Cesarean scar pregnancy

INTRODUCTION — Cesarean scar pregnancy (CSP) refers to a pregnancy that is implanted on or in a scar from a prior cesarean birth. A pregnancy implanted on or in a myomectomy scar (also referred to as intramural pregnancy) can also occur; for the purposes of this topic, the term CSP will be used for all hysterotomy scar pregnancies. If unrecognized or inadequately managed, CSP can lead to severe fetal and maternal morbidity (eg, uterine rupture, hemorrhage) and mortality.

CSP is likely a precursor to, and shares common histology with, placenta accreta spectrum (PAS). The two conditions may represent a continuum of the same disease, with CSP being a diagnosis of the first (and early second) trimester, and PAS being diagnosed later in pregnancy (second trimester and beyond). Like PAS, the rate of CSP parallels that of cesarean birth and is expected to rise as the rate of cesarean birth increases and as the diagnostic accuracy of imaging for CSP improves.

By contrast, CSP is distinct from ectopic pregnancy as CSP occurs within uterine cavity and can result in a live born infant, which rarely, if ever, occurs in an ectopic pregnancy. Thus, treatments appropriate for ectopic pregnancy should not be used indiscriminately to treat or manage CSP.

This topic will focus on issues related to CSP. The diagnosis and management of PAS, as well as other abnormally implanted pregnancies (eg, abdominal, cervical, tubal, heterotopic), are discussed in detail separately.

●PAS (see "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences" and "Placenta accreta spectrum: Management")

●Abdominal pregnancy (see "Abdominal pregnancy")

●Cervical pregnancy (see "Cervical pregnancy: Diagnosis and management")

●Tubal and heterotopic pregnancies (see "Ectopic pregnancy: Epidemiology, risk factors, and anatomic sites" and "Ectopic pregnancy: Clinical manifestations and diagnosis" and "Ectopic pregnancy: Choosing a treatment")

CLASSIFICATION — There are two main types of CSPs [1]:

●"On-the-scar" (type 1) – Implantation of the CSP on the well-healed scar of a previous cesarean birth (also termed "endogenous" implantation).

●"In-the-niche" (type 2) – Implantation of the CSP within the defect or "niche" of an incompletely healed scar (also termed "niche pregnancy," or "exogenous" implantation).

Clinical outcomes appear to be more favorable for "on-the-scar" compared with "in-the-niche" CSP; however, most ultrasound reports and published studies do not specify CSP type. (See 'Natural history' below and 'Expectant management' below.)

Other classification systems have been described [2,3].

INCIDENCE — CSP occurs in approximately 1 in 2000 pregnancies and accounts for approximately 6 percent of abnormally implanted pregnancies among patients with a prior cesarean birth [4-6]. The incidence appears to be rising; in a 10-year review at one institution in China, the number of patients with CSPs increased from three in the initial five years (2004 to 2008) to 19 in the following five years (2009 to 2013) [7].

It is the author's impression that, like placenta accreta spectrum (PAS), incidence correlates with the number of previous cesarean deliveries; however, there are no data to support or refute this theory. There are also no data regarding the incidence of CSP by type (ie, "on-the-scar" [type 1] versus "in-the-niche" [type 2]). (See 'Classification' above.)

The incidence of pregnancies implanted at the site of a prior myomectomy scar is rare [8,9].

Multiple gestations with a coexistent CSP have also been described [10].

PATHOGENESIS — The mechanism for implantation of a CSP is unclear; various theories include: the endogenous migration of the embryo through either a wedge defect in the lower uterine segment or a microscopic fistula within the scar [11,12]; invasion of placental villi into the uterine wall at a point of scar dehiscence [13-15]; low oxygen tension of scar tissue attracting implantation of the fertilized oocyte [16].

Some experts believe that CSP is a precursor to, and shares a common histology with, placenta accreta spectrum (PAS) and that they are a continuum of the same disease [17-19]. Both involve the placenta attaching to or invading the myometrium, almost always in an area of scarring caused by previous uterine surgery. While PAS is typically a diagnosis made in the second trimester, some experts believe that sonographic findings may be present in the first trimester, further strengthening the association of PAS with CSP [20-24].

Representative studies supporting this theory include:

●In a literature review including seven studies with 551 pregnant patients at risk for PAS, at least one ultrasound marker suggestive of PAS (eg, low anterior implantation of the gestational sac, placental lacunae, reduced myometrial thickness) was detected in 91 percent of patients with a confirmed diagnosis [23]. Low anterior implantation of the placenta was the most commonly seen early ultrasound sign suggestive of PAS (83 percent of cases), although its individual predictive accuracy was not high.

●In one series including 10 patients with a diagnosis of CSP made prior to 10 weeks gestation, all patients exhibited signs of PAS by the second trimester [17].

