Early sonographic signs of placenta accreta spectrum disorder: A pictorial review

Placental accreta spectrum and caesarean scar pregnancy

The finding of a caesarean scar pregnancy (CSP) may be one of the first signs of PAS on a first-trimester ultrasound as demonstrated in Figure 1. As caesarean section rates continue to rise, there is a corresponding increase in the incidence of CSPs being identified in women with previous caesarean deliveries. Several studies have described CSPs as a precursor of PAS.^3,11–13 Timor-Tritsch et al. ¹¹ explored the natural progression of CSP in ten women, diagnosed prior to 10 weeks’ gestation, who chose to continue their pregnancies. All ten patients had developed sonographic features of morbidly adherent placenta in the second trimester, underwent caesarean hysterectomy at delivery and placenta percreta was confirmed at histological examination. Based on this, the authors proposed that PAS is a continuum of CSP, eventually resulting in a morbidly adherent placenta that is classically seen with advancing gestation.

OPEN IN VIEWER

Figure 1.

A case of a 40-year-old patient with one previous caesarean section who was diagnosed with a CSP at 6 weeks of gestation. She was asymptomatic at this time and chose to continue the pregnancy. She was delivered at 34 weeks via caesarean section and hysterectomy. (a) Ultrasound sonography (USS) at 6 weeks demonstrates a low-lying gestational sac located within the caesarean niche. Indicates caesarean niche. Identifies the internal os. (b) Transabdominal USS illustrating a low-lying gestational, extending into the caesarean niche. (c) Thin anterior myometrial wall thickness at 6 weeks. (d) Transabdominal USS at 31 weeks featuring placental bulge, lacunae and loss of clear zone.

There is currently no universally accepted definition or standardised approach for recognizing and reporting the sonographic features of CSP. In 2022, Jordans et al. ¹⁴ proposed a sonographic reporting system for the diagnosis of CSP within the first 12 weeks of gestation, developed using a modified Delphi method. CSP was defined as a pregnancy implanted in, or in close contact with, the caesarean section niche. The consensus identified 10 basic and 8 advanced sonographic features to guide assessment, as outlined in Table 2.

OPEN IN VIEWER

Table 2.

Modified from Jordans et al. ¹⁴

Basic assessment

Advanced assessment

Sagittal plane: • Location of gestation sac • Presence of cardiac activity • Location of placenta in relation to the uterine scar • Presence of placenta praevia • Invasion of placenta or trophoblast into the myometrium • Presence of caesarean niche • Anterior myometrial thickness • Adjacent myometrial thickness • Bulging of gestational sac beyond the serosa towards the bladder or bowels • Exact amount of protrusion of gestation sac beyond the uterine cavity line and serosal line

Sagittal plane (colour Doppler): • Circular flow around the gestational sac • Lining of endometrium covering the caesarean niche may be relevant to detect PAS • Location of placental/trophoblast tissue • Placental ingrowth and its relation to the myometrium/bladder • Distance between the vessels of the placenta and serosa (indicates chance of developing PAS) • Use of pulsed Doppler and 3D Doppler in the research setting Transverse plane: • Distance between the gestation sac and the uterine arteries • Level of protrusion in relation to outer serosal contour

To facilitate accurate description of CSP location relative to the endometrium, Jordans et al. ¹⁴ proposed a classification system categorizing CSP into three types. This system is based on two imaginary reference lines: the serosal line (SL), drawn along the outer surface of the lower anterior uterine wall, and the uterine cavity line (UCL), drawn at the endometrial–myometrial junction. The types are defined as follows (Figure 2):

This classification not only supports standardization in CSP diagnosis and reporting but also provides prognostic value, aiding in more informed counselling and management decisions.

OPEN IN VIEWER

Figure 2.

Red line = UCL, blue line = SL. (a) Type 1 CSP–gestational sac crosses the UCL. (b) Type 2 CSP–gestational sac embedded within the myometrium and not crossing either the UCL or SL. (c) Type 3 CSP–gestational sac extends beyond the SL.

