Cesarean scar pregnancy occurs when a gestational sac implants at the site of a previous hysterotomy scar. With an incidence of 1 in 1,800 to 1 in 2,200 pregnancies, cesarean scar pregnancies represent 6% of all ectopic pregnancies in women with prior cesarean delivery.1,2 Presentation of cesarean scar pregnancy may highly vary. Diagnosis is made by ultrasonographic visualization of a mass embedded in the hysterotomy scar, an empty uterine cavity, and thinning of a visible defect in the myometrium between the bladder and the sac on transvaginal ultrasonogram.3
Given the risk of life-threatening complications, pregnancy termination is generally recommended. Numerous treatment modalities have been described, including expectant management, administration of methotrexate, “compressive” therapy with an intrauterine balloon, surgical intervention with resection of the cesarean scar pregnancy through vaginal or abdominal approaches, and many others in different combinations. There is no consensus on the optimal treatment modality.3
There are limited data on subsequent pregnancy outcomes after cesarean scar pregnancy. Uncomplicated viable term intrauterine pregnancies have been reported after cesarean scar pregnancies. However, cesarean scar pregnancy can recur, and risk may be increased for uterine rupture and morbidly adherent placenta in future pregnancies. There remain few data to guide treatment decisions and properly counsel patients with a history of cesarean scar pregnancy on their risks in future pregnancies.4,5
To add to the available literature on the treatment modalities of cesarean scar pregnancy, immediate therapeutic outcomes, and subsequent fertility effects, we present a case series describing our experience in treatment and follow-up of women with a cesarean scar pregnancy at Yale-New Haven Hospital during a 5-year period.
MATERIALS AND METHODS
This is a series of all cases of cesarean scar pregnancy managed at a single tertiary care center (Yale-New Haven Hospital, New Haven, Connecticut) from May 2013 to March 2018. The patients with a diagnosis of cesarean scar pregnancy were identified retrospectively using an R4 Perinatal Reporting System database query for the diagnoses “cesarean scar pregnancy” and “cesarean scar ectopic.” Electronic medical records were subsequently reviewed to collect the data on patients' demographics; detailed medical, surgical, and social history; symptoms; imaging and laboratory parameters at the time of cesarean scar pregnancy diagnosis and during treatment; treatment modalities; and subsequent immediate and remote outcomes. New cases of cesarean scar pregnancy were diagnosed during the study period and included in the study for prospective follow-up. The patients in the study represented the university clinic and private community groups population serving all of southern Connecticut. In all cases, the diagnoses were either initially made or confirmed at the maternal-fetal medicine ultrasound unit. The study was approved by the Yale institutional review board.
All diagnoses were made based on the patient's history of prior cesarean delivery, positive pregnancy test, presence of a gestational sac in the area of the scar, and otherwise empty uterine cavity on transvaginal ultrasonogram (Fig. 1).
Only descriptive statistics were used for this case series. Normally distributed continuous data were described using mean±SD. For data that were not normally distributed, median and range were used.
Over the course of the study, 30 cases of cesarean scar pregnancy were diagnosed in 26 patients (patient 3 had one recurrence, patient 7 had three recurrences). Detailed characteristics of each case and composite data are depicted in Tables 1 and 2, respectively. Notably, there was an increase in the number of diagnosed cases of cesarean scar pregnancy from 2013 to 2016 (Table 3), which may be explained by the higher detection rate in our facility as well as a possibly the true increase in the incidence of the cesarean scar pregnancy.
There were 12 Hispanic patients (46%), 11 non-Hispanic white patients (42%), and three non-Hispanic black patients (12%). This seems to be higher than Hispanic representation in New Haven County and the city of New Haven based on 2016 Census Bureau data (17.6% and 27% Hispanics in the county and the city, respectively). Patients in the Hispanic and non-Hispanic white groups had similar rates of obesity, smoking, total number of pregnancies, and number of previous cesarean deliveries, although the number of cases was too small for meaningful statistical comparison. Additional clinical features of the cohort are summarized in Table 2.
