Cesarean Scar Ectopic Pregnancies

Etiology, Diagnosis, and Management

Michael A. Rotas, MD, Shoshana Haberman, MD, PhD, and Michael Levgur, MD

OBJECTIVE: To clarify the appropriate way to diagnose cessful in the management of cesarean delivery scar
and treat an ectopic pregnancy in the uterine scar of a pregnancy. Because subsequent pregnancies may be
prior cesarean delivery. complicated by uterine rupture, the uterine scar should
be evaluated before, as well as during, these pregnancies.
DATA SOURCES: Articles written in English that were
(Obstet Gynecol 2006;107:1373–81)
published from January 1966 to August 2005 and quoted
in the computerized database MEDLINE/PubMed re-

P
trieved by using the words “cesarean section,” “cesarean regnancy in the scar from a cesarean delivery is
delivery,” “cesarean section scar pregnancy,” and “ec- located outside the uterine cavity and is completely
topic pregnancy.” Additional articles were obtained from surrounded by myometrium and fibrous tissue of the
reference lists of pertinent case reports and reviews. scar in the prior low uterine segment. The recognized
METHODS OF STUDY SELECTION: Fifty-nine articles long-term risks of cesarean delivery are subsequent
that met the inclusion criteria provided data on the ectopic pregnancies, uterine rupture, and placental dis-
clinical presentation, diagnosis, and treatment modalities orders in future pregnancies such as abruptio placentae,
of 112 cases of cesarean delivery scar pregnancies. placenta previa, and placenta accreta, which is the most
TABULATION, INTEGRATION, AND RESULTS: Review serious condition.1,2 However, endometrial and myome-
of the 112 cases revealed a considerable increase in the trial disruption and scarring subsequent to cesarean
incidence of this condition over the last decade, with a delivery also may predispose to implantation in the
current range of 1:1,800 to 1:2,216 normal pregnancies. uterine scar, which is even more dangerous than pla-
More than half (52%) of the reported cases had only one centa accreta. Invasion of the myometrium early in the
prior cesarean delivery. The mean gestational age was 7.5
first trimester may lead to uterine rupture and profuse
ⴞ 2.5 weeks, and the most frequent symptom was
bleeding as the pregnancy advances.3
painless vaginal bleeding. Endovaginal ultrasonography
was the diagnostic method in most cases, with a sensi-
There is minimal awareness of the possibility of
tivity of 84.6% (95% confidence interval 0.763– 0.905). gestation in a previous cesarean scar, which is often
Expectant management of 6 patients resulted in uterine misdiagnosed as a cervical or aborting pregnancy.
rupture that required hysterectomy in 3 patients. Dilation Because suspicion is low, diagnosis of an early preg-
and curettage was associated with severe maternal mor- nancy in a prior cesarean scar may be delayed, and
bidity. Wedge resection and repair of the implantation potentially catastrophic complications may ensue.
site via laparotomy or laparoscopy were successful in 11 We recently encountered a case of a pregnancy in
of 12 patients. Simultaneous administration of systemic a cesarean scar, which triggered a thorough search of
and intragestational methotrexate to 5 women, all with the medical literature to ascertain the most effective
␤-hCG exceeding 10,000 milli-International Units/mL re- approach to this form of ectopic pregnancy. Because
quired no further treatment. we were impressed with the exponential increase in
CONCLUSION: Surgical treatment or combined sys- the number of cases reported over the last 5 years, we
temic and intragestational methotrexate were both suc- proceeded with a systematic review of the topic. This
article outlines the etiology and the predisposing risk
From the Department of Obstetrics and Gynecology, Maimonides Medical factors and updates our knowledge of available treat-
Center, Brooklyn, New York. ments for this life-threatening condition.
Corresponding author: Michael Levgur, MD, Maimonides Medical Center, 967
48th street, Brooklyn, NY 11219; e-mail: mlevgur@maimonidesmed.org.
SOURCES
© 2006 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins. The primary investigator and a medical librarian
ISSN: 0029-7844/06 searched the computerized database MEDLINE/