Measurement of anterior myometrial thickness, defined as the distance from the CSP to the serosal surface, should be performed upon identification of CSP. This parameter has emerged as a highly predictive first-trimester sonographic marker for PAS. The relevance and implications of this finding are explored in greater detail in the following sections of this manuscript.

Advanced imaging settings (Table 2) incorporate colour Doppler to evaluate vascular flow surrounding the gestational sac and to assess placental/trophoblastic tissue as well as the endometrium. Although no causal relationship has been established between first-trimester diagnostic ultrasound and adverse fetal outcomes, ¹⁵ the absence of evidence does not equate to evidence of absence. Accordingly, the As Low As Reasonably Achievable principle (ALARA) should be followed. ¹⁶ This entails using colour Doppler only when clinically indicated and for the shortest duration necessary. In the context of diagnosing CSP, the use of colour Doppler is justified, given the significant clinical benefit of early detection. The International Society of Ultrasound in Obstetrics and Gynaecology (ISUOG) recommends maintaining a thermal index (TI) of ⩽1.0 and limiting exposure time, ideally to under 5–10 minutes. Furthermore, fetal safety is unlikely to be compromised if the embryo lies outside the Doppler beam. ¹⁷ Therefore, when assessing placental and myometrial structures, the Doppler box should be positioned to avoid inclusion of the gestational sac or embryo.

If transvaginal ultrasound indicates abnormal placentation, it is crucial to assess the location of the gestational sac in relation to the endometrial cavity (Figure 3). This assessment is vital for guiding patient counselling, as pregnancies located outside the endometrial cavity are at a higher risk of growing outwards, leading to rupture at early gestations and therefore poorer patient outcomes.

OPEN IN VIEWER

Figure 3.

Transvaginal ultrasound demonstrating a low-lying gestational sac implanted within the caesarean niche at 5 weeks.

Timor-Tritsch et al. ¹⁸ emphasised the importance of distinguishing between a low-lying gestational sac as a CSP or a continuing intrauterine pregnancy. They proposed that this differentiation could be achieved by assessing the relative location of the centre of the gestational sac in relation to the midpoint of the uterus on a sagittal plane, measured between the external cervical os and the uterine fundus. Their findings indicated that most CSPs are located distal to this midpoint, whereas pregnancies likely to progress as intrauterine are typically located proximal to it. Figure 4 demonstrates this concept on a case of CSP (Figure 4(a)) and an intrauterine pregnancy (Figure 4(b)).

OPEN IN VIEWER

Figure 4.

Applying the principle from the Timor-Tritsch study, in (a) the gestational sac is distal to the midpoint of the uterus, and this pregnancy was ultimately diagnosed as a CSP, whereas (b) is of an early intrauterine pregnancy, and the gestational sac is above the midpoint of the uterus.

If assessment of the entire endometrial cavity is difficult through transvaginal scanning, then a transabdominal ultrasound is often the most effective method for obtaining a clear sagittal view of the entire uterus. The importance of the transabdominal scan is demonstrated in the case shown in Figure 5. A 45-year-old woman with a history of two previous caesarean sections presented at 9 weeks of gestation to a tertiary centre for an early pregnancy scan. The initial impression from the transvaginal ultrasound was an intrauterine pregnancy that was low in the cavity, with trophoblastic tissue adjacent to the niche, increasing the possibility of PAS. However, the transvaginal ultrasound did not assess the endometrial cavity in length, and a subsequent transabdominal scan revealed that the pregnancy was in fact not within the endometrial cavity. While information from the transvaginal imaging alone, in this case, may have included counselling the patient regarding continuing the pregnancy, including delivery with a caesarean section hysterectomy, the additional information from the transabdominal imaging indicated an increased risk of rupture as the pregnancy progresses. The patient had already opted for a termination. She underwent a suction dilation and curettage, which was complicated by a 1500-ml blood loss. The significant bleeding supports the ultrasound finding of early PAS. This case highlights the importance of interpreting ultrasound findings by evaluating the entire endometrial cavity, which may require transabdominal imaging. Without this comprehensive assessment, a pregnancy located outside the endometrial cavity may go unrecognised.

OPEN IN VIEWER

Figure 5.