At the time of the diagnosis, the majority of patients were asymptomatic. Four patients had light vaginal bleeding, three patients had varying degrees of lower abdominal pain, and three patients reported both. None of the patients had abnormal vital signs or were hemodynamically unstable at the time of the diagnosis. Two patients (patients 18 and 24) had an initial presentation with severe pain concerning for uterine rupture; however, the pain subsided spontaneously and the diagnosis of uterine rupture was not made.
Pregnancy characteristics including mean gestational age, median serum human chorionic gonadotropin (hCG) level, mean gestational sac diameter as well as presence of fetal heartbeat are outlined in Table 1. Of note, patient 13 was initially concerning for molar pregnancy with an initial hCG level of greater than 100,000 milliunits/mL. Three patients (patients 11, 23, and 29) initially were diagnosed with viable, normally located intrauterine pregnancies and proceeded with medical termination of pregnancy, which failed. Concern for retained products of conception in these patients triggered further ultrasound evaluation, which was then diagnostic of cesarean scar pregnancy. In an additional six patients, the initial ultrasonogram was nondiagnostic of cesarean scar pregnancy (suspected cervical ectopic pregnancy, concern for nonviable intrauterine pregnancy, pregnancy of unknown location) but triggered further ultrasonographic evaluation at the maternal-fetal medicine ultrasound unit.
Most patients were treated by a multidisciplinary team that included at least a maternal-fetal medicine subspecialist and a gynecologist. All patients were counselled regarding the risks of continuation of the cesarean scar pregnancy and available treatment options including expectant management; conservative treatment with local and systemic methotrexate; hysteroscopy with curettage; and cesarean scar pregnancy resection through laparotomy, hysterectomy, uterine artery embolization, or an intrauterine balloon. All patients eventually opted for termination of pregnancy, except for those who already appeared to be undergoing spontaneous abortion and who opted for expectant management (three patients). Table 4 outlines treatment and subsequent outcomes for each cesarean scar pregnancy case. Table 5 summarizes all treatment modalities used in this study.
The following therapeutic methods were applied as an initial treatment modality:
- Expectant management was planned in three patients. In one of the patients (patient 2) spontaneous resolution of the pregnancy did not occur and the patient underwent hysteroscopy and curettage for persistent heterogenous material in the area of the scar. Notably, this patient's starting hCG was 37,300 milliunits/mL, higher than the other two expectedly managed patients.
- Systemic methotrexate only was used in four patients and was given at a dose of 25 mg intramuscularly. Fetal heart beat was present in one of these four cases at the time of treatment. The hCG level among these patients was 2,781–15,700 milliunits/mL. In one patient (patient 15), the cesarean scar pregnancy failed to resolve, and a repeat dose of systemic methotrexate was required, followed by an unsuccessful attempt of hysteroscopic removal of the pregnancy. Finally, this patient underwent definitive treatment with a total abdominal hysterectomy. Notably, this was a second episode of cesarean scar pregnancy for this patient. Records of the first cesarean scar pregnancy in this patient were not available for review.
- Systemic and local methotrexate was used in 12 patients. When used locally, 25 mg methotrexate were injected under transvaginal ultrasound guidance into both the placenta and the gestational sac. One patient (patient 20) was complicated by retained products of conception and required a second dose of systemic methotrexate simultaneously with hysteroscopic removal of the pregnancy. Notably, this patient was initially erroneously diagnosed with a viable normally located pregnancy and underwent attempted medical termination of pregnancy elsewhere.
- Potassium chloride injection was used before systemic and local methotrexate injection in three patients, one of whom required subsequent hysteroscopic resection of the cesarean scar pregnancy (patient 4). To achieve asystole, 3–6 mL of concentrated solution of potassium chloride (2 mEq/mL) was injected into the fetus.
- Potassium chloride injection followed by laparotomy with a wedge resection of the pregnancy was used in one patient (patient 3). Products of conception at the ballooning cesarean scar were removed through Pfannenstiel laparotomy after careful dissection of the thin myometrial layer overlying the sac. The resultant uterine defect was closed in three layers. A Penrose drain was used as a uterine tourniquet for hemostasis throughout the procedure. Surgery was uncomplicated, and the patient was discharged home on postoperative day 2. Notably this patient had a subsequent cesarean scar pregnancy recurrence.
