –5.7±1.5, 1 CL −11.3±1.2, and >1 CL −12.2±2.1 (P=0.006); gestational weeks 7 to 9: 0 CL −5.8±1.5, 1 CL −10.4±1.3, and >1 CL −11.6±2.0 (P=0.028); and borderline significantly different during gestational weeks 10 to 12: 0 CL compared with cohorts 1CL and >1 CL combined (P=0.063; Figure 5B). In addition, the area above the curves for the overall decline in Alx% during the first trimester was significantly different among the 3 cohorts (P=0.017) and for 0 CL compared with cohorts 1 and >1 CL combined (P=0.005; Figure 5B). For analysis of AIx% in relation to HR, see Tables S9 and S10, Results and Discussion in the Data Supplement.

Figure 5.

Figure 5. Augmentation index in pregnancies conceived with or without IVF. A, Absolute values of augmentation index before, during, and after pregnancy and (B) changes in augmentation index from prepregnancy baseline in women who conceived with or without assisted reproduction. BP indicates before pregnancy; CL, corpus luteum; and PP, postpartum. +P<0.017 vs BP by Wilcoxon signed-rank test with Bonferroni correction. *P<0.05 among the 3 cohorts at 5 to 6 and 7 to 9 gestational wks, *P<0.063 for 0 CL compared with 1 and >1 CL cohorts during 10 to 12 gestational wks by nonparametric ANOVA.

Global Arterial Compliance

gAC estimated using SVR derived from aVaD_CO in the equation is shown in Table S12 and Figure 6. Again, because we hypothesized that the potential impact of the CL would be the greatest in the first trimester, we analyzed gAC across the 3 cohorts for the first 4 visits. In the full mixed model, the group by time interaction was not significant. In the reduced model, both time (P=0.005) and group (P=0.006) effects were significant. After combining the 1 and >1 CL cohorts, which were similar, and comparing this combined cohort with the 0 CL group, a borderline significant group by time interaction emerged (P=0.078; Figure 6A). gAC was significantly increased at all 3 time points in the first trimester relative to prepregnancy baseline for the 1 CL cohort (P<0.017), at gestational weeks 7 to 9 for the >1 CL cohort (P<0.017), but at no time points for the 0 CL group (Figure 6A). The rise of gAC in the 0 CL cohort when compared with the 1 and >1 CL cohorts combined was less at 10 to 12 gestational weeks (P=0.09). Moreover, the area under the curve for the overall increase in gAC during the first trimester was also significantly different among the 3 cohorts (P=0.045; Figure 6B), and for the cohorts 1 and >1 CL combined relative to the 0 CL group (P=0.017; Figure 6B). The first trimester rise in gAC using SVR derived from the bVbD_CO was also attenuated in the 0 CL group (Figure S5; Results in the Data Supplement).

Figure 6.

Figure 6. Global arterial compliance in pregnancies conceived with or without IVF. A, Absolute values of global arterial compliance before, during, and after pregnancy and (B) changes in global arterial compliance from prepregnancy baseline in women who conceived with or without assisted reproduction. Global arterial compliance calculated based on SVR derived from aVaD_CO and MAP. BP indicates before pregnancy; CL, corpus luteum; and PP, postpartum. +P<0.017 vs BP by Wilcoxon signed-rank test with Bonferroni correction. *P=0.09 for 0 CL compared with 1 and >1 CL combined during gestational wks 10 to 12 by nonparametric ANOVA.

Discussion

In this study, several lines of experimental evidence converged revealing a generalized compromise of maternal cardiovascular function during early pregnancy in women who conceived by IVF in the absence of a CL. The gestational rise in CO, cardiac index, left atrial dimension, and mitral E wave velocity were attenuated during the first trimester in comparison to gravid women with 1 or >1 CL, who were generally similar to each other. In addition, several indices of arterial compliance and wave reflections were also subdued during the first trimester in the 0 CL cohort; specifically, the gestational increases of gAC and peripheral/central pulse pressure ratio, and decrease in aortic augmentation index. These circulatory deficiencies complement the impairment of gestational rise in CAC as reflected by an attenuated decrease in carotid-to-femoral (cf) pulse wave velocity noted previously in this same cohort of women.7 Moreover, the findings herein corroborate and add to the growing list of circulatory perturbations identified in a separate cohort of women who conceived by IVF without a CL, which we also previously reported.18 Of potential clinical relevance is that these impaired cardiovascular adaptations in early pregnancy are associated with an increased risk for preeclampsia and preeclampsia with severe features observed in women who conceived by IVF in the absence of a CL.7 Thus, we propose that secretory product(s) of the CL may play an important role in establishing the maternal cardiovascular adaptations of early pregnancy and affording protection against preeclampsia. As a logical extension, because CL product(s) most likely contribute to the cardiovascular changes during the secretory phase, which are comparable to early pregnancy, but of lesser magnitude,19,20 women conceiving by IVF without a CL may also enter into pregnancy with an inadequately preconditioned cardiovascular system.

