- Email: [email protected]
Plasma levels of microparticles at 24 weeks of gestation do not predict subsequent pregnancy complications
Plasma levels of microparticles at 24 weeks of gestation do not predict subsequent pregnancy complications Ophira Salomon, M.D.,a Ben-Zion Katz, Ph.D.,b Rima Dardik, Ph.D.,a Tami Livnat, Ph.D.,a David M. Steinberg, Ph.D.,c Reuven Achiron, M.D.,d and Uri Seligsohn, M.D.a a Amalia Biron Research Institute of Thrombosis and Hemostasis, Sheba Medical Center, Tel Hashomer and Sackler Faculty of Medicine; b Hematology Institute, Tel Aviv Sourasky Medical Center, Tel Aviv and Sackler Faculty of Medicine; c Department of Statistics and Operations Research, Faculty of Exact Sciences; and d Department of Obstetrics and Gynecology, Sheba Medical Center, and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Objective: To discern whether plasma levels of microparticles (MPs) measured at 24 weeks of gestation predict late complications of pregnancy. Design: Secondary analysis of samples obtained prospectively. Setting: Large academic medical center. Patient(s): Two hundred sixty-two healthy women selected from 642 nulliparous women with singleton pregnancies. Intervention(s): Sampling for blood cell MPs and thrombophilias at 24 weeks of gestation and measurements of blood flow resistance in uterine, placental, and umbilical arteries at 24 and 31 to 33 weeks of gestation. Main Outcome Measure(s): Relationship between levels of MPs and late pregnancy complications, thrombophilias, and blood flow resistance. Result(s): Flow cytometry only detected MPs derived from endothelial cells (CD31þ) and platelet (CD41þ). No statistically significant correlation was found between levels of CD31þ or CD41þ MPs and subsequent occurrence of pregnancy-induced hypertension, preeclampsia, intrauterine growth restriction, or small for gestational age infants. Nor was there a statistically significant correlation with blood flow resistance parameters at 24 weeks of gestation (except for the left uterine artery) or at 31 to 33 weeks of gestation. Levels of these MPs in thrombophilic and nonthrombophilic women were similar. Conclusion(s): Levels of circulating MPs at 24 weeks of gestation had no predictive value for subsequent development of pregnancy-induced hypertension, preeclampsia, intrauterine growth restriction, or small for gestational age infants. (Fertil Steril 2009;92:682–7. 2009 by American Society for Reproductive Medicine.) Key Words: Microparticles, thrombophilia, intrauterine growth restriction, small for gestational age, preeclampsia, fetomaternal circulation
Circulating microparticles (MP) are 0.1—1.0 mm fragments that bud off endothelial cells, platelets, monocytes, granulocytes, or red blood cells after apoptosis or activation by cytokines. These MPs possess proinflammatory, procoagulant, and diverse other physiologic and pathologic properties and have become in recent years a subject of intense research. Elevated plasma levels of MPs were found in patients with coronary artery disease, thrombotic angiopathies, systemic lupus erythematosus, antiphospholipid syndrome, preeclampsia, and recurrent fetal loss (1, 2). In women with preeclampsia, plasma levels of endothelial cell–derived MPs were shown to be increased compared with women during normal pregnancy or women with gestational hypertension (3, 4). The level of granulocyte-derived MPs Received March 30, 2008; revised and accepted June 11, 2008; published online August 11, 2008. O.S. has nothing to disclose. B-Z.K. has nothing to disclose. R.D. has nothing to disclose. T.L. has nothing to disclose. D.M.S. has nothing to disclose. R.A. has nothing to disclose. U.S. has nothing to disclose. Reprint requests: Ophira Salomon, M.D., Amalia Biron Thrombosis and Hemostasis Research Institute, Sheba Medical Center, Tel-Hashomer, Israel (FAX: 9723-535-1568; E-mail: [email protected]).
682
was also increased in preeclampsia (5), and an excess of syncytiotrophoblast-derived MPs was observed in women with early onset preeclampsia (6). In contrast, in women with preeclampsia (7, 8), the decreased levels of platelet-derived MPs were attributed to their consumption by fibrin deposited in abnormal placental vasculature (8). One study examined the effect of MPs isolated from patients with preeclampsia on endothelial cell function in an ex vivo model. This study showed that myometrial arteries obtained during caesarian section from healthy women that were exposed to MPs caused an impaired endothelial-dependent relaxation by bradykinin (9). Another study of omental arteries harvested during caesarian sections and of aortic rings or mesenteric arteries in mice revealed a complex response after exposure to MPs from women with preeclampsia; leukocyte-derived MPs had a vasoconstrictive effect whereas platelet MPs induced a vasodilator effect (10). In women with recurrent fetal loss, two studies described an increased number of circulating MPs measured as procoagulants by a capture on immobilized annexin 5 assay (11) or by endothelial cell markers evaluated by flow cytometry (12). Because the sampling of MPs in both studies was at least 2 to 4 months after fetal loss, it was suggested that an elevated
Fertility and Sterility Vol. 92, No. 2, August 2009 Copyright ª2009 American Society for Reproductive Medicine, Published by Elsevier Inc.
