Background
Alterations in circulating angiogenic factors are associated with the diagnosis of preeclampsia and correlate with adverse perinatal outcomes during the third trimester.
Objective
Analysis of the sequential levels of plasma angiogenic factors among patients admitted for evaluation of preeclampsia.
Study Design
We performed an observational study among women with singleton pregnancies admitted to Beth Israel Deaconess Medical Center, Boston, Massachusetts, for evaluation of preeclampsia at less than 37 weeks of gestation. Plasma samples were collected on admission and daily for the first 3 days and then weekly until delivery. Doppler ultrasound was performed on admission (within 48 hours) and then weekly (within 24 hours of blood collection) to evaluate uteroplacental and umbilical blood flows. Maternal demographics, hospital course, mode of delivery, diagnosis of hypertensive disorder, adverse maternal outcomes (elevated liver function enzymes, low platelet count, pulmonary edema, cerebral hemorrhage, convulsion, acute renal insufficiency, or maternal death), and adverse fetal/neonatal outcomes (small for gestational age, abnormal umbilical artery Doppler, fetal death, and neonatal death) were recorded. Circulating angiogenic factors (soluble fms-like tyrosine kinase and placental growth factor were measured on automated platform in a single batch after delivery and in a blinded fashion. Data are presented as median (25th to 75th centile), mean, or proportions as appropriate.
Results
During the study period, data from 100 women were analyzed for the study, and 43 had adverse outcomes. Women with adverse outcomes had lower gestational age of delivery, higher systolic and diastolic blood pressures during hospitalization, and lower birthweight and placental weight (all P < .01). These patients had higher soluble fms-like tyrosine kinase and soluble fms-like tyrosine kinase/placental growth factor ratio on admission and continued to have an increase in levels throughout hospital course. The median (25th to 75th) soluble fms-like tyrosine kinase/placental growth factor ratio among patients with adverse outcomes was 205.9 (72.5, 453.1) versus 47.5 (9.7, 87.0) among women without adverse outcomes ( P < .001). The median (25th to 75th) absolute change per day in soluble fms-like tyrosine kinase levels (pg/mL) was 491.0 pg/mL (120.3, 1587.2) among women with adverse outcomes versus 81.3 pg/mL (−177.9, 449.0) among women without adverse outcomes ( P = .01). Similarly the absolute change per day for soluble fms-like tyrosine kinase/placental growth factor ratio was 15.1 (1.8, 58.1) versus 2.7 (−0.6, 8.3) between the two groups ( P = .004). The mean (range) days from admission to delivery was 6 (0–35) among subjects with soluble fms-like tyrosine kinase/placental growth factor ratio ≥85 and 14 (0–39) below a ratio of 85 ( P < .001). The positive predictive value for plasma soluble fms-like tyrosine kinase/placental growth factor ratio ≥85 at admission for indicated delivery within 2 weeks was 91% (83–99%). Admission plasma soluble fms-like tyrosine kinase/placental growth factor ratio positively correlated with pre-delivery uterine artery resistive index ( r = 0.35; P = .004).
Conclusion
Among women admitted for evaluation of preeclampsia, women at risk for adverse pregnancy outcomes have higher soluble fms-like tyrosine kinase/placental growth factor ratio on admission, which continued to rise until delivery. Women with high soluble fms-like tyrosine kinase/placental growth factor ratios delivered sooner than women with low soluble fms-like tyrosine kinase/placental growth factor levels. These data support the hypothesis that targeting angiogenic imbalance in preeclampsia may lead to prolongation of pregnancy.
Preeclampsia (PE) affects 3–10% of pregnancies and is responsible for significant maternal morbidity and mortality worldwide. It is currently believed that abnormal trophoblastic invasion of the maternal spiral arteries is the primary placental pathology that in turn leads to maternal endothelial dysfunction resulting in systemic manifestations of disease. PE can present with a range of signs including only hypertension and proteinuria to severe maternal outcomes such as hemolysis, renal failure, abruption, pulmonary edema, and rarely, maternal death. PE is also associated with fetal adverse outcomes such as severe growth restriction and fetal death. Because of this array of clinical presentations, PE is considered to be heterogeneous disease with different pathophysiologic mechanisms explaining its different clinical presentations. However, recent studies have revealed that PE-related adverse outcomes are largely mediated through dysregulation of angiogenic factors. Studies have shown that women having adverse outcomes related to hypertension especially at preterm gestational ages have high levels of circulating antiangiogenic protein (sFlt1) and low levels of proangiogenic protein (placental growth factor, PlGF). Studies have also shown that among women with suspected preeclampsia, triage levels of angiogenic factors are strongly correlated with duration of pregnancy. Circulating levels of angiogenic factors added to ultrasound parameters have also been shown to predict severe intrauterine fetal growth restriction, a disorder that shares similar placental pathological features with preeclampsia.
