|Year : 2019 | Volume
| Issue : 4 | Page : 1359-1364
Evaluation of placenta accreta index to predict placental invasion in patients with placenta previa
Hamed ElLakwa1, Zakaria Sanad1, Haitham A Hamza1, Nareman Elhamamy2, Mohamed El-Sheikh3, Amr Samak4
1 Department of Gynecology and Obstetrics, Faculty of Medicine, Menoufia University, Menoufia, Egypt
2 Department of Gynecology and Obstetrics, Faculty of Medicine, Tanta University, Menoufia, Egypt
3 Department of Gynecology and Obstetrics, Shebin Elkom Teaching Hospital, Menoufia, Egypt
4 Department of General Surgery, El-Mokatem Health Insurance Hospital, Cairo, Egypt
|Date of Submission||12-Jun-2019|
|Date of Decision||03-Aug-2019|
|Date of Acceptance||21-Aug-2019|
|Date of Web Publication||31-Dec-2019|
Department of Gynecology and Obstetrics, Shebin Elkom Teaching Hospital, Shebin Elkom, Menoufia 32500
Source of Support: None, Conflict of Interest: None
To evaluate the role of placenta accreta index in predicting placental invasion in patients with placenta previa.
Placenta accreta placenta is a condition in which placenta is adherent to the uterine wall because of myometrial invasion.
Patients and methods
In this prospective study, 60 pregnant women with placenta previa were enrolled at the Obstetrics Clinics of Menoufia University Hospitals in the period from January 2016 to December 2017. The placenta was scanned using gray-scale ultrasound and color Doppler. Placenta accreta index was formulated for each patient. Definitive diagnosis was made at delivery when the myometrium invasion by the placenta was seen. The hysterectomy specimens were sent for pathological confirmation of placental invasion. Data were collected and entered using Statistical Packages for the Social Science program for statistical analysis.
Nineteen patients were found to have placenta accreta. All the parameters of placenta accreta index were statistically significant. The total score was significantly higher in accreta patients (P < 0.001). Receiver operating characteristic curve showed that score 4 was the best cut-off point to diagnose placenta accreta. The number of patients that had index above 4 was 18 patients. Seventeen patients of them were proved to have placenta accreta (P < 0.001).
Placenta accreta index score 4 can be used as a cut-off value for prediction of placental invasion in patients with placenta previa.
Keywords: morbidly adherent placenta, placenta accrete, placenta accreta index, placenta previa, placental invasion
|How to cite this article:|
ElLakwa H, Sanad Z, Hamza HA, Elhamamy N, El-Sheikh M, Samak A. Evaluation of placenta accreta index to predict placental invasion in patients with placenta previa. Menoufia Med J 2019;32:1359-64
|How to cite this URL:|
ElLakwa H, Sanad Z, Hamza HA, Elhamamy N, El-Sheikh M, Samak A. Evaluation of placenta accreta index to predict placental invasion in patients with placenta previa. Menoufia Med J [serial online] 2019 [cited 2020 Jun 6];32:1359-64. Available from: http://www.mmj.eg.net/text.asp?2019/32/4/1359/274230
| Introduction|| |
Morbidly adherent placenta is a condition in which all or part of placenta is adherent to the uterine wall because of myometrial invasion . Morbidly adherent placenta includes placenta accreta, increta, and percreta as it penetrates through the decidua basalis then through the myometrium. For ease of description the term accreta is used for all these conditions .
A high presenting part or an abnormal lie, painless and unprovoked bleeding should raise the suspicion of placenta previa irrespective of previous imaging results however, this condition is often diagnosed during cesarean section (CS), upon placental removal with unfavorable maternal outcome . Attempts to remove the placenta can cause severe uterine bleeding. An accurate prenatal diagnosis is required to reduce the risk of maternal/fetal morbidity and mortality . Antenatal diagnosis of placental invasion has the potential to improve maternal and fetal outcomes .
Predelivery knowledge of morbidly adherent placenta allows for multidisciplinary planning and delivery before the onset of labor and/or vaginal bleeding . This knowledge has lowered overall maternal morbidity rates, including less blood loss, as well as fewer transfusion requirements and intraoperative urologic injuries .
Sonography with gray scale and color Doppler imaging is the recommended first-line modality for diagnosing morbidly adherent placenta. Myometrial involvement greater than 1 mm with large placental lakes on Doppler ultrasound predicts myometrial invasion . The diagnosis of morbidly adherent placenta by ultrasound involves a number of different variables. These markers include an inability to visualize the normal retroplacental clear zone, irregularity and attenuation of the uterine-bladder interface, retroplacental myometrial thickness, presence of intraplacental lacunar spaces and bridging vessels between the placenta and bladder wall when using color Doppler .