0015-0282/09/$36.00 doi:10.1016/j.fertnstert.2008.06.015
number of MPs may represent an acquired chronic thrombophilia that is associated with complications of pregnancy and that such a finding may have prognostic significance (11, 13). Our study examined a cohort of nulliparous women from the 24 week of gestation until after delivery to determine whether an increased level of MPs from various blood cells could predict pregnancy-induced hypertension (PIH), preeclampsia (PE), intrauterine growth restriction (IUGR), or small for gestational age (SGA) newborns. We also examined the relationship between the level of MPs and blood flow resistance in the fetomaternal circulation and the common thrombophilias. MATERIAL AND METHODS Patients, Definitions, and Tests We selected 262 women whose mean age was 28 years (standard deviation [SD] ¼ 2.8; range: 18 to 38 years) from a cohort of 642 women who had been prospectively studied for the relationship between common thrombophilias (factor V Leiden, prothrombin G20210A, increased titer of anticardiolipin, lupus anticoagulant) and pregnancy complications (14). The original study had included consecutive unrelated, healthy, nulliparous women with spontaneous singleton pregnancies. All women were followed from 24 weeks of gestation until after delivery. The frequencies of IUGR, SGA, PIH, or PE, as defined in our previous study (14), were determined. Fetal weight was estimated by ultrasound examinations at 24 weeks and at 31 to 33 weeks of gestation. We defined IUGR as 10th percentile or less of the estimated weight of fetuses of the entire study group of 642 women. We defined SGA as birth weight at the 10th percentile or less of the whole study group. Preeclampsia was defined as systolic blood pressure R140 mm Hg or diastolic blood pressure R90 mm Hg on at least two occasions, and when proteinuria of R300 mg per day was found. Blood flow resistance was measured in the left and right uterine, placental, and umbilical arteries at 24 and at 31 to 33 weeks of gestation by multigate Doppler ultrasonography (15). The test results for common hereditary or acquired thrombophilias in the present study were extracted from the data of our previous study (14). The institutional review board of Sheba Medical Center approved the study, and informed consent was obtained from all of the participants. Analysis of Microparticles by Flow Cytometry At 24 weeks of gestation, blood was obtained from the antecubital vein by using plastic syringes and 21 gauge needles and was introduced into tubes containing 3.2% buffered sodium citrate. The blood samples were centrifuged within 3 hours at 1500 g for 10 minutes at room temperature; the supernatants were centrifuged again at 10,000 g for 3 minutes. This yielded platelet-poor plasma, which was frozen in aliquots at 80 C until analyzed. The levels of MPs derived from endothelial cells, platelets, monocytes, granulocytes, and red cells were measured by flow cytometry. Samples were labeled by 5 mL of the following monoclonal antibodies: CD31 (PECAM-1) (IQ Products, Groningen, the NetherFertility and Sterility
lands) as a marker of endothelial cells; CD41 (aIIb) (IQ Products) as a marker of platelets; CD14 (endotoxin receptor) (IQ Products) as a marker of monocytes; CD45 (Beckman Coulter, Inc., Marseille, France) as a marker of leukocytes; CD235a (glycophorin A) (Beckman Coulter, Inc., Marseille, France) as a marker of red cells; and an anti-tissue factor monoclonal antibody (American Diagnostics, Stanford, CT). The monoclonal antibodies were conjugated with fluorescein isothiocyanate, phycoerythrin, or allophycocyanin. The presence of phosphatidylserine in MPs was examined by two fluorescein-conjugated annexin 5 preparations (IQ Products; and Bender Med Systems, Vienna, Austria). Beads with an average diameter of 3 mm (Sigma, St. Louis, MO) were added to samples before staining with the monoclonal antibodies for verification of size and amount of MPs. Samples were incubated for 15 minutes with the monoclonal antibodies and then were diluted with 0.5 mL of phosphatebuffered saline. Flow cytometry acquisition was performed by FACS Calibur (Becton-Dickinson, San Jose, CA), at a rate of 150 to 300 events per second, and the data were analyzed using CellQuest software (Becton-Dickinson). Daily calibration with fluorescent beads was carried out to ensure that fluctuations were less than 2%. Also, isotype-matched control values for each fluorophore were subtracted from examined samples. Levels of MPs positive for each marker were expressed as a percentage of all MPs measured. Flow cytometry analyses were performed by one of the authors (BZK), who was blinded to the clinical and laboratory data. Analysis of Procoagulant Function of MPs Thrombin generation (TG) was measured with a fluorometer (Fluoroskan Ascent; Labsystem, Helsinki, Finland) as previously described elsewhere (16). The MPs were isolated from the platelet-poor plasma of women at 24 weeks of gestation by ultracentrifugation at 200,000 g for 2 hours at 4 C. Supernatants were discarded, and the pellets were dissolved in 100 mL of a reaction buffer containing 20 mM HEPES, 140 mM NaCl at pH 7.3, and 5 mg/mL of bovine serum albumin (BSA) (Sigma). For the assay of TG, 20 mL of MPs suspension or 20 mL of reaction buffer were added to 80 mL of factor XI–deficient plasma in microtiter plates (Greiner Bio-1, Frickenhausen, Germany) in the presence of 1.75 mg/mL recombinant factor VIIa (rFVIIa) (NovoNordisk, Copenhagen, Denmark). The reactions were initiated by adding 20 mL of HEPES/BSA containing 100 mM CaCl2 and 5 mM fluorogenic substrate Z-GGR-AMC (Bachem, Bubendorf, Switzerland). Each sample was tested in duplicates on at least three occasions. Thrombin generation was expressed by endogenous thrombin potential (ETP) (the area under the thrombin generation curve), by the lag time for initial TG, and by the height of thrombin peak. These TG parameters were derived from a plot that was recorded by a computer program attached to the fluorometer (16). Statistical Methods Levels of MPs were adjusted for asymmetric distribution and for differences related to the day of laboratory measurement. 683
FIGURE 1 Flow cytometry plots of microparticles (MPs). The MPs were directly labeled with phycoerythrin conjugated antiCD41 monoclonal antibodies (samples 1–3, left panel) or fluorescein isothiocyanate conjugated anti-CD31 monoclonal antibodies (samples 4–6, right panel). Each panel depicts representative samples from patients containing CD41þ (samples 2, 3) or CD31þ (samples 5, 6) MPs as well as negative samples (samples 1, 4). The percentage of positive MPs is depicted in each micrograph. Isotype control ranges are marked by dashed lines.
Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
The adjustments were made by first converting the positive MPs percentages to a logarithmic scale, subtracting off daily averages on that scale, and adding the overall log scale average. The adjusted MP values on the log scale were converted back to the original scale for presentation in graphs. For example, consider the result on MPs positive for CD31 for one of the women who was among those in the first flow cytometry 684
Salomon et al.
group of samples analyzed. This woman had a measurement of 0.8% positive MPs corresponding to 0.10 on the log scale. The average logarithm of percentage of CD31þ MPs among all women in the first group was 0.1798, whereas the overall average for all women in the study was 0.2102. The adjusted log measurement for this woman was thus 0.10 (0.1798) þ 0.2102 ¼ 0.29. The result for this woman was adjusted
Microparticles and complications of pregnancy
Vol. 92, No. 2, August 2009
FIGURE 2 Thrombin generation (TG) induced by concentrated microparticles (MPs). (A) Representative curves of TG induced in factor XI–deficient plasma containing rFVIIa (1.75 mg/mL) in the absence or presence of MPs. No TG is observed without addition of MPs (flat curve). Addition of CD31þ MPs, CD41þ MPs, or MPs negative for both CD31 and CD41 from three women, respectively, induced indistinguishable pronounced TG. (B) Endogenous thrombin potential (ETP) and peak height (mean SE) in four women negative for CD41 and CD31, four women with CD 41þ MPs, and four women with CD31þ MPs. Notably, TG was abolished in all samples by annexin 5 (1.25 mg/mL).
