Association Between Prepregnancy Blood Pressure and Reproductive Outcomes of In Vitro Fertilization

Abstract

BACKGROUND:

The latest updated 2025 and 2017 American College of Cardiology and the American Heart Association guidelines lowered the diagnostic threshold for hypertension to 130/80 mm Hg. Whether the new classification for hypertension has implications for reproductive outcomes remains uncertain.

METHODS:

This retrospective cohort study was conducted at the Reproductive Medicine Center of Shandong University in China. Women who underwent the initial embryo transfer of their first in vitro fertilization cycle were categorized into the normal blood pressure (BP), elevated BP, stage 1 hypertension, and stage 2 hypertension groups based on BP levels measured just before in vitro fertilization treatment. We examined associations of prepregnancy BP and reproductive outcomes.

RESULTS:

This study included 43 629 women who received in vitro fertilization treatment. The rate of live birth was lower in women with stage 1 and stage 2 hypertensions (46.1% and 41.4%, respectively) compared with women with normal BP (49.2%), with the adjusted relative ratios of 0.97 (95% CI, 0.937–0.996; P=0.027) and 0.91 (95% CI, 0.85–0.98; P=0.009), respectively. Compared with normal BP, both stage 1 and stage 2 hypertension were associated with higher risks of pregnancy loss, preeclampsia, and preterm delivery. Elevated BP was associated with a higher risk of gestational hypertension. Optimal BP cutoffs for adverse reproductive outcomes were consistent with the diagnostic threshold for stage 1 hypertension.

CONCLUSIONS:

Compared with normal BP, prepregnancy stage 1 and stage 2 hypertension were associated with a lower rate of live birth after in vitro fertilization treatment and increased risks of pregnancy complications.

Novelty and Relevance.

What Is New?

• Prepregnancy stage 1 and 2 hypertension diagnosed by the 2025 American College of Cardiology/American Heart Association guideline was associated with worse reproductive outcomes compared with normal blood pressure.

What Is Relevant?

• Hypertension in women of reproductive age was not only associated with an increased risk of long-term cardiovascular diseases but also with adverse reproductive outcomes.

Clinical/Pathophysiological Implications?

• Further studies on intervention strategies for hypertension before pregnancy, particularly for stage 1 hypertension, are warranted. Reproductive endocrinologists need to work closely with cardiologists to manage hypertension in women of childbearing age.

Prepregnancy hypertension is associated with increased risks of adverse reproductive outcomes, including preterm delivery, stillbirth, fetal growth restriction, placental abruption, and cesarean delivery.^1–3 Hypertension is typically defined as a systolic blood pressure (SBP) ≥140 mm Hg or a diastolic blood pressure (DBP) ≥90 mm Hg.⁴ The latest updated 2025 recommendation by the American College of Cardiology (ACC) and the American Heart Association (AHA) endorse the continued use of the 2017 ACC/AHA guideline for the diagnostic threshold for hypertension,^5,6 which proposed elevated blood pressure (BP; SBP of 120–129 mm Hg and DBP <80 mm Hg) and stage 1 hypertension (SBP of 130–139 mm Hg or DBP of 80–89 mm Hg) while classifying traditional hypertension as stage 2 hypertension. The new classification framework, however, was primarily based on the long-term cardiovascular disease risks of the general population, with limited consideration of reproductive health and the population seeking pregnancy. It was still unclear whether the new classification for hypertension has implications for reproductive outcomes. Several studies evaluated the association between hypertension in early gestation (<20 weeks of gestation) and the risks of obstetric complications in women with natural conception.^2,3,7,8 However, the association between prepregnancy hypertension and reproductive outcomes was rarely studied.

Women undergoing in vitro fertilization (IVF) treatment serve as a good model for assessing the effect of prepregnancy BP on reproductive outcomes. In this setting, BP is measured uniformly before initiating IVF treatment. Approximately 40% to 50% of women could achieve pregnancy within a short period due to the high efficacy of IVF treatment.^9–11 Thus, BP levels measured shortly before conception, rather than those during early gestation or several months to 1 year before conception, can be used to analyze the association with reproductive outcomes. In addition, among women undergoing IVF, systemic follow-ups throughout pregnancy are routinely performed, and all pregnancy outcomes (including first-trimester pregnancy loss) can be captured.

