Evaluation of the topographic classification of placenta accreta spectrum: protocol for an ongoing prospective multicenter study

Abstract

Background

Placenta accreta spectrum (PAS) is a life-threatening obstetric condition associated with increasing cesarean delivery rates worldwide. Existing classifications, such as the World Health Organization International Statistical Classification of Diseases and Health-Related Problems and the International Federation of Gynecology and Obstetrics system, describe depth of invasion and histopathological features but do not adequately predict surgical outcomes or guide individualized management. The PAS topographic classification describes the anatomical extent of uterine wall remodeling and the presence of uterovesical adhesions, allowing surgical teams to anticipate intraoperative complexity, organ involvement, and appropriate therapeutic strategies. Although successfully applied in selected centers, its broader evaluation across diverse healthcare settings is lacking.

Methods

We designed a prospective, multicenter, international cohort study enrolling patients with a high prenatal suspicion of PAS. Eligible patients are aged ≥ 18 years, undergoing surgery after 20 weeks’ gestation, and managed by multidisciplinary PAS teams familiar with the topographic classification. Standardized prenatal ultrasound staging and intraoperative surgical staging are required, with photographic and video documentation. Surgical strategies include one-step conservative surgery, total hysterectomy, and modified subtotal hysterectomy, guided by intraoperative classification. The primary outcome is intraoperative blood loss, measured using a standardized protocol. Secondary outcomes include intraoperative complications, operative time, treatment type, and usability assessments of the classification through structured surveys. Data are collected in REDCap with external review of imaging records.

Discussion

This study will provide the first prospective, multicenter evaluation of the PAS topographic classification, assessing its correlation with clinical outcomes and its feasibility across hospitals with varying resources and surgical expertise. Preliminary results indicate successful implementation in both high- and low-volume centers, with promising adoption of prenatal ultrasound staging and surgical protocols. The collaborative, image-based, and open-access methodology aims to strengthen the reliability of PAS research by harmonizing surgical strategies and allowing external supervision. Ultimately, this project will generate robust multicenter data to inform individualized management strategies and support the integration of the topographic classification into routine clinical practice worldwide.

Trial registration

ClinicalTrials.gov Identifier: NCT05922397. Registered on 21 May 2023.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-025-08423-0.

Introduction

Placenta accreta spectrum (PAS) is a potentially life-threatening condition characterized by abnormal attachment of part of placenta to the uterine wall requiring a surgical procedure to remove [1]. The prevalence of PAS has increased over the past decades following a rapid rise in cesarean delivery (CD) rates, worldwide. The main risk factor for the development of PAS is implantation and placentation of a subsequent pregnancy inside the scar of a previous CD. The majority of patients with a high probability of PAS at birth present antenatally with a history of multiple CDs and a low-lying placenta or placenta previa [1]. While cesarean hysterectomy remains the most frequently used management strategy, alternative surgical approaches may be appropriate in selected clinical scenarios, including conservative surgical procedures and expectant management (commonly referred to as “leaving the placenta in situ”) [2].

Traditionally, PAS has been classified according to the depth of placental invasion into the uterine wall, with three histological grades: accreta, increta, and percreta [3]. These three severity grades were incorporated into the World Health Organization (WHO) International Statistical Classification of Diseases (ICD) and Health-Related Problems [4] in 2015, facilitating the search for specific PAS codes in medical records and population-scale surveillance [5], but also allowing for an overdiagnosis of approximately 40% [6]. In 2018, the International Federation of Gynecology and Obstetrics (FIGO) updated this classification by adding clinical and histopathologic criteria to each of these grades [7]. Although the clinical features included in this new classification have improved the description of the severity of the condition compared to the WHO ICD, the FIGO classification does not adequately predict surgical outcomes [8]. Overall, both classifications fail to provide the managing team with guidance on the likelihood of intraoperative complications, the degree of surgical difficulty, or the most appropriate treatment strategy for each case, particularly in low-resource health systems [1, 2, 9].

The topographic classification of PAS [10] proposes descriptions of the uterine area affected by PAS at laparotomy and evaluates the presence or absence of dense adhesions between the uterus and bladder [11–13]. This enables the surgical team to assess surgical complexity, the risk of injury to adjacent organs, and the likelihood of success of different surgical approaches [9, 14]. It is based on detailed anatomical knowledge of uterine vascular pedicles [15] and the relationships between different uterine segments and neighboring vascular, urinary, and musculoskeletal structures [16]. The topographic classification has been applied in several low- and middle-income countries for more than a decade. In 2020, we described how its use can help to identify patients who would benefit from a one-step conservative surgery [11].

Although originally designed for intraoperative application through surgical staging, recent studies have described how detailed pre-operative ultrasonographic assessment correlates closely with surgical findings within each category of this classification [16–18].

Despite evidence of its usefulness in different populations, many PAS multidisciplinary teams in specialist centers in both high and low-resources health systems remain unfamiliar with this classification [19]. Its implementation may appear challenging, as it requires proficiency in multiple surgical techniques and adherence to quality processes in PAS care, such as effective communication between imaging and surgical teams, standardized outcome audit protocol, and external supervision during the learning phase.

A prospective, multicenter, and international evaluation of the topographic classification of PAS is therefore required, including analysis of the correlation between classification categories and clinical outcomes, as well as an assessment of user perspectives regarding its applicability. In this article we describe the protocol and the preliminary results of an ongoing multicenter study, with the aim of facilitating understanding of the project for maternity hospitals interested in participating.

Methodology and study design

We designed a prospective, multicenter, multicohort study in which patients with a high prenatal suspicion of PAS are managed according to the recommendations of the topographic classification [20]. We will evaluate clinical outcomes across all categories of the classification, as well as the perceived usefulness and usability of the system by PAS teams responsible for patient management. Figure 1 presents a summary of the general characteristics of the study.

Fig. 1.

General methodological characteristics of the study. Detailed information about the methodology of this study is available at https://clinicaltrials.gov/study/NCT05922397 and recordings of virtual discussion sessions on the research protocol and the included interventions are available at l

Topographic classification

The topographic classification (Table 1) has been previously described [10], and the correlation between its categories and clinical outcomes has been demonstrated in multicenter retrospective studies [11, 12, 21, 22]. Furthermore, its role in guiding treatment selection has been reported in a feasibility study for a randomized controlled trial [14]. In brief, the classification describes the uterine areas associated with PAS. Each category identifies the involved uterine wall (anterior, lateral, or posterior) and whether the lesion is located above or below the vesicouterine peritoneal reflection. The final component of the classification is whether the surgeon can separate the posterior wall bladder from the anterior and lateral walls of the lower uterine segment (LUS). To enhance patient safety, the topographic classification can only be fully applied after exposing the anterior and lateral LUS walls and assessing the posterior wall. This requires the dissection of the LUS to separate the bladder serosa from the uterine serosa and an exploration of the parametrial space (surgical staging). If, for any reason outside this protocol (during the use of the classification in “real-life” scenarios), the treating team decides not to dissect the bladder in a patient with a high suspicion of PAS (such as in cases of leaving the placenta in situ or following vaginal delivery), the patient may be classified as a “suspected PAS case” or as having “PAS of uncertain type.” Since the lack of knowledge regarding the topography of the lesion limits the ability to define morbidity risk—and considering that all PAS surgeons should be capable of mobilizing the bladder as part of the surgical management of these patients—the topographic classification strongly recommends performing surgical staging (see below) before determining the type of treatment to be applied.