A
100
Positive CD 31 MPs Positive CD 41 MPs
90
Thrombin(nM)
80
Negative CD 41/CD 31 MPs
70 60 50 40
No MPs
30 20 10 0
0
10
20
30
40
50
60
70
Time (min)
nM/min
B 2000 1800 1600 1400 1200 1000 800 600 400 200 0
140 120
ETP
CD41/31 negative
CD41/31 negative +annexin
CD41 positive
CD41 positive +annexin
CD31 positive
CD31 positive +annexin
Peak height
nM
100 80 60 40 20 0 CD41/31 negative
CD41/31 negative +annexin
CD41 positive
CD41 positive +annexin
CD31 positive
CD31 positive +annexin
Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
upward by 0.39 to account for the generally low measurements in the first measurement group by converting back to the original scale; the adjusted measurement was 100.29 ¼ 1.95. These adjusted levels were compared across different groups of women using nonparametric tests: Wilcoxon for two groups and Kruskal-Wallis for more than two groups. Pearson correlation coefficients were computed relating the adjusted MP levels to blood flow resistance measurement. Regression analysis was used to test for a possible relationship of MPs levels to newborn weight, adjusting for gestational age at birth, gender, and blood flow resistance in the left uterine artery at 24 weeks of gestation. Women who developed preFertility and Sterility
eclampsia or had a lupus anticoagulant or a high titer of anticardiolipin were oversampled to increase their proportion in the sample. We defined IUGR and SGA as being in the lowest weight deciles at 31 to 33 weeks of gestation and at birth, respectively, after applying regression analysis to adjust weight for gestational age, sex, and smoking to the entire cohort of 642 women from our previous study (14). RESULTS The MPs that were negative for the specific markers examined by flow cytometry were indistinguishable from isotype control values, whereas MPs that were positive for markers 685
TABLE 1 The relationship between complications of pregnancy and percentage of CD31D or CD41D microparticles after adjustment for daily differences. Adjusted CD31a
Adjusted CD41a
Complication
n
Median
Quartiles
P
n
Median
Quartiles
P
No PIH or PE PIH PE No IUGR IUGR No SGA SGA
227 13 22 209 32 234 26
1.51 1.48 0.97 1.48 1.55 1.48 1.48
0.65, 4.07 0.52, 3.89 0.32, 1.62 0.66, 4.90 0.58, 2.16 0.66, 3.98 0.60, 2.95
.19
226 14 22 208 33 234 26
1.78 1.07 1.23 1.95 1.43 1.76 1.57
0.77, 4.74 0.65, 3.36 0.73, 3.06 0.81, 5.20 0.64, 2.43 0.73, 4.63 0.82, 3.36
.42
.23 .73
.10 .92
Abbreviations: IUGR, intrauterine growth restriction; PE, preeclampsia; PIH, pregnancy induced hypertension; SGA, small for gestational age. a Median relates to percentages of microparticles adjusted to a daily average on a log scale. Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
(e.g., CD31 and CD41) displayed increased fluorescence intensity (Fig. 1). In most plasma samples examined, MPs were negative for annexin 5, tissue factor, CD45, CD235a, and CD14 (not shown). The annexin 5 negativity supposedly indicated that the MPs did not harbor phosphatidylserine. However, when MPs from women who were CD31 positive, CD41 positive, or CD31 and CD41 negative were concentrated by ultracentrifugation, they displayed normal TG by an assay that is sensitive to the presence of phosphatidylserine (Fig. 2) (16, 17). Adding annexin 5 to these samples abolished TG, thus proving the presence of phosphatidylserine. These findings indicated that MPs exhibited a procoagulant function and that phosphatidylserine was present on MPs after all but at very low concentrations undetectable by flow cytometry. Because the flow cytometry analyses disclosed that MPs were negative for CD14, CD45, CD235a, anti-tissue factor, and annexin 5, analyses were only based on CD31 and CD41 markers. The adjusted levels of CD31þ MPs (of endothelial origin) and the levels of CD41 (of platelet origin) in women with IUGR, SGA, PIH, or PE showed no statistically significant difference from the levels in women who did not develop these complications (Table 1). Blood flow resistance measured at 24 weeks of gestation in the placental, umbilical, or right uterine artery had no statistically significant correlation with the levels of MPs (data not shown). In contrast, blood flow resistance in the left uterine artery was negatively correlated with the levels of CD31þ MPs (r ¼ 0.18, P¼.005). However, blood flow resistance measurements in all blood vessels examined at 31 to 33 weeks of gestation showed no statistically significant correlation with the levels of CD31þ or CD41þ MPs sampled at 24 weeks of gestation. The levels of CD31þ and CD41þ MPs at 24 weeks of gestation showed no statistically significant difference between women with and without inherited or acquired thrombo686
Salomon et al.
philias—that is, factor V Leiden, prothrombin G20210A, or antiphospholipid antibodies (Fig. 3). DISCUSSION In a previous study of 642 nulliparous women with singleton pregnancies, we found no significant relationship between common acquired or inherited thrombophilias and the occurrence of PIH, PE, IUGR, or SGA or with resistance to blood flow in the fetomaternal compartment (14). In the present study, we examined a subset of our original study group to determine whether elevated levels of circulating MPs at 24 weeks of gestation might predict the development of the aforementioned adverse outcomes. Our reasoning was that MPs, which shed from platelets, granulocytes, monocytes, or syncytiotrophoblasts after inflammatory cytokine stimulation or apoptosis, can aggravate vascular dysfunction and promote thrombosis by expressing tissue factor, phosphatidylserine, and mediators of inflammation (for review, see reference 2). Moreover, increased levels of MPs assessed in women several months after recurrent fetal loss suggested that this finding may reflect a chronic hypercoagulable state (11). Our flow cytometry analyses revealed that the only discernible MPs were of endothelial cell and platelet origin (syncytiotrophoblast MPs were not measured). Levels of CD31þ and CD41þ MPs of endothelial cell and platelet origins, respectively, had no statistically significant relationship with the occurrence of PIH, PE, IUGR, or SGA, nor was there a statistically significant relationship with the acquired or inherited thrombophilias or with blood flow resistance in the fetomaternal compartments at 24 and 31 to 33 weeks of gestation. A dubiously significant relationship between MPs levels and blood flow resistance in the left uterine artery was observed at 24 weeks but not at 31 to 33 weeks of gestation. One limitation of our study was the rather small number of patients with pregnancy complications who were examined.
