Intramural ectopic pregnancy: An individual patient data systematic review.

The Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) [4] guidelines were used to conduct a systematic review of the medical literature. We selected cases of IMP published in peer-reviewed, PUBMED-indexed, and Google Scholar journals. All articles on IMP in human subjects available in English and French languages were included. As described by Memtsa and colleagues [1] and Auer-Schmidt and colleagues [5] , we defined IMP as a unique clinical entity with clinicopathologic features distinct from those of cesarean scar and cervical EPs. Cases of cesarean scar and cervical EPs were therefore excluded. Case reports of EP outside the uterus were also excluded. In April 2022, we performed a comprehensive search of the PUBMED database. The following terms were used and relevant citations assessed: “intramural pregnancy,” “intramural ectopic pregnancy,” and “intramyometrial pregnancy.” The terms “human” and “case report” were used as filters. A total of 65 articles published between 1965 and 2021 were selected. Google Scholar was also queried using the same terms and yielded an additional 36 articles. All citations identified were selected for abstract review. Articles not related to IMP or published in languages other than English or French were excluded. The remaining publications were selected for a preliminary evaluation during which incomplete and out-of-print articles were excluded. We then proceeded to the full-text evaluation and those considered relevant were included for final review. The initial review of abstracts was carried out independently by two individuals. During the final review, relevant data were extracted from case descriptions. Those include the first author’s name, country & year of publication, maternal & gestational ages, mode of conception, clinical presentation, obstetrics & gynecologic history, preliminary ultrasound diagnosis, ultrasound, MRI, CT, hysteroscopy & laparoscopy findings, and management ( Table 1 ). Table 1 Detailed characteristics of included studies [1] , [2] , [3] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] , [41] , [42] , [43] , [44] , [45] , [46] , [47] , [48] , [49] , [50] , [51] , [52] , [53] , [54] , [55] , [56] , [57] , [58] , [59] , [60] , [61] , [62] , [63] , [64] , [65] , [66] , [67] , [68] , [69] , [70] , [71] , [72] , [73] . Table 1 Detailed characteristics of included studies [1] , [2] , [3] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] , [41] , [42] , [43] , [44] , [45] , [46] , [47] , [48] , [49] , [50] , [51] , [52] , [53] , [54] , [55] , [56] , [57] , [58] , [59] , [60] , [61] , [62] , [63] , [64] , [65] , [66] , [67] , [68] , [69] , [70] , [71] , [72] , [73] .

A total of 65 and 36 articles were selected from the PUBMED and Google Scholar indices, respectively, for a total of 101 articles (including nine case series). After removing a duplicate study, twelve additional studies were excluded after abstract review, as the findings were unrelated to IMP. Six articles were written in languages other than French and English. Ten additional articles were excluded because they were either incomplete or the full text was out of print. 72 articles comprising 87 individual case reports were reviewed in detail. Five did not meet the inclusion criteria. A total of 82 case studies (patients) were included in the review ( Fig. 1 ). Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram illustrating the database search and selection process for cases and articles included in the study. Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram illustrating the database search and selection process for cases and articles included in the study. Patients’ ages ranged from 19 to 44 years (mean 32.07 years, SD ± 5.67 years) and the reported GA ranged from 4 weeks to 37 weeks [mean 9.27 weeks, SD ± 5.68 weeks]. A total of 11 patients (13.41%) conceived through assisted reproduction, 10 of which involved embryo transfer. A history of spontaneous and/or induced abortion was reported in 37 patients (45.12%). With the exception of two, all the remaining patients included in the analysis had at least one risk factor for IMP. History of curettage was the most common risk factor, reported in 30 (36.58%) patients, followed by history of salpingectomy in 10 patients (12.19%), assisted reproduction techniques with embryo transfer (ART-ET) in 10 patients (12.19%), and previous cesarean delivery in 9 patients (10.97%) [ Fig. 2 ]. Fig. 2 Bar graph illustrating the most common risk factors for intramural ectopic pregnancy (IMP). ART+ET = assisted reproductive techniques with