Placenta accreta spectrum. Current diagnostic issues

Introduction

Placenta accreta is a condition that pathomorphologically determines the complete or partial absence of the decidual membrane (decidua basalis), leading to the increment or germination of the muscle layer of the uterus by placental villi [1].

The term “placenta accreta” has been known to the world since 1937, when Irving and Hertig defined it as “abnormal partial or complete attachment of the placenta to the uterine wall” for the first time, publishing an article in which they presented a literature review (86 cases) and their data describing the clinical picture and histopathology of 20 cases of placenta accreta [2].

The “garland of victory” in the description of this placentation anomaly still belongs to the Canadian specialist Forster, who in 1927 was the first to publish a case of placenta accreta diagnosis in PubMed, presenting a histological description of the invasion of placental villi into the myometrium [3]. However, back in 1885, McDonald used the term “pathologically attached placenta” in an article devoted to the prevention of sepsis with placenta retention in the uterus after childbirth [4].

Over the past decades, there has been a significant evolution not only in the frequency of occurrence of this placentation pathology but also in diagnostic approaches and algorithms for the management of such patients.

“The true placenta increment is a very rare phenomenon” – this was the opinion of the leading figure of Russian obstetrics Malinovsky, expressed in 1955 on the pages of the publication “Operative Obstetrics” [5]. “Back in 1741, an experienced midwife Justina Zigemundin wrote that it could take many years before it would be possible to see a really grown placenta”, the author of the manual commented on his opinion [5].

According to Breen et al., “incremented” placentas are extremely rare and occur in one case out of 7000 births [6]. Over a 20-year period, Morison investigated 67 cases where the uterus was removed due to placenta accretion, with 645,000 births in Northern Ireland [7].

From the middle of the 20th century and till present, the frequency of abnormal placenta ingrowth has increased tenfold: from 1:25,000–50,000 births to 1:500–1000 at present, and in specialized centers — up to 1:343 [8]. Such dynamics of the process are explained by the progressive increase in the frequency of operative abdominal labor. The International Federation of Obstetricians and Gynecologists (FIGO) publishes strong epidemiological evidence that placenta accreta has now essentially become an iatrogenic condition as a result of an increase in the frequency of cesarean sections worldwide [9].

Placenta accreta spectrum (PAS) is the most severe form of all variants of placental attachment abnormalities with a high risk of uterine bleeding, which, according to WHO, occupies a leading position among all causes of maternal mortality [10]. At the same time, the direct share of PAS in the structure of maternal mortality reaches 7% [11].

The avalanche-like increase in the number of abnormal placenta attachment, accompanied by the above problems, contributed to the creation of an International Society called “International Society for Abnormally Invasive Placenta (IS-AIP)” with the formation of the European Working Group on Abnormally Invasive Placenta, (EW-AIP), uniting obstetricians and gynecologists, specialists radiation diagnostics, pathologists, anesthesiologists, and researchers of various profiles [12].

Due to the high risk of massive blood loss and maternal mortality, this pathology of placentation is an urgent problem of modern obstetrics, and optimization of diagnostic algorithms is its most important aspect.

Evolution of opinions on the PAS causes

One of the first hypotheses of PAS formation is based on a theoretical primary defect in the biology of the trophoblast, leading to excessive invasion of the endometrium and myometrium. At the beginning of the twentieth century, Ernst Bumm noted: “Speaking of placenta accreta, it is suggested that due to inflammatory processes, the spongy layer of decidua serotina, which under normal conditions separates at the slightest sipping, can become abnormally dense and capable of resistance and that this circumstance complicates the natural separation of the afterbirth” [13]. The overwhelming majority of specialists in those years agreed with the hypothesis that endometrial inflammatory processes are at the heart of the formation of PAS.

