Operative laparoscopy for unruptured ectopic pregnancy in a caesarean scar


  • Y-L Wang,

    Corresponding author
    1. Department of Obstetrics and Gynecology, Mackay Memorial Hospital and Mackay Medicine, Nursing and Management College, Taipei, Taiwan
    Search for more papers by this author
  • T-H Su,

    1. Department of Obstetrics and Gynecology, Mackay Memorial Hospital and Mackay Medicine, Nursing and Management College, Taipei, Taiwan
    Search for more papers by this author
  • H-S Chen

    1. Department of Obstetrics and Gynecology, Mackay Memorial Hospital and Mackay Medicine, Nursing and Management College, Taipei, Taiwan
    Search for more papers by this author

Dr Y-L Wang, Department of Obstetrics and Gynecology, Mackay Memorial Hospital and Mackay Medicine, Nursing, and Management College, 92, Section 2, Chung-Shan North Road, Taipei 10449, Taiwan. Email K9850316@ms24.hinet.net


Eight women diagnosed by transvaginal ultrasonography with unruptured caesarean scar pregnancy underwent operative laparoscopy as an alternative treatment to laparotomy. The ultrasonographic diagnosis of caesarean scar pregnancy was confirmed in all women at laparoscopy. None of the women required conversion of the procedure to laparotomy. The total operative time ranged from 75 to 120 minutes. The total blood loss was limited, ranging from 50 to 200 ml. All women tolerated the operation well and had uneventful recoveries. Our results show that in the hands of a well-trained operator, laparoscopy appears to be a reasonable alternative for the management of an unruptured caesarean scar pregnancy.


Implantation within the scar from a previous caesarean section is one of the rarest forms of ectopic pregnancy.1 Because there are very few reports of such cases, little is known about the optimal management.1 The natural history of this condition is unknown, but uterine rupture and haemorrhage, even in the first trimester, seem likely if the pregnancy is allowed to continue. Such a complication is extremely dangerous and may require hysterectomy with consequent loss of fertility.2

Ultrasound examination may permit early, accurate diagnosis of caesarean scar pregnancy. Intervention at this point may then allow preservation of the uterus and fertility without a high risk of maternal complications.3–6 We present our experience with eight women with caesarean scar pregnancy diagnosed using transvaginal ultrasound scan and treated by laparoscopy as an alternative to laparotomy.

Materials and methods

From August 2003 to July 2005, a total of 446 ectopic pregnancies were diagnosed at Mackay Memorial Hospital. Among these, eight women had an unruptured ectopic pregnancy in a lower segment caesarean section scar. All diagnoses were made by the author using transvaginal ultrasonography. Four women had undergone suction dilatation and curettage at local clinics before being transferred to our hospital for further management. The other four had not yet had any treatment.

The diagnosis of caesarean scar pregnancy was made if all the following sonographic criteria were met: (1) the uterus was empty, with a clearly demonstrated endometrium; (2) the cervical canal was empty; (3) the gestational sac was located at the anterior part of the uterine isthmus and (4) the gestational sac, with or without cardiac activity, was embedded in and surrounded by the myometrium and the fibrous tissue of the scar, and it was separate from the endometrial cavity or fallopian tube (Figure 1).

Figure 1.

Ultrasound image showing a well-encapsulated, bulging mass with a regular gestational sac containing a yolk sac and living embryo located within the anterior uterine isthmus, in the location of the previous caesarean scar.

Laparoscopy was performed to confirm the diagnosis and to remove the gestational products. The defect in the uterus was then repaired by extracorporeal suturing.

