A RETROSPECTIVE STUDY OF OUR NINE YEARS OF EXPERIENCE WITH ABNORMAL PLACENTATION — RISK FACTORS AND PREG-NANCY OUTCOMES

: Purpose: Placenta previa and morbidly adherent placenta are life-threatening conditions that necessitate close monitoring during pregnancy and can result in increased ante-and intrapartum complications. This study aims to evaluate the frequency of placental disorders in our population, reveal the impact of the known risk factors on maternal morbidity, and study new predisposing conditions and their effects on pregnancy outcomes . Materials and Methods: We performed a single-centered retrospective study in which we analyzed 330 women diagnosed with abnormal placentation—placenta previa and placenta accreta spectrum (PAS). We included all the women that gave birth at Nadezhda Hospital, from March 2013 to August 2021. We processed the data us-ing the chi-square test, Cramer’s V test, and cross-section analysis. Results: The frequency of placenta previa in our population is higher than the worldwide reported data. We were unable to demonstrate the connection between previous uterine manipulations (c-section, D&C, operative hys-teroscopy, etc .) and the occurrence of higher-grade placenta previa in our population. However, the more surgeries performed, the more complications during pregnancy and delivery occur (antepartum bleeding, increased blood loss, surgical management of postpartum hemorrhage, peri-partum hysterectomy). Pregnancies following assisted reproductive techniques (ART) are at a three-times higher risk of placenta previa. Endometriosis determines a greater risk of developing an abnormally adherent placenta. Conclusion: More prospective studies need to be performed to define the exact risk factors for abnormal placentation that affect the Bulgarian population. A better understanding of the condition will lead to better management of higher-risk pregnancies, thus reducing maternal and fetal morbidity.


INTRODUCTION
Abnormal placentation is a serious obstetrical disorder that increases maternal and fetal morbidity and mortality.A placenta previa is diagnosed when the placenta develops in close proximity (less than 2 centimeters) to the internal cervical os or covers it to some extent [1].'Minor' placenta previa is defined when the edge of the placenta lies adjacent to the internal os or it ends exactly at it.When the os is partially or fully covered, it is defined as 'major' placenta previa [1].When there is abnormal penetration into the myometrium (some authors describe it as abnormal myometrial invasion), we refer to this condition as the placenta accreta spectrum (PAS).Depending on the depth of villous invasion there are three subtypes -accreta, increta, and percreta and each one might be focal or can engage the whole surface of the placenta [1,2].Although it was described for the first time around 400 B.C. by Hippocrates [3], the etiology of placenta previa is still unknown.In the past decades, many risk factors that contribute to abnormal placentation have been revealed -maternal age, parity, cigarette smoking, previous cesarean sections, and uterine manipulations (myomectomy, D&C, etc.) [4,5].Some new factors, such as assisted reproduction techniques and endometriosis, have been studied lately [6,7].The frequency rate of placenta previa varies widely from country to country due to the subjectivity of diagnosis and the lack of a unitary registration system.The overall occurrence rate is 5,2 per 1000 pregnancies, with Asian studies having the highest prevalence-12,2:1000 [4,5].There is no registry form in Bulgaria, so the prevalence of the condition is unknown.In "Nadezhda" Hospital, the rate of placenta previa is 2,27%, of which 13,6 % are defined as 'major' placenta praevia, and the cases with both placenta previa and PAS represent 0,09% of the studied women.In our country, these conditions are highly associated with preterm delivery and prematurity, delivery complications such as increased blood loss, and the need for subsequent surgical management of peripartum hemorrhage.When delivery of unsuspected placenta praevia or PAS takes place at a tertiary health center there is a great risk of a fatal outcome for the mother and the baby.