In this large-scale cohort study of women undergoing IVF treatment, we investigated the relationship between maternal prepregnancy BP classified by the 2025 ACC/AHA guideline and reproductive outcomes. In addition, we sought to determine the optimal cutoff values for SBP and DBP with regard to reproductive outcomes.

Methods

Data Availability

The data that support the findings of this study are available from the corresponding author on reasonable request.

Data Sources and Study Population

This single-center cohort study was approved by the ethics committee of the Reproductive Medicine Center of Shandong University. The report of findings followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines. Data were extracted from the clinical database, and informed consent was waived due to the retrospective nature and deidentification of participants.

Women who underwent their initial embryo transfer of the first cycle of IVF with or without intracytoplasmic sperm injection between January 2017 and August 2023 were included in the study. Follow-up of pregnancy outcomes was completed on June 30, 2024. Exclusion criteria were: (1) women who underwent preimplantation genetic testing cycles, in vitro maturation cycles, or cycles with donated oocytes; (2) women with the diagnosis of uterine abnormalities; (3) women who had a history of recurrent spontaneous miscarriage (defined as ≥2 spontaneous miscarriage before 20 weeks of gestation); (4) women with missing data on SBP or DBP; and (5) women with incomplete follow-up information.

Exposures

Before the initiation of ovarian stimulation, resting BP was measured by trained nurses using an automatic BP monitor (Mibobo; Shenzhen Raycome Health Technology Co, Ltd, Shenzhen, China). Women were instructed to avoid caffeine and exercise for at least 30 minutes before measurement and advised to empty their bladders and rest quietly for at least 5 minutes. Measurement was taken in the upright seated position with the cuff placed on the bare upper arm at heart level. An appropriately sized cuff was selected according to the upper arm circumference. BP was measured twice at a 5-minute interval and then the readings were averaged. Mean arterial pressure (MAP) was calculated as one-third of SBP plus two-thirds of DBP (MAP=SBP/3+2×DBP/3).

According to the 2025 ACC/AHA guideline, women were categorized into 1 of the following 4 groups: the normal BP group (SBP<120 mm Hg and DBP<80 mm Hg), the elevated BP group (120≤SBP <130 mm Hg and DBP<80 mm Hg), the stage 1 hypertension group (130≤SBP<140 mm Hg or 80≤DBP<90 mm Hg), and the stage 2 hypertension group (SBP≥140 mm Hg or DBP≥90 mm Hg).⁶

Women meeting the diagnosis of stage 2 hypertension were referred to cardiologists for consultation on hypertension treatment. Women diagnosed with stage 1 hypertension or elevated BP were verbally advised to make lifestyle modifications, including restricting sodium intake, keeping a healthy diet and physical activity, and losing weight for women with obesity and overweight.

Outcomes

The main outcome was live birth after the initial embryo transfer, which was defined as delivery of at least 1 live neonate at 24 weeks of gestation or later. The secondary outcomes included biochemical pregnancy, clinical pregnancy, ongoing pregnancy, pregnancy loss, ectopic pregnancy, maternal complications, and neonatal complications (the definitions of secondary outcomes were provided in Table S1).

Statistical Analysis

Continuous variables that followed a normal distribution were expressed as mean±SD and compared using ANOVA. Continuous variables that did not meet the criteria of normality were expressed as median and interquartile range (interquartile range) or presented as box-and-whisker plot (median, 25th and 75th percentiles as the limits of the box, and fifth and 95th percentiles as the limits of the whiskers), and compared using Kruskal-Wallis tests. Categorical variables were presented as frequencies (percentages) and analyzed using the Pearson χ² test or Fisher exact test, as appropriate.