Operative procedure

Laparoscopy was performed under general anaesthesia with the woman in the 15° Trendelenburg position. A Foley catheter was inserted preoperatively to empty the bladder and enable continuous monitoring of urine output. A Verres needle was inserted through a small incision just inferior to the umbilicus and a pneumoperitoneum created by insufflating carbon dioxide to a maximal pressure of 20 mmHg. A 10-mm operative trocar was then inserted into the abdominal cavity, and a laparoscope with attached video camera was passed through the cannula to visualise the intra-abdominal organs. A 5-mm trocar was inserted into the pelvic cavity in the midline suprapubically, and two additional 5-mm trocars were inserted at the level of the anterior superior iliac spine, lateral to the epigastric blood vessels. After placement of the trocars, the intra-abdominal pressure was decreased to 15 mmHg. Although intending to manage the condition laparoscopically, we were prepared to perform immediate laparoscopically assisted vaginal hysterectomy or to convert to open laparotomy if serious bleeding developed.

The serosa was incised and the bladder pushed down to give access to the lower uterine segment. In each case, a mass with a thin wall of myometrium was seen (Figure 2). Dilute vasopressin (1 unit/ml) was used for haemostasis. We injected 5–10 ml of vasopressin solution into the myometrium at one or more sites with an 18-gauge spinal needle placed directly through the abdominal wall and waited until blanching occurred. A transverse incision was then made over the most prominent area of the mass, revealing in each case a dark red gestational sac which was removed using grasping forceps. The resulting space in the myometrium was cleaned using suction irrigation, and haemostasis was achieved using Wolf bipolar forceps at 20 W. One layer of interrupted 2-0 polyglactin sutures was placed in the uterine wall using the extracorporeal method (Figure 3). The gestational tissue was removed in an endobag.

Figure 2.

A mass with a thin myometrium and a protruding gestational sac in the area of the previous caesarean section scar.

Figure 3.

A single layer of interrupted 2-0 vicryl sutures placed in the uterine wall.


The women’s ages ranged from 28 to 42 years. All previous caesarean sections had been performed by transverse incision of the lower segment. Two women had had three previous caesarean sections, four had had two and four had had only one. The interval between caesarean section and the scar pregnancy ranged from 1 to 7 years (Table 1). None of the women had a history of heart, lung, liver, kidney or other chronic disease.

Table 1.  Clinical data of women undergoing laparoscopy for caesarean scar pregnancy
CaseAge (years)Previous lower segment caesarean section (n)Time interval from last caesarean section (years)Gestational age (weeks)Fetal cardiac activityInitial human chorionic gonadotrophin (iu/l)Presenting symptomsBlood loss (ml)Blood transfusionOperation time (minutes)
1282Unknown8Yes53 723Pain and vaginal bleeding50No120
2423310Yes27 674Profuse bleeding after pregnancy interruption100Yes90
33631UnknownUnknown29.70Intermittent vaginal bleeding and lower abdominal pain after pregnancy interruption100No120
43214.57No18 584Pain and vaginal bleeding100No80
5322Unknown8No78 055Profuse bleeding after pregnancy interruption200Yes80
642259No60 309Pain and vaginal bleeding200No95
729138No32 301Pain and vaginal bleeding100No120
84127UnknownNo3101.8Profuse bleeding after pregnancy interruption50No75

Laparoscopy was successful in treating all eight women. None required conversion to laparotomy. The total operative time ranged from 75 to 120 minutes. The total blood loss was limited, ranging from 50 to 200 ml, although two women had previously required blood transfusion because of profuse bleeding after suction curettage at local clinics. Pathology examination in each case revealed blood clots, chorionic villi and prominent decidual tissue, all consistent with ectopic pregnancy. Patient 6 chose to undergo tubal ligation. Patient 3 successfully conceived again 3 months after laparoscopy and went on to deliver a healthy term baby by lower segment caesarean section. All women were discharged on the second day after laparoscopy.


Although caesarean section is a very common procedure, caesarean scar pregnancy occurs very rarely. The incidence seems to be increasing, however, possibly because of the increased performance of caesarean section and more widespread use of transvaginal ultrasound scan as a diagnostic method.7 Early intervention is important as a scar pregnancy may result in uterine rupture and haemorrhage. If discovered at that late point, it may be impossible to preserve the uterus. Sonography is a first-line diagnostic tool for caesarean scar pregnancy. However, it can be difficult to differentiate a scar pregnancy from a miscarriage in progress or a cervicoisthmic pregnancy. Serial transvaginal ultrasound scans should be performed in all women with possible gestational pathology.