MATERIALS AND METHODS:
Data was collected from the delivery records, antenatal ultrasound protocols, MRI protocols, surgery protocols, and histopathological results.We obtained and analyzed the information of every woman that had delivered her child in "Nadezhda" Hospital, from March 2013 until August 2021, and was diagnosed ante-or intrapartum with abnormal placentation.In that period, 14 527 women gave birth at our hospital.Of them, 330 had varying degrees of placenta previa, and 14 were concomitant with a PAS disorder.As inclusion criteria, we defined the following requirements: ultrasound findings of the placenta placed at less than 2 centimeters from the internal os; persistence of the findings after 32 weeks of gestation; MRI scans showing abnormal invasion of the placenta into the myometrium.We excluded all the women who had been diagnosed with placenta praevia early in the pregnancy, which resolved during the third trimester.We analyzed the data using 16 criteria: age, parity, cigarette smoking, previous C-section, previous myomectomy, ART, endometriosis, etc.Some of them are well-known risk factors related to abnormal placentation, and others are understudied conditions that might contribute to placenta praevia or PAS.We used descriptive analysis, chi-square test, and Cramer's V test for analyzing the criteria concerning only the studied patient.Cross-section analysis we used for the statistical data processing of the whole population of 14 527 women.

RESULTS: Well-known risk factors
We performed descriptive analysis on 330 women diagnosed with abnormal placentation for the well-known risk factors: parity, number of fetuses, cigarette smoking, and previous cesarean delivery.The age range of the patients was from 21 to 49 years, with a mean age of 34.8 y.For nearly 80% of the women, this was their first pregnancy.12.1% of the multiparous women had a previous normal delivery, and 8.2% of them had undergone cesarean delivery in the previous pregnancy.Due to the lack of information on cigarette smoking, no conclusion can be stated about the importance of this risk factor.(Table 1).
As for the number of fetuses, it gives an impression that only 1% of pregnancies were multifetal.It is known that twin pregnancy is a risk factor for placenta praevia due to the larger surface of the developing one or two placentas [8].Most of the ultrasound-diagnosed placenta praevia resolved after 32 weeks of gestation, however, there is a tendency that dichorionic twin pregnancies are at greater risk compared to monochorionic or singleton pregnancies [9].We performed a cross-sectional analysis encompassing all 14 527 women that gave birth at the hospital for the revised period, of which 940 carried twins, and only 5 of them were diagnosed with placenta praevia.The result showed that multifetal pregnancy is not a risk factor for the occurrence of placenta previa for our population: OR < 1 (95% CI: 0,0897 to 0.5274, p=0.0007).

Previous uterine manipulation as a risk factor for abnormal placentation
We decided to study the implication of previous uterine manipulation as a risk factor in our population.For a more accurate analysis, we divided this group into two subgroups.In the first subgroup, we included all women that had experienced one or more uterine manipulations for discontinuation of pregnancy due to the following reasons: spontaneous pregnancy loss, medically indicated termination of pregnancy up to 26 weeks of gestation, and intentional discontinuation of pregnancy.(Table 2).The second subgroup consisted of women that had undergone one or more uterine manipulations -D&C, operative hysteroscopy, hysterotomy, cervical conization, or trachelectomy, related to any of the following uterine Table 2. Distribution according to the number and reasons for miscarriages pathologies -abnormal bleeding, uterine fibroids, adenomyosis, endometrial or cervical polyps, intrauterine adhesions, dysmorphic uterus, operative delivery.(Table 3).

Table 3. Distribution according to previous uterine manipulations
We performed chi-square and Cramer's V tests for the data analysis, and we conclude that there is no statistical relationship between the occurrence of a higher grade placenta previa and the number and the reasons for the occurrence of miscarriages (χ2=10.814,df=6, p=0.094; χ2=11.518,df=12, p=0.485), nor is there a relationship between the number and the type of the performed uterine manipulations and the mentioned placental pathology (χ2=18.070,df=12, p=0.114).There is, however, a significant relationship between the prior uterine manipulations There is not even one case of PAS disorder among the women who had induced abortions.On the contrary, in the group with medically indicated abortion, the risk of abnormal placentation is almost 3 times higher (11 %) compared to the women with spontaneous miscarriages (4 %).
(Tables 4 and 5).Another significant finding is that with the increase in the number of preexisting uterine surgeries, the chances of complications arising out of delivery are getting highergreater blood loss, the need for surgical hemostasis, peripartum hysterectomy (χ2=27.205,df=6, p=0.000,Cramer's=0.203).Six women (5,7 %) of the group with prior uterine manipulations underwent peripartum hysterectomies.From the same group, 8,6 % (9 women) needed Bakrii Balloon application for intraoperative coping with excessive blood loss.Compared to the group with no previous uterine surgeries, it is 3 times higher (6 women, 2,8 %).
Every myometrial or endometrial scarring is a risk factor not only for abnormal placentation but for bleeding during pregnancy (χ2=16.888,df=6, p=0.010,Cramers =0.160).Twelve of the women experienced genital bleeding during the whole pregnancy, 41 of the women had bleeding before 24 weeks of gestation, and 40 women had it in the second half of the pregnancy.Prior operative hysteroscopies seem to have a strong impact on this complication.(Table 6).