Table 1.

Placenta accreta spectrum topographic classification

PAS type	Uterine walls affected and relation with peritoneal reflection	Clinical features		Source of blood supply to the PAS area	Recommended vascular procedures	Proposed treatment
		US	Surgical
0	Upper anterior. Placenta detaches from the uterus	PAS US signs above the midpoint of the posterior bladder wall with half-full bladder1. US views: TA longitudinal (up–down, right–left)	Anterior lower uterine segment is free of abnormalities. During retrovesical dissection, healthy myometrium above the cervix level is rapidly encountered	UA	None	OSCS
1	Upper anterior. Placenta does not detach from the uterus			UA, SVA branches	NFV dissection and ligature	OSCS
2U	Upper lateral (parametrial)	PAS US signs in the upper lateral uterine wall. US views: TA transverse (up–down, right–left)	Obvious abnormality in the upper lateral uterine wall upon opening the parametrial space.	ObtA, IIA collaterals, UA	NFV dissection/ligature	OSCS
2 L	Lower lateral (parametrial)	PAS US signs in the lower lateral uterine wall US views: TV longitudinal (right–to–left) and TV transverse (down–to–up)	Bulging of the lower lateral uterine wall observed upon opening the parametrial space; the inferior margin of the bulge is not identifiable	IIA collaterals, CA, VA	IRAA, Bilateral CIA temporary occlusion	Total hysterectomy after ureteral tunneling
3	Lower anterior	PAS US signs in the anterior uterine segment, below the midpoint of the posterior bladder wall US views: TV longitudinal midline and right–to–left evaluation	Abnormality (bulging and hypervascularization) in the most caudal portion of the LUS and cervix. Evident during retrovesical dissection and bladder mobilization. It can be suspected before fetal extraction but is confirmed after delivery, upon exteriorization of the uterus and completion of the retrovesical dissection, by evaluating the cephalic myometrium toward the cervix	SVA, IVA, UA, CA, VA	IRAA, Bilateral CIA temporary occlusion	OSCS or total hysterectomy
4	Lower anterior + Fibrosis	Dense adhesions (fibrosis) are difficult to identify on US. Typically seen in exuberant lesions with distortion of the LUS and marked cervical vascularization	Separation of the bladder from the uterus in the LUS, particularly near the cervix, is extremely challenging even for experienced PAS surgical teams. This difficulty may be suspected during retrovesical dissection prior to fetal delivery, but it must be reassessed once the baby is delivered and the uterus is exteriorized. Bladder traction with Allis clamps at a 45-degree angle to the horizontal plane, combined with counter-traction on the LUS and a lateral-to-medial dissection approach, is often insufficient to adequately develop the retrovesical space	SVA, IVA, UA, CA, VA	IRAA, Bilateral CIA temporary occlusion	MSTH2
5U	Upper posterior	Difficult to identify in US in cases of placenta previa. Occasionally, depending on the acoustic window, an abnormality at the upper posterior uteroplacental interface may be observed and this topography suspected; however, physical conditions for US access to this uterine region are limited.	After fetal delivery and uterine exteriorization, an abnormality of the posterior uterine wall is identified. In upper lesions, the lower margin of the bulging is usually well defined and easily recognized, whereas lower lesions typically extend into the pouch of Douglas	UA, OA, vicariant flow from IMA	UA ligature	OSCS
5 L	Lower posterior	US views: TV longitudinal (right–to–left) can identify abnormalities in the posterior wall of the LUS.		UA, ARA collaterals	Collaterals ligature. IRAA?	Total hysterectomy

In each category of the topographic classification, some arterial pedicles are identified that provide most of the blood supply to the PAS area and that determine the recommended vascular procedures and the type of treatment necessary (One Step Conservative Surgery [OSCS], Total hysterectomy or Modified Subtotal Hysterectomy [MSTH])

Patients with an abnormal appearance of the uterine segment at laparotomy (bulging, purple coloration, etc.), but without vesicouterine vascular connections, in whom surgical dissection is performed without difficulty, are included in this classification as Type 0. These are cases with “uterine dehiscence” or “uterine window” in which PAS-suggestive ultrasound signs can be observed. Only after gross pathological examination of the surgical specimen can these cases be distinguished from true PAS, as slight manual traction allows separation of the myometrial area with abnormal appearance from the underlying placenta

Some patients have a combination of a large dehiscence with a smaller area where retrovesical dissection is cumbersome and/or the placenta does not detach ("mild PAS"). There are also patients with coexistence of PAS in different topographies of the uterus, which is described as “Synchronous PAS”. A patient may have a lesion in the lower part of the LUS (Type 3) and in the upper part of the lateral uterine wall (Type 2U), its topographic classification would be Type 3 + Type 2U

PAS placenta accreta spectrum, US ultrasonography, TA transabdominal, TV transvaginal, LUS lower uterine segment, UA uterine artery, SVA superior vesical artery, NFV Newly formed vessels, ObtA obturator artery, IIA internal iliac artery, CA cervical artery VA vaginal arteries, IVA inferior vesical artery, OA ovarian artery, IMA inferior mesenteric artery, ARA anterior rectal artery, IRAA infrarenal aortic artery, CIA common iliac artery

¹ A fully distended bladder is not advisable, as it distorts the anatomy and may lead to misinterpretation of placental localization. Conversely, an empty bladder limits visualization of the vesicouterine space. We recommend emptying the bladder half an hour before the ultrasound examination, followed by the immediate intake of 500 mL of water

² Other treatment options include total hysterectomy with partial bladder resection or leaving a portion of the uterus attached to the bladder. In both situations, the integrity of the urinary tract is compromised in a highly sensitive area (near the trigone), and substantial blood loss occurs. These approaches, as well as leaving the placenta in situ, are not recommended according to the topographic classification

Attempting to evaluate the severity of PAS based solely on the initial view at laparotomy (before bladder dissection) frequently leads to misinterpretation of the surgical risks. In particular, the surgical difficulty can be overestimated in cases with exuberant lesions over the peritoneal reflection, which may require only brief retrovesical dissections to reach healthy caudal myometrium and thus qualify for conservative management. By contrast, the surgical complexity can be underestimated in cases where the uterus appears grossly normal but where complex remodeling lesions following the scarification process (major dehiscence) and vascular changes associated with PAS are “concealed” by the bladder and peritoneal reflection. In the latter scenario, an overconfident surgeon may perform a transplacental fetal extraction followed by attempts at placental delivery, resulting in massive hemorrhage from anterior or low lateral lesions with no possible rapid solution since the bladder has not been dissected to achieve bleeding control (Fig. 2).