An important question is why the normal gestational rise in CO (or CI) during the first trimester was attenuated in the women who conceived by IVF without a CL (Tables S4, S5; Figure 1; Figures S3 and S4, and Results in the Data Supplement). To briefly summarize here, the attenuated response was because of subdued gestational increases of HR more than stroke volume and impaired reduction in both the steady and pulsatile components of arterial load (see Discussion in the Data Supplement).

As previously described,21–23 left atrial dimension progressively increased during pregnancy in all 3 participant cohorts beginning in the first trimester reflecting elevated preload (Table S6; Figure 2). However, the gestational rise in LAD was attenuated during the first trimester in the 0 CL cohort when compared with the women in the control group conceiving without assisted reproduction who had 1 CL (the change in LAD during the first trimester of the >1 CL cohort was intermediate). Left ventricular end diastolic dimension (LVEDD) also progressively increased during pregnancy consistent with earlier reports21–23 but was comparable in the 3 cohorts (Table S6). Thus, the delayed rise of LAD in the 0 CL cohort noted during the first trimester was not mirrored by a similar delayed rise in LVEDD. One possible explanation for this apparent discrepancy is that in normal pregnancy LAD may be a more sensitive indicator of preload because the left atrium is more compliant as reflected by the often observed larger gestational increases in LAD than LVEDD (relative to prepregnancy).21–23

Compared with the nonpregnant condition, mitral E wave velocity was reported to be increased at least during the first half of normal pregnancy by many22,24–26 but not all27–29 investigators. In our study, mitral E wave velocity was increased throughout most of the gestation in both the 1 and >1 CL cohorts but remained unchanged or even declined in the 0 CL cohort (Table S6; Figure 3). Although Ogueh et al29 did not observe a rise in E wave velocity in control pregnancies spontaneously conceived without IVF (1 CL), they did report that E wave velocity was significantly lower in 4 egg donor recipient pregnancies without a CL, in comparison to the control pregnancies at 16 gestational weeks. The deficits in mitral E wave velocity and LAD during early pregnancy in women conceiving by IVF without a CL may have reflected attenuation of the gestational increase in preload and impairment of LV compliance during early diastole. Although we did not detect any differences in the gestational decline of hemoglobin concentration among the 3 cohorts to suggest reduced plasma volume expansion in the 0 CL group (data not shown), we did not directly measure plasma volume. Moreover, comparable hemodilution did not preclude the possibility of deficient vasodilation of liver sinusoids and consequent reduced mobilization of the splanchnic blood reservoir in early pregnancy, thereby compromising preload in the 0 CL cohort.30

Several indices of arterial compliance and wave reflections were also perturbed during early pregnancy in women conceiving by IVF in the absence of a CL. We previously reported that PPAmp ratio (peripheral/central pulse pressure ratio) rapidly rose during the first trimester of normal pregnancy,13 a finding that was recapitulated in the current work in the 1 CL cohort (Table S8; Figure 4). This normal pattern of change was also observed in the >1 CL cohort; however, in the women conceiving by IVF without a CL, PPAmp demonstrated a significantly delayed increase during the first trimester (Table S8; Figure 4). Consistent with this observation, the decline in augmentation index was attenuated in the 0 CL cohort compared with the 1 and >1 CL cohorts during the first trimester, which were similar (Table S9; Figure 5). Typically, AIx% falls during normal pregnancy.13 Indeed, we observed that PPAmp ratio and AIx% were inversely correlated (y=−1.29×AIx%+166.0; R2=0.85; P<0.001) as previously reported by O’Rourke and Adji.31 PPAmp ratio and AIx% are indices of wave reflections, insofar as major factors affecting change in these variables are the magnitude of reflected waves returning from peripheral sites of reflection, and the timing of their return to the heart during the cardiac cycle. In turn, the magnitude of the reflected waves and timing of their return are affected by age, HR, arterial compliance, and diastolic pressure.32–35 In our study, multiple regression models were constructed with the change of PPAmp or AIx% during the first trimester from prepregnant levels as dependent variables (Tables S13 and S14). Age, HR, diastolic pressure, and carotid-femoral PWV or gAC were included in the models as covariates. Of the 4 covariates, only HR and gAC were significantly associated with PPAmp and AIx% (all P<0.001). Taken together, the 4 covariates accounted for between 31 and 45% (R2 values) of the variability in PPAmp and AIx% noted in this study (both P<0.001). For further analysis of PPAmp and Alx% in relation to HR, see Results and Discussion in the Data Supplement.

