Clinical significance of retained products of conception in placenta previa: a retrospective analysis.

BACKGROUND: Retained products of conception (RPOC) often cause severe postpartum hemorrhage (PPH) but the clinical significance of RPOC in placenta previa is unclear. This study aimed to investigate the clinical significance of RPOC in women with placenta previa. The primary outcome was to evaluate risk factors of RPOC and the secondary outcome was to consider risk factors of severe PPH. METHODS: Singleton pregnant women with placenta previa who underwent cesarean section (CS) and placenta removal during the operation at the National Defense Medical College Hospital between January 2004 and December 2021 were identified. A retrospective analysis was performed to examine the frequency and risk factors of RPOC and the association of RPOC with severe PPH in pregnant women with placenta previa. RESULTS: This study included 335 pregnant women. Among these, 24 (7.2%) pregnant women developed RPOC. Pregnant women with prior CS (Odds Ratio (OR) 5.98; 95% Confidence Interval (CI) 2.35-15.20, p < 0.01), major previa (OR 3.15; 95% CI 1.19-8.32, p < 0.01), and placenta accreta spectrum (PAS) (OR 92.7; 95% CI 18.39-467.22, p < 0.01) were more frequent in the RPOC group. Multivariate analysis revealed that prior CS (OR 10.70; 95% CI 3.47-33.00, p < 0.01,) and PAS (OR 140.32; 95% CI 23.84-825.79, p < 0.01) were risk factors for RPOC. In pregnant women who have placenta previa with RPOC or without RPOC, the ratio of severe PPH were 58.3% and 4.5%, respectively (p < 0.01). Furthermore, the occurrence of prior CS (OR 9.23; 95% CI 4.02-21.20, p < 0.01), major previa (OR 11.35; 95% CI 3.35-38.38, p < 0.01), placenta at the anterior wall (OR 3.44; 95% CI 1.40-8.44, p = 0.01), PAS (OR 16.47; 95% CI 4.66-58.26, p < 0.01), and RPOC (OR 29.70; 95% CI 11.23-78.55, p < 0.01) was more in pregnant women with severe PPH. In the multivariate analysis for severe PPH, prior CS (OR 4.71; 95% CI 1.29-17.13, p = 0.02), major previa (OR 7.50; 95% CI 1.98-28.43, p < 0.01), and RPOC (OR 13.26; 95% CI 3.61-48.63, p < 0.01) were identified as risk factors. CONCLUSIONS: Prior CS and PAS were identified as risk factors for RPOC in placenta previa and RPOC is closely associated with severe PPH. Therefore, a new strategy for RPOC in placenta previa is needed.

Hemorrhagic morbidity in nulliparous patients with placenta previa without placenta accrete spectrum disorders.

Background: Placental adhesion spectrum (PAS) is a disease in which the trophoblast invades the myometrium, and is a well-known high-risk condition associated with placental previa. Aim: The morbidity of nulliparous women with placenta previa without PAS disorders is unknown. Patients and Methods: The data from nulliparous women who underwent cesarean delivery were collected retrospectively. The women were dichotomized into malpresentation (MP) and placenta previa groups. The placenta previa group was categorized into previa (PS) and low-lying (LL) groups. When the placenta covers the internal cervical os, it is called placenta previa, when the placenta is near the cervical os, it is called the low-lying placenta. Their maternal hemorrhagic morbidity and neonatal outcomes were analyzed and adjusted using multivariate analysis based on univariate analysis. Results: A total of 1269 women were enrolled: 781 women in the MP group and 488 women in the PP-LL group. Regarding packed red blood cell transfusion, PP and LL had adjusted odds ratio (aOR) of 14.7 (95% confidence interval (CI): 6.6 - 32.5), and 11.3 (95% CI: 4.9 - 26) during admission, and 51.2 (95% CI: 22.1 - 122.7) and 10.3 (95% CI: 3.9 - 26.6) during operation, respectively. For intensive care unit admission, PS and LL had aOR of 15.9 (95% CI: 6.5 - 39.1) and 3.5 (95% CI: 1.1 - 10.9), respectively. No women had cesarean hysterectomy, major surgical complications, or maternal death. Conclusion: Despite placenta previa without PAS disorders, maternal hemorrhagic morbidity was significantly increased. Thus, our results highlight the need for resources for those women with evidence of placenta previa including a low-lying placenta, even if those women do not meet PAS disorder criteria. In addition, placenta previa without PAS disorder was not associated with critical maternal complications.