In 1986, the textbook “Obstetrics” edited by Bodyazhina et al. (the desktop book of many generations of Russian medical specialists) [14] postulated the relationship between “... the dense attachment of the placenta (placenta adhaerens), sometimes even its true increment (accretion placentae)...” and placenta previa, when “... there are a number of peculiar morphological changes in the lower segment and other parts of the uterus ... when chorionic villi penetrate into the wall of the uterus in the area of its isthmus much deeper than into the body of the uterus ...”. It is noteworthy that there is no relationship between the abnormal invasion of the chorion and the “scar” on the uterus after cesarean section, which is so relevant in modern obstetrics.

The modern point of view on the pathogenesis of PAS development is based on the opinion about the pathological morphofunctional transformation of the endometrium (inability to normal decidualization) as a result of traumatic damage to the uterus (cesarean section, myomectomy, curettage), which secondarily initiates chronic endomyometritis and, as a consequence, abnormal conditions for adequate implantation of the blastocyst [15]. This, in turn, creates conditions for the formation of one of the variants of ectopic pregnancy (pregnancy in the scar) with a high risk of developing abnormal placental attachment in the future [16].

Evolution of PAS classifications

Over the past decade, the variants of terminology and classification of this pathology have been revised. Currently, a new and very accurate terminology has been introduced worldwide, placenta accreta spectrum disorders (PAS), covering the entire spectrum of abnormal placentation disorders: abnormally adhesive placenta (placenta accreta) and abnormally invasive placenta (AIP, including placenta accreta + increta + percreta) [17].

For many years, the traditional PAS classification has been widely used (depending on the depth of villi invasion into the myometrium) [18][19]:

• placenta accreta — partial tight attachment, in which the chorionic villi reach the myometrium without damaging the integrity of the muscle fibers;
• placenta increta — true ingrowth of villi into the thickness of the myometrium, which is accompanied by damage to its structure;
• placenta percreta — associated with the germination of villi of the entire thickness of the myometrium up to the parametrium and peritoneum, less often — neighboring organs (bladder, intestines).

In 2018, a guideline of FIGO on surgical and conservative management of PAS was published, which included an updated clinical and diagnostic classification. These recommendations postulate the following 6 stages of an abnormally invasive placenta [20]:

Stage 1 — normal attachment/separation of the placenta;

Stage 2 — the placenta does not germinate through the uterine wall, but manual separation of the placenta (partial focal ingrowth) is required;

Stage 3 — the placenta does not germinate through the uterine wall, but manual separation of the placenta (diffuse ingrowth) is required;

Stage 4 — the placenta grows through the uterine wall, but the bladder is anatomically surgically separated from the uterus with a cesarean section;

Stage 5 — the placenta grows through the uterine wall, the bladder cannot be anatomically separated from the uterine wall;

Stage 6 — the placenta grows through the uterine wall and infiltrates the parametrium or any organ other than the bladder.

At the moment, the FIGO version [20] has been proposed as the “final” version of the PAS classification, which includes the following variants: PAS 1 — abnormal attachment of the placenta (placenta adherent or creta); PAS 2 — abnormal invasion of the placenta (increta); PAS 3 — abnormal invasion of the placenta (percreta): 3a — germination is limited to the serous lining of the uterus; 3b — with invasion of the bladder; 3c — invasion of other tissues/pelvic organs.

Evolution of diagnostics

Over the past decades, the algorithm for the diagnosis of “Placenta accreta” has radically changed.

At the time of Forster, Irving and Hertig [2][3], pathology was identified solely by the fact of a complicated course of the afterbirth (difficulties up to the impossibility of separating the placenta, accompanied by profuse uterine bleeding) or postpartum periods (abundant uterine bleeding). “As long as there is no bleeding, the true increment of the placenta does not cause any symptoms and is not recognized”, Academician of the Russian Academy of Sciences Persianinov wrote in 1974 [18].

Currently, the main method of diagnosing placental ingrowth is echography (ultrasound scanning, ultrasound diagnostics) [19–21]. The first experience of ultrasound PAS diagnostics dates back to 1982, when a team of American specialists diagnosed placenta previa with an abnormal (absent) hypoechoic retroplacental zone in a multipara Philippine patient (with a history of one cesarean section) at 25 weeks of gestation against the background of uterine bleeding. The ultrasound PAS diagnosis was confirmed by a morphohistologic examination of the uterus extirpated due to progressive bleeding [22].