Fig. 2.

Overestimation or underestimation of lesion severity when relying only on findings at laparotomy. The appearance of the uterine segment at laparotomy, before bladder dissection, may lead to an inaccurate assessment of PAS severity. In some cases, a normal-looking anterior uterine wall (A) suggests absence of PAS, but after bladder dissection and mobilization, abnormal myometrial areas hidden behind the bladder become evident (B), complicating conservative procedures and increasing the risk of bleeding during retrovesical dissection. In other cases, exuberant bulging and hypervascularization are apparent at first sight (C), yet retrovesical dissection proves straightforward, and after ligating only a few vesicouterine pedicles (arrows), healthy caudal myometrium is quickly exposed (asterisk in D), identifying candidates for conservative surgery (D). The example of an iceberg, with only a small portion of its ice mass visible above the water surface while the greater volume remains hidden beneath it (E), or that of a rock with a large mass above the ground and only a small portion below (F), serves as a reminder that surgical exploration beneath the vesicouterine peritoneal fold is essential before making any assessment regarding the severity of PAS

Diagnosis of PAS

The definitive diagnosis of PAS is established when the placenta cannot be completely separated from the uterus at birth [1, 23]. In all included patients, gentle traction will be applied to the placenta, and whether complete removal was possible will be documented by video or still photography (Fig. 3). All surgical strategies included in this protocol allow for specimen evaluation before leaving the operating room.

Fig. 3.

Diagnosis of PAS. The definitive diagnosis of PAS is established when the placenta cannot be completely separated from the surgical specimen before leaving the operating room. In all included patients, firm traction will be applied to the placenta, and whether complete removal was possible will be documented by video or still photography. A–C One-step conservative surgery specimen. The maternal surface of the placenta, with a portion of adherent uterine segment (A), is examined, and the surgeon applies traction to separate both structures (B). Failure to achieve separation (C) confirms the diagnosis of PAS. D–G Total hysterectomy specimen. The uterus is opened longitudinally along the midline of the anterior wall (D–E), and traction is applied to the placenta to detach it from the myometrium F. Inability to separate the placenta (G) confirms PAS

Sample size

Considering the heterogeneity of PAS phenotypes and the relatively low incidence of severe forms (e.g., low lateral lesions [type 2 L] or thick and diffuse adhesions between the bladder and the LUS [type 4]) [11], we plan to recruit 100 patients in each of the two most common categories: high anterior (type 1) and low anterior (type 3). Based on the relative frequency of less common phenotypes (posterior [type 5] and lateral [type 2]) in previous reports, the recognition of a significant number of uterine dehiscence’s misdiagnosed as PAS during ultrasound evaluation [24], and a projected 15% attrition rate during recruitment, we established a total sample size of 326 patients.

Participating maternity hospitals

All maternity hospitals regularly managing patients with PAS are invited to participate, provided they have a dedicated PAS multidisciplinary team familiar with the topographic classification and offering individualized management options, including one-step conservative surgery, total hysterectomy, and modified subtotal hysterectomy. A formal invitation was sent to members of both international PAS societies (Panamerican society for PAS [PAS-2] and international society for PAS [IS-PAS]). In addition, open-access weekly webinars were conducted during the first semester of 2024 to disseminate the protocol and review planned interventions. Recordings of these webinars are available on YouTube [25].

Participating hospitals are required to follow the same prenatal preoperative and surgical protocols and to share clinical information and imaging records with the coordinating research group, receiving periodic feedback. No minimum annual case volume is required, given the pragmatic design of the study, which aims to assess the usefulness and usability of the topographic classification across diverse settings, including high-volume centers (> 10 cases/year) and low-volume centers, as well as hospitals located in countries with varying World Bank income classifications. Once the participating hospitals have signed the cooperation agreement with the principal investigator, one or more virtual meetings will be held to share study information, resolve doubts, and confirm that the hospital master and uses the surgical protocols proposed in this study. Three of the presentations used during these initial meetings are included as supplementary material for this article.

Patient eligibility

Eligible patients are > 18 years old, with prenatal signs suggestive of a high-risk of PAS, with detailed prenatal and preoperative ultrasound assessments, and managed by the local multidisciplinary team at the participating hospital, with surgery after 20 weeks of gestational age. Patients may be included whether admitted for urgent (except for active heavy bleeding) or scheduled delivery.

Prenatal evaluation protocol

All included patients must undergo ultrasonographic assessment according to the standardized protocol provided in the supplementary material (Fig. 4) [18]. Images supporting the prenatal suspicion of PAS must be stored for external review by three independent experts. Recommended documentation includes six short video clips (10-20 s each) and four static images. Each local team will perform a preoperative ultrasound staging using both transabdominal and transvaginal scans, assign one category of the topographic classification, and develop a surgical plan. Ultrasonographic findings should be illustrated using a schematic representation of the uterus in three planes, depicting the placental location and the area affected by PAS, particularly in relation to the bladder and cervix (Fig. 5). This information must be uploaded to the database before surgery.

Fig. 4.

Protocolized ultrasound evaluation in a patient at high risk for placenta accreta spectrum. All included patients must undergo ultrasonographic assessment according to the standardized protocol provided in the Supplementary Material. The evaluation should be performed with a full bladder, and four static images plus six short video clips (10-20 s each) must be obtained. All images should be de-identified, and their storage is recommended as part of the internal debriefing activities of the local PAS team. The purpose of these images and videos is to allow the local PAS team to justify its ultrasonographic diagnosis by documenting the lesion in a specific uterine wall. In addition, these materials may be reviewed by the primary investigator group and by the readers of this article once the study is completed. The four static images (A–D) include: two transabdominal views (A and B), one in grayscale (A) and another with color Doppler (B), both in the longitudinal midline plane showing the bladder, pubic bone, and placental position in relation to the cervix (A), and the vascular structures at the vesicouterine interface B. Two transvaginal views (C and D) are also required, one in grayscale (C) and another with color Doppler (D), both in the longitudinal midline plane, showing the cervical canal and lower anterior uterine segment (C), as well as subplacental vascularization toward the cervix D. The six short video clips (E–J) consist of three transabdominal views (E–G) and three transvaginal views (H–J). The transabdominal views are: longitudinal from cranial to caudal (E), longitudinal from right to left (F), and transverse from cranial to caudal G. The transvaginal views are: longitudinal from right to left (H), transverse from caudal to cranial (I), and transverse from right to left J. In all videos, identification of orientation structures such as the filled bladder, cervical canal, pubic bone, and iliac vessels should be facilitated

Fig. 5.