In addition to the gestational rise in HR, which abetted the increase in PPAmp and reduction in Alx% during the first trimester in the 1 and >1 CL cohorts, another contributing factor was the increased arterial compliance as reflected by decreased cfPWV and increased gAC (Table S12; Figure 6; Figure S5, and Results).7,13 Gestational increases of arterial compliance reduced the magnitude and velocity of the return of reflected waves to the heart, so they arrived later in the cardiac cycle (Tables S9 and S10). Indeed, in all 3 cohorts during the prepregnancy and 3 first trimester visits, PPAmp correlated negatively with cfPWV (y=−7.0×cfPWV+193.7; r2=0.092; P<0.001) and positively with gAC (y=11.7×gAC+130.7; r2=0.24; P>0.001), while AIx% correlated positively with cfPWV (y=5.46×PWV−21.82; r2=0.11; P<0.001) and negatively with gAC (y=−9.38×gAC+27.99; r2=0.30; P<0.001). The correlations with cfPWV were weaker than for gAC consistent with the multiple regression analyses described above. Thus, in addition to gestational elevations of HR, increased arterial compliance contributed to reducing pulsatile arterial load and left ventricular wasted energy during the first trimester of pregnancy in the 1 and >1 CL cohorts (Table S11). Moreover, increase in CAC prevented P1 height from rising during pregnancy, that is, the ascending aortic pressure wave generated by left ventricular ejection of stroke volume, which increases during gestation (Tables S4, S5, and S9). In contrast, the fall in cfPWV7 and rise in gAC (Table S12; Figures 6 and S5) were subdued during the first trimester in the 0 CL cohort, which likely contributed to the attenuated rise in PPAmp and decline in AIx% (Tables S8 and S9; Figures 4 and 5).

PPAmp and Alx% are also negatively and positively correlated, respectively, with central diastolic (or mean) pressure.33,35 Evaluation of PPAmp and AIx% in relation to DP for the 3 participant cohorts is presented in the Discussion in the Data Supplement.

The participants in the 0 CL cohort were significantly older, on average by 4 years, and had a greater body mass index by 3 to 4 kg/m2, compared with the other 2 participant cohorts (Table). To our knowledge, whether a modest elevation in maternal age or BMI per se can lead to impairment of the gestational changes in CO, LAD, E wave velocity, PPAmp, AIx%, and gAC as we observed in the 0 CL cohort is unknown.36,37 As such, we cannot exclude this possibility. Nevertheless, we used a linear mixed model to assess the impact of maternal age and BMI on the change in CO during the first trimester from the prepregnancy baseline among the 3 cohorts. Neither the maternal age–gestational age interaction nor the main effect of maternal age was significant (P=0.90 and P=0.86, respectively). The BMI-gestational age interaction also did not reach statistical significance (P=0.49), but as expected, with advancing gestational age and increasing BMI, CO increased (both main effects P<0.001). These results suggest the modestly older maternal age and increased BMI of the 0 CL cohort were unlikely to have accounted for the subdued gestational rise of CO during the first trimester in these women. In addition, it is noteworthy that the cardiovascular parameters mostly recovered except for E wave velocity at the end of the first or beginning of the second trimester in the 0 CL cohort, when one would have expected any detrimental consequences of prepregnancy factors on circulatory function like elevated maternal age, BMI, or infertility to persist throughout the pregnancy. Rather, the impairment of cardiovascular function during the first trimester in the 0 CL cohort was perhaps more consistent with the absence of circulating vasoactive factor(s) from the CL, which normally peak at that time of gestation. Thereafter, the CL regresses, albeit not completely, and hormones secreted by the developing placenta may have mediated the recovery of cardiovascular function from the earlier deficit in the 0 CL cohort.38,39 Again, however, E wave velocity was an exception, showing little, if any recovery.