Microparticles and complications of pregnancy
Vol. 92, No. 2, August 2009
FIGURE 3 Relationship between levels of CD31þ or CD41þ MPs and inherited or acquired thrombophilia. Levels of CD31þ MPs are shown in the upper panel, and CD41þ MPs are shown in the lower panel. Levels of MPs are expressed as log percent after adjusting for day to day differences.
Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
Nevertheless, the consecutive prospective nature of our study removed potential sampling bias. Another limitation was our inability to retest the patients at the time of pregnancy complications. It is notable that the relationship between levels of MPs during normal pregnancies or complicated pregnancies is not at all clear. The MPs tested during normal pregnancy by a procoagulant assay have been shown to be increased (8) or decreased (18); in some studies, the level of MPs of platelet origin have been found to be similar in preeclamptic patients and women during normal pregnancy (3–5) but were found to be decreased in other studies (7, 8); and levels of MPs of endothelial cell origin in women with preeclampsia have been shown to be either increased (3, 4) or normal (5, 8). Our study does not explain these discrepancies because we measured the levels of MPs at the 24th week of gestation, Fertility and Sterility
not at the time of pregnancy complication. Measurement of MPs at 24 weeks of gestation does not predict subsequent development of PIH, PE, IUGR, or SGA. REFERENCES 1. Morel O, Toti F, Hugel B, Bakouboula B, Camoin-Jau L, DignatGeorge F, et al. Procoagulant microparticles disturbing the vascular homeostasis equation? Arterioscler Thromb Vasc Biol 2006;26:2594–604. 2. Lynch SF, Ludlam CA. Plasma microparticles and vascular disorders. Br J Haematol 2007;137:36–48. 3. Gonzalez-Quintero VH, Jimenez JJ, Jy W, Mauro LM, Hortman L, O’Sullivan MJ, et al. Elevated plasma endothelial microparticles in preeclmpsia. Am J Obstet Gynecol 2003;189:589–93. 4. Gonzalez-Quintero VH, Smarkusky LP, Jimenez JJ, Mauro LM, Jy W, Hortsman LL, et al. Elevated plasma endothelial microparticles: preeclampsia versus gestational hypertension. Am J Obstet Gynecol 2004;191:1418–24. 5. VanWijk MJ, Nieuwland R, Boer K, van der Post JAM, vanBavel EF, Sturk A. Microparticle subpopulations are increased in preeclampsia: possible involvement in vascular dysfunction? Am J Obstet Gynecol 2002;187:450–6. 6. Goswami D, Tannetta DS, Magee LA, Fuchisawa A, Redman CW, Sargent IL, et al. Excess syncytiotrophoblast microparticles shedding is a feature of early-onset pre-eclampsia, but not normotensive intrauterine growth restriction. Placenta 2006;27:56–61. 7. Harlow FH, Brown MA, Brighton TA, Smith SL, Trickett AE, Kwan YL, et al. Platelet activation in the hypertensive disorders of pregnancy. Am J Obstet Gynecol 2002;187:688–95. 8. Bretelle F, Sabatier F, Desprez D, Camoin L, Grunebaum L, Combes V, et al. Circulating microparticles: a marker of procoagulant state in normal pregnancy and pregnancy complicated by preeclampsia or intrauterine growth restriction. Thromb Haemost 2003;89:486–92. 9. VanWijk MJ, Svedas E, Boer K, Nieuwland R, vanBavel EF, Kublickiene KR. Isolated microparticles, but not whole plasma, from women with preeclampsia impair endothelium-dependent relaxation in isolated myometrial arteries from healthy pregnant women. Am J Obstet Gynecol 2002;187:1686–93. 10. Meziani F, Tesse A, David E, Martinez MC, Wangesteen R, Schneider F, et al. Shed membrane particles from preeclamptic women generate vascular wall inflammation and blunt vascular contractility. Am J Pathol 2006;169:1473–83. 11. Laude I, Rongieres-Bertrand C, Boyer-Neumann C, Wolf M, Mairovitz V, Hugel B, et al. Circulating procoagulant microparticles in women with unexplained pregnancy loss: a new insight. Thromb Haemost 2001;85:18–21. 12. Carp H, Dardik R, Lubetsky A, Salomon O, Eskaraev R, Rosenthal E, et al. Prevalence of circulating procoagulant microparticles in women with recurrent miscarriage: a case control study. Hum Reprod 2004;19:191–5. 13. Greer IA. Thrombophilia: implications for pregnancy outcome. Thromb Res 2003;109:73–81. 14. Salomon O, Seligsohn U, Steinberg DM, Zalel Y, Lerner A, Rosenterg N, et al. The common prothrombotic factors in nulliparous women do not compromise blood flow in the feto-maternal circulation and are not associated with preeclampsia or intrauterine growth restriction. Am J Obstet Gynecol 2004;191:2002–9. 15. Yagel S, Anteby EY, Shen O, Cohen SM, Friedman Z, Achiron R. Simultaneous multigate spectral Doppler imaging of the umbilical artery and placental vessels: novel ultrasound technology. Ultrasound Obstet Gynecol 1999;14:256–61. 16. Livnat T, Zivelin A, Martinowitz U, Salomon O, Seligsohn U. Prerequisites for recombinant factor VIIa-induced thrombin generation in plasma deficient in factor VIII, IX or XI. J Thromb Haemost 2006;4:192–200. 17. Livnat T, Martinowitz A, Zivelin A, Seligsohn U. Effects of factor VIII inhibitor bypassing activity (FEIBA), recombinant factor VIIa or both on thrombin generation in normal and haemophilia A plasma. Haemophilia. Published online March 25, 2008. 18. Desprez D, Zobairi F, Aucouturier JS, Leymarie F, Freyssinet JM, Grunebaum L, et al. Evolution of circulating procoagulant microparticles during normal pregnancy. Blood Coag Fibrinol 2008;19:179–81.
