American Journal of Obstetrics and Gynecology (2006) 195, 255–9

www.ajog.org

Evaluation of placenta growth factor and soluble

Fms-like tyrosine kinase 1 receptor levels in mild
and severe preeclampsia
Christopher J. Robinson, MD,a Donna D. Johnson, MD,a,* Eugene Y. Chang, MD,a
D. Michael Armstrong, MD,a Wei Wang, MSPHb

Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology,a Departments of Biostatistics,
Bioinformatics, and Epidemiology,b Medical University of South Carolina, Charleston, SC

Received for publication September 14, 2005; revised December 13, 2005; accepted December 22, 2005

KEY WORDS Objective: The purpose of this study was to determine if maternal serum concentrations of pla-
Pregnancy centa growth factor (PlGF) and soluble Fms-like tyrosine kinase 1 receptor (s-Flt1) are more
Preeclampsia abnormal in patients with severe preeclampsia compared with mild preeclampsia.
Placenta growth factor Study design: Serum samples were collected from 32 control patients and 80 patients with mild or
(PlGF) severe preeclampsia. PlGF and s-Flt1 concentrations were quantitated by enzyme-linked immu-
Soluble Fms-like nosorbent assay (ELISA). Results are expressed as median (Q1-Q3) unless stated otherwise. After
tyrosine kinase normalization, serum markers were compared using one-way analysis of covariance (ANCOVA).
1 receptor (s-Flt1) Results: Patients with preeclampsia had decreased levels of PlGF (75.1 G 14 vs 391 G 54 pg/mL,
P ! .0001) and elevated s-Flt1 concentration (1081 G 108 vs 100.1 G 26.9 pg/mL, P ! .0001)
compared with the respective controls (mean G SEM). PlGF concentration was lower in patients
with mild preeclampsia compared with severe, respectively (67 pg/mL [39-158] vs 24 pg/mL [4-57],
P ! .02). s-Flt1 was not different between mild and severe preeclampsia (674 pg/mL [211-1297] vs
1015 pg/mL [731-1948], P = .08).
Conclusion: PlGF and s-Flt1 serum levels are abnormal in patients with preeclampsia com-
pared with controls, but only PlGF is more abnormal in severe preeclampsia compared with mild
preeclampsia.
Ó 2006 Mosby, Inc. All rights reserved.

Supported by National Institutes of Health National Institute of
Child Health and Human Development grant number: 5 R03 HD
41031-02, Institutional Review Board Project # HR 9983.
Presented at the Twenty-fifth Society of Maternal Fetal Medicine
Meeting, Reno, NV, February 7-12, 2005.
* Reprint requests: Donna D. Johnson, MD, Department of
Obstetrics and Gynecology, Medical University of South Carolina,
96 Jonathan Lucas St, PO Box 250619, Charleston, SC 29439.
E-mail: johnsodo@musc.edu
Preeclampsia is a multisystem disorder that occurs
in 5% of all pregnancies and is a major contributor to
maternal and neonatal morbidity and mortality.1,2 The
hallmark of the disease is hypertension and proteinuria.
The etiology of preeclampsia remains speculative and
is most likely multifactorial; however, endothelial dys-
function likely contributes to the multisystem organ
dysfunction.3-6

0002-9378/$ - see front matter Ó 2006 Mosby, Inc. All rights reserved.
doi:10.1016/j.ajog.2005.12.049
256
Table I Population demographics
Characteristic Control (n = 32)
Age (y) 24 (22-27)
Race
White (%) 29
Black (%) 52
Hispanic (%) 19
Gestational age at delivery (wk) 38 (36-39)
Primiparous (%) 23.0
Mean arterial pressure (mm Hg) 86 (81-96)
Birth weight (g) 3403 (3105-3622)
Apgar !6 at 5 minutes (n) 0 2
Data presented as median (Q1-Q3) unless otherwise indicated.
* Analysis by Kruskal-Wallis test.
y
Analysis by c2.
z
Analysis by Fisher exact test.
Recently, growth factors that activate the endothe-
lium have received attention. Vascular endothelial growth
factor (VEGF) is the most potent stimulant of vascular
endothelium and permeability studied to date.7-9 To
confer its biological properties, VEGF must covalently
dimerize before it can bind to its endothelial receptor
and activate endothelial cells. Two proteins, placenta
growth factor and soluble fms-like tyrosine kinase 1 re-
ceptor (s-Flt1), can modulate VEGF biological activity.