Women conceiving by FET in a medicated cycle without a CL are at increased risk for developing preeclampsia or preeclampsia with severe features compared with women conceiving by FET in a modified natural cycle with 1 CL, and matched for maternal age and BMI.7 In the present study, the detriments in cardiovascular function during the first trimester in women conceiving without a CL were also observed in the uncomplicated pregnancies after excluding those with adverse outcomes as exemplified by aVaD_CO and PPAmp shown in Figures S6 and S7, respectively. As such, absence of the CL and the resulting adverse impact on early pregnancy cardiovascular function may be factors predisposing women to develop preeclampsia. Possibly, a healthy placenta or maternal factors can compensate to overcome this predisposition. Alternatively, a second deleterious event may be needed to elicit disease. In other investigations powered for prediction of obstetric outcomes, perturbed circulatory adaptations during early gestation were associated with adverse pregnancy outcomes including preeclampsia.2,8,9 It should also be noted that the detriments in cardiovascular function during the first trimester in women conceiving without a CL were also observed in singleton pregnancies after excluding twins, and in both nulliparous and multiparous subgroups (data not shown).

Circulating relaxin is one plausible candidate hormone, which is absent in pregnant women without a CL, because relaxin is secreted solely from the CL throughout human pregnancy,40–42 is a potent vasodilator,11,43,44 and mediates maternal vasodilation and increases arterial compliance in the gravid rat model.4,5 That is, the absence of circulating relaxin, which was below the limit of assay detection throughout pregnancy in all women without CL in this study (data not shown) may have contributed to the cardiovascular dysfunction in the first trimester and higher preeclampsia risk as previously reported.7 However, correlations between circulating relaxin concentration and CO, SVR, or HR were weak in the 1 CL and >1 CL cohorts (data not shown) suggesting that relaxin is not the mediator of these physiological gestational changes, or perhaps acts permissively exerting a dichotomous dose response with full effect regardless of the circulating concentration as in the 1 and >1 CL cohorts, and no effect, if completely absent as in the case of the 0 CL cohort. On the contrary, perhaps the limited number of participants precluded significant correlations. Conceivably, other unidentified CL factors(s) may mediate maternal cardiovascular changes during early pregnancy in women. Proof of this concept requires further research, and ultimately perhaps, demonstration that replacement of the missing CL factor(s) in women conceiving by IVF in the absence of a CL restores the cardiovascular deficits and reduces the incidence of preeclampsia.

Last, it is noteworthy that in the prepregnant and postpartum state, women who conceived with assisted reproduction regardless of the IVF protocol, generally demonstrated elevated left ventricular mass and thicker wall dimensions, and lower cardiac index compared with women conceiving without assisted reproduction (Table S6; Figure S4, Results and Discussion in the Data Supplement). Thus, both IVF groups (0 CL and >1 CL cohorts) had similar prepregnancy differences in cardiovascular function compared with 1 CL. But, only the 0 CL cohort manifested attenuated changes in cardiovascular adaptations during early pregnancy and increased rates of preeclampsia and preeclampsia with severe features.7 This divergence would seem to mitigate against these prepregnancy factors as being the (only) reason for the attenuated gestational changes in the first trimester and increased rates of preeclampsia and preeclampsia with severe features in the 0 CL cohort.45

This study of maternal physiology in women conceiving by IVF has several significant strengths. First and foremost, it provides new insights into the cardiovascular function of normal and IVF pregnancies with implications for pregnancies complicated by preeclampsia. Second, comprehensive, serial investigation of cardiovascular function throughout pregnancy in women conceiving by IVF in the context of CL status is, to our knowledge, unprecedented. Finally, another strength is the numerous serial assessments of multiple cardiovascular parameters during pregnancy including a prepregnant baseline control performed in the absence of the CL.

The comprehensive, longitudinal design of our study limited the number of participants to ≈20 per cohort. Thus, our investigation was underpowered to conduct covariate analyses, for example, for adverse obstetric outcomes, although the overall pattern of change in cardiovascular parameters during the first trimester for the 3 cohorts was similar regardless of the inclusion or exclusion of those women with adverse outcomes (eg, Figures S6 and S7). Subfertile participants who ultimately conceived without assisted reproduction were not included as an additional control group for the 0 CL cohort. As mentioned above, however, the 0 CL cohort manifested recovery of most cardiovascular parameters at the end of the first or beginning of the first trimester. This finding was inconsistent with a confounding influence of subfertility or reduced ovarian reserve, which would persist throughout gestation. Rather, it implicated missing circulating CL factor(s) in the first trimester that normally predominate before the corpus luteal-placental shift, after which placental factor(s) circulate in sufficient amounts to exert cardiovascular actions.

Perspectives

Increased preeclampsia risk was associated with frozen embryo transfer cycles in the absence of a CL, but there was no difference in the occurrence of preeclampsia between modified natural cycle FET and spontaneous conception (both with 1 CL).7 In parallel, gestational increases of CAC were impaired in the absence of a CL during the first trimester.7 In the present work, we provided evidence for a more generalized compromise of cardiovascular function in women conceiving by IVF without a CL. Conceivably, the impairment of cardiovascular adaptations in early pregnancy and the increased risk of preeclampsia when the CL is absent could be mitigated by using more physiological IVF protocols, which lead to development of a CL such as a modified natural cycle, or by replacing the missing circulating CL product(s). Absence of circulating CL factor(s), perturbed maternal circulatory adaptation to pregnancy, and increased preeclampsia risk may be linked, but additional investigation is needed to establish causality.