Modified Poisson regression (Poisson regression with a robust error variance) was used to estimate adjusted relative ratios (aRRs) and 95% CIs for the associations between prepregnancy BP and dichotomized outcomes because it could directly estimate relative risk and is less prone to convergence problems.^12,13 We constructed 3 nested regression models by incrementally controlling for covariates to evaluate the robustness and potential confounding of the associations. Model 1 was unadjusted. Model 2 was adjusted for a core set of demographic and clinical characteristics, including age (continuous), body mass index (continuous), endometrial thickness (continuous), type of embryo transferred (fresh or frozen), number of oocytes retrieved (continuous), stage of embryo transferred (cleavage stage or blastocyst), primary infertility (yes or no), and a diagnosis of polycystic ovary syndrome (yes or no). Model 3 was further adjusted for additional reproductive parameters, including antral follicle count (continuous), anti-Müllerian hormone (continuous), duration of infertility (continuous), type of fertilization (IVF, intracytoplasmic sperm injection, or rescue intracytoplasmic sperm injection), and number of embryos transferred (1 or 2). Model 3 served as the primary analytic model. Covariates included in each model were selected based on clinical relevance and prior literature reports.^14,15 Kaplan-Meier curves were plotted to visualize the distribution of the cumulative probability of delivery outcomes over time across BP categories for women with ongoing pregnancies beyond 12 weeks of gestation (using gestational weeks as the time scale, delivery was coded as the event of interest, and pregnancy loss was treated as a censored observation).

BP was also modeled as a continuous variable. The average effects on reproductive outcomes were estimated per 10-mm Hg increment in SBP and DBP and per 5-mm Hg increment in MAP. To describe the trends and potential inflection points between BP (SBP, DBP, and MAP, respectively) and reproductive outcomes, smoothed curves were initially plotted using a moving average with a window size of 3 as reported in the literature.¹⁶ Specifically, SBP, DBP, and MAP were first divided by 5 mm Hg intervals, and the rates of reproductive pregnancy outcomes were calculated within each interval. To reduce random fluctuations and better visualize overall trends, we calculated the average rates across every 3 adjacent intervals. Point estimates and corresponding 95% CIs were calculated using the Wald method for binomial proportions and displayed on the figures. We fitted restricted cubic spline models based on modified Poisson regression (4 knots at the fifth, 35th, 65th, and 95th percentiles) to estimate the adjusted associations of SBP, DBP, and MAP with reproductive outcomes, respectively. The nonlinearity was evaluated using Wald statistics. Subsequently, regression analyses were conducted using SBP, DBP, and MAP categorized according to the observed thresholds. We additionally performed secondary analyses restricted to women achieving a singleton live birth to examine the robustness of the results and performed stratified analyses according to the embryo transfer type (fresh and frozen embryo transfer) or the developmental stage of transferred embryo(s; cleavage-stage and blastocyst-stage).

A 2-sided P<0.05 was considered statistically significant for the main outcome of live birth. No adjustment was made for multiple comparisons; therefore, analyses of the secondary outcomes should be interpreted as exploratory. All statistical analyses were conducted using R software (version 4.4.2).

Results

Baseline Characteristics of Participants

A total of 43 629 eligible women were included in the final analysis (Figure 1). Of them, 30 743 (70.5%) had normal BP, 6206 (14.2%) had elevated BP, 5594 (12.8%) had stage 1 hypertension, and 1086 (2.5%) had stage 2 hypertension. Baseline characteristics are presented in Table 1. The mean age at baseline was 32.3 years (SD, 4.7 years). Compared with women with normal BP, women with elevated BP or hypertension had higher body mass index and were more likely to be diagnosed with polycystic ovary syndrome (Table 1). The median (interquartile range) endometrial thickness before embryo transfer and median (interquartile range) number of oocytes retrieved of each BP category were numerically similar, but the differences across the groups were statistically significant (Table 2), and the distributions are illustrated by a box-and-whisker plot (Figures S1 and S2).

Figure 1.

Study flow chart. DBP indicates diastolic blood pressure; IVF-ET, in vitro fertilization and embryo transfer; and SBP, systolic blood pressure.