Currently, the prenatal PAS diagnosis is based on echographic signs of two-dimensional seroscale scanning in combination with Color Doppler Imaging (CDI) and three-dimensional echography with an energy Doppler option [22]. However, with all the indisputable advantages of modern ultrasound technologies, echography remains an absolutely “operator-dependent” method, therefore, largely subjective, determined by the experience of a specialist in detecting this pathology with an instrument.

Nevertheless, a systematic review and meta-analysis of ultrasound studies in 3907 pregnant women at risk of PAS showed high efficiency of echography with a sensitivity of 90.7% (95% CI 87.2–93.6), specificity of 96.94% (95% CI 96.3–97.5) with a diagnostic odds ratio (DOS) of 98.59% (95% CI 48.8–199.0) [24].

In order to optimize ultrasound identification, level diagnostic errors, and unify ultrasound protocols, EW-AIP proposed the following terms for specific ultrasound PAS signs and their descriptions [25]:

• uterovesical hypervascularization;
• retroplacental hypervascularization;
• focal exophytic structure in the lumen of the bladder;
• loss/discontinuity of the bladder wall;
• “thick” lower segment of the placenta (normally, the thickness of the placenta approximately corresponds to the number of weeks);
• transverse course of the vessels between the uterus and the bladder wall (“vascular bridges” from the myometrium) through the serous cover into the bladder wall (so-called “rail sign”);
• discontinuity of the hyperechoic line at the border of the uterus and bladder;
• vessels feeding placental lacunae;
• presence of more than 2 lacunae in the thickness of the placental bed (on the maternal surface), including large (more than 2 cm of irregular shape) containing a turbulent flow visible in the gray scale (so-called symptom of “Swiss cheese”);
• protrusion of placental fragments (“placental bulge”);
• thinning of the myometrium (thickness less than 1 mm or not localized);
• loss of a clear hypoechoic zone between the placenta and the myometrium (“clear zone”);
• in the CDI mode, the presence of areas with a lacunar type of blood flow in the thickness of the myometrium is determined;
• vascular lakes in the myometrium with the presence of turbulent blood flow (pulsation velocity of more than 15 cm/s);
• visualization of blood flow in the area of the “scar” on the uterus;
• vascular invasion of the cervix;
• in the 3D mode, a complex of numerous randomly located heterogeneous convoluted vessels of the placenta is determined.

In 2015, the Placenta Accreta Index (PAI) was developed to predict abnormal placentation (PAS) in a population of high-risk women. This group consisted of pregnant women with placenta previa who had previously undergone cesarean section. The index was calculated taking into account the number of previously transferred cesarean sections in combination with a set of ultrasound criteria (vascular lacunae, thickness of the myometrium in the lower uterine segment, placenta location relative to the internal pharynx, color loci), estimated in points. According to the authors of the methodology, with a total of 6 points, the PAS probability was 83%, 7 points — 91%, and 8 points — 96% [26].

With the development of ultrasound technologies, the diagnostic capabilities of echography in detecting PAS are being improved and developed. Thus, there is a single piece of information in the literature on the possibility of using ultrasound elastography (echoelastography) in the diagnosis of PAS and placenta previa. It has been shown that the velocity of shear wave propagation is higher with placenta previa. However, due to the small sample size (43 patients), it was not possible for the authors to determine statistically significant differences in elastographic parameters during placental presentation with and without abnormal placental attachment [27].

It should be noted that the additional use of stereoscopic imaging of blood flow (LumiFlowTM) technology allows visualizing the involvement of the bladder wall of a pregnant woman in a conglomerate of abnormal placentation tissues (“rail sign”) from the myometrium through the serous cover into the wall of the urinary belly, that is, to identify placenta percreta (PAS 3b) [28].