Acknowledgments

We gratefully acknowledge all study participants and the following colleagues for their invaluable contributions to this research: Kevin Bishop, ARNP and Lynn Musselman, BA, CHRC, Recruitment Coordinators; Elaine Whidden, ARNP, Research Coordinator; Jessica L. Cline, BS, Julie Bailes, BS, and T.J. Arndt, MPH, CPH, Data Managers; Mathew S. Evron, BS, echo analysis assistant; and Elizabeth R. Currin, PhD, Administrative Assistant. We also thank Charles E. Wood, PhD, Maureen Keller-Wood, PhD, and Sanjeev G. Shroff, PhD for helpful discussion. M.S. Segal, W.W. Nichols, and K.P. Conrad performed study design; M.D. Lingis and J.J. Larocca performed data collection; J.W. Petersen performed echocardiogram analysis; K.-H. Chiu, J. Liu, X. Zhai, M.D. Lingis, M. Li, and Y.-Y. Chi performed data management and statistical analysis; R.S. Williams and A. Rhoton-Vlasak performed clinical liaison and patients for the study. The first draft of the manuscript was written by K.P. Conrad. All co-authors contributed to the revised manuscript and approved the final version.

Footnotes

The online-only Data Supplement is available with this article at https://www.ahajournals.org/doi/suppl/10.1161/HYPERTENSIONAHA.119.13015.

Correspondence to Kirk P. Conrad, Departments of Physiology and Functional Genomics, and Obstetrics and Gynecology, University of Florida College of Medicine, 1600 SW Archer Rd, PO Box 100274, Gainesville, FL 32610-0274. Email

