PlGF, sFlt-1 and sFlt-1/PlGF ratio: Promising markers for predicting pre-eclampsia

Author(s): Ola H El-Demerdash, Mona M Zaki, Mohamed O El Maraghy, Doaa M A Elzoghby, Marwa A Abdel-Wahed* and Ahmed M Mamdouh Objective: To evaluate placental growth factor (PlGF) and soluble fms-like tyrosine kinase-1 (sFlt-1) as early predictors of pre-eclampsia. Methods: Cohort study at Department of Obstetrics and Gynecology, Ain Shams University Hospitals, enrolling eighty pregnant women between 14 and 19 weeks’ gestation, sera were collected, and stored at -80°C. Thirty women develop ... Abstract View Full Article View DOI: 10.17352/jgro.000052

Inclusion criteria were singleton normotensive pregnant women during their second trimester (between 14 and 19 weeks' gestation) and maternal age more than 18 years. All included subjects were Egyptian. Exclusion criteria were renal diseases, systemic lupus erythematosus (or any other autoimmune disease), chronic corticosteroid therapy, women with multiple pregnancies, major fetal congenital anomalies or chromosomal abnormalities.
All pregnant women included in this study were subjected to full history taking and thorough clinical examination with special emphasis on edema and blood pressure measurement, which was taken in a semirecumbent position, with a supported arm and appropriately sized cuff using a manual sphygmomanometer.
First blood samples were collected on enrollment centrifuged at 1000 x g for 15 minutes after complete blood clotting and aliquots were stored at -80°C. Second blood samples were collected on admission during the third trimester. Laboratory investigations done in the second trimester for exclusion of any clinical abnormality included complete blood count (CBC), prothrombin time (PT), international normalized ratio (INR), random serum glucose (RSG), renal function tests (BUN, creatinine), liver function tests including AST and ALT, and urine protein testing by dipstick in a random urine sample. These tests were repeated in the third trimester to assess the clinical severity of pre-eclampsia.
The analysis of CBC was done using Coulter LH 750 Cell Counter, measurement of PT was performed using Stago Compact Autoanalyzer using Neoplastine CI Plus supplied by Diagnostica Stago and serum chemistry was performed using Beckman Coulter AU480 Autoanalyzer. Serum sFlt-1 and PlGF were quantifi ed by chemiluminescence immunoassay technique using reagents provided by EIAAB Science Co., Gaoxin Road, Wuhan, China. The used analyzer reader for chemiluminescence absorbance was Promega E7031 Glomax.

Sample size calculation
Based on the published data of 88% average diagnostic sensitivity and 82% average diagnostic specifi city for sFlt-1, PlGF and sFlt-1/PIGF ratio [9]. A minimum total sample size of 188 participants to obtain a minimum of 15 pre-eclamptic patients had achieved a power of 90%. The assumed prevalence of the disease was 0.08 and the target level of signifi cance was 0.05 [10]. The calculation was done using PASS program version 15.

Statistical analysis
Statistical analysis was done using SPSS software version (V. 23.0, IBM Corp., USA, 2015). The values for the biochemical markers were expressed as mean and standard deviation in case of parametric data and as median and interquartile ranges in case of skewed data, while categorical variables were summarized using frequency measures. Student´s t test, Wilcoxon's rank sum test (Mann-Whitney U) and Chi-square test were used for comparative analysis.
Receiver operator characteristic (ROC) curves were constructed and optimal cut-off values for the biomarkers were established by the best sensitivity and specifi city. Multi-ROC curve analysis was used to assess the diagnostic performance of the combined use of more than one test, where the right angle at the upper left corner is the best diagnostic threshold (cut-off) of the parameter being varied.
Odds ratio measures the association between an exposure and an outcome and how many times the risk was present among diseased individuals compared to that among nondiseased ones. In all statistical analyses, P<0.05 was considered signifi cant.

