Materials and Methods

This is a retrospective case series of 26 patients between 6-14 weeks postmenstrual age referred to NYU Langone Medical Center over a period of 3 years (2009 through 2011 and evaluated in 2011) with diagnosed or suspected CSP. The diagnosis, treatment, and follow-up of all patients were performed in the ultrasound facility without anesthesia. Twenty-two of the 26 patients had demonstrable fetal heart activity at the time of ultrasound examination in our institution. One patient was referred after she had undergone elective termination of a 7-week pregnancy. However, we subsequently diagnosed that the pregnancy had not been located within the uterine cavity and was located in the hysterotomy scar. One patient was referred because of an arteriovenous (A-V) malformation in the scar of a CD. Two patients presented with CSP with embryos/fetuses without heart activity. Two patients were referred for a second opinion. Twelve women had been treated with various doses (25-50 mg) of intramuscular MTX prior to referral to our institution. Since MTX was not effective in causing cessation of fetal heart activity in these patients they were referred for additional treatment.

In the presence of a positive pregnancy test, a CSP was diagnosed by transvaginal ultrasound using the following criteria:

• 1
  

  Visualization of an empty uterine cavity as well as an empty endocervical canal ( Figure 1 , A and B).

  
  

  
  

  
  

  

  
  

  
  

  Transvaginal sonographic criteria for diagnosis of cesarean scar pregnancy

  
  

  A, Empty uterine cavity with gestational sac ( arrow ) between cavity and cervix (Cx). B, Power Doppler of blood vessels surrounding gestational sac. C, Gestational sac embedded in scar. Thin (1-3 mm) or lack of myometrium ( arrow ) between sac and bladder. D, Triangular shape of sac (on sagittal plane) assuming shape of niche. E-G, Prominent, richly vascular area in site of previous cesarean delivery scar highlighted by power Doppler in patient presenting with bleeding and positive serum human chorionic gonadotropin test. Arrows point to vascular malformation.

  
  

  Timor-Tritsch. Diagnosis and treatment of cesarean scar pregnancy. Am J Obstet Gynecol 2012.

  
  
  
  
• 2
  

  Detection of the placenta and/or a gestational sac embedded in the hysterotomy scar ( Figure 1 , C).

  
  
• 3
  

  In early gestations (≤8 weeks), a triangular gestational sac that fills the niche of the scar ( Figure 1 , D); at ≥8 postmenstrual weeks this shape may become rounded or even oval.

  
  
• 4
  

  A thin (1-3 mm) or absent myometrial layer between the gestational sac and the bladder ( Figure 1 , C).

  
  
• 5
  

  A closed and empty cervical canal.

  
  
• 6
  

  The presence of embryonic/fetal pole and/or yolk sac with or without heart activity.

  
  
• 7
  

  The presence of a prominent and at times rich vascular pattern at or in the area of a CD scar in the presence of a positive pregnancy test ( Figure 1 , E-G).

All these criteria had to be present to diagnose CSP. Some of the above criteria were derived from the literature (items 1, 4, and 5) or generated and modified by our group (items 2, 3, 6, and 7).

Sonographic diagnosis and a baseline serum human chorionic gonadotropin (hCG) concentration were determined. In addition, 3-dimensional (3D) ultrasound data sets using a 4- to 8-MHz transvaginal probe (Voluson 730; General Electric Medical Systems, Milwaukee, WI) were obtained. Volume of the chorionic sac site and power Doppler was used serially after the injection of MTX and compared to baseline information obtained before the local injection of MTX. Power Doppler settings were 0.9 kHz pulse repetition frequency and 200 MHz filter (standardized for all examinations). Chorionic sac volume and vascularization were analyzed offline using a software system (4DView; General Electric Medical Systems). The placenta/gestational sac complex volume (mL) was calculated using the manual segmentation procedure (Virtual Organ Computer-aided Analysis [VOCAL] 4DView; General Electric Medical Systems) ( Figure 2 , A). The outer boundaries of the segmentation, or in other words the perimeter of the gestational sac, were followed to define the sac size. This area/volume also included the vascular “ring.” Six rotational steps (60 degrees apart) were used to define sac volume. The sensitivity for defining the vascularization index (VI) was mentioned above. The VI was calculated using the same software ( Figure 2 , B). The VI is the number of color flow–containing voxels divided by the total number of voxels contained within the volume expressed as a percent value ( Figure 2 , C). The mean VI for patients undergoing hysterectomies was compared to those who were not treated by hysterectomy. Sonographic examinations were repeated for 3 weeks at weekly intervals at first, and subsequently, bimonthly, until the site of the sac was barely visible and the VI declined (usually <3%). We also required that the area of the gestational sac site did not show any more color Doppler signals with a pulse repetition frequency as low as 0.3 kHz.