embryo transfer. Fig. 2 Bar graph illustrating the most common risk factors for intramural ectopic pregnancy (IMP). ART+ET = assisted reproductive techniques with embryo transfer. IMP rupture necessitating emergency laparotomy was reported in 12 patients (14.63%). GA at the time of rupture ranged between 5 weeks, 6 days and 26 weeks, 0 days (mean: 15.11 weeks, SD ± 6.18 weeks). Clinical presentations and complications are summarized in Fig. 3 . Fig. 3 Bar graph illustrating the most common clinical presentations of IMP. Abd = abdominal; RPOC = retained products of conception; Rt = right. Fig. 3 Bar graph illustrating the most common clinical presentations of IMP. Abd = abdominal; RPOC = retained products of conception; Rt = right. The diagnosis of IMP was confirmed on pathology [22 cases (26.83%)], Magnetic Resonance Imaging (MRI) [14 cases (17.07%)], exploratory laparotomy [12 cases (14.63%)], and diagnostic laparoscopy [10 cases (12.19%)] ( Fig. 4 ). The proposed diagnostic steps are presented in Fig. 5 . Fig. 4 Bar graph illustrating modalities commonly used to confirm a diagnosis of IMP. CEUS = contrast-enhanced ultrasound; TVS = transvaginal ultrasound; Uss = ultrasound. Fig. 4 Fig. 5 Diagnostic algorithm for suspected IMP. Fig. 5 Bar graph illustrating modalities commonly used to confirm a diagnosis of IMP. CEUS = contrast-enhanced ultrasound; TVS = transvaginal ultrasound; Uss = ultrasound. Diagnostic algorithm for suspected IMP. A presumptive diagnosis of IMP was established by sonography in 29 cases, though ultrasound findings were not reported in 16 cases (19.51%). Among cases in which ultrasound findings were included, 13 (19.7%) were wrongly diagnosed as either angular, cornual, or interstitial pregnancy ( Fig. 6 ) and eventually proven to be IMP. Fig. 6 Bar graph demonstrating diagnoses of suspected IMP following an ultrasound evaluation. EP = ectopic pregnancy; GTD/GTN = gestational trophoblastic disease/gestational trophoblastic neoplasm; IUP = intrauterine pregnancy. Fig. 6 Bar graph demonstrating diagnoses of suspected IMP following an ultrasound evaluation. EP = ectopic pregnancy; GTD/GTN = gestational trophoblastic disease/gestational trophoblastic neoplasm; IUP = intrauterine pregnancy. Management of IMP involved laparoscopic (21 cases [25.61%]) or laparotomic (19 cases [23.17%]) resection of the pregnancy mass followed by surgical repair. Hysterectomy and medical therapy were performed in 11(13.41%) and 19(23.17%) patients respectively. Only 3(3.65%) patients had a successful conservative approach with the IMP resolving without an intervention ( Fig. 7 ). Fig. 7 Bar graph illustrating management strategies for IMP. Bil = bilateral; Uss = ultrasound. Fig. 7 Bar graph illustrating management strategies for IMP. Bil = bilateral; Uss = ultrasound. Various imaging modalities were used to evaluate suspected IMP. Ultrasound was performed in the vast majority (80 patients [97.56%]). MRI and Computed Tomography (CT) scan were used less frequently (18 cases [21.95%] and 4 cases [4.88%], respectively). Laparoscopy (diagnostic and surgical) and diagnostic hysteroscopy were performed in 36 patients (43.9%) and 22(26.83%) respectively. The only two cases in which ultrasound wasn’t performed were reported in 1965 by McGowan [24] , when diagnostic ultrasound wasn’t widely available in clinical practice.

The objective of this individual patient data systematic review was to identify and synthesize what is known about IMPs, including etiology and pathophysiology, common clinical presentations, imaging features, laparoscopic and hysteroscopic findings, and management.

IMP is a rare but potentially lethal clinical entity. A significant proportion of patients are asymptomatic and have no known risk factors. Correlation of clinical history and imaging findings is vital to establishing a prompt diagnosis and reducing the risk of a catastrophic outcome. Ultrasound plays a key role in establishing a diagnosis of IMP. However, in the setting of a positive pregnancy test, a GS or mass-like amorphous echoes partially or completely surrounded by myometrium is virtually pathognomonic. A serosal surface of the myometrium less than 3 mm thick in the body of the uterus should also raise suspicion for IMP. In some cases, MRI may be indicated to establish a definitive diagnosis and for surgical planning. There are no defined clinical practice guidelines for the management of IMP. However, medical and/or surgical managements are nearly always required. Recognition of classic clinical and radiologic findings helps improve diagnostic accuracy and reduce morbidity and mortality among women presenting with IMP.

