ativacao plaquetaria10set2012

7/29/2019 ativacao plaquetaria10set2012

1/57

EVALUATION OF PLATELET ACTIVATION BY FLOW CYTOMETRY, IN

PREECLAMPSIA

FREITAS, Letcia GonalvesMARTINS-FILHO, Olindo Assis

CARVALHO, Maria das Graas

SATHLER-AVELAR, Renato

DUSSE, Luci Maria SantAna

ABSTRACT

Preeclampsia (PE) is a serious complication of pregnancy associated with high

morbidity and mortality, maternal-fetal, whose etiology has not been

established. In pure form, it is characterized by the appearance, in normal

pregnant women after 20 weeks of pregnancy, hypertension and proteinuria.

The hemostatic and inflammatory changes associated with normal pregnancy

are more exacerbated in PE. Studies suggest that pronounced platelet

activation and aggregation contribute to the hypercoagulable state in this

disease. The aim of this study was to evaluate the expression of markers of

platelet activation in PE. Were evaluated 97 women, 35 of which were pregnant

women with PE (15 with severe-PEG and 20 mild-PEL), 31 normotensive

pregnant women (GN) and 31 nonpregnant women (CNG). Were evaluated the

mean fluorescence intensity (MFI) of biomarkers CD41a, CD61, CD42a, CD62P

and the ratio between them as well as the percentage of platelets CD62P +. The

IMF of CD62P and the percentage of CD62P+ platelet, in the three groups did

not differ (p = 0.67 and p = 0.38, respectively). After subdividing the group of

pregnant women with PE was not obtained difference (p = 0.69 for the IMF

CD62P and p = 0.54 for CD62P+ platelet). The MFI of CD41a, in pregnant

women with PE was lower than in CNG (p = 0.004) and GN compared to CNG

(p = 0.007). The MFI of CD61 on PEG group was lower than the CNG (p =

0.008). The MFI of CD41a and CD61 were lower in the PE group than in the

CNG, and the GN compared to CNG (p = 0.00 for both). Both were lower in

PEG and PEL compared to CNG (p = 0.00 for both). The MFI of CD42a, even


2/57

after the division of the PE group did not differ (p = 0.44 and p = 0.57,

respectively). The ratios between the IMF of CD61/CD42a and CD41a/CD62P

were lower in NG than in the CNG (p = 0.002 and p = 0.007, respectively) and

in the group of PE and PEG compared to CNG (p = 0.001 for both). The reason

CD42a/CD41a MFI was greater in PE and PEG compared to CNG (p = 0.001

for both) and in GN compared to CNG (p = 0.002). The ratios of the MFI

CD61/CD41a, CD61/CD62P and CD42a/CD62P before (p = 0.29, p = .07 and p

= .73, respectively) and after (p = 0.42, p = 0,14, p = 0.69, respectively)

subdividing the PE group did not differ. Pregnancy is accompanied by changes

in expression of CD41a and CD61 on the platelet surface, and reduced platelet

counts.

Keywords: Preeclampsia, CD41a, CD61, CD42a, CD62P.

INTRODUCTION

Primary hemostasis and platelets

Platelets are cytoplasmic fragments of megakaryocytes. These are present

in the bone marrow and their maturation and differentiation depends on

thrombopoietin and interleukin (IL) IL-3, IL-6 and IL-11. The cytoplasmic

fragmentation of megakaryocytes occurs in the bone marrow and platelets are

released into the circulation, isolated or grouped (HARTWIG et al. 1995). For a

long time the platelets were considered mere onlookers on hemostasis (JURK &

KEHREL, 2005). Currently, it is clear that they play an essential role in this

process and have other important functions such as the maintenance of

vascular tone through the secretion of various substances including serotonin,thromboxane (TXA 2) and prostaglandin (HARTWIG et al. 1995).Thus, platelets

act as contributors and modulators of various disorders, including coronary

artery disease, deep vein thrombosis, myeloproliferative disorders, atrial

fibrillation, cancer, cerebrovascular accident (CVA), infection with human

immunodeficiency virus (HIV), kidney disease, diabetes, inflammatory bowel

disease and multiple sclerosis (HARTWING & ITALIAN-JR, 1995; GURNEY et

al., 2002; FARIAS & Bo, 2008; SHEREMATA et al., 2008; GILS Van et al.,2009; HESS & GRANT, 2011 ). The role of platelets in the hemostatic process


3/57

includes their recruitment from the bloodstream to the subendothelial matrix

occurs when vascular damage and release of tissue factor (TF). When

activated, platelets quickly express glycoproteins (GP) surface that interact with

other platelet, vascular endothelial and inflammatory cells (Hartwig et al. 1995;

Lowenberg et al., 2010). A succession of events including contact, adhesion,

activation, secretion and aggregation culminates in the formation of platelet

plug, which temporarily prevents blood loss (JURK & KEHREL, 2005).

Subsequently, platelets also contribute to secondary hemostasis by activating

the coagulation cascade occurs in which the fibrin clot formation, plugging more

efficiently the injured site.

Platelet activation

Platelet activation occurs primarily through two mechanisms: release of

soluble substances (proteases, metabolites of arachidonic acid, peroxides,

platelet-activating factor and cathepsin G in pathological conditions) and by

direct contact with inflammatory cells. The activation results in an increase in

expression of adhesion molecules surface starting GPIIb/IIIa on platelets and

leukocytes CD11b/CD18. These leukocyte GP establish connections with

platelets (PANASLUK et al. 2005). P-selectin and GPIIb/IIIa are biomarkers of

platelet activation most commonly used in research. Besides these, the

lysosomal protein (CD63), glycoproteins p24 (CD9), GPIIa (CD29), GPIV

(CD36), GPIb (CD42b), GPIb (CD42c), GPIa (CD49b), GPIC (CD49f) and

adhesion molecule platelet-endothelial cell type 1 (PECAM-1 or CD31) have

been studied (Lazarus et al. 1995).

P-selectin (CD62P)

The P-selectin mediates platelet adherence of activated platelets to

neutrophils and monocytes. Monoclonal antibodies directed against these GP

only bind to platelets degranuladas and not those in the inactive state

(Michelson, 1996; Michelson et al. 2000). P-selectin is synthesized by

endothelial cells and megakaryocytes and then stored in platelet granules andWeibel-Palade bodies of endothelial cells (Gurney et al. 2002; YANG et al.


4/57

2,009; Lowenberg et al. 2,010). When exposed on the platelet surface, P-

selectin average interactions between the endothelium, platelets and

leukocytes, and initially stabilizes the interaction GPIIb/IIIa-fibrinogen to the

platelet plug formation (Gurney et al. 2002; Theoret et al. 2006). However, this

biomarker is unstable and releases from the platelet surface rapidly, while still

functioning in plasma. The soluble P-selectin (sCD62P) is known to induce a

procoagulant state in monocytes and is considered a potent activator of T cells,

endothelium, platelets, essential for stabilizing an arterial thrombus (Jurk &

KEHREL, 2005).

Increased plasma levels of this biomarker has been observed in

thrombotic thrombocytopenic purpura, hemolytic uremic syndrome, diabetes,

cancer, systemic inflammation, atherosclerosis, acute coronary syndrome, and

other cardiovascular events in smokers (GURNEY et al., 2002; Lowenberg et al.

, 2010). It is assumed that P-selectin presents important role as a connective

factor between inflammation and thrombosis. Moreover, their plasma levels are

elevated a few hours after the occurrence of venous thrombosis and remain

high for many months (Masopust et al., 2011).

Glycoprotein GPIIIa (CD61)

The GPIIIa is a transmembrane GP, 3 integrin expressed on platelets,

megakaryocytes, osteoclasts and endothelium. This GP is associated with

GPIIb to form the GPIIb/IIIa complex, which mediates platelet aggregation and

CD51 to form the CD51/CD61 complex, a vitronectin receptor. In addition, it

binds to fibrinogen, fibronectin, thrombospondin and von Willebrand factor

(vWF) to mediate the process of accession (di MINNO et al., 2009).

Glycoprotein IIb/IIIa (CD41a)

The GPIIb/IIIa was first described in the 1980s triggered by events in

platelet activation. This GP is activated after a signal initiated with intra-platelet

complex formation FvW-GPIb/IX/V and is able to bind to fibrinogen, fibronectin,

vWF, vitronectin and thrombospondin (Michelson, 1996; GURNEY et al. 2002;Delgado et al. 2003). The increased plasma levels of these GP was useful to


5/57

evaluate the pro-thrombotic activity in various clinical conditions, including

coronary angioplasty, sickle cell disease in remission and during painful crisis

and hypertension caused by pregnancy, especially in preeclampsia-PE (Tomer,

2004). Clinically, platelet aggregation inhibitors that target GPIIb/IIIa are

effective to reduce thrombotic disorders during coronary intervention (Freedman

& LOSCALZO, 2002, Theoret et al. 2006).

GPIX glycoprotein (CD42a)

The GPIX form the complex GPIb/IX/V receptor for vWF, essential in the

process of platelet adhesion to subendothelial vascular when a injury occurs. In

contrast to antibodies that generally have higher affinity for platelet receptorswhen platelets are activated, CD42 has a higher affinity for platelets in an

inactive state. This probably occurs due to complex translocation to the

membrane connected to the channel system (Michelson, 1996).

A new theory of coagulation and the role of tissue factor and platelet

surface

The coagulation theory, proposed in the 1960s by MaFCarlane, Davieand Ratnoff, was revised and a new model was proposed by Hoffman & Monroe

in 2007. The model consists of three stages: initiation, amplification and

propagation, coordinated by FT (expressed in tissue cells, monocytes and

activated cells endotelials) and platelet surface. Then, when a blood vessel is

injured, a large number of platelets are recruited to the site in order to provide

sufficient area for a burst of thrombin generation and fibrin clot formation

effective in buffering the injury and ensuring hemostasis. The platelet surface is

specialized to coordinate the connection of the tenase and prothrombinase

complex (FXa, FVa).