Despite the developed classifications and ultrasound semiotics for various variations of abnormal placental attachment, the issues of PAS diagnosis continue to be discussed. So, according to Jauniaux et al. [19], no ultrasound signs or combinations of signs specific for determining the depth of placental invasion have been detected so far. Magnetic resonance imaging (MRI) is currently widely used in order to solve this problem. MRI technology is considered to be a useful tool for assessing the depth of invasion in cases of suspected placenta percreta, as well as in the diagnosis of complex cases, such as the predominant location of the placenta previa along the posterior wall of the uterus [29][30].

The main signs of placenta accreta according to MRI data are to a certain extent consistent with the ultrasound PAS criteria [31][32]:

• placental heterogeneity;
• strengthening of the vascular pattern;
• retroplacental shadow;
• swelling of the uterine wall;
• thinning of the myometrium;
• “dark tapes” — irregular intra-placental stripes;
• vascular lacunae;
• involvement of related bodies;
• cervical-trigonal hyperplasia, which occurs due to increased blood flow through the basins of the vaginal-uterine anastomoses extending from the internal genital artery between the Lieutaud body of the bladder and the cervix.

The MRI diagnostics has the greatest prognostic significance in determining the depth of invasion of placental villi, especially in the case of placenta percreta, as well as in cases of suspected ingrowth into the posterior wall of the uterus and into the parametria [33][34].

The sensitivity and specificity of MRI in placental growth vary from 75 to 100% and from 65 to 100%, respectively [35][36]. The sensitivity of MRI for the diagnosis of placenta accreta is 94.4% (95% confidence interval (CI) — 15.8–99.9), placenta increta — 100% (95% CI — 75.3–100), placenta percreta — 86.5% (95% CI — 74.2–94.4). The specificity of MRI for the diagnosis of placenta accrete is 98.8% (95% CI — 70.7–100), placenta increta — 97.3% (95% CI — 93.3–99.3), placenta percreta — 96.8% (95% CI — 93.5–98.7) [37][38].

A promising interest is the diagnostic algorithm “MAPI-RADS”, developed by Russian medical specialists [35]. The system takes into account the MRI diagnostic criteria, divided into two groups. The main criteria (diagnostic value of 2 points) are as follows:

• bulging of the uterine wall with prolapse of the lower uterine segment;
• thinning of the myometrium in the area of attachment of the placenta; the presence of vascular “lacunae” in the structure of the placenta;
• presence of vascular “tapes”;
• retroplacental hypointensive “shadow”.

Additional criteria (1 point — for the first and second, 3 points — for the 3^rd criteria):

• vascular fullness and varicose veins of the uterine wall;
• central/marginal placenta previa;
• spread of the placenta beyond the uterine wall.

Further, when summing up the scores obtained during interpretation, the “class” of pathological attachment of the placenta is set [35]:

• Class 1 (0 points): normal placenta, with a very low (less than 2%) risk of PAS;
• Class 2 (from 1 to 2 points): low (less than 10%) risk of PAS, possible tight attachment of the placenta (placenta adhaerens);
• Class 3 (from 3 to 10 points): high (more than 90%) risk of “ingrown” placenta (placenta accreta);
• Class 4 (from 11 to 12 points): high (more than 90%) risk of “ingrown” placenta (placenta increta);
• Class 5 (from 13 to 15 points): high risk (more than 95%) risk of “deeply ingrown” placenta (placenta percreta).

Two imaging technologies, Ultrasound and MRI, each with a very high diagnostic potential [24][29–33], at the present stage of development of obstetrics have completed the “competition” for the “garland of victory” in the question of their role in the detection, diagnosis, and verification of the PAS degree.

The FIGO Expert Group on the diagnosis and Treatment of PAS postulated [25] the following:

• Echography is an inexpensive, widely available imaging method, so it should be a priority in the diagnosis of PAS;
• MRI does not play a primary role in the diagnosis of PAS, but it may not be necessary to detail the topography of areas that are difficult to assess during an ultrasound examination.

