Placenta accreta is a severe pregnancy complication that may be associated with massive and potentially life-threatening intrapartum and postpartum hemorrhage.1 Wu et al reported an increase in the incidence to 1:533 births for the period from 1982 to 2002, suggesting that it is mainly the result of the increasing rate of cesarean delivery.2
Placenta accreta is defined as abnormal adherences or in-growths or both of the placenta to the uterine wall, such that clear separation of the placenta does not occur after the delivery of the newborn.3 Excessive trophoblastic invasion or a deficiency of decidua or both have been proposed to be responsible for the development of placenta accreta.4,5 The exact mechanism for the abnormal adherence and in-growth of placental tissues has not been fully elucidated.
Tissue damage resulting from a previous instrumentation is thought to play a role, because women presenting with placenta accreta have often had previous cesarean deliveries or curettage or both.6 Defective decidualization, abnormal maternal vascular remodeling, excessive trophoblastic invasion, or a combination thereof are considered to be the consequences of previous instrumentation.6
Given the uncertainty in the pathophysiology of placenta accreta and the conflicting data regarding the role of defective decidualization compared with excessive trophoblastic invasion, we investigated in vitro the effect of decidual injury on trophoblastic invasion potential using a decidual sample with an incision in it. We aimed to isolate the role of decidual injury by using normal trophoblast cells obtained from first-trimester legal abortions. Additionally, we aimed to investigate whether a precise and complete re-approximating of the incised decidual edges modifies the extent of in vitro invasion.
MATERIALS AND METHODS
This study was conducted from May 2009 to September 2010 in the Labor and Delivery Ward and the Laboratory for Research in Reproductive Sciences of the Department of Obstetrics and Gynecology at Ha'Emek Medical Center in Afula, Israel, a university teaching hospital.
Human trophoblast cells were collected from first-trimester (6 to 9 weeks) legal abortions, after informed consent was obtained from the participating patients. Only healthy women who had a surgical termination of pregnancy were included.
Trophoblast cells were isolated according to a protocol generously provided by G. Desoye (Clinic of Obstetrics and Gynecology, Medical University of Graz, Auenbruggerplatz 14, A-8036 Graz, Austria). Briefly, placenta were washed in saline, then the minced trophoblastic villi were digested by 0.25% trypsin and DNase I (both Sigma-Aldrich), and cytotrophoblasts separated from blood cells and decidua on a discontinuous Percoll gradient (Sigma-Aldrich). Contaminating leukocytes were removed by immunopurification with anti-CD45RB (DAKO) coupled to magnetic particles. We verified the purity of trophoblast cells using immunohistochemistry with specific antibodies to cytokeratin 7 (positive) and vimentin (negative), commonly used as an indication of trophoblast purity. This method supplies a 95–98% purity of trophoblasts, including all trophoblast subgroups.7 Cells were cultured in M-199 medium supplemented with 1.5% fetal calf serum (Beit-HaEmek, Israel) and 1% penicillin/streptomycin until completeness of the experiment.
Decidual tissue was retrieved from healthy, low-risk women who underwent an elective cesarean delivery at term because of abnormal presentation or macrosomia. The specimens were retrieved from the uterine incision site. Women with a previous uterine scar, multiple gestation, or malformed fetus or with an intrauterine fetal death were excluded. Specimens obtained were immediately transferred to the laboratory in a sterile box containing normal saline and were processed immediately.
All specimens were collected under aseptic conditions in the operating suite. A sample was taken from the decidua by scissors, after removal of the placenta. After tissue samples were collected (in sterile saline), they immediately were immersed in prewarmed medium without serum M199 (Beit-Haemek, Israel). The remaining tissue handling was performed under a laminar flow hood. All specimens were washed with M199 to remove the remaining blood before the beginning of the incubation period. No significant difference was found between the viability (as tested by trypan blue) after overnight incubation compared with 72 hours of incubation between samples (93±4.6% and 91±5.5%, mean±standard error [SE], respectively).
Each dissected decidual sample was divided into four similar-sized samples (approximately 0.5×0.5 cm2). The first slice was not treated, the second was incised with a surgical blade (no. 11) to mimic an in vivo injury, the third was incised and immediately repaired with a medical adhesive material (Histoacryl), and the fourth slice was covered with Histoacryl only, to exclude any effect of the adhesive material on the decidua. All specimens were culture in the bottom of a transwell plate with M199 medium.