To the best of our knowledge, this is the largest and most comprehensive review of IMP to date, including a total of 82 cases across 27 countries ( Table 1 ). The etiology and pathophysiology of IMP remain to be definitively established. It has been postulated that IMP results from increased lytic activity of syncytiotrophoblasts with resultant defective decidualization, which allows the conceptus to penetrate the myometrium or implant in the serosa following external migration [2] , [30] , [39] . Auer-Schmidt and colleagues [5] described three contributory factors that may lead to IMP. First is a false tract between the endometrium and the myometrium, most often secondary to prior uterine/endometrial trauma during instrumentation or surgery. The second is in vitro-fertilization with embryo transfer (IVF-ET), during which embryos are mistakenly placed into the myometrium through the false tract. The third factor is adenomyosis, which enhances myometrial receptivity thereby increasing the likelihood of myometrial implantation. Other authors suggest that intrauterine trauma during difficult embryo transfer results in a false passage [2] , [71] , [74] . Risk factors associated with the pathogenesis of IMP [ Fig. 2 ] were history of curettage (36.58%), history of cesarean section (10.97%), history of salpingectomy (12.19% patients), ART-ET (12.19%), history of myomectomy (9.75%), and adenomyosis (4.88%). Interestingly, two reported cases had no predisposing risk factors [13] , [37] . To the best of our knowledge, this study is the first to identify a possible association between a history of salpingectomy and the development of IMP. Among the 10 patients (12.19%) with IMP who had a history of salpingectomy, eight (80%) had no other risk factor. The underlying pathophysiology of a potential link between IMP and salpingectomy remains to be elucidated. Mellouli and colleagues [58] reported a case of IMP following salpingectomy without providing more information on the possibility of a connection. Similarly, You and colleagues [33] reported a case of IMP implanted on a scar where the uterus was previously perforated by a dislodged intrauterine device (IUD). IMPs present with a range of nonspecific signs and symptoms. Factors that influence the clinical course include the extent of myometrial involvement, the GA at the time of diagnosis, and the location of the GS [1] . Vaginal bleeding and lower abdominal pain are the most common initial symptoms ( Fig. 3 ). However, in our study, 18 patients (21.69%) were asymptomatic. IMP rupture presents as acute abdominal pain and/or signs of hypovolemic or hemorrhagic shock. This is most common in gestations exceeding 12 weeks GA [1] , [24] , [35] , [40] , [46] , [56] , [62] , [64] . Notably, although IMPs persisting beyond 12 weeks are uncommon and exceedingly rare beyond the second trimester, cases of IMPs with fetal survival have been reported; all of which required a cesarean section [34] , [75] . Two cases of biopsy-confirmed IMP with negative beta-HCG have been reported [28] , [76] . Dousias and colleagues [28] reported a patient with five-month menorrhagia who underwent a myomectomy after a diagnosis of “intramural myoma” was made on transvaginal ultrasound. Post-operative pathology findings were however those of an IMP. Similarly, Hsieh and colleagues [76] reported a woman with a five-month history of vaginal spotting who underwent curettage six months prior. Ultrasound showed an intramural cyst with embryo-liked components. The diagnosis of IMP was made postoperatively after the biopsied cyst was confirmed to be an IMP. It is possible that in both cases, the IMP spontaneously got demise, but failed to resorb completely. Post-partum diagnosis of IMP has also been reported [1] . In this case, part of the retained placental products was partially embedded in the myometrium. In our opinion, this description is identical to that of placenta percreta. A clear differentiation of partial IMP [as described by Memtsa et al. [1] ] from placenta percreta is necessary for clarification. Other types of EPs and spontaneous abortions present with symptoms similar to those of IMP ( Fig. 8 ). To avoid false positive findings, it is important to consider other factors such as clinical history, physical examination, laboratory and imaging findings while making diagnostic assessments. In the absence of advanced diagnostic tools, especially ultrasound imaging, early diagnosis of IMP is challenging. Historically, cases were diagnosed following laparotomy. In developed countries, ultrasound and MRI now play a central role in