Placenta accreta index consisted of group of variable parameters found to offer significant improvement in prediction of morbidly adherent placenta. Each parameter was weighted to create a nine-point scale in which a score of 0–9 provided a probability of invasion that ranged from 2 to 96%, respectively .
The aim of this study was to evaluate the role of placenta accreta index in predicting placental invasion in patients with placenta previa.
| Patients and Methods|| |
With a margin error of 5%, confidence interval of 80%, population size of 20 000 and response distribution of 10%, 60 female patients with present history of placenta previa were included in our study from the Obstetrics and Gynecology Clinic of Menoufia University Hospitals and Shebin Elkom teaching hospital after 34 gestational weeks [Figure 1]. Patients with severe bleeding were excluded from the study.
After the approval of ethics committee and informed consent from all patients included in the study, patients were assessed by full history taking including age of the patient, obstetrics history as number of pregnancies, CS, etc. Full clinical examinations were performed on the patients.
For each patient, the whole placenta was scanned using both gray-scale ultrasound and color flow mapping. Transabdominal ultrasound examination was performed after 34 weeks with the bladder partially filled which allows optimal visualization of the uterine serosa and the bladder wall. Ultrasound detects fetal viability, fetal maturity, fetal age, and location of the placenta. In placenta accreta patients, ultrasound may show complete loss of the retroplacental sonolucent zone, irregular retroplacental sonolucent zone, thinning or disruption of the hyperechoic uterine serosa–bladder interface, the presence of focal exophytic masses invading the urinary bladder and/or the presence of abnormal placental lacunae. Abnormal placental lacunae were graded into; grade 0 if not seen, grade 1 if present and generally small, grade 2 if present and tending to be larger and more irregular, and grade 3 if many throughout the placenta and appearing large and bizarre .
Doppler was used to assess abnormal vasculature. The diagnosis of placenta accreta was regarded positive when any one of these color Doppler criteria was present; diffuse or focal lacunar flow pattern, sonolucent vascular lakes with turbulent flow, hypervascularity of the uterine bladder interface and/or markedly dilated vessels over the peripheral subplacental region.
Placenta accreta index used the radiological findings to offer significant improvement in prediction of placenta accreta [Table 1]. Each parameter was weighted to create a nine-point scale in which a score of 0–9 provided a probability of invasion that ranged from 2 to 96%, respectively .
All the pregnancies enrolled in this study were delivered by CS after 34 weeks and before 37 weeks. The hysterectomy specimens were sent for pathological confirmation of the presence of morbidly adherent placenta.
Data were collected and entered to the computer using Statistical Package for the Social Science program, version 13.0 (SPSS Inc., Chicago, Illinois, USA) for statistical analysis. Data entered as numerical or categorical, as appropriate. Quantitative data were shown as mean, SD, and range. Qualitative data were expressed as frequency and percentage. Statistical analysis was done using χ2 test and χ2 goodness-of-fit test for qualitative data and two-sample t test for quantitative data. The cut-off value of the scoring system and the corresponding sensitivity, specificity were calculated. P value was considered to be of statistical significance if it was less than 0.05.
| Results|| |
Placenta accreta was found in 19 (31.67%) patients during the time of CS. Patients with placenta accreta were significantly older than nonaccreta placenta previa patients (P = 0.005). Number of pregnancies was significantly higher in accreta than nonaccreta patients (P < 0.001). Also the decreased number of labors was associated with decreased incidence of placenta accreta (P < 0.001). There were no significant differences between accreta and nonaccreta patients in terms of number of abortions, the previous history of antepartum hemorrhage and the previous history of normal vaginal delivery. The number of previous CS was significantly higher in nonaccreta patients (P < 0.001) while the number of previous D&C was significantly higher in accreta patients (P = 0.001). The previous history of placenta previa was significantly higher in accreta patients (P < 0.001) [Table 2].