Analysis of risk factors for massive intraoperative bleeding in patients with placenta accreta spectrum.

BACKGROUND: To analyze relevant factors for massive postpartum hemorrhage in women with placenta accreta spectrum in order to improve the ability to identify those at risk for intraoperative bleeding and improve outcome. METHODS: This study is a retrospective study and based on data from Hospital electronic medical record. Placenta accreta patients who delivered by cesarean section at Peking University Third Hospital from September 2017 to December 2019 were selected and included. According to the amount of intraoperative bleeding, they were categoried into the massive bleeding group (bleeding volume ≥ 2000 mL, 68 cases) and non-massive bleeding group (bleeding volume < 2000 mL, 99 cases). Univariate analysis and multivariate logistic regression were used to analyze the correlations between related risk factors or ultrasound imaging characteristics and the severity of bleeding during operation. RESULTS: (1) There were statistically significant differences in gravidity, parity, number of prior cesarean deliveries and placenta accreta ultrasound scores (P < 0.05) between the two groups of patients. (2) Among the ultrasonographic indicators, the disappearance of the post-placental clear space, the emergence of cross-border blood vessels in the region of subplacental vascularity, interruption or disappearance of the bladder line, and the presence of the cervical blood sinus had the most significant correlation with hemorrhage during PAS (P < 0.05). CONCLUSION: The presence of cervical blood sinus, interruption or disappearance of bladder line, the disappearance of the post-placental clear space and abnormal subplacental vascularity are independent risk factors for massive hemorrhage during PAS. We should pay more attention to these indicators in prenatal ultrasound examination in order to reduce the intraoperative bleeding and improve maternal outcomes.

Intraoperative blood volume loss according to gestational age at delivery among pregnant women with placenta accreta spectrum (PAS): an 11-year experience in Songklanagarind Hospital.

A retrospective study was conducted to evaluate the intraoperative blood volume loss in pregnant women with PAS according to gestational age at delivery. A total of 116 women were enrolled, 39 (33.6%) had an intraoperative massive blood loss (>5000 ml). The massive haemorrhage group had statistically significantly higher percentages of increta and percreta type than the non-massive haemorrhage group (94.9 vs. 67.5%, p < .001). Multiple linear regression analysis showed a decreasing trend of intraoperative blood loss after 34 weeks' gestation with the nadir period between 35 and 36+6 weeks' gestation, especially from 36-36+6 weeks' gestation which was statistically significant, p <.05. The perinatal morbidities from 36-36+6 weeks were not statistically significantly different from 37 weeks' gestation. Therefore, we recommend that pregnant women with PAS and stable clinical symptoms should be scheduled for caesarean hysterectomy from 36-36+6 weeks' gestation.Impact statementWhat is already known on this subject? Massive obstetric haemorrhage from PAS disorders is the main concern for caesarean hysterectomy among these patients as it leads to secondary complications including coagulopathy, multisystem organ failure, and death.What do the results of this study add? The amount of intraoperative blood loss in pregnant women who underwent caesarean hysterectomy due to PAS, was lowest from 36-36+6 weeks' gestation.What are the implications of these findings for clinical practice and/or further research? We recommend that pregnant women with PAS and stable clinical symptoms should be scheduled for caesarean hysterectomy from 36-36+6 weeks' gestation.

Scoring system for the prediction of the severity of placenta accrete spectrum in women with placenta previa: a prospective observational study.

PURPOSE: The clinical outcomes are significantly different in accreta, increta, and percreta. There is currently no scoring system that can preoperatively distinguish its severity in an at-risk population. The aim of this study is to establish a scoring system for the prediction of the severity of placenta accrete spectrum (PAS) in women with placenta previa. METHODS: A prospective observational study was conducted in patients with placenta previa who delivered at a Chinese tertiary care center between June 12, 2016 and June 30, 2018. Optimal scaling regression was performed to determine the parameters which really contribute to the prediction of PAS, and calculate percentage of contribution. RESULTS: Among 392 cases with placenta previa, 79, 53, and 28 had been surgically and/or histologically confirmed as accreta, increta, or percreta, respectively. Seven parameters were scheduled for the estimated scores for PAS, and five of them were finally entered into the predictive model. Their percentage of contribution was as follows: placental lacunas (19%), vascularity at the uterus-bladder interface (17.5%), myometrial thickness and hypoechoic retroplacental zone (25.6%), bladder line (22.6%), and previous caesarean sections (15.3%). The thresholds of scores for the prediction of accreta, increta, and percreta yielded 2.25-6.2, 6.2-8.95, and â‰§ 8.95, respectively, with the positive and negative predictive value, and false positive rates of the scoring system were 96.68%, 95.44%, and 3.32%, respectively. CONCLUSIONS: The scoring system can predict the severity of PAS in women with placenta previa. This will help identify the actual high-risk patients and improve their treatment.