diagnosis. Fig. 8 Diagram illustrating locations and incidence of uncommon sites of ectopic pregnancy [Based on a model used by Chukus et al. [77] ]. Fig. 8 Diagram illustrating locations and incidence of uncommon sites of ectopic pregnancy [Based on a model used by Chukus et al. [77] ]. Ultrasound is a safe, non-invasive and widely available diagnostic modality; mostly used as the first-line diagnostic tool. In our study, 29 cases (35.36%) of IMP were correctly diagnosed or strongly suspected after the initial ultrasound. The most commonly described ultrasound features were as follows: 1. Empty uterus and cervical canal with the endometrial cavity not connected to the GS. 2. GS (with/without a fetal pole), a mass or amorphous echoes partially or completely surrounded by the myometrium. 3. Thin myometrial serosal surface usually measuring 3 mm or less. 4. Asymmetrically enlarged uterus with distorted contour. 5. High myometrial arcuate or peripheral vascular flow with low resistance on Doppler ultrasound, sometimes described as “ring of fire” Empty uterus and cervical canal with the endometrial cavity not connected to the GS. GS (with/without a fetal pole), a mass or amorphous echoes partially or completely surrounded by the myometrium. Thin myometrial serosal surface usually measuring 3 mm or less. Asymmetrically enlarged uterus with distorted contour. High myometrial arcuate or peripheral vascular flow with low resistance on Doppler ultrasound, sometimes described as “ring of fire” In addition, a sinus tract connecting the GS to the endometrium was visualized using high-resolution 2D [20] and 3D [20] , [49] transvaginal ultrasounds. IMPs that are collapsed, ruptured, or presenting with a GS without a fetal pole can easily mimic other conditions such as degenerating leiomyoma (12, 23, 27; 28) or gestational trophoblastic disease (GTD) [21] , [43] , [57] , [61] . Sherer and colleagues [73] reported a case of intramural choriocarcinoma confirmed on biopsy. Contrast-enhanced ultrasound (CEUS) has recently been used to confirm a suspected diagnosis of IMP. In a recent study, Liu and colleagues [78] described the following hallmark findings: 1. Early enhancement of the myometrial implantation site (about nine seconds post-injection), and that of a vessel-like area in the myometrium, indicating the myometrial origin of the GS blood supply. 2. The contrast agent enhanced for a long duration around the GS (between the eleventh to the twenty-eighth second after contrast agent’s injection), and washed out later than the myometrium. 3. Clear delineation with minimal enhancement of the myometrial boundary between the GS and the endometrium during the late enhancement phase and the washout stage (about 29 s post-injection). Early enhancement of the myometrial implantation site (about nine seconds post-injection), and that of a vessel-like area in the myometrium, indicating the myometrial origin of the GS blood supply. The contrast agent enhanced for a long duration around the GS (between the eleventh to the twenty-eighth second after contrast agent’s injection), and washed out later than the myometrium. Clear delineation with minimal enhancement of the myometrial boundary between the GS and the endometrium during the late enhancement phase and the washout stage (about 29 s post-injection). These findings are from a single case, and may vary with the GA. More studies need to be conducted to increase the knowledge about the diagnostic use of CEUS in IMP. 2D sonography cannot always clearly distinguish IMPs from interstitial or cornual EPs. However, with 3D sonography, there is a more accurate localization of the GS in relation to the uterine cornu or interstitium, and the endometrial cavity [2] , [14] , [49] . Furthermore, 3-D ultrasound in surface rendering mode enables a very clear visualization of the endometrial–myometrial junction, which facilitates the diagnosis of partial IMPs [79] . In our study, four (4.88%) cases involved the use of 3D ultrasound in their diagnostic evaluation [10] , [20] , [49] , [52] . Ruptured IMP typically presents with hemoperitoneum, characterized as a hypoechoic fluid collection with low-level internal echoes on ultrasound. The GS may be seen attached to or detached from the uterus, sometimes with the embryo or fetus floating within the abdomen. MRI can be used to supplement or confirm the diagnosis following an ultrasound. Because of its excellent spatial resolution, it clearly demonstrates the endometrium-myometrium border and the relationship between the endometrial cavity and the GS. It has been postulated as the gold