Schematic draft of ultrasonographic findings. To improve communication between the prenatal diagnostic team and the surgical team, we recommend creating a schematic drawing that integrates the interpretation of ultrasonographic findings in three planes: longitudinal (sagittal), transverse (axial), and coronal. These diagrams should emphasize the relationship of the placenta and the area of placental abnormal implantation with the anterior, lateral, and posterior uterine walls, as well as with the bladder, cervix, and other pelvic structures identifiable on ultrasound, such as the iliac vessels in cases of low lateral bulging. A template is provided to guide the sonologist in preparing the diagram (A), along with an example drawing displaying three views of a low anterior lesion (Type 3) with cervical hypervascularization (B), where the planned treatment was total hysterectomy due to the likely absence of sufficient healthy residual caudal myometrium for uterine reconstruction. This surgical plan must be reassessed after intraoperative staging, before selecting the final surgical approach

Surgical management protocol

Initial management of all patients follows surgical staging principles [9, 11, 14, 26, 27], including exposure of the anterior and lateral uterine walls (Fig. 6). The proposed sequence of interventions may be adapted depending on intraoperative findings:

Fig. 6.

Surgical staging and graphic documentation of intraoperative findings. In all cases, the initial surgical approach must consist of “surgical staging”, exposing the anterior, lateral, and posterior uterine walls before selecting the definitive treatment. Participating hospitals are requested to store photographs or short videos of the key surgical steps that justify the topographic classification category assigned to each patient, along with the corresponding clinical outcomes (type of surgery performed, blood loss, complications, etc.). The first view of the anterior uterine wall upon laparotomy (A) is insufficient for decision-making. It is recommended to expose the lateral uterine walls by applying traction to the round ligament (asterisk in B) and incising the anterior leaf of the broad ligament (B, C), followed by digital opening of the parametrial space (B), which allows confirmation or exclusion of more severe lesions (low parametrial or type 2 L). To expose the anterior uterine wall, the retrovesical space should be digitally assessed using the Pelosi maneuver or retrovesical bypass (D), after which the bladder (asterisk in E) is dissected and mobilized by applying anterocaudal traction at 45° with Allis clamps, followed by sequential ligation of the vesicouterine pedicles E. Once the bladder has been dissected, the fetus is delivered through a high segmental hysterotomy using the Ward maneuver F. After fetal delivery and externalization of the uterus, the lower anterior (asterisk in G) and posterior (asterisk in H) uterine walls are further assessed. If sufficient healthy myometrium is preserved caudal to the lesion (asterisk in I), one-step conservative surgery is performed (I-L) by ligating the colpouterine pedicles and en-bloc resection of the placenta together with the abnormal myometrium (asterisk in J), followed by uterine reconstruction K. If adequate residual tissue is not present, hysterectomy is indicated (M). If the bladder has been dissected, most of the procedure (one-step conservative surgery or hysterectomy) can be performed using a uterine tourniquet (asterisk) placed caudal to the area of abnormal myometrium (L, M) to avoid extra blood loss

1. Traction of the round ligament and opening of the anterior leaf of the broad ligament, followed by digital opening of the parametrium.

2. Exploration of the medial paravesical space from the parametrial entry.

3. Gentle digital exploration of the retrovesical space through the medial paravesical spaces (Pelosi maneuver or retrovesical bypass).

4. Retrovesical dissection with ligation of vesicouterine pedicles, assisted by vesical traction with Allis forceps angled 45° to the horizontal plane.

Once the anterior and lateral uterine walls are exposed, the surgical team assigns a topographic classification category and establishes the operative strategy. If complete bladder dissection is not possible before fetal delivery, the fetus is delivered through a transverse incision above the site of abnormal bulging and hypervascularization, over an area of macroscopically normal myometrium. If the surgeon encounters normally inserted placenta in this region, the Ward maneuver [28] is applied (dissecting the normal placenta between the incision site and the nearest placental edge) (Fig. 7). After fetal extraction, the uterus is exteriorized through the transverse abdominal incision, the posterior uterine wall is inspected, bladder dissection is completed, and the planned surgical strategy is implemented (one-step conservative surgery, total hysterectomy, or modified subtotal hysterectomy). The recommended approach for performing uterine sparing surgery [14], modified subtotal hysterectomy [22], selecting patients eligible for vascular interventions [12], and determining the type of vascular intervention to be used in cases of severe lesions [29] was discussed with each local group, in accordance with previously published recommendations.

Fig. 7.

Ward maneuver. In cases of anterior placenta previa, fetal extraction through the upper uterine segment can be performed without disrupting the placenta using the Ward maneuver. After completing intraoperative staging and ruling out a low lateral lesion (type 2 L) or dense adhesions between the uterus and bladder (type 4), an incision is made in the myometrium 1–2 cm above the area of abnormal bulging or hypervascularization A. After cutting the myometrium and before incising the placenta, the myometrial incision is extended (B), and the surgeon introduces a hand between the placenta and myometrium to detach the normally inserted portion of the placenta (C), locating the nearest upper placental edge D. Upon palpation of the amniotic membranes, manual rupture is performed, and the amniotic fluid is drained E. Once the detached placental flap is exteriorized, access to the fetus is achieved, and delivery is performed in the manner most convenient for the surgeon F. In these images, a breech delivery is shown (F and G), with visualization of the detached placental edge (asterisk), which does not bleed actively because the placenta has not been lacerated. After fetal extraction, the uterus can be easily exteriorized through the transverse laparotomy (H), and the edges of the incised myometrium can be clamped with ring forceps to minimize local bleeding (I) while the procedure is completed

This protocol does not include patients managed by leaving the placenta in situ. Additional interventions such as urinary stenting or aortic balloon occlusion are reserved for severe lesions (type 2 L or type 4) [21, 22] or unexpected massive hemorrhage, as determined by intraoperative staging. PowerPoint presentations with additional information on surgical staging, guidance on selecting a surgical technique based on intraoperative findings, and details on one-step conservative surgery are provided as supplementary material to this article.

Data collection

In addition to standard clinical variables typically reported in PAS surgical studies (e.g. intraoperative blood loss, complication rate, operative time, treatment type, others detailed in supplementary material), the study incorporates photographic and video documentation of ultrasound findings, intraoperative staging, macroscopic specimen analysis, and quantified blood loss.

• Primary outcome: intraoperative blood loss, measured using a standardized protocol (Fig. 8), including:

  1. Evaluation of suction devices (subtracting preoperative volume, amniotic fluid, and irrigation fluids).
  2. Weighing blood-soaked sponges and surgical drapes (subtracting dry weights).
  3. Measuring blood collected in a calibrated drape beneath the patient in a low lithotomy position [30].

