- Top of page
Pre-eclampsia (PE) and intrauterine growth restriction (IUGR) are major causes of maternal perinatal morbidity and mortality1. It is increasingly becoming accepted that early-onset and late-onset PE should be regarded as different forms of the disease. Early-onset PE is commonly associated with IUGR, abnormal uterine and umbilical artery Doppler evaluation, and adverse maternal and neonatal outcomes1–4. In contrast, late-onset PE is mostly associated with mild maternal disease and a low rate of fetal involvement, and perinatal outcomes are usually favorable2–4.
Prediction of PE and/or IUGR has been a major clinical and research issue for the past 20 years2. So far, the best test for predicting PE or IUGR at 20–24 weeks of gestation is uterine artery Doppler examination, with reported sensitivities for these conditions ranging from 30 to 80%5. Although most studies have focused on the prediction of PE or IUGR occurring at any time during pregnancy, it seems likely that sensitivity is much higher for early-onset forms5.
Recent evidence suggests that angiogenic factors are involved in the pathogenesis of PE and their assessment could be useful for the early prediction of the disease6–11. Placental growth factor (PlGF) is expressed by trophoblast cells at high levels and is involved in the regulation of placental vascular development12. Circulating soluble fms-like tyrosine kinase 1 (sFlt1) is a potent PlGF antagonist that prevents its interaction with cell receptors. Increased sFlt1 levels and decreased circulating PlGF levels have been demonstrated in PE3, 6, 12, as well as in isolated IUGR3, 13, 14. Differences in the levels of these factors with respect to healthy pregnancies are considerably more pronounced in early-onset cases, whereas in late-onset disease there is substantial overlap with controls3. Several studies have shown that abnormal levels of angiogenic and antiangiogenic factors may be detected 5–14 weeks before the clinical onset of PE and even in the first trimester7–11. One recent study suggested that the increase in antiangiogenic factors occurred earlier in PE that developed before term than in term PE8.
If late-onset PE/IUGR constitutes a heterogeneous condition with minimal or no placental involvement, tests based on the identification of signs of abnormal placentation may continue to be of limited value in predicting this entity. We investigated the existence of differences in the prediction of early- vs. late-onset PE/IUGR by uterine artery Doppler evaluation and measurement of maternal serum PlGF and sFlt1 at 24 weeks' gestation.
- Top of page
This study provides evidence that PlGF measurement may be a useful second-trimester screening test for early-onset PE/IUGR, alone or in combination with sFlt1 or uterine artery Doppler examination. In contrast, the data do not support the use of these parameters to predict late-onset PE/IUGR.
In this study, PlGF showed high sensitivity in predicting early-onset PE/IUGR, which warrants its evaluation in large prospective studies. Uterine artery Doppler and sFlt1 measurements showed a higher degree of overlap with controls and consequently their predictive values were lower, even for early-onset cases. However, the highest accuracy was achieved by the combination of PlGF with either uterine artery Doppler or sFlt1 measurement, suggesting the potential benefit of combining the tests.
The findings of our study are consistent with previously reported data21–23. In a prospective study conducted in 122 high-risk pregnancies with a PE prevalence of 11%, second-trimester PlGF had the highest predictive value for PE whereas uterine artery Doppler was slightly less effective21. In another study, Parra et al. prospectively evaluated PlGF and sFlt1 together with other biochemical markers in the first and second trimester in 170 low-risk pregnancies22. Second-trimester PlGF levels were independently associated with PE, although PlGF did not improve the predictive value of uterine artery Doppler results alone. Savvidou et al. evaluated sFlt1 but not PlGF at 23–25 weeks. In agreement with our results, these authors found considerable overlap with controls and a relatively poor predictive value23. However, although the sample size was too small to allow definitive conclusions to be drawn, their findings suggest higher levels in early-onset PE. Finally, Muller et al. found no correlation between uterine artery Doppler velocimetry and circulating angiogenic factor levels, and suggested a potential improvement in the prediction of PE/IUGR by combining both markers24.
Prediction of late-onset PE/IUGR was very poor for all the parameters evaluated. This finding may be partially explained by a longer period between the time of screening and disease onset compared with early-onset cases7–9. However, we have previously reported that, after clinical onset, late-onset PE and IUGR are associated with milder changes in PlGF and sFlt1 levels, and considerable overlap with controls compared with early-onset cases3. Moreover, virtually all cases of early-onset PE/IUGR have abnormal uterine artery Doppler findings, compared with less than half of late-onset patients3. Finally, early-onset PE is consistently associated with abnormal placental morphology, whereas placentas from late-onset cases are morphologically similar to those from healthy pregnancies25.
The rate of late-onset PE in our population was low in comparison to that in some other countries. This may be explained by the existence of different risk factors for cardiovascular disease in south European countries, such as a lower rate of black ethnicity26, lower body mass index27, lower rate of atherogenic lipid profile, and higher vitamin C and E diet intake28, compared with other populations with a higher prevalence of PE2, 29, 30. In addition, although very unlikely, an extra ultrasound examination in women with pregnancies with abnormal uterine artery Doppler findings could result in earlier diagnosis and delivery of fetuses with IUGR thereby preventing clinical onset of PE. However, in spite of many theories, the precise reasons for differences in cardiovascular disease rate in different countries have been extremely difficult to identity. Therefore, it is difficult establish the exact reasons for the differences in the prevalence of late-onset PE in our population.
This study supports the notion that early-onset PE and IUGR are placenta-mediated diseases that share important similarities. Changes in second-trimester levels of angiogenic factors and in uterine artery Doppler PI values were similar in both conditions. These results are in line with changes in maternal plasma angiogenic (PlGF) and antiangiogenic (sFlt1) protein levels occurring after clinical onset of both conditions3, 13, and with placental morphometric studies showing that placentas from pregnancies complicated by IUGR and PE share similar structural and cellular abnormalities4, 31. The question of why placental disease is associated with IUGR alone in some cases, and with maternal hypertension in others, remains to be elucidated.
In conclusion, this study shows that measurement of maternal serum angiogenic factors and uterine artery Doppler examination might constitute useful screening tools for early PE/IUGR, but not for late-onset cases. However, these results must be interpreted with caution as this is an exploratory study with inherent limitations owing to the small sample size. Therefore, larger prospective studies are warranted to further assess whether the combined use of these tests could improve the early detection of PE and IUGR.