Matrigel (1 mg/mL; BD Biosciences) diluted in serum-free media was added to the upper chamber of a 24-well transwell plate (8-micron pores; Corning). A total of 105 first-trimester human trophoblast cells in 150 microliters of media were added to the upper chamber and 500 microliters of media added to the lower chamber (containing the deciduas explants). Cells (105/well) were incubated at 37°C for 48 hours. Five wells were used. The first well, without decidua, was used as a control for examining trophoblastic invasion potential (model 1). The second well, with an intact decidua underneath the cells, was used to explore the effect of the decidua on the extent of trophoblastic invasion and to serve as a control for incised decidua (model 2). The third well, with an incised decidua underneath the cells, was used to investigate the extent of change in trophoblastic invasion resulting from the effect of injury to the decidua. The fourth well contained an incised decidua as in model 3; however, in this model the incised edges were subsequently re-approximated completely with Histoacryl underneath the cells. This model was used to investigate trophoblastic invasion through a repaired decidua. Decidual incised edges were glued together with Histoacryl because applying sutures was technically complicated (model 4). In the fifth well, an intact and normal decidua covered with Histoacryl underneath the cells was used to investigate whether the glue had any effect on the potential of invasion (model 5). Models 2–5 were applied each time with decidua taken from the same woman, and models 1–5 were examined with the same cohort of trophoblasts.
Cells seeded simultaneously in the same media in a well without transwell served as reference of total seeded cells. After incubation, noninvaded cells on top of the transwell were scraped off with a cotton swab, and the amount of invaded cells in the lower well as well as the amount of total seeded cells was evaluated with an XTT reagent kit. The percent of invasion was calculated as: absorbance of invaded cell×100=invasion index (%). Invasion was expressed as invasion index (% of control). The mean and standard deviation of the invasion index were then calculated.
The present study was approved by the ethical committee of Ha'Emek Medical Center in compliance with the Helsinki declaration. Signed consent was obtained from each of the participating women.
For each decidual sample retrieved from a distinct woman, four invasion tests (models 2–5) were performed in addition to the first control model performed without a decidua with the same trophoblastic cells population. The mean and standard deviation of the invasion index from all samples were then calculated for each model. The Friedman test was used to test whether there were significant differences between the models and the Wilcoxon signed rank test was used to test differences between model pairs. Significance level was set at P<.05 and Bonferroni corrections were made for multiple testing.
Decidual samples were obtained from eight women who underwent an elective cesarean delivery. Demographic and obstetric characteristics of the eight women are presented in Table 1. Each sample from each woman was submitted to the four invasion tests (intact, incised, incised and glued, and intact glued decidua) in addition to the control trophoblastic invasion test performed in the absence of decidua.
The mean invasion index for model 1 (control with no decidua) was 57.2% (±6.6%). Adding an intact decidua (model 2) increased significantly the invasion index to 69.9% (±5.1%) compared with no decidua (P=.01). The mean invasion index was 83.3% (±8.1%) for model 3 (incised deciduas), 66.6% (±8.2%) for model 4 (incised edges re-approximated with Histoacryl), and 58.3% (±11.3%) for model 5 (intact and normal decidua, covered with Histoacryl). Figure 1 illustrates the mean invasion index of each model. The highest invasion index was observed in the incised unapproximated deciduas (model 3; Table 2). There was no significant difference (P=.4) in the invasion indexes of the intact deciduas (model 2), the incised and glued deciduas (model 4), and the glued intact deciduas (model 5).
The exact pathogenesis of placenta accreta is unknown. A proposed hypothesis includes a mal-development of decidua, excessive trophoblastic invasion, or a combination of both. In this study we showed, in vitro, that an induced sharp decidual incision, imitating the in vivo process, that is, cesarean delivery, significantly increased the invasion potential of the trophoblastic cells. Using the same cohort of trophoblast cells, we showed that by changing the decidual features, we were able to modify the rate of trophoblast invasion, emphasizing the role of decidua on the invasion potential.
Generally, placenta accreta has been diagnosed on hysterectomy specimens when an area of accretion showed chorionic villi in direct contact with the myometrium and an absence of decidua.6,8 This finding may be focal in some cases where the decidua is present in areas adjacent to the foci of accreta. This decidual maldevelopment in placenta accreta is usually associated with previous instrumentation as in the case of previous cesarean deliveries or uterine curettages.9
Tseng and Chou hypothesized that the abnormal expression of growth, angiogenesis, and invasion-related factors in the trophoblast populations are the main factors responsible for the occurrence of placenta accreta.9 In addition, Cohen et al reported that the cytotrophoblast secretes factors that favor invasion, whereas decidua seems to not have a major role in regulating cytotrophoblast invasion in vitro.10 Data from Tantbirojn et al explained invasion of larger vessels in the outer myometrium and near the serosa to be determined by access rather than a preexisting defect in trophoblastic growth that would produce uncontrolled invasion through the entire depth of the myometrium in cases of accreta.6 They propose that increta and percreta more likely arise as a restult of dehiscence of a scar, which gives cells from the trophoblast column better access to large outer myometrial vessels.6 These observations clarify, although indirectly, our subsequent findings that showed that complete re-approximation of the incised edges of the decidua caused the incised decidua to behave similarly to intact decidua while restricting once again the extent of the invasiveness. We used medical adhesive material, which, according to the results of model 5 in this study, has no effect in modifying the potential of invasion. These results may point to the importance of careful closure and precise re-approximation of all layers of the uterine incision, particularly the decidua, not only the myometrium, during a cesarean delivery that may possibly restrict trophoblastic invasion through a complete or focal absence in the next pregnancy. However, only in vivo clinical studies may confirm this suggestion.
