A prospective study published in Hypertension Research evaluates whether cumulative home blood pressure load provides superior prediction of low birth weight (LBW) compared with average home blood pressure (HBP) readings. The research suggests that monitoring both the height and duration of elevated readings early in pregnancy may improve risk stratification for LBW.
Background
LBW affects about one in 15 newborns and is associated with heightened lifetime risks of hypertension, diabetes, cardiovascular disease, and chronic kidney disease. Many drivers of LBW are potentially modifiable during the first trimester, making early detection crucial. Blood pressure naturally fluctuates during pregnancy, with dips in early gestation followed by rises later on. Consequently, simple averages may miss risk signals.
Home blood pressure monitoring captures repeated readings, while cumulative blood pressure load combines how high readings are with how long they exceed a threshold—concepts linked to organ damage in adults. Notably, some Asian populations display lower early-pregnancy pressures yet higher LBW incidence, underscoring the need for pregnancy-specific targets and further research to determine whether cumulative load enhances risk stratification.
Study design and methods
This analysis used data from the BOSHI study (Babies and Their Parents’ Longitudinal Observation in Suzuki Memorial Hospital on the Intrauterine Period) conducted in Sendai, Japan. Eligible participants were pregnant women with singleton gestations who recorded at least four HBP measurements between 10 weeks 0 days and 15 weeks 6 days.
Home BP measurements were taken each morning after micturition and a brief rest. Daily systolic BP (SBP) and diastolic BP (DBP) values were linked to compute a cumulative load by calculating the percentage above a threshold set at the cohort’s median average BP (thresholds derived after excluding participants with chronic hypertension). Separate analyses were conducted for SBP and DBP. An elevation was defined as above 20% of the threshold, with 10% and 30% also tested in sensitivity analyses.
The primary outcome was LBW, defined as birth weight <2,500 grams. Poisson regression estimated risk ratios (RRs), adjusting for maternal age, body mass index (BMI), smoking, primiparity, and history of hypertensive disorders of pregnancy (HDP). Additional sensitivity analyses varied elevation cutoffs and restricted cohorts (e.g., excluding chronic hypertension, requiring eight or more measurements). Model comparison used the Quasi-likelihood under the Independence model Criterion (QIC). Analyses employed R and SAS, with multiple imputation for missing data.
Results
Among 729 women (mean age 31.2 years; about half primiparous), the median birth weight was 3,060 grams and LBW occurred in 47 infants (6.5%). Median cumulative loads were 49% for SBP and 51% for DBP; median home SBP/DBP averages were 103.9/61.8 mmHg. Cumulative load correlated strongly with average BP (Spearman’s rho 0.964 for SBP; 0.956 for DBP).
Key associations
Isolated elevated cumulative SBP load, even when average SBP was normal, associated with higher LBW risk (RR 2.86; 95% CI, 1.33–6.17).
High average SBP showed a similar pattern (RR 3.57; 95% CI, 1.38–9.24).
For DBP, isolated elevated cumulative-load (RR 2.22; 95% CI, 1.08–4.58) and high average DBP (RR 3.35; 95% CI, 1.08–10.34) were linked to greater LBW risk.
Nonlinear patterns emerged
LBW risk rose as cumulative SBP load exceeded roughly 20%, peaking near 40%. DBP-related risk increased up to about 40% before plateauing. Models based on cumulative load provided a better fit (lower QIC) than models based solely on average BP across classifications.
Sensitivity analyses
When data were stratified to match primary sample-size distributions, the significant rise in LBW risk appeared mainly in the high-average SBP category. In tertile analyses, SBP-related risk rose in the second and third tertiles relative to the first, while DBP-related risk rose in the highest tertile. Varying elevation thresholds to 10% or 30% preserved associations. Excluding women with chronic hypertension or including only those with eight or more measurements yielded similar effect sizes with improved model fit.
Clinical interpretation
No LBW cases occurred in the subset with cumulative DBP load below 100% or cumulative SBP load below 90%, suggesting that extremely low cumulative loads align with lower risk. The best-fitting thresholds observed were around 20% for SBP and 30% for DBP, supporting a prespecified SBP cutoff and a potentially optimal DBP cutoff near 30%. Clinically, brief clusters of morning readings above a threshold may signal risk not fully captured by mean values, indicating that early, consistent home monitoring can be a practical tool for clinicians and patients.
Conclusions
Elevated cumulative BP load in early pregnancy was associated with LBW even when mean home BP remained within guideline-normal ranges. Counting days above a threshold may refine risk stratification beyond averages alone and could be integrated into digital hypertension programs for pregnancy. The findings support obtaining home BP readings in the first trimester and exploring mechanisms such as placental growth factor, soluble endoglin, and tyrosine kinase-1. Validation in diverse populations and trials testing threshold-based feedback are needed to translate these insights into routine practice and to improve maternal and neonatal outcomes.
Related topics
