Monitoring maternal blood pressure variations in pregnancy: gestational age-specific reference percentiles and Z-scores

BackgroundMaternal blood pressure (BP) varies greatly during pregnancy in response to hemodynamic changes, which has led to debate surrounding the use of a single diagnostic threshold. Previous studies have generated "reference ranges" of BP in pregnancy, yet they lack implementation and translation to clinical practice. This study aimed to generate gestational age-specific references and analyze BP Z-scores to further explore cardiovascular dynamics in pregnancy and their potential clinical implications. MethodsRepeated measurements of BP from 2 to 44 gestational weeks were extracted from the obstetric files of 1,875 mothers from the French EDEN cohort. Percentiles of systolic (SBP), diastolic (DBP), and mean arterial pressure (MAP) were modeled using Generalized Additive Models for Location, Scale, and Shape (GAMLSS) as a function of gestational age. They were generated from a "low-risk" reference population defined by reduced cardiovascular risk factors and no hypertensive disorders. For each woman in the overall sample, Z-scores of BP were calculated relative to the "low-risk" percentiles, to assess how BP deviated from the expected trajectory across gestation. BP Z-score trajectories according to hypertensive disorders of pregnancy categories (chronic hypertension, gestational hypertension, and preeclampsia) were then plotted and compared. ResultsA U-shaped trend was observed in overall and "low-risk" percentiles of SBP and MAP, with a nadir around 25 weeks and an increase in the last trimester. In the overall sample, SBPs 95th percentile curve remained below the 140mmHg diagnostic threshold between 15 and 35 weeks of gestation. The Z-score trajectories of women presenting with hypertensive disorders start to diverge as early as the first 5 gestational weeks, well-before their diagnoses (median age: 35-36 weeks). BP evolution differed according to type of hypertensive disorder, for example, a steep increase (>1 standard deviation) from early in pregnancy among preeclamptic women. ConclusionThis percentiles-to-Z-score approach can position individual risk of mothers while considering the natural variation of blood pressure across pregnancy. Our results question the applicability of a non-time-specific threshold to these dynamics. Beyond their potential clinical applications, these references can be used in further research to examine "abnormal" cardiovascular trajectories and their consequences for future maternal and offspring health.