- Hysteroscopy with curettage was performed as an adjunct treatment modality for retained products of conception in four patients. The procedure was performed by an experienced gynecologic surgeon under abdominal ultrasound guidance. The trophoblastic tissue was first identified with direct visualization at the area of prior cesarean scar with a diagnostic hysteroscope and evacuated with either sharp or suction curettage with or without ultrasound guidance. Chorionic villi were identified by pathologic examination of the retrieved tissue in all patients.
- Systemic methotrexate with uterine artery embolization followed by scheduled hysteroscopy and curettage was used in a patient whose initial presentation was concerning for molar pregnancy with heavy vaginal bleeding (patient 22). Bilateral uterine artery embolization with Gelfoam was performed under fluoroscopy guidance by the interventional radiology team. The final pathologic examination was not consistent with molar pregnancy.
- Uterine artery embolization in conjunction with systemic and local methotrexate was used in one patient.
- Uterine balloon was used in four patients. Placement was performed in the outpatient ultrasound unit or in the operating room under sedation as a result of the patient's intolerance of pelvic examinations. Under transabdominal ultrasound guidance, a double-balloon cervical ripening catheter with stylet was advanced into the endometrial cavity. The distal and proximal balloons were sequentially inflated with 10 mL and 8–20 mL of sterile saline, respectively. Correct placement and compression of the gestational sac were confirmed with transabdominal ultrasonogram (Fig. 2). Patients were monitored for 1 hour. A repeat abdominal ultrasonogram was done. Forty-eight to 72 hours later the balloons were sequentially deflated and removed, and patients were monitored for a total 60 minutes to ensure absence of vaginal bleeding. The technique was adopted from Timor-Tritsch et al.6 Repeat ultrasound evaluation was performed to assess the appearance of the gestational sac. Patients were then followed with serial ultrasonograms and hCG levels. None of these patients required any additional treatments.
The median number of days from the time of the diagnosis to treatment was 1 (range 0–14 days). Patient 19 declined intervention for 2 weeks because it was a highly desired pregnancy.
After the treatment was executed, patients were followed closely with frequent outpatient visits, repeat ultrasound evaluations, and hCG levels. All patients were asked to come back for β-hCG level in 7 days and a follow-up scan within 1 week after the main treatment was initiated. In 16 patients β-hCG levels were followed to levels less than 10 milliunits/mL. In six patients, the last measured hCG level was 11–87 milliunit/mL; in the remaining cases, patients did not follow-up as instructed. Most of the patients continued serial ultrasound examinations to confirm resolution or near resolution of the ectopic pregnancy. Cesarean scar pregnancy was considered resolved in women whose hCG level was undetectable and no mass or gestational sac was appreciated at the site of the cesarean delivery scar on transvaginal ultrasonography. One patient (patient 7) with three recurrences of cesarean scar pregnancy did not follow-up after each of the treatments after declining surgical intervention each time.
None of the patients had any immediate adverse effects of the therapy they opted for. Four patients were subsequently seen in the emergency department with vaginal bleeding; however, their blood counts remained stable and they were discharged home without intervention. Two other patients were admitted for overnight observation in the setting of lower abdominal pain and discharged home the next day without additional interventions. None of the patients required blood transfusion.
In four patients (patients 2, 4, 15, and 20), hysteroscopic removal of retained products of conception was required and was performed 82, 48, 60, and 57 days after the diagnosis of cesarean scar pregnancy, respectively. Initial treatment plans for those patients were expectant management, potassium chloride with systemic and local methotrexate administration, systemic methotrexate only, and systemic and local methotrexate, respectively. In all four patients, serum hCG levels were trending down and the ultrasonographic size of the mass at the level of cesarean scar plateaued. In patient 15, as mentioned previously, hysteroscopic evacuation of the uterus (60 days after the diagnosis) failed to achieve complete removal of products and an interval abdominal hysterectomy was undertaken (69 days after the diagnosis). There was no definitive correlation between the highest level of hCG and need for unplanned surgical intervention.