Results
The descriptive and comparative statistics of the demographic data are shown in tables 1-3.
The diagnostic utility of serum sFlt-1, PlGF and sFlt-1/PlGF ratio in discriminating pre-eclamptic from healthy women during the second and third trimesters using ROC curve analysis are shown in tables 4,5 and fi gures 1, 2.
Multi-ROC curve analysis of the combined use of PlGF and sFlt-1/PlGF ratio during the second trimester revealed that the best cut-off value of PlGF was 328 pg/mL while that of the sFlt-1/PlGF ratio was 40. At these cut-off values, the diagnostic sensitivity, diagnostic specifi city, positive predictive value (PPV), negative predictive value (NPV) and diagnostic effi ciency were 100%, respectively ( Figure 1).
Multi-ROC curve analysis of the combined use of sFlt-1 and sFlt-1/PlGF ratio during the third trimester revealed that the best cut-off value of sFlt-1 was 4,840 pg/mL with a sFlt-1/PlGF ratio of 51. At these cut-off values the diagnostic sensitivity, diagnostic specifi city, PPV, NPV and diagnostic effi ciency were 100%, respectively ( Figure 2).
In an attempt to determine the Odds ratio for the occurrence of pre-eclampsia and not just to discriminate between preeclamptic and healthy women, all pregnant women included in our study during the second trimester were distributed into four quartiles according to their serum levels of sFlt-1, PIGF and sFlt-1/PlGF ratio. Table 2 Table 6 demonstrates the Odds ratio for the possible occurrence of pre-eclampsia during the second trimester. It revealed that pregnant women in "Q4" had a 5.46 higher risk  for the development of pre-eclampsia than pregnant women in "Q3". Pregnant women in "Q3" had a 4.05 higher risk for the development of pre-eclampsia than pregnant women in "Q2". Finally, there was no signifi cant difference as regards the Odds risk between "Q2" and "Q1".