Evaluation of volume and vascular supply of cesarean scar pregnancy

Evaluation used 3-dimensional (3D) transvaginal ultrasound with Virtual Organ Computer-aided Analysis (VOCAL) software (General Electric Medical Systems, Milwaukee, WI). A, 3D segmentation of sac perimeter drawn around outer boundaries of color ring resulting in sac volume. B, 3D angiographic rendering of vascularization around gestational sac. C, 3D angiographic measurement of vascularization index representing percent blood flow containing units (voxels) over outlined grayscale units.

Timor-Tritsch. Diagnosis and treatment of cesarean scar pregnancy. Am J Obstet Gynecol 2012.

Patients were counseled about the risks of the condition and management alternatives, including potential benefits and risks (known and unknown). The need to adhere to a follow-up period was specified. Patients signed a written informed consent for treatment.

If interventional treatment was recommended as an option, this consisted of a real-time, transvaginal ultrasound-guided puncture and MTX injection into the chorionic sac. An automated, spring-loaded device (Labotect Co, Göttingen, Germany) was attached to the transvaginal transducer (SL400; Siemens, Erlangen, Germany). The procedure represented a slight modification of the puncture injection approach previously reported by the authors. We used a 20-gauge needle. Under ultrasound guidance, the area of the embryonic/fetal heart was identified for the placement of the tip of the needle.

After confirming the placement of the needle, 25 mg of MTX in 1 mL of solution was injected slowly. The intragestational sac dose administered was 25 mg, and an additional 25 mg was injected outside the gestational sac as the needle was withdrawn, preferably the placental site if that area was in the needle tract.

The patient underwent another sonographic examination 60-90 minutes after the procedure to confirm cessation of fetal heart activity and to identify local bleeding. The patient also received an additional intramuscular injection of 25 mg MTX (for a total, combined dose of 75 mg) before discharge from our unit. Patients were asked to return in 24-48 hours for a follow-up scan. As for the number of CD before the CSP, of the 26 patients, 15 had 1, 9 had 2, and 2 had 3 CD.

One patient had 2 chorionic sacs (twin gestation) in the scar, but only 1 gestational sac had detectable embryonic heart activity (an intragestational sac injection was performed in the sac with cardiac activity) since the other sac did not contain viable embryo. One patient had 3 consecutive CSP. All 3 were treated according to the same protocol and counted as 3 separate cases.

The protocol for follow-up included evaluation of the outcome: (1) a weekly serum hCG determination for 3 consecutive weeks, and 1 determination bimonthly until this hormone was undetectable; and (2) determination of the gestational sac volume and the area vascularization at the above intervals using the previously described techniques. Patients were asked not to have vaginal intercourse until the resolution of the CSP. This was judged by sonographic examination.

Analysis of the data was as follows: values of the serum hCG, sac volume, and VI were tabulated for each patient entering them into an Excel spreadsheet (Microsoft, Redmond, WA) on the day they were obtained. These values were used to generate graphic representation, as a function of the days following treatment.

Results

Clinical details of the patients are summarized in Table 1 . Of the 26 patients, 2 of them (patients 4 and 15 in Table 1 ) were referred to us for a second opinion. They each had 1 prior CD and presented at 9 and 14 weeks, respectively. After the diagnosis of CSP ( Figure 3 ) and counseling, both patients opted to continue their pregnancies (after being informed of the risk of a possible placenta accreta). Both patients had uterine rupture with profuse bleeding at 15 and 17 weeks, respectively, requiring massive blood transfusions and hysterectomies.

