It has been proposed that placental angiogenesis is defective in preeclampsia, as evidenced by failure of the cytotrophoblast to convert from an epithelial to an endothelial phenotype (referred to as pseudovasculogenesis) thus causing variable degree of invasion of the maternal spiral arteries [3]. Furthermore, placenta isolated from women with early onset preeclampsia was associated with abnormal morphology compared to those with late onset preeclampsia [4] which suggests that placental ischemia is an early event relating to maternal endothelial dysfunction [5]. Recently serum soluble Fms-like tyrosine kinase 1 (sFlt1) expression has been noted to be elevated in preeclampsia [4]. sFlt-1 acts as an antagonist to pro-angiogenic factors Vascular Endothelial Growth Factor (VEGF) and Placental Growth Factor (PlGF) as it binds both VEGF and PlGF, thus reducing free circulating levels of these factors [5]. Maynard et al. (2003) demonstrated that excess sFlt1 in patients with preeclampsia causes endothelial dysfunction and produces a syndrome of nephrotic range characterized by proteinuria, hypertension and glomerular endotheliosis when administered in rats [5]. The effects circulating factors like sFlt1 and PlGF exert on placental structural changes and vasculopathies remains unclear although raised levels of these biomarkers have frequently been associated with the presence of the placenta.

We observed lower PlGF but higher sFlt-1 levels in GH women in comparison to normotensive during antepartum and intrapartum, with correlation analysis confirming an inverse relationship. Salahudin et al., (2007) reported similar findings in that sFlt-1 levels in GH women were significantly higher than normotensive women in a prevalence case control study done in the United States [11]. Findings by Maynard et al., (2003) and Levine et al., (2005) both observed reduction in PlGF and an increase in sFlt-1 in preeclamptic women [5,12]. While the latter studies proposed this for preeclampsia, this study proves these biomarkers are also significantly altered in GH. Furthermore, the inverse relationship between sFlt-1 and PlGF observed in our study is representative of findings from an ethnically diverse population. We acknowledge that results from this study present a contrasting finding compared to a previous cross sectional study. In the latter study, while no absolute values of the biomarkers were discussed, analysis for sFlt1:PlGF ratios demonstrated no significant differences between GH and normotensive mothers except at 33 through 36 weeks of pregnancy [13]. This contrast could be due to variations in the study population (for example: age, parity, ethnicity, smoking status) or on the actual onset of GH versus the diagnosis of GH.

PlGF levels in GH and normotensive women during the postpartum period were similar and this can be attributed to the fact that PlGF is primarily expressed by the placenta which is no longer present at postpartum. Our study showed that sFlt-1 levels remain significantly higher in GH women compared to normotensive women during postpartum period. This finding raises important questions on the effect of persistently raised postpartum levels of sFlt-1, may have on the period of surveillance and continuation of anti-hypertensive drugs. Cohort studies that have used population-based pregnancy databases consistently identify a clinically significant association of GH and preeclampsia with later hypertensive disorders [14]. This clearly defines a need for long term follow-up for the development of essential hypertension in GH women. A recent study by Berks et al., 2009, on preeclamptic women in the Netherlands indicates that it can take up to 2 years for hypertension and proteinuria to resolve [15]. Findings from these studies suggest that it may be necessary to further evaluate the effect prolonged exposure to high levels of sFlt-1 may have and it's subsequent clinical utility in the management of GH cases.

The findings for villous capillarization and intervillous space were not significantly different among the two groups investigated (normotensive and GH). Our morphological findings were consistent with that of Mayhew et al., (2004) who reported that in cases of preeclampsia not accompanied with intrauterine growth restriction, placental morphometry is similar to that of control (normotensive) group [16]. It is hypothesized that, in GH cases there is increased anti-angiogenic property thus causing defective angiogenesis, in turn leading to decreased villous capillarization [5,12]. However, increased villous capillarization can also occur in GH women as a form of a compensatory mechanism. Hypoxic experiments with placental tissue have previously demonstrated an increase in vascular endothelial growth factor expressions [17]. These angiogenic factors may thus bind to receptors to exert their effect and promote angiogenesis by increasing capillarization explaining the lack of morphometric differences between normotensive and GH as seen in our study.

Dr Vishna Devi Nadarajah BSc (Hons) (Biochem), PhD

Prof Dr John Paul Judson MBBS, DHA

Prof Dr Sivalingam Nalliah MBBS, MRCOG, FRCOG

Dr Farouk Abdullah MBBS, MRCOG, FRCOG