Table 1.

Demographics and Clinical Characteristics of Participants Across Prepregnancy BP Categories

Table 2.

Frequencies of In Vitro Fertilization Treatment Outcomes Across Prepregnancy BP Categories

Reproductive Outcomes Across BP Categories

The live birth rate decreased in a dose-dependent manner with the increase of BP (49.2% for normal BP, 47.9% for elevated BP, 46.1% for stage 1 hypertension, and 41.4% for stage 2 hypertension, P<0.001; Table 3). Multivariable regression results from the fully adjusted model (model 3) showed the aRR in stage 1 and stage 2 hypertension group for live birth were 0.97 ([95% CI, 0.937–0.996]; P=0.027) and 0.91 ([95% CI, 0.85–0.98]; P=0.009), respectively, compared with the normal BP group (Table 3). The results from unadjusted and partially adjusted models (models 1 and 2) were consistent with the main results (Table S2).

Table 3.

Frequencies and Relative Ratios of Reproductive Outcomes Across Prepregnancy BP Categories

We did not observe a significant association between stage 1 or stage 2 hypertension and the rate of singleton live birth after adjustment (Table 3). Compared with the normal BP, stage 2 hypertension was significantly associated with a decreased rate of clinical pregnancy (57.8% versus 51.6%; aRR, 0.92 [95% CI, 0.87–0.98]; P=0.007; Table 3) and an increased risk of biochemical pregnancy loss (11.3% versus 14.8%; aRR, 1.30 [95% CI, 1.08–1.58]; P=0.006; Tables 4). Both stage 1 (aRR, 1.06 [95% CI, 1.01–1.13]; P=0.033) and stage 2 (aRR, 1.14 [95% CI, 1.02–1.28]; P=0.025) hypertension groups were significantly associated with a higher risk of total pregnancy loss compared with the normal BP (Table 4).

Table 4.

Frequencies and Relative Ratios of Maternal and Neonatal Complications Across Prepregnancy BP Categories

Compared with normal BP, both stage 1 and stage 2 hypertension were significantly associated with higher risks of preterm delivery (aRR, 1.15 [95% CI, 1.04–1.27]; P=0.005 for stage 1 hypertension; aRR, 1.40 [95% CI, 1.16–1.69]; P=0.001 for stage 2 hypertension; Table 4). A significant association was observed between stage 1 hypertension and higher risks of gestational hypertension (aRR, 2.90 [95% CI, 2.54–3.32]; P<0.001) and preeclampsia (aRR, 2.33 [95% CI, 1.62–3.33]; P<0.001; Table 4). The singleton birthweight differed across BP categories (P=0.001), and the distribution of values within each group was illustrated in a box-and-whisker plot (Figure S3). Compared with normal BP, stage 2 hypertension was significantly associated with higher risks of low birth weight (aRR, 1.58 [95% CI, 1.08–2.31]; P=0.019) and small for gestational age (aRR, 2.20 [95% CI, 1.52–3.19]; P<0.001) for singleton infants (Table 4). Both stage 1 and stage 2 hypertension were associated with a lower risk of large for gestational age (Table 4). For women with ongoing pregnancies, Kaplan-Meier cumulative incidence curves representing time-to-delivery across BP categories were presented in Figure S4. Women with stage 2 hypertension exhibited shorter gestational age compared with women with normal BP.

Effects of SBP, DBP, and MAP as Continuous Variables on Reproductive Outcomes

The adjusted RR for live birth per 10-mm Hg increase in SBP was 0.99 ([95% CI, 0.98–0.99]; P=0.001) and in DBP was 0.99 ([95% CI, 0.98–1.00]; P=0.090). In addition, each 5-mm Hg increase in MAP was also associated with a 1% lower chance of live birth (aRR, 0.99 [95% CI, 0.988–0.998]; P=0.010; Table S3). SBP was negatively related to clinical pregnancy rate, while the associations were not statistically significant for DBP and MAP. SBP, DBP, and MAP were all positively associated with the risks of total pregnancy loss, preterm delivery, gestational hypertension, and preeclampsia (Table S3).