Evaluation of platelets by flow cytometry

Flow cytometry (FC) using monoclonal antibodies is presently the most

sensitive technique to detect the increased expression of surface molecules on

activated platelets (Harlow et al. 2002). The research involving platelet

evaluation using whole blood, washed platelets and platelet rich plasma (PRP).


6/57

However, a major limitation of these studies is to activate in vitrothat may occur

during sample processing (Michelson, 1996).

The evaluation of platelet by FC presents broad clinical applicability,

including assessing hyperreactivity and/or activation of circulating platelets;

evaluate platelet hyporeactivity in neonates with intraventricular hemorrhage;

assess MPP in several clinical conditions, evaluate the platelet-monocyte

aggregate as sensitive markers of platelet activation in acute coronary

syndromes; predict risk of the occurrence of acute and subacute ischemic

events after angioplasty; assess platelet function in diseases such as acute

coronary syndrome, stroke and angioplasty; diagnose the disease pool stock;

determine the rate of thrombopoiesis; diagnose idiopathic thrombocytopenia

and purpura posttransfusion; diagnose diseases due to deficiency of GP platelet

surface; assess reticulated platelets; determine the flow of Ca2+ platelet;

diagnose neonatal thrombocytopenia alloimmune; evaluate alloimmunization in

patients multitransfused (about 15% are due to platelet antibodies); monitor

pharmacotherapy with antagonists of GPIIb/IIIa, heparin-induced

thrombocytopenia diagnose and monitor the quality of platelet concentrates

stored in blood banks (Lazarus et al. 1995; Michelson, 1996; Michelson et al. ,

2000; BROWN & WITTWER, 2000; CASTRO & Gourley, 2010).

Platelet changes during pregnancy

During normal pregnancy, occur significant hemostatic and inflammatory

changes in women, which favor a hypercoagulable state that seeks to protect

them from a potential postpartum hemorrhage (Robb et al., 2010). Thus, there

is an increase in the concentration of various procoagulant and reduction ofnatural anticoagulants (Perry & MARTIN, 1992).

The platelet count during pregnancy may remain unchanged (Ahmed et

al. 1993), due to a lower dilution effect (SEJENY et al. 1975, CUNNINGHAM &

Pritchard 1978; SILL et al. 1985) or increase (MOR et al. 1960). It is known that

during pregnancy the blood volume increases in the first quarter and at 34

weeks, when it reaches the maximum value, about 40 to 50% higher than in

nonpregnant women (CUNNINGHAM & PRITCHARD, 1978). This increase is

caused by the expansion of plasma volume, probably due to alterations in the


7/57

renin-angiotensin-aldosterone system and eicosanoids, due to increased

concentrations of progesterone and estrogen (Mabie & Sibai, 1991). These

amendments are intended to enhance the ability of the kidneys to reabsorb

water and sodium (REMUZZI & RUGGENENTI, 1991). The hypervolemia

protects the mother and fetus from the harmful effects of the reduction in

venous return and cardiac output in the upright position during pregnancy and

blood loss during and after childbirth (CUNNINGHAM & PRITCHARD, 1978).

Studies suggest that occur during gestation platelet activation and

consumption, although its useful life does not change (Ahmed et al. 1993;

JANES & GOODALL, 1994; JANES et al. 1995). Pregnancy also provides

significant changes in the plasma membrane of platelets, such as increased

enzyme activity that catalyzes the hydrolysis of adenosine triphosphate

(ATPase), the decrease in membrane fluidity and the concentration of

cholesterol (RABINI et al. 1995).

A platelet hyperactivation, mediated by Ca2+ elevation is observed from

the 16th week of pregnancy and normalizes after six weeks of delivery.

Moreover, the susceptibility of platelets to the secretion of adenosine

triphosphate (ATP) and granules containing PF-4 and -thromboglobulin, is

progressive, suggesting an increase in turnover and activation of the platelets

(Morrison et al. 1985; Holthe et al. 2004).

Platelet changes in preeclampsia

Preeclampsia (PE), in pure form, is characterized by the appearance, in

normal pregnant women, hypertension and proteinuria after 20 weeks of

gestation. The PE is a serious complication of pregnancy and it is oftenassociated with considerable morbidity and mortality maternal-fetal (ACOG,

2002). According to the World Health Organization (2005), this disease is the

most common complication of pregnancy, occurring in 3-5% of pregnant women

worldwide. In developing countries, this percentage rises to 10%, due to

attendance often inadequate or absent, resulting in more than 60,000 maternal

deaths per year.

These disease has no cure, except for the removal of the placenta andmay develop into more serious conditions such as eclampsia, in which we


8/57

observe a neurological disorder, manifested by seizures and coma may also

occur stroke and bleeding (Gibson et al. 1982), HELLP syndrome (haemolysis,

elevated liver enzyme activity, low platelet); hemostatic serious complications

(disseminated intravascular coagulation-DIC) and renal (Sibai et al., 1993a).

The ability of blood oxygenation in areas of slow flow in pregnant patients with

PE is reduced to 50%, contributing to prematurity, low fetal weight and

increased neonatal mortality (SIEKMANN & Heilmann, 1989).

Despite numerous studies, the etiology of PE remains unclear and there

is no specific therapy for the disease. The clinical management is restricted to

the administration of antihypertensive drugs and the prevention of eclampsia

and the anticonvulsant therapy with magnesium sulfate and administration of

corticosteroids to accelerate fetal lung maturation (Grill et al. 2009). Ongoing

studies evaluating a new drug that would act in the angiogenic balance by

altering placental receptors, especially the growth factor receptor vascular

endothelial 121 (VEGFR-121). It is recognized that this therapy contribute to

moderate manifestations of PE, while that prolong the gestation until full

development of the fetus (Li et al. 2007).

A better understanding of the role of platelets in the pathogenesis of PE

would allow other therapeutic possibilities, including the use of antiplatelets

(Kazmi et al., 2011). Platelets in patients with PE can release an excess amount

of TXA2, exacerbating the disease. The administration of low-dose aspirin was

tested in pre-eclamptic pregnancy, to reduce the biosynthesis of TXA2 without

compromising prostaglandin production, assuming the prevention of

vasoconstriction and the hemostatic problems in PE (Gibson et al., 1982).

Sibai et al. (1993b) conducted a large study of 3135 normotensive

pregnant women to see if treatment with aspirin was effective in reducing theincidence of PE and maternal-fetal morbidity. Observed a significant reduction

of disease incidence, and this occurred in 69 (4.6%) of patients who received a

daily dose of 60 mg of aspirin and 94 (6.3%) of those who received placebo.

There was no difference in the occurrence of complications in newborns in both

groups, although there was an increased risk of abrupt placenta.

A multicenter study involving 9364 pregnant women evaluated the

efficacy of aspirin, administered under similar conditions to the previous study.The authors concluded that this therapy is not associated with the incidence of


9/57

intrauterine growth restriction or neonatal death. However, due to the risk of

maternal perinatal bleeding, this therapy is only justified in pregnant women at

high risk of early manifestation of the disease (CLASP, 1994).

Later, Sibai (1998) investigated alternatives to reduce the incidence and

severity of PE from the assessment made earlier randomized studies using

magnesium salts and zinc, dietary supplementation with fish oil and Ca 2+ and

low-dose aspirin. Concluded that these alternatives do not confer benefits or

they are minimal, which was later confirmed (NATIONAL HIGH BLOOD

PRESSURE EDUCATION PROGRAM WORKING GROUP ON HIGH BLOOD

PRESSURE IN PREGNANCY, 2000; ACOG, 2002; Sibai et al., 2005).

The platelet counts normally decreases in PE, but only about 18% of

patients develop thrombocytopenia (Perry & MARTIN, 1992; Ahmed et al. 1993;

JANES et al. 1995). Some mechanisms have been proposed to explain the

thrombocytopenia, as thrombin generation in the presence of circulating

immune and vascular injury, platelet aggregation and destruction mediated

immune mechanisms (PERRY & Martin, 1992).

Literature data show that hypertensive disorders in pregnancy occur with

platelet activation, increased levels of fibrinogen and adhesion molecules

derived from vascular endothelium (MACEY et al., 2010). Platelet activation

occurs in a conformational change of GPIIb/IIIa, located on the platelet surface,

favoring binding to fibrinogen, platelet aggregation and release the contents of

their granules. The degranulation contributes to placental and systemic vascular

changes, especially the release of vasoactive substances, mitogenic and

mioproliferativas such as TXA2, growth factors and platelet-derived

transformation and -thromboglobulin (JANES et al., 1995).

Although admittedly that the PE take place with inflammatory andcoagulation, platelet activation accompanied by the studies of literature are

controversial, with some showing an increase of markers of activation, while

others do not (Holthe et al., 2005; ACAR et al . Lok et al. 2,007; Robb et al.,

2010).

Changes in maternal platelet PE also can affect the newborn. A study in

pregnant women with PE and healthy pregnant women and their newborns,

showed an increased expression of CD62P, CD63, CD41, CD9 and CD36 inhealthy pregnant women and their newborns after in vitro activation of platelets


10/57

with thrombin. Newborns of women with PE, however, showed lower expression

of CD62P, CD63, CD9 and CD36. Furthermore, platelets were less responsive

to activation in vitro. These findings suggest that PE influences the expression

of platelet surface GP in pregnant women and their newborns, which may have

altered platelet function, contributing to an additional risk of bleeding in

thrombocytopenic neonates (Huhne et al., 1996).