First, VEGF may form a bond with VEGF to form
a homodimer or with placenta growth factor (PlGF)
to form a heterodimer. VEGF homodimers are much
more potent stimulants to the endothelial cells than the
VEGF-PlGF heterodimers. In fact, PlGF homodimers
compete with the same receptors as the more potent
VEGF dimers and only weakly activate endothelial
cells.10 Next, VEGF and PlGF may circulate bound to
s-Flt1. Only the free VEGF and PlGF is bioavailable.11,12
So, PlGF and s-Flt1 receptor are both important regula-
tors of VEGF biological activity.13,14
Whether VEGF serum levels are altered in pree-
clampsia remains debateable, but the factors that mod-
ulate the biological activity of VEGF are clearly altered.
PlGF is significantly decreased in preeclamptic patients
compared with controls.15-17 S-Flt1 receptor is increased
in patients with preeclampsia. The serologic changes in
PlGF and s-Flt1 precede the onset of clinical symp-
toms.17,18 The hypothesis to be tested in this study was
that PlGF and s-Flt1 would be more altered in
maternal serum in patients with severe preeclampsia
compared with mild preeclampsia.
Material and methods
The Institutional Review Board at the Medical Univer-
sity of South Carolina approved this study. Control
patients were healthy nonsmokers with an uncomplicated
Robinson et al
Mild PE (n = 27) Severe PE (n = 53) P value
24 (19-32) 24 (22-32) NS*
22 39 NSy
59 54
19 7
37 (36-39) 32 (31-36) ! .001*
35.6 64.4 .025y
113 (102-120) 119 (112-136) ! .001*
2991 (2690-3450) 1796 (1245-2515) ! .001*
5 NSz
antenatal course. Patients who met the criteria for mild
or severe preeclampsia as defined by the American
College of Obstetricians and Gynecologists were en-
rolled.19 Hemolysis, elevated liver enzymes, and low
platelets (HELLP) syndrome was defined as an aspar-
tate amniotransferase (AST) greater than 70 IU/L
and platelet count less than 100,000. The definition of
HELLP syndrome used in this study is limited in
that peripheral smears were not obtained to support in-
travascular hemolysis. Some authors have defined this
as ‘‘partial’’ HELLP syndrome.20 If a history of
chronic hypertension, collagen vascular disease, multi-
ple gestations, renal disease, diabetes, current urinary
tract infection, or tobacco abuse was obtained, the pa-
tient was excluded. Both controls and patients with ei-
ther mild or severe preeclampsia were enrolled into the
study before spontaneous labor, induction of labor, or
rupture of fetal membranes. A subanalysis of serum
PlGF and s-Flt1 in patients with HELLP syndrome
was compared with patients with severe preeclampsia.
Whole blood was obtained from participants by
venipuncture and collected in a serum-separator tube.
The serum was removed after centrifugation at 10,000
RPM for 10 minutes. Serum samples were aliquotted
and stored at
70(C for 4 months before assay. Assays
were performed in batches every 4 months.
Quantikine human PlGF and s-Flt1 ELISA (R&D
Systems, Minneapolis, MN) kits were used to measure
the serum concentrations in duplicate. A microplate reader
was utilized to detect the optical density at 450 nm with
optical correction at 540 nm for each serum sample. PlGF
and s-Flt1 concentrations were determined by compari-
son to standard curve measurements performed under
identical lab conditions at the time of assay. Concentra-
tions of PlGF and s-Flt1 were reported as picograms per
milliliter. The lab personnel were blinded to the clinical
information regarding each subject.
Robinson et al
Table II Analysis of covariance model results
Log- Square
transformed root-transformed
ANCOVA/P value PlGF s-Flt1
Group
Control vs mild vs severe ! .001 ! .001
Control vs mild or severe ! .001 ! .001
Mild vs severe ! .03 NS
Gestational age at delivery NS NS
Mean arterial pressure NS NS
Birth weight NS NS
Primiparous NS NS
The PlGF ELISA measured free PlGF. The s-Flt1
ELISA measured only the soluble form of the VEGFR1
receptor. The minimal detection, intra-assay, and inter-
assay variance of PlGF ELISA were 7 pg/mL, 7.0%,
and 12%, respectively. The minimal detection, intra-
assay, and interassay variance of the s-Flt1 ELISA were
5.0 pg/mL, 4%, and 8%, respectively.