Because of the rarity of this particular ectopic implantation, there are no universal treatment guidelines.8,9 From 1978 to March 2006, 91 cases have been reported in the English literature. Medical treatment options, including systemic and ultrasound-guided local methotrexate, potassium chloride and hyperosmolar glucose, have been used successfully,6,10–13 but they do have disadvantages. Methotrexate, for example, has reportedly been successful in 29 of 40 cases. However, a subsequent caesarean scar pregnancy has occurred after methotrexate.14 One drawback, then, is that medical treatment leaves the original scar in place, a scar which has already demonstrated a predisposition to ectopic implantation. Surgery would still be necessary if conservative treatment fails. Medical treatment also results in only a relatively slow decline in β-human chorionic gonadotrophin levels. In addition, there is the potential for massive bleeding and uterine rupture before the condition resolves.

Curettage is contraindicated because the trophoblastic tissue is outside the uterine cavity and thus is unreachable by the curette, in addition to which, it might result in rupture and severe haemorrhage. In a review by Arslan et al.,15 uterine curettage was either unsuccessful or caused complications in eight of nine women. The four women in our series who had undergone pregnancy interruption with suction dilatation and curettage at outside clinics all presented with profuse vaginal bleeding. Two required blood transfusions.

Only surgery offers the opportunity to remove the pregnancy and simultaneously repair the defect. According to Fylstra’s review,2 termination of the pregnancy by either laparotomy or hysterotomy with repair of the accompanying uterine scar dehiscence is probably the best treatment for caesarean scar pregnancy. Vial et al.5 have also suggested surgical resection of the old scar and a new closure be offered even if recurrence is thought to be unlikely.

The effectiveness and safety of operative laparoscopy in the treatment of reproductive and gynaecologic lesions is well established. At our hospital, laparoscopically assisted vaginal hysterectomy is the most common procedure for removal of the uterus. Given our experience with this procedure, we felt it was safe to attempt to manage scar pregnancies laparoscopically, knowing that if intractable bleeding intervened, we could immediately perform a laparoscopically assisted vaginal hysterectomy. Conversion to open laparotomy would have been the procedure of last resort.

Lee et al.9 claim to have performed the first successful laparoscopic resection of a caesarean scar pregnancy in a woman who had undergone dilatation and curettage for termination of pregnancy. However, there was not enough information available in that report (e.g. gestational age or results of ultrasound scan before the original curettage abortion) to determine if it was a true ectopic pregnancy or if it was uterine perforation at the old caesarean scar site at by the curettage resulted in an intrauterine pregnancy passing through the rupture, thus mimicking ectopic implantation. Four of our women had previously undergone suction curettage, so the same question might be raised, although all the four women fulfilled the ultrasonic criteria for a scar pregnancy. In addition, both the laparoscopic and pathology findings were consistent with implantation in the scar. The other four had had no previous intervention, so there can be little doubt about the diagnosis in those women.

Vasopressin is an effective haemostatic agent in vaginal surgery.16 Bryman et al.17 demonstrated the contractile effect of vasopressin on the cervix in vitro. We relied on local injection of vasopressin and the tamponade effect of the pneumoperitoneum to decrease bleeding during the procedure. The blood loss was in fact minimal, with transfusions required only for the two women who had already had substantial bleeding related to procedures performed elsewhere.

In summary, our cases series demonstrates the value of laparoscopy in treatment of caesarean scar pregnancy. This procedure avoids open laparotomy and preserves the woman’s reproductive capacity. While more invasive than conservative medical treatment with methotrexate, for example, laparoscopy also allows revision of the scar, which it is to be hoped will reduce the risk of a recurrent ectopic pregnancy in the same location. In the hands of a well-trained operator, laparoscopy appears to be a reasonable approach, as long as facilities are available for an immediate change in the procedure should uterine rupture and massive bleeding intervene.