Trajectory of blood pressure change during pregnancy and the role of pre-gravid blood pressure: a functional data analysis approach

Abstract

The study aims to examine the blood pressure (BP) trajectory during pregnancy and its association with pre-gravid BP level. In a pre-conception cohort study, newly-married women in Liuyang, China underwent pre-gravid measurements and were followed throughout the pregnancy. BP was measured at pre-conception and again throughout pregnancy. The functional principal component analysis was used to examine the trajectory of BP changes during pregnancy. A total of 1282 women with a singleton pregnancy who had both pre-conception and gestational BP measurements performed were included in the final analysis. The results showed that BP decreased significantly in early pregnancy and increased thereafter, without BP drop around 20 weeks of gestation. Pre-gravid BP level was inversely associated with the BP drop in early pregnancy, such that women with higher pre-gravid BP had greater BP drop at the beginning, while women with the lowest pre-gravid BP level demonstrated no obvious BP drop throughout the entire pregnancy.

Figure 1

Distribution of the timing of measurements and number of antenatal visits.

Figure 2

Estimated trajectories of blood pressure change from pre-gravid level and the eigenfunctions. Top panels showed the trajectories of per cent blood pressure change from the pre-gravid level, estimated by functional principal component analysis. Bottom panels showed the first three or four eigenfunctions. Order of eigenfunctions and the percentage of the variation explained by each eigenfunction were expressed as φ_i (%).

Figure 3

Estimated trajectories of blood pressure change from pre-gravid level, stratified by the pre-gravid mean arterial pressure tertile. Stratifications were created according to the pre-gravid mean arterial pressure tertiles. Different colors signified different tertiles. Women in the highest pre-gravid MAP tertile had the largest BP drop at the beginning of pregnancy, while women in the lowest MAP tertile showed no BP drop at the beginning. Women in the highest tertile showed DBP drop around 15 weeks of gestation. Women in the lowest tertile showed some BP fluctuations at the before 15 weeks of gestation.

Figure 4

Estimated trajectories of blood pressure change from pre-gravid level, stratified by the number of antenatal visit. The sample was divided by the median of antenatal visits. Subgroup analysis showed that both subgroups had similar trajectories and trends, partly overlapped, although fluctuations were present. No major selection bias existed.