References

  • 1. Coutifaris C. “Freeze only”–an evolving standard in clinical in vitro fertilization.N Engl J Med. 2016; 375:577–579. doi: 10.1056/NEJMe1606213CrossrefMedlineGoogle Scholar
  • 2. Conrad KP, Baker VL. Corpus luteal contribution to maternal pregnancy physiology and outcomes in assisted reproductive technologies.Am J Physiol Regul Integr Comp Physiol. 2013; 304:R69–R72. doi: 10.1152/ajpregu.00239.2012CrossrefMedlineGoogle Scholar
  • 3. Conrad KP. Maternal vasodilation in pregnancy: the emerging role of relaxin.Am J Physiol Regul Integr Comp Physiol. 2011; 301:R267–R275. doi: 10.1152/ajpregu.00156.2011CrossrefMedlineGoogle Scholar
  • 4. Novak J, Danielson LA, Kerchner LJ, Sherwood OD, Ramirez RJ, Moalli PA, Conrad KP. Relaxin is essential for renal vasodilation during pregnancy in conscious rats.J Clin Invest. 2001; 107:1469–1475. doi: 10.1172/JCI11975CrossrefMedlineGoogle Scholar
  • 5. Debrah DO, Novak J, Matthews JE, Ramirez RJ, Shroff SG, Conrad KP. Relaxin is essential for systemic vasodilation and increased global arterial compliance during early pregnancy in conscious rats.Endocrinology. 2006; 147:5126–5131. doi: 10.1210/en.2006-0567CrossrefMedlineGoogle Scholar
  • 6. Smith MC, Murdoch AP, Danielson LA, Conrad KP, Davison JM. Relaxin has a role in establishing a renal response in pregnancy.Fertil Steril. 2006; 86:253–255. doi: 10.1016/j.fertnstert.2005.11.070CrossrefMedlineGoogle Scholar
  • 7. von Versen-Höynck F, Schaub AM, Chi YY, Chiu KH, Liu J, Lingis M, Stan Williams R, Rhoton-Vlasak A, Nichols WW, Fleischmann RR, Zhang W, Winn VD, Segal MS, Conrad KP, Baker VL. Increased preeclampsia risk and reduced aortic compliance with in vitro fertilization cycles in the absence of a corpus luteum.Hypertension. 2019; 73:640–649. doi: 10.1161/HYPERTENSIONAHA.118.12043LinkGoogle Scholar
  • 8. De Paco C, Kametas N, Rencoret G, Strobl I, Nicolaides KH. Maternal cardiac output between 11 and 13 weeks of gestation in the prediction of preeclampsia and small for gestational age.Obstet Gynecol. 2008; 111(2pt 1):292–300. doi: 10.1097/01.AOG.0000298622.22494.0cCrossrefMedlineGoogle Scholar
  • 9. Khaw A, Kametas NA, Turan OM, Bamfo JE, Nicolaides KH. Maternal cardiac function and uterine artery doppler at 11-14 weeks in the prediction of pre-eclampsia in nulliparous women.BJOG. 2008; 115:369–376. doi: 10.1111/j.1471-0528.2007.01577.xCrossrefMedlineGoogle Scholar
  • 10. Gasse C, Boutin A, Coté M, Chaillet N, Bujold E, Demers S. First-trimester mean arterial blood pressure and the risk of preeclampsia: The Great Obstetrical Syndromes (GOS) study.Pregnancy Hypertens. 2018; 12:178–182. doi: 10.1016/j.preghy.2017.11.005CrossrefMedlineGoogle Scholar
  • 11. Conrad KP, Debrah DO, Novak J, Danielson LA, Shroff SG. Relaxin modifies systemic arterial resistance and compliance in conscious, nonpregnant rats.Endocrinology. 2004; 145:3289–3296. doi: 10.1210/en.2003-1612CrossrefMedlineGoogle Scholar
  • 12. Petersen JW, Liu J, Chi YY, Lingis M, Williams RS, Rhoton-Vlasak A, Segal MS, Conrad KP. Comparison of multiple non-invasive methods of measuring cardiac output during pregnancy reveals marked heterogeneity in the magnitude of cardiac output change between women.Physiol Rep. 2017; 5:e13223.CrossrefMedlineGoogle Scholar
  • 13. Rodriguez C, Chi YY, Chiu KH, Zhai X, Lingis M, Williams RS, Rhoton-Vlasak A, Nichols WW, Petersen JW, Segal MS, Conrad KP, Mohandas R. Wave reflections and global arterial compliance during normal human pregnancy.Physiol Rep. 2018; 6:e13947. doi: 10.14814/phy2.13947CrossrefMedlineGoogle Scholar
  • 14. Khalil A, Garcia-Mandujano R, Maiz N, Elkhouli M, Nicolaides KH. Longitudinal changes in maternal hemodynamics in a population at risk for pre-eclampsia.Ultrasound Obstet Gynecol. 2014; 44:197–204. doi: 10.1002/uog.13367CrossrefMedlineGoogle Scholar
  • 15. Katsipi I, Stylianou K, Petrakis I, Passam A, Vardaki E, Parthenakis F, Makrygiannakis A, Daphnis E, Kyriazis J. The use of pulse wave velocity in predicting pre-eclampsia in high-risk women.Hypertens Res. 2014; 37:733–740. doi: 10.1038/hr.2014.62CrossrefMedlineGoogle Scholar
  • 16. Butlin M, Qasem A. Large artery stiffness assessment using sphygmoCor technology.Pulse (Basel). 2017; 4:180–192. doi: 10.1159/000452448CrossrefMedlineGoogle Scholar
  • 17. Nichols WW, O’Rourke MF, Vlachopoulos C. Pressure Pulse Waveform Analysis McDonald’s Blood Flow in Arteries. 6th ed. Boca Raton: CRC Press;2011:595–638.Google Scholar