Use of ART as a risk factor for abnormal placentation
In recent years, more and more attention has been drawn to the role of the emerging use of assisted reproductive techniques in obstetrical complications.It has been proven that women who are conceived by ART are at a greater risk of developing preeclampsia, hypertension during pregnancy, placental abruption, and placenta previa [6].We decided to study the impact of ART on our patients.We included all the women that achieved pregnancy by IVF or ICSI methods, and we performed a cross-sectional analysis.Of the 330 women with placenta previa, 144 were con-ceived by ART -5 % of all the women who were conceived with ART and gave birth at our hospital.Most of the women (70,1%) were in the age range of 31-40 years old.89 of them did not have previous miscarriages (61.8%) or surgical interventions of the uterus (53.5%, 77).46.6% of them were diagnosed with idiopathic infertility.We performed a cross-sectional test encompassing all 14 527 women, and we found that women conceived with ART are exposed to a three-times higher risk of the occurrence of placenta praevia OR=3,3 ( 95% CI: 2,6410-4,1116, p 0,0001) and RR 3,18 (95% CI: 2,5678 -3.9347).Similar results can find in the meta-analyzes of M.Karami et al. -OR was 2.67 (95% CI: 2.01, 3.34), and RR was 3.62 (95% CI: 0.21, 7.03), which proves the universal importance of ART [10].Most certainly, this is a collective factor that reflects the importance of the age of the woman, different reproductive techniques that had been used, accompanying gynecological conditions that demand conservative or surgical treatment, and other factors that are still undiscovered [10,11].

Preexisting endometriosis as a risk factor for abnormal placentation
Endometriosis is a chronic condition that affects around 10% to 15 % of women of reproductive age [12], and around 30 % of the women diagnosed with this pathology suffer from reproductive failure [13].It is clear now that this condition induces major endometrial changes that can lead to preterm birth, bleeding during the pregnancy, growth restriction of the fetus, and placental abnormalities [7,14].This dependency appears to be more pronounced during the first pregnancy.In the population that we investigated, 21 of the women (6.36%) had been previously diagnosed with endometriosis, and 16 of them (76.1%) had conceived with IVF.We observed that women with more than one uterine manipulation had been diagnosed with endometriosis more frequently (fig.1).

Fig. 1. Distribution of the women due to previous uterine surgeries and endometriosis
We performed chi-square and V'Cramer tests and we found a statistically significant relationship between endometriosis and the occurrence of PAS disorders.(χ 2 =16.757, df=3, p=0.001,Cramer's=0.225).The same conclusion we found in the meta-analyzes performed by S. Matsuzaki et al. [15] Not only did they prove this relationship, but they stated that most surely there is an association between endometriosis, the need for ART, and the occurrence of placenta praevia and PAS disorders in these patients.

