1. Introduction
Preeclampsia (PE) is a hypertensive disease that complicates approximately 3-5% of all pregnancies, being one of the leading causes of maternal morbidity and mortality1–10, as well as adverse perinatal and neonatal outcomes11–14. This condition is part of the most important obstetrical syndromes15,16 and multiple etiologies17–23 have been proposed to play a role in its pathophysiology, including an imbalance of angiogenic and anti-angiogenic factors24–40, systemic maternal inflammation 41–43, endothelial dysfunction44,45, metabolic syndrome46–51, vascular disorders of the placenta52–54, abnormal placentation55–58, and utero-placental ischemia59–61. Indeed, PE is characterized by a systemic maternal vascular dysfunction associated with an abnormal placentation that is, in part, attributable to abnormal remodeling of the spiral arteries during early pregnancy62–69. An abnormal placentation is associated with the increased release of cellular debris from the trophoblast into the maternal circulation that contributes to systemic inflammation, endothelial dysfunction, and the clinical manifestation of the disease70–75. The only effective treatment of PE is the delivery of the fetus, thus removing the deleterious effects of the placenta on maternal physiology67,76.
Early identification of women at risk for PE would allow for the development and evaluation of timely and appropriate intervention strategies to limit short- and long-term adverse outcomes77,78. Given that interventions such as aspirin administration during the first trimester of pregnancy, have demonstrated a reduction of its incidence, the development of accurate methods for identifying women at risk of developing PE is a recognized clinical need79–82. Several multiparametric predictive algorithms have been previously reported in literature83,84, and are based on several combinations of maternal risk factors, uterine artery Doppler pulsatility indices, and different blood-borne biomarkers85–87; however, these algorithms have not been universally adopted and accepted for routine obstetric clinical care78,88–94. The development of more accurate, inexpensive and effective risk assessment algorithms may increase the adoption of such testing in clinical care and improve patient management and disease outcome.
Recently, we have identified gingival crevicular fluid (GCF) as a source of surrogate biomarkers of placental function95–97. GCF is a serum exudate that originates in the gingival sulcus as a result of periodontal inflammation and contains a variety of biological cell types and molecular markers of systemic and local origin98. Thus, the determination of the concentration of such biomarkers in GCF may serve as a minimally invasive source of biomarkers for the prediction of placenta-originated diseases99,100. Among those biomarkers, placental alkaline phosphatase (PLAP) has been linked to perinatal diseases such as preterm delivery and PE101,102. PLAP is a membrane-bound glycoprotein expressed by the maternal microvillous membrane of the syncytiotrophoblast103–105. The concentration of PLAP in maternal blood increases throughout gestation in normal pregnancy103, and has been implicated in regulating fetal/maternal metabolism, the transport of nutrients, and placental differentiation106,107. Moreover, in a recent case-control study, we reported a significantly higher concentrations of GCF-PLAP in pregnant women with PE compared to those with a normal pregnancy, even after adjusting for smoking status, body mass index, and periodontal diagnosis99.
The rationale to utilize GCF for the prediction of PE also relies on the association between periodontal disease and development of hypertensive disorders of pregnancy108–111. A recent overview of systematic reviews suggested an association between periodontal disease and PE development (odds ratio [OR] 2.2; 95% confidence interval [CI], 1.4 to 3.4), after the analysis of 15 studies that comprised 5,111 pregnant women112. In addition, periodontal bacteria, such as Porphyromona gingivalis , Fusobacterium nucleatum , and Treponema denticola , have been found to be significantly enriched in the placentae of women affected by hypertensive disorders compared to those of healthy controls113–115, with increased expression of Toll-like receptor 2 in the placentae of patients with PE115; this finding suggests that periodontal bacteremia could stimulate the placental tissue, inducing the systemic release of pro-inflammatory cytokines.
Based on this previous considerations, the aims of the present study are (1) to determine whether GCF-PLAP concentrations are increased during early pregnancy in patients who will subsequently develop PE and (2) to assess the diagnostic performance of GCF-PLAP concentrations when combined with other maternal clinical parameters for the identification of patients who will develop PE.