Management of Placenta Accreta in Pregnancy with Placenta Previa

Placenta accreta is a life-threatening obstetrical condition. Prenatal prediction of placenta accreta helps to minimize clinical complications. Placenta previa is one of the most important factors associated with placenta accreta. In our prospective cohort study, ultrasound finding of loss of the retroplacental hypoechoic clear zone was found to be a single predictor of placenta accreta in women with placenta previa (odds ratio, 15.6; 95% confidence interval, 2.1–114.6; p < 0.01). In addition, we have devised a novel scoring system for predicting placenta accreta in pregnancy with placenta previa, yielding 91.3% sensitivity, 98.0% specificity, 87.5% positive predictive value, and 98.7% negative predictive value. Planned preterm cesarean hysterectomy with the placenta left in situ is generally recommended for women with suspicion of placenta accreta. If the women have a desire for future fertility, conservative approach may be considered.


Introduction
Placenta accreta is a life-threatening obstetrical condition. Clinical complications of placenta accreta involve massive hemorrhage, damage to adjacent organs, cesarean hysterectomy, and maternal death. Placenta previa is one of the most important risk factors for placenta accreta [1]. Prenatal prediction of placenta accreta helps to minimize clinical complications by enabling obstetricians to plan for resources that may be required during cesarean delivery, including obstetric anesthesia, appropriate surgical expertise, available blood products, and interventional radiology for uterine artery embolization [2,3]. Therefore, accurate prenatal prediction of placenta accreta in women with placenta previa is important. This review will focus on the recent knowledge regarding the prenatal diagnosis and management of placenta accreta in women with placenta previa.

Ultrasonography
It has been reported that ultrasonography is the most useful method for diagnosing placenta accreta. Previous studies have demonstrated that ultrasound (US) findings involving the presence of placental lacunae (PL) [4], an anterior myometrial thickening [5], loss of the retroplacental hypoechoic clear zone (LCZ) [6], anomalies of the bladder-myometrium interface [7], and the presence of turbulent blood flow (TBF) in the arteries radiating from the placenta toward the uterine serosa, detected by color Doppler [8] are associated with placenta accreta. However, these previous results were based on retrospective studies for pregnant women with and without placenta previa.
Recently, we demonstrated that a novel US finding referred to as an "irregular sign" might be useful for predicting placenta accreta in women with placenta previa [11]. The irregular sign is defined as the irregularity of the border between the placenta and the myometrium around internal uterine os observed by transvaginal ultrasonography (Figure 1). The presence of irregular sign yielded 56.5% sensitivity, 99.3% specificity, 92.9% PPV, and 93.8% NPV for predicting placenta accreta in women with placenta previa [11].

Magnetic resonance imaging
Some investigators suggested that MRI was a useful tool for diagnosing placenta accreta prenatally [12,13]. MRI findings suggestive of placenta accreta included indistinctness or the absence of myometrial wall at the placental site, loss of the thin T2 dark uteroplacental interface, a nodular interface between the placenta and the uterus, a mass effect of the placenta on the uterus causing uterine outer bulge, heterogeneous signal intensity within the placenta, dark intraplacental bands on T2-weighted images, and abnormal dilated venous lakes within the placenta [13][14][15]. MRI provided advantages in diagnosing placenta accreta in women who had posterior placental location [16]. We found that MRI was selected as a significant finding in predicting placenta accreta in pregnant women with placenta previa by univariate analyses in the prospective study (OR, 24.6; 95% CI, 4.7-129.2; p < 0.01) [9].
However, other investigators claimed that MRI was more expensive and invasive examination than US; therefore, US should be the diagnostic modality of first choice for placenta accreta, and MRI was not always necessary [17]. Dynamic contrast-enhanced MRI with gadolinium (Gd) is not recommended at any time during pregnancy, because the use of Gd increases the risk of a broad set of rheumatological, inflammatory, or infiltrative skin conditions and of stillbirth or neonatal death [18].