DISCUSSION:
This research aimed to establish the frequency of abnormal placentation in the patient of Nadezhda hospital, analyze the risk factors that have the greatest impact on the occurrence of the placenta praevia and PAS disorders, and suggest some new risk factors.To describe the contingent of our hospital, we must say that these are mostly women between 30 and 40 years old, most of them were primiparas, and only 8,2 % had prior operative delivery.We could not manage to prove that previous c-section and cigarette smoking increase the frequency of placenta previa, as it is stated in several studies [1,2,4,6].
In terms of multiple pregnancies, we've done crosssectional analyses of all women who gave birth in the hospital, and we've proven that twin pregnancy is not a risk factor.In contrast to our findings, in the retrospective study of M. Weis et al., the authors show a significantly higher risk of placenta previa in dichorionic twin pregnancies compared with singleton pregnancies [9].
We obtained quite controversial results on previous uterine surgeries as a risk factor for abnormal placentation.
We have failed to prove that previous myomectomies, cesarean sections, operative hysteroscopy, or other uterine manipulations can lead to a higher degree of placenta previa.In their meta-analysis, A. Faiz and C. Ananth clearly outlined the relationship between prior manipulation of the uterus (c-section and abortions) and the higher chances of placenta previa [8].We only succeeded in finding a significant relationship between previous surgeries performed for abortion or uterine pathology and PAS disorders.An interesting fact is that women with a previously medically indicated abortion are three times more likely to suffer from abnormal placentation.The aforementioned metaanalysis indicates that women with spontaneous abortions have a higher risk of placenta previa, followed by women with induced abortions [8].In our population, not even one of the women who had induced abortion had PAS abnormality.When there are previous uterine lesions, there is a significant increase in the severity of intraoperative complications -mostly excessive bleeding.The greater the number of manipulations, the greater the severity of the complication.
An intriguing new risk factor for abnormal placentation is ART.The results from the papers of Romundstad L. B. [16], Karami M. [10], and Slavov S. [11], show that the implication of ART as a risk factor for abnormal placentation cannot be underestimated -between 2 and 6 times greater chances of placenta praevia.We performed cross-sectional analyses, and we confirmed their results.In our population, women who conceived with ART are at three times higher risk of developing placenta previa.We assume that the results we present are of high quality and that Nadezhda Hospital is one of the major reproductive centers in our country.We were unable to determine why ART has such a large impact on abnormal placentation -is it because women who require assisted reproduction are older, have various diseases of the reproductive system, or it is due to different techniques of embryo transfer or endometrial priming, which affect the placentation.Certain risk factors have not yet been identified.In the retrospective study of S. Korosec et al. [17], they analyze the outcome of pregnancy after the transfer of fresh and frozen embryos.The authors provide evidence that pregnancies after fresh embryo transfer have a higher rate of developing placenta previa development and third-trimester bleeding.In our study, we found that women with a history of operative hysteroscopy are more likely to bleed before delivery.
We managed to prove that even a condition like endometriosis, which is so well known, can have an impact on placentation and increase the rate of PAS disorders.In the observational study of N. Conti et al., they exhibit an increased rate of antepartum bleeding and placental complication during the first pregnancy of women with endometriosis [14].

CONCLUSION:
Placenta praevia and placenta accrete spectrum disorders are complications of pregnancy that affect the wellbeing of the mother during pregnancy and maternal and neonatal morbidity and mortality.Despite the unknown etiology of abnormal placentation, we should make efforts for a better and earlier diagnosis of the condition.The increase in understanding of the evolution of gynecological diseases allows us to develop knowledge of their etiology and the ability to define new risk factors so we can build good prevention strategies and health systems.New prospective research should focus on defining the risk factors that have the greatest impact on increasing the incidence of placenta praevia and PAS.Recognizing a patient who is at risk of adverse maternal and neonatal outcomes can lead to better triage of the patients, transferring them to a tertiary care hospital where timely and adequate intervention can be done so that complications for both the mother and the neonate could be minimalized.

Abbreviations
Abbreviations: PAS -placenta accrete spectrum ART-assisted reproductive technology MRI -magnetic resonance imaging D&C -dilation and curettage IVF -in vitro fertilisation ICSI -intracytoplasmic sperm injection

Table 1 .
Distribution according to well-known risk factors

Table 4 .
Distribution according to the number of miscarriages and PAS disorder

Table 6 .
Distribution of the patient with antepartum bleeding according to the degree of placenta previa, prior miscarriages and surgeries of the uterus

Table 5 .
Distribution according to the reason for the occurrence of miscarriage and PAS disorder