It is important to recall the difficulty of diagnosing PE, established based

on blood pressure and proteinuria. Currently, no laboratory marker to submit a

favorable cost-effectiveness ratio was proposed for the diagnosis of this

disease. Understanding the role of platelets in the pathophysiology of PE which

results in the need for early termination of pregnancy, in most cases,

endangering the life of the mother and/or baby, is certainly of great value.

AIMS

The aim of this study was to evaluate the expression of platelet

membrane glycoproteins (P-selectin, GPIIb/IIIa, GPIIIa and GPIX) by flow

cytometry in samples from women with preeclampsia, and normotensive

nonpregnant women.

Material and methods

This research was previously approved by the Ethics Committee of the

Federal University of Minas Gerais, by the Ethics in Research of Santa Casa of

Belo Horizonte, the Research Center of Maternity Odete Valadares/Research

Ethics Committee of the Hospital Foundation the State of Minas Gerais, BeloHorizonte, the Ethics Committee in Research of Basic Health Unit Family

Guanabara/Betim and the Ethics Committee of the Hospital Municipal Odilon

Behrens, Belo Horizonte. The clarification of the objectives of the research was

done by researchers to all women in the study and signed a consent form was

obtained in all cases. A clinical record containing important data for analysis of

the results was completed in all cases.

Casuistry


11/57

In this study, the assessment of platelet surface markers included 85

women, 35 women (41.18%) with PE, of which 15 (17.65%) had the severe

form-PEG of the disease and 20 (23.53%) to mild form-PEL, 25 (29.41%) and

25 normotensive pregnant women-GN (29.41%) non-pregnant women-CNG.

The inclusion criteria for PEG group were:

a. Systolic/diastolic blood pressure greater than or equal to 160 x 110

mmHg, measured in two steps with an interval of at least 2 hours and after rest;

b. Proteinuria greater than 2g in 24-hour urine or larger than (+2) by semi-

quantitative strips method in unic urine sample;

c. Being in the third trimester of pregnancy.

The presence of some of the clinical symptoms and clinical and laboratory

findings were also considered by the obstetric team to define the diagnosis of

severe PE as:

a. Epigastric pain in the upper abdomen;

b. Vision changes;

c. Exacerbation of deep tendon reflexes, and checked two reflexes (patellar

and upper limbs);

d. Headache;

e. Behavioral changes;

f. Dyspnea and signs of pulmonary congestion;

g. Urine volume of less than 500mL or 100mL in 24 hours in 4 hours (2

steps);

h. Thrombocytopenia (platelet count


12/57

The inclusion criteria for GN group were:

a. Systolic / diastolic blood pressure at or below 120 x 80 mmHg and no

history of hypertension and PE;

b. No proteinuria;

c. Being in the third trimester of pregnancy.

The inclusion criteria for CNG group were:

a. Systolic/diastolic blood pressure at or below 120 x 80 mmHg and no

history of hypertension and PE;

b. No proteinuria;

Exclusion criteria common to all three groups were:

a. Obesity;

b. Presence of intercurrent diseases such as coagulation disorders,

cardiovascular diseases, autoimmune diseases, liver diseases, renal and

infectious, diabetes, cancer and chronic hypertension;

c. Labor advanced;

d. Presence of bleeding of any kind.

The diagnosis of PE was made by the obstetric team of the. The group

members CNG and GN were selected at the Hospital Municipal Odilon Behrens

and UBSF Guanabara/Betim. The women were matched according to age and

gestational age and belonged to the same social class.

Biological samples

Were collected 3.5 ml of blood of all women participating in the study.

The blood was collected in sodium citrate 3.2%, using vacuum system and

disposable material. Blood samples were processed within two hours after

venipuncture and the reading done immediately.

For the group of women with PE, the result of the blood test was

obtained from medical records. For groups GN and CNG, in addition to 3.5 mLof citrated blood were collected 4 ml of blood in ethylenediaminetetraacetic acid


13/57

(EDTA) to blood count. This was done using the electronic cell counter, Cell-

Dyn 3500R (Abbott Diagnostics).

Method

The study of the expression of markers of platelet activation was

performed according to the original protocol defined by Tomer (2004), Noronha

et al. (2007) and Assinger et al. (2011), brief: The sample of citrated whole

blood previously centrifuged at 800rpm for 10 minutes to obtain PRP. To fix

platelets, 400L aliquots of PRP were dripped into 1.600L of fixative solution

(paraformaldehyde 10g/L sodium cacodylate at 10.2 g/L, sodium chloride 6.63

g/L. The solution pH was adjusted to 7.2 to 7.4) under vortexing and placed in a

4 to 8 C overnight. The platelets were washed with phosphate buffered solution

(0.015 M PBS, pH = 7.2) and centrifuged for 10 minutes at 2.200rpm. The

supernatant was discarded. The pellet was resuspended gently with PBS and

platelet count was performed using the Cell-Dyn hematology counter 3500R

(Abbott Diagnostics).

The fixed platelet suspension was adjusted to 5x106 platelets/mL.

Aliquots of 100mL of this suspension (500,000 platelets) were marked with

monoclonal antibodies (Pharmingen Becton Dickinson ) directed against

platelet GP as described in Table 1. The tubes contained an internal control for

self-fluorescence (background) in which the platelet suspension was incubated

in the absence of monoclonal antibodies. These samples were homogenised by

vortexing and then incubated at room temperature and under incubation for 30

minutes. After, the samples were washed with PBS, homogenized by vortexing,

and centrifuged for 10 minutes at 2.200rpm. The supernatant was discardedand the pellet was resuspended in 200L PBS.

Table 1Monoclonal antibodies labeled with fluorochromes used for analysis ofplatelet glycoproteins.

Sample Reagent/antibody Fluorocrom Volum ClonePhenotype

target

1 PBS - 5L - -


14/57

2 Anti-CD41a PE 4L HIP8 GPIIb/IIIa

3 Anti-CD42a FITC 4L ALMA.16 GPIX

4 Anti-CD61 PE 3L VI-PL2 GPIIIa

5Anti-CD42a

Anti-CD62P

FITC

PE

4L

5L

ALMA.16

AK-4

GPIX

P-selectina

FITC: fluorescein isotiocianate; PE: ficoeritrin.

A total of 100,000 events were acquired per sample using the flow

cytometer FACScalibur (Becton Dickinson). The software CellQuest was

used for acquisition and data analysis.

The different strategies employed for analysis of platelet profile markers

are represented in Figure 1. The selective analysis of platelets was determined

by combination of fluorescence 1 (FL1) 2 or fluorescence (FL2) versusSSC.

After selecting the population of platelets were used one-dimensional

histograms of fluorescence to establish the mean fluorescence intensity (MFI)

for CD41a, CD42a, CD61 and CD61P. The percentage of CD62P+ platelet

population was defined from the combination anti-CD42a FITC versus anti-

CD62P PE. Results were expressed as percentage of platelet CD42a+/CD62P+,

corresponding to the upper right quadrant double-positive.


15/57

Figure 2Strategies used for determining the mean fluorescence intensity of markers

of platelet activation and the percentage of platelets CD62P+. Figures obtained using the

Flow Jo software.

SSC/Granulosidade

FL2/Anti-CD41a PE

FL2/Anti-CD41a PE

FL2/Anti-CD61 PE

FL2/Anti-CD61 PE

FL1/Anti-CD42a FITC

FL1/Anti-CD42a FITC

FL1/Anti-CD42a FITCFL1/Anti-CD42a FITC

FL2

/Anti-CD62PPE

Nmerodeplaquetas

Plaquetas

Plaquetas

Plaquetas

Plaquetas

IMF=197,5

IMF=978,6

IMF=410,8

5,2%94,8%


16/57

Defining the type of citrated blood sample

After completing the standardization of the technique to evaluate the

expression of markers of platelet activation and the establishment of strategy

analysis, were avalued the possible platelet activation in vitro at the time of

venipuncture. Thus, were assessed whether it would be necessary to dismiss a

small volume of blood (containing activators) before collecting the sample to be

used in the experiments.

Were collected two tubes of blood in sodium citrate (3.5 mL) of four

women, one pregnant woman with severe PE, a PE with mild and two

nonpregnant. The tubes were numbered as "1" and "2" in accordance with the

order of collecting and protocols defined been made to the two tubes in parallel.

The results showed no difference in the parameters evaluated (p> 0.05) in the

tubes "1" and "2". Thus, we chose to not despise a small volume of blood

before collection tube containing sodium citrate.

Statistical Analysis

The statistical analysis was performed using SPSS (version 13.0). Data

normality was tested by the Shapiro-Wilk method. To set up a blood sample

would be used in the first or second collection tube, were used the paired T-test

for normal data and Wilcoxon test for data that did not follow normality. For

continuous variables with normal distribution, comparison of means between

three or more groups was performed using ANOVA. When significant

differences were found in the ANOVA, were used the multiple comparison testtwo at Least Square Difference (LSD). The comparison of results between two

groups was made by Student's T-test. For continuous variables with non-normal

distribution, comparison of medians between three or more groups was

performed by Kruskal-Wallis. The comparison of results between two groups

was performed by Mann-Whitney test, with Bonferroni correction.


17/57

RESULTS

The clinical characteristics of the three groups of participants in this study

are presented in Tables 2 and 3.

Table 2Clinical characteristics of the study participants.

PE (n=35) GN (n=31) CNG (n=31) p1 p2 p3 p4

Idade (anos)

IEP (meses)

29 7

93 69

26 6

36 41

25 6

-

0,02*

-

0,06

0,16

0,01*

-

0,35

-

IMC (Kg/m ) 28 6 25 5 24 4 0,01* 0,04* 0,00* 0,38

IG (sem.) 33 3 34 4 - - 0,17 - -

GPG (Kg) 12 8 12 8 - - 0,71 - -

PA sist.