VOL. 107, NO. 6, JUNE 2006 OBSTETRICS & GYNECOLOGY 1373

PubMed using the medical heading search words “ce- Etiology
sarean section,” “cesarean delivery,” “cesarean section In cesarean scar pregnancy, invasion of the conceptus
scar,” and “ectopic pregnancy.” The search was limited into the myometrium is believed to occur through a
to human subjects and the English language for the microscopic dehiscence or a defect in the scar second-
period January 1966 through August 2005. Two non- ary to poor vascularization of the lower uterine seg-
English articles were reviewed, but their abstracts did ment with fibrosis and incomplete healing.4 This
not include adequate information on the treatment mechanism is similar to the one generally accepted
applied. In addition, articles were retrieved from refer- for intramural pregnancy, where dehiscence or tract
ence lists of pertinent case reports and reviews. formation follows uterine manipulation such as curet-
tage, myomectomy, hysteroscopy, metroplasty, and
STUDY SELECTION
manual removal of placenta.9,10 Such a tract or defects
Fifty-nine articles met the search criteria, including 54 in the cesarean scar can be detected by endovaginal
retrieved from MEDLINE database and 5 from cross- ultrasonography years after a cesarean delivery, and
referencing. Four of these articles were case series that
their size and depth can be measured at sonohyster-
included 8, 12, 15, and 18 cases, respectively. The
ography.11 Defects and thinned uterine myometrium
total number of cases reported was 112. All selected
were confirmed with endovaginal ultrasonography in
publications provided information on clinical presen-
50% of 47 women 3 or more months after a low
tation and diagnosis, with emphasis on success, fail-
segment cesarean delivery.12 Cesarean delivery scar
ure, and complications of different treatments.
defect was also diagnosed in the nonpregnant state by
RESULTS the presence of triangular-shaped fluid collection
within the incision site.13,14 According to Jurkovic et
Epidemiology
al,4 the scar in the majority of postpartum patients is
Recent literature suggests that cesarean delivery scar
well healed. However, particularly after many cesar-
pregnancy is more common than previously thought.
eans, the scar surface is increased, and the anterior
Its incidence ranges from 1:1,800 to 1:2,216 pregnan-
uterine wall may be deficient because of poor vascu-
cies, and it constitutes 6.1% of all ectopic pregnancies
larity, fibrosis, and impaired healing. Consequently
with a history of at least one cesarean delivery.4,5 This
the likelihood of implantation into such a scar is
incidence exceeds somewhat that of cervical pregnancy,
increased. Indeed, 72% of patients with pregnancy in
which occurs in 1:2,000 to 1:18,000 pregnancies.6
the uterine scar reported by this author had more than
Larsen and Solomon7 reported the first case of
cesarean scar pregnancy in 1978, and the first review 2 cesarean deliveries.4 Our search did not support this
article by Fylstra3 summarized 18 cases accumulated notion inasmuch as, out of 75 women whose obstetric
up to 2002. Since then, 94 more cases were reported, histories were available, 39 (52.0%) had only one
for a total of 112 cases, summarized in 4 series and previous cesarean delivery, 27 (36.0 %) had two, and
several case reports. The recent increase in the num- 9 (12.0 %) had more than two. Other authors also
ber of cases may reflect the great increase in the supported the notion that the number of cesarean
number of cesarean deliveries worldwide, but also deliveries has no impact.15
may be directly related to improved diagnostic accu- Maymon et al16 reported on an interesting asso-
racy and high index of suspicion.8,4 ciation between cesarean deliveries for breech presen-
The mean age of patients presenting with cesarean tation and subsequent scar pregnancies. This was also
scar pregnancy was 33.4 ⫾ 5.7 years and was not affected acknowledged by others.17–20 The hypothesis is that
by the number of prior cesarean deliveries (Table 1). The many cesarean deliveries for this indication are elec-
mean gestational age at presentation was 7.5 ⫾ 2.5 weeks. tive, and thus there is a poorly developed lower
uterine segment that may lead to faulty healing and,
consequently, implantation within the scar. The indi-
Table 1. Mean Age and Number of Prior cations for cesarean delivery that preceded the abnor-
Cesarean Deliveries in 75 Women With mal implantation were documented in 35 of the 112
Cesarean Scar Pregnancies reported cases and are shown in Table 2, with breech
Age (y, mean ⴞ SD) Previous Cesareans n % presentation the most frequent. Another factor con-
33.1 ⫾ 6.3 1 39 52 tributing to the recently increased incidence of these
33.7 ⫾ 4.8 2 29 27 abnormal implantations may be the change in surgi-
33.7 ⫾ 5.3 3 or more 9 12 cal technique for repairing the uterine incision. In the
SD, standard deviation. past a double-layer closure was performed, with su-

1374 Rotas et al Pregnancy in Uterine Scar OBSTETRICS & GYNECOLOGY

Table 2. Indications for Prior Cesarean Delivery in Table 3. Symptoms of Cesarean Scar Pregnancies
35 Women With Gestation in Uterine in 57 Women
Section Scar Presenting Symptom n %
Indication n % Asymptomatic 21 36.8
Breech presentation 11 31.4 Painless vaginal bleeding 22 38.6
Fetal distress 8 22.9 Abdominal pain and bleeding 9 15.8
Arrest of labor 7 20 Abdominal pain 5 8.8
Preeclampsia 4 11.4
Placenta previa 3 8.6
Twin gestation 2 5.7 were incorrectly diagnosed as cervical pregnancies or
incomplete abortions. Generally ultrasonography can
detect an enlarged scar in the lower uterine segment
with an embedded mass (Fig. 1). A very thin myome-
tures inverting the first layer by the second one. A
trium may be visualized between the bladder and the
single noninverting running suture, as commonly
gestational sac (Fig. 2).16,22 In 10 of 15 patients re-
used today, may lead to impaired postoperative heal-
ported in one case series, the myometrial thickness at
ing and creation of defects within the scar.16 However,
the implantation site ranged between 2 and 5 mm.22
none of the reports provided information about the
Many authors combined endovaginal with transab-
technique used.
dominal ultrasonography to permit imaging of fine
Diagnosis details of the ectopic pregnancy and its surroundings
and to obtain a panoramic view of the uterus (Fig.
Because a delay in the diagnosis of pregnancy in the
3).16,23 Vial et al19 proposed the following sonographic
uterine scar may result in rupture, a prompt and
criteria for the diagnosis of this condition, which were
accurate diagnosis is crucial.21 The diagnosis should
accepted later by Fylstra3 and Godin et al24: 1) The
rely on the patient’s history and clinical manifesta-
trophoblast is located between the bladder and the
tions, including abdominal pain and bleeding, which
anterior uterine wall; 2) fetal parts are not present in
may range from spotting to life-threatening hemor-
the uterine cavity; 3) on a sagittal uterine view that
rhage. According to Seow et al,5 history per se may be
runs through the amniotic sac, no myometrium is
helpful in differentiating this condition from other
seen between the gestational sac and the urinary
forms of pregnancy failure. Spontaneous or inevitable
bladder, as illustrated by the lack of continuity of the
abortions begin with more extensive bleeding from
anterior uterine wall.
the detached chorionic sac. Furthermore, most abort-
To avoid confusion with the expulsion of the
ing patients complain of cramping or lower abdomi-
conceptus in abortion or a cervicoisthmic implanta-
nal pain and sometimes exhibit cervical motion or
adnexal tenderness, in contrast to pregnancies in a
uterine scar, which present with only mild or moder-
ate lower abdominal pain.
Of the 112 reported cases summarized in this
review, we could retrieve 57 (50.9%) that provided
information on the presenting symptoms. Thirty-six
had either vaginal bleeding, ranging from spotting to
severe hemorrhage, bleeding, and abdominal cramps,
or low abdominal pain as the only manifestation. The
remaining 21 women were asymptomatic and diag-
nosed by ultrasonography only after referral for fur-
ther evaluation to rule out an ectopic pregnancy
(Table 3).
Several techniques have been used to diagnose
uterine scar pregnancies. Endovaginal ultrasound ex-
amination was the primary diagnostic modality in all
of the published cases except for the first one reported Fig. 1. Endovaginal ultrasonography demonstrating a preg-
by Larsen and Solomon.7 It correctly diagnosed 94 of nancy implanted in the low uterine segment. Large white
the 111 cases, for a sensitivity of 84.6% (95% confi- arrow, gestational sac; white arrowhead, cervix.
dence interval 0.763– 0.905). The remaining 17 cases Rotas. Pregnancy in Uterine Scar. Obstet Gynecol 2006.

