Interpretation
This is the first study to evaluate the potential of calcium-based antacids to reduce the risk of preeclampsia. Several mechanisms through which calcium intake could prevent preeclampsia have been proposed. Calcium may protect endothelial cells from endothelial activation by multiple activators, including necrotic trophoblast debris and inflammatory cytokines, and prevent hypertension by increasing the production of nitric oxide, which is a potent vasodilator.38 Additionally, low serum calcium could increase blood pressure by inducing parathyroid release, which stimulates renin release, resulting in increased intracellular calcium leading to vasoconstriction.39,40 Indeed, calcium supplementation has been shown to substantially decrease the risk of preeclampsia based on data from randomized controlled trials, although publication bias cannot be excluded.6 Only a few observational studies on the association between calcium supplementation and hypertensive disorders of pregnancy have been conducted, indicating that beneficial effects may only apply to women with a diet low in calcium or observing no effect at all.41-43 In exploratory analyses in the current study, we did not observe a decreased risk of late-onset preeclampsia associated with calcium-based antacid use among women at high risk of developing preeclampsia or among women who did not use calcium supplementation either.
In line with previous studies,11-13 we did not find the risk of late-onset preeclampsia to be reduced after use of PPIs, neither in early or mid-pregnancy or in late pregnancy. The latter, however, is in contrast with the results from Hastie et al.,11 who observed a reduced risk of preterm preeclampsia (delivery <37 weeks of gestation; adjusted odds ratio 0.63 [95% CI 0.41-0.96]) and early preeclampsia (delivery <34 weeks of gestation; adjusted odds ratio 0.41 [95% CI 0.20-0.82]) associated with PPI use after gestational week 28. In addition to a different outcome definition applied in our study (i.e. late-onset preeclampsia: any preeclampsia diagnosed after 33+6 weeks of gestation), this difference in results could be explained by the analytical approach. The extended Cox model applied in our study accounts for PPI use to be a function of time, whereas the standard time-independent model used by Hastie et al. may be biased by immortal time,27,44,45 leading to overestimation of the protective effects of PPI use.
Mechanistically, PPIs may be more likely to prevent preterm preeclampsia than term preeclampsia, in which the relative difference in sFlt-1 levels between those affected and those who are not is smaller than in preterm preeclampsia.46 Therefore, PPI use may prevent the development of early-onset preeclampsia in particular. We excluded cases with this subtype, however, due to etiologic and prognostic heterogeneity,18 as small numbers prevented us from including these cases as a separate outcome group.
Instead of the hypothesized decreased risk, we actually observed an increased risk of late-onset preeclampsia associated with use of low-dose calcium-based antacids in gestational weeks 0-16 or use of PPIs in gestational weeks 17-33, in particular among women at high risk of developing preeclampsia. We did not identify other studies with similar results on calcium in early pregnancy, but slightly increased risks of overall preeclampsia associated with PPI use in pregnancy were also observed in other studies.11-13 However, these results may be biased by confounding by indication, as illustrated by the attenuated RR in comparisons with an H2RA-exposed group in a nationwide cohort study in South Korea.13Furthermore, reverse causation cannot be excluded, since epigastric pain is one of the symptoms consistent with preeclampsia.47