The use of pressure balloons in the treatment of first trimester cesarean scar pregnancy.

Cesarean scar pregnancy (CSP) is among the most severe complications of cesarean delivery. CSP refers to the abnormal implantation of the gestational sac in the area of the prior cesarean delivery (CD), potentially leading to severe hemorrhage, uterine rupture, or development of placenta accreta spectrum disorders (PAS). The management of women with CSP has not been standardized yet. In women who opted for termination, discussion about the treatments should consider maternal symptoms, gestational age at intervention, and the future reproductive risk. A multitude of treatments, either medical or surgical, for CSP has been reported in the published literature. The present review aims to provide up-to-date information on a recently introduced minimally invasive treatments for CSP, including the single and double balloon catheter. The methodology of using the single or double catheter is described in a step-by-step fashion illustrated by pictures as well as video recordings. Both catheters have their deserved place to be used as a primary method for terminating scar pregnancies as well as using them as adjuncts to other treatments. They were successfully used by multiple individual practitioners and institutions due to their simplicity and low complication rates. The rare, but possible post-procedure complications such as recurrent CSP and enhanced myometrial vascularity are also mentioned.

Anterior uterine incarceration complicated by placenta previa and placenta accreta spectrum disorder: a case report.

Background: Uterine incarceration is a rare obstetric complication that is always associated with retroversion and prone to misdiagnosis. Pelvic examination and imaging methods including ultrasound and magnetic resonance imaging (MRI) are used as the primary diagnostic tool. We present an asymptomatic anterior uterine incarceration complicated by placenta previa and placenta accreta spectrum (PAS) disorder, which could be diagnosed during the pregnancy periods, but was first diagnosed during the cesarean section (CS) and got the surgeons into trouble. Case Description: A 28-year-old woman, gravidity 4, parity 1, was hospitalized due to placenta previa and PAS disorder diagnosed by ultrasound and MRI at 35.6 weeks of gestation. She had not experienced any discomfort. Given her history of a previous CS, she underwent a well-prepared cesarean delivery for the termination of the pregnancy. The patient had a series of periodical ultrasound and MRI examinations in which placenta previa and placenta accreta were described in disregard of the abnormal location of cervix; consequently, interior uterine incarceration was first diagnosed during the surgery, which caused significant difficulties in the operation which lasted 3 hours and 21 minutes. The patient developed severe hemorrhaging and lost approximately 5,000 mL of blood. Fortunately, she delivered a health male infant weighing 3,440 grams with quite good Apgar scores. During the follow-up, maternal and child health was confirmed. Conclusions: With regard to patients who have undergone previous pelvic surgery, doctors need to pay close attention to the position of the cervix, the pelvic adhesion situation during the prenatal examination, and be on alert for uterine incarceration. Uterine incarceration can be accurately recognized by periodic sonography and MRI if the radiologist is aware of this unusual condition. We do suggest that special are must be taken to avoid unnecessary trauma by misdiagnosis because of lack of awareness.

Placenta accreta spectrum disorders in twin pregnancies as an under reported clinical entity: a case series and systematic review.

Recent reports suggested a potential association between twin pregnancy and the occurrence of placenta accreta spectrum (PAS) disorders. Despite this, scarce data on PAS disorders in twins has been reported in the published literature. We present a series of twelve twin pregnancies complicated by PAS from two large institutions over 5 years. A systematic review of the literature was also conducted in order to find studies reporting on the risk factors, prenatal diagnosis using ultrasound and clinical outcomes of PAS in twin pregnancies.

The environmental risk factors prior to conception associated with placenta accreta spectrum: An umbrella review.