While a number of studies exists evaluating the measurement of circulating angiogenic factors in third trimester or triage setting, there is paucity of data about the concentration of these proteins among women with established preeclampsia. In a small, retrospective study of 34 patients with PE, Schaarschmidt et al found that sFlt1 concentration and consequently the sFlt1/PlGF ratio increased from admission to delivery in all cases of PE and HELLP syndrome. Such data will be important given the recent studies showing that reduction of sFlt1 by an external apheresis device resulted in prolongation of pregnancy. More recently, pravastatin has been used in a few patients with severe preterm PE to prolong pregnancy. Little is known about admission levels and the change in sFlt1 levels among patients admitted to the hospital who could be eligible for such potential therapeutic interventions.
To answer these questions, we set up a prospective cohort study enrolling patients admitted for evaluation of preeclampsia and measured sequential levels of angiogenic factors and uterine and umbilical Doppler velocimetry studies. The purpose of our study was to evaluate the sequential measurement of angiogenic factors in prediction of adverse maternal, fetal and neonatal outcomes and to see if angiogenic factor levels correlate with umbilical and uterine blood flows.
Material and Methods
Ethics statement
The study was approved by the Beth Israel Deaconess Medical Center (BIDMC) Institutional Review Board (IRB), and all patients provided written informed consent.
Study population
Our patient population included all pregnant patients with singleton pregnancies admitted for evaluation of PE at Beth Israel Deaconess Medical Center from July 2013 until October 2014. Inclusion criteria included women with singleton pregnancies, less than 37 weeks of gestation and at least 18 years of age. Exclusion criteria included multiple gestation and fetuses with chromosomal or congenital anomalies. Discarded plasma samples were obtained daily from the clinical laboratory during the initial 72 hours of admission. Then, blood samples were collected weekly from the hospital clinical laboratory until delivery or discharge. If a discarded sample was not available, blood was drawn directly from the patient. Once collected the samples were centrifuged at 3000 RPM for 10 minutes at −4 ° C, labeled with the study number, aliquotted and stored at −80 ° C for later analysis. Within 48 hours of admission, Doppler studies were performed at the Center for Maternal Fetal Medicine by one specialist using a 5- to 8-MHz curvilinear transabdominal probe and a General Electric Voluson E8 machine (GE Medical Systems), and this exam was repeated on a weekly basis until delivery. Clinical data were ascertained prospectively and included age, race, height, weight, and smoking status, gestational age at time of admission, past medical and obstetrical history, as well as clinical information including blood pressure and laboratory values. All data was entered into a password protected database in real time. All pregnancy outcomes were recorded including complications of delivery, gestational age and birthweight at delivery. Since this was an exploratory study, no formal power analysis to determine sample size was performed.
Diagnosis and outcomes
PE was diagnosed based on the “ACOG Task Force on Hypertension in Pregnancy 2013” criteria : women known to be normotensive who developed a BP ≥140 mm Hg systolic or ≥90 mm Hg diastolic on 2 occasions at least 4 hours apart after the 20th week GA and proteinuria ≥300 mg/24 hour urine collection or protein/creatinine ratio ≥0.3. In the absence of proteinuria, preeclampsia was diagnosed as hypertension with new onset of thrombocytopenia (platelet count <100,000/microliter), elevated liver transaminases (twice the normal range), renal insufficiency (creatinine level >1.1 mg/dL), pulmonary edema and/or new onset of cerebral or visual disturbances. Preeclampsia was considered to have severe features in the presence of any of the following: BP ≥160 mm Hg systolic and ≥110 mm Hg diastolic on 2 occasions at least 4 hours apart unless antihypertensive treatment was initiated, thrombocytopenia, elevated liver enzymes, severe persistent right upper quadrant pain or epigastric pain unresponsive to treatment, progressive renal dysfunction, pulmonary edema, and/or new onset cerebral or visual disturbances. Gestational hypertension (gestational HTN) was defined as new onset hypertension after the 20th week of gestation without the presence of proteinuria. Chronic hypertension (chronic HTN) was defined as hypertension prior to the 20 th week. Superimposed PE was defined as chronic hypertension with development of proteinuria after the 20th week or women with proteinuria prior to the 20th week presenting with exacerbation of blood pressures and/or other signs or symptoms such as elevated liver enzymes, thrombocytopenia, creatinine level >1.1 mg/dL, pulmonary edema, and cerebral or visual disturbances. All diagnoses were confirmed by 2 maternal fetal medicine specialists (KB and SR).