687
Objective: To discern whether plasma levels of microparticles (MPs) measured at 24 weeks of gestation predict late complications of pregnancy. Design: Secondary analysis of samples obtained prospectively. Setting: Large academic medical center. Patient(s): Two hundred sixty-two healthy women selected from 642 nulliparous women with singleton pregnancies. Intervention(s): Sampling for blood cell MPs and thrombophilias at 24 weeks of gestation and measurements of blood flow resistance in uterine, placental, and umbilical arteries at 24 and 31 to 33 weeks of gestation. Main Outcome Measure(s): Relationship between levels of MPs and late pregnancy complications, thrombophilias, and blood flow resistance. Result(s): Flow cytometry only detected MPs derived from endothelial cells (CD31þ) and platelet (CD41þ). No statistically significant correlation was found between levels of CD31þ or CD41þ MPs and subsequent occurrence of pregnancy-induced hypertension, preeclampsia, intrauterine growth restriction, or small for gestational age infants. Nor was there a statistically significant correlation with blood flow resistance parameters at 24 weeks of gestation (except for the left uterine artery) or at 31 to 33 weeks of gestation. Levels of these MPs in thrombophilic and nonthrombophilic women were similar. Conclusion(s): Levels of circulating MPs at 24 weeks of gestation had no predictive value for subsequent development of pregnancy-induced hypertension, preeclampsia, intrauterine growth restriction, or small for gestational age infants. (Fertil Steril 2009;92:682–7. 2009 by American Society for Reproductive Medicine.) Key Words: Microparticles, thrombophilia, intrauterine growth restriction, small for gestational age, preeclampsia, fetomaternal circulation
Circulating microparticles (MP) are 0.1—1.0 mm fragments that bud off endothelial cells, platelets, monocytes, granulocytes, or red blood cells after apoptosis or activation by cytokines. These MPs possess proinflammatory, procoagulant, and diverse other physiologic and pathologic properties and have become in recent years a subject of intense research. Elevated plasma levels of MPs were found in patients with coronary artery disease, thrombotic angiopathies, systemic lupus erythematosus, antiphospholipid syndrome, preeclampsia, and recurrent fetal loss (1, 2). In women with preeclampsia, plasma levels of endothelial cell–derived MPs were shown to be increased compared with women during normal pregnancy or women with gestational hypertension (3, 4). The level of granulocyte-derived MPs Received March 30, 2008; revised and accepted June 11, 2008; published online August 11, 2008. O.S. has nothing to disclose. B-Z.K. has nothing to disclose. R.D. has nothing to disclose. T.L. has nothing to disclose. D.M.S. has nothing to disclose. R.A. has nothing to disclose. U.S. has nothing to disclose. Reprint requests: Ophira Salomon, M.D., Amalia Biron Thrombosis and Hemostasis Research Institute, Sheba Medical Center, Tel-Hashomer, Israel (FAX: 9723-535-1568; E-mail: [email protected]).
682
was also increased in preeclampsia (5), and an excess of syncytiotrophoblast-derived MPs was observed in women with early onset preeclampsia (6). In contrast, in women with preeclampsia (7, 8), the decreased levels of platelet-derived MPs were attributed to their consumption by fibrin deposited in abnormal placental vasculature (8). One study examined the effect of MPs isolated from patients with preeclampsia on endothelial cell function in an ex vivo model. This study showed that myometrial arteries obtained during caesarian section from healthy women that were exposed to MPs caused an impaired endothelial-dependent relaxation by bradykinin (9). Another study of omental arteries harvested during caesarian sections and of aortic rings or mesenteric arteries in mice revealed a complex response after exposure to MPs from women with preeclampsia; leukocyte-derived MPs had a vasoconstrictive effect whereas platelet MPs induced a vasodilator effect (10). In women with recurrent fetal loss, two studies described an increased number of circulating MPs measured as procoagulants by a capture on immobilized annexin 5 assay (11) or by endothelial cell markers evaluated by flow cytometry (12). Because the sampling of MPs in both studies was at least 2 to 4 months after fetal loss, it was suggested that an elevated
Fertility and Sterility Vol. 92, No. 2, August 2009 Copyright ª2009 American Society for Reproductive Medicine, Published by Elsevier Inc.