BACKGROUND: The present umbrella review evaluated the environmental risk factors prior to conception associated with placenta accrete spectrum(PAS) based on meta-analyses and systematic reviews. MATERIALS AND METHODS: We searched PubMed, Scopus, and Web of Science until October 14, 2021. All meta-analyses that had focused on assessing the risk factors associated with the PAS were included. We calculated summary effect estimates, 95% CI, heterogeneity I², 95% prediction interval, small-study effects, excess significance biases, and sensitive analysis. The quality of the meta-analyses was evaluated with A Measurement Tool to Assess Systematic Reviews 2 (AMSTAR 2). RESULT: The risk factor of in-vitro fertilization (IVF) (OR 5.03, 95% CI: 3.34, 7.56) was graded as suggestive evidence (class III). The multiple gestation (OR 1.90, 95% CI: 1.26, 2.88) was graded as the risk factor with weak evidence (class IV). Hypertension disorders (OR 0.5, 95% CI: 0.30, 0.82), low socioeconomic status (SES) (OR 0.51, 95% CI: 0.37, 0.71) (class IV) and male fetus (OR 0.79, 95% CI: 0.74, 0.84) (class III) were as the protective factors (class IV). The methodological quality of four of the included meta-analyses, based on AMSTAR 2, was low and three meta-analyses were critically low. CONCLUSION: The multifetal gestation and IVF were environmental risk factors for the PAS, while hypertension disorders, low SES, and male fetus were the protective factors.

Performance comparison of ultrasonography and magnetic resonance imaging in their diagnostic accuracy of placenta accreta spectrum disorders: a systematic review and meta-analysis.

OBJECTIVES: Accurate prenatal diagnosis of placenta accrete spectrum disorder (PAS) remains a challenge, and the reported diagnostic value of ultrasonography (US) and magnetic resonance imaging (MRI) varies widely. This study aims to systematically evaluate the diagnostic accuracy of US as compared with MRI in the detection of PAS within the identical patient population. METHODS: Medline, EMBASE, Google scholar and Cochrane library were searched. Pooled sensitivity, specificity, diagnostic odds ratio (DOR) and the area under the summary receiver operating characteristic (SROC) curve were calculated. Subgroup analysis was also performed to elucidate the heterogeneity of results. RESULTS: A total of 18 articles comprising 861 pregnancies were included in the study. The overall diagnostic accuracy of US for identification of PAS was as follows: sensitivity [0.90 (0.86-0.93)], specificity [0.83 (0.79-0.86)], DOR [39.5 (19.6-79.7)]. The overall diagnostic accuracy of MRI for identification of PAS was as follows: sensitivity [0.89 (0.85-0.92)], specificity [0.87 (0.83-0.89)], DOR [37.4 (17.0-82.3)]. The pooled sensitivity (p = 0.808) and specificity (p = 0.413) between US and MRI are not significantly different. SROC analysis revealed that there was no statistical difference (p = 0.552) in US and MRI for the overall predictive accuracy of PAS. Furthermore, in the subgroup analysis of between retrospective and prospective studies, between earlier and most recent studies, there was no statistical difference (p > 0.05) in diagnostic accuracy of US and MRI for the detection of PAS. CONCLUSIONS: Both ultrasonography (US) and magnetic resonance imaging (MRI) showed comparable accuracy in the prenatal diagnosis of placenta accrete spectrum disorder (PAS). Routine employment of MRI with relatively high expense in the prenatal identification of PAS should not be recommended.

The maximum length of T2-dark intraplacental bands may help predict intraoperative haemorrhage in pregnant women with placenta accreta spectrum (PAS).