All eligible patients were managed with standard of care. Briefly, antihypertensives were administered to maintain systolic blood pressure mostly below 160 and diastolic below 110 mm Hg; magnesium sulfate was administered in cases of preeclampsia with severe features for seizure prophylaxis. Magnesium was also administered for neuroprotection if the gestational age was less than 31 6/7 and at risk for imminent delivery. Corticosteroids were administered for fetal lung maturity in pregnancies at or beyond 23 5/7 weeks up until 34 0/7 weeks (betamethasone 12 mg/day for 2 days).
Adverse maternal outcomes included: elevated liver function tests (aspartate aminotransferase [AST] or alanine aminotransferase [ALT] [≥80 U/L]), low platelet count (≤100 K/μL), placental abruption (clinical or pathological diagnosis), pulmonary edema, cerebral hemorrhage, convulsion (in the absence of a preexisting seizure disorder), acute renal insufficiency (creatinine >1.1 mg/dL), or maternal death. The adverse fetal/neonatal outcomes include small for gestational age birthweight (≤10th percentile for gestational age) with or without abnormal umbilical artery Doppler (absent or reverse flow), fetal death, and neonatal death.
Angiogenic factor measurements
The samples were analyzed in a single batch for measurement of angiogenic factors in blinded fashion after delivery and after all the outcomes were achieved by all the patients. Assays for sFlt1 and PlGF were performed at the clinical laboratory at BIDMC, Boston, Massachusetts, using the commercially available assays on an automated platform Elecsys (Roche Diagnostics, Penzberg, Germany) and have been previously published. The interassay coefficient of variation for sFlt1 and PlGF immunoassays ranged from 2.6% to 3.0% and 2.0% to 2.4%, respectively. The assay operators were blinded to the clinical information of the participants and the treating physicians were unaware of the test results of sFlt1 and PlGF values. Normal angiogenic profile was defined as patients with sFlt1/PlGF ratio of <85 and abnormal angiogenic profile was defined as sFlt1/PlGF ratio of ≥85.
Doppler measurements
Doppler examination was performed based on the guidelines published in ISUOG 2013. Examination was performed using a 5- to 8-MHz curvilinear transabdominal probe and a General Electric Voluson E8 machine by one sonographer trained to perform these measurements. The sonographer was aware of the clinical indication for the scan but not the course of disease, treatment or angiogenic factor levels. Data were entered prospectively into the database.
Color Doppler was used to optimize the insonation by the pulsed Doppler examination. The angle of insonation for all the Doppler measurements was kept closest to 0° with an angle of correction up to 30°. The measurements were based on 3 to 5 symmetrical waveforms. All measurements were recorded during fetal quiescence. For the umbilical Doppler, the waveform velocities were obtained from a free-floating loop. For the uterine artery Doppler, the transducer was placed on the lower abdomen at the groin area, and color Doppler was applied to identify the uterine artery, at the apparent crossover over the external iliac artery. The resistance index was measured bilaterally and recorded. In the data reporting of the Doppler studies, the independent variable “umbilical artery” was defined abnormal if S/D ratio corrected for GA was elevated, and end diastolic flow was either absent or reversed. For the uterine artery, high-resistance Doppler was defined by mean RI >0.70 in both uterine arteries.
Statistical analysis
Characteristics of patients with and without adverse outcomes at baseline and at delivery were summarized as medians (quartile 1, quartile 3) and compared by use of Mann-Whitney U tests and or number (percentage) by use of χ 2 test. Admission sFlt1/PlGF ratios of patients with and without adverse outcomes were presented as medians (quartile 1, quartile 3) and compared by use of Mann-Whitney U tests. Change in angiogenic factors by adverse outcome groups were presented in a similar manner. Spearman correlation coefficients were used to measure the degree of association between sFlt1/PlGF ratio and ultrasound findings on admission and before delivery. Spaghetti plots were used to show sFlt1, PIGF and sFlt1/PIGF levels and duration of pregnancy with normal and abnormal angiogenic profile at admission. Error bars (median ± standard error) were used to visualize angiogenic levels over time among patients with and without adverse outcomes. All P values were 2-sided, and values of P < .05 were considered to indicate statistical significance. All statistical analyses were performed with SAS version 9.4 (Cary, NC).