Images (a) and (b) are transvaginal scans showing a low-lying gestational sac with abnormal placentation at the site of the caesarean niche. The operator measured the anterior myometrium as >5 mm; however, in retrospective review of these images, it is likely this was trophoblastic tissue around the gestational sac with no overlying myometrium. Images (c) and (d) are transabdominal scans that provide additional context, clearly demonstrating that the gestational sac is entirely within the caesarean niche and not within the endometrial cavity, thereby increasing the risk of rupture as the pregnancy advances. Highlights the endometrium, clearly seen as empty in images (c) and (d).

It is also crucial to determine whether a low-lying gestational sac will continue to develop into the endometrial cavity or grow outward, leading to uterine rupture. This determination is often challenging with a single ultrasound scan. Consequently, serial scanning is essential in these cases, as it provides the only reliable method for monitoring and detecting the progression of the pregnancy. Figures 6–8 illustrate how serial scans can be instrumental in tracking the clinical course of a low-lying gestational sac. The case presented in Figure 6 progressed to severe PAS, necessitating a hysterectomy in the second trimester. In contrast, the case in Figure 7 developed into an intrauterine pregnancy. Notably, the patient in Figure 7 did not have a history of caesarean section, placing them at a lower risk of developing PAS.

OPEN IN VIEWER

Figure 6.

(a) Low-lying gestational sac at 13 weeks with trophoblastic tissue invading into the myometrium and extending to the cervix. (b) Placental lacunae at 13 weeks . (c) TA scan at 16 weeks showing pregnancy and placental tissue growing outside the endometrial cavity().

OPEN IN VIEWER

Figure 7.

(a) Transabdominal ultrasound at 12 weeks demonstrating a low-lying gestational sac in contact with the endometrial cavity, appearing to protrude into the myometrium in a patient with no prior history of caesarean section. (b) On transvaginal imaging, the anterior myometrium is very thin. However, as can also be seen in image (c), the placenta is posterior and has no morphological features of PAS. In addition, there is no disruption of the cervix. (d) Placenta low-lying at 20 weeks and remained praevia; otherwise, the placenta remained morphologically normal throughout the pregnancy. The patient was delivered via caesarean section at the end of the third trimester, with no clinical features of PAS evident.

OPEN IN VIEWER

Figure 8.

(a) The gestational sac appears to be located well within the endometrial cavity. Caesarean niche is visible below the gestational sac and does not appear to be involved in this image, with the gestational sac giving the impression that the pregnancy will continue as intrauterine. (b) Growth of trophoblastic tissue into the caesarean niche at 10 weeks. Even at this early gestation, advanced signs of PAS, such as lacunae, can be seen. (c) and (d) Abnormal trophoblast containing vascular lacunae and bulging into the surrounding myometrium at 12 weeks.

Serial scanning is particularly important in patients with risk factors for PAS, as early ultrasound signs may be subtle or absent in the first trimester, particularly before 8 weeks, and may therefore be overlooked or not recognised. Conducting serial scans allows for the identification of these cases in the late first or early second trimester, when abnormal placental invasion may become more apparent. Given that many women undergo a nuchal translucency scan at 12–13 weeks, this presents an optimal opportunity for clinicians to recognize features suggestive of PAS. It is therefore essential for clinicians to be familiar with these ultrasound findings, so they can identify PAS in at-risk patients, even if the initial scan appeared normal.

Figure 8 illustrates the case of a patient with a history of one previous caesarean section, whose pregnancy initially appeared to be a normal intrauterine pregnancy at 6 weeks. However, despite this early assessment, the patient experienced persistent vaginal bleeding. Serial ultrasounds subsequently revealed progressive invasion of trophoblastic tissue through the uterine serosa and into the cervical tissue by 10 weeks, raising suspicion of PAS, which was confirmed at 12 weeks. The patient ultimately underwent a wedge resection in the second trimester. A wedge resection is performed by first injecting a vasoconstrictor, such as vasopressin, into the myometrium around the ectopic pregnancy to minimize intraoperative bleeding. ¹⁹ An incision is then made through the myometrium to access and remove the gestational tissue, after which the affected area is excised in a wedge-shaped fashion and the bleeding points are sealed with cautery. ¹⁹ The defect is then oversewn. The procedure can be performed either laparoscopically or through open surgery, depending on patient and surgeon factors. This technique facilitates removal of the ectopic pregnancy while preserving healthy uterine tissue, thereby maintaining fertility potential. The case described in Figure 8 underscores the critical role of serial scans, particularly in symptomatic patients, in facilitating early diagnosis and appropriate management of PAS.