●In a retrospective study including 58 cases of CSP and PAS where histologic specimens were independently reviewed by two pathologists who were blinded to each other's findings, they agreed that, on the basis of histology, it was impossible to differentiate between the two diseases [18].

The pathogenesis and ultrasound findings of PAS are discussed in detail separately. (See "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences", section on 'Pathogenesis' and "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences", section on 'First-trimester ultrasound examination'.)

NATURAL HISTORY — CSPs are associated with high morbidity early in pregnancy (eg, hemorrhage, uterine rupture, placenta accreta spectrum [PAS]) and even maternal mortality. However, some CSPs may continue to a viable gestation. Series describing outcomes of expectantly managed CSPs all involve small case numbers and high hysterectomy rates.

●Uterine rupture – CSPs are predisposed to early (ie, first- or second-trimester) uterine rupture [25]. In a meta-analysis including 17 studies (69 patients) evaluating the outcomes of patients with CSP that were managed expectantly, early uterine rupture in patients with and without embryonic/fetal cardiac activity occurred in 9.9 to 13.4 percent (95% CI 2.9-20.4 and 2.7-30.3) of patients, respectively [25].

In our experience, uterine rupture occurs more frequently in those implanted "in-the-niche" (ie, type 2) rather than "on-the-scar" (type 1).

●Delivery of a live born neonate and hysterectomy rates – Rates of delivery of live born neonates and hysterectomy vary.

In one series including 10 patients with a diagnosis of CSP made prior to 10 weeks gestation and described above (see 'Pathogenesis' above), nine patients (90 percent) delivered live born neonates between 32 and 37 weeks of gestation all of whom delivered by cesarean hysterectomy; mean blood loss was 1560 mL (range 300 to 6000 mL) and two patients required a blood transfusion [17]. One patient experienced cervical shortening and intractable vaginal bleeding prompting termination (with hysterectomy) at 20 weeks gestation. Placenta percreta was the histopathological diagnosis in all 10 cases.

Lower rates have been reported [26,27]. In a retrospective study including eight patients diagnosed with a live CSP diagnosed prior to 14 weeks gestation and choosing expectant management, five patients (63 percent) delivered a live-born neonate between 29 and 36 weeks of gestation, of whom three (60 percent) required cesarean-hysterectomy [27]. The remaining three patients experienced a subsequent fetal demise; no patients with fetal demise required hysterectomy.

Outcomes may be more favorable for "on the scar" compared with "in-the-niche" CSPs, but data are limited as most studies do not specify CSP type. (See 'Expectant management' below.)

●Enhanced myometrial vascularity – CSP may be associated with the development of enhanced myometrial vascularity (EMV; a distinct entity from arteriovenous malformation, which occurs in nonpregnant patients) [28-30], and treatments that leave the pregnancy in place (eg, balloon catheters) may increase this risk (see 'Medical therapy' below). In one retrospective study including 60 patients with CSP, five patients (8 percent) with persistent and/or severe bleeding were subsequently diagnosed with an EMV; further management with uterine artery embolization (UAE; four patients) or hysterectomy (one patient) was required [31].

Other outcomes associated with CSPs that progress to, or near, term are discussed in detail separately. (See "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences", section on 'Consequences'.)

RISK FACTORS

History of CSP — The risk of recurrent CSP may be more common than previously thought and ranges from 5 to 40 percent; this is based mostly on single case reports or series with few patients [32,33]. The true rate is unknown.

In one meta-analysis including 44 studies (3598 patients) evaluating subsequent reproductive outcomes in patients with a prior CSP, recurrence occurred in 17.6 percent of patients [32]. When stratified according to management type, recurrence of CSP was more common in patients undergoing surgical compared with nonsurgical management (90 of 428 patients [21 percent] versus 21 of 138 patients [15 percent], respectively). One limitation of this study is that subgroup analyses according to the type of surgical (eg, uterine curettage, uterine resection) or nonsurgical (eg, systematic methotrexate [MTX], local injection of MTX or potassium chloride [KCl], uterine artery embolization) procedure were not performed.

Other — Other factors that may contribute to the risk of CSP in patients with a prior cesarean birth include [4,8,34]:

●Other previous uterine surgery (eg, dilation and curettage [D&C], endometrial ablation)

●Manual removal of the placenta

●In vitro fertilization

CLINICAL PRESENTATION — Approximately one-third of patients with CSP are asymptomatic at the time of ultrasound diagnosis [35].

For patients with symptoms, vaginal bleeding is the most common symptom and usually presents in the late first or early second trimester. However, symptoms can range from vaginal bleeding with or without abdominal/pelvic pain to uterine rupture with hypovolemic shock [34,36-38].

In one retrospective study including 22 patients with CSP, the presenting symptom was vaginal bleeding alone (86 percent), vaginal bleeding with abdominal pain (9 percent), or abdominal pain alone (4.5 percent) [7]. This is consistent with a subsequent retrospective study including 20 patients with CSP in which most patients (90 percent) presented with either vaginal bleeding and/or abdominal pain; two patients (10 percent) were asymptomatic at the time of diagnosis [39].