There was no significant difference between accreta and nonaccreta patients in terms of time of CS. The placenta previa type was centralis in all accreta patients while in nonaccreta patients, it was lateralis in 14 patients, marginalis in 10 patients and centralis in 17 patients (P < 0.001). Preterm labor was found in 15 (78.95%) patients in accreta group and six (14.36%) patients in nonaccreta group (P < 0.001). Hysterectomy was done in 14 (73.68%) patients in accreta group and four (9.76%) patients in nonaccreta group (P < 0.001). Urological complications occurred in five (26.32%) patients in accreta group and one (2.44%) patient in nonaccreta group (P = 0.004). Blood transfusion was needed in 16 (84.21%) patients in accreta group and seven (17.07%) patients in nonaccreta group (P < 0.001). ICU transfer was needed in 10 (52.63%) patients in accreta group and two (4.88%) patients in nonaccreta group (P < 0.001). By pathological study, placenta accreta was found in three (21.43%) patients while placenta percreta was found in six patients. Placenta increta was found in five (35.71%) patients [Table 3].
All the parameters of placenta accreta index were statistically significant. The total score was significantly higher in accreta patients (P < 0.001) [Table 4].
By performing receiver operating characteristic (ROC) curve of placenta accreta index [Figure 2] as a diagnostic marker, the area under the curve was 0.935. Score 4 was found to be the best cut-off point to diagnose placenta accreta. The number of patients that had index above 4 was 18 patients. Seventeen patients of them were proved to have placenta accreta while it was found in two patients with index below or equal to 4. The difference was significant (P < 0.001). The sensitivity, specificity, positive predictive value, and negative predictive value of the index are shown in [Table 5].
|Figure 2: ROC curve of placenta accreta index in diagnosis of placenta accreta. ROC, receiver operating characteristic.|
Click here to view
| Discussion|| |
This study evaluated the placenta accreta index as a predictive value for the proper diagnosis of PAS, also to improve the consistency and allow appropriate comparison of different imaging markers. The published literature is difficult to interpret because of several problems in the definition, terminology, and diagnosis of this disorder.
In this study, we found that the identification of placenta accreta among women was associated with similar risk factors and adverse maternal outcomes. History of prior cesarean, prior uterine surgery, and placenta previa in the current pregnancy were all significantly more common in the accreta group.
The outcomes with placenta accreta were significantly more likely to have a major hemorrhagic morbidity including postpartum hemorrhage, blood transfusion, hysterectomy, and maternal ICU admission. Furthermore, in subsequent pregnancies, we found an increased risk of morbidly adherent placenta following index pregnancies with occult placenta accreta compared with a retained placenta without accreta (42 vs. 19%).
This study showed the maternal age is older than nonaccreta group and it is significant risk factor for placenta accreta spectrum (PSA) as proved by Vinograd et al. . In this study 31% of PAS cases were associated with placenta previa. This is consistent with the findings of Oyelese and Smulian .
In our study, there were statistically significant differences between adherent and nonadherent placenta in relation to age, gravidity, and parity. In contrary, Cali et al.  reported no statistically significant difference between MAP and nonadherent in relation to age, parity, and gravidity.
Several authors have attempted to make an antenatal diagnosis of placenta accrete sonographically. Very few authors reported on the use of transvaginal sonography in PAS outside the location and follow-up of placenta previa accreta. Thus, most case reports and cohort studies described ultrasound images obtained transabdominally from PAS diagnosed in the late second and early third trimesters. This, and the fact that the resolution of ultrasound imaging has improved over the last three decades, may explain the wide variation in terminology used to describe the prenatal ultrasound features associated with PAS .
In this study the analysis was performed using a ROC curve, which indicated that the combination of the smallest sagittal myometrial thickness, intraplacental lacunae, and bridging vessels, in addition to the number of previous cesarean deliveries and placental location, generates an area under the curve of 0.87 (95% confidence interval, 0.80–0.95). Each parameter was weighted to create a nine-point scale in which a score of 0–9 (placenta accreta index) provided a probability of invasion that ranged from 2 to 96%, respectively. In our study, score 4 was found to be the best cut-off point to diagnose placenta accreta. The number of patients that had index above 4 was 18 patients. Seventeen patients of them were proved to have placenta accreta while it was found in two patients with index below or equal to 4. The difference was significant.
A similarly designed study done by Weiniger et al.  where 92 cases of suspected accreta found that the area under the ROC curve was 0.85, with contribution from three variables: placenta previa, number of previous cesarean deliveries, and ultrasound suspicion. These studies indicate that combining diagnostic features associated with PAS disorders through mathematical modeling may improve accuracy of prenatal diagnosis compared with ultrasound alone. However, like most single center studies, these may have overestimated accuracy because they are conducted in centers specialized in prenatal diagnostics, and the overall number of cases of PAS disorders included in these series is small. This study also did not differentiate between adherent and invasive cases, limiting the use of the data in clinical practice.