Two untreated CSPs with subsequent uterine rupture and hysterectomy

A, 3-Dimensional power Doppler angiogram at 9 weeks of patient 4 in Table 1 . B, 2-Dimensional color Doppler ultrasound image at 14 postmenstrual weeks of patient 15 in Table 1 .

CSP , cesarean scar pregnancy cases.

Timor-Tritsch. Diagnosis and treatment of cesarean scar pregnancy. Am J Obstet Gynecol 2012.

Patient 10 in Table 1 was scheduled to have intragestational sac MTX injection of a CSP at 6 weeks and 1 day, but slightly bled prior to the scheduled procedure. The patient was treated by tamponade with a 5-mL balloon catheter inserted into the cervix and inflated until bleeding ceased. The next morning, there was absence of detectable fetal heart rate, and no additional treatment was given. Six weeks later, involution of the scar site was noted.

On the day of referral, 2 patients (patients 23 and 24 in Table 1 ) had detectable embryonic/fetal cardiac activity and were scheduled for treatment, but the following day (when the procedure was scheduled), fetal cardiac activity had ceased. No treatment was administered. Patient 23 received intramuscular MTX prior to referral, while for patient 24, the fetal cardiac activity ceased without MTX administration. These patients were followed up according to the protocol described above.

Patient 9 in Table 1 had a complex clinical course. This 33-year-old patient had 6 pregnancies, 4 deliveries, and 1 abortion, and presented to the emergency room with vaginal bleeding 67 days after an attempted elective termination of pregnancy at 7 weeks of gestation at another institution (the pathology report described the presence of chorionic villi). The patient had 2 previous CD and 2 normal vaginal deliveries at term. At presentation, the serum hCG was 55 mIU/mL, and sonographic examination at our center revealed an empty uterine cavity, a clearly imaged hysterotomy scar niche ( Figure 4 , A), and a richly vascularized anterior uterine wall (which was double in thickness compared to the posterior wall) ( Figure 4 , B). We considered that the images were consistent with the diagnosis of placenta accreta or percreta that was left untouched during the termination procedure. The pregnancy was in close proximity to the hysterotomy scar. We managed this condition by administering intramuscular MTX (80 mg) on day 81 after her initial dilatation and curettage (D&C) on the first day under our care. This injection was administered with the suspicion that the patient may have had residual gestational trophoblastic disease. On follow-up the hCG serum concentration became nondetectable 2 weeks (on the 100th day) from the time of the initial surgical intervention. The VI and placental volume showed a decrease in magnitude on the 105th day. However, the patient developed severe vaginal bleeding. A hysterectomy and uterine artery embolization were offered, but declined by the patient. A repeat sonographic examination demonstrated an increase in the VI. The ultrasound image was suspicious for the presence of an A-V malformation ( Video Clips 1 and 2 ). Vaginal bleeding persisted, and on the 155th day bilateral uterine artery embolization was performed. Vaginal bleeding decreased, but there was persistence of the prominent vessel in the lower anterior uterine wall ( Figure 4 , C). The peak systolic velocity within the vascular structure was 45.3 cm/s, consistent with an A-V vascular malformation ( Figure 4 , D). Five days later, the patient underwent a hysterectomy with an uneventful recovery. The sequence of events is illustrated in Table 2 .

Placenta percreta in case no. 9 from Table 1

A, Sagittal section of uterus. Anatomy is outlined by dotted lines and annotations indicate placental location, cesarean section (C/S) niche, empty uterine cavity, and cervical canal. B, 3-Dimensional power Doppler image of vascularization. C, After 140-144 days large dilated blood vessel is seen. Inlay represents color flow of vessel. D, Peak systolic velocity of 45.3 cm/s was measured in vessel.

Timor-Tritsch. Diagnosis and treatment of cesarean scar pregnancy. Am J Obstet Gynecol 2012.