DIAGNOSIS

Definitive diagnosis — CSP is a diagnosis made based on first- (or early second-) trimester transvaginal ultrasound (TVUS) findings of a pregnancy implanted on or in a prior hysterotomy scar/niche and confirmed at the time of surgery with histologic findings consistent with placenta accreta spectrum (PAS) disorder. (See "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences", section on 'Postpartum placental histology'.)

TVUS combined with color Doppler evaluation is the preferred imaging modality in patients with CSP and has superior resolution compared with transabdominal ultrasound.

Findings suggestive of the diagnosis — Early ultrasound findings suggestive of CSP in a patient with a prior uterine scar include:

●Low anterior position of the gestational sac at ≤7 weeks – A low anterior position of the gestational sac is an ultrasound marker for CSP only at very early gestational ages (ie, between five and seven weeks of gestation) [40]. After seven to nine weeks of gestation, the gestational sac "pushes" into the uterine cavity (leaving behind the implanted placenta) as this is the only available place for the sac (with the embryo/fetus) to expand into.

The following method can be used to determine location of the gestation sac (image 1) [41]:

•On a panoramic, longitudinal, sagittal scan, divide the uterus in half by an imaginary line perpendicular to its longitudinal axis and determine the location of the gestational sac related to this line.

•Mark the center of the gestational sac.

•If the center of the sac is above the aforementioned line (closer to the uterine fundus), it is most likely an intrauterine pregnancy.

•If the center of the sac is below the line (closer to the cervix), it is a CSP (or, in rare cases, a cervical pregnancy). (See 'Differential diagnosis' below.)

When applied at or before seven weeks of gestation, this method has reliable statistical performance (sensitivity: 93 percent, specificity: 98.9 percent, positive predictive value: 96.4 percent, negative predictive value: 97.9 percent) [41].

An exception is for patients with a prior classical incision or fundal incision (during myomectomy). For such patients, the CSP is likely to develop in the upper, rather than lower, portion of the uterus.

●Other findings – Other findings are consistent with PAS and include one or more of the following [40,42,43]:

•Low-lying anterior placenta or placenta previa – A low-lying anterior placenta or placenta previa (image 2) increases the risk of persistent placenta previa and PAS in patients with a prior cesarean birth. As with other patients with PAS, attention should be directed at determining the extent of invasion into adjacent pelvic structures (eg, bladder, cervix, parametria), if present [44,45].

•Placental lacunae (image 3).

•Thin (<3 mm) or no intervening myometrium/clear space between the placenta and bladder or anterior uterine serosa (image 4). This distance is also termed residual myometrial thickness [RMT] and is the only quantifiable ultrasound marker of CSP.

•Irregular or disrupted bladder line or anteriorly bulging sac into the bladder (image 5).

•Increased vascularity at the bladder/placental interphase (image 6).

When present, these findings (with the exception of the low anterior position of the gestation sac) become increasingly evident as the pregnancy progresses from five to seven weeks of gestation to the end of the first trimester [46,47]. (See "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences", section on 'Ultrasound findings'.)

Other imaging techniques — Three-dimensional ultrasound rarely contributes to the diagnosis of CSP; however, if expertly used, it may aid in the diagnosis in some patients [48,49].

There is no evidence to support the routine use of other imaging modalities, such as pelvic magnetic resonance imaging (MRI), to diagnosis CSP [50]. However, in cases in which ultrasound imaging is indeterminate or the ultrasound images are suboptimal, MRI may be used for further evaluation.

DIFFERENTIAL DIAGNOSIS

●Cervical pregnancy – Cervical pregnancy is a rare form of ectopic pregnancy and can be differentiated from CSP by the finding of a gestational sac or placenta within the cervix on transvaginal ultrasound (TVUS). (See "Cervical pregnancy: Diagnosis and management", section on 'Transvaginal ultrasound'.)

Unlike CSP, cervical pregnancy almost always occurs in patients without a history of a prior cesarean birth. In our experience, patients with a history of prior uterine surgery and with a finding of gestational sac in a low anterior position and/or a low-lying anterior placenta have a CSP and not a cervical pregnancy.

●Early pregnancy loss – In patients undergoing an early pregnancy loss (ie, miscarriage, spontaneous abortion), a single ultrasound may demonstrate a gestational sac or placenta in the lower uterine segment; fetal cardiac activity is usually absent. In patients with a prior uterine scar, this may be confused with a CSP. However, for patients undergoing a pregnancy loss, subsequent ultrasound examinations will typically show the pregnancy even lower in the uterus, in the cervical canal, or an empty uterus (in the case of a complete abortion). Additional ultrasound findings that support a diagnosis of early pregnancy loss are discussed separately. (See "Pregnancy loss (miscarriage): Ultrasound diagnosis".)