Anterior myometrial wall thickness

To assess whether CSPs implanted on a dehiscent scar ‘niche’ behave differently compared to those on top of a well-healed scar, Agten et al. ⁴ conducted a retrospective study of 17 patients diagnosed with CSP between 5 and 9 weeks of gestation who opted to continue their pregnancies. They found that all patients with CSP implanted in a caesarean niche displayed signs of PAS on ultrasound and required hysterectomy. They also demonstrated significantly lower myometrial thickness of <2 mm in the patients who developed PAS. There were no sonographic signs of PAS in the CSP group implanted on a well-healed scar, though one patient did have a hysterectomy for intractable haemorrhage, and partial accreta was subsequently histologically confirmed. The myometrial thickness for this patient was also 2 mm in the first-trimester ultrasound. It can be concluded from this study that patients with CSPs who have a myometrial thickness of 2 mm or less on first-trimester ultrasound are at higher risk of developing PAS.

In a patient with a low-lying gestational sac and a history of caesarean delivery, the smallest anterior myometrial thickness, measuring the distance from the CSP to the serosa, has emerged as a highly predictive first-trimester sonographic indicator for PAS. The reliability of this parameter lies in its consistent association with PAS across multiple studies,^3,12,20 with Rac et al. ¹² finding it to be the only biometric measurement that significantly improved detection of PAS.

The exact threshold below which the anterior myometrial wall thickness can improve diagnostic accuracy for PAS has been the subject of numerous investigations. As previously mentioned, Agten demonstrated that a myometrial wall thickness of <2 mm was significantly associated with PAS. Further research by Moschos et al. ²⁰ examined various thresholds and found, in their retrospective study of 56 patients, that a threshold of <5 mm was 100% sensitive but only 74.3% specific, a threshold of <3 mm was 88.9% sensitive but achieved 100% specificity and a threshold of <4 mm was 94.9% sensitive and 97.1% specific. This highlights the challenges in selecting the most appropriate threshold to balance sensitivity and specificity in early diagnosis of PAS.

Though a consensus has not yet been reached regarding the optimal threshold for anterior myometrial wall thickness in diagnosing PAS, a measurement of <5 mm should raise concern. Patients with this finding should receive adequate counselling and careful monitoring with subsequent ultrasound for the potential development of PAS.

Ultrasound features across all gestations

Several ultrasound indicators for PAS have been suggested, but many lack clear definitions and show varying levels of sensitivity and specificity. To improve the consistency of these ultrasound markers, the European Working Group on Abnormally Invasive Placenta (EW-AIP) developed a Delphi consensus proposal in 2015. ²¹ This proposal aimed to standardize the ultrasound descriptions for each sign. The EW-AIP consensus statement has not specified the sonographic signs for gestational age.

In 2023, Jauniaux et al. ²² revisited the existing criteria using a modified Delphi process. Their study reassessed the previously established signs and introduced eight additional markers identified through a comprehensive literature review. While consensus was reached on seven of the original ultrasound signs, agreement was not achieved for any of the newly proposed signs. This ongoing heterogeneity in expert opinion underscores the persistent diagnostic challenges associated with PAS.

Table 3 presents the sonographic features proposed in the EW-AIP consensus, with the seven signs that achieved consensus in the 2023 study highlighted in bold.

OPEN IN VIEWER

Table 3.

Standardised features of PAS were modified from the EW-API consensus statement.