According to a systematic review in 2014, it was concluded that, since MRI is not a screening method, it should be considered as an auxiliary and, if PAS is suspected, used after ultrasound examination [39].

The last decade has been marked by a significant interest of researchers in the search for alternative imaging technologies (ultrasound, MRI) for PAS diagnostics. Various serum markers were considered potentially significant tools for solving this problem [40–44]:

1. Placental lactogen.
2. Troponin 1.
3. Creatine kinase.
4. Pro-brain natriuretic peptide (pro-BNP).
5. Extracellular fetal DNA.
6. Cell-free placental m
7. PAPP-a.
8. β-HCG.
9. α-FP.
10. Matrix metalloproteinase (MMP-9, MMP-2).
11. Matrix metalloproteinase inhibitor (TIMP-1, TIMP-2).
12. Kisspeptin 1.

The above-mentioned list of markers is far from complete and will probably be updated in the future. However, according to the Experts of the FIGO group on the diagnosis and treatment of PAS [20][25], after confirming the results in prospective studies, these biomarkers can only be used as a supplement to ultrasound screening, but not as an independent tool for diagnosing PAS.

Conclusion

As it follows from the above-mentioned information, the diagnosis of abnormal placenta attachment (ingrowth, increment) at the present stage of development of obstetrics does not present significant difficulties. Thus, echography makes it possible to form a group of patients with a high PAS risk, revealing indirect signs of abnormal placentation, already in the first trimester of gestation. Chen et al. report a successful “diagnosis” of placenta accreta at the 9^th week of pregnancy [45]. It is known that both ultrasound and MRI have a high diagnostic potential in the PAS diagnosis [24][26][33][34]. However, as recent foreign studies have shown, PAS is not diagnosed in 2/3 of cases before delivery, while in 1/3 of cases, it occurs in specialized multidisciplinary hospitals [cit. 23]. According to Eller et al., in 28% of cases, the prenatally formulated PAS diagnosis had no histopathological confirmation [46].

The paradox of this situation is explained by the following.

First, the prenatal PAS diagnosis is to a certain extent subjective and is determined by the specialist's experience in detecting this pathology, which may be limited by the rarity of its occurrence, the lack of training programs, and the class of equipment used [23].

Second, a number of questions remain relevant at present, but have not received a definitive answer. Who can perform ultrasound diagnostics in order to detect PAS? Is a non-selective screening study by analogy sufficient in this case, as in the case of the formation of risk groups for chromosomal abnormalities? Or is selective ultrasound screening necessary in a group of women with a history of cesarean section? When is it necessary to conduct this study? Should it be performed in terms of early prenatal screening (11–13+6 weeks, according to the Order of the Ministry of Health of the Russian Federation No. 1130n dated October 20, 2020)? Or, taking into account the close relationship between PAS and the variant of ectopic pregnancy (cicatricial pregnancy — cesarean scar pregnancy, which requires the earliest possible diagnosis, at 7–8 weeks [16]), the formation of patients’ cohort for selective screening echography should be carried out at the same time?

However, even such an algorithm has its own questions and peculiar “pitfalls” requiring reflection and solutions. For example, if the absolute logic of the PAS formation chain from the moment of the blastocyst implantation into the “scar” uterus zone was practically recognized, what about those cases when the patient did not have any cesarean section in the anamnesis, but nevertheless PAS was formed (with full placenta previa [47] in particular cases)?

At last, who should conduct the research? In this case, the answer is on the surface. The purposeful search for PAS should be carried out by a trained expert-level specialist, an obstetrician-gynecologist, competent in ultrasound diagnostics and using high-resolution modern equipment.

Thus, the topic of abnormal placenta attachment diagnostics, despite significant achievements in the study of this problem, requires further development. This is confirmed by the clinical recommendations currently in the project format “Pathological attachment of the placenta (placenta accreta)”, developed by the Russian Society of Obstetricians and Gynecologists.

Authors’ contribution:

The contribution of the authors in writing the work is equivalent.

Conflict of interest. Authors declares no conflict of interest.