Urinary protein was quantified from a 24-hour sam-
ple. To collect the sample, the first urine collection was
discarded. All urine samples were collected and stored in
an opaque plastic container in the refrigerator. Urine
collection was completed 24 hours after the first sample
had been discarded. The protein in the sample was
determined by spectrophotometry on a Synchron LX-20
(Beckman Coulter, Fullerton, CA) and reported in
mg/dL. Total protein was calculated based on the volume
of the specimen.
Descriptive statistics were presented as median and
quantiles for continuous variables and percentage for
categorical data. P values displayed in Table I were esti-
mated by Kruskal-Wallis test, and for all categorical
data, by chi-square test.
The primary comparison (control vs mild preeclamp-
sia vs severe preeclampsia) was tested by fitting an
analysis of covariance model to PlGF and s-Flt1, with
gestational age, mean arterial pressure, birth weight, and
the 5-minute Apgar score as covariates and primiparity
as a factor (Table II). In addition, further preplanned
group comparisons were performed given an overall sig-
nificant test. The secondary comparison for a subgroup
of preeclampsia patients (mild vs severe vs HELLP) was
explored by one-way analysis of variance.
Because of skewed distributions of the data examined
by graphic technique and normality test, a log transfor-
mation was performed on PlGF and a square root
transformation on s-Flt1. Analyses of statistical signif-
icance with parametric methods were then conducted
on the transformed measures. All tests with P ! .05
(2-tailed) were considered statistically significant. SAS,
version 9.1 (SAS Institute, Inc, Cary, NC) was used
for all analyses.
257
Figure 1 A, Boxplot (top and bottom of box represent the
third and first quartile, respectively, horizontal line within
the box represents median value, and upper and lower ex-
tremes are represented by vertical bars from the top and bot-
tom of the box) of PlGF and B, s-Flt1 serum concentrations
in controls versus mild versus severe preeclampsia. Both
PlGF and s-Flt1 are altered in patients with preeclampsia com-
pared to controls but only PlGF serum levels are different in
patients with mild versus severe disease. Data were analyzed
using ANCOVA. *P value based on log-transformed PlGF
data; #P value based on square-root transformed s-Flt1 data.
Results
Demographics are shown in Table I. As expected,
patients with severe preeclampsia were delivered at an
earlier gestational age, more likely to be nulliparous,
had a higher mean arterial pressure, and delivered smaller
infants than patients in the control and mild pree-
clamptic group.
Patients with preeclampsia (n = 80) had decreased
levels of PlGF when compared with controls (n = 32)
(75.1 G 14 vs 391 G 54 pg/mL, P ! .0001). s-Flt1 con-
centration was elevated in the preeclamptic population
compared with controls (1081 G 108 vs 100.1 G 26.9
pg/mL, P ! .0001). PlGF serum concentration in severe
preeclampsia (n = 53) was 3-fold lower than in mild
preeclampsia (n = 27). Although s-Flt1 serum concen-
tration was one-and-a-half-fold higher in severe pree-
clampsia than in mild preeclampsia, this difference was
not significantly different (Figure 1). PlGF and s-Flt
concentrations were not significantly different in the
patients with severe preeclampsia and HELLP syndrome
(n = 11), respectively (PlGF 35.5 G 13.8, P = NS; s-Flt1
1422 G 324, P = NS) (Figure 2). When examining all
patients with preeclampsia, a significant correlation was
258
Figure 2 Boxplot of PlGF (A) and s-Flt1 (B) serum concen-
trations in severe preeclampsia versus HELLP syndrome. No
significant difference in PlGF or s-Flt1 was demonstrated
between severe preeclampsia and HELLP syndrome. Data were
analyzed using ANOVA. *P value based on log-transformed
PlGF data; #P value based on square-root transformed s-Flt1
data.
observed between s-Flt1 and proteinuria (P = .42, P ! 0.01)
but not between PlGF and proteinuria using Spearman’s
correlation.
Comment
Given the information that is known about the biolog-
ical action of VEGF, it is only natural to study regula-
tory factors of VEGF in preeclamptic patients. Several
investigators have demonstrated that PlGF serum levels
are significantly decreased in patients with preeclampsia
and our study confirms these findings.15-17 Our data
demonstrate that serum PlGF is lower in patients with
severe preeclampsia compared with mild preeclampsia.
However, PlGF levels are not different between patients
with severe preeclampsia and HELLP syndrome.