• Imaging: Documentation must justify the chosen classification category during both ultrasound (Figs. 4 and 5) and surgical staging (Fig. 6).

• Participant feedback on the use of topographic classification for PAS: After six months of participating in the project, a survey will capture participant perceptions of the classification’s usefulness and implementation challenges at their institution.

Fig. 8.

Objective intraoperative blood loss quantification. In all patients included in this study, objective measurement of intraoperative blood loss will be performed, including: (A) evaluation of suction devices (subtracting preoperative volume, amniotic fluid, and irrigation fluids –tape marc in the canister-); (B) weighing of blood-soaked sponges and surgical drapes (subtracting dry weights); and (C) measurement of blood collected in the drainage bag placed under the patient’s pelvis when the low lithotomy position is used. The calibrated under-buttock bag has markings at 50 or 100 mL intervals, which facilitate rapid determination of its content. If a non-calibrated bag is used, it should be weighed at the end of surgery in a manner similar to the evaluation of blood-soaked surgical sponges (1 g = 1 mL).The process of blood loss measurement should be documented with photographs, as shown in this figure

A step-by-step guide describing the data collection process for this study can be found in the supplementary material.

Ethical considerations

This is a prospective observational study without post-discharge follow-up. Since participating hospitals already apply the topographic classification, the protocol does not alter patient management; it only standardizes the reporting of outcomes retrievable from medical records. The protocol promotes quality-improvement processes within PAS teams, including documentation of prenatal (ultrasound), intraoperative, and pathological (gross analysis of the surgical specimen) findings. All shared information must be de-identified. Local hospitals are responsible for obtaining patient consent to capture intraoperative photographs and share deidentified data.

Data will be stored securely at the Clinical Research Center, Fundación Valle de Lili (Cali, Colombia). Clinical variables will be recorded in REDCap, while imaging files will be archived in LogicalDOC, both with restricted access to the principal investigators. Ultrasound images will be reviewed by four expert sonographers for quality and classification assignment. Surgical and pathological images will be reviewed by the principal investigators, who will provide feedback to participating centers.

A cooperation agreement will be signed between each hospital and the coordinating research center. (Fundación Valle de Lili) The protocol is registered at ClinicalTrials.gov [20] under the number NCT05922397 and has been approved by the ethics committee of Fundación Valle de Lili (protocol No. 2023.2125).

Data collection period

The study began in Colombia on May 21, 2023, and patient recruitment is planned to conclude on June 1, 2027; however, this date may be extended if the expected sample size has not been reached.

Analysis plan

Descriptive analyses will be performed by stratifying patients according to the topographic PAS classification (T0, T1, T2-upper, T2-lower, T3, T4, T5-upper, T5-lower). Categorical variables will be evaluated using absolute and relative frequencies, while continuous variables will be assessed for normality with the Kolmogorov-Smirnov test. Normally distributed variables will be reported as mean and standard deviation, and non-normally distributed variables as median and interquartile range (Q1–Q3).

For inferential analyses, comparisons of continuous variables across the eight subgroups will be conducted using one-way ANOVA or, if non-normal, the Kruskal-Wallis test. Post-hoc analyses will be applied when appropriate. Categorical variables will be compared across subgroups using the Chi-square test. A p-value < 0.05 will be considered statistically significant.

No validated scale currently exists to assess the acceptability of new clinical classifications, such as the proposed PAS topographic classification. After an exhaustive literature search, we selected the Ottawa Acceptability of Decision Rules Instrument [31], a Likert-type tool widely used to evaluate rule acceptability among physicians. Since no Spanish version was available, we performed a direct translation of its content and included open-ended questions, which will be analyzed and tabulated by the principal investigator group. Minor adaptations were also made in our survey to the participants to some items to align with the study context (survey available at supplementary Material).

In June 2025, the first interim analysis was performed and the clinical results of the patients included up to that date, as well as the methodological difficulties encountered will be presented in this article,

Preliminary results

The study was registered at ClinicalTrials.gov on May 21, 2023, and the first patient was enrolled in Colombia on May 29, 2023. The second participating hospital was in Turkey, where the first patient was enrolled on April 23, 2024. Following the virtual dissemination activities conducted between January and June 2024, additional hospitals initiated the process of obtaining local approval, and between August 2024 and June 2025, seven new hospitals joined the study. By June 2025, five more hospitals had initiated participation, coinciding with the submission of this manuscript.

As of June 1, 2025, data from 51 patients (9 hospitals, 9 countries, 3 continents) had been included. Seven patients were excluded after quality analysis of the uploaded data: one because the surgery was performed at 19 weeks of gestation; two because they were classified as Type 2 L without supporting evidence of that topography in the shared images; and four because she was classified as Type 0 with suspected dehiscence prior to surgery (during ultrasound staging), yet no ultrasound signs of PAS were demonstrated in the shared images.

The preliminary clinical results of the 44 patients included in the preliminary analysis are described in Table 2. Most patients had lesions in the lower anterior wall (Type 3, n = 26, 59%).

Table 2.

Preliminary clinical results of placenta accreta spectrum patients according to the topographic classification