VOL. 107, NO. 6, JUNE 2006 Rotas et al Pregnancy in Uterine Scar 1375
 endovaginal probe. Maymon et al25 was opposed to
this technique because it may lead to vaginal bleeding
and even to uterine rupture.
Color Doppler imaging and 3-dimensional power
Doppler ultrasonography may enhance the diagnostic
capability of endovaginal ultrasonography by evalu-
ating the flow, resistance, and pulsatility indices in the
peritrophoblastic vasculature.26 –29 High velocity and
low impedance surrounding an ectopic gestational sac
are consistent with viable early pregnancy.4,5 On
pulsed Doppler examination, flow waveforms of high
velocity (peak velocity ⬎ 20 cm/s) and low imped-
ance (pulsatility index ⬍ 1) have been reported in
cases of cesarean scar gestations.4
A new technique of color Doppler imaging,
termed 3-dimensional vocal imaging system, can
quantify changes in uterine neovascularization sur-
Fig. 2. Endovaginal ultrasonography demonstrating the
gestational sac of a cesarean scar pregnancy separated from rounding a uterine scar pregnancy. This technique is
the bladder with a thin layer of myometrium. Large white particularly helpful in monitoring the response to the
arrow, gestational sac; white arrowhead, cervix; small primary treatment such as uterine artery embolization.29
white arrow, urinary bladder. Few authors used magnetic resonance imaging as
Rotas. Pregnancy in Uterine Scar. Obstet Gynecol 2006.
an adjuvant to endovaginal ultrasonography30 –32 to
improve intraoperative orientation.21 Nevertheless,
some authors advocate the use of magnetic resonance
imaging only if endovaginal ultrasound examination
fails to identify the typical findings of a cesarean scar
pregnancy.16
Finally, endoscopic modalities, such as cystos-
copy, were used to rule out bladder penetration.32
Using hysteroscopy, Roberts et al33 described a
salmon red appearance of the lesion.

Treatment Modalities
Because cesarean scar pregnancy is rare, experience is
based mainly on case series, and thus no therapeutic
Fig. 3. Transabdominal ultrasonography showing a pan-
protocols have been established universally. In most
oramic view of the uterus, bladder, and the cesarean scar cases, modality of treatment selection was based on
pregnancy. Large white arrow, gestational sac; white arrow- severity of symptoms, ␤-hCG levels, and surgical
head, cervix; small white arrow, urinary bladder; ENDO, experience.
endometrium.
Rotas. Pregnancy in Uterine Scar. Obstet Gynecol 2006.
Expectant Management
The notion that the sac of the pregnancy in the scar is
4
tion, Jurkovich et al reiterated the importance of the connected to the uterine cavity, and thus progression to
absence of healthy myometrium between the bladder term pregnancy is feasible, led some authors to recom-
and the gestational sac, while adding the following mend expectant management.19 However, this ap-
criteria: 1) On Doppler imaging, the sac is well proach may result in uterine rupture.5,16,34 Of the 6
perfused in contrast to the avascular appearance of an patients observed expectantly, 3 had uterine rupture and
aborting gestational sac; 2) the negative “sliding or- severe hemorrhage and disseminated intravascular co-
gans sign,” defined as the nondisplacement of the agulation that mandated hysterectomy (Table 4).21,34
gestational sac from its position at the level of the Severe bleeding complicated the remaining three
internal os when gentle pressure is applied by the cases, which was controlled with salvage treatments.4

1376 Rotas et al Pregnancy in Uterine Scar OBSTETRICS & GYNECOLOGY

Table 4. Various Treatments Provided to 112 Women With Cesarean Scar Gestation
Treatment Mode Mean ␤-hCG* Successful (n) Complicated (n) Hysterectomy† (n) Total (n)
Observation N/A 2 4 3 6
Laparotomy 16,700 8 1 0 9
Laparoscopy 14,457 3 0 0 3
Hysteroscopy N/A 1 0 0 1
Dilation and curettage 17,044 5 16 3 21
Systemic methotrexate 10,347 7 9 1 16
Local methotrexate 19,456 8 7 0 15
Combined methotrexate 24,124 6 0 0 6
Trichosanthin and mifepristone N/A 6 9 3 15
Other N/A 9 11 1 20
Total N/A 55 57 11 112
* Serum levels in milli-International Units per milliliter.
†
Included in the complicated cases.