(mmHg)

160 10 110 20 120 10 0,00* 0,01* 0,00* 0,00*

PA diast.

(mmHg)

100 10 70 20 80 10 0,00* 0,00* 0,00* 0,01*

IMC: ndice de massa corporal; IEP: intervalo entre partos; IG: idade gestacional; sem.: semanas; GPG:ganho de peso na gestao; PA: presso arterial; sist.: sistlica; diast.: diastlica; PE: grupo degestantes com PE; NG: grupo de gestantes normotensas; CNG: grupo de mulheres no gestantes.*p0,05. Os dados paramtricos so representados como mdiadesvio padro (ANOVA/LSD/Teste t-Student). Os dados no-paramtricos so apresentados como medianaintervalo interquartil (Kruskal-Wallis/Mann-Whitney/Correo de Bonferroni).p1: PE x GN x CNG; p2: PE x GN; p3: PE x CNG; p4: GN x CNG.

Clinical characteristics of group participants GN and PE were obtained

from medical records and antenatal card. Of the 67 women who participated in

the study, 56 (83.6%) made at least three visits for antenatal care. Data from

the CNG group participants were obtained during the interview.

The mean age of the members of the PE group was higher compared to

the CNG (p = 0.01) and the mean body mass index (BMI), as compared with

the GN (p = 0.04 ) and CNG (p = 0.00) group.


18/57

There was no difference between the mean gestational age (GA) (p =

0.17) and the mean weight gain during pregnancy (WDP) (p = 0.71) and calving

interval (CI) (p = 0.16), compared to pregnant women with PE and GN.

The median values of systolic (p = 0.00) and diastolic blood pressure (p =

0.00) were different when comparing the three groups.

A similar analysis was performed after subdivision group severe PE

(PEG) and PE mild (PEL) as shown in Table 3.

Table 3Clinical characteristics of the study participants after subdivision of the PEgroup as severe and mild.

PEG (n=15) PEL (n=20) p p p p p p

Idade (anos)

IEP (meses)

31 6

127 66

28 8

54 52

0,02*

0,02*

0,16

0,02

0,02*

0,012**

0,00*

-

0,36

-

0,08

-

IMC (Kg/m2) 26 5 30 6 0,00* 0,03* 0,86 0,37 0,00* 0,00*

IG (sem.) 33 3 34 3 0,29 - - - - -

GPG (Kg) 14 5 11 10 0,64 - - - - -

PA sist.

(mmHg)

160 20 160 10 0,00* 0,01* 0,00* 0,00* 0,00* 0,00*

PA diast.

(mmHg)

110 20 100 11 0,00* 0,05* 0,00* 0,00* 0,00* 0,00*

IMC: ndice de massa corporal; IEP: intervalo entre partos; IG: idade gestacional; sem.: semanas; GPG: ganhode peso na gestao; PA: presso arterial; sist.: sistlica; diast.: diastlica. PEG: grupo de gestantes com PE

grave; PEL: grupo com PE leve; NG: grupo de gestantes normotensas; CNG: grupo de mulheres nogestantes.*p0,05. **p0,017 (Correo de Bonferroni). Os dados paramtricos so representados como mdiadesviopadro (ANOVA/LSD/Teste t-Student). Os dados no-paramtricos so apresentados como medianaintervalointerquartil (Kruskal-Wallis/Mann-Whitney/Correo de Bonferroni).p1: PEG x PEL x GN x CNG; p2: PEG x PEL; p3: PEG x GN; p4: PEG x CNG, p5: PEL x GN, p6: PEL x CNG.

The average age of the group members was superior in PEG than in GN

(p = 0.02) and CNG (p = 0.00) groups. The average BMI of women with severe

PE was lower than those with mild PE (p = 0.03), however, pregnant women


19/57

with mild PE had lower mean BMI compared to the GN (p = 0.00) and CNG (p =

0.00).

There was no difference between the mean gestational age (p = 0.29)

and the mean GPG (p = 0.64), compared to pregnant women with severe or

mild PE and normotensive. The median IEP with severe group were higher than

in the GN group (p = 0.012). The median values of systolic (p = 0.00) and

diastolic (p = 0.00) were different when comparing the four groups.

Other clinical parameters were evaluated and tested for their association

with the occurrence of PE. Only the variable "number of pregnancies" was

significant (p = 0.00). Residual analysis is shown in Table 4.

Table 4Residue analysis for the variable "number of pregnancies" PEG groups, PEL,NG and CNG.

Grups

Gestacional number

0 1 2

PEG -1,7 -0,5 1,7

PEL -2,1* 1,1 0,4

GN -2,8* -1,0 2,9*

CNG 5,9* 0,4 -4,5*

*Localizao da diferena: Resduo 1,96 e Resduo -1,96.PEG: grupo de gestantes com PE grave; PEL: grupo degestantes com PE leve; NG: grupo de gestantesnormotensas; CNG: grupo de mulheres no gestantes.

We found greater number of nulliparous women in the group CNG.Regarding the occurrence of two or more pregnancies was higher in the group

NG and smaller in CNG group. No group stood out in the event of a pregnancy.


20/57

Evaluation of platelet count and markers of platelet activation

In Table 5 are the results of platelet count and expression of biomarkers

of platelet activation obtained for pregnant women with PE (PEG and PEL), and

NG CNG.

Table 5Platelet count and platelet activation markers obtained for PE groups (PEGand PEL), GN and CNG.

Parameters PE (n=35) PEG (n=15) PEL (n=20) GN (n=25)#

CNG (n=25)#

N de plaquetas x

103/L

238,4

(81,7)

223,7

(94,9)

249,4

(70,9)

219,3

(53,6)

286,2

(67,0)

IMF CD41a 278,3

(178,0)

253,0

(182,5)

297,3

(176,8)

239,9

(239,5)

364,9

(426,1)

IMF CD61 903,9

(292,3)

855,7

(323,0)

940,0

(270,8)

888,0

(339,4)

1330,6

(441,1)

IMF CD42a 446,0

(155,6)

463,3

(170,0)

433,0

(147,1)

439,0

(152,3)

486,2

(108,9)

IMF CD62P 152,2

(99,4)

150,3

(55,9)

157,8

(110,6)

154,1

(126,2)

163,1

(118,3)

CD62P (%) 1,4

(2,1)

1,4

(2,0)

1,2

(2,1)

1,2

(3,9)

1,6

(4,0)

IMF: intensidade mdia de fluorescncia.#Para N de plaquetas x 103/L considerar n=31. Os dados paramtricos so representadoscomo mdia (desvio padro). Os dados no-paramtricos so apresentados como mediana(intervalo interquartil).

The Figures 3 and 4 illustrate the results for the platelet count and

expression of biomarkers of platelet activation in the PE, and GN CNG and after

subdivision of the group of women with PE in severe and mild.


21/57

Figure 3 Platelet count, mean fluorescence intensity of CD41a, CD61, CD42a, CD62P

and percentage of CD62P+

platelets in the PE, GN and CNG.

A comparison of the mean platelet count showed a lower value in the

group of women with PE compared to CNG (p = 0.01) and in the GN group

compared to CNG (p = 0.00). There was no difference between the mean PE

and GN groups (p = 0.26).

Plaquetas (/L)

PE GN CNG0

200000

400000

600000*

*

CD41a (IMF)

PE GN CNG0

500

1000

1500

*

*

CD61 (IMF)

PE GN CNG0

1000

2000

3000

*

*

CD42a (IMF)

PE GN CNG0

200

400

600

800

1000

p=0,44

CD62P (IMF)

PE GN CNG0

200

400

600

800

p=0,67

CD62P (%)

PE GN CNG0

5

10

15

p=0,38

*p0,05 ou p0,017 (Correo de Bonferroni). Para os dados paramtricos foram realizados os testes ANOVA/LSD.Para os dados no-paramtricos foram realizados os testes Kruskal-Wallis/Mann-Whitney/Correo de Bonferroni.


22/57

For the median MFI of CD41a and the average MFI of CD61 were

obtained in the group with lower values in PE compared to CNG (p = 0.00 for

both), and GN group compared to CNG (p = 0.01 and p = 0.00, respectively).

There was no difference between the groups GN and PE (p = 0.10 and p =

0.87, respectively).

A statistical comparison of the average MFI of CD42a and CD62P

medians of the IMF, as well as the percentage of CD62P+ platelets, the three

groups showed no significant difference (p = 0.44, p = 0.67 and p = 0.38 ,

respectively).

Plaquetas (/L)

PEG PEL GN CNG0

200000

400000

600000

* *

CD41a (IMF)

PEG PEL GN CNG0

500

1000

1500

*

*

CD61 (IMF)

PEG PEL GN CNG0

500

1000

1500

2000

2500

* **

CD42a (IMF)

PEG PEL GN CNG0

200

400

600

800

1000

p=0,57

CD62P (IMF)

PEG PEL GN CNG0

200

400

600

800

p=0,70

CD62P (%)

PEG PEL GN CNG0

20

40

60

80

p=0,54

*p0,05 ou p0,008 (Correo de Bonferroni). Para os dados paramtricos foram realizados os testesANOVA/LSD. Para os dados no-paramtricos foram realizados os testes Kruskal-Wallis/Mann-Whitne /Corre o de Bonferroni.


23/57

Figure 4Platelet count, mean fluorescence intensity of CD41a, CD61, CD42a, CD62P

percentage of platelets and CD62P + for PEG, PEL, GN and CNG groups.