0015-0282/09/$36.00 doi:10.1016/j.fertnstert.2008.06.015
number of MPs may represent an acquired chronic thrombophilia that is associated with complications of pregnancy and that such a finding may have prognostic significance (11, 13). Our study examined a cohort of nulliparous women from the 24 week of gestation until after delivery to determine whether an increased level of MPs from various blood cells could predict pregnancy-induced hypertension (PIH), preeclampsia (PE), intrauterine growth restriction (IUGR), or small for gestational age (SGA) newborns. We also examined the relationship between the level of MPs and blood flow resistance in the fetomaternal circulation and the common thrombophilias. MATERIAL AND METHODS Patients, Definitions, and Tests We selected 262 women whose mean age was 28 years (standard deviation [SD] ¼ 2.8; range: 18 to 38 years) from a cohort of 642 women who had been prospectively studied for the relationship between common thrombophilias (factor V Leiden, prothrombin G20210A, increased titer of anticardiolipin, lupus anticoagulant) and pregnancy complications (14). The original study had included consecutive unrelated, healthy, nulliparous women with spontaneous singleton pregnancies. All women were followed from 24 weeks of gestation until after delivery. The frequencies of IUGR, SGA, PIH, or PE, as defined in our previous study (14), were determined. Fetal weight was estimated by ultrasound examinations at 24 weeks and at 31 to 33 weeks of gestation. We defined IUGR as 10th percentile or less of the estimated weight of fetuses of the entire study group of 642 women. We defined SGA as birth weight at the 10th percentile or less of the whole study group. Preeclampsia was defined as systolic blood pressure R140 mm Hg or diastolic blood pressure R90 mm Hg on at least two occasions, and when proteinuria of R300 mg per day was found. Blood flow resistance was measured in the left and right uterine, placental, and umbilical arteries at 24 and at 31 to 33 weeks of gestation by multigate Doppler ultrasonography (15). The test results for common hereditary or acquired thrombophilias in the present study were extracted from the data of our previous study (14). The institutional review board of Sheba Medical Center approved the study, and informed consent was obtained from all of the participants. Analysis of Microparticles by Flow Cytometry At 24 weeks of gestation, blood was obtained from the antecubital vein by using plastic syringes and 21 gauge needles and was introduced into tubes containing 3.2% buffered sodium citrate. The blood samples were centrifuged within 3 hours at 1500 g for 10 minutes at room temperature; the supernatants were centrifuged again at 10,000 g for 3 minutes. This yielded platelet-poor plasma, which was frozen in aliquots at 80 C until analyzed. The levels of MPs derived from endothelial cells, platelets, monocytes, granulocytes, and red cells were measured by flow cytometry. Samples were labeled by 5 mL of the following monoclonal antibodies: CD31 (PECAM-1) (IQ Products, Groningen, the NetherFertility and Sterility
lands) as a marker of endothelial cells; CD41 (aIIb) (IQ Products) as a marker of platelets; CD14 (endotoxin receptor) (IQ Products) as a marker of monocytes; CD45 (Beckman Coulter, Inc., Marseille, France) as a marker of leukocytes; CD235a (glycophorin A) (Beckman Coulter, Inc., Marseille, France) as a marker of red cells; and an anti-tissue factor monoclonal antibody (American Diagnostics, Stanford, CT). The monoclonal antibodies were conjugated with fluorescein isothiocyanate, phycoerythrin, or allophycocyanin. The presence of phosphatidylserine in MPs was examined by two fluorescein-conjugated annexin 5 preparations (IQ Products; and Bender Med Systems, Vienna, Austria). Beads with an average diameter of 3 mm (Sigma, St. Louis, MO) were added to samples before staining with the monoclonal antibodies for verification of size and amount of MPs. Samples were incubated for 15 minutes with the monoclonal antibodies and then were diluted with 0.5 mL of phosphatebuffered saline. Flow cytometry acquisition was performed by FACS Calibur (Becton-Dickinson, San Jose, CA), at a rate of 150 to 300 events per second, and the data were analyzed using CellQuest software (Becton-Dickinson). Daily calibration with fluorescent beads was carried out to ensure that fluctuations were less than 2%. Also, isotype-matched control values for each fluorophore were subtracted from examined samples. Levels of MPs positive for each marker were expressed as a percentage of all MPs measured. Flow cytometry analyses were performed by one of the authors (BZK), who was blinded to the clinical and laboratory data. Analysis of Procoagulant Function of MPs Thrombin generation (TG) was measured with a fluorometer (Fluoroskan Ascent; Labsystem, Helsinki, Finland) as previously described elsewhere (16). The MPs were isolated from the platelet-poor plasma of women at 24 weeks of gestation by ultracentrifugation at 200,000 g for 2 hours at 4 C. Supernatants were discarded, and the pellets were dissolved in 100 mL of a reaction buffer containing 20 mM HEPES, 140 mM NaCl at pH 7.3, and 5 mg/mL of bovine serum albumin (BSA) (Sigma). For the assay of TG, 20 mL of MPs suspension or 20 mL of reaction buffer were added to 80 mL of factor XI–deficient plasma in microtiter plates (Greiner Bio-1, Frickenhausen, Germany) in the presence of 1.75 mg/mL recombinant factor VIIa (rFVIIa) (NovoNordisk, Copenhagen, Denmark). The reactions were initiated by adding 20 mL of HEPES/BSA containing 100 mM CaCl2 and 5 mM fluorogenic substrate Z-GGR-AMC (Bachem, Bubendorf, Switzerland). Each sample was tested in duplicates on at least three occasions. Thrombin generation was expressed by endogenous thrombin potential (ETP) (the area under the thrombin generation curve), by the lag time for initial TG, and by the height of thrombin peak. These TG parameters were derived from a plot that was recorded by a computer program attached to the fluorometer (16). Statistical Methods Levels of MPs were adjusted for asymmetric distribution and for differences related to the day of laboratory measurement. 683
FIGURE 1 Flow cytometry plots of microparticles (MPs). The MPs were directly labeled with phycoerythrin conjugated antiCD41 monoclonal antibodies (samples 1–3, left panel) or fluorescein isothiocyanate conjugated anti-CD31 monoclonal antibodies (samples 4–6, right panel). Each panel depicts representative samples from patients containing CD41þ (samples 2, 3) or CD31þ (samples 5, 6) MPs as well as negative samples (samples 1, 4). The percentage of positive MPs is depicted in each micrograph. Isotype control ranges are marked by dashed lines.
Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
The adjustments were made by first converting the positive MPs percentages to a logarithmic scale, subtracting off daily averages on that scale, and adding the overall log scale average. The adjusted MP values on the log scale were converted back to the original scale for presentation in graphs. For example, consider the result on MPs positive for CD31 for one of the women who was among those in the first flow cytometry 684
Salomon et al.
group of samples analyzed. This woman had a measurement of 0.8% positive MPs corresponding to 0.10 on the log scale. The average logarithm of percentage of CD31þ MPs among all women in the first group was 0.1798, whereas the overall average for all women in the study was 0.2102. The adjusted log measurement for this woman was thus 0.10 (0.1798) þ 0.2102 ¼ 0.29. The result for this woman was adjusted
Microparticles and complications of pregnancy
Vol. 92, No. 2, August 2009
FIGURE 2 Thrombin generation (TG) induced by concentrated microparticles (MPs). (A) Representative curves of TG induced in factor XI–deficient plasma containing rFVIIa (1.75 mg/mL) in the absence or presence of MPs. No TG is observed without addition of MPs (flat curve). Addition of CD31þ MPs, CD41þ MPs, or MPs negative for both CD31 and CD41 from three women, respectively, induced indistinguishable pronounced TG. (B) Endogenous thrombin potential (ETP) and peak height (mean SE) in four women negative for CD41 and CD31, four women with CD 41þ MPs, and four women with CD31þ MPs. Notably, TG was abolished in all samples by annexin 5 (1.25 mg/mL).