Surgical Management Selective Uterine Artery Embolization

Laparotomy and Laparoscopy This invasive radiologic procedure has recently
A wedge excision of the gestational mass was done via gained wide acceptance as a conservative method for
elective laparotomy in 9 cases (Table 4). Eight had no various obstetric and gynecologic conditions such as
sequelae, and one resulted in severe hemorrhage and postpartum hemorrhage, uterine myomas, and cervi-
later endometritis.19,27,32,35,36 The follow-up period was cal pregnancies.46,47 Some authors have considered it
shorter than with medical treatment, and the risk of as the only alternative to hysterectomy for control of
uterine rupture or recurrence at the site of repair was bleeding in cases of trophoblastic tissue invading the
less likely.19,25 The same procedure was also per- vesicouterine space.48 Of the series published by
formed laparoscopically in 3 cases, all with uneventful Sugaware et al,49 3 women underwent uterine artery
postoperative course.26,37,38 embolization primarily, followed by systemic or local
methotrexate administration. Severe bleeding re-
Hysteroscopy quired a dilation and curettage in 2 patients, and 1
This minimally invasive procedure was recently used patient needed a second embolization. In another
in the management of pregnancy in a cesarean deliv- similar case, uterine artery embolization coupled with
ery scar.39 Hysteroscopy enabled the identification of systemic methotrexate injection failed, as demon-
the embryonic sac and the distribution of vessels at strated by increased peritrophoblastic vascularity in a
the implantation site. The sac was separated from the follow-up 3-dimentional ultrasound scan.50
uterine wall with the operative hysteroscopy and the
vessels were electrocoagulated to assure hemostasis. Medical Management
Systemic Methotrexate
Dilation and Curettage Given intramuscularly, this drug has been used exten-
This procedure was not only suboptimal but also sively as first-line treatment in cases of tubal and
created risks because the trophoblastic tissue is lo- cervical pregnancies if gestational age is less than 9
cated outside the uterine cavity and thus unreachable, weeks, fetal pole size does not exceed 10 mm in size,
except in the rare case of a connection between the no embryonic cardiac activity is seen, and serum
gestational sac and the uterine cavity, as demon- ␤-hCG levels are less than 10,000 milli-International
strated on endovaginal ultrasonography.40 Our litera- Units/mL.51 This technique was initially employed for
ture search identified 21 cases (Table 2) that were the management of cesarean uterine scar as an ad-
managed primarily by dilation and curettage; only 5 junct to hysterotomy or other procedures. Later,
(23.8%) were uncomplicated and required no further however, it was used as a primary therapy, given in
treatment. Of the remaining 16 (76.1%), 3 had severe single or multiple intramuscular injections in cases
hemorrhage that necessitated hysterectomy,5,22,41 and diagnosed early.52 Our search retrieved 16 cases that
the rest required systemic methotrexate42 or laparot- were treated only by systemic intramuscular metho-
omy and excision of the mass.20,43,44 Other hemostatic trexate; in 5 (36%), all with baseline ␤-hCG level less
measures that were used included tamponade with an than 5,000 milli-International Units/mL, it led to a
intracavitary Foley catheter4,5 or cervical cerclage.45 complete and uncomplicated resolution within a few