●Intrauterine pregnancy – While an intrauterine pregnancy with an unusually low implantation of the gestational sac may raise suspicion for a CSP, CSP can essentially be ruled out in patients without a history of previous cesarean birth.

In a literature review including 751 cases of CSP, misdiagnosis of CSP as a cervical ectopic pregnancy or miscarriage occurred in 13.6 percent of patients; the delay in diagnosis potentially resulted in a large number of additional complications (eg, hemorrhage, shock, hemoperitoneum) with some requiring hysterectomy for definitive surgical management [51].

MANAGEMENT — This section describes the management of CSP in the first and early second trimesters. At later gestations, these pregnancies are managed the same as placenta accreta spectrum (PAS), which is discussed in detail separately. (See "Placenta accreta spectrum: Management".)

As with PAS, patients with CSP should transferred, whenever possible, to a center of excellence for PAS, or a tertiary care hospital in which maternal-fetal medicine specialists, a variety of treatment options (eg, uterine artery embolization [UAE]), and blood bank services are available.

Hemodynamically unstable patients — A patient with hemorrhage and existing or impending hemodynamic instability requires immediate surgical intervention (eg, wedge resection, gravid hysterectomy) and/or a minimally invasive procedure (ie, UAE). (See 'Surgical therapy' below and 'Adjunctive therapy' below.)

Hemodynamically stable patients — In hemodynamically stable patients, management options include termination of pregnancy (medical or surgical) or continuation of the pregnancy. The optimal management is unclear as there are an insufficient number of reported cases on which to base a specific treatment recommendation. Shared decision-making is essential and guided by factors such as CSP type (see 'Classification' above), gestational age, desire for future fertility, and experience of the physician treating the patient.

Surgical compared with medical therapy may be associated with higher success rates (ie, resolution of the pregnancy). In a systematic review including 3127 patients with CSP (mean gestational age: eight weeks), patients undergoing surgical (ie, hysterectomy, surgical resection, or dilation and curettage [D&C]) compared with medical (ie, intragestational methotrexate [MTX]) therapy had higher rates of pregnancy resolution (83 versus 60 percent); of the surgical procedures, D&C had the lowest success rates (76 percent) [52]. While rates of hemorrhage were also higher for those managed surgically compared with medically (18 versus 7 percent), use of adjuvant therapies (ie, systemic MTX, UAE) reduced rates of hemorrhage, in most cases.

Clinicians and patients should consider the following factors when choosing a treatment approach.

Patients with an embryonic/fetal demise — CSPs with embryonic/fetal demise may be managed expectantly or with medical or surgical therapy [25,53]. In our practice, we prefer expectant management, especially for those patients in the first trimester, as spontaneous resolution of the pregnancy may occur.

In the retrospective study described above including patients with CSP diagnosed prior to 14 weeks gestation (see 'Natural history' above), 10 patients presented with embryonic demise; seven patients underwent an intervention (ie, ultrasound-guided D&C, intragestational injection of potassium chloride (KCl), systemic MTX, misoprostol followed by D&C) and three were expectantly managed [27]. One patient in the intervention group required hysterectomy for bleeding; no patients in the expectant management required hysterectomy.

For patients choosing expectant management, weekly follow-up with serum human chorionic gonadotropin (hCG) and transvaginal ultrasound (TVUS) are performed until the hCG is undetectable and the pregnancy recedes completely on imaging, which can take several months to occur.

For patients with incomplete resolution (eg, persistently elevated hCG levels, incomplete resolution on TVUS), or for patients who desire to hasten the resolution of the demised CSP, treatment with systemic MTX or uterine aspiration may be performed (see 'Surgical therapy' below and 'Medical therapy' below). The diagnosis of embryonic/fetal demise by ultrasonography is discussed in detail separately. (See "Pregnancy loss (miscarriage): Ultrasound diagnosis".)

Patients with a live pregnancy — Patients with a live CSP may also be managed expectantly or with medical or surgical therapy (algorithm 1). While expectant management of a CSP with a live gestation is associated with severe maternal morbidity (eg, hemorrhage, early uterine rupture, hysterectomy, PAS, maternal death) [25-27,54,55], some patients will deliver a live born neonate at, or near, term (see 'Natural history' above); thus, some patients may choose to continue the pregnancy [53]. However, expert groups, including the Society of Maternal-Fetal Medicine, generally advise against this approach [56,57].