In our study, sensitivity of score 4 of placenta accreta index was 89.5% while the specificity was 97.6%. Warshak et al.  reported that the criteria of abnormal placenta detected by gray-scale ultrasound had sensitivity of 86% and negative predictive value of 92%. They concluded that visualization of Lacunae has the highest sensitivity in diagnosis of placenta accrete.
Comstock and Bronsteen  reported that out of the mentioned criteria for diagnosing placenta accreta using ultrasonographic examination during the second and third trimesters, intraplacental lacunae were found to be the most reliable one, with 93% sensitivity and 93% positive predictive value.
Histopathological confirmation was done in 14 cases in accreta group who had hysterectomy. Pathology confirmed placenta accreta in three patients, placenta increta in five patients and placenta percreta in six patients. The remaining five patients had focal accreta so diagnosis of placenta accreta was based on intraoperative surgeon's finding with no need for pathological confirmation. In the other group, four patients had hysterectomy for uncontrolled bleeding from placenta bed and the pathology reported no sign of placental invasion.
| Conclusion|| |
Placenta accreta index score 4 can be used as a predictive cut off value for prediction of placental invasion in patients with placenta previa.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Aggarwal R, Suneja A, Vaid NB, Yadav P, Sharma A, Mishra K. Morbidly adherent placenta: a critical review. J Obstet Gynaecol India 2012; 62
Eller A, Bennett M, Sharshiner M, Masheter C, Soisson A, Dodson M et al
. Maternal morbidity in cases of placenta accreta managed by a multidisciplinary care team compared with standard obstetric care. Obstet Gynecol 2011; 117
Omokanye L, Olatinwo A, Salaudeen A, Ajiboye A, Durowade K. A 5-year review of pattern of placenta previa in Ilorin, Nigeria. Int J Health Sci (Qassim) 2017; 11
Fujisaki M, Furukawa S, Maki Y, Oohashi M, Doi K, Sameshima H. Maternal morbidity in women with placenta previa managed with prediction of morbidly adherent placenta by ultrasonography. J Pregnancy 2017; 2017
Buca D, Liberati M, Calì G, Forlani F, Caisutti C, Flacco M, et al
. Influence of prenatal diagnosis of abnormally invasive placenta on maternal outcome: systematic review and meta-analysis. Ultrasound Obstet Gynecol 2018; 52
Abuhamad A. Morbidly adherent placenta. Semin Perinatol 2013; 37
Goh W, Zalud I. Placenta accreta: diagnosis, management and the molecular biology of the morbidly adherent placenta. J Matern Fetal Neonatal Med 2015; 29
Rac M, Dashe J, Wells C, Moschos E, McIntire D, Twickler D. Ultrasound predictors of placental invasion: The Placenta Accreta Index. Am J Obstet Gynecol 2015; 212
Finberg H, Williams J. Placenta accreta: prospective sonographic diagnosis in patients with placenta previa and prior cesarean section. J Ultrasound Med 1992; 11
Vinograd A, Wainstock T, Mazor M, Beer-Weisel R, Klaitman V, Dukler D, et al
. Placenta accreta is an independent risk factor for late pre-term birth and perinatal mortality. J Matern Fetal Neonatal Med2015; 28
Oyelese Y, Smulian C. Placenta previa, placenta accreta, and vasa previa. Am J Obstet Gynecol 2006; 107
Cali G, Giambanco L, Pucchio G, Forlani F. Morbidly adherent placenta: evaluation of ultrasound diagnostic criteria and differentiation of placenta accreta from percreta. Ultrasound Obstet Gynecol 2013; 41
Jauniaux E, Collins S, Burton G. Placenta accreta spectrum: pathophysiology and evidence-based anatomy for prenatal ultrasound imaging. Am J Obstetr Gynecol 2018; 218
Weiniger C, Einav S, Deutsch L, Ginosar Y, Ezra Y, Eid L. Outcomes of prospectively collected consecutive cases of antenatal- suspected placenta accreta. Int J Obstet Anesth 2013; 22
Warshak C, Eskander R, Hull A, Scioscia A, Mattery R Benirschke K, Resnik R. Accuracy of ultrasonography and magnetic resonance imaging in the diagnosis of placenta accreta. Obstet Gynecol 2006; 108 (3 Part 1
Comstock C, Bronsteen R. The antenatal diagnosis of placenta accreta. BJOG 2014; 121
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]