The term placenta accreta is used to describe any placental implantation in which there is abnormally firm adherence to the uterine wall. As a consequence of partial or total absence of decidua basalis and imperfect development of the fibrinoid layer, placental villi are attached to the myometrium in placenta accreta, actually invade the myometrium in increta, or penetrate the myometrium in placenta percreta. Although an uncommon condition, abnormal placentation assumes considerable significance clinically because of morbidity and, at times, mortality from severe haemorrhage, uterine perforation and infection (1).

Spontaneous uterine rupture early in course of pregnancy is a very rare complication, more so, in case of patients treated by invitro fertilization (IVF). High risk factors are previously scarred uterus, dilatation and curettage, advanced maternal age and abnormal placentation (2). We present here an interesting case of this potentially life-threatening complication in an IVF conceived patient where placenta accreta led to uterine rupture in second trimester necessitating hysterectomy due to uncontrolled hemorrhage.

A 40-year old infertile woman was referred to our center for IVF program (oocyte donation and embryo transfer). She had two previous histories of dilatation and curettage, one after spontaneous abortion and another for diagnostic purposes.

Hysteroscopy was done prior to IVF cycle which revealed a band of adhesion running from anterior to posterior wall for which adhesiolysis was done in the same session. The patient conceived in the second cycle of embryo transfer (oocyte donation). Pregnancy was confirmed by beta-human chorionic gonandotrophin (β–hCG) levels and later on, by ultrasound which confirmed single intrauterine gestational sac with a beating heart. She was followed up regularly for her obstetric examination with no other risk factors except for advanced maternal age.

Level two obstetric ultrasound examination was done at eighteen weeks of gestation which ruled out gross congenital anomalies and suggested grade 0 left fundo-anterior placenta with normal invasion and normal amniotic fluid. In the nineteenth weeks of gestation, patient presented in the emergency department of our hospital with acute abdominal pain and signs of early shock (pulse rate 180/min, blood pressure 90/60 mmHg, hemoglobin 7.5 g% and hematocrit of 24%).

Initially, the patient was revived from shock with adequate fluid replacement and other basic management procedures including central line placement for cental venous pressure (CVP) monitoring. Once the patient's condition was stable, she was sent for ultrasound and CT scan evaluations. Ultrasound examination revealed fresh intrauterine fetal death along with hemoperitoneum with no other gross finding.

CT scan was done to rule out other causes of heamoperitoneum. Findings were suggestive of weakening of the uterine musculature at the left fundal region with fetus inside the cavity. Patient was taken up for emergency laprotomy and a heamoperitoneum of 3000 ml was found along with a dead fetus coming out of a large uterine fundal defect. Female fetus of approximately 700 g was extracted but the placental tissue could not be separated from the myometrium. Subtotal hysterectomy was performed in view of massive haemorrhage, while all other intraperitoneal organs were found to be intact. The patient was transfused one unit of packed cell intraoperatively and two units thereafter. She was intensively monitored postoperatively and was discharged on the sixth post-operative day in a perfectly stable condition.

Uterus (Figure 1) sent for biopsy and pathology report showed chorionic villi reaching the myometrium with absent decidua basilis which were consistent with the diagnosis of placenta accreta (Figure 2).

Review of the literature suggested incidence of abnormal placentation, including placenta percreta, to be varying between 1 in 540 and 1 in 93000 with an average of 1 in 700. Recently, incidence of placenta accreta is on rise due to increased number of caesarean section in modern obstetrics (1).

Spontaneous rupture of the uterus due to placenta percreta is one of the most urgent obstetrical complications resulting in rapid exsanguination and high mortality which is more commonly seen in the third trimester and is very rare in second trimester (3). It is rarely recognized intrapartum and is very difficult to diagnose.

Suspicion arises in case of unexplained elevation of alpha-fetoprotein (AFP) (4). Ultrasound (findings in percreta) and MRI can give more details leading to further suspicion of such an abnormal placenta. We could not perform MRI and AFP in our patient as placental pathology was not suspected for the disease in the first place.

The precise etiology of all cases of placenta accreta is unknown, however, there are known factors that increase the risk. Greatest among them are scarring of endometrial cavity with previous caesarean section, uterine curettage, myomectomy, Asherman's syndrome, iatrogenic uterine perforation and advanced maternal age. These risk factors are frequently seen in IVF/embryo transfer candidates.

In most cases, placenta accreta is caused by a combination of factors and its occurrence is it unlikely to be attributed to a single factor.

In contrast to many reports, in which a placenta percreta was diagnosed in patients with previously scarred uterus (5, 6), our patient had no history of previous caesarean section. However, less significant factors like advanced maternal age and history of curettage may have been the cause in this patient. Spontaneous uterine rupture in early gestation is extremely rare. Cases of spontaneous uterine rupture as a result of placenta percreta during early pregnancy have been reported during 10^th to 20^th week of gestation, but amongst them only few involved healthy patients with nonscarred uterus (2, 7).

Important differtial diagnoses in suspected cases when placenta could not be separated postpartum include retained incarcerated placenta, adherent placenta where there is a line of cleavage and choriocarcinoma. Histopathology has been found to be the gold standard of making the final diagnosis (4, 7).

Therefore, it may be concluded that one has to be very careful and keep the possibility of placenta accreta while dealing with IVF conceived pregnancies as the incidence of above discussed risk factors in most of infertile patients is high. Ultrasonography and colour Doppler study for the diagnosis of site of placentation along with maternal serum alpha fetoprotein measurement in doubtful cases of placenta accreta seems to benecessary as early diagnosis renders conservative treatment possible.

Informed Consent: Informed consent of the patient was obtained beforehand.