It was observed a reduction in mean platelet count in the PEG groupcompared to CNG (p = 0.01). There was no difference between the groups

means PEL and PEG (p = 0.28), PEG, PEL and GN (p = 0.84 and p = 0.13,

respectively) and between groups PEL and CNG (p = 0, 66).

For the median MFI of CD41a expression was observed in the lowest

PEG group compared to CNG (p = 0.00). The comparison between both

groups: PEG and PEL (p = 0.48), PEG and GN (p = 0.69), PEL and GN (p =

0.73), PEL and CNG (p = 0.10) no difference.The average MFI of CD61 was lower in the PEG and PEL groups

compared to CNG (p = 0.00 for both). The comparison between both groups:

PEG and PEL (p = 0.49), PEG and GN (p = 0.79), PEL and GN (p = 0.63) did

not differ.

The comparison of the average MFI of CD42a, the medians of CD62P

and the percentage of CD62P+ platelets in the four groups did not differ (p =

0.57, p = 0.70 and p = 0.54, respectively).The Table 6 shows the ratios between the IMF markers of platelet

surface obtained for pregnant women with PE (PEG and PEL), GN and CNG.

Table 7MFI ratios between the markers of platelet surface obtained for PE groups(PEG and PEL) GN and CNG.

Paameters PE (n=35) PEG (n=15) PEL (n=20) GN (n=25) CNG (n=25)

IMF CD61/CD41a 3,6(2,3)

4,2(12,1)

3,3(1,9)

3,1(4,2)

2,8(3,8)

IMF CD61/CD42a2,0

(0,9)1,8

(0,70)2,3

(0,7)1,9

(1,1)2,7

(0,7)

IMF CD42a/CD41a1,7

(2,0)1,8

(8,0)1,6

(1,0)1,7

(1,1)1,2

(1,0)

IMF CD61/CD62P5,2

(2,6)5,4

(2,5)5,4

(2,1)4,8

(2,5)7,6

(4,2)

IMF CD42a/CD62P2,6

(2,0)3,0

(1,6)2,5

(1,6)2,5

(1,3)2,8

(1,3)


24/57

IMF CD41a/CD62P1,2

(1,5)1,6

(1,3)1,7

(1,2)1,4

(1,0)2,6

(1,6)

IMF: intensidade mdia de fluorescncia.Os dados so apresentados como mediana (intervalo interquartil).

Figures 5 and 6 illustrate the ratios of MFIs markers of platelet surface, in

the PE, GN and CNG and after subdivision of the group of women with PE

(PEG and PEL), respectively.

CD61/CD41a

PE GN CNG0

10

20

30

40

p=0,29

CD61/CD42a

PE GN CNG0

2

4

6

**

CD42a/CD41a

PE GN CNG0

5

10

15

20

*

*

CD61/CD62P

PE GN CNG0

5

10

15

20

p=0,07

CD42a/CD62P

PE GN CNG0

2

4

6

8

p=0,73

CD41a/CD62P

PE GN CNG0

2

4

6

8

*

*

*p0,05 ou p0,017 (Correo de Bonferroni). Foram realizados os testes Kruskal-Wallis/Mann-Whitney/Correode Bonferroni.


25/57

Figure 5Ratio of the mean intensities of fluorescence CD61/CD41a, CD61/CD42a,

CD42a/CD41a, CD61/CD62P, CD42a/CD62P CD41a/CD62P and PE groups, GN and CNG.

The median ratio between the IMF the CD61/CD42a and CD41a/CD62P

were lower in the PE group compared to CNG (p = 0.00 and p = 0.015,

respectively) and GN compared to CNG (p = 0.00 and p = 0.01). The

comparison of medians between groups PE and GN showed no significant

difference (p = 0.57 and p = 0.56, respectively).

For the median ratio of MFI values obtained CD42a/CD41a were higher

in the PE group compared to CNG (p = 0.00) and in the GN group compared to

CNG (p = 0.00). The comparison of medians between groups PE and GN

showed no significant difference (p = 0.57).

The median ratios of the IMF CD61/CD41a, CD61/CD62P and

CD42a/CD62P, in the three groups showed no significant difference (p = 0.29, p

= 0.07 and p = 0.73, respectively ).

CD61/CD41a

PEG PEL GN CNG0

10

20

30

40

p=0,42

CD61/CD42a

PEG PEL GN CNG0

2

4

6

* *

CD42a/CD41a

PEG PEL GN CNG0

5

10

15

20

*

*

CD61/CD62P

PEG PEL GN CNG0

5

10

15

20

p=0,14


26/57

Figure 6 Ratio of the mean intensities of fluorescence CD61/CD41a, CD61/CD42a,

CD42a/CD41a, CD61/CD62P, CD42a/CD62P CD41a/CD62P and PEG groups, PEL, GN and

CNG.

The median ratio of CD61/CD42a MFI was lower in the PEG group

compared to CNG (p = 0.00). The comparison between both groups: PEG and

PEL (p = 0.10), PEG and NG (p = 0.67), PEL and GN (p = 0.23), PEL and CNG

(p = 0.10) not significantly different.

However, for the median ratio of CD42a/CD41a MFI values were

obtained in the upper PEG group compared to CNG (p = 0.00). The comparisonbetween both groups: PEG and PEL (p = 0.10), PEG and GN (p = 0.67), PEL

and GN (p = 0.23), PEL and CNG (p = 0.10) not significantly different.

The median values of the ratio of the CD41a/CD62P MFI was higher in

the CNG group compared to GN (p = 0.01) as previously described. The

comparison between both groups: PEG and PEL (p = 0.67), PEG and GN (p =

0.94), PEG and CNG (p = 0.10), PEL and GN (p = 0.41), PEL and CNG (p =

0.06) showed no significant difference.The median ratios of the CD61/CD41a, CD61/CD62P and CD42a/CD62P

IMF in four groups showed no significant difference (p = 0.42, p = 0.14 and p =

0.69, respectively).

DISCUSSION

The PE is a complex disease whose etiology is not yet fully known. The

development of this disease is associated with an aberrant trophoblastic

CD42a/CD62P

PEG PEL GN CNG0

2

4

6

8

p=0,69

CD41a/CD62P

PEG PEL GN CNG0

2

4

6

8*


27/57

invasion in early pregnancy, which may support the hypothesis that platelets are

activated at an early stage of pregnancy. There is evidence that platelet

activation precedes the onset of clinical symptoms (FALCONER et al., 1987;

VANTROYEN & VANSTRAELEN, 2002).

Although some help tests are part of the monitoring of pregnant women

with suspected PE, the diagnosis is mainly based on clinical data, blood

pressure measurement and determination of proteinuria (GRILL et al., 2009).

The blood pressure is known to be subject to postural changes and emotional.

Proteinuria is routinely detected by dipstick or measured quantitatively in urine

sample or isolated 24 hours, tests also have limitations. Thus, a major limitation

to studies involving the PE is the difficulty of diagnosis, which can lead to

erroneous conclusions and justify the variability of results found in literature.

The new understanding of the process of coagulation and platelet

surface highlights the FT as essential to trigger the cascade sequence of

reactions that culminates in the formation of fibrin clot. Coat et al. (1992) found

that the platelet membrane of pregnant women with PE presents an abnormal

lipid composition. It is known that platelet activation is associated with the

conformational change of GPIIb/IIIa, located on the platelet surface, favoring

binding to fibrinogen, platelet aggregation and release the contents of their

granules. Thus, platelet activation could contribute to placental and systemic

vascular changes observed in PE, especially the release of vasoactive

substances, mitogenic and mioproliferativas (JANES et al., 1995).

Some studies about platelet activation in patients with PE (in the

presence and absence of proteinuria) and normotensive and nonpregnant

women concluded that pregnant women with PE and proteinuria had evidence

of platelet activation and degranulation, increased platelets attached tofibrinogen, increased expression of CD63 (marker of platelet activation) and

plasma levels of-thromboglobulin. There was a moderate correlation between

the ratio of the fibrinogen-bound platelets and CD63 expression in all groups

(JANES & GOODALL, 1994; JANES et al. 1995).

Similar investigation conducted by Harlow et al. (2002) revealed that

there was increased expression of CD62P and CD63 in the group with PE.

These researchers proposed that platelet activation is an important factor in thepathophysiology of PE, but the platelet count is not able to predict the


28/57

occurrence of the disease in all pregnant women. Bagamery et al. (2005) also

evaluated the levels of CD63, however concluded that there is greater platelet

activation in pregnant women with PE.

However, the in vitro platelet hyperactivity in PE is seemingly

contradictory, since aggregation is reduced compared to various agonists,

reflecting the depletion of platelets resulting from continuous activation. The

reduced secretion of ATP also confirms the hypothesis hyperstimulation

(Louden et al. 1991). If the platelet hyperfunction is cause or effect of PE still

requires further studies.

Lok et al. (2007) evaluated platelet activation from the release of platelet

microparticle (MPP) expressing P-selectin and the plasma concentration of

sCD62P. They found an increased concentration of sCD62P in PE, although it

did not differ from normotensive pregnant women, although the number of MPP

expressing P-selectin is higher in PE.

In further studies, Lok et al. (2008) investigated whether levels of MPP

are associated with the severity of PE. They observed that during normal

pregnancy the number of circulating initially reduces MPP, normalizing later.

The number of MPP derived from placenta samples increased gradually

throughout pregnancy in normal and PE decreased due to thrombocytopenia.

There was also an increased number of monocyte-derived microparticles in PE,

which could reflect the activation of these cells resulting from systemic

inflammatory state. Admitted that the release of inflammatory mediators from

activated platelets trigger inflammatory responses in endothelial cells and

monocytes.