  • 18. von Versen-Höynck F, Narasimhan P, Selamet Tierney ES, Martinez N, Conrad KP, Baker VL, Winn VD. Absent or excessive corpus luteum number is associated with altered maternal vascular health in early pregnancy.Hypertension. 2019; 73:680–690. doi: 10.1161/HYPERTENSIONAHA.118.12046LinkGoogle Scholar
  • 19. Chapman AB, Zamudio S, Woodmansee W, Merouani A, Osorio F, Johnson A, Moore LG, Dahms T, Coffin C, Abraham WT, Schrier RW. Systemic and renal hemodynamic changes in the luteal phase of the menstrual cycle mimic early pregnancy.Am J Physiol. 1997; 273:F777–F782. doi: 10.1152/ajprenal.1997.273.5.F777MedlineGoogle Scholar
  • 20. Conrad KP, Stillman IE, Lindheimer MD. The kidney in normal pregnancy and preeclampsia.Taylor R. N., Roberts J. M., Cunningham F. G. and Lindheimer M. D., eds. In: Chesley’s Hypertensive Disorders in Pregnancy. 4th ed. San Diego: Academic Press; 2015:335–377.Google Scholar
  • 21. Robson SC, Hunter S, Boys RJ, Dunlop W. Serial study of factors influencing changes in cardiac output during human pregnancy.Am J Physiol. 1989; 256(4 pt 2):H1060–H1065. doi: 10.1152/ajpheart.1989.256.4.H1060MedlineGoogle Scholar
  • 22. Kametas NA, McAuliffe F, Hancock J, Chambers J, Nicolaides KH. Maternal left ventricular mass and diastolic function during pregnancy.Ultrasound Obstet Gynecol. 2001; 18:460–466. doi: 10.1046/j.0960-7692.2001.00573.xCrossrefMedlineGoogle Scholar
  • 23. Melchiorre K, Sharma R, Thilaganathan B. Cardiac structure and function in normal pregnancy.Curr Opin Obstet Gynecol. 2012; 24:413–421. doi: 10.1097/GCO.0b013e328359826fCrossrefMedlineGoogle Scholar
  • 24. Estensen ME, Beitnes JO, Grindheim G, Aaberge L, Smiseth OA, Henriksen T, Aakhus S. Altered maternal left ventricular contractility and function during normal pregnancy.Ultrasound Obstet Gynecol. 2013; 41:659–666. doi: 10.1002/uog.12296CrossrefMedlineGoogle Scholar
  • 25. Fok WY, Chan LY, Wong JT, Yu CM, Lau TK. Left ventricular diastolic function during normal pregnancy: assessment by spectral tissue doppler imaging.Ultrasound Obstet Gynecol. 2006; 28:789–793. doi: 10.1002/uog.3849CrossrefMedlineGoogle Scholar
  • 26. Mesa A, Jessurun C, Hernandez A, Adam K, Brown D, Vaughn WK, Wilansky S. Left ventricular diastolic function in normal human pregnancy.Circulation. 1999; 99:511–517. doi: 10.1161/01.cir.99.4.511LinkGoogle Scholar
  • 27. Zentner D, du Plessis M, Brennecke S, Wong J, Grigg L, Harrap SB. Deterioration in cardiac systolic and diastolic function late in normal human pregnancy.Clin Sci (Lond). 2009; 116:599–606. doi: 10.1042/CS20080142CrossrefMedlineGoogle Scholar
  • 28. Simmons LA, Gillin AG, Jeremy RW. Structural and functional changes in left ventricle during normotensive and preeclamptic pregnancy.Am J Physiol Heart Circ Physiol. 2002; 283:H1627–H1633. doi: 10.1152/ajpheart.00966.2001CrossrefMedlineGoogle Scholar
  • 29. Ogueh O, Brookes C, Johnson MR. A longitudinal study of the maternal cardiovascular adaptation to spontaneous and assisted conception pregnancies.Hypertens Pregnancy. 2009; 28:273–289. doi: 10.1080/10641950802601203CrossrefMedlineGoogle Scholar
  • 30. Conrad KP. G-Protein-coupled receptors as potential drug candidates in preeclampsia: targeting the relaxin/insulin-like family peptide receptor 1 for treatment and prevention.Hum Reprod Update. 2016; 22:647–664. doi: 10.1093/humupd/dmw021CrossrefMedlineGoogle Scholar
  • 31. O’Rourke MF, Adji A. Basis for use of central blood pressure measurement in office clinical practice.J Am Soc Hypertens. 2008; 2:28–38. doi: 10.1016/j.jash.2007.08.006CrossrefMedlineGoogle Scholar
  • 32. Nichols WW, O’Rourke MF, Vlachopoulos C. McDonald’s Blood Flow in Arteries:Theoretical, Experimental and Clinical Principles. 6th ed. Boca Raton, FL: CRC Press; 2011.Google Scholar
  • 33. Wilkinson IB, Franklin SS, Hall IR, Tyrrell S, Cockcroft JR. Pressure amplification explains why pulse pressure is unrelated to risk in young subjects.Hypertension. 2001; 38:1461–1466.LinkGoogle Scholar
  • 34. Wilkinson IB, MacCallum H, Flint L, Cockcroft JR, Newby DE, Webb DJ. The influence of heart rate on augmentation index and central arterial pressure in humans.J Physiol. 2000; 525(pt 1):263–270. doi: 10.1111/j.1469-7793.2000.t01-1-00263.xCrossrefMedlineGoogle Scholar
