Three women attended our early pregnancy unit with amenorrhea of more than 5 weeks, mild lower abdominal pain and slight vaginal bleeding. Both transabdominal and transvaginal ultrasound findings were compatible with a Cesarean scar pregnancy (CSP) according to the CSP criteria proposed by Jurkovic et al.1 and Vial et al.2, namely: (1) an empty uterine cavity; (2) an anterior location of the gestational sac at the level of the internal os covering the visible or presumed site of the previous lower uterine segment of the Cesarean scar; (3) evidence of functional trophoblastic/placental circulation on Doppler scans; and (4) the presence of trophoblast between the bladder and the anterior uterine wall as a sign of deep implantation. Termination of pregnancy was suggested, and therapeutic options (including exploratory laparotomy, laparoscopy, suction evacuation, and medical therapy) were offered. All patients gave informed consent to surgical evacuation. Before the evacuation, facilities for hemostasis and emergency laparotomy were prepared. Evacuation procedures were performed under transabdominal ultrasound guidance. After transcervical dilatation using a Hegar dilator (Lawton, Munich, Germany), a suction vacurette was carefully introduced into the uterus to remove gestational villi and blood clots.
The clinical history of the three new cases is summarized in Table 1 (Cases 1–3). In our first patient, evacuation was performed 2 days after the diagnosis was made. There was massive uterine bleeding immediately after the gestational tissue was removed. A 20-G Foley catheter was inserted at the level of the implantation site, with 60 mL saline introduced to achieve homeostasis by compression. Estimated blood loss was 700–800 mL. Bleeding stopped after the Foley compression, and the catheter was deflated the next day. One week after the evacuation, the beta-human chorionic gonadotropin (β-hCG) level was 7482 IU/L, and ultrasound images revealed a heterogeneous mass, measuring 5.6 × 4.7 cm at the site of the previous sac location. Residual trophoblast and hematoma were suspected and laparoscopy was performed. After adhesions had been dissected off and the bladder base pushed forward, a uterine defect of approximately 2 cm in length with dark reddish tissues was discovered and repaired by one layer of interrupted endoscopic sutures. Histopathological examination of the removed tissue revealed degenerated chorionic villi and trophoblasts with blood clots, which were consistent with ectopic pregnancy. Two weeks later subsequent ultrasound scans and β-hCG levels were normal.
Table 1. Characteristics of patients with Cesarean scar pregnancy (CSP) accurately diagnosed and primarily treated with evacuation therapy: summary of the literature
A case of heterotopic CSP: one sac implanted in the Cesarean scar and one in the uterine cavity. β-hCG, β-human chorionic gonadotropin; CRL, crown–rump length; C/S, Cesarean section; D&C, dilatation and curettage; FHB, fetal heart beat; GA, gestational age (weeks by last menstrual period); GS, gestational sac; modified GA, gestational age estimated on the basis of ultrasound measurements of the CRL/GS; NA, not mentioned.
The second patient was admitted because of vaginal spotting and intermittent low abdominal pain. She had a history of irregular periods and two previous low-segment Cesarean sections. Ultrasound demonstrated a uterine wedge defect over the lower uterine segment and an empty uterine cavity. No gestational sac was found. The β-hCG level was 10 320 IU/L. A diagnosis of pregnancy with unknown location was made, and she was advised to stay in contact with the hospital and report any changes in her condition. One week later, β-hCG was 11 522 IU/L and ultrasound examination revealed a 5-mm sac, with no fetal parts and compatible with a gestational age of 4–5 weeks, embedded in the wide-open wedge defect connected to the uterine cavity (Figure 1). Gestational tissue was successfully evacuated with minimal blood loss. Two weeks after surgery, ultrasound examination revealed a persistent uterine defect without gestational tissue.
The third patient was referred because of a missed menstrual period and intermittent vaginal spotting. She had three previous Cesarean deliveries. A transvaginal ultrasound scan revealed an empty uterus and a 12-mm gestational sac with no fetal parts, implanting in a wide-open wedge defect in the anterior uterine isthmic area. Instrumental removal of gestational tissue was uneventful and the blood loss was less than 100 mL. Two weeks after surgery ultrasound examination showed a uterine defect without gestational tissue.
CSP is the rarest form of ectopic pregnancy. The differential diagnosis between cervico-isthmic pregnancy, spontaneous miscarriage in progress, and CSP is considered to be difficult3. Since the first case of CSP was reported in 19784, about 40 case reports or series have been published. Apparently, increasing rates of CSP may reflect both a high Cesarean section rate and more accurate diagnoses5. Transvaginal sonography and unequivocal diagnostic criteria facilitate early detection of CSP. Some authors consider CSP to be more common than cervical pregnancy6.