A
100
Positive CD 31 MPs Positive CD 41 MPs
90
Thrombin(nM)
80
Negative CD 41/CD 31 MPs
70 60 50 40
No MPs
30 20 10 0
0
10
20
30
40
50
60
70
Time (min)
nM/min
B 2000 1800 1600 1400 1200 1000 800 600 400 200 0
140 120
ETP
CD41/31 negative
CD41/31 negative +annexin
CD41 positive
CD41 positive +annexin
CD31 positive
CD31 positive +annexin
Peak height
nM
100 80 60 40 20 0 CD41/31 negative
CD41/31 negative +annexin
CD41 positive
CD41 positive +annexin
CD31 positive
CD31 positive +annexin
Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
upward by 0.39 to account for the generally low measurements in the first measurement group by converting back to the original scale; the adjusted measurement was 100.29 ¼ 1.95. These adjusted levels were compared across different groups of women using nonparametric tests: Wilcoxon for two groups and Kruskal-Wallis for more than two groups. Pearson correlation coefficients were computed relating the adjusted MP levels to blood flow resistance measurement. Regression analysis was used to test for a possible relationship of MPs levels to newborn weight, adjusting for gestational age at birth, gender, and blood flow resistance in the left uterine artery at 24 weeks of gestation. Women who developed preFertility and Sterility
eclampsia or had a lupus anticoagulant or a high titer of anticardiolipin were oversampled to increase their proportion in the sample. We defined IUGR and SGA as being in the lowest weight deciles at 31 to 33 weeks of gestation and at birth, respectively, after applying regression analysis to adjust weight for gestational age, sex, and smoking to the entire cohort of 642 women from our previous study (14). RESULTS The MPs that were negative for the specific markers examined by flow cytometry were indistinguishable from isotype control values, whereas MPs that were positive for markers 685
TABLE 1 The relationship between complications of pregnancy and percentage of CD31D or CD41D microparticles after adjustment for daily differences. Adjusted CD31a
Adjusted CD41a
Complication
n
Median
Quartiles
P
n
Median
Quartiles
P
No PIH or PE PIH PE No IUGR IUGR No SGA SGA
227 13 22 209 32 234 26
1.51 1.48 0.97 1.48 1.55 1.48 1.48
0.65, 4.07 0.52, 3.89 0.32, 1.62 0.66, 4.90 0.58, 2.16 0.66, 3.98 0.60, 2.95
.19
226 14 22 208 33 234 26
1.78 1.07 1.23 1.95 1.43 1.76 1.57
0.77, 4.74 0.65, 3.36 0.73, 3.06 0.81, 5.20 0.64, 2.43 0.73, 4.63 0.82, 3.36
.42
.23 .73
.10 .92
Abbreviations: IUGR, intrauterine growth restriction; PE, preeclampsia; PIH, pregnancy induced hypertension; SGA, small for gestational age. a Median relates to percentages of microparticles adjusted to a daily average on a log scale. Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
(e.g., CD31 and CD41) displayed increased fluorescence intensity (Fig. 1). In most plasma samples examined, MPs were negative for annexin 5, tissue factor, CD45, CD235a, and CD14 (not shown). The annexin 5 negativity supposedly indicated that the MPs did not harbor phosphatidylserine. However, when MPs from women who were CD31 positive, CD41 positive, or CD31 and CD41 negative were concentrated by ultracentrifugation, they displayed normal TG by an assay that is sensitive to the presence of phosphatidylserine (Fig. 2) (16, 17). Adding annexin 5 to these samples abolished TG, thus proving the presence of phosphatidylserine. These findings indicated that MPs exhibited a procoagulant function and that phosphatidylserine was present on MPs after all but at very low concentrations undetectable by flow cytometry. Because the flow cytometry analyses disclosed that MPs were negative for CD14, CD45, CD235a, anti-tissue factor, and annexin 5, analyses were only based on CD31 and CD41 markers. The adjusted levels of CD31þ MPs (of endothelial origin) and the levels of CD41 (of platelet origin) in women with IUGR, SGA, PIH, or PE showed no statistically significant difference from the levels in women who did not develop these complications (Table 1). Blood flow resistance measured at 24 weeks of gestation in the placental, umbilical, or right uterine artery had no statistically significant correlation with the levels of MPs (data not shown). In contrast, blood flow resistance in the left uterine artery was negatively correlated with the levels of CD31þ MPs (r ¼ 0.18, P¼.005). However, blood flow resistance measurements in all blood vessels examined at 31 to 33 weeks of gestation showed no statistically significant correlation with the levels of CD31þ or CD41þ MPs sampled at 24 weeks of gestation. The levels of CD31þ and CD41þ MPs at 24 weeks of gestation showed no statistically significant difference between women with and without inherited or acquired thrombo686
Salomon et al.
philias—that is, factor V Leiden, prothrombin G20210A, or antiphospholipid antibodies (Fig. 3). DISCUSSION In a previous study of 642 nulliparous women with singleton pregnancies, we found no significant relationship between common acquired or inherited thrombophilias and the occurrence of PIH, PE, IUGR, or SGA or with resistance to blood flow in the fetomaternal compartment (14). In the present study, we examined a subset of our original study group to determine whether elevated levels of circulating MPs at 24 weeks of gestation might predict the development of the aforementioned adverse outcomes. Our reasoning was that MPs, which shed from platelets, granulocytes, monocytes, or syncytiotrophoblasts after inflammatory cytokine stimulation or apoptosis, can aggravate vascular dysfunction and promote thrombosis by expressing tissue factor, phosphatidylserine, and mediators of inflammation (for review, see reference 2). Moreover, increased levels of MPs assessed in women several months after recurrent fetal loss suggested that this finding may reflect a chronic hypercoagulable state (11). Our flow cytometry analyses revealed that the only discernible MPs were of endothelial cell and platelet origin (syncytiotrophoblast MPs were not measured). Levels of CD31þ and CD41þ MPs of endothelial cell and platelet origins, respectively, had no statistically significant relationship with the occurrence of PIH, PE, IUGR, or SGA, nor was there a statistically significant relationship with the acquired or inherited thrombophilias or with blood flow resistance in the fetomaternal compartments at 24 and 31 to 33 weeks of gestation. A dubiously significant relationship between MPs levels and blood flow resistance in the left uterine artery was observed at 24 weeks but not at 31 to 33 weeks of gestation. One limitation of our study was the rather small number of patients with pregnancy complications who were examined.