Of 25 patients who preserved fertility after the initial cesarean scar pregnancy episode, there were 10 conceptions in eight patients (Table 6; Fig. 3). Three conceptions resulted in full-term deliveries through repeat cesarean delivery, none of which were complicated by abnormal placentation or uterine rupture. Notably, patient 8 had a full-term pregnancy, which was delivered by scheduled cesarean delivery. The surgery was complicated by intraoperative cardiac arrest. She also had massive hemorrhage that required hysterectomy. Ultimately, the patient recovered with some remaining neurologic deficit. The delivery occurred in an outside hospital and documentation was not suggestive of abnormal placentation or uterine rupture. Patient 3 had one recurrence of cesarean scar pregnancy, and patient 7 had three recurrences of cesarean scar pregnancy. There was one miscarriage of a spontaneous twin pregnancy. One normally located intrauterine pregnancy is currently ongoing and is in the third trimester. To summarize, of 10 conceptions, four were recurrences of cesarean scar pregnancy (40%) and six were normally located viable intrauterine pregnancies (60%), of which one was a spontaneous miscarriage (10% of all conceptions).
As mentioned previously, patient 15’s cesarean scar pregnancy was a recurrence; however, the initial episode of her cesarean scar pregnancy was treated at another institution and thus that episode was not included in the analysis. With this patient included, recurrence occurred in three patients out of 26 (11.5%), and six patients had a viable subsequent pregnancy (23%). The median number of months between the diagnosis of cesarean scar pregnancy and the onset of subsequent pregnancy was 8 (range 6–35 months).
Since the first description of cesarean scar pregnancy in 1978, its frequency has increased dramatically, which correlates with the cesarean delivery rate uptrend.3,7 We describe a series of 30 cases of cesarean scar pregnancies in 26 patients diagnosed during a 5-year period.
In our study the following patient characteristics were notable: a large proportion of women were of Hispanic origin and most women had body mass index above normal. Ethnic predisposition and obesity, along with specific surgical techniques during cesarean delivery (suture material, type of hysterotomy closure and others), need to be further studied as potential risk factors for cesarean scar pregnancy.
In three patients, an erroneous diagnosis of a normally located pregnancy was made before referral to our unit. These patients attempted elective medical termination of pregnancy, which was unsuccessful. As a result of the risk of severe morbidity in undiagnosed cases of cesarean scar pregnancy, a high level of suspicion for cesarean scar pregnancy should be present in all patients with a history of cesarean delivery.
A systematic review by Timor-Tritsch and Monteagudo identified 31 different treatment modalities for cesarean scar pregnancy described in the literature.7 A recent systematic review recommended five treatment modalities (transvaginal resection; laparoscopy; uterine artery embolization combined with dilatation, curettage, and hysteroscopy; uterine artery embolization in combination with dilatation and curettage; and hysteroscopy) as the most effective and safe.8 Another systematic review failed to identify the leading method for treatment of cesarean scar pregnancy supporting the need for further studies in this field.9 The vast majority of studies included in these systematic reviews originated from Asian countries with only few case series performed in Europe and the United States.10 Thus, there is still the need for further comprehensive reviews of the cases and treatment strategies of cesarean scar pregnancies in the U.S. population.
A broad spectrum of options represents a real challenge for the health care provider. The treatment of choice should account for clinical presentation, gestational age, laboratory trends, imaging characteristics, level of suspicion for uterine rupture, and patients' desires including plans for future fertility. Availability of the appropriate equipment, qualified staff, and patient's access to care and compliance also play a key role in the treatment choice.