VOL. 107, NO. 6, JUNE 2006 Rotas et al Pregnancy in Uterine Scar 1377
months.4,5,23,25 Another 5 women received multiple Sac Aspiration
doses of intramuscular methotrexate alternating with Fine-needle aspiration under sonographic guidance
leucovorin. Treatment was successful in 3 of the was attempted for small-sized cesarean scar pregnan-
women; in the remaining 2 women, it was compli- cies. Of the 5 reported cases, 2 resolved, including
cated by hemorrhage that was managed by laparot- one heterotopic in vitro fertilization pregnancy.25,60
omy and wedge resection in one and hysterectomy in The remaining 3 required additional therapy with
the other.53 In the remaining 6 women, the ␤-hCG systemic methotrexate.25,33
levels ranged from 6,000 to 48,000 milli-International
Units/mL, and additional interventions were necessary.
These included direct intragestational methotrexate in- Follow-up of Therapy and Future Fertility
jection,54 dilation and curettage,40,48 uterine artery em- Most uterine scar pregnancies managed medically
bolization,28,48,55 and Foley balloon tamponade.56 resolved within 3–9 months.3–5,25 Continuation of
cardiac activity or growth of the sac indicated failure.
All authors agreed on the protocol for posttherapy
Local Methotrexate
follow-up, which includes weekly ␤-hCG measure-
First reported by Godin et al,24 direct intragestational
ments until undetected and monthly ultrasound eval-
injection of methotrexate appeared to be effective
uations until no products of conception are visual-
because of high concentration in the sac.57 Fifteen of
ized.4 Seow et al5 suggested serial color Doppler
the 112 patients reported in the literature were man-
endovaginal ultrasound examinations to identify per-
aged initially by this method (Table 2). Eight (53.3%)
pregnancies with an initial ␤-hCG level ranging be- sistence of high velocity, low impedance, and turbu-
tween 14,086 and 93,000 milli-International lent flow that heralds risk of uterine rupture, even if
Units/mL resolved, but the process took several ␤-hCG levels decline. Chou et al29 recommended
months. No additional interventions were needed and 3-dimensional imaging with simultaneous display of
no complications ensued.4,5 The remaining 7 patients the volume of the sac and its surrounding spatial
had a persistent gestational sac or suffered massive vascular network and blood flow to quantify changes
bleeding4 and, therefore, required additional metho- in uterine neovascularization.
trexate, systemically5 or in multiple intragestational As to future fertility, information is available for
injections.58 27 patients previously treated for cesarean delivery
scar pregnancy. One case series consisted of 12
women, 7 of whom conceived and delivered 8 babies,
Combined Systemic and Local Methotrexate 4 singletons and 2 sets of twins.63 One of these patients
Five women with ␤-hCG levels from 12,000 to 46,000 was treated with dilation and curettage, which caused
milli-International Units/mL received systemic and profuse bleeding that was controlled with Foley bal-
local intragestational methotrexate simultaneous- loon tamponade. Three months later the patient
ly.5,25,59 None required any additional therapy (Table conceived, but despite monthly sonographic follow-
4). None of the known adverse effects of methotrex- up, the uterus ruptured at 38 weeks. The newborn was
ate, such as pneumonitis, alopecia, nausea, or stoma- a stillborn, and the mother died from hypovolemic
titis, occurred in these women. shock. Three other patients who received intragesta-
tional methotrexate conceived within 3 years and had
Local Embryocides an elective cesarean delivery. The fifth patient had in
Potassium chloride has been used in cases of hetero- vitro fertilization and conceived triplets. However, the
topic pregnancies in which one pregnancy was im- pregnancy implanted in the scar and therefore was
planted in a uterine scar even though extravasation to terminated with methotrexate. The same patient con-
the amniotic sac of the adjacent intrauterine gestation ceived again, this time a twin pregnancy, and deliv-
is a concern.60 It was successfully used in 3 such cases, ered by cesarean. A similar patient with heterotopic
allowing the remaining intrauterine pregnancy to pregnancy underwent aspiration of the cesarean scar
progress to term.17,61,62 Two other embryocides used pregnancy, whereas the other twin was delivered at
included hyperosmolar glucose33 and crystalline tri- 32 weeks. However, massive hemorrhage from pla-
chosanthin followed by mifepristone. The latter was centa accreta required cesarean hysterectomy. The
given to 15 patients. It failed in 9 of them and last patient in this case series received intragestational
required methotrexate and dilation and curettage. In methotrexate, conceived subsequently, and had a
3 of those patients, hysterectomy was finally neces- cesarean delivery complicated by placenta accreta
sary to control the hemorrhage.22 and disseminated intravascular coagulation. Of the

1378 Rotas et al Pregnancy in Uterine Scar OBSTETRICS & GYNECOLOGY

remaining 15 patients in the published literature, 13 repair of the defect via laparotomy or laparoscopy
