The soluble fms-like tyrosine kinase (sFlt-1)/placental growth factor (PlGF) ratio is a reliable tool in the assessment of preeclampsia. We tested the hypothesis that the sFlt-1/PlGF ratio is able to identify women at risk for imminent delivery. We characterized the sFlt-1/PlGF ratio in different types of hypertensive pregnancy disorders.
We investigated 388 singleton pregnancies with normal pregnancy outcome, 164 with PE, 36 with gestational hypertension, and 42 with chronic hypertension. sFlt-1 and PlGF were measured in serum samples.
Patients with preeclampsia had a significantly increased sFlt-1/PlGF ratio as compared with controls and with patients with chronic and gestational hypertension in <34 weeks and ≥34 weeks ( P < .001). Time to delivery was significantly reduced in women with preeclampsia in the highest quartile of the sFlt-1/PlGF ratio ( P < .001).
The sFlt-1/PlGF ratio allows the identification of women at risk for imminent delivery and is a reliable tool to discriminate between different types of pregnancy-related hypertensive disorders.
Preeclampsia (PE), a multisystem disorder in pregnancy, is still a leading cause of maternal and fetal morbidity and mortality. Occurring with an incidence of 2-8% worldwide, PE represents one of the major contributors of preterm birth, accounting for 15% of all preterm deliveries. PE is defined as the new onset of hypertension (≥140/90 mm Hg) and proteinuria (≥300 mg per 24 hours) after 20 weeks of gestation.
Other hypertensive disorders in pregnancy comprise gestational hypertension (GH) and chronic hypertension (chrHTN). However, these definitions do not take into account the clinical variety of the disease as well as its impact on mother and child. Moreover, the current classification does not consider the fact that the speed of the clinical disease progression can vary dramatically between patients with different subtypes of PE.
Recently several modifications of this simple definition of a complex disease have tried to cope with the clinical diversity of the maternal syndrome. Actual definition proposals try to implement early-onset PE as a severity criterion as well as include a definition of severe hypertension apart from mild hypertension. It is known that the perinatal risk and the maternal complications are the greater, the earlier PE occurs. However, the late preterm birth also carries an increased risk of perinatal morbidity, and maternal complications are also described in late PE.
The reliable identification of high-risk PE patients is crucial because intensified monitoring and referral to specialized perinatal care centers substantially reduce maternal and fetal morbidity. Quick and reliable detection of the disease allows expeditious intervention with steroids for fetal lung maturity, magnesium sulfate for seizure prophylaxis, and antihypertensive therapy.
A multitude of case-control as well as prospective studies have highlighted the role of the measurement of soluble fms-like tyrosine kinase (sFlt-1) and placental growth factor (PlGF) in the peripheral blood of pregnant women as a diagnostic as well as a predictive test for PE. Other laboratory markers for prediction or diagnosis of PE, such as soluble endoglin or free fetal DNA, have been investigated. We have shown recently that an elevated sFlt-1/PlGF ratio below the preliminary cutoff of 85 as assessed by the Elecsys assays allows the determination of PE in the clinical context. However, the putative predictive ability of different levels of the sFlt-1/PlGF ratio in women with clinical PE regarding pregnancy outcome is still unknown.
The aim of the work presented was to characterize the sFlt-1/PlGF ratio in different types of hypertensive pregnancy disorders including early- and late-onset preeclampsia. We tested the hypothesis that the level of the sFlt-1/PlGF ratio is predictive for a risk for imminent delivery and has thereby putative eligibility as a prognostic marker. Furthermore, we performed correlation analysis between the sFlt-1/PlGF ratio and clinical and laboratory markers of PE/HELLP (hemolysis, elevated liver enzymes, and low platelet count) syndrome.
Materials and Methods
Singleton pregnancies were enrolled at 7 European medical centers. An identical study protocol and data collection form were used at each center. The local ethics committees and institutional review boards approved the study, and all subjects gave their written informed consent before participation. A total of 630 individuals were enrolled in the study: 388 singleton pregnancies with normal pregnancy outcome, 164 singleton pregnancies with PE outcome, 36 subjects with GH, and 42 patients with chrHTN ( Table 1 ).
|Variable||<34 wks||≥34 wks||All|
|Visits for analysis a||69||95||164|
|Visits for analysis b||291||176||467|
|Visits for analysis b||11||26||37|
|Visits for analysis b||28||26||54|
a For every visit of the PE/HELLP group, there is 1 paired gestational age–matched control. For the subgroup controls, GH, and chrHTN, only 1 sample of an individual is measured per gestational age window;
PE was defined according to the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Briefly, hypertension was defined as the repeated measurement of systolic blood pressure of 140 mm Hg or greater (Korotkoff I) and diastolic blood pressure of 90 mm Hg or greater (Korotkoff V).