In our series, 8.3% of patients treated with systemic and local methotrexate, 25% of patients received systemic methotrexate only, and 33% of patients treated expectantly required additional surgical procedures for retained products of conception. These findings are similar to the systematic review in which the use of systemic methotrexate alone or in combination with local methotrexate was associated with the need for additional treatment in 15% (systemic and local) and 25% (systemic only).8 In light of these findings, every patient who opts for conservative treatment with methotrexate or chooses expectant management should be counseled on the risks of severe complications or need for additional procedures in the future. Uterine artery embolization was used only twice in our series and was instrumental in treatment of active bleeding in one patient. This method in conjunction with other treatment modalities is worth closer consideration for treatment or prophylaxis of bleeding in the setting of cesarean scar pregnancy in specific clinical scenarios. The use of an intrauterine double balloon has emerged as an effective conservative method with no major complications associated with its use.6,11
Although serial serum hCG measurement is a widely accepted method of ectopic pregnancy follow-up, it does not always reflect the resolution of cesarean scar pregnancy. In our group, four patients had retained products of conception in the setting of downtrending hCG. This may be explained by the retention of products of conception within the uterus, which still confers a risk for complications such as infection and bleeding.12 At our institution we continue to perform serial ultrasonograms to confirm resolution of pregnancy.
The exact recurrence risk of cesarean scar pregnancy is unknown with some studies quoting 4–15.6%.13,14 In our study 30% of women had a subsequent pregnancy. Forty percent of conceptions were recurrent cesarean scar pregnancies, and the rest were normally located intrauterine pregnancies. The high rate of recurrence warrants appropriate counseling of the patients on future fertility risks. Health care providers should have a high level of suspicion for cesarean scar pregnancy recurrence in patients with a history of cesarean scar pregnancies. The small number of subsequent pregnancies in our study did not allow for a meaningful association between the treatment modality and subsequent pregnancy outcome.
Given the complexity of the cesarean scar pregnancy diagnosis, in 2015, Yale-New Haven Hospital developed institutional guidelines outlining the diagnostic criteria, required laboratory testing, counseling points, and treatment options for patients with cesarean scar pregnancy. Furthermore, it included a list of health care providers from maternal-fetal medicine and benign gynecology divisions with experience in diagnosis and treatment of cesarean scar pregnancy.
Given the increasing frequency of cesarean scar pregnancy, health care providers should be prepared to diagnose this disorder with ultrasonography and refer to facilities with expertise in suspicious cases and for treatment. Treatment of cesarean scar pregnancy may be complicated, multistep, and require a multidisciplinary approach. Regardless of the treatment of choice, close follow-up with clinical assessment, serial measurement of quantitative hCG levels, and serial ultrasonographic examinations are required given the risk of remote complications and a potential need for additional treatment modalities. With multiple treatment options available, the decision on the modality of choice in every case should be tailored to a specific clinical scenario. Diagnosis of cesarean scar pregnancy puts the patient at risk of complications during this and subsequent pregnancies and requires thorough counseling and close follow-up. Despite the potential for catastrophic consequences after a cesarean scar pregnancy, the case series presented here demonstrates that safe outcomes are, indeed, achievable in a consistent manner in the appropriate setting. Specifically, early and accurate diagnosis by skilled ultrasonographers, a team of experienced clinicians who are invested in optimizing the care of patients with cesarean scar pregnancies, and care at a facility with access to multidisciplinary expertise are all recommended to maximize the likelihood of safe outcomes for these high-risk patients.
The small number of patients and lack of uniformity in treatment modalities are the main limitations of our study. For this reason, we are unable to identify the single optimal method for cesarean scar pregnancy treatment. Additionally, some cesarean scar pregnancies and subsequent pregnancies could have been managed by general obstetrics and gynecology care providers and would not be included in the study. Also, some of the cases of cesarean scar pregnancy may have been missed during early pregnancy and subsequently treated by maternal-fetal medicine providers as pregnancies complicated by morbidly adherent placentas at later gestational ages.7,15–18 Additionally, some cesarean scar pregnancies erroneously diagnosed as normally located intrauterine pregnancies could have been missed if they spontaneously regressed and resulted in a miscarriage. Several patients in our study did not follow-up as planned and were lost to follow-up. Further large case–control studies are needed to illicit the role of ethnicity, obesity, and smoking in predisposition to cesarean scar pregnancy.