Graphs for Trends in Reproductive Outcomes and the Cutoff Value of BP

The visual trend graphs revealed inflection points at 130 mm Hg for SBP, 80 mm Hg for DBP, and 90 mm Hg for MAP. Beyond these thresholds, the rates of live birth, clinical pregnancy, and ongoing pregnancy declined markedly (Figure 2; Figures S5).

Figure 2.

Rates of reproductive outcomes across prepregnancy systolic and diastolic blood pressure (SBP and DBP) categories. Smoothed curves present the trends in rates of biochemical pregnancy, clinical pregnancy, ongoing pregnancy, live birth (A and C), as well as biochemical pregnancy loss, clinical pregnancy loss, first-trimester loss, second-trimester loss, and total pregnancy loss (B and D) across systolic (A and B) and diastolic (C and D) pressure categories, using a moving average with a window size of 3. The vertical lines indicate the cutoff value of 130 mm Hg for SBP and 80 mm Hg for DBP.

The results of adjusted restricted cubic spline curves showed nonlinear associations between SBP, DBP, and MAP and the reproductive outcomes of IVF (P<0.050 for nonlinearity; Figures S6 through S8). However, the pregnancy loss rate exhibited a continuous upward tendency as SBP, DBP, and MAP increased (Figure 2; Figure S5), and the adjusted restricted cubic spline curves also indicated a linear trend (Figures S6 through S8). The results of regression analysis based on these thresholds were presented in Tables S4 through S6.

Secondary Analyses

The results of stratified analyses by the embryo transfer type or by the developmental stage of transferred embryo(s) were consistent with the main results (Tables S7 through S10). In the secondary analyses restricted to singleton live birth, the results for pregnancy complications remained consistent with the main findings (Tables S11 and S12). The results of secondary analysis restricted to twin live births showed a similar trend (Tables S13 and S14).

Discussion

Main Findings

Compared with normal BP, prepregnancy stage 1 and stage 2 hypertension were associated with a lower rate of live birth after IVF treatment. The lowered rate of live birth was more closely related to the increase in the SBP. Both stage 1 and stage 2 hypertension were associated with a higher risk of pregnancy complications. The optimal cutoff values of SBP and DBP for adverse reproductive outcomes were consistent with the diagnostic threshold for stage 1 hypertension.

Interpretation of Results and Comparison of Findings From Previous Studies

We found a lower rate of live birth in women with stage 1 and stage 2 hypertension compared with women with normal BP, which was more closely related to SBP level. Few studies have evaluated the association of prepregnancy hypertension with reproductive outcomes and reported conflicting results. A retrospective cohort study reported a reduced live birth rate after IVF treatment among women with prepregnancy stage 2 hypertension, with no association observed for DBP.¹⁴ In contrast, another study found a negative association between DBP and live birth rate in fresh embryo transfer cycles.¹⁵ Although SBP and DBP were highly correlated, previous studies suggested that the endothelial dysfunction, which plays an important role in the establishment and maintenance of pregnancy,^17–19 may be more strongly linked to SBP.^14,20

Based on the advantage of the early follow-ups of IVF pregnancies, we observed a dose-dependent relationship between higher BP levels (SBP, DBP, and MAP) and an increased risk of first-trimester pregnancy loss. The findings addressed the gap in the effect of hypertension on early pregnancy that was usually missed in the studies among women with natural conceptions. The underlying mechanisms remain uncertain. However, endothelial dysfunction, inflammation, and oxidative stress, which are generally recognized as important pathophysiological mechanisms of hypertension, have also been implicated in early pregnancy loss.^18,19,21,22

Our findings suggested that both stage 1 and stage 2 hypertension were associated with elevated risks of multiple maternal and neonatal complications compared with normal BP. Consistent with previous studies among women with natural conception,^23–25 we found that women with stage 2 hypertension had higher risks of preterm delivery and small for gestational age compared with those with normal BP. A previous study involving 2239 women undergoing IVF found that stage 1 hypertension was associated with an increased risk of preeclampsia,²⁶ consistent with our findings. In contrast to previous studies in naturally conceived populations,^2,3,27 we did not observe an association between prepregnancy hypertension and the risk of gestational diabetes. This finding should be interpreted with caution, as discrepancies across studies may be explained by differences in sample size, population heterogeneity, and residual confounding.