  • 35. Wilkinson IB, MacCallum H, Hupperetz PC, van Thoor CJ, Cockcroft JR, Webb DJ. Changes in the derived central pressure waveform and pulse pressure in response to angiotensin II and noradrenaline in man.J Physiol. 2001; 530(pt 3):541–550. doi: 10.1111/j.1469-7793.2001.0541k.xCrossrefMedlineGoogle Scholar
  • 36. Vinayagam D, Thilaganathan B, Stirrup O, Mantovani E, Khalil A. Maternal hemodynamics in normal pregnancy: reference ranges and role of maternal characteristics.Ultrasound Obstet Gynecol. 2018; 51:665–671. doi: 10.1002/uog.17504CrossrefMedlineGoogle Scholar
  • 37. Guy GP, Ling HZ, Garcia P, Poon LC, Nicolaides KH. Maternal cardiovascular function at 35-37 weeks’ gestation: relation to maternal characteristics.Ultrasound Obstet Gynecol. 2017; 49:39–45. doi: 10.1002/uog.17311CrossrefMedlineGoogle Scholar
  • 38. Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, Schisterman EF, Thadhani R, Sachs BP, Epstein FH, Sibai BM, Sukhatme VP, Karumanchi SA. Circulating angiogenic factors and the risk of preeclampsia.N Engl J Med. 2004; 350:672–683. doi: 10.1056/NEJMoa031884CrossrefMedlineGoogle Scholar
  • 39. Than NG, Balogh A, Romero R, Kárpáti E, Erez O, Szilágyi A, Kovalszky I, Sammar M, Gizurarson S, Matkó J, Závodszky P, Papp Z, Meiri H. Placental protein 13 (PP13) - A placental immunoregulatory galectin protecting pregnancy.Front Immunol. 2014; 5:348. doi: 10.3389/fimmu.2014.00348CrossrefMedlineGoogle Scholar
  • 40. Sherwood OD. Relaxin.Knobil E., Neill JD., Greenwald G S., Markert CL. and Pfaff DW., eds. In: The Physiology of Reproduction. 2nd ed. New York, NY: Raven Press; 1994:861–1008.Google Scholar
  • 41. Johnson MR, Abdalla H, Allman AC, Wren ME, Kirkland A, Lightman SL. Relaxin levels in ovum donation pregnancies.Fertil Steril. 1991; 56:59–61.CrossrefMedlineGoogle Scholar
  • 42. von Versen-Höynck F, Strauch NK, Liu J, Chi YY, Keller-Woods M, Conrad KP, Baker VL. Effect of mode of conception on maternal serum relaxin, creatinine, and sodium concentrations in an infertile population.Reprod Sci. 2019; 26:412–419. doi: 10.1177/1933719118776792CrossrefMedlineGoogle Scholar
  • 43. Danielson LA, Sherwood OD, Conrad KP. Relaxin is a potent renal vasodilator in conscious rats.J Clin Invest. 1999; 103:525–533. doi: 10.1172/JCI5630CrossrefMedlineGoogle Scholar
  • 44. Smith MC, Danielson LA, Conrad KP, Davison JM. Influence of recombinant human relaxin on renal hemodynamics in healthy volunteers.J Am Soc Nephrol. 2006; 17:3192–3197. doi: 10.1681/ASN.2005090950CrossrefMedlineGoogle Scholar
  • 45. Foo FL, Mahendru AA, Masini G, Fraser A, Cacciatore S, MacIntyre DA, McEniery CM, Wilkinson IB, Bennett PR, Lees CC. Association between prepregnancy cardiovascular function and subsequent preeclampsia or fetal growth restriction.Hypertension. 2018; 72:442–450. doi: 10.1161/HYPERTENSIONAHA.118.11092LinkGoogle Scholar

Novelty and Significance

What Is New?

  • Physiological transformation of the maternal cardiovascular system from the prepregnant to pregnant state was impaired in women conceiving by in vitro fertilization in the absence of a corpus luteum (CL).

  • Multiple cardiovascular parameters were compromised in the first trimester indicating a generalized impairment.

What Is Relevant?

  • CL factor(s) secreted into the maternal circulation play a role in establishing maternal cardiovascular adaptations during early pregnancy.

  • Cardiovascular dysregulation during early pregnancy in women conceiving by autologous frozen embryo transfer or donor eggs in a medicated cycle without a CL may be a predisposing factor for developing preeclampsia or preeclampsia with severe features, which occurs more frequently in these pregnancies.

Summary

Evidence is provided for an important role of circulating CL factor(s) in establishing the maternal cardiovascular transformation during early pregnancy, insofar as in pregnancies without a CL, the cardiovascular transformation was impaired. After the corpus luteal-placental shift, placental factors likely contributed to partial or full recovery of most cardiovascular parameters.

Whether absent circulating CL factor(s), cardiovascular dysregulation in the first trimester, and enhanced preeclampsia risk in pregnancies without a CL are causally linked requires further investigation. Ultimately, replacing the critical missing CL factor(s) may prove beneficial.

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.