In this study we favored first-trimester trophoblastic cells over term trophoblasts because trophoblastic invasion is normally limited in time to the first trimester, although deciduas were retrieved from term pregnancies.10–12 Trophoblast cells from the first trimester have a naturally high invasion potential regardless of the existence of a target tissue. They act as carcinogenic cells. Our group and others have described this quality in previous reports.13–15 This may explain why more than half of trophoblast cells in model 1 migrated through the porous membrane of the upper well into the lower well without a target tissue. In contrast, decidual explants in this study were retrieved from term pregnancy. Although decidua from term pregnancy may differ from that of early pregnancy, we assumed that the dimensions of the decidual microimplants retrieved during terminations of pregnancy in the first trimester would not be large enough for producing the designed model and completing the experiment. Additionally, we used both model 1 (without decidua) and model 2 (with intact decidua) to examine the effect of a term decidua in attenuating the invasion potential. Model 2 showed a significantly higher invasion index. Finally the subsequent comparison across all models was performed with the same term decidua.
It has been reported that as in cases of third-trimester placenta accreta diagnosed after hysterectomies, accreta diagnosed in the first trimester also exhibits an anatomic defect of decidua.6 Owing to the rarity of this condition, it was not feasible to obtain trophoblasts from women diagnosed with placenta accreta in the first trimester and who also wished to terminate their pregnancy. We were unable, therefore, to obtain such trophoblasts to determine whether they demonstrate a different invasive potential. Still, Earl et al reported that the immunophenotype of extravillous trophoblastic populations in placenta accreta is identical to that seen in normal placenta, suggesting that it is unlikely that overactive trophoblastic invasion plays a major role in the pathogenesis of placenta accreta, and the absence of decidua is of greater importance in the pathogenesis.4
Without a purely anatomic absence of decidua or a lasting absence that alters trophoblastic behavior that is responsible for the development of placenta accreta, it is still to be determined why this anatomic defect exists only among a selected group of women and whether this lasting absence is preventable in vivo as well.
1. Faranesh R, Romano S, Shalev E, Salim R. Suggested approach for management of placenta percreta invading the urinary bladder. Obstet Gynecol 2007;110(2 pt 2):512–5.
2. Wu S, Kocherginsky M, Hibbard JU. Abnormal placentation: twenty-year analysis. Am J Obstet Gynecol 2005;192:1458–61.
3. Gersell DJ, Kraus FT. Diseases of the placenta. In: Kurman RJ, editor. Blaustein's pathology of the female genital tract. 5th ed. New York (NY): Springer; 2002:1111–3.
4. Earl U, Bulmer JN, Briones A. Placenta accreta: an immunohistological study of trophoblast populations. Placenta 1987;8:273–82.
5. Hutton L, Yang SS, Bernstein J. Placenta accreta: a 26-year clinicopathologic review (1956–1981). N Y State J Med 1983;83:857–66.
6. Tantbirojn P, Crum CP, Parast MM. Pathophysiology of placenta creta: the role of decidua and extravillous trophoblast. Placenta 2008;29:639–45.
7. Blaschitz A, Weiss U, Dohr G, Desoye G. Antibody reaction patterns in first trimester placenta: implications for trophoblast isolation and purity screening. Placenta 2000;21:733–41.
8. Benirschke K, Kaufmann P, Baergen RN. Pathology of the human placenta. 5th ed. New York (NY): Springer-Verlag; 2006.
9. Tseng JJ, Chou MM. Differential expression of growth-, angiogenesis- and invasion-related factors in the development of placenta accreta. Taiwan J Obstet Gynecol 2006;45:100–6.
10. Cohen M, Wuillemin C, Irion O, Bischof P. Role of decidua in trophoblastic invasion. Neuro Endocrinol Lett 2010;31:193–7.
11. Robillard PY. Interest in preeclampsia for researchers in reproduction. Reprod Immunol 2002;53:279–87.
12. Smith GC. First trimester origins of fetal growth impairment. Semin Perinatol 2004;28:41–50.
13. Staun-Ram E, Goldman S, Gabarin D, Shalev E. Expression and importance of matrix metalloproteinase 2 and 9 (MMP-2 and -9) in human trophoblast invasion. Reprod Biol Endocrinol 2004;2:59.
14. Goldman S, Shalev E. Difference in progesterone-receptor isoforms ratio between early and late first-trimester human trophoblast is associated with differential cell invasion and matrix metalloproteinase 2 expression. Biol Reprod 2006;74:13–22.
© 2011 The American College of Obstetricians and Gynecologists
15. Cohen M, Bischof P. Factors regulating trophoblast invasion. Gynecol Obstet Invest 2007;64:126–30.