In addition, we found that the optimal BP cutoffs for reproductive outcomes aligned with the diagnostic threshold for stage 1 hypertension, suggesting that endothelial dysfunction at this BP level may similarly affect multiple organ systems. Previous studies have suggested that stage 1 hypertension may represent the threshold at which endothelial dysfunction accumulated sufficiently to initiate vascular remodeling and subsequently an increased risk of cardiovascular diseases.^28–31 At this threshold, endothelial dysfunction may also reduce uterine perfusion—a proposed mechanism of placental insufficiency—and thereby potentially contribute to an increased risk of pregnancy loss and a decreased likelihood of achieving live birth.^17,32–34 Few studies have identified optimal BP thresholds for reproductive outcomes. A previous study involving 2418 women undergoing fresh embryo transfer reported SBP and DBP cutoffs of 119.5 and 69.5 mm Hg, respectively, for predicting live birth rate using receiver operating characteristic curve analysis.¹⁵ Differences in study populations and analytical approaches may account for the discrepancy in findings.

Strengths and Limitations

One of the strengths of this study was a comprehensive follow-up throughout the entire pregnancy, which allowed us to capture all reproductive outcomes, including first-trimester pregnancy loss and the maternal and neonatal complications. In addition, the large sample size enabled more precise effect estimates, adequate adjustment for potential confounders, and sufficient statistical power to detect the difference in obstetric complications.

This study has several limitations. First, as an observational study, causal inferences cannot be drawn. Second, as a single-center study, the potential of center-specific effects cannot be excluded. Third, despite adjusting for multiple important confounders, residual or unknown confounders may still influence the results. In particular, the absence of information on the use of antihypertensive medication may potentially lead to underestimation of the associations. Some women categorized as having a normal BP level may have been taking antihypertensive therapy and may still have an increased risk of adverse reproductive outcomes,³⁵ potentially attenuating the differences between the hypertension and normal BP groups. Moreover, antihypertensive therapy during pregnancy may improve perinatal outcomes,³⁶ and may, therefore, further dilute the between-group differences. Nonetheless, the observation of significant associations between hypertension and adverse reproductive outcomes, despite the existence of these likely negative biases, further supported our findings.

Conclusions

Compared with normal BP, prepregnancy stage 1 and stage 2 hypertension were associated with a lower rate of live birth after IVF treatment. Both stage 1 and stage 2 hypertension were related to an increased risk of pregnancy complications. The optimal cutoff values for BP were consistent with the diagnostic threshold for stage 1 hypertension.

Perspectives

The 2025 ACC/AHA guideline recommends starting antihypertensive therapy for all patients with stage 2 hypertension and for individuals with stage 1 hypertension who have a high cardiovascular disease risk or failed to achieve BP control through lifestyle modification. These practice-changing recommendations suggest a trend toward earlier detection and more active intervention in hypertension management. Moreover, with the lowered diagnostic threshold for hypertension, an increased number of younger individuals, including women of reproductive age, are identified as hypertensive. Our study addressed the gap between prepregnancy BP and reproductive health. The findings have implications for public health on awareness of the importance of prepregnancy hypertension, since even stage 1 hypertension may adversely affect reproductive outcomes. Whether antihypertensive strategies before pregnancy, particularly for stage 1 hypertension, could improve reproductive outcomes warrants further research.

Article Information

Acknowledgments

The authors acknowledge the contributions of all team members and the generous support from participants and medical staff during data collection.

Sources of Funding

This study was supported by the National Key Research and Development Program of China (2022YFC2703502 and 2023YFC2705502) and the National Natural Science Foundation of China (82495194 and 82421004).

Disclosures

None.

Supplemental Material

Tables S1–S14

Figures S1–S8