carried the pregnancies to term and had elective emerges as a safe therapy, particularly in advanced
cesarean delivery, one had recurrence of the preg- pregnancies with ␤-hCG levels exceeding 15,000
nancy within the scar,58 and the last one had 2 milli-International Units/mL. Laparoscopic approach
recurrent spontaneous abortions. seems reasonable as long as the appropriate expertise
and facilities are available should rupture or massive
CONCLUSION bleeding occur.
Pregnancy in cesarean delivery uterine scar in the first Dilation and curettage was complicated by severe
trimester has been encountered more commonly over hemorrhage in 76.1% of the patients and hysterec-
the last decade, and uterine scar may no longer be the tomy in 14.2 % of them and, therefore, should not be
most infrequent site for ectopic implantation. This a first-line therapeutic option. Systemic methotrexate
increase in incidence may be attributed to the liberal may be effective for patients with ␤-hCG levels lower
use of endovaginal ultrasonography in the first trimes- than 5,000 milli-International Units/mL. For levels
ter and to the worldwide increase in the number of exceeding 5,000 milli-International Units/mL, its si-
cesarean deliveries. Endometrial and myometrial dis- multaneous employment with direct intragestational
ruption and scarring caused by the cesarean incision injection of methotrexate was effective in all 5 cases
are the main predisposing factors. From our review, it reported. The above technique may be combined
seems that the number of cesarean deliveries does not with potassium chloride injection used as an embryo-
play a role, inasmuch as more than half (52%) of the cide or with uterine artery embolization to minimize
reported cases had only one operation. A poorly hemorrhage. There are several disadvantages to these
developed low uterine segment at the time of the conservative treatments: slow decline in ␤-hCG lev-
cesarean delivery, such as in breech presentation, els, possible massive bleeding or uterine rupture, and
may predispose to incomplete healing of the scar and risk for future recurrent implantation.
to a subsequent implantation of pregnancy in it. In Patients with history of a pregnancy in a cesarean
fact, breech presentation was the most common delivery scar should be advised of the risk for future
(31.4%) indication for cesarean delivery in the cases rupture of the pregnant uterus. Uterine rupture and
we reviewed. placenta accreta are serious complications that may
Surprisingly, pain as a presenting symptom was occur even if the initial treatment was successful.63 In
not as frequent as expected, because one third of the general, the risk of uterine rupture in women with
patients who were included in this review were com- prior cesarean is 17-fold higher than in the absence of
pletely asymptomatic, and approximately 40% had a uterine scar (0.3–1.7%).64 However, after a cesarean
only painless vaginal bleeding. Sonography was the scar pregnancy, this risk is even higher because of
method used for diagnosis in these patients to ascer- thinning of the scar. Susceptibility to rupture and its
tain localization and size of the conceptus and its timing are unpredictable.5 Maymon et al16 recom-
viability.3,4,16,20 Performed in the first few weeks of mended preconceptional sonohysterography in
conception, endovaginal ultrasonography, with a sen- women with prior cesarean scar gestation to detect
sitivity of 84.6%, has dramatically reduced maternal any defect in the scar. Others recommended repair of
morbidity, enabling medical management in an in- the scar before any attempt at subsequent conception
creasing number of cases. Both 3-dimentional Dopp- or at least the use of contraception for 1–2 years.23,32
ler sonography and magnetic resonance imaging are The next pregnancy should be delivered by cesarean
adjunctive methods in management and follow-up. before the onset of labor because elasticity of the scar
The rarity of this condition explains the absence of cannot adapt to rapid uterine enlargement in late third
universal guidelines for management. Although sev- trimester.5 Careful survey for placenta accreta has also
eral interventions have been used to maintain uterine been advocated. If present, a cesarean hysterectomy at
integrity, none has been universally accepted or 32–34 weeks is recommended.5
found completely reliable.4 In summary, we reviewed all published case
In more than half of the reported cases, compli- series and case reports on pregnancy in the uterine
cations occurred, ranging from bleeding to uterine scar of a prior cesarean delivery. This review is
rupture that frequently necessitated additional ther- limited because it relies on experience from anecdotal
apy or hysterectomy. It seems that expectant manage- cases. We are aware of the fact that treatment policies
ment is not justified because rupture of the scar and should not be based on such reports. However, this
hemorrhage may occur even in the first trimester. review is important because it provides a summary of
Wedge resection of the gestation in the scar and the different therapeutic modalities, treatment fail-