1. Rotas MA, Haberman S, Levgur M. Cesarean scar ectopic pregnancies: etiology, diagnosis, and management. Obstet Gynecol 2006;107:1373–81.
2. Seow KM, Huang LW, Lin YH, Lin MY, Tsai YL, Hwang JL. Cesarean scar pregnancy: issues in management. Ultrasound Obstet Gynecol 2004;23:247–53.
3. Ash A, Smith A, Maxwell D. Caesarean scar pregnancy. BJOG 2007;114:253–63.
4. Ben Nagi J, Helmy S, Ofili-Yebovi D, Yazbek J, Sawyer E, Jurkovic D. Reproductive outcomes of women with a previous history of Caesarean scar ectopic pregnancies. Hum Reprod 2007;22:2012–5.
5. Seow KM, Hwang JL, Tsai YL, Huang LW, Lin YH, Hsieh BC. Subsequent pregnancy outcome after conservative treatment of a previous cesarean scar pregnancy. Acta Obstet Gynecol Scand 2004;83:1167–72.
6. Timor-Tritsch IE, Monteagudo A, Bennett TA, Foley C, Ramos J, Kaelin Agten A. A new minimally invasive treatment for cesarean scar pregnancy and cervical pregnancy. Am J Obstet Gynecol 2016;215:351.e1–8.
7. Timor-Tritsch IE, Monteagudo A. Unforeseen consequences of the increasing rate of cesarean deliveries: early placenta accreta and cesarean scar pregnancy. A review. Am J Obstet Gynecol 2012;207:14–29.
8. Birch Petersen K, Hoffmann E, Rifbjerg Larsen C, Svarre Nielsen H. Cesarean scar pregnancy: a systematic review of treatment studies. Fertil Steril 2016;105:958–67.
9. Gonzalez N, Tulandi T. Cesarean scar pregnancy: a systematic review. J Minim Invasive Gynecol 2017;24:731–8.
10. Timor-Tritsch IE, Khatib N, Monteagudo A, Ramos J, Berg R, Kovács S. Cesarean scar pregnancies: experience of 60 cases. J Ultrasound Med 2015;34:601–10.
11. Timor-Tritsch IE, Monteagudo A, Kaelin Agten A. Recap-Minimally invasive treatment for cesarean scar pregnancy using a double-balloon catheter: additional suggestions to the technique. Am J Obstet Gynecol 2017;217:496–7.
12. Timor-Tritsch IE, Monteagudo A, Santos R, Tsymbal T, Pineda G, Arslan AA. The diagnosis, treatment, and follow-up of cesarean scar pregnancy. Am J Obstet Gynecol 2012;207:44.e1–13.
13. Maymon R, Svirsky R, Smorgick N, Mendlovic S, Halperin R, Gilad K, et al. Fertility performance and obstetric outcomes among women with previous cesarean scar pregnancy. J Ultrasound Med 2011;30:1179–84.
14. Wang Q, Peng HL, He L, Zhao X. Reproductive outcomes after previous cesarean scar pregnancy: follow up of 189 women. Taiwan J Obstet Gynecol 2015;54:551–3.
15. Pekar-Zlotin M, Melcer Y, Levinsohn-Tavor O, Tovbin J, Vaknin Z, Maymon R. Cesarean scar pregnancy and morbidly adherent placenta: different or similar? Isr Med Assoc J 2017;19:168–71.
16. Timor-Tritsch IE, Monteagudo A, Cali G, Vintzileos A, Viscarello R, Al-Khan A, et al. Cesarean scar pregnancy is a precursor of morbidly adherent placenta. Ultrasound Obstet Gynecol 2014;44:346–53.
17. Timor-Tritsch IE, Monteagudo A, Cali G, Palacios-Jaraquemada JM, Maymon R, Arslan AA, et al. Cesarean scar pregnancy and early placenta accreta share common histology. Ultrasound Obstet Gynecol 2014;43:383–95.
18. Usta IM, Hobeika EM, Musa AA, Gabriel GE, Nassar AH. Placenta previa-accreta: risk factors and complications. Am J Obstet Gynecol 2005;193:1045–9.