Expectant management — We counsel patients undergoing expectant management that outcomes appear to be more favorable for patients in whom the CSP is "on-the-scar" (type 1) rather than "in-the-niche" (type 2) and for those with a myometrial thickness ≥3 mm rather than <3 mm. However, data regarding myometrial thickness vary [58,59], and there is no universally accepted threshold. In one retrospective study including 17 patients with CSP diagnosed prior to nine weeks gestation, patients with "on-the-scar" (type 1) compared with "in-the-niche" (type 2) CSP were delivered at more advanced gestational ages (38 weeks [median; range 37 to 39 weeks] versus 34 weeks [median; range 20 to 36 weeks]) and had lower rates of cesarean-hysterectomy (1 of 6 patients [17 percent] versus 11 of 11 patients [100 percent]) [1]. Myometrial thickness was less in patients requiring cesarean-hysterectomy compared with patients not requiring hysterectomy (1 mm [range 0 to 2 mm] versus 5 mm [range 4 to 9 mm]). While there was a trend towards lower blood loss in patients with "on-the-scar" compared with "in-the-niche" CSP (median: 700 mL versus 1200 mL), this was not statistically significant.

When expectant management is performed, patients are managed in the same manner as patients with PAS and a repeat cesarean birth is planned between 34 0/7 and 35 6/7 weeks of gestation [57]. This is discussed in detail separately. (See "Placenta accreta spectrum: Management".)

Pregnancy termination — For patients who choose to terminate their pregnancy (eg, risks outweigh benefits, pregnancy is undesired), treatment is time-sensitive. As with patients without CSP, the morbidity associated with pregnancy termination increases with increasing gestational age. Management (ie, surgical or medical) is effective and depends on the following factors:

●Is this a second-trimester CSP? – For patients in their second trimester, we suggest surgical rather than medical management. In our experience, surgical treatment of second-trimester CSP has a higher success rate than medical treatment. (See 'Surgical therapy' below and 'Adjunctive therapy' below.)

●Is future pregnancy desired? If future pregnancy is not desired, and for patients who prefer definitive surgical management, gravid hysterectomy may be performed. (See 'Surgical therapy' below.)

Specific therapies

Surgical therapy

●Operative resection –Operative resection (eg, wedge resection) of the pregnancy can be performed via laparoscopy [60], hysteroscopy, or (less commonly) laparotomy; laparoscopic-assisted operative hysteroscopic management has also been described [61,62]. Operative resection may also be performed as subsequent management after medical therapy (eg, intragestational injection of MTX or KCl (see 'Medical therapy' below)). These procedures should be performed only by highly experienced surgeons.

An advantage of resection over other therapies is that the scar can be excised and the uterus reapproximated.

Operative resection is also likely to be curative; however, monitoring serum hCG levels weekly until undetectable is reasonable to confirm resolution.

The following is a description of a wedge resection performed via laparotomy in a hemodynamically stable patient (movie 1) [63]:

•A Pfannenstiel incision is made, and the abdomen entered.

•The uterus is grasped with a uterine elevating forceps and delivered through the skin incision.

•The peritoneum is incised along the medial aspect of the round ligament; the vesicouterine fold is identified, and both index fingers are used to dissect the posterior bladder from the cervix until the cesarean scar is adequately exposed.

•0-delayed absorbable (Vicryl) suture is placed at the 12, 3 and 9 o'clock positions (and includes the anterior wall of the upper vagina) to occlude the colpo-uterine vessels and reduce blood flow to the CSP.

•A scalpel with a number 5 blade is used to cut the uterine tissue around the CSP; approximately 5 mm of healthy myometrial borders are included in the incision.

•If bleeding occurs from the sectioned myometrium, 3-0 absorbable suture (rather than electrocautery) is used to obtain hemostasis.

•The inside of the uterine cavity is wiped with a gauge sponge to remove any remaining pregnancy tissue.

•The uterus is closed with 2-0 delayed absorbable suture in a running stitch.

•Plasma fibrinogen level is checked; if the level is <200 mg/dL, fibrinogen is replaced. (See "Disorders of fibrinogen", section on 'Fibrinogen concentrate: Dosing and monitoring'.)

●Ultrasound-guided suction aspiration – Suction aspiration (ie, suction D&C) is typically performed for patients in the early first trimester (five to seven weeks of gestation) with use of a transcervical balloon catheter if heavy bleeding occurs [64,65]. (See 'Medical therapy' below.)

In one prospective study including 18 cases of CSP (gestational age range: 4 to 23 weeks of gestation), suction curettage was successful in treating five of eight cases (63 percent); three patients (37 percent) experienced significant bleeding (500 to 1000 mL) requiring the insertion of an intrauterine balloon catheter to tamponade the uterine cavity to achieve hemostasis [64].

●Gravid hysterectomy – Gravid hysterectomy should only be performed in patients in whom future childbearing is not desired or in those with life-threatening hemorrhage, and may be the preferred approach for patients with second-trimester CSPs. Gravid hysterectomy may be performed with or without prior UAE. (See 'Hemodynamically unstable patients' above and 'Adjunctive therapy' below and "Overview of second-trimester pregnancy termination", section on 'Placental abnormalities'.)