Microparticles and complications of pregnancy
Vol. 92, No. 2, August 2009
FIGURE 3 Relationship between levels of CD31þ or CD41þ MPs and inherited or acquired thrombophilia. Levels of CD31þ MPs are shown in the upper panel, and CD41þ MPs are shown in the lower panel. Levels of MPs are expressed as log percent after adjusting for day to day differences.
Salomon. Microparticles and complications of pregnancy. Fertil Steril 2009.
Nevertheless, the consecutive prospective nature of our study removed potential sampling bias. Another limitation was our inability to retest the patients at the time of pregnancy complications. It is notable that the relationship between levels of MPs during normal pregnancies or complicated pregnancies is not at all clear. The MPs tested during normal pregnancy by a procoagulant assay have been shown to be increased (8) or decreased (18); in some studies, the level of MPs of platelet origin have been found to be similar in preeclamptic patients and women during normal pregnancy (3–5) but were found to be decreased in other studies (7, 8); and levels of MPs of endothelial cell origin in women with preeclampsia have been shown to be either increased (3, 4) or normal (5, 8). Our study does not explain these discrepancies because we measured the levels of MPs at the 24th week of gestation, Fertility and Sterility
not at the time of pregnancy complication. Measurement of MPs at 24 weeks of gestation does not predict subsequent development of PIH, PE, IUGR, or SGA. REFERENCES 1. Morel O, Toti F, Hugel B, Bakouboula B, Camoin-Jau L, DignatGeorge F, et al. Procoagulant microparticles disturbing the vascular homeostasis equation? Arterioscler Thromb Vasc Biol 2006;26:2594–604. 2. Lynch SF, Ludlam CA. Plasma microparticles and vascular disorders. Br J Haematol 2007;137:36–48. 3. Gonzalez-Quintero VH, Jimenez JJ, Jy W, Mauro LM, Hortman L, O’Sullivan MJ, et al. Elevated plasma endothelial microparticles in preeclmpsia. Am J Obstet Gynecol 2003;189:589–93. 4. Gonzalez-Quintero VH, Smarkusky LP, Jimenez JJ, Mauro LM, Jy W, Hortsman LL, et al. Elevated plasma endothelial microparticles: preeclampsia versus gestational hypertension. Am J Obstet Gynecol 2004;191:1418–24. 5. VanWijk MJ, Nieuwland R, Boer K, van der Post JAM, vanBavel EF, Sturk A. Microparticle subpopulations are increased in preeclampsia: possible involvement in vascular dysfunction? Am J Obstet Gynecol 2002;187:450–6. 6. Goswami D, Tannetta DS, Magee LA, Fuchisawa A, Redman CW, Sargent IL, et al. Excess syncytiotrophoblast microparticles shedding is a feature of early-onset pre-eclampsia, but not normotensive intrauterine growth restriction. Placenta 2006;27:56–61. 7. Harlow FH, Brown MA, Brighton TA, Smith SL, Trickett AE, Kwan YL, et al. Platelet activation in the hypertensive disorders of pregnancy. Am J Obstet Gynecol 2002;187:688–95. 8. Bretelle F, Sabatier F, Desprez D, Camoin L, Grunebaum L, Combes V, et al. Circulating microparticles: a marker of procoagulant state in normal pregnancy and pregnancy complicated by preeclampsia or intrauterine growth restriction. Thromb Haemost 2003;89:486–92. 9. VanWijk MJ, Svedas E, Boer K, Nieuwland R, vanBavel EF, Kublickiene KR. Isolated microparticles, but not whole plasma, from women with preeclampsia impair endothelium-dependent relaxation in isolated myometrial arteries from healthy pregnant women. Am J Obstet Gynecol 2002;187:1686–93. 10. Meziani F, Tesse A, David E, Martinez MC, Wangesteen R, Schneider F, et al. Shed membrane particles from preeclamptic women generate vascular wall inflammation and blunt vascular contractility. Am J Pathol 2006;169:1473–83. 11. Laude I, Rongieres-Bertrand C, Boyer-Neumann C, Wolf M, Mairovitz V, Hugel B, et al. Circulating procoagulant microparticles in women with unexplained pregnancy loss: a new insight. Thromb Haemost 2001;85:18–21. 12. Carp H, Dardik R, Lubetsky A, Salomon O, Eskaraev R, Rosenthal E, et al. Prevalence of circulating procoagulant microparticles in women with recurrent miscarriage: a case control study. Hum Reprod 2004;19:191–5. 13. Greer IA. Thrombophilia: implications for pregnancy outcome. Thromb Res 2003;109:73–81. 14. Salomon O, Seligsohn U, Steinberg DM, Zalel Y, Lerner A, Rosenterg N, et al. The common prothrombotic factors in nulliparous women do not compromise blood flow in the feto-maternal circulation and are not associated with preeclampsia or intrauterine growth restriction. Am J Obstet Gynecol 2004;191:2002–9. 15. Yagel S, Anteby EY, Shen O, Cohen SM, Friedman Z, Achiron R. Simultaneous multigate spectral Doppler imaging of the umbilical artery and placental vessels: novel ultrasound technology. Ultrasound Obstet Gynecol 1999;14:256–61. 16. Livnat T, Zivelin A, Martinowitz U, Salomon O, Seligsohn U. Prerequisites for recombinant factor VIIa-induced thrombin generation in plasma deficient in factor VIII, IX or XI. J Thromb Haemost 2006;4:192–200. 17. Livnat T, Martinowitz A, Zivelin A, Seligsohn U. Effects of factor VIII inhibitor bypassing activity (FEIBA), recombinant factor VIIa or both on thrombin generation in normal and haemophilia A plasma. Haemophilia. Published online March 25, 2008. 18. Desprez D, Zobairi F, Aucouturier JS, Leymarie F, Freyssinet JM, Grunebaum L, et al. Evolution of circulating procoagulant microparticles during normal pregnancy. Blood Coag Fibrinol 2008;19:179–81.
687