Paradoxically, Robb et al. (2010) evaluated several markers of platelet

activation in PE and did not receive sharp expression of PECAM-1 (CD31),CD61, CD42a, CD62P and CD63. There was an increased expression of

sCD62P throughout normal pregnancy, as in PE and subsequent reduction in

the postpartum period. The CD31 marker was the best that was associated with

PE. They concluded that platelet activation is increased during pregnancy, but

is not altered in PE. However, it did not exclude the contribution of platelets to

the development of PE, assuming the increased activation in the maternal-fetal

interface.


29/57

Holthe et al. (2004) determined the state of platelet activation (from the

markers CD61, CD42a, CD62P, and CD63 ligand activated fibrinogen receptor

type 1-PAC-1) in patients with PE, normotensive and nonpregnant women at

baseline and after activation in vitro. Observed that in normal pregnancy or PE

is greater platelet activation, and that the increased expression of CD63 is more

significant in PE.

In subsequent studies, Holthe et al. (2005) evaluated microparticles,

platelet-platelet aggregates and expression of P-selectin. Observed that in PE

(in the presence or absence of agonist) is smaller proportion of microparticles

than in normotensive pregnant women because they are more easily removed

from circulation. The microparticles and platelets in PE expressing P-selectin in

larger quantity, which determines a procoagulant state. Because P-selectin

mediates binding between platelets, leukocytes and endothelium, their

presence is associated with thrombogenicity (Holthe et al. 2005; LOK et al.

2007).

Acar et al. (2007) investigated the association between the PE and the

expression of soluble GPV, considered a biomarker of platelet activation. They

found no difference between women with PE and normotensive, although the

platelet count was lower in the group with PE. They proposed that there is likely

a subpopulation of pregnant women with PE in which platelet activation, if it

occurs, is a secondary event. The primary cause of hypercoagulability would

stimulate the occurrence of inflammation, endothelial dysfunction and

overactive leukocytes.

It is known that the PE preferably covers the ends of childbearing age

(Gibson et al. 1982; HAWFIELD & Freedman, 2009; ZHONG et al., 2010). The

analysis of the age of the members of the study (Table 3) revealed that theaverage age of the group of nonpregnant women was lower than that of the

group with PE (p = 0.01) and the group with severe PE (p = 0.00 ). The average

age of pregnant women with severe PE was higher compared to normotensive

pregnant women (p = 0.02) (Table 4). Knowing that platelet function does not

vary in the age of the study participants (minimum 19 and maximum: 36 years),

it can be inferred that the age difference observed in the study groups did not

affect the results.


30/57

A comparison of PE pregnancy (PEG separately and PEL) and

normotensive pregnancy showed no difference between the mean gestational

age (p = 0.17 and p = 0.30, respectively). There was also no difference between

the median and normotensive pregnant women with PE compared to EPO (p =

0.16). However, the medians of EPO in severe PE group were higher than in

normotensive pregnant women (p = 0.012), which at first suggests a period

longer calving favors the occurrence of PE, being more evident in severe cases

disease (Young et al., 2010).

The average BMI of the patients with PE was higher than that of

normotensive pregnant women (p = 0.04), although included in the group of

"over-weight". Pregnant women with PE also had higher mean BMI for

nonpregnant women (p = 0.00), as was expected. The average BMI of women

with severe PE was lower than that of women with mild PE (p = 0.03).

Interestingly, women with mild PE had lower mean BMI compared to GN groups

(p = 0.00) and CNG (p = 0.00). There was no difference in the median of GPG

between the groups PE (p = 0.71), PEG and PEL (p = 0.64) compared to

normotensive pregnant women.

The presence of edema was observed, in varying degrees, in 16 patients

(48%) with PE, and +1 in 04 pregnant women (12%), +2 in 10 (30%), +3 in 01

(3%) and +4 01 (3%). However, no association was found between the intensity

of edema and severity of PE (p = 0.25). Edema was once considered a

parameter for diagnosis of PE. However, because it is a very common finding in

pregnancy, this is no longer a diagnostic criterion.

Platelet Markers

Care in the pre-analytical

The evaluation of in vitro platelet is complex, whereas the habitat of

platelets is the circulating blood. Once the blood is collected, platelets tend to

become activated and aggregate together. The ideal way to collect the blood

sample for evaluating platelets in vitro to avoid or minimize platelet activation, is

discussed. Van Ierssel et al. (2010) propose that the ideal is to disregard the

first few milliliters of blood leaving the vessel, in order to discard the tissue


31/57

factor released by puncturing the tissues and the vessel wall with the needle

collection, which could induce platelet activation. Moreover, van Berrs et al.

(2009) prioritize shorter tourniquet, preventing platelet activation, microparticle

formation and hemolysis area dammed by garrote.

In the present study, after careful evaluation of interferences in the act of

collecting platelet activation, the choice was made to collect the pipe sodium

citrate (used for evaluating platelet in vitro) through accurate puncture, in which

the blood vessel is reached immediately and the tourniquet is released as soon

as the blood begins to flow. The tubes used had the inner walls coated with

silicone, to prevent platelet activation induced by negative charge on the glass.

The tubes were centrifuged at maximum two hours after collection at low speed

(800rpm) for 10 minutes in order to pellet the erythrocytes and leukocytes and

platelets remain in suspension in the plasma. Immediately after centrifugation,

PRP was removed and added to the fixative solution.

Platelet count

The change in the number of circulating platelets during pregnancy is

controversial. Ahmed et al. (1993) showed that the number of platelets is not

changed during pregnancy. However, other studies have shown a decrease in

circulating platelets due to a dilution effect (SEJENY et al. 1975,

CUNNINGHAM & Pritchard 1978; SILL et al. 1 985) and one showed an

increase of platelet counts (MOR et al., 1960).

In the present study, the comparison of the mean platelet count showed

a lower value in pregnant women with PE compared to the group of

nonpregnant women (p = 0.01). The comparison of the average number ofplatelets in the group of normotensive pregnant women compared to

nonpregnant women showed a reduction in platelet count in normotensive

pregnant women (p = 0.00). No difference was observed comparing the mean

platelet count of pregnant women with normotensive and PE (p = 0.26) (Figure

15). To subdivide the group of women with PE (severe and mild) also observed

a reduction in mean platelet count in the group with severe PE compared to

nonpregnant women (p = 0.01) (Figure 16).


32/57

Several studies have shown a reduction in the number of circulating

platelets in pregnant women with PE compared with normotensive patients

(REDMAN et al., 1978; GILES & INGLIS, 1981; Neiger et al., 1992, Fitzgerald

et al., 1996; JREMO et al. 2000; Edelstam et al. 2001; Holthe et al. 2004;

Holthe et al. 2005; LOK et al. 2007). Redman et al. (1978) and Holthe et al.

(2004) admitted that the constant platelet activation in PE may result in an

increase in platelet consumption, which may exceed the capacity of production

of the bone marrow, resulting in reduced number of circulating platelets.

McCrae (2010) reported that a thrombocytopenia occurs in 50% of cases of PE,

however, Fitzgerald et al. (1996), admitted that thrombocytopenia is a less

common finding, occurring between 11 and 29% of pregnant women. Fallahian

& Nabaie (2005) found that the average number of platelets is significantly

reduced 3-6 weeks before delivery in pre-eclamptic pregnancy, but still within

the reference range. At delivery there is a significant reduction. Admitted even if

a mild or subclinical thrombocytopenia during the second half of pregnancy may

precede the PE, this parameter may be useful as a predictor of the disease.

However, Laskin et al. (2011) recognized that the platelet count is less sensitive

to aid the diagnosis of PE and should not be used in isolation, although it is a

simple, quick and inexpensive.

Platelet count less than 100 x 103/L is a sign of serious illness. If

delivery is not made, these levels are still decreasing and there is an increased

risk of maternal hemorrhage. It is unclear if the newborn may also develop

thrombocytopenia (NATIONAL HIGH BLOOD PRESSURE EDUCATION

PROGRAM WORKING GROUP ON HIGH BLOOD PRESSURE IN

PREGNANCY, 2000). Several mechanisms have been proposed to explain the

thrombocytopenia. Among them, the generation of thrombin in the presence ofcirculating immune and vascular injury; result of platelet aggregation, and the

resulting destruction mediated by immunological mechanisms (PERRY &

Martin, 1992).

In the present study, no reduction was obtained in the platelet count in

women with PE compared to normotensive. However, 07 patients (20%) with

PE, and 05 (33.3%) and severe PE 02 (10%) PE light, showed values below the

limit of the reference range (150 x 103

/L) whereas 04 (11.4%) had very low


33/57

levels (99, 102, 112 and 118 x 103/L). Of these, two progressed to HELLP

syndrome.

In concordance with the results obtained in this study Janes & Goodall

(1994), Konijnenberg et al. (1997a), Makuyana et al. (2002) and Santos & Son

(2004) also found no difference in platelet counts between women with PE and

normotensive pregnant women.

McCrae (2010) reported that a possible explanation for the change in

platelet count in studies involving pregnant women with PE, normotensive

pregnant women and nonpregnant women would thrombocytopenia from other

causes such as hypersplenism, autoimmune diseases and several platelet

changes. However, in this study these variables were considered as exclusion

criteria and probably did not affect the evaluation of this parameter.

A comparison of the mean platelet count of the patients with mild and

severe forms of PE in this study showed no difference (p = 0.23). In agreement

with this result, Neiger et al. (1992) and Ceyhan et al. (2006) also reported that

the platelet count did not differ in these groups.

Contrary to the results of this study, stn et al. (2007), Canzoneri et al.

(2009) and McCrae (2010), found a reduction of platelet count in the group with

severe regarding shape and light in normotensive pregnant women, suggesting

a correlation between the platelet count and severity of PE. McCrae (2010)

admitted that the intensity of thrombocytopenia is directly associated with

disease severity.