Medical therapy — Medical therapy involves the use of injectable medical agents or local pressure with simple devices such as balloon catheters [57]. These are often office-based procedures and patients may not require hospitalization.

In our practice, we use systemic MTX as an adjunct to all of the following medical therapies; however, there are no high-quality data to support this practice and adverse events can occur. (See 'Adjunctive therapy' below.)

●Intragestational injection of MTX – Ultrasound-guided (transabdominal or transvaginal) intragestational injection of MTX is an effective treatment for CSP. However, there is insufficient evidence to suggest that this approach is more effective or, conversely, associated with more complications compared with other treatments [66,67].

In the prospective study including 18 patients with CSP described above (see 'Surgical therapy' above), local injection of 25 mg MTX into the exocoelomic cavity was successful in five of seven cases (71 percent); two of the seven cases (20 percent percent) required emergency surgery to control hemorrhage over 1000 mL [64]. Higher rates of successful resolution have been reported [66].

A slow resolution of the pregnancy should be expected. In a study of 101 patients with CSP treated with ultrasound-guided MTX injection, the mean time to hCG resolution was 40 days (range 21 to 140 days) [67].

●Intragestational injection of KCl – Ultrasound-guided (transabdominal or transvaginal) KCl injection (5 mEq into the gestational sac) into a CSP with embryonic/fetal cardiac activity has also been described [31,68]. This regimen may also be used as a secondary procedure if other treatments (eg, UAE, balloon catheters) do not result in cessation of fetal heart activity. (See "Induced fetal demise", section on 'Intracardiac potassium chloride injection'.)

●Transcervical insertion of balloon catheters – For patients with a CSP ≥6 0/7 and ≤8 6/7 weeks of gestation, balloon catheters are an efficient, safe, well-tolerated, and minimally invasive treatment option [69-71]. In a retrospective study including 18 patients with type 1 CSP, management with a double cervical ripening balloon catheter resulted in successful treatment in all cases; there were no cases of maternal hemorrhage [69]. The majority of patients also received systemic methotrexate.

At later gestational ages (ie, >8 6/7 weeks), cervical ripening balloon catheter use is limited as the pregnancy may be too large to be sufficiently compressed by the balloon.

These catheters can effectively stop fetal cardiac activity and prevent possible bleeding complications (by applying direct pressure to the pregnancy), can be used with or without other treatments (eg, uterine aspiration [64,65], intragestational injection of MTX [69] or KCl), and are familiar to obstetricians-gynecologists who use the same cervical ripening catheters for labor induction. In the small retrospective study described above, nine different obstetrician gynecologists performed the procedure without complication and all reported confidence in using this technique [69].

The various balloon catheters adapted for use in the treatment of CSP include the single Foley balloon catheter [72,73] or the double cervical ripening balloon catheter (ie, double-balloon Cook catheter (picture 1)) [65,74-77]. These are described in detail separately. (See "Induction of labor: Techniques for preinduction cervical ripening", section on 'Balloon catheter'.)

•The procedure is performed as follows:

-Antibiotic prophylaxis: To decrease the risk of infection, antibiotics are routinely administered, although there is no high-quality evidence to support this practice. In our practice we prescribe azithromycin 500 mg at the time of the procedure, followed by 250 mg daily for four days. However, the optimal dosing regimen has not been established and various other protocol may be used.

-Pain control: We administer ibuprofen 800 mg 30 to 60 minutes prior to the procedure. A paracervical block, using 2 to 3 ml 1% lidocaine, is typically performed for patients who have never had a previous vaginal birth or an intrauterine surgical procedure. Intravenous analgesia may also be used for patients requiring additional pain control

-Hegar dilators may be required to achieve an easy passage of the double balloon catheter, which has a slightly larger diameter than a Foley catheter.

-Insertion of the catheter and inflation of the intrauterine (anchor) balloon with saline is performed under transabdominal ultrasound guidance; a small amount of urine in the bladder helps to visualize the advancing catheter. The lower (pressure) balloon is inflated opposite the gestational sac under transvaginal ultrasound guidance (image 7).

The amount of fluid used to fill the balloons varies, and saline is progressively added until the gestational sac collapses.

-At the end of the procedure, the catheter is fastened to the patient's thigh and left in place for 24 to 48 hours. Patients are typically discharged home with instructions on when to return (eg, heavy bleeding, fever, severe pain); inpatient observation may be appropriate for some patients.

Before deflating the balloon(s), embryonic/fetal cardiac activity is rechecked to confirm cessation. If cardiac activity is still present, the balloon may be inflated to a higher pressure for an additional 24 to 48 hours; adjuvant therapies (eg, intragestational MTX or KCl, systemic MTX) may be administered, but are rarely required.

If cardiac activity has ceased, the balloon catheter is removed; our technique for removal is shown in the following algorithm (algorithm 2).