VOL. 107, NO. 6, JUNE 2006 Rotas et al Pregnancy in Uterine Scar 1379
ures, and possible complications. Considering the 18. Tan G, Chong YS, Biswas A. Caesarean scar pregnancy: a
diagnosis to consider carefully in patients with risk factors. Ann
rarity of pregnancy in a cesarean delivery scar, it Acad Med Singapore 2005;34:216–9.
would be of great importance to report even individ-
19. Vial Y, Petignat P, Hohlfeld P. Pregnancy in a cesarean scar.
ual cases, particularly those with treatment failures or Ultrasound Obstet Gynecol 2000;16:592–3.
complications, so that eventually universal treatment 20. Neiger R, Weldon K, Means N. Intramural pregnancy in a
guidelines can be established. cesarean section scar: a case report. J Reprod Med 1998;43:
999–1001.
21. Einenkel J, Stumpp P, Kosling S, Horn LC, Hockel M. A
REFERENCES misdiagnosed case of caesarean scar pregnancy. Arch Gynecol
1. Hemminki E, Merilainen J. Long-term effects of cesarean Obstet 2005;271:178–81.
sections: ectopic pregnancies and placental problems. Am J 22. Weimin W, Wenqing L. Effect of early pregnancy on a
Obstet Gynecol 1996;174:1569–74. previous lower segment cesarean section scar. Int J Gynaecol
2. Chazotte C, Cohen WR. Catastrophic complications of previ- Obstet 2002;77:201–7.
ous cesarean section. Am J Obstet Gynecol 1990;163:738–42. 23. Ravhon A, Ben-Chetrit A, Rabinowitz R, Neuman M, Beller
3. Fylstra DL. Ectopic pregnancy within a cesarean scar: a U. Successful methotrexate treatment of a viable pregnancy
review. Obstet Gynecol Surv 2002;57:537–43. within a thin uterine scar. Br J Obstet Gynaecol 1997;104:
4. Jurkovic D, Hillaby K, Woelfer B, Lawrence A, Salim R, Elson 628–9.
CJ. First trimester diagnosis and management of pregnancies 24. Godin PA, Bassil S, Donnez J. An ectopic pregnancy develop-
implanted into the lower uterine segment cesarean section scar. ing in a previous caesarian section scar [published erratum
Ultrasound Obstet Gynecol 2003;21:220–7. appears in Fertil Steril 1997;68:187]. Fertil Steril 1997;67:
5. Seow KM, Huang LW, Lin YH, Lin MY, Tsai YL, Hwang JL. 398–400.
Cesarean scar pregnancy: issues in management. Ultrasound 25. Maymon R, Halperin R, Mendlovic S, Schneider D, Vankinz,
Obstet Gynecol 2004;23:247–53. Herman A, Pansky M. Ectopic pregnancies in caesarean
6. Yankowitz J, Leake J, Huggins G, Gazaway P, Gates E. section scars: the 8-year experience of one medical centre.
Cervical ectopic pregnancy: review of the literature and report Hum Reprod 2004;19:278–84.
of a case treated by single-dose methotrexate therapy. Obstet 26. Wang CJ, Yuen LT, Yen CF, Lee CL, Soong YK. Three-
Gynecol Surv 1990;45:405–14. dimensional power Doppler ultrasound diagnosis and laparo-
7. Larsen JV, Solomon MH. Pregnancy in a uterine scar sacculus: scopic management of a pregnancy in a previous cesarean scar.
an unusual cause of postabortal hemorrhage. A case report. S J Laparoendosc Adv Surg Tech A 2004;14:399–402.
Afr Med J 1978;53:142–3. 27. Shih JC. Cesarean scar pregnancy: diagnosis with three-dimen-
8. Leitch CR, Walker JJ. The rise in cesarean section rate: the sional (3D) ultrasound and 3D power Doppler. Ultrasound
same indications but a lower threshold. Obstet Gynecol Surv Obstet Gynecol 2004;23:306–7.
1999;54:19-20. 28. Imbar T, Bloom A, Ushakov F, Yagel S. Uterine artery
9. McGowan L. Intramural pregnancy. JAMA 1965;192:637–8. embolization to control hemorrhage after termination of preg-
nancy implanted in a cesarean delivery scar. J Ultrasound Med
10. Fait G, Goyert G, Sundareson A, Pickens A Jr. Intramural 2003;22:1111–5.
pregnancy with fetal survival: Case history and discussion of
etiologic factors. Obstet Gynecol 1987;70:472–4. 29. Chou MM, Hwang JI, Tseng JJ, Huang YF, Ho ES. Cesarean
scar pregnancy: quantitative assessment of uterine neovascu-
11. Rozenberg P, Goffinet F, Philippe HJ, Nisand I. Thickness of larization with 3-dimensional color power Doppler imaging
the lower uterine segment: its influence in the management of and successful treatment with uterine artery embolization. Am
patients with previous cesarean sections. Eur J Obstet Gynecol J Obstet Gynecol 2004;190:866–8.
Reprod Biol 1999;87:39–45.
30. Shufaro Y, Nadjari M. Implantation of a gestational sac in a
12. Chen HY, Chen SJ, Hsieh FJ. Observation of cesarean section cesarean section scar. Fertil Steril 2001;75:1217.
scar by transvaginal ultrasonography. Ultrasound Med Biol
1990;16:443–7. 31. Marcus S, Cheng E, Goff B. Extrauterine pregnancy resulting
from early uterine rupture. Obstet Gynecol 1999;94:804-5.
13. Armstrong V, Hansen WF, Van Voorhis BJ, Syrop CH.
Detection of cesarean scars by transvaginal ultrasound. Obstet 32. Valley MT, Pierce JG, Daniel TB, Kaunitz AM. Cesarean scar
Gynecol 2003;101:61–5. pregnancy: imaging and treatment with conservative surgery.
Obstet Gynecol 1998;91:838–40.
14. Monteagudo A, Carreno C, Timor-Tritsch IE. Saline infusion
sonohysterography in nonpregnant women with previous 33. Roberts H, Kohlenber C, Lanzarone V, Murray H. Ectopic
cesarean delivery: the “niche” in the scar. J Ultrasound Med pregnancy in lower segment uterine scar. Aust N Z J Obstet
2001;20:1105–15. Gynaecol 1998;38:114–6.
15. Chuang J, Seow KM, Cheng WC, Tsai YL, Hwang JL. 34. Herman A, Weinraub Z, Avrech O, Maymon R, Ron-El R,
Conservative treatment of ectopic pregnancy in a caesarean Bukovsky Y. Follow up and outcome of isthmic pregnancy
section scar. BJOG 2003;110:869–70. located in a previous caesarean section scar. Br J Obstet
16. Maymon R, Halperin R, Mendlovic S, Schneider D, Herman Gynaecol 1995;102:839–41.
A. Ectopic pregnancies in a caesarean scar: review of the 35. Fylstra DL, Pound-Chang T, Miller MG, et al. Ectopic preg-
medical approach to an iatrogenic complication. Hum Reprod nancy within a cesarean delivery scar: a case report. Am J
Update 2004;10:515–23. Obstet Gynecol 2002;187:302–4.
17. Hartung J, Meckies J. Management of a case of uterine scar 36. Seow KM, Hwang JL, Tsai YL. Ultrasound diagnosis of a
pregnancy by transabdominal potassium chloride injection. pregnancy in a cesarean section scar. Ultrasound Obstet
Ultrasound Obstet Gynecol 2003;21:94–5. Gynecol 2001;18:547–9.