Comparison of platelet count of pregnant women with severe and mild

compared to normotensive pregnant women showed no difference (p = 0.84

and p = 0.13, respectively). Similar results were obtained by Ceyhan et al.

(2006). However, stn et al. (2007) observed a reduction in the number ofplatelets in pregnant women with severe PE compared with normotensive

patients (p = 0.00). For the mild form of the disease, these researchers found

no difference and proposed that the change in the number of platelets in this

case could be subclinical.

Comparison of platelet count in normotensive pregnant and non-pregnant

women in the present study revealed that this was lower in normotensive

pregnant women, which is consistent with several studies (SEJENY et al., 1975;GIBSON et al., 1982 ; Holthe et al. 2004; Holthe et al. 2005; CEYHAN et al. and


34/57

2006 McCrae, 2010). However, Giles & Inglis (1981) found no difference

between these groups.

A limitation of this study concerning the platelet count, that is, the group

with PE, was obtained in three different cell counters. Eleven women with PE

(31.4%) were selected in the maternity Odete Valadares, 18 (51.5%) in the

maternity Hilda Brando of Santa Casa de Belo Horizonte and 06 (17.1%) in the

Maternity Hospital Municipal Odilon Behrens. Diso addition, the platelet count

for the control groups (GN and CNG) was performed at another counter

hematological laboratory. Although the four clinical laboratories are embedded

in programs of quality control, we can not rule out differences in the accuracy of

platelet count between the automated counters used.

Markers expression of platelet activation

The importance of platelets in the pathophysiology of PE stems from the

fact that this disease is a deficiency in the production of PG and TXA2

biosynthesis excessive. There is also activation and aggregation of platelets

followed by a stimulation of the coagulation cascade resulting in the formation of

thrombi in placental vessels and vital organs maternal (CLASP, 1994;

NATIONAL HIGH BLOOD PRESSURE EDUCATION PROGRAM WORKING

GROUP ON HIGH BLOOD PRESSURE IN PREGNANCY , 2000; ACOG, 2002;

Sibai et al. 2005; YOUNG et al. 2,010; Kazmi et al. May 2011).

Markers of platelet surface

In the FC method, the data can be analyzed in two ways: on thepercentage of particles that express antigens on the surface, or by mean

fluorescence intensity (MFI) of the particle population that expresses this

antigen. The first strategy is useful when you know that the biomarker is

restricted to a subpulation and second, when dealing with constituent

biomarkers. The MFI values enable the distinction of which are at the extremes

of the range of linearity (Konijnenberg et al. 1997a, b).

The expression of surface markers results of the present study wasmade in MFI except for P-selectin, also expressed in percent. Only two studies


35/57

evaluated the expression of P-selectin by MFI (STAR et al. Konijnenberg et al.

1997). The markers of platelet activation were evaluated to P-selectin (CD62P),

GPIIb/IIIa (CD41a), GPIIIa (CD61) and GPIX (CD42a).

P-selectin (CD62P)

A statistical comparison of the medians of the MFI and CD62P

percentage of CD62P+ platelets in the three groups of women evaluated in this

study showed no significant difference (p = 0.67 and p = 0.38, respectively)

(Figure 15). After subdividing the group of pregnant women with PE as severe

and mild neither difference was obtained comparing the four groups (p = 0.70

for the IMF CD62P and p = 0.54 for platelet CD62P+) (Figure 16).

Experiments in vitro suggest that the maximal expression of P-selectin

requires stimulation with agonists such as thrombin (JANES & GOODALL,

1994). It should be noted that in this study, the expression of P-selectin was

investigated without prior platelet activation in vitro. This decision was made

starting from the premise that platelets would already be activated in vivo in PE,

as suggested by the literature (CLASP, 1994; NATIONAL HIGH BLOOD

PRESSURE EDUCATION PROGRAM WORKING GROUP ON HIGH BLOOD

PRESSURE IN PREGNANCY, 2000; Yoneyama et al . 2001; ACOG, 2002;

Sibai et al. 2005; MACEY et al. and YOUNG et al. 2,010; Kazmi et al. May

2011).

Star et al. (1997) evaluated the expression of P-selectin in normotensive

pregnant and nonpregnant women, before and after activation of platelets with

thrombin and TXA2 analogues. In the absence of agonist, no difference was

observed between the two groups, as in the present study. In the presence ofagonist, platelet activation was lower in normotensive pregnant women

compared to nonpregnant women.

In agreement with the results obtained in this study, Gatti et al. (1994),

Harlow et al. (2002) and Macey et al. (2010) found no difference in the

expression of P-selectin between normotensive pregnant and nonpregnant

women. Harlow et al. (2002) found that, on average, 0.61 1.15% of platelets in

normotensive pregnant women are CD62P+

, unlike nonpregnant women (0.43


36/57

1.13%), but this difference was not significant. These researchers found no

correlation between the intensity of platelet activation and severity of PE.

The results of this study are in agreement with those observed by Robb

et al. (2010). These researchers reported no difference in the percentage of

platelets expressing P-selectin among pregnant women with PE and

normotensive pregnant women. They concluded that platelet activation

increases during pregnancy, but is not affected by the disease.

Contrary to the results of this study, Goodall & Janes (1994),

Konijnenberg et al. (1997a), Harlow et al. (2002), Holthe et al. (2004) and

Macey et al. (2010) found an increase in platelet activation in patients with PE

compared to normotensive pregnant women and nonpregnant women.

Yoneyama et al. (2001) and Tomer (2004) also obtained conflicting results

when evaluating pregnant patients with PE and normotensive pregnant women.

Janes & Goodall (1994) evaluated the platelets before and after in vitro

stimulation with ADP and epinephrine, but the employee was the CD63 marker.

At baseline, these researchers observed a significantly reduced expression of

CD63 (about 0.25% of platelets) in nonpregnant women compared to

normotensive pregnant women (0.53%). They also reported that pregnant

women with PE had higher levels of this GP (0.65%) compared to

normotensive.

Konijnenberg et al. (1997a) found a higher percentage of platelets

expressing P-selectin in the group of pregnant women with PE (4.0%, 1.0 to

20.2) compared to normotensive pregnant women (0%, -1.0 to 5, 8). This

occurred in 04 patients (40%) normotensive and to a greater extent, 10 (100%)

in pre-eclamptic pregnancy. However, a critical analysis of this study reveals

that the results did not show uniform distribution in the two groups, with extremevalues. The expression of P-selectin showed great variation even among

healthy controls. Thus, one might question the definition of the percentage of

CD62P+ platelets considered indicative of the presence of increased platelet

activation in a given population. These researchers concluded that platelets are

more activated during pregnancy, and this activation is most evident in PE.

However, they admitted that only a subpopulation of platelets appears to be

activated. They admitted, though, that it is possible that a greater number ofplatelets are activated in PE, but these are rapidly cleared from the circulation.


37/57

They also underlined that this subpopulation of platelets is activated (even

modestly) only if the GP membrane have been previously activated by thrombin

in vitro.

In parallel to the percentage Konijnenberg et al. (1997a), we investigated

the expression of P-selectin in MFI. Reported that the average MFI for P-

selectin was 10.3, ranging from 7.5 to 11.5, in normotensive pregnant women

and 11.3, ranging from 8.0 to 15.3, in patients with PE. There was great

variability in the data groups, but there was no significant difference between

them. They concluded that this parameter is limited to indicate the presence of

platelet activation in normal pregnancy and in PE. Finally, admitted that other

mechanisms, different platelet activation may also be involved in the

pathophysiology of PE. It should be noted that these researchers present the

results only after stimulation with agonists.

Harlow et al. (2002) compared the expression of P-selectin in

normotensive pregnant women and pre-eclamptic and concluded that this was

higher in pre-eclamptic (1.35 1.22% versus 0.61 1.15%, p = 0,02). However,

the increased expression of this marker was observed in 53% of pregnant

women with PE and 25% of normotensive pregnant women, which shows that

platelet activation is not exclusive and does not occur in all pregnant women

with PE. Janes et al. (1995) using CD63 obtained results similar to those

reported by Harlow et al. (2002). These researchers concluded that it is unlikely

that platelet activation is a constant event in PE and their in vitro evaluation as a

predictor of disease has limited value.

Macey et al. (2010) obtained a percentage of CD62P+ platelets higher in

women with PE (0.8 2.6%) compared to normotensive pregnant women (0.3

0.5%), suggesting that platelets are activated in this disease. However, theextent to which platelets are activated in normal pregnancy and in PE remains

uncertain.

Yoneyama et al. (2001) obtained an increase in the percentage of

CD62P+ platelets in pregnant women with PE (7.8 1.2%) compared to

normotensive pregnant women (4.7 0.7%), differing from the results obtained

in the present study.

Goodall & Janes (1994) and Holthe et al. (2004) reported that the basallevels of CD62P, after stimulation with ADP, are high in healthy pregnant


38/57

women compared to nonpregnant women. Robb et al. (2011) also observed this

increase, even at baseline.

It should be noted that most studies in the literature using whole blood

samples with the justification that would easily activated platelets in the steps of

centrifugation, separation and attachment. Centrifugation also result in loss of

platelets, which would be even more significant for those of larger, more

activated (JANES et al. 1995). In this study, we chose to use the PRP fixed in

paraformaldehyde and sodium cacodylate to minimize possible interference of

erythrocytes in the analysis by FC. The possible activation during manipulation

of the sample was minimized by fast attachment of platelets after separation of

the PRP.

It should also be noted that P-selectin is unstable and releases from the

platelet surface rapidly, while the plasma remains in a soluble and functional but

not detectable by FC. It is known that platelets which have lost the P-selectin

may remain activated even in the circulation, although not express this more GP

(Michelson, 1996; Konijnenberg et al. 1997; Michelson et al. 2000).