Patient 26 in Table 1 was referred to us for vaginal bleeding and a positive pregnancy test. On transvaginal ultrasound an A-V malformation was seen at the site of her previous CD scar ( Figure 1 , E-G). This patient did not have any surgical intervention for this pregnancy and was promptly treated by emergency uterine artery embolization to stop the bleeding. Two other patients had no demonstrable embryonic/fetal cardiac activity on the day of their scheduled MTX injection thus were not treated at all.

In only 1 patient (patient 3 in Table 1 ) was the CSP the result of in vitro fertilization and transfer of 2 embryos. Nineteen patients (6-9 weeks of gestation) underwent successful local injection of 50 mg of MTX and all showed evidence of embryonic/fetal cardiac activity. One patient had 3 prior CD. Typically, patients had prolonged, intermittent vaginal spotting for 2-3 weeks following the procedure. During the follow-up period, most women resumed menses before resolution of the gestational sac volume and vascularization. No side effects were seen related to the MTX treatments.

Of interest, 1 patient with 2 previous CD underwent intragestational sac MTX injection of 50 mg at 7 postmenstrual weeks for a CSP, and subsequently returned 10 weeks later with a second CSP at 6 postmenstrual weeks. She underwent again intragestational sac MTX injection. It is noteworthy that the first CSP was a dichorionic twin gestation with 1 empty sac (blighted ovum?), and an additional gestational sac containing an embryo. This same patient returned again, 4 months after her second CSP similarly treated with a third CSP at 5 postmenstrual weeks and 6 days. She was treated again as per our described protocol with good outcome.

A small number of clots from the uterine cavity were aspirated on day 26 in patient 25 after continuous spotting was reported.

The following observations were noted regarding the hCG serum concentrations, the gestational sac volume, and the VI:

• 1
  

  Serum hCG: in 13 of the 19 injected cases after an initial plateau or a small temporary increase in the serum hCG concentrations, the values decreased slowly and became nondetectable (cutoff was <3 mIU/L) 41-100 days following MTX injection ( Figure 5 ).

  
  

  
  

  
  

  

  
  

  
  

  Graph of serum hCG as function of days post injection

  
  

  After initial increase most levels dropped to undetectable levels by day 40-60.

  
  

  hCG , human chorionic gonadotropin.

  
  

  Timor-Tritsch. Diagnosis and treatment of cesarean scar pregnancy. Am J Obstet Gynecol 2012.

  
  
  
  
• 2
  

  Gestational sac volume: in 12 of the cases the gestational sac volume increased or plateaued after MTX injection, and this was followed by a slow decrease in volumes ( Figure 6 ). However, the area of involution was visible even >5 months’ posttreatment.

  
  

  
  

  
  

  

  
  

  
  

  Graph of gestational sac volumes as a function of days post injection

  
  

  Timor-Tritsch. Diagnosis and treatment of cesarean scar pregnancy. Am J Obstet Gynecol 2012.

  
  
  
  
• 3
  

  VI: in 14 of the cases after an initial increase or brief plateau in the VI, a slow but steady decline was observed to what was considered to be minimal values (<3%). Color Doppler did not demonstrate vascularization 30-140 days from the MTX injection ( Figure 7 ).

  
  

  
  

  
  

  

  
  

  
  

  VI as function of time after intragestational sac injection of methotrexate

  
  

  VI increased after injection and steadily dropped thereafter.

  
  

  VI, vascularization index.

  
  

  Timor-Tritsch. Diagnosis and treatment of cesarean scar pregnancy. Am J Obstet Gynecol 2012.

  
  

The interquartile ranges for the serum hCG concentrations, the sac volumes, and VI are presented in Table 3 .

TABLE 3

Mean, SE, and interquartile range for serum hCG, volume of gestational sac, and VI

Measure	Mean	SE	25-75% range
hCG, mIU/mL	4334.6	1114.1	9.0–4677.0
Volume, mL	18.1	3.1	2.4–24.6
VI	18.9	2.1	7.0–26.4

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Where should I have my baby?
2. Biomarkers of progestin therapy resistance and endometrial hyperplasia progression
3. Discussion: ‘Moderate ambient level of carbon monoxide and risk of preeclampsia’ by Zhai et al
4. Smoking during pregnancy influences the maternal immune response in mice and humans

Stay updated, free articles. Join our Telegram channel

Join