Diagnostic scoring systems
Diagnostic scoring systems may be more useful than a single US finding for predicting placenta accreta in women with a history of previous CS, placenta previa, or low-lying placenta [19][20][21]. These scoring systems include several US findings suggestive of placenta accreta, the presence of placenta previa, and the number of prior CS. Optimal cut-off values of these scores determined in the retrospective or prospective studies yielded 72.0-94.2% sensitivity, 52.5-85.0% specificity, 63.4-70.0% PPV, and 86.0-100% NPV for the prediction of placenta accreta [19][20][21]. We devised a novel scoring system for predicting placenta accreta in women with placenta previa, and in a prospective cohort study evaluated the diagnostic efficacy of this scoring system named the placenta previa with adherent placenta (PPAP) score [11]. The PPAP score is composed of two categories: (1) past history of CS, surgical abortion, and/or uterine surgery and (2) US and MRI findings. Each category is graded as 0, 1, 2, or 4 points, yielding a total score between 0 and 24 ( Table 1).
Women with placenta previa who had PPAP score ≥ 8 were considered to be at a high risk for placenta accreta. The PPAP score yielded 91.3% sensitivity, 98.0% specificity, 87.5% PPV, and 98.7% NPV for predicting placenta accreta in women with placenta previa [11]. However, the PPAP scoring failed to predict two women with placenta accreta. One woman had a history of myomectomy and PPAP score of 6 and another had a past history of myometrium resection for adenomyosis and PPAP score of 4. The presence of a history of myomectomy and myometrium resection for adenomyosis seems to increase the risk for placenta accreta in women who had PPAP score < 8. Special attention should be paid to the risk of placenta accreta regardless of US or MRI findings.

The management of placenta accreta in women with placenta previa
All pregnant women with placenta previa suspected to have placenta accreta should be managed at specialized tertiary centers [22]. Their deliveries should be performed by an experienced medical team consisted of obstetric surgeons, urologists, general surgeons, and gynecologic oncologists [23]. Planned cesarean hysterectomy decreases the morbidity and mortality rates in women complicated by placenta accreta [3]. The timing of delivery in such women must be individualized; however, a recent study suggested that delivery at 34 weeks of gestation (GW) in stable women with placenta accreta optimized the outcomes of both mothers and neonates [24].
The anesthesiologists should assess which anesthetic techniques are used before delivery. Both general and regional anesthetic techniques are available, and the decision of which type of technique to be used should be made on an individual basis [23]. Preoperative cystoscopy with placement of ureteral stents may help prevent accidental urinary tract injury. In addition, sufficient amount of blood products should be available in the operating room.
Planned preterm cesarean hysterectomy with the placenta left in situ is generally recommended for women with suspicion of placenta accreta, because forced removal of the placenta causes massive hemorrhage. Midline vertical incision may be considered because it provides sufficient exposure if hysterectomy is needed. CS using transverse uterine fundal incision method is often used to avoid the placenta and allow delivery of the infant.
Hysterectomy is performed in the usual fashion. In some cases with anterior placenta accreta, especially in cases with placenta percreta, partial resection of the bladder wall is necessary. On the other hand, if the women have a strong desire for future fertility, conservative approach, i.e. leaving the placenta in situ, may be considered. However, a review, which summarized the conservative management of 60 women with placenta accreta, showed that infection occurred in 11 of the 60 women (18%), bleeding in 21 (35%), and disseminated intravascular coagulation in 4 (7%) [25]. Therefore, this conservative approach should be considered only when women are willing to accept the risks involved in this approach.
There has been lack of sufficient evidence for beneficial effects of prophylactic catheter placement for balloon occlusion or artery embolization [26][27][28] as well as treatment with methotrexate [29][30][31] . Therefore, a firm recommendation on the use of these procedures cannot be made [23,31]. It is difficult to establish evidence-based management strategies for placenta accreta in pregnancy with placenta previa. Therefore, clinicians should manage these women by suitable approaches in each medical institution. Figure 2 shows an algorithm used in the Kobe University Hospital for the management of pregnant women with placenta previa. All pregnant women with placenta previa receive workup for placenta accreta in inpatient or outpatient care. If women have bleeding, they are hospitalized immediately and receive intravenous administration of tocolytic agents such as magnesium sulfate or β-stimulant. Even if they do not have bleeding, they are hospitalized at 32-34GW. Women have a PPAP score ≥ 8, who are suspected of having placenta accreta, receive both preoperative internal iliac artery occlusion balloon catheters placement and elective CS at 35-37GW. Figure 3 shows a flow algorithm used in the Kobe University Hospital for the preoperative preparation and operative procedures for women with suspicion of placenta accreta. Women with placenta previa had a PPAP score ≥ 8, who were suspected of having placenta accreta, received preoperative internal iliac artery occlusion balloon catheters placement. After fetal delivery by a CS using transverse uterine fundal incision method, the internal iliac artery occlusion balloon catheters were inflated. After occlusion of the artery, local injection of oxytocin into the myometrium and uterine massage were performed to induce spontaneous placental separation. If placental separation did not occur at all and women did not have a desire for future fertility, cesarean hysterectomy was performed. When the placenta was not partially separated, partial resection of uterine wall or removal of placenta using advanced bipolar was performed.