2D	3D
Greyscale: • Loss of clear zone • Abnormal placental lacunae • Bladder wall interruption • Myometrial thinning • Placental bulge • Focal exophytic mass 2D colour Doppler: • Uterovesical hypervascularity • Subplacental hypervascularity • Bridging vessels • Placental lacunae feeder vessels	• Intraplacental hypervascularity • Placental bulge • Focal exophytic mass • Uterovesical hypervascularity • Bridging vessels

Many of the ultrasound signs of PAS are subtle in the early first trimester (<9 weeks’ gestation) and may be easily overlooked. A review of our cases indicates that the most consistently observed early features include a low-lying gestational sac, a gestational sac positioned outside the endometrial cavity and a thin anterior myometrium. Other PAS-related ultrasound findings, which are discussed in detail below, may be present but are often inconsistent or only subtly detectable at this stage. Therefore, it is crucial that clinicians are not falsely reassured by the absence of these features on early pregnancy ultrasound, as PAS may still develop and become more apparent in subsequent scans.

Many of these signs described by EW-AIP can be observed in the late first trimester (9 weeks’ gestation and above) and early second trimester. They include placental lacunae, myometrial thinning, subplacental hypervascularity, placental bulge, bladder wall interruption and bridging vessels (Figures 9-12).

OPEN IN VIEWER

Figure 9.

(a) No measurable overlying myometrium at 9 weeks. (b) Thin (<1 mm) anterior myometrium at 5 weeks.

OPEN IN VIEWER

Figure 10.

Images depicting placental lacunae at various gestational ages. (a) 10 weeks. (b) 12 weeks. (c) 16 weeks. (d) 23 weeks.

OPEN IN VIEWER

Figure 11.

Images of subplacental hypervascularity in the first and early second trimester. (a) 9 weeks. (b) 10 weeks. (c) 16 weeks. Also demonstrates a feeder vessel.

OPEN IN VIEWER

Figure 12.

(a) USS at 16 weeks highlighting placental bulge and hypervascularity in a patient with a history of one previous caesarean section. (b) USS of the same patient at 23 weeks demonstrating a significant placental bulge and lacunae.

To evaluate the diagnostic accuracy of these signs in the late first trimester (11–14 weeks), Cali et al. conducted a retrospective analysis of prospectively collected data from 188 women at risk for PAS (defined by placenta praevia and at least one previous caesarean delivery) who underwent an ultrasound between 11 and 14 weeks. They assessed four specific signs: loss of the clear zone, placenta lacunae, bladder wall interruption and uterovesical hypervascularity. ¹³ The summarised findings are presented in Table 4.

OPEN IN VIEWER

Table 4.

Summarised from Cali et al.

Sign	Sensitivity %	Specificity %
Loss of clear zone	84.3	81.9
Placenta lacunae	78.3	81
Bladder wall interruption	75.9	99.1
Uterovesical hypervascularity	50.6	100

These findings underscore the overall accuracy of these four ultrasound signs in diagnosing PAS during the first trimester. Utilizing multiple signs enhances diagnostic precision. Moreover, the study achieved a 90% accuracy rate in detecting placenta percreta through the 11 to 14-week ultrasound scans, highlighting the effectiveness of early pregnancy screening in obtaining critical diagnostic insights for PAS and the positive implications this has on achieving safe patient outcomes.

Role of magnetic resonance imaging in placental accreta spectrum

Antenatal magnetic resonance imaging (MRI) serves as a valuable complement to ultrasound in the diagnosis of PAS, as it offers precise anatomical descriptions of the placenta’s invasion area. ¹² In addition, MRI is a safe imaging modality during pregnancy, with no teratogenic effects on the developing fetus at any gestational age. In 2019, the Society of Abdominal Radiology (SAR) and the European Society of Urogenital Radiology (ESUR) issued a joint statement outlining recommended MRI findings for PAS to standardize the diagnostic criteria and reporting practices, enhancing consistency and accuracy in its identification. ²³

A 2022 systematic review of 18 articles compared the diagnostic accuracy of ultrasound and MRI for PAS, finding no statistically significant difference between the two imaging modalities. ²⁴ Given that MRI is more expensive and can be an uncomfortable experience for patients, ultrasound remains the preferred imaging modality in assessing and diagnosing PAS, with MRI serving best as an adjunctive tool.