Discussion
The recent screening studies for pre-eclampsia using sFlt-1 and PlGF were performed during the third trimester [11][12][13]. Meanwhile, conducting our study in the second trimester might be an additive privilege, which enlights the main value of this current study in early prediction of the disease and hence early follow up and proper management could be done.
Our study revealed signifi cantly higher serum levels of sFlt-1 and sFlt-1/PIGF ratio in pre-eclamptic women in early second trimester as compared to the normotensive pregnant women, these being associated with signifi cantly lower PlGF levels. This is in agreement with the fi ndings of a prospective longitudinal study carried out from 15 weeks' gestation onward [14]. Similarly, another prospective study measured serum sFlt-1 and PlGF levels at 10, 18, 26 and 35 weeks of gestation [15]. Both confi rmed that increased serum sFlt-1 levels and decreased serum PlGF levels, early in pregnancy, are associated with a higher risk of pre-eclampsia.
The connection between circulating sFlt-1 and PlGF in preeclampsia was explored by previous studies suggesting that upregulation of the anti-angiogenic sFlt-1 in pre-eclampsia leads to binding of the angiogenic factor PlGF, with a consequent decrease in circulating free PlGF levels [6,16].
Our study revealed a persistent highly signifi cant increase in serum sFlt-1 levels with lowering of PlGF levels and raised sFlt-1/PIGF ratio among the studied pre-eclamptic patients during the third trimester of pregnancy. This highlights the potential role of sFlt-1, PlGF and sFlt-1/PIGF ratio as promising biomarkers for diagnosis of pre-eclampsia. Similar results recorded that pre-eclamptic women had higher serum levels of both sFlt-1 and sFlt-1/PlGF ratio and lower serum levels of PlGF compared to the values found in control women [17]. In this respect, sFlt-1 and PlGF seem to be important etiological factors in the pathogenesis of pre-eclampsia, as the elevated serum levels of sFlt-1 modify endothelial integrity of blood vessels, hence causing hepatic edema as well as the hypertension and proteinuria encountered in pre-eclamptic patients. Also, blood-brain-barrier damage may occur leading     to brain edema. This aberrant angiogenesis and hypertension are the hallmarks of pre-eclampsia [16].
In context with the scenario of multi-system affection in preeclampsia, our current study revealed statistically signifi cantly higher serum levels of AST, ALT, BUN, and creatinine in preeclamptic patients compared to controls during the third trimester. It was postulated that elevated serum transaminases might be attributed to the profound systemic vasoconstriction with generalized damage to the endothelium of the liver, which likely occurs following the release of toxic factors from the diseased placenta, namely von Willebrand antigen, cellular fi bronectin, soluble tissue factor, platelet-derived growth factor and endothelin [18,19]. Moreover, the elevated serum levels of sFlt1 modify the endothelial integrity of blood vessels, causing hepatic edema which may also contribute to the elevation of hepatic transaminases. Concerning the elevated BUN and creatinine levels, this could be explained by the damage infl icted on the kidneys by excess of anti-angiogenic factors such as sFlt-1 over angiogenic mediators such as VEGF and PlGF in the systemic circulation [16]. A parallel imbalance in the renal circulation would be expected, thus affecting renal function [18,19].
Our second trimester results revealed that sFlt-1 at a cut-off level of 2,330 pg/mL had a relatively low diagnostic sensitivity of 66.7% and 64% diagnostic specifi city in discriminating pre-eclamptic subjects from the healthy pregnant women.
Meanwhile, PlGF showed a better diagnostic performance at a cut-off level of 92 pg/mL, with a diagnostic sensitivity of 83.3%, diagnostic specifi city 84%. These fi ndings are comparable with the results of a previous study showing that serum sFlt-1 at a cut-off value of 3,198 pg/mL could discriminate pre-eclamptic women from healthy pregnant women with 88% diagnostic sensitivity and 83.6% diagnostic specifi city. At a cut-off value of 138 pg/mL, PlGF could predict pre-eclampsia with a diagnostic sensitivity of 85.5% and a diagnostic specifi city of 77.2% [9].
The sFlt-1/PlGF ratio was the best tool for discriminating pre-eclamptic women from the healthy pregnant women early in their second trimester in our study. Our ratio's best cutoff (32.6) had 93.3% diagnostic sensitivity and 84% diagnostic specifi city. This is much close to the reported cut-off value of sFlt-1/PIGF ratio at 38.46 that predicted pre-eclampsia with a diagnostic sensitivity and specifi city of 88.5%, respectively, during the second trimester (24-28 weeks of gestation) [6].  [20].
Concerning the diagnostic performance of sFlt-1/PlGF ratio during the third trimester, our best cutoff value of 51 was more valuable in discriminating the pre-eclamptic from healthy women with a diagnostic sensitivity of 96.7% and a diagnostic specifi city of 94%. Our results agree with an earlier study, suggesting that sFlt-1/PIGF ratio at a cut-off value 45 could diagnose pre-eclampsia during the third trimester with a diagnostic sensitivity of 97% and a diagnostic specifi city of 95% [21].
Our results were also compared to a recent study which suggested two cut-off values for sFlt-1/PlGF ratio in the diagnosis of pre-eclampsia at 20-37 weeks of gestation; the fi rst cut-off value focusing on high diagnostic sensitivity at 20-33 weeks of gestation and the second one focusing on high diagnostic specifi city at 34 weeks of gestation till delivery. These cut-offs were ≤33 and ≥85 with yielded diagnostic sensitivity/ specifi city of 95%/94% and 88%/99.5%, respectively [22,23].
In our study, the combined use of sFlt-1/PlGF ratio at a cut-off value 40 together with PlGF at a cutoff value of 328 pg/mL during the second trimester improved the diagnostic sensitivity, specifi city, PPV, NPV and diagnostic effi ciency to reach 100%, respectively in multi-ROC curve analysis performed in pre-eclamptic patients versus healthy controls. Similarly, the combined use of sFlt1/PlGF ratio at a cutoff value of 51 and sFlt-1 at a cut-off value of 4,840 pg/mL during the third trimester also made the same achievement in discriminating pre-eclamptic patients from healthy controls. Our results support the conclusion reporting that combinations of markers generally led to an increase in sensitivity and/or specifi city compared with single markers [24].
The Odds ratio for the occurrence of pre-eclampsia was 5.4 in Q4 versus Q3 women and 4.05 higher in Q3 versus Q2 women. Women in Q2 and Q1 had similar risk for developing pre-eclampsia.
The Odds ratio of sFlt-1 and PlGF was evaluated separately in a previous study, assuming that the Odds ratio for developing pre-eclampsia increased progressively among pregnant women in the quartiles with high serum levels of sFlt-1 and sFlt-1/ PlGF ratio or with low serum levels of PlGF. The researchers highlighted that the prediction of pre-eclampsia using standard diagnostic criteria as hypertension and proteinuria may misdiagnose a signifi cant number of cases and they added that women whose rate of change in sFlt-1 and PlGF was in the most abnormal quartile had greater Odds of developing preeclampsia [15,25].
In conclusion, serum sFlt-1, PlGF and sFlt-1/PlGF ratio can be considered promising biomarkers for prediction of pr-eclampsia. The excellent diagnostic performance (100% diagnostic effi ciency) of the combined use of PlGF and sFlt- 1/PlGF ratio using their multi-ROC cut-offs (328 pg/mL and a ratio of 40, respectively) calls for their inclusion in the laboratory work-up of pregnant females, especially high-risk ones, for early prediction of the disease during the second trimester. Meanwhile, the combined use of sFlt-1 and sFlt-1/PlGF ratio at cut-offs 4,840 pg/mL and 51, respectively, is 100% diagnostic of the disease during the third trimester.