After catheter removal, the patient is closely monitored for any heavy bleeding; the patient should be observed for approximately one hour and can be discharged home with instructions to return for an ultrasound examination within two or three days. Serial ultrasound examinations and hCG levels are typically followed until complete resolution of the CSP.

Adjunctive therapy

●UAE – UAE can be used alone or in combination with other therapies (eg, suction aspiration, MTX, operative hysteroscopy) [5,6,8,11,12,52,57,61,64,78-80]. Patients in whom future fertility is desired should be counseled regarding the risks of pregnancy after UAE, which may be dependent on the embolizing agent(s) used [80,81]. (See "Uterine fibroids (leiomyomas): Treatment with uterine artery embolization", section on 'Reproductive outcomes'.)

UAE may be superior to systemic MTX as an adjunctive treatment to suction curettage [79]. In addition, shorter treatment intervals between UAE and suction aspiration appear to have more favorable outcomes. In a retrospective study including 314 patients with CSP who received UAE followed by suction D&C at time intervals of ≤24, 24 to 48 hours, 48 to 72 hours, and >72 hours, rates of intraoperative hemorrhage ≥200 mL increased (5, 7.8, 10, and 19.4 respectively) as did rates of other complications (6.7, 11.7,15.7, and 25.8 percent, respectively; adjusted odds ratio [OR] at >72 hours: 3.37, 95% CI 1.4-8.09) [82]. Intrauterine balloon tamponade or conversion to emergency laparotomy or laparoscopy occurred in 31 of the 314 patients (10 percent).

●Systemic MTX – We use systemic MTX as an adjunct to all of the above medical therapies; however, there are no high-quality data to support this practice and adverse events (eg, pancytopenia, nephrotoxicity) can occur [83].

In the systematic review discussed above (see 'Hemodynamically stable patients' above), systemic MTX followed by D&C or surgical (ie, hysteroscopic, laparoscopic) excision was associated with resolution of the pregnancy in 82 and 87 percent of patients, respectively [52]. MTX was typically administered seven days prior to the surgical procedure and resulted in the reduction of blood flow around the scar on Doppler examination and a decline in serum hCG concentrations. Adverse events occurrent in 5 percent of patients and included pelvic infection/fever, elevation of liver enzymes, uterine perforation, and bone marrow suppression.

We do not use systemic MTX alone for the treatment of CSP. This is discussed in detail below. (See 'Treatments not used' below.)

Treatments not used — Expert groups recommend not using the following treatments:

●Systemic MTX alone in patients with a live gestion– As with ectopic pregnancy, treatment of a pregnancy with fetal cardiac activity with systemic MTX is associated with high rates of treatment failure [57,76,84,85]. Systemic MTX may be used as an adjunct to other treatments and for CSPs with embryonic/fetal demise.

●Misoprostol – While misoprostol may be used for medical termination of first-trimester pregnancies, we do not support its use in treating CSP; the deeply implanted placenta with CSPs appear to prevent the abortive effect of misoprostol, leading to increased rates of treatment failure and severe complications [85,86].

●Sharp curettage — Expert groups recommend against sharp curettage given the higher risk of complications (eg, perforation, hemorrhage, need for reintervention) [57]; profuse hemorrhage may occur as a result of the deeply implanted placental vessels and scant (or no) myometrium to stop the bleeding vessels severed by sharp curetting. In addition, residual tissue following the D&C may result in subsequent complications, such as enhanced, or extremely enhanced, myometrial vascularity [29,87].

While successful termination of CSP has been reported using sharp curettage [88-91], we do not use this approach in our practice.

FUTURE PREGNANCIES — Subsequent pregnancies following CSP have been reported [32,92-95]; such patients are at risk for recurrent scar implantation, abnormal placentation, and uterine rupture (resulting in maternal or fetal death). (See 'History of CSP' above.)

Role of ultrasound for screening — Pregnant patients with a previous cesarean birth and other high-risk patients (eg, history of prior CSP) may benefit from early ultrasound screening in a subsequent pregnancy. In our practice, such patients are counseled to contact their obstetrician gynecologist soon after their positive pregnancy test and undergo a transvaginal ultrasound (TVUS) as early as five to seven weeks of gestation. As this is well before most obstetric practices schedule a first prenatal visit, this requires very early communication between the patient and their physician with swift scheduling of the TVUS. In addition, the ultrasound should be performed by an expert sonologist or physician with experience in diagnosing CSP. Initiation of such measures may not only maximize management options but also may optimize treatment outcomes [95-98].

