Proposal for standardized ultrasound descriptors of abnormally invasive placenta (AIP)

Abnormally invasive placenta (AIP) is a clinical term used to describe a placenta that does not separate spontaneously at delivery and cannot be removed without causing abnormally high blood loss1. It encompasses the histopathological diagnoses of placenta accreta, placenta increta and placenta percreta. It is a spectrum disorder, ranging from placentae containing a small area of abnormally adherent tissue (focal accreta) to those which have invaded into the adjacent viscera (percreta). It is potentially life-threatening, as forced removal of an AIP can lead to catastrophic maternal hemorrhage2, 3. In developed countries, AIP is the most common reason cited for Cesarean hysterectomy4. Maternal mortality for the most severe end of the AIP spectrum (placenta percreta) has been reported to be as high as 7%5, although this information was published in 1996 and may now be an overestimate given the advances in perinatal care and facilities. Prior Cesarean section (CS), other uterine surgery, assisted reproduction techniques and placenta previa are all risk factors for AIP and their prevalence has increased steadily over the last few decades6-8. The incidence of AIP has increased from 1:25 000 in the 1950s to 1:25008, 9 in the 1980s, paralleling the rise in CS rates, and AIP rates as high as 1:533 (USA)7 and 1:588 (Canada)10 have been reported recently. If these trends continue, it has been estimated that, by 2020, the USA will have a 56% CS rate, accounting for an additional 4504 cases of AIP and 130 maternal deaths annually11. Maternal mortality and morbidity are reduced when AIP is diagnosed antenatally and women deliver in a tertiary care hospital with a multidisciplinary care team12-14. Currently, diagnosis relies on ‘typical’ sonographic findings15, 16, such as ‘placental lacunae’ and ‘loss of the retroplacental clear zone’17. Magnetic resonance imaging, although employed widely in cases of suspected AIP, has yet to be demonstrated clearly to improve management or pregnancy outcome15. Irrespective of the imaging modality, diagnosis of AIP is subjective, with accuracy depending on the training and level of experience of the operator. Several studies have assessed the predictive value of different ultrasound markers of AIP. However, the performance of these markers shows considerable variability among studies using the same signs16. These differences have been attributed previously to a combination of limited sample size, retrospective design and variability of study inclusion criteria and eventual diagnosis of AIP16. Furthermore, as with all diagnostic techniques reliant on subjective opinion, the recorded presence or absence of each sign will be influenced by the operator's interpretation of what constitutes that marker. This is particularly important to clinicians, who may not have much experience with ultrasonography of the placenta or diagnosing AIP. Additionally, there is no published consensus on the definition of the ultrasound markers used commonly for AIP. Many signs have been described under different names, and in other cases the same term has been used for different findings. The aim of our study, therefore, was to provide unified definitions of ultrasound markers used commonly for AIP (‘ultrasound descriptors of AIP’). The ‘European Working Group on Abnormally Invasive Placenta’ (www.EW-AIP.org) is an international non-profit group, currently consisting of 29 obstetricians, gynecologists, pathologists, anesthesiologists and basic science researchers from 11 European countries. The aim of the group is to advance diagnosis and treatment and to promote research and knowledge on AIP. To improve comparability of future studies, to increase diagnostic capabilities and to facilitate international collaboration, the EW-AIP here proposes standardized definitions of the AIP imaging descriptors. These standardized definitions were produced by analysis of all 23 studies included in a recent systematic review of the antenatal sonographic diagnosis of AIP16 (Appendix S1). The exact wording used to describe the ultrasound signs of AIP was extracted, the descriptions were grouped according to ultrasound modality (two-dimensional (2D) grayscale ultrasound, 2D color Doppler and three-dimensional (3D) ultrasound) and synonymous or identical terms were unified under a common heading (Table S1). Following discussion by an expert panel (EW-AIP members present at the 7th EW-AIP meeting in Nancy, November 2014) the various wordings were unified into a set of 11 descriptors, six for 2D grayscale ultrasound, four for 2D color Doppler and one for 3D power Doppler (Table 1). The occurrence of each descriptor in the 23 papers used, grouped according to ultrasound modality, are listed in Table S2. Four of the papers18-21 included descriptions considered by the expert panel to be insufficiently comprehensive or specific and were not included in the unifying descriptors. Compound signs (e.g. ‘uterovesical hypervascularity AND bridging vessels’) were divided and included in the individual descriptors. Loss of ‘clear zone’ (Figure 1) Abnormal placental lacunae (Figure 2) Bladder wall interruption (Figure 3) Myometrial thinning (Figure 4) Placental bulge (Figure 5) Focal exophytic mass (Figure 