Proteinuria was defined as the excretion of 300 mg protein or greater per day in the 24 hour urine collection or a repeated dipstick of 1+ or greater. PE cases were classified as early onset (<34 weeks of gestation) or late onset (≥34 weeks), Furthermore, PE was defined as severe in contrast to mild, when hypertension was 160/110 mm Hg or greater, proteinuria 5 g or greater per 24 hours, or the occurrence of organ failure (kidney, lung) or neurologic symptoms were observed.
HELLP syndrome, superimposed PE, chrHTN, and intrauterine growth restriction were defined as previously published. In this study, patients with any form of PE or HELLP were combined in the PE/HELLP group (PE/HELLP) for analysis.
We analyzed 2 gestational age windows: 24+0 to 33+6 weeks (<34 weeks) and 34+0 weeks or greater (≥34 weeks). In the PE/HELLP group, the sample obtained after diagnosis was used for allocation to the respective gestational age window and analysis. In case of repeated samples, the last sample before the delivery was analyzed to relate the sFlt-1/PlGF ratio in the sample closest to delivery of that PE patient. Patients of the PE/HELLP group were pairwise matched by gestational age to a healthy control (without GH, chrHTN, or intrauterine growth restriction of the fetus), resulting in the same sample size of PE/HELLP and control group.
Patients who had chrHTN or developed GH in pregnancy were included in the respective groups and appropriate gestational age windows. For the GH and chrHTN groups, only 1 visit was selected randomly per patient and gestational age window. Consequently, every patient contributed at most 1 visit per window. Patients with chrHTN or GH who developed superimposed PE in the later course of pregnancy were analyzed as part of the PE/HELLP group.
For time-to-delivery analysis, the correlation of the sFlt-1/PlGF ratio with the remaining pregnancy duration was performed. According to the standard operating procedures of the respective center, maternal and/or fetal indications for expeditious delivery (<48 hours) before 34 weeks were severe PE plus 1 additional criteria of clinical worsening. Immediate delivery was also considered in severe late-onset PE, whereas in the milder forms, expectant management was recommended until 37 weeks. After 37 weeks of gestation, no prolongation of pregnancy was performed in PE/HELLP patients.
Investigators were blinded to sFlt-1 and PlGF levels, which excluded an influence of this information on decision making and the defining time point of delivery.
Samples and immunoassays
Serum samples were collected according to a common standard operating procedure at each center. Maternal blood was collected by venipuncture in tubes without anticoagulant. After clotting, the samples were centrifuged with 2000 × g and serum was pipetted, and stored at –80°C until testing. The sFlt-1 and PIGF concentrations of each sample were determined in parallel. For each sample sFlt-1/PIGF ratio was calculated. Single measurements were performed for sFlt-1 and PIGF on the fully automated Roche (Penzberg, Germany) Elecsys system (Elecsys PIGF, human PIGF, and Elecsys sFlt-1, sFlt-1) as described previously.
Basic statistics (mean, median, SD, quartiles, and range) were performed for sFlt-1, PIGF, and the sFlt-1/PIGF ratio. To compare clinical subgroups, descriptive statistics (quartiles, mean ± SEM) and according box plots were generated. For the statistical comparison of continuous variables (eg, marker levels in the respective subgroups), time-to-delivery and nonparametric tests (Wilcoxon rank-sum test) were applied.
To explore dependencies between marker values and the time until delivery, survival analysis was applied to the PE/HELLP group. Subgroups within the PE/HELLP group were built by dichotomizing at the third quartile (Q3) of sFlt-1/PlGF. Hazard ratios (adjusted for gestational age) by Cox regression, survival estimates (Kaplan-Meier), and P values of log-rank tests were shown for illustrating the differences of risk between high- and low-level marker groups.
In the PE/HELLP group, relations between laboratory parameters (aspartate aminotransferase [AST], aminotransferase [ALT], platelets) or clinical parameters (systolic blood pressure [BP], diastolic BP) and marker values are reported by Pearson correlation coefficients, if appropriate after transformation of values, and illustrated by scatter plots. All P values are 2 tailed.