Antenatal diagnosis of non‐previa placenta increta with histological confirmation

A 36-year-old woman, gravida 4 para 3, was referred to our unit at 20 weeks' gestation because of a complicated obstetric history. She had had three previous vaginal deliveries, all of which had been associated with primary postpartum hemorrhage. In the second and third deliveries, the placenta was noted to be adherent at the third stage of labor. Manual removal of the placenta resulted in profuse hemorrhage, requiring massive blood transfusion and intensive care admission. In the third pregnancy, partial/focal placenta accreta was confirmed on histopathological examination of the placenta. On ultrasound examination in the current pregnancy (ATL HDI 5000, Philips, Bothell, WA, USA), the placenta was noted to be bilobed, with one lobe located posteriorly and the other anteriorly, and with the cord insertion between the lobes. Placental lacunae were not seen in the anterior lobe. However, tongues of placental tissue could be seen interrupting the placental–uterine wall interface in the anterior, lateral and posterior walls of the uterus (Figures 1 and 2a). Communicating vessels were seen running directly between the placental substance and the uterine musculature at these interface-interruption sites on Doppler examination of both posterior (Figure 2a) and anterior (Figure 2b) walls. Placenta increta was diagnosed. On gray-scale ultrasound, the placental–uterine wall interface is interrupted by tongues (arrows) of placental (PL) tissue in the anterior (a) and lateral (b) uterine (UT) walls. L, left; LAT, lateral. On color Doppler imaging, communicating blood vessels can be seen at the interface-interruption sites between the placenta (PL) and the uterine (UT) wall posteriorly (a) and anteriorly (b). Since the couple had completed their family, an elective Cesarean hysterectomy was planned and this was performed uneventfully at 37 weeks' gestation. The placenta was bilobed and adherent to the uterine wall both anteriorly and posteriorly, with histological evidence of increta. Abnormal placentation varies in the degree of myometrial invasion from placenta accreta, to increta and percreta. Detection rates recorded in the literature of antenatal sonographic detection of placenta accreta vary, ranging from 33% (4/12)1 to 100% (5/5)2. The diagnostic criteria suggested in the literature for the sonographic diagnosis of placenta accreta, increta or percreta include loss of hypoechoic retroplacental myometrial zone, presence of placental lacunae, disruption of the posterior bladder wall–uterine interface and, in the case of placenta percreta3, presence of focal exophytic masses. The sensitivity of visualization of placental lacunae, absence of the retroplacental clear zone, and interruption of the posterior bladder–uterine wall interface in the detection of placenta accreta, respectively, at the second- or third-trimester ultrasound scans is reported to be 93%, 73% and 20% at a false positive rate of 54.4% in the second trimester4. With the use of color and power Doppler, the sensitivity and specificity of antenatal diagnosis range from 82.4% to 100%, and 92% to 96.8%, respectively2, 5, 6. Findings on Doppler imaging include a pattern of turbulent blood flow extending from the placenta into the surrounding tissue2, vessels crossing the placenta–bladder interface5, placental lacunar flow, interface hypervascularity and subplacental vascularity6. Magnetic resonance imaging appears to add information only when the placenta is not well depicted on ultrasound, for example in posterior placenta accreta5. In a previous report of non-previa placenta accreta, the diagnosis was based on the absence of a decidual interface and unusual dilatation of uterine vessels under the placental implantation site. It was also noted that the basal plate of the placenta appeared to float inside the uterine vessels7. It is known that placental vascular growth occurs as early as 21 days postconception and continues throughout pregnancy. The uterine vasculature undergoes three main adaptative changes to cope with the increased demand on blood supply: vasodilatation, increased permeability, and growth and development of new vessels8. Moreover, maternal arterial blood entering the placenta ceases to be enclosed within proper arteries. The smooth muscle is lost and the endothelial lining of the maternal blood spaces is replaced by giant trophoblastic cells9. The trophoblasts produce a number of angiogenic and vasoactive substances to regulate placental vessel formation and maternal vascular adaptation to pregnancy. One of these is human chorionic gonadotropin8, which has an invasion-stimulating action on the trophoblastic cells and a direct angiogenic effect. Invasion of uterine musculature by the placental tissue would be expected to be accompanied by vessel and sinus formation (due to loss of the smooth muscle lining of vessels) and increased vascularity of the placental implantation site in the uterine wall. All these could explain the gray-scale and Doppler findings reported in association with placenta accreta2-6: the placental lacunae or sinuses and the turbulent blood flow pattern within, and the hypervascularity of the tissue interface and the placental implantation site. These changes have been reported even in the first trimester10, 11 and, though they may be exaggerated in placenta accreta or placenta percreta4, they are adaptive changes that are not specific to such conditions. In a normal pregnancy, there is a separation between the trophoblastic (or placental) vessels and the vessels in the uterine musculature because of the presence of the decidua basalis. However, placenta accreta is characterized by the absence of decidua basalis, either as a cause or as a consequence of trophoblastic invasion12. Blood vessels can therefore cross the placental–uterine interface (or uterine–bladder or other tissue interface, in the case of placenta percreta5, 13) along the invading tongues of trophoblastic cells, as in this case. In the present case, there were interruptions to the placental–uterine interface in both anterior (Figure 1) and posterior (Figure 2a) walls of the uterus, consistent with the diagnosis of placenta accreta. The additional finding of vessels crossing the placental–uterine wall interface essentially completed the diagnosis (Figure 2). These vessels being lined with trophoblastic cells instead of endothelium and smooth muscle could also explain the severe bleeding often observed following attempts at placental separation when these vessels are torn, and the better maternal outcome when placenta accreta is treated conservatively13, 14. However, when placenta accreta is managed conservatively without histological confirmation, the case is ‘unproven’15. Specific ultrasound diagnostic criteria correlating with the pathology may be required to avoid false-positive and the false-negative diagnoses. In conclusion, both gray-scale and Doppler ultrasound imaging are valuable tools for the antenatal diagnosis of placenta accreta. In addition to the specific sign of tissue interface interruption, the finding of abnormal communicating vessels crossing the tissue interface on Doppler examination may be a useful sign when placenta accreta/increta/percreta is suspected. It may be particularly useful in non-previa placenta accreta and when the placenta lies in the posterior or lateral wall of the uterus, as in such cases gray-scale imaging may not be able to demonstrate the interruption of the tissue interface as well as it can when the placenta lies in the anterior wall.