6) Uterovesical hypervascularity (Figure 7) Subplacental hypervascularity (Figure 8) Bridging vessels (Figure 9) Placental lacunae feeder vessels (Figure 10) Intraplacental hypervascularity (Figure 11) During the meeting in Nancy, and in the following discussions among all EW-AIP members, importance was placed on defining each sign unambiguously, irrespective of opinions regarding the predictive value of each descriptor. The unified descriptors were augmented by images that the expert panel agreed were characteristic; these images, accompanied by examples of normal appearance where appropriate, of the unified descriptors are provided here (Figures 1-11 and S1–S11) and a description of technical aspects of the sonographic examination of AIP, with suggestions for obtaining such images in cases suspicious for AIP, are provided in Appendix S2. AIP is a clinically relevant, difficult-to-manage problem with rising incidence worldwide22. Accurate antenatal diagnosis, the basis for appropriate risk assessment and delivery planning, improves maternal outcome12-14, but is currently dependent on subjective interpretation of imaging findings. Until now, there has been no agreed terminology for these findings. We have identified and analyzed terms commonly used in the literature and unified them. We propose standardized unambiguous definitions of these AIP descriptors and accompany them with characteristic ultrasound images. These descriptors for AIP should be useful for clinical use, education, teaching and future research projects. In addition to using common terminology, describing precisely the location of the placenta and the part suspicious for abnormal invasion (the topography of AIP) should be considered a standard requirement for describing affected cases. Maternal mortality and morbidity associated with AIP are reduced when cases are delivered in a tertiary referral center with an experienced multidisciplinary team12-14, 23. Referral to such a team depends on the prenatal diagnosis of AIP by the primary healthcare providers. By defining clearly the sonographic signs of AIP we hope to facilitate this referral process. Furthermore, the rarity of this condition necessitates collaboration between centers, both nationally and internationally. Ensuring that all investigators are reporting the same ultrasound findings when referring to a specific sign will improve homogeneity of data collection, making results more valid. Based on these new descriptors we are currently collecting evidence and expert opinions regarding the predictive value of these signs, aiming to develop guidelines for the diagnosis and management of AIP. P.C. is supported by a research grant (RVO-VFN64165) from the Ministry of Health of the Czech Republic. Thorsten Braun, Charitè University, Berlin, Germany; Pavel Calda, Department of Obstetrics and Gynecology, General Faculty Hospital, Charles University, Prague, Czech Republic; Kinga Chalubinski, University of Vienna, Vienna, Austria; Frederic Chantraine, Université de Liège, Liège, Belgium; Sally Collins, John Radcliffe Hospital, Oxford, UK; Johannes Duvekot, Erasmus MC-University Medical Centre, Rotterdam, The Netherlands; Jean-Michel Foidart, Université de Liège, Liège, Belgium; Reynir Tómas Geirsson, University of Iceland, Reykjavik, Iceland; Hildur Hardardottir, University of Iceland, Reykjavik, Iceland; Wolfgang Henrich, Charitè University, Berlin, Germany; Gilles Kayem, Louis-Mourier, Columbes, France; Charlotte Krebs-Albrechtsen, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark; Jens Langhoff-Roos, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark; Louis Marcellin, Groupe Hospitalier Cochin-Broca-Hôtel Dieu, Université Paris Descartes, Assistance Publique-Hôpitaux de Paris, Paris, France; Pasquale Martinelli, University of Naples Federico II, Naples, Italy; Olivier Morel, Maternité Régionale Universitaire de Nancy, Nancy, France; Maddalena Morlando, University of Naples Federico II, Naples, Italy; Athanasios Mousiolis, Alexandra University Hospital, Athens, Greece; Carine Munaut, Université de Liège, Liège, Belgium; Michelle Nisolle, Université de Liège, Liège, Belgium; Per Olofsson, Skåne University Hospital, Malmo, Sweden; Jorma Paarvonen, Helsinki University Hospital, Helsinki, Finland; Philippe Petit, Université de Liège, Liège, Belgium; Babett Ramsauer, Vivantes Clinics Neukölln, Berlin, Germany; Loic Sentilhes, Angers University, Angers, France; Vedran Stefanovic, Helsinki University Hospital, Helsinki, Finland; Minna Tikkanen, Helsinki University Hospital, Helsinki, Finland; Vassilis Tsatsaris, Maternite Port Royal Hospital Cochin, Paris, France; Boris Tutschek, Prenatal Zürich, Zürich, Switzerland, and Medical Faculty, Heinrich Heine University, Düsseldorf, Germany; Heleen van Beekhuizen, Erasmus MC-University Medical Centre Rotterdam, Rotterdam, The Netherlands; Katharina von Weizsäcker, Charité University, Berlin, Germany. Appendix S1 The 23 studies of pregnancies at risk for invasive placentation analyzed by the European Working Group on Abnormally Invasive Placenta (EW-AIP) to produce the proposed standardized definitions of the abnormally invasive placenta imaging descriptors Appendix S2 Technical aspects to consider when performing an ultrasound scan for AIP markers Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.