​

Placenta accreta spectrum (PAS) is a complex placentation disorder associated with high maternal morbidity; complications of PAS include hemorrhage, blood transfusion, multiple organ failure, and death (1). The incidence of PAS has been increasing steadily in response to the increase in cesarean delivery rate (2). Available evidence supports planned preterm cesarean hysterectomy with the placenta left in situ as the standard treatment of PAS (3).

However, hysterectomy is traumatic to many women due to its operative sequences, impact on fertility, and disruption of self-image. Therefore, several conservative management options were proposed as an alternative to hysterectomy (4). Although many of conservative approaches yielded satisfactory results, their implementation as a part of standard protocols has been limited (5). There is primarily because evidence supporting most of these approaches is limited to case series, which is insufficient to support their safety. As a sequence, clinical trials are challenged by the lack of the margin of safety that would support ethical rationale of future studies. Availability of large multicenter studies is anticipated to provide robust evidence regarding optimal management of PAS and appropriate patient selection for conservative management. 

 

​

​

 

Aim of this project is to study diagnosis and management approaches of PAS and to assess safety and efficacy of different conservative approaches compared to planned hysterectomy. We aim at improving selection process and patient counselling for women who would like to consider alternatives to hysterectomy. To achieve these objectives, creation of an international database collected by PAS-experienced centers that represent all continents would promote conduction of large studies that provide higher level of evidence on different options of management of PAS.  

 

 

​

 

This study will include centers that present Europe, South America, Asia, and Africa. Data will be retrospectively collected from January 1st, 2010 to December 31st, 2019. Each center is anticipated to recruit at least 25 patients to participate in the study.  All women who were diagnosed with PAS during this period will be considered in the study.

 

Inclusion criteria:

• Pregnant women diagnosed with PAS, aged between 18 to 48 years.

• Women should be delivered by the corresponding center.

 

Exclusion criteria:

• Inadequate information and follow-up (e.g. single antenatal visit)

• Authorization to use anonymous patient data for research purposes.

 

After institutional review board (IRB) approves the study, data will be collected using an excel spreadsheet designed for this study or an online form (as per center preference). Data include patient age, parity, body mass index, smoking, ethnicity, previous gynecologic surgeries, obstetric complications, gestational age at diagnosis and delivery (in weeks), method of diagnosis, administration of antenatal steroids, indication of delivery, predelivery and postdelivery hemoglobin (g/dl), operative management, placental site, degree of placental invasion, type of cesarean incision and its relation to the placenta, intraoperative and postoperative complications, intraoperative blood loss in ml, transfusion of blood products, and neonatal outcomes. Data will not include any identifiable information.

The study meets the criteria of accelerated IRB review as it conveys minimal risk to research participants.

 

Statistical analysis

​

Data will be described using (mean, median, standard deviation, range) in the final sample. Machine learning method is superior to traditional statistical methods as it provides robust and automatic estimation of complex relationships between different variables and clinical outcomes. Data will be utilized as Xi and Yi where Xi presents input (features) and Yi presents dependent variables (outcomes). Functional regression is based on support vector machine by regressing the outcomes  on inputs . Model Validation will be performed via bootstrap estimation to evaluate the predictive ability of the functional regression models. Data will be split to training data (approximately 63% of the data) to create prediction model where bootstrapping will be applied, and testing data where prediction model will be validated. Machine learning models will be created using python 3.8®.

 

​

​

​

1. Jauniaux E, Collins S, Burton GJ. Placenta accreta spectrum: pathophysiology and evidence-based anatomy for prenatal ultrasound imaging. American journal of obstetrics and gynecology 2018;218:75-87.

2. Silver RM, Barbour KD. Placenta accreta spectrum: accreta, increta, and percreta. Obstetrics and Gynecology Clinics 2015;42:381-402.

3. Silver RM, Branch DW. Placenta accreta spectrum. New England Journal of Medicine 2018;378:1529-36.

4. Sentilhes L, Kayem G, Chandraharan E, et al. FIGO consensus guidelines on placenta accreta spectrum disorders: conservative management. International Journal of Gynecology & Obstetrics 2018;140:291-8.

5. Jauniaux E, Alfirevic Z, Bhide A, et al. Placenta Praevia and Placenta Accreta: Diagnosis and Management: Green-top Guideline No. 27a. BJOG: an international journal of obstetrics and gynaecology 2019;126:e1.