PURPOSE: To investigate the relationship between the maximum length of T2-dark intraplacental bands (MLTIB) and intraoperative haemorrhage in pregnant women with placenta accreta spectrum (PAS). METHODS: Between February 2018 and February 2021, 86 pregnant women with PAS who delivered in Taizhou Hospital of Zhejiang Province and underwent preoperative magnetic resonance imaging (MRI) examination were retrospectively recruited. The presence of T2-dark intraplacental bands, placental/uterine bulge, loss of retroplacental T2-hypointense line, myometrial thinning, bladder wall interruption, focal exophytic mass, and abnormal vascularization of placental bed were recorded, and the MLTIB was measured. The relative risk ratios of the MRI findings and intraoperative bleeding were measured. Receiver operating characteristic (ROC) analysis was used to evaluate the ability of the MLTIB to help predict intraoperative haemorrhage in pregnant women with PAS. RESULTS: Of the 86 pregnant women, 32 had intraoperative blood loss ≥ 1000 ml; of these, 18 had intraoperative blood loss ≥ 2000 ml. Abnormal vascularization of placental bed was associated with the highest relative risk ratio for the detection of intraoperative haemorrhage (RR = 10.66), followed by the presence of T2-dark intraplacental bands (RR = 8.02). The optimal cut-off of the MLTIB for predicting intraoperative haemorrhage (≥ 1000 ml) in pregnant women with PAS was 28.95 mm, and the AUC was 0.91 (sensitivity: 84%; specificity: 91%). The optimal cut-off of the MLTIB for predicting massive intraoperative haemorrhage (≥ 2000 ml) was 35.65 mm, and the AUC was 0.94 (sensitivity: 89%; specificity: 85%). CONCLUSION: MLTIB was related to intraoperative haemorrhage in pregnant women with PAS. An MLTIB greater than 28.95 mm is an effective predictor of intraoperative haemorrhage. An MLTIB of 35.65 mm or greater strongly suggests the possibility of massive intraoperative haemorrhage.

A dual-path neural network fusing dual-sequence magnetic resonance image features for detection of placenta accrete spectrum (PAS) disorder.

With the increase of various risk factors such as cesarean section and abortion, placenta accrete spectrum (PAS) disorder is happening more frequently year by year. Therefore, prenatal prediction of PAS is of crucial practical significance. Magnetic resonance imaging (MRI) quality will not be affected by fetal position, maternal size, amniotic fluid volume, etc., which has gradually become an important means for prenatal diagnosis of PAS. In clinical practice, T2-weighted imaging (T2WI) magnetic resonance (MR) images are used to reflect the placental signal and T1-weighted imaging (T1WI) MR images are used to reflect bleeding, both plays a key role in the diagnosis of PAS. However, it is difficult for traditional MR image analysis methods to extract multi-sequence MR image features simultaneously and assign corresponding weights to predict PAS according to their importance. To address this problem, we propose a dual-path neural network fused with a multi-head attention module to detect PAS. The model first uses a dual-path neural network to extract T2WI and T1WI MR image features separately, and then combines these features. The multi-head attention module learns multiple different attention weights to focus on different aspects of the placental image to generate highly discriminative final features. The experimental results on the dataset we constructed demonstrate a superior performance of the proposed method over state-of-the-art techniques in prenatal diagnosis of PAS. Specifically, the model we trained achieves 88.6% accuracy and 89.9% F1-score on the independent validation set, which shows a clear advantage over methods that only use a single sequence of MR images.

Radical surgery including the role of an interval hysterectomy.

The incidence of placenta accrete spectrum (PAS) disorders is increasing worldwide. Pregnancies complicated by PAS are at a high risk of intrapartum surgical complications, mainly due to severe maternal hemorrhage, potentially leading to death, thus highlighting the need for a tailored an appropriate surgical management for these women. Despite its clinical relevance, there are still unanswered questions regarding the surgical management of women with PAS. Hysterectomy has been considered as the gold standard for the surgical treatment of these women. However, the surgical approach has not yet been standardized, and several conservative surgical procedures such as the Triple P Procedure are also being performed for PAS. Interventional radiology techniques have been demonstrated to reduce the risk of severe blood loss in women with postpartum hemorrhage, but their role in the management of women with PAS has not yet been fully defined. The aim of this chapter is to provide an up-to-date insight on the radical surgical approach to adopt during cesarean delivery in pregnancies complicated by PAS disorders.

Placenta accreta spectrum: Management and outcomes in a tertiary centre in India: An observational cross-sectional study.

Our observational cross-sectional study looked at the risk factors, diagnosis, management and outcomes of placenta accrete spectrum at the Christian Medical College and Hospital, Vellore, India, between January 2013 and December 2018. A total of 21 cases of placenta accrete spectrum are described among whom a preop diagnosis was available in 14 cases. A previous history of Caesarean section and placenta previa was present in 90%. Caesarean hysterectomy was carried out in 80%, but none of those managed conservatively required interval hysterectomy. Urinary tract injury was the most common surgical complication, seen in over 50%. The mean blood loss was 3.5 l and 14 patients required intensive care unit admission, but no maternal mortality ensued. Thus, we conclude that the conservative management in carefully selected cases is feasible.