1380 Rotas et al Pregnancy in Uterine Scar OBSTETRICS & GYNECOLOGY

37. Lee CL, Wang CJ, Chao A, Yen CH, Soong YK. Laparoscopic 52. Haimov-Kochman R, Sciaky-Tamir Y, Yanai N, Yagel S.
management of an ectopic pregnancy in a previous caesarean Conservative management of two ectopic pregnancies
section scar. Hum Reprod 1999;14:1234–6. implanted in previous uterine scars. Ultrasound Obstet
38. Wang YL, Su TH, Chen HS. Laparoscopic management of an Gynecol 2002;19:616–9.
ectopic pregnancy in a lower segment cesarean section scar: a 53. Lam PM, Lo KW, Lau TK. Unsuccessful medical treatment of
review and case report. J Minim Invasive Gynecol 2005;12: cesarean scar ectopic pregnancy with systemic methotrexate: a
73–9. report of two cases. Acta Obstet Gynecol Scand 2004;83:
39. Wang CJ, Yuen LT, Chao AS, Lee CL, Yen CH, Soong YK. 108–11.
Caesarean scar pregnancy successfully treated by operative 54. Hwu YM, Hsu CY, Yang HY. Conservative treatment of
hysteroscopy and suction curettage. BJOG 2005;112:839–40. caesarean scar pregnancy with transvaginal needle aspiration
40. Graesslin O, Dedecker F Jr,Quereux C, Gabriel R. Conserva- of the embryo. BJOG 2005;112:841–2.
tive treatment of ectopic pregnancy in a cesarean scar. Obstet 55. Marchiole P, Gorlero F, de Caro G, Podesta M, Valenzano M.
Gynecol 2005;105:869–71. Intramural pregnancy embedded in a previous Cesarean sec-
41. Liang HS, Jeng CJ, Sheen TC, Lee FK, Yang YC, Tzeng CR. tion scar treated conservatively. Ultrasound Obstet Gynecol
First-trimester uterine rupture from a placenta percreta: a case 2004;23:307–9.
report. J Reprod Med 2003;48:474–8. 56. Chuang J, Seow KM, Cheng WC, Tsai YL, Hwang JL.
42. Padovan P, Lauri F, Marchetti M. Intrauterine ectopic preg- Conservative treatment of ectopic pregnancy in a caesarean
nancy: a case report. Clin Exp Obstet Gynecol 1998;25:79–80. section scar. BJOG 2003;110:869–70.
43. Li SP, Wang W, Tang XL, Wang Y. Cesarean scar pregnancy: 57. Schiff E, Shalev E, Bustan M, Tsabari A, Mashiach S, Weiner
a case report. Chin Med J 2004;117:316–7. E. Pharmacokinetics of methotrexate after local tubal injection
44. Yang MJ, Jeng MH. Combination of transarterial embolization for conservative treatment of ectopic pregnancy. Fertil Steril
of uterine arteries and conservative surgical treatment for 1992;57:688–90.
pregnancy in a cesarean section scar: a report of 3 cases. 58. Hasewaga J, Ichizuka K, Matsuoka R, Otsuki K, Sekizawa A,
J Reprod Med 2003;48:213–6. Okai T. Limitations of conservative treatment for repeat
45. Reyftmann L, Vernhet H, Boulot P. Management of massive cesarean scar pregnancy. Ultrasound Obstet Gynecol 2005;25:
uterine bleeding in a cesarean scar pregnancy. Int J Gynaecol 310–1.
Obstet 2005;89:154–5. 59. Nawroth F, Foth D, Wilhelm L, Schmidt T, Warm M, Romer
46. Lobel SM, Meyerovitz MF, Benson CC, Goff B, Bengtson JM. T. Conservative treatment of ectopic pregnancy in a cesarean
Preoperative angiographic uterine artery embolization in the section scar with methotrexate: a case report. Eur J Obstet
management of cervical pregnancy. Obstet Gynecol 1990;76: Gynecol Reprod Biol 2001;99:135–7.
938–41. 60. Hsieh BC, Hwang Jl, Pan HS, Huang SC, Chen CY, Chen PH.
47. Saliken JC, Normore WJ, Pattinson HA, Wood S. Emboliza- Heterotopic caesarean scar pregnancy combined with intra-
tion of the uterine arteries before termination of a 15-week uterine pregnancy successfully treated with embryo aspiration
cervical pregnancy. Can Assoc Radiol J 1994;45:399–401. for selective embryo reduction: case report. Hum Reprod
2004;19:285–7.
48. Ghezzi F, Lagana D, Franchi M, Fugazzola C, Bolis P. Con-
servative treatment by chemotherapy and uterine arteries 61. Salomon LJ, Fernandez H, Chauveaud A, Doumerc S, Fryd-
embolization of a cesarean scar pregnancy. Eur J Obstet man R. . Successful management of a heterotopic Caesarean
Gynecol Reprod Biol 2002;103:88–91. scar pregnancy: potassium chloride injection with preservation
of the intrauterine gestation: case report. Hum Reprod 2003;
49. Sugawara J, Senoo M, Chisaka H, Yaegashi N, Okamura K.
18:189–91.
Successful conservative treatment of a cesarean scar pregnancy
with uterine artery embolization. Tohoku J Exp Med 2005; 62. Yazicioglou PM, Turgut S, Madazli R, Aygun M, Cebi Z,
206:261–5. Sonmez S. An unusual case of heterotopic twin pregnancy
managed successfully with selective feticide. Ultrasound
50. Chou MM, Hwang JI, Tseng JJ, Huang YF, Ho ES. Cesarean
Obstet Gynecol 2004;23:626–7.
scar pregnancy: quantitative assessment of uterine neovascu-
larization with 3-dimensional color power Doppler imaging 63. Seow KM, Hwang Jl, Tsai YL, Huang LW, Lin YH, Hsieh BC.
and successful treatment with uterine artery embolization. Subsequent pregnancy outcome after conservative treatment
Am J Obstet Gynecol 2004;190:866–8. of a previous cesarean scar pregnancy. Acta Obstet Gynecol
51. Hung TH, Shau WY, Hsieh TT, Hsu JJ, Soong YK, Jeng CJ. Scand 2004;83:1167–72.
Prognostic factors for an unsatisfactory primary methotrexate 64. Gregory KD, Korst LM, Cane P, Platt LD, Kahn K. Vaginal
treatment of cervical pregnancy: a quantitative review. Hum birth after cesarean and uterine rupture rates in California.
Reprod 1998;13:2636–42. Obstet Gynecol 1999;94:985–9.

VOL. 107, NO. 6, JUNE 2006 Rotas et al Pregnancy in Uterine Scar 1381