Results
Clinical characteristics and adverse outcomes
During the study period, a total of 103 patients were enrolled, 3 patients were excluded (1 patient withdrew, 1 patient had multiple gestation, and 1 was beyond 37 weeks GA); a total of 100 patients were analyzed. Among 100 patients, 53 had diagnosis of PE, 11 had chronic HTN, 27 had superimposed PE, and 9 had gestational HTN. One patient had missing angiogenic factors on admission and 2 additional patients had missing information on fetal/neonatal outcomes.
Adverse outcomes occurred in 43 of 100 subjects enrolled in the study. Baseline characteristics of all patients and patients with and without adverse outcomes are described in Table 1 . Patients with adverse outcomes presented at a significantly earlier gestational age, and although within normal range had higher ALT and lower platelet counts on admission. There was no significant difference in the BMI or race. There was also no significant difference in admission systolic or diastolic blood pressures among patients with and without adverse outcomes.
| Variable | All patients | No adverse outcome | Adverse outcome | P value |
|---|---|---|---|---|
| n | 100 | 57 | 43 | |
| Baseline (admission) | ||||
| Gestational age (weeks) | 33 (31, 35) | 34 (32, 36) | 30 (29, 35) | <.001 a |
| Age (years) | 33 (30, 36) | 33 (31, 36) | 33 (29, 36) | .64 |
| Body mass index (kg/m 2 ) | 31.6 (28.5, 37.3) | 33.0 (28.5, 39.3) | 31.4 (28.5, 34.2) | .33 |
| Nulliparous (%) | 57 (57.0) | 28 (49.1) | 29 (67.4) | .07 |
| Current smoker (%) | 1 (1.0) | 0 (0.0) | 1 (2.3) | .25 |
| Race (%) | .80 | |||
| White/Caucasian | 55 (55.0) | 32 (56.1) | 23 (53.5) | |
| Black/African American | 17 (17.0) | 11 (19.3) | 6 (14.0) | |
| Asian | 6 (6.0) | 3 (5.3) | 3 (7.0) | |
| Other/unknown | 22 (22.0) | 11 (19.3) | 11 (25.6) | |
| History of preeclampsia (%) | 20 (20.2) | 14 (24.6) | 6 (14.3) | .21 |
| History of chronic hypertension (%) | 40 (40.0) | 26 (45.6) | 14 (32.6) | .19 |
| History of diabetes (%) | 11 (11.0) | 8 (14.0) | 3 (7.0) | .26 |
| Admission clinical findings | ||||
| Highest SBP (mm Hg) | 122 (116, 136) | 121 (112, 136) | 122 (116, 140) | .60 |
| Highest DBP (mm Hg) | 80 (70, 84) | 79 (70, 82) | 80 (70, 90) | .40 |
| Proteinuria (%) | 5 (18.5) | 2 (13.3) | 3 (25.0) | .44 |
| ALT (U/L) | 17 (13, 26) | 15 (12, 20) | 21 (14, 53) | .007 a |
| Creatinine (mg/dL) | 0.6 (0.5, 0.7) | 0.6 (0.5, 0.7) | 0.6 (0.5, 0.7) | .22 |
| Uric acid (mg/dL) | 5.3 (4.3, 6.4) | 4.9 (4.2, 5.9) | 5.8 (4.4, 6.5) | .05 |
| Platelet count (K/μL) | 224 (190, 269) | 246 (205, 279) | 213 (165, 250) | .002 a |
| Admission angiogenic factors | ||||
| sFlt1 (pg/mL) | 6605 (3692, 10274) | 5595 (2821, 8193) | 9136 (5724, 12161) | <.001 a |
| PlGF (pg/mL) | 88 (37, 177) | 125 (65, 266) | 49 (21, 94) | <.001 a |
| sFlt1/PlGF ratio | 73.8 (22.4, 228.8) | 47.5 (9.7, 87.0) | 205.9 (72.5, 453.1) | <.001 a |
| sFlt1/PlGF ≥85 (%) | 48 (48.5) | 17 (29.8) | 31 (73.8) | <.001 a |
| Hospital course | ||||
| Highest SBP (mm Hg) | 167 (155, 180) | 162 (151, 177) | 172 (160, 185) | .006 a |
| Highest DBP (mm Hg) | 100 (94, 107) | 99 (92, 105) | 105 (96, 110) | .004 a |