Assessment of the massive hemorrhage in placenta accreta spectrum with magnetic resonance imaging.

The aim of the present study was to evaluate whether MRI features are able to predict massive hemorrhage of patients with placenta accreta spectrum (PAS). A total of 40 patients with suspected PAS after ultrasound examination were subjected to MRI. Of these, 29 patients were confirmed as having PAS. MRI data were analyzed independently by two radiologists in a blinded manner. Inter-observer agreement was determined. The 29 confirmed patients were divided into two groups (moderate and massive hemorrhage) according to the estimated blood loss (EBL) and blood transfusion, and the MRI features were compared between the two groups. The EBL, as well as blood transfusion, between the patients with and without each MRI feature were compared. The inter-observer agreement between the two radiologists for the 11 MRI features had statistical significance (P<0.05). Intra-placental thick dark bands and markedly heterogeneous placenta were the most important MRI features in predicting massive hemorrhage and blood transfusion (P<0.05). The difference in EBL between the patients with and without focal defect of the uteroplacental interface (UPI) was significant (P<0.05). The differences in blood transfusion between the patients with and without myometrial thinning, disruption of the inner layer of the UPI, increased placental vascularity and increased vascularity at the UPI were significant (P<0.05). These results indicate that MRI features may predict massive hemorrhage of patients with PAS, which may be helpful for pre-operative preparation of PAS patients.

Uterine conservation in placenta accrete spectrum (PAS) disorders: A retrospective case series: Is expectant management beneficial in reducing maternal morbidity?

OBJECTIVE: To examine whether expectant management confers any benefit on operative morbidity for the management of placenta accrete spectrum (PAS) disorders. STUDY DESIGN: This was a single center retrospective cohort study at a tertiary referral center In Singapore. Women with PAS disorder between January 2006 and December 2017 were identified from the hospital register. Antenatal features, surgical factors and post-operative morbidity were compared between women having caesarean hysterectomy, those having caesarean section with placental removal and women having expectant management, defined as caesarean section with retention of placenta using the student's t and Chi square tests. The natural course, complications and preservation of fertility were examined for women having expectant management. RESULTS: Ninety women with PAS were included. The incidence of PAS was 0.064 %. Mean gestational age (GA) at diagnosis was 26.4 weeks. Elective and emergency deliveries were performed at 36.7 and 32.4 weeks respectively (p = <0.0001). Caesarean hysterectomy, Caesarean section with placenta removal and expectant management (EM) were performed in 51(56.7 %),16(17.8 %) and 23(25.6 %) women respectively. The mean blood loss (MBL) and surgical time for EM were significantly lower than those for caesarean hysterectomy 0.52 L vs 3.17 L (p < 0.0001) and 70.8 min vs 171.6 min (p < 0.0001). The advantage of lower blood loss with expectant management persisted even after blood loss at delayed hysterectomy was considered (1284.09 mL vs 3168.72 mL (p-value <0.0001)). Uterine preservation with EM was successful in 61 % (14/23) women. Although nine women (39 %) in this group needed hysterectomy most complications were minor and presented within three months. The mean follow up overall was 13 months. CONCLUSION: Traditionally caesarean hysterectomy has been the main surgical approach for PAS. Our study consolidates existing evidence for expectant management being an option for a select group of patients to avoid complications associated with hysterectomy and allow uterine preservation.

Ultrasound accuracy in prenatal diagnosis of abnormal placentation of posterior placenta previa.

OBJECTIVES: To evaluate the accuracy of ultrasound in prenatal diagnosis of Placenta accrete spectrum disorders in patients with posterior placenta previa, and to assess the impact of prenatal diagnosis in our population. STUDY DESIGN: We prospectively enrolled 198 women with posterior placenta previa from 2011 to 2017. We performed transabdominal and transvaginal ultrasound examinations (Grey-scale and colour/power Doppler). The diagnosis of placenta accreta spectrum disorders was based on detection of at least two of the following criteria: loss of retroplacental clear zone, interruption of uterine serosa-bladder wall interface, turbulent placental lacunae with high velocity flow, myometrial thickness <1 mm, increased vascularity of uterine serosa-bladder wall interface, loss of vascular arch parallel to basal plate and/or irregular intraplacental vascularization. Definitive diagnosis was made at delivery with Caesarean section. Furthermore, we