standard for diagnosing IMP [56] , [80] . In our review, MRI was used to evaluate IMP in only 18 cases (21.95%). This could be explained by the scarcity of MRI in many parts of the world, especially in developing countries [81] , [82] , [83] . Furthermore, there are many cases that were reported before diagnostic MRI became commercially available. On MRI, the GS is typically hyperintense on T2 weighted, and isointense or hypointense on T1 weighted images [12] , [68] , [70] , [71] . In addition, the GS may be surrounded by tortuous vessels, which appear as tubular structures of high signal intensity [25] or signal voids [9] . Intramural GS or mass enhances on gadolinium contrast and the level and the distribution of the enhancement varies with the content. [68] , [25] . Furthermore, a connecting tract between the endometrial cavity and the GS presumably due to endometrial trauma can also be visualized and was reported in two cases [70] , [25] . CT scan utilization was reported in four cases. Although CT scan provides good spatial resolution, it should only be used under certain circumstances: a) when MRI is unavailable, b) when there is a very high index of suspicion, c) when only an amorphous mass or GS with a non-viable embryonic pole is seen on ultrasound. This is because of its high radiation dose and potential teratogenicity, especially in the first 8 weeks during organogenesis. In this study, CT findings were similar to those of USS and MRI. In addition, a mass with unclear flocculent enhancement in the uterine wall, reaching deep into the myometrium [43] and a mass with an obscured boundary [53] have been described. As preoperative ultrasonography, CT and MRI cannot exclude other types of EPs or GTD in some cases. In these situations, the diagnosis of IMP will be made using invasive methods such as hysteroscopy, diagnostic laparoscopy and postoperative biopsy. Diagnostic hysteroscopy was carried out in 22 (26.83%) cases after an equivocal transvaginal ultrasound scan in most cases. Recurrently reported findings include: 1. Empty uterus with no GS visualized (100%) 2. Visualization of the Fallopian tubes’ ostia, excluding tubal pregnancies (50%). 3. Uterine adhesions or fibrous tissues from previous instrumentation (13.64%) 4. Sometimes decidual hyperplasia/endometrial thickening or thinning. (31.82%) 5. Small bulge into the endometrial cavity. (18.18%) Empty uterus with no GS visualized (100%) Visualization of the Fallopian tubes’ ostia, excluding tubal pregnancies (50%). Uterine adhesions or fibrous tissues from previous instrumentation (13.64%) Sometimes decidual hyperplasia/endometrial thickening or thinning. (31.82%) Small bulge into the endometrial cavity. (18.18%) Furthermore, Auer-Schmidt and colleagues [5] describe a false tract connecting the GS to the endometrial cavity visualized on hysteroscopy. Diagnostic laparoscopy is usually the last in the chain of diagnostic evaluations before therapeutic surgical intervention. Findings include an asymmetrically enlarged uterus with a bulging or protruding mass. The mass is usually highly vascularized and covered by a very thin serosa that sometimes reveals the GS ( Fig. 9 ). In addition, hemoperitoneum can be seen in cases of slowly leaking or ruptured IMP. Fig. 9 Laparoscopy showing an IMP medial (black arrow) to the round ligament (white arrow). Note the hypervascularity of the bulging mass and the thin overlying serosa [from Ashraf et al. [23] ]. Reused with permission. Fig. 9 Laparoscopy showing an IMP medial (black arrow) to the round ligament (white arrow). Note the hypervascularity of the bulging mass and the thin overlying serosa [from Ashraf et al. [23] ]. Reused with permission. In cases where a questionable mass instead of a GS sac was visualized on imaging, the gold standard for diagnosis was histopathology after surgery. Findings were chorionic villi (with or without degenerative changes) surrounded by myometrial smooth muscles infiltrated by trophoblastic cells, plus no identifiable fallopian tubes. The management of IMP depends on the clinical condition, age of the patient, plans for future children, serum β-hCG value, size and location of the mass as well as hemodynamic status [84] . GA also plays a key role in treatment decisions. Reported options include conservative/expectant treatment, local or systemic methotrexate, local potassium chloride (KCl) injection and surgery. Historically, the treatment option was surgery usually with hysterectomy as IMPs were diagnosed after they had ruptured or intraoperatively during a diagnostic exploratory laparotomy. Advances in imaging with ultrasound and MRI have made it possible to diagnose IMP at a very early stage. Successful conservative management in which no treatment was required for the IMP to resolve was reported in three cases [1] , [19] , [49] . However, the most commonly used management was the surgical approach; including laparoscopic or laparotomic GS resection or hysterectomy. 