Variable		Categories of topographic classification. n = 44
		Type 0 Dehiscence n = 1	Type 1 Upper anterior n = 4	Type 2U Upper lateral n = 1	Type 2 L Lower lateral n = 3	Type 3 Lower anterior n = 26	Type 4 Lower anterior + UVDA n = 7	Type 5U Upper posterior n = 2	Type 5 L Lower posterior n = 0
Maternal age (years) *		40 (40–40)	35 (31–37)	39 (39–39)	23 (23–27)	36 (28–40)	30 (28–36)	35 (28–41)	NA
Pregnancies *		3 (3–3)	4 (3–5)	4 (4–4)	2 (1–4)	3 (3–4)	3 (2–3)	3 (2–3)	NA
Previous c-sections *		2 (2–2)	1 (1–2)	2 (2–2)	1 (1–1)	2 (1–2)	1 (1–2)	1 (0–2)	NA
Placenta previa, n (%)		1 (100)	4 (100)	0	3 (100)	25 (96)	6 (86)	2 (100)	NA
Gestational age at surgery (weeks) *		34 (34–34)	35 (33–36)	21 (21–21)	35 (34–35)	35 (34–36)	34 (33–35)	36 (36–36)	NA
Ultrasound staging, n (%)	Type 1	1 (100)	3 (75)	0	0	5 (19.2)	0	0	0
	Type 2U	0	0	0	0	0	0	0	0
	Type 2 L	0	0	1 (100)	1 (33)	0	0	0	0
	Type 3	0	1 (25)	0	1 (33)	20 (76.9)	5 (71)	1 (50)	0
	Type 4	0	0	0	1 (33)	1 (3.8)	2 (29)	0	0
	Type 5U	0	0	0	0	0	0	1 (50)	0
	Type 5 L	0	0	0	0	0	0	0	0
Blood loss volume (mL) *		378 (378, 378)	970 (720, 1255)	1560 (1560, 1560)	2000 (1800, 2200)	1600 (855, 2500)	3500 (2500, 3700)	750 (500, 1000)	NA
Transfusion, n (%)		0	2 (50%)	1 (100%)	3 (100%)	17 (65%)	6 (86%)	0	0
Number of PRBCU transfused *		NA	1 (1–1)	1(1–1)	2 (2–2)	2 (2–3)	5 (3–7)	NA	NA
Surgical time (minutes) *		113 (113–113)	110 (89–120)	172 (172–172)	129 (95–190)	107 (70–170)	220 (200–232)	94 (81–107)	NA
Postoperative hospital stay (days) *		2 (2–2)	3 (2–5)	2 (2–2)	5 (5–6)	4 (3–6)	7 (4–10)	4 (3–5)	NA
Type of treatment	OSCS	1 (100)	4 (100)	1 (100)	0	18 (69)	0 (0)	1 (50)	0
	TH	0	0	0	2 (67)	7 (26.9)	4 (57.1)	0	0
	MSTH	0	0	0	1 (33)	0	3 (42.8)	0	0
	Other	0	0	0	0	1 (3.8)	0	1 (50)	0
Bladder tear, n (%)		0	1 (25)	0	0	6 (23)	5 (71.4)	0	0
Reintervention, n (%)		0	0	0	1 (33)	0	1 (14.2)	0	0
Vascular intervention used, n (%)	Aortic occlusion	0	0	0	0	5 (19.2)	5 (71.4)	0	0
	IIA balloon occlusion	0	0	0	0	0	1 (14.3)	0	0
	IIA ligature	0	0	0	0	1 (3.8)	0	0	0

* Median (Q1 - Q3)

PAS placenta accreta spectrum, UVDA utero-vesical dense adhesion, PRBCU packed red blood cells units, OSCS one step conservative surgery, TH total hysterectomy, MSTH modified subtotal hysterectomy

Multiple challenges were identified during early data entry (Table 3), prompting the introduction of more frequent monitoring meetings (every 3 months) to improve subsequent data quality. Some clarifications on the use of topographic classification for the participants hospitals and minor changes in methodology were necessary and were included in updated versions of the protocol available at www.clinicaltrials.gov [20].

Table 3.

Challenges encountered during the design and implementation phases of the project