| Proteinuria (%) | 78 (83.9) | 42 (79.3) | 36 (90.0) | .16 |
| Betamethasone (%) | 53 (53.0) | 21 (36.8) | 32 (74.4) | <.001 a |
| Antihypertensive (%) | 61 (61.0) | 32 (56.1) | 29 (67.4) | .25 |
| Gestational age at delivery (weeks) | 35 (32, 37) | 36 (34, 37) | 33 (30, 35) | <.001 a |
| Preterm delivery <37 weeks (%) | 71 (72.5) | 34 (61.8) | 37 (86.1) | .008 a |
| Birthweight (grams) | 2348 (1515, 2850) | 2710 (2355, 3035) | 1525 (1050, 2025) | <.001 a |
| Placental weight (grams) | 311 (228, 394) | 373 (284, 454) | 245 (180, 338) | <.001 a |
During the course of hospitalization, patients with adverse outcomes had significantly higher systolic and diastolic blood pressures, received more betamethasone for fetal lung maturity, had lower gestational age of delivery and a larger proportion of patients delivering before the 37 weeks of gestation (all P < .01) ( Table 1 ). These patients also had lower birthweight babies and lower placental weight (both P < .001).
Angiogenic factors and adverse outcomes
At admission, women with adverse outcomes had higher sFlt1 (pg/mL) median (25th to 75th centile) 9136 (5724, 12161) versus 5595 (2821, 8193) and lower PlGF (pg/mL) median (25th to 75th centile) 49 (21, 94) versus 125 (65, 266) (all P values <.001) ( Table 1 ). We then calculated the sFlt1/PlGF ratio and found that subjects who developed adverse outcomes had higher sFlt1/ PlGF ratio, median (25th to 75th centile) was 205.9 (72.5, 453.1) versus 47.5 (9.7, 87.0) compared with patients without adverse outcomes ( P < .001), Tables 1 and 2 . The differences in the ratio between the two groups were statistically significant even after adjustment of gestational age at presentation ( P unadjusted was <.001 and P adjusted was .001). Similarly, patients with individual adverse outcomes had higher sFlt1/PlGF ratio, although data were limited by the small number of individual outcomes ( Table 2 ).
| Outcome | No adverse outcome | Adverse outcome | P value | ||
|---|---|---|---|---|---|
| n | sFlt1/PIGF ratio Median (Q1, Q3) | n | sFlt1/PIGF ratio Median (Q1, Q3) | ||
| Any outcome b | 57 | 47.5 (9.7, 87.0) | 42 | 205.9 (72.5, 453.1) | <.001 a |
| Maternal | |||||
| ELLP | 94 | 70.1 (22.1, 225.3) | 5 | 177.0 (170.0, 330.3) | .047 a |
| Transaminitis | 82 | 65.2 (20.7, 145.8) | 17 | 241.2 (170.0, 304.4) | .002 a |
| Thrombocytopenia | 92 | 70.1 (22.3, 220.1) | 7 | 177.0 (114.3, 475.9) | .07 |
| DIC | 99 | 73.8 (22.4, 228.8) | 0 | n/a | n/a |
| Abruption | 94 | 73.2 (22.4, 225.3) | 5 | 103.4 (32.0, 330.3) | .61 |
| Eclampsia | 98 | 77.9 (22.4, 228.8) | 1 | n/a | n/a |
| Cerebral hemorrhage | 99 | 73.8 (22.4, 228.8) | 0 | n/a | n/a |
| ARF | 94 | 73.2 (22.1, 225.3) | 5 | 114.3 (68.0, 280.7) | .29 |
| Pulmonary edema | 96 | 73.2 (22.3, 235.0) | 3 | 170.0 (32.0, 197.0) | .74 |
| Maternal death | 99 | 73.8 (22.4, 228.8) | 0 | n/a | n/a |
| Fetal/neonatal | |||||
| SGA ≤10% | 79 | 72.3 (22.1, 202.1) | 18 | 143.0 (34.1, 562.0) | .08 |
| Abnormal umbilical Doppler | 88 | 65.9 (22.0, 187.3) | 11 | 295.1 (72.5, 1007.7) | .003 a |
| Neonatal death (<28 days) | 96 | 77.9 (23.5, 227.1) | 1 | n/a | n/a |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