Because of the high risk of uterine rupture, invasive placenta and profuse uterine bleeding, termination of CSP is recommended3, 5, 7. There is no consensus about the optimal method of terminating a CSP. Some authors consider evacuation to be contraindicated in CSP because of its propensity to induce uterine perforation and devastating hemorrhage8, 9. In contrast, others have reported patients successfully treated with surgical evacuation1, 10, 11. CSP is so rare that it is hardly feasible to perform a randomized controlled study to evaluate the efficacy of different therapies.
In a comprehensive electronic search of the MEDLINE Ovid system of English literature published from January 1966 to June 2004, we found 37 case reports and four series dealing with Cesarean scar pregnancy. In 11 reports there were 20 cases in which evacuation was performed as the primary treatment. Eleven cases appeared to have been accurately diagnosed as CSP before treatment. Including our three cases, Table 1 summarizes the clinical characteristics of the 14 correctly diagnosed cases1, 10–12. In our summary we have used the following definitions: gestational age is estimated on the basis of ultrasound measurements of crown–rump length13 or gestational sac14 if data were available; ‘sac bulging’ is the protrusion of a gestational sac that changes the outer uterine contour in the low anterior segment and, if so, part of the trophoblastic tissue is located between the bladder and the anterior uterine wall. A deep sac implantation is suspected in these cases.
The rate of satisfactory evacuation therapy was 64% (9/14) in the cases correctly diagnosed before surgery. The rate of unsuccessful treatment was 80% (4/5) in cases ≥ 7 gestational weeks vs. 11% (1/9) < 7 gestational weeks. Two patients (Cases 1 and 4) with sac bulging had profuse bleeding and retained gestational tissue, whereas three cases (Cases 2, 3 and 14) with sacs within the uterine canal were uncomplicated.
Table 2 is a summary of CSP patients initially diagnosed with other conditions who subsequently underwent evacuation therapy4, 8, 9, 15–18. The ascertainment of CSP was made by subsequent interventions. Massive uterine bleeding occurred in all cases. Seven patients needed an emergency laparotomy or laparoscopy and in two cases ultimate management was by hysterectomy. To some extent, without correct diagnosis and adequate pre-surgery preparation, primary evacuation for CSP is detrimental.
Table 2. Characteristics of patients with Cesarean scar pregnancy initially under wrong diagnoses and treated with evacuation therapy: summary of the literature
Sac location: according to ultrasound imaging or the description of findings on subsequent intervention.
Bulging sac, the protrusion of a gestational sac changing the outer contour in the low anterior uterine segment. β-hCG, β-human chorionic gonadotropin; CRL, crown–rump length; C/S, Cesarean section; D&C, dilatation and curettage; FHB, fetal heart beat; GA, gestational age (weeks by last menstrual period); GS, gestational sac; modified GA, gestational age estimated on the basis of ultrasound measurements of the CRL/GS; NA, no figure available or not mentioned; SM, spontaneous miscarriage; UP, unwanted pregnancy; UAE, uterine artery embolization.
Vial et al.2 proposed two types of implantation in CSP. The first type involves superficial implantation on a scar with subsequent progression into the cervico-isthmic space and uterine cavity. The second involves deep implantation directly into the myometrium or inner surface of the visceral serosa, leading to uterine rupture and life-threatening bleeding. We believe that a bulging sac is a sonographic sign of deep implantation involving the outer myometrium, which predisposes to the second type of sac implantation proposed by Vial et al. Under such circumstances, gestational tissue is not reachable by the curette or suction vacurette, and the myometrium may tear and cause intractable bleeding during curettage.
Evacuation treatment is a convenient conservative procedure. However, there are two uncertainties about this method. The first is that uterine scar discontinuity persists after evacuation. Uterine rupture in the next pregnancy is therefore possible if the defect is not repaired. Although several cases of uneventful pregnancies after D&C have been reported1, Seow et al.18 reported that one patient, primarily treated with evacuation in her previous scar pregnancy, died at 38 weeks' gestation in her subsequent pregnancy because of uterine rupture. The second is that it is difficult to predict massive bleeding during evacuation. A deep sac implantation might be associated with myometrium involvement and a higher risk of profuse hemorrhage. Salvage intervention should be fully prepared.
In conclusion, evacuation therapy is a feasible, albeit not universally applicable, treatment modality of CSP. It is suitable for CSP patients with gestational age below 7 weeks in whom there is no indication of deep sac implantation.
The authors would like to thank Dr Peter Rowley for his assistance in linguistic correctness.