More recently, s-Flt1 has been shown to be elevated
in patients with preeclampsia and our findings are
similar. Our data demonstrate that s-Flt1 levels are
higher in patients with severe preeclampsia though not
statistically significant compared with mild preeclampsia.
Furthermore, s-Flt1 levels are not different between
patients with severe preeclampsia and HELLP syndrome.
Robinson et al
The data in this observational study clearly demon-
strate that maternal serum concentrations of PlGF and
s-Flt1 receptor are abnormal in patients with preeclamp-
sia that are otherwise a healthy, nonsmoking popula-
tion. Only serum levels of PlGF are more abnormal
between severe and mild disease. To determine if PlGF
serum levels decrease and s-Flt1 increase as the disease
progresses, the more ideal study design would be to
follow patients longitudinally. Efforts made toward a
longitudinal study were thwarted as most patients in this
study either presented with severe preeclampsia and
were delivered expeditiously or had mild preeclampsia
at term and were induced. The next acceptable study
design would be a case-controlled study. Control pa-
tients were initially matched with patients with pree-
clampsia by gestational age. Patients who are delivered
at a preterm gestational age similar to patients with
severe preeclampsia have either an obstetric problem or
medical problem that required delivery. These problems
such as chronic hypertension, diabetes, or multiple
gestations lead to even more confounding variables, so
these patients were excluded for controls.
Gestational age does affect maternal serum levels of
PlGF and s-Flt. Notably, the samples from the severe
preeclampsia population were obtained at an earlier
gestational age when compared with either the control
or mild preeclampsia patients. PlGF serum concentra-
tions peak at 26 to 30 weeks and then decline as term
approaches.21 S-Flt1 has a stable concentration until
33 to 36 weeks and then increases about 145 pg/mL
per week.17 Patients with severe disease in this study
would be expected to have higher PlGF and lower
s-Flt1 levels than controls and patients with mild pree-
clampsia if gestational age were a major factor in our
population. The opposite observation was found.
The mean birth weight of the patients with severe
preeclampsia is at the 27th percentile for gestational
age.22 Fifteen of 53 infants born to mothers with severe
preeclampsia were small for gestational age (SGA).
Patients with severe preeclampsia are known to have a
higher incidence of SGA.23 In a similar study, PlGF
and s-Flt1 levels were more deranged in patients with
severe preeclampsia and SGA infants.17 So, PlGF and
s-Flt1 may also be affected by abnormal fetal growth.
If PlGF and s-Flt1 levels were simply markers of pla-
cental function both serum levels should decrease, but
s-Flt1 actually increases.
The pathophysiology of severe preeclampsia and
HELLP syndrome is debated in maternal fetal medicine.
These two disorders may have similar etiologies or may
differ greatly. In this study, the levels of PlGF and s-Flt1
are not different between the two populations. These
data suggest that if PlGF and s-Flt1 play a role in the
development of preeclampsia, their role may not differ
in patients with HELLP syndrome or severe preeclamp-
sia. Although our number of HELLP patients are small,
Robinson et al
using a power calculation with a 2-tailed alpha = 0.05,
197 patients are needed to show a statistically difference
between patients with HELLP and severe preeclampsia.
Although a larger number of patients may demonstrate
a statistical difference, the known biological activity of
PlGF and s-Flt1 would be more difficult to correlate.
We specifically chose to look at proteinuria for
several reasons. Hypertension and proteinuria are the
hallmarks of preeclampsia. Proteinuria is known to be
associated with increased permeability in the renal
basement membrane. Serial measurements of 24-hour
urine protein excretion are used to assess of the pro-
gression of preeclampsia. Animal studies demonstrate
that as the concentration of these growth factors are
manipulated the amount of proteinuria is altered.12 Our
data suggest that as the serum levels of s-Flt1 become
more abnormal, proteinuria increases. PlGF levels did
not correlate with the amount of proteinuria, so it is bi-
ologically plausible that s-Flt1 may play a role in the
proteinuria associated with preeclampsia. Because
s-Flt1 correlates with the degree of proteinuria, s-Flt1
should intuitively differentiate between mild and severe
preeclampsia. However, in this study most patients
with proteinuria had less than 5 g and did not meet
the criteria for severe preeclampsia using this clinical
parameter.
Many researchers have spent years trying to find the
cause of preeclampsia but the mechanism remains elu-
sive. It is possible that diminished maternal serum levels
of PlGF and increased levels of s-Flt1 may contribute to
increased vascular permeability. However, it is just as
likely these serum proteins are just markers for the
disease and have no role in the mechanism.