Identified/Potential Challenge	Action Plan
Some hospitals interested in participating have no prior knowledge of how to apply the PAS topographic classification or surgical alternatives to hysterectomy.	The MRG, independently from this study, continued with on-site and virtual training activities in various hospitals. After completing four supervised procedures, these hospitals maintained virtual interactions as needed and, after variable intervals ranging from 3 to 12 months, chose to participate once they had mastered the interventions included in this protocol and routinely performed auditing and supervisory activities for each of their patients.
	Weekly meetings were held over a six-month period to disseminate the specific details of the study and to review real PAS clinical cases, illustrating the management proposed by the topographic classification for each category. Those recording meetings are available on YouTube (https://www.youtube.com/@PAStopographyofficial/videos).
Potential multiplicity of surgical approaches across participating hospitals, introducing bias into the results.	All hospitals interested in participating, after obtaining approval from their local ethics committee and signing a cooperation agreement with the project coordinating hospital (Fundacion Valle de Lili, Cali, Colombia), held a virtual meeting with the MRG to discuss all study-related activities. Each local group presented one or more of their operated cases, and their surgical technique was compared with the one proposed by the PAS topographic classification. Only hospitals that agreed to apply both the ultrasonographic and surgical staging as outlined in this protocol, and to provide individualized management according to the recommendations of the topographic classification, were admitted. For example, groups that routinely apply the “leaving the placenta in situ” approach declined participation; however, in most cases, collaboration with the local investigative group continued outside this protocol with the aim of sharing experiences and training in techniques such as OSCS.
	Participating hospitals are required to submit photographs of the procedures performed in each case. When clinical outcomes diverged from what is expected for a given topographic classification category, the local team was required to provide the reasons for such deviation. For instance, a type 0 lesion (dehiscence) is not expected to result in heavy bleeding; if this occurred, the local team was obliged to explain the underlying causes identified during their internal monitoring activities.
	The specific features of each category of the topographic classification, practical tips for its application in resource-limited hospitals, as well as the recommended surgical technique for each scenario, have been extensively described in previous publications. These resources are available to hospitals interested in participating and can be consulted at the following links.PAS topographic classification: https://drive.google.com/drive/folders/1fIgYovSroyY8y6KQEbL5eHsKZjTEpIKp?usp=drive_link PAS type 0: https://drive.google.com/drive/folders/1qyDwZHkca24szCToN-bYN9CiUqcY7WVO?usp=drive_link PAS type 1: https://drive.google.com/drive/folders/1oDUlIy92vpo4gubPcfvxb0Z0b0QKMP3m?usp=drive_link PAS type 2: https://drive.google.com/drive/folders/14R6XTgB2MSamjyqxgrEyQGEybPtvPCua?usp=drive_link PAS type 4: https://drive.google.com/drive/folders/14R6XTgB2MSamjyqxgrEyQGEybPtvPCua?usp=drive_link Key concepts that justify the use of topographic classification and recommendations for its implementation in hospitals where previously only hysterectomy was performed.: https://drive.google.com/drive/folders/1Ps6yoGeAtwmCieiL4Zn7l2FBpnqak2CM?usp=drive_link Ultrasound staging for PAS: https://drive.google.com/drive/folders/1dmHUncix7fIJKCUMzrebqqJDeNH-aA3e?usp=drive_link Surgical staging for PAS: https://drive.google.com/drive/folders/1JOOzsNeLZ8gGuZVIK7LGJbUDlpCk8ddj?usp=drive_link Individualized use of vascular interventions in PAS: https://drive.google.com/drive/folders/1qslMnuDTd40cDSKDNixYn0Apa_ikXCq5?usp=drive_link
Risk of bias in ultrasound staging when the diagnosis is assigned after knowledge of the surgical findings.	Each local prenatal diagnostic group was asked to upload ultrasound staging data prior to surgery. At a minimum, the suspected topographic classification category and the proposed surgical plan had to be defined before and upload to RedCap database the scheduled surgery date. The required images (a schematic drawing of ultrasonographic findings, four static photographs, and six short video clips, Figs. 3 and 4) could be uploaded within two days postoperatively.
	The images and videos provided by the local PAS teams were reviewed by three sonologists from the MRG. They evaluated whether the submitted ultrasonographic images supported the selected topographic classification category and justified the proposed surgical plan.
Incomplete data entry for some variables during the initial cases.	The study coordinator reviewed all included cases on a weekly basis and provided feedback to each participating PAS team regarding missing information in one or more variables.
	Interim analyses were scheduled after the inclusion of 100, 200, and 300 patients, with an independent evaluator assessing data quality.
Data entry errors identified in several included cases.	Illogical or clearly erroneous data were reported immediately during weekly reviews. The principal investigator at each participating hospital will be asked directly to correct any erroneous information immediately.
	Monthly, the MRG provided feedback to the participating PAS teams on potential inconsistencies identified from patient-level data reviews.
Inability to share patient-related imaging material in certain hospitals	In some hospitals, the local ethics committee did not authorize sharing of patient photographs or videos. In such cases, ultrasonographic, surgical, and gross specimen images were reviewed by the MRG via videoconference with the local PAS team, using a slide presentation without transferring files outside the hospital’s information system.
	If such an arrangement was not approved by the local ethics committee, subgroup analyses will be performed depending on whether graphical material could be assessed or not.
Variability in the annual number of PAS patients managed across hospitals.	Since no standardized definition exists for the minimum number of patients required for a hospital to be classified as a “high-volume PAS center”, this study accepted all referral centers for PAS within each region or country where patients with this condition are routinely managed. A subgroup analysis is planned, separating hospitals in quartile 1 of annual case volume from those in quartiles 2–4.
Risk of inclusion of patients without PAS	Only patients with prenatal suspicion of PAS will be included. Patients with a prenatal suspicion of uterine dehiscence (prenatal staging = Type 0) will not be considered. In some cases, patients initially suspected of having uterine dehiscence may ultimately receive a final diagnosis of PAS; however, such cases will not be included in this study, which is focused exclusively on patients in whom prenatal evaluation suggests PAS At the beginning of the study, this distinction was not entirely clear, and data from four patients with ultrasound staging suggesting dehiscence (Type 0) were uploaded to the REDCap database. These cases were subsequently excluded from the analysis.
	Although some patients with placenta previa and a prior cesarean delivery, in whom PAS is suspected, may ultimately present with a placenta that separates from the uterus with gentle traction—classified as PAS type 0 in the topographic classification or as uterine dehiscence—these patients typically display abnormal features of the uterine segment during surgical staging (such as thinning with bulging or marked hypervascularization of the anterior lower uterine segment) and show ultrasound findings suggestive of PAS on prenatal evaluation. Local PAS teams are required to justify their assessment with shared imaging, which will be reviewed by the MRG.
	Given the challenges in diagnosing PAS following vaginal delivery, only patients with a definitive diagnosis established at cesarean section and confirmed by pathological gross examination of the surgical specimen will be included in this protocol.
	Some patients with a ultrasound suspicion of PAS (ultrasound staging) may in fact present with a uterine dehiscence, with the placenta detaching either during traction of the surgical specimen (gross analysis) or intraoperatively (surgical staging). These patients will be categorized as Type 0 and will be included in the protocol.
Possibility that, after initiating participation, local teams apply surgical techniques different from those proposed in the protocol.	Given that the validity of a multicenter surgical study depends on the application of uniform surgical techniques across all participating centers, emphasis was placed throughout all study phases (design, dissemination, inter-institutional agreements, kickoff meeting, and follow-up) that hospitals wishing to participate must adhere to the surgical recommendations established in this protocol and embedded in the topographic classification. Minor deviations will be accepted, such as the strategy for vascular ligation (suture or advanced bipolar energy), use or omission of ureteral stents for lesion different to type 2 lower (in which we strongly recommend its use), methods of aortic occlusion in severe cases (endovascular balloon, clamping or manual compression), choice of skin and fascial incision, anesthetic technique, and postoperative follow-up protocols. Major deviations from the proposed surgical approach—based on lesion topography and uterine vascular anatomy—resulted in exclusion of the patient from the database. If recurrent, the local group’s participation will be reconsidered. Examples of major deviations include: visual blood loss quantification; prophylactic ligation or occlusion of uterine or internal iliac arteries; fundal fetal extraction before intraoperative staging; transplacental fetal extraction without use of the Ward maneuver; use of “leaving the placenta in situ” as definitive management; omission of colpouterine pedicle ligation before en-bloc resection of abnormal myometrium and placenta; and placental traction prior to complete surgical staging due to a deceptively normal uterine external appearance during laparotomy.
Not all patients managed by the participating PAS team are included in the study.	In hospitals where all suspected PAS patients are managed by the same team, inclusion of all patients was expected, although a 25% loss rate was permitted due to multiple factors.
	In hospitals where not all patients were managed by the participating team, all patients treated according to the concepts in this protocol—at minimum, those operated on by the participating team—were expected to be included, with an allowable loss of up to 25%.
Challenges in categorizing patients with lesions across multiple uterine walls (multiple topographies)	Some patients may present lesions involving multiple uterine locations (e.g., lesions in the lower anterior wall and in the upper lateral wall: Type 3 + Type 2U), and it was not initially clear in which group they should be included for the analysis of clinical outcomes. In general, the topography associated with greater surgical difficulty and morbidity defines both the treatment approach and the clinical outcomes. Participating hospitals were therefore advised to categorize these patients by listing first the topography most closely related to surgical difficulty and the treatment chosen, followed by the additional topographies identified. For example, in a patient with PAS involving the lower anterior wall without fibrosis between the uterus and bladder (Type 3), coexisting with a low lateral lesion (Type 2 L), the greatest surgical difficulty is determined by the low lateral lesion, which would likely preclude conservative surgery. This patient would therefore be analyzed in the Type 2 L group and described as: Type 2 L + Type 3.
	Defining the PAS category that determines the type of management and the clinical outcome in each case is challenging when extensive lesions or multifocal (“Synchronous PAS”) are present. However, it is almost always possible to attribute the clinical outcome to a specific category, usually to the most “severe” one. Lesions located below the peritoneal reflection (Type 2 L, Type 3, Type 4, and Type 5 L) are generally associated with worse outcomes. Type 4 and Type 2 L lesions pose the greatest surgical challenges and account for the highest proportion of morbidity in cases with mixed lesions. For the statistical analysis, it was essential to individualize the most relevant lesion in each patient with mixed lesions. Following this preliminary analysis, each local group was asked to apply this approach to their previous and future cases, with minor supervised modifications to the database.
Risk of bias due to patient exclusion	Following the preliminary analysis of the first 51 patients enrolled, 7 cases were excluded. This raises concerns regarding the validity of the results once the total expected population is completed. To prevent recurrence, feedback was provided to all participating hospitals on the findings of the initial analysis, with emphasis on the difficulties outlined in this table. The planned interim analyses and weekly monitoring of enrolled cases are expected to reduce the frequency of patient exclusions. After the final analysis of this study, detailed information on all included patients will be published, without omitting any additional protocol deviations that may have occurred.