11 (13.41%) had a hysterectomy of which one had biopsy confirmed choriocarcinoma [73] . Successful treatment with Uterine artery embolization (UAE) [29] and hysteroscopic guided curettage [5] have been reported. After the surgery, methylene blue instillation was used in one case to confirm the presence of a fistulous tract communicating between the endometrial cavity and the IMP [10] . It is important to note that no major complication was reported with the various surgical treatments. Medical management in the majority of cases, consisted of the use of local or systemic methotrexate (MTX). The successful use of local injection of KCl as a single agent [30] or with MTX [2] has been reported. Chemotherapeutic agents [Etoposide, Methotrexate, Actinomycin D, cyclophosphamide (MEA-CO)] have been described in cases of suspected intramural GTD/neoplasm pending the final pathology report [43] , [57] . Medical treatment requires follow-up for a long duration and in many instances fails, especially at advanced GA, necessitating surgery. The success of medical treatment reduces as the pregnancy progresses. In our study, out of the 19 who had medical treatment, 6 (31.58%) eventually underwent surgery owing to failure; for a success rate of about 68.42%. Treatment options that can be considered hybrid; involving the use of both surgical and medical management have also been described. After the surgical excision of the mass, local or systemic MTX is administered to discourage the growth of residual trophoblastic cells [8] , [16] , [20] , [63] . On the other hand, medical treatment can be given to shrink the mass or the GS before surgery is performed, hence, reducing the amount of bleeding and the size of uterine incision. However, in this study, most surgeries performed after a medical approach were as a result of treatment failure. Chida and colleagues described a case where bilateral UAE was performed three days prior to surgery [3] . Despite not expatiating the rationale for this approach, it is possible that using UAE to cut the blood supply of the IMP prior to surgery would make the mass shrink and reduce the amount of bleeding during surgery. Because of the increased risk of IMP recurrence, women with preserved uterus following IMP removal surgeries should be advised to do an ultrasound scan as soon as the GS can be visualized (5–6 weeks) in all future pregnancies [1] to enable early diagnosis and treatment.

An ectopic pregnancy (EP) is a developing gestation outside the uterus, most commonly in the fallopian tubes. Intramural pregnancy (IMP) refers to a very uncommon EP located within the uterine wall, partially or completely surrounded by myometrium, and separate from the uterine cavity and fallopian tubes or the round ligaments. The trophoblast invades beyond the endometrial-myometrial junction, with the gestational sac (GS) partially or completely implanted within the uterine myometrium. It is the rarest type of EP [1] , and accounts for approximately 1% of all EPs [2] . It was first reported by Theodore Doderlein in 1913 on a woman with adenomyosis. IMPs can easily be misdiagnosed, especially at an advanced gestational age (GA), when the endometrial lining/cavity becomes effaced. There are various hypotheses about its etiology and pathophysiology, including the creation of a false tract secondary to trauma to the endometrial cavity. Nonetheless, IMP cases without prior uterine trauma or surgery have been reported. Urgent recognition and management of IMP is vital, as delayed diagnosis can result in rupture with life-threatening hemorrhage. The mortality rate is nearly 2.5% [3] . There have only been a few reported cases or small series of IMPs in the literature. However, there is still limited awareness and knowledge among sonographers and interpreting physicians (radiologists, obstetricians, gynecologists, emergency physicians). Furthermore, no established clinical practice guidelines exist to direct diagnosis, management and follow-up. We reviewed individual case reports on IMP and summarized the clinical and imaging features as well as management strategies for this rare and potentially devastating obstetric condition.

All three authors have NO conflict of interest to declare.