References
1. Walker J. Pre-eclampsia. Lancet 1999;356:1260-5.
2. Roberts JM, Cooper DW. Pathogenesis and genetics of pre-
eclampsia. Lancet 2001;357:53-6.
3. Roberts JM, Taylor RN, Musci TJ, Rodgers GM, Hubel CA,
McLaughlin MK. Preeclampsia: an endothelial cell disorder. Am
J Obstet Gynecol 1989;161:1200-4.
4. Dekker GA, van Geijn HP. Endothelial dysfunction in pre-
eclampsia. Part I. J Perinat Med 1996;24:99-117.
5. Dekker GA, van Geijn HP. Endothelial dysfunction in pre-
eclampsia. Part II. J Perinat Med 1996;24:119-39.
259
6. VanWijk MJ, Kublickiene K, Boer K, VanBavel E. Vascular func-
tion in preeclampsia. Cardiovasc Res 2000;47:38-48.
7. Carmeliet P. Mechanisms of angiogenesis and arteriogenesis. Nat
Med 2000;6:389-95.
8. Ferrara N. VEGF: an update on biological and therapeutic
aspects. Curr Opin Biotechnol 2000;11:617-24.
9. Zachary I, Gliki G. Signaling transduction mechanisms mediating
biological actions of the vascular endothelial growth factor family.
Cardiovasc Res 2001;49:568-81.
10. De Falco S, Gigante B, Persico MG. Structure and function of pla-
cental growth factor. Trends Cardiovasc Med 2002;12:241-6.
11. Koga K, Osuga Y, Yoshino O, Hirota Y, Ruimeng X, Hirata T,
et al. Elevated serum soluble vascular endothelial growth factor
receptor 1 levels in women with preeclampsia. J Clin Endocrinol
Metab 2003;88:2348-51.
12. Sugimoto H, Hamano Y, Charytan D, Cosgrove D, Kieran M,
Sudhakar A, et al. Neutralization of circulating vascular endothe-
lial growth factor (VEGF) by anti-VEGF antibodies and soluble
VEGF receptor 1 induces proteinurea. J Biol Chem 2003;278:
12605-8.
13. Persico MG, Vincenti V, DiPalma T. Structure, expression, and
receptor-binding properties of placenta growth factor. Curr Top
Microbiol Immunol 1999;237:31-40.
14. Eriksson U, Alitalo K. Structure, expression, and receptor-binding
properties of novel vascular endothelial growth factors. Curr Top
Microbiol Immunol 1999;237:41-57.
15. Reuvekamp A, Velsing-Aarts FV, Poulina IE, Capello JJ, Duits AJ.
Selective deficit of angiogenic growth factors characterises pre-
gnancies complicated by preeclampsia. BJOG 1999;106:1019-22.
16. Torry DS, Wang HS, Wang TH, Caudle MP, Torry RJ. Pree-
clampsia is associated with reduced serum levels of placenta
growth factor. Am J Obstet Gynecol 1998;179:1539-44.
17. Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF,
et al. Circulating angiogenic factors and the risk of preeclampsia.
N Engl J Med 2004;350:672-83.
18. Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et al.
Excess placental soluble fms-like tyrosine kinase 1 may contribute
to endothelial cell dysfunction, hypertension, and proteinurea in
preeclampsia. J Clin Invest 2003;111:649-58.
19. American College of Obstetricians and Gynecologists. ACOG
Practice Bulletin No. 33. Diagnosis and management of pree-
clampsia and eclampsia. Obstet Gynecol 2002;99:159-67.
20. Sibai SM. The HELLP syndrome (hemolysis, elevated liver
enzymes, and low platelet count): much ado about nothing? Am
J Obstet Gynecol 1990;162:311-6.
21. Tidwell SC, Ho HN, Chiu WH, Torry RJ, Torry DS. Low mater-
nal serum levels of placenta growth factor as an antecedent of clin-
ical preeclampsia. Am J Obstet Gynecol 2001;184:1267-72.
22. Hadlock FP, Deter RL, Harrist RB, Park SK. Estimating fetal age:
computer assisted analysis of multiple fetal growth parameters.
Radiology 1984;152:497-505.
23. Xiao R, Sorensen TK, Williams MA, Luthy DA. Influence of pre-
eclampsia on fetal growth. J Matern Fetal Neonatal Med 2003;13:
157-62.