PAS placenta accreta spectrum, MRG main research group, OSCS one step conservative surgery

An analysis of the utility and usability of the PAS topographic classification has not yet been performed, as it will be conducted once all the hospital participants have been using it within this project for 6 months.

Discussion

PAS presents with multiple clinical phenotypes and is frequently associated with both false-positive and false-negative prenatal diagnoses. The need to provide individualized management for this condition was the rationale for developing the topographic classification [10]. Although this classification has demonstrated usefulness in hospitals accustomed to its application [11–14], its evaluation across diverse scenarios is essential, including settings with variable resource availability, annual case volume, duration of classification use, and access to the different therapeutic options recommended within the system.

During preliminary study, several maternity hospitals with no prior experience on using the topographic classification expressed interest in participating [32]. The core research group, therefore, established collaborative partnerships, providing either in-person or virtual support to these PAS teams as they incorporated individualized management strategies guided by the classification [32–34]. Although this activity was conducted outside the original study protocol, it facilitated the inclusion of new centers once they had acquired proficiency with the required procedures.

This study underscores the importance of linking a classification system to specific therapeutic strategies tailored to patient characteristics. Otherwise, subdividing the disease would serve only to describe the anomaly without guiding the clinical team in assessing surgical risk or selecting the most appropriate management option. Our preliminary results also indicate that both preoperative ultrasound imaging and surgical staging can be successfully implemented even in centres with limited resources and a low annual patient volume (Table 2).

The study also highlights several methodological aspects required to integrate new knowledge gained in recent years into surgical research in PAS. The high rate of misdiagnosis [24, 35] raises concerns about potential biases in retrospective single-center reports, which sometimes include patients without risk factors or those diagnosed after vaginal delivery. Photographic documentation of both prenatal and intraoperative staging is critical for external supervision and for standardizing surgical protocols [23]. The only sustainable way to collect and evaluate multicenter surgical data in PAS is through harmonization of surgical management across participating centres.

While variability in surgical practice and the “surgeon factor” can never be completely eliminated [36], the willingness of participating hospitals to adapt their surgical approach and accept external feedback is essential for producing high-quality PAS research, as it facilitates supervision of the quality of the intervention performed—one of the key elements for increasing the reliability of surgical research [37]. This study will provide valuable information for maternity hospitals seeking to implement individualized management strategies and, importantly, will generate an image-based dataset that will be accessible to readers, enabling secondary analyses and the potential for future individual patient data meta-analyses.

The collaborative, open, and transparent methodology adopted here also allows the identification of previously unreported challenges in surgical research on PAS, providing a foundation for improvement in future studies. Inevitably, several difficulties were encountered early in the project (Table 3). These have been addressed through amendments to the research protocol, thereby generating information that will be valuable for other investigators planning new studies in this field.

Conclusion

The prospective multicenter cohort study aimed at evaluating the topographic classification of PAS is currently underway and open to enrollment of additional participating hospitals. The implementation of standardized, protocol-based management across centers, coupled with supervision of surgical procedures and real-time data entry, provides an opportunity to generate high-quality scientific evidence and to facilitate individualized management of patients at high risk of placenta accreta spectrum.

Abbreviations

PAS

Placenta accreta spectrum

FIGO

International Federation of Gynecology and Obstetrics

PAS-2

Panamerican society for PAS

IS-PAS

international society for PAS

Authors’ contributions

Authors’ contributions:•Conception of the work: Jose Miguel Palacios-Jaraquemada, Albaro Jose Nieto-Calvache•Design of the work: Albaro Jose Nieto-Calvache, Jose Miguel Palacios-Jaraquemada, Juan Sebastian Galindo-Sanchez, Alejandro Solo Nieto-Calvache, Maria Alejandra Suarez-Revelo, Rozi Aryananda, Eric Jauniaux, Ahmed Hussein•Acquisition, analysis, or interpretation of data: Albaro Jose Nieto-Calvache, Jose Miguel Palacios-Jaraquemada, Juan Sebastian Galindo-Sanchez, Juan Pablo Benavides, Nicolas Basanta, Maria Alejandra Suarez-Revelo, Conrado Coutinho, Alejandro Solo Nieto-Calvache, Rozi Aryananda, Eric Jauniaux, Ahmed Hussein, Donald Brennan, Egle Savukyne, Erbil Karaman, George Daskalakis, Helena Bartels, Hubert Huras, Jean Marie Pellegrinelli, Magdalena Kolak, Andrzej Jaworowski, Nestor Javier Pavón, Otto Henrique May Feuerschuette, Ozhan M. Turan, Pele Koutroumanis, Marianna K. Theodora, Amparo Morales, Rita Lopez, Roman Shmakov, Tamara Yarygina, Valeria Lombardi, Ylva Vladic-Stijernholm, Juan Manuel Burgos-Luna, Adriana Messa.•Drafted the work: Albaro Jose Nieto-Calvache, Jose Miguel Palacios-Jaraquemada, Eric Jauniaux•Substantively revised: Albaro Jose Nieto-Calvache, Jose Miguel Palacios-Jaraquemada, Juan Pablo Benavides, Maria Alejandra Suarez-Revelo, Conrado Coutinho, Rozi Aryananda, Eric Jauniaux, Ahmed Hussein, Alejandro Solo Nieto-Calvache, Donald Brennan, Egle Savukyne, Erbil Karaman, Helena Bartels, Hubert Huras, Jean Marie Pellegrinelli, Magdalena Kolak, Nestor Javier Pavón, Otto Henrique May Feuerschuette, Ozhan M. Turan, Pele Koutroumanis, Rita Lopez, Roman Shmakov, Tamara Yarygina, Valeria Lombardi, Ylva Vladic-Stijernholm, Juan Sebastian Galindo-Sanchez, Juan Manuel Burgos-Luna•All the authors have approved the submitted version and have agreed both to be personally accountable for the author’s own contributions and to ensure that questions related to the accuracy or integrity of any part of the work, even ones in which the author was not personally involved, are appropriately investigated, resolved, and the resolution documented in the literature.

Funding

This article did not receive funding from any grant. Logistical and statistical support from the Clinical Research Center at Fundación Valle de Lili was essential for the planning and execution of this study. Participating hospitals did not receive financial support.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board (IRB) of Fundación Valle de Lili (Cali, Colombia) protocol Number 2023.2125.

Consent for publication

All images and videos used in this paper and in the accompanying supplementary materials were obtained with the written informed consent of the patients treated at the Valle de Lili Foundation, Cali, Colombia and at Dr. Soetomo Hospital, Surabaya, Indonesia.

Competing interests

The authors declare no competing interests.