Lower severe maternal morbidity (eg, first-trimester hemorrhage, need for blood transfusion, uterine rupture, emergency hysterectomy) has been demonstrated in patients diagnosed and treated at earlier compared with later gestations. In one meta-analysis including 36 studies (724 patients) with CSP and undergoing immediate treatment, those diagnosed ≤9 weeks compared with >9 weeks had lower rates of composite adverse maternal outcomes (5.9 versus 32.4 percent, odds ratio [OR] 0.14, 95 % CI 0.1-0.4) [95]. Future studies are needed to assess the cost-effectiveness of such measures [97].

REGISTRY — An international CSP registry has been established to better understand the disease [99]. Information is available online.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Obstetric hemorrhage" and "Society guideline links: Ectopic pregnancy".)

SUMMARY AND RECOMMENDATIONS

●Terminology – Cesarean scar pregnancy (CSP) refers to a pregnancy that is implanted on ("on-the-scar" [type 1]) or in ("in-the-niche" [type 2]) a scar from a prior cesarean birth (or rarely from a prior myomectomy). If unrecognized or inadequately managed, CSP can lead to severe maternal and fetal outcomes (eg, uterine rupture, hemorrhage, death). (See 'Introduction' above and 'Classification' above.)

●Pathogenesis – CSP is a likely a precursor to, and shares a common histology with, placenta accreta spectrum (PAS). The two are likely a continuum of the same disease. (See 'Pathogenesis' above and "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences".)

●Natural history – CSPs are associated with high maternal and fetal morbidity (eg, hemorrhage, uterine rupture, PAS) and even mortality. However, some CSPs may continue to a viable gestation. Clinical outcomes appear to be more favorable for "on-the-scar" versus "in-the-niche" CSPs; however, data are limited as CSP type is often not specified in ultrasound reports or published studies. (See 'Natural history' above.)

●Diagnosis

•CSP is a diagnosis made based on first- (or early second-) trimester transvaginal ultrasound (TVUS) findings of a pregnancy implanted on or in a prior hysterotomy scar/niche and confirmed at the time of surgery with histologic findings consistent with PAS disorder. (See 'Definitive diagnosis' above.)

•Other early ultrasound findings suggestive of CSP in a patient with a prior uterine scar include a low anterior position of the gestational sac at ≤7 weeks; low-lying anterior placenta or placenta previa; placental lacunae; thin (<3 mm) or no intervening myometrium/clear space (ie, residual myometrial thickness) between the placenta and bladder or anterior uterine serosa; irregular or disrupted bladder line or anteriorly bulging sac into the bladder; and increased vascularity at the bladder/placental interphase. (See 'Findings suggestive of the diagnosis' above.)

●Management – All patients with CSP should be transferred, whenever possible, to a center of excellence for PAS, or a tertiary care hospital in which maternal-fetal medicine specialists, a variety of treatment options (eg, uterine artery embolization [UAE]), and blood bank services are available. (See 'Management' above.)

•A patient with hemorrhage and existing or impending hemodynamic instability requires immediate surgical intervention (eg, wedge resection, gravid hysterectomy) and/or a minimally invasive procedure (ie, UAE). (See 'Hemodynamically unstable patients' above.)

•For hemodynamically stable patients, the optimal management (ie, surgical or medical termination, expectant management) is unclear. While expert groups generally advise pregnancy termination, some pregnancies will continue to, or close to, term and thus some patients will choose to continue the pregnancy. Shared decision-making is essential and guided by factors such as CSP type, gestational age, desire for future fertility, and experience of the physician treating the patient. (See 'Hemodynamically stable patients' above.)

-Patients with an embryonic/fetal demise can be managed expectantly or with medical or surgical therapy. (See 'Patients with an embryonic/fetal demise' above.)

-For patients with a desired, live pregnancy, outcomes appear to be more favorable for patients in whom the CSP is "on-the-scar" (type 1) and with a myometrial thickness ≥3 mm (algorithm 1). Thus, we offer expectant management to such patients and manage similarly to patients with PAS. Outcomes appear to be less favorable for patients in whom the CSP is "in-the-niche" (type 2) or with a myometrial thickness <3 mm. (See 'Expectant management' above and "Placenta accreta spectrum: Management".)

-For patients with a live pregnancy choosing termination at more advanced (ie, second trimester) gestational ages, we suggest surgical rather than medical therapies (algorithm 1) (Grade 2C). For patients at earlier gestations, medical or surgical therapies can be used. (See 'Pregnancy termination' above.)

•UAE and systemic methotrexate (MTX) are frequently used as adjunctive therapy to both medical and surgical treatments; patients should be counseled regarding the benefits and risks of such treatment. (See 'Adjunctive therapy' above.)

•We do not use of systemic MTX alone, misoprostol, or sharp uterine curettage for the management of CSP. (See 'Treatments not used' above.)

●Screening – Pregnant patients with a prior cesarean birth and other risk factors for CSP (eg, prior CSP) may benefit from early (ie, five to seven weeks of gestation) ultrasound screening to determine if abnormal implantation has occurred. (See 'Role of ultrasound for screening' above.)