GPIIb/IIIa (CD41a)

In the present study, values below the median MFI of CD41a, in pregnant

women with PE compared to nonpregnant women (p = 0.00) and in

normotensive pregnant women compared to nonpregnant women (p = 0.01) (

Figure 15). Making up the subdivision of the group of women with EP, we

observed a reduction in the median MFI of CD41a in pregnant women with

severe PE compared to nonpregnant women (p = 0.00). There was no

difference between the medians of the groups PE and normotensive pregnantwomen (p = 0.10), severe PE and mild PE (p = 0.48), severe PE and

normotensive pregnant women (p = 0.69), and mild PE pregnant normotensive

(p = 0.73), mild PE and nonpregnant women (p = 0.10) (Figure 16)

Admittedly, circulating platelets are at different stages of activation.

Platelets can be activated as a result of vascular lesions with release of FT and

activation process can be reversible or with such intensity that culminates in

platelet aggregation in injured site. Obviously, only those who still circulatingplatelets can be evaluated in vitro in peripheral blood samples for FC.


39/57

Konijnenberg et al. (1997b) proposed that is understandable that most of the

platelets are identified in PE still in the native state, not activated, since the

activated already be entrapped in platelet aggregation and no longer in the

circulation.

The results of this study, which was obtained from the smaller MFI

CD41a in the PE and severe PE could be justified by the premise Konijnenberg

et al. (1997b), previously cited.

Only three other studies that evaluated the GPIIb/IIIa inhibitors during

pregnancy were found in the literature. Contrary to the results of this work, Star

et al. (1997) and Sheu et al. (2002) found no difference in the expression of

GPIIb/IIIa when compared normotensive pregnant and nonpregnant women.

Khne et al. (1996) in turn, also found no difference in expression of GP in

patients with PE compared to normotensive pregnant women.

It should be noted that the interaction antibody-GPIIb/IIIa can be affected

by in vitro manipulation of the sample, or, more commonly, by the use of

fasteners and platelet inhibitors, which may mask the expected levels of platelet

activation (JANES et al., 1994). Differences in pre-analytical stage could explain

the discrepancy between our results and those found in the literature.

GPIIIa (CD61)

The average MFI of CD61 was lower in pregnant women with PE

compared to non-pregnant women and in normotensive pregnant women

compared to nonpregnant women (p = 0.00 for both) (Figure 15). The average

MFI of CD61 was lower in the groups with severe and mild PE compared to

nonpregnant women (p = 0.00 for both). The comparison between both groups:mild PE and severe PE (p = 0.49), severe PE and normotensive pregnant

women (p = 0.79), mild PE and normotensive pregnant women (p = 0.63)

showed no significant difference ( Figure 16). These results were similar to

those obtained for CD41a (GPIIb/IIIa) that was expected since both recognize

the same glycoprotein complex, confirming that these two biomarkers are

equivalent. The hypothesis that the platelets are identified in the FC less active,

since that would already be trapped in the activated platelet aggregate and nolonger in the circulation, may also be used to explain these findings.


40/57

Paradoxically the results obtained in the present study, Holthe et al.

(2004) observed that the density of CD61 on the platelet surface levels at

baseline and after stimulation with agonist was higher in PE compared to

normotensive pregnant women and nonpregnant women. No difference was

found between normotensive pregnant and nonpregnant women. These

researchers proposed that platelets are activated during pregnancy and more

intensely in PE, highlighting the importance of platelets in the pathogenesis of

this disease.

GPIX (CD42a)

A comparison of mean MFI of CD42a the three groups (Figure 15) and

after division of the PE group showed no difference (p = 0.44 and p = 0.57,

respectively) (Figure 16).

In accordance with the results of this study, Star et al. (1997) and Holthe

et al. (2004) also observed no difference in the expression of CD42a between

normotensive pregnant women and non-pregnant, even after activation with

agonist. Conversely, Holthe et al. (2004) had higher expression GPiX in

pregnant women with PE compared to normotensive pregnant women.

The setting is a variable that needs to be controlled because the affinity

of binding to the antibody-dependent activation can be reduced compared to

platelets unfixed. An argument in favor of immediate fixation is that platelet

activation is time dependent in vitro to various platelet surface antigens such as

CD42, CD41 and CD62P (Michelson, 1996). Unfortunately, for most markers

exocytosis, the reactions are reversible and suffer interference from sample

handling, fixing and use of inhibitors, which can mask the levels of plateletactivation expected (JANES & GOODALL, 1994). Possibly, this may be one

reason for the discrepancy between studies.

Ratio biomarkers platelet surface

The platelet GP superficies have important functions in the process of

hemostasis and coagulation, especially with regard to the interaction betweenplatelets, leukocytes, endothelial cells, collagen and fibrinogen (Lowenberg et


41/57

al., 2010). Like most markers of platelet activation in this study is constitutive,

we chose to evaluate the existence of the predominance of GP determined from

the ratio between these groups of pregnant women with PE, normotensive

pregnant women and nonpregnant women.

The median ratio of MFI CD61/CD42a (GPIIIa and GPIX) and

CD41a/CD62P (GPIIb/IIIa and P-selectin) revealed significantly lower in the

group of normotensive pregnant women compared to nonpregnant women (p =

0.00 p = 0.01). The comparison of medians between groups PE and

normotensive pregnant women showed no significant difference (p = 0.57 and p

= 0.56, respectively) (Figure 17). For the reason CD61/CD42a, there was a

median lower in the group of PE and PE severe compared to nonpregnant

women (p = 0.00 for both) (Figure 18).

Analysis of these results suggest that in pregnant women, both

normotensive and preeclamptic less likely expressing platelet GPIIIa and,

therefore, less amount of GPIIb/IIIa complex in the circulation. This result allows

further inferred that in pregnancy the platelets are mainly aggregated and

probably adhered on vascular sites.

It is known that fibrinogen levels are elevated in normal pregnancy,

integrating a number of haemostatic abnormalities resulting in a

hypercoagulable state in this physiological condition. Hypercoagulability is

increasing throughout gestation and aims to prepare the woman's body for the

hemostatic challenge associated with the delivery of the placenta and

simultaneous rupture of numerous blood vessels. The formation of clots in the

extremities of injured vasculature of the uterine wall is initiated promptly,

preventing excessive bleeding after childbirth (STIRLING et al., 1984; WANG et

al., 2002).One hypothesis to explain the activation of the GPIIb/IIIa receptors and

consequently the aggregation of platelets could be an increased plasma

fibrinogen in pregnancy (Bonnar et al. 1970). The levels are also significantly

elevated thrombin, altering the interactions of cells endotelias and increasing

the permeability of the endothelium. The formed thrombin activates platelets,

leukocytes and endothelial cells (Wang et al. 2002). Hoffman & Monroe (2007)

admitted that platelet aggregates are retained in small vessels or determine theactivation of coagulation proteins and the formation of fibrin clot. Physiologically,


42/57

a discrete fibrin deposition in small capillaries placenta is admitted as a way to

prevent rupture of the same and ensure the integrity of placental vascularization

(FLETCHER et al., 1979; CARON et al., 1990).

Assessing the median ratio of MFI CD42a/CD41a (GPIX and GPIIb/IIIa)

was higher than in group PE and severe PE compared to the group of non-

pregnant women and the group of normotensive pregnant women compared to

nonpregnant women ( p = 0.00 for both) (Figures 17 and 18).

The reasoning for the interpretation of this result is similar to that done for

CD61/CD42a and CD41a/CD62P. However, the GPIIb/IIIa is now in the

denominator, resulting in a numerically higher ratio. The predominance of

GPIIb/IIIa is understandable since it is the first to be platelet glycoprotein and is

expressed in larger quantities. This glycoprotein binds to a multitude of

adhesive molecules, is fundamental in the process of platelet aggregation

(Lowenberg et al., 2010).

A statistical comparison of the ratios of median MFI of CD61/CD41a

(GPIIIa and GPIIb/IIIa) of CD61/CD62P (GPIIIa and P-selectin) and

CD42a/CD62P (GPIX and P-selectin) before (p = 0,29, p = 0.07 and p = 0.73,

respectively) and after (p = 0.42, p = 0.14 and p = 0.69, respectively) the

subdivision of the PE group, showed no difference between groups assessed

(Figures 17 and 18). No studies in the literature that compared the ratios of

these glycoproteins.

CONCLUSION

The results obtained in this study did not demonstrate an increasedexpression of markers of platelet activation in PE. However, one can not rule

out a role of this activation in the pathophysiology of the disease, since it is

possible to admit the exacerbation of platelet activation in the maternal-fetal

interface.

Although studies involving the uterine circulation are not very common,

due to the difficulty of obtaining samples for laboratory evaluation, it has been

reported a pronounced activation of coagulation in this location (Higgins et al.,1998).


43/57

It should be noted that the discrepancies of the results of studies in the

literature concerning the expression of markers of platelet activation may result

from methodological differences. Although the assessment of markers of

platelet activation is increased in the study of PE, there is still no consensus on

the optimal method for assessing these parameters, including the definition of

anticoagulant to be used in the collection of blood sample, the sample type

(whole blood or PRP), the specification of antibodies and the use or not of

fixative in processing the sample. Another important aspect refers to the wide

variability in the general population, the expression of markers of platelet

activation, which hinders the establishment of cut-off for these parameters.

Certainly, advances in research of these markers by FC and standardization of

the pre-analytical and analytical methods will contribute to greater

understanding of platelet activation in the pathophysiology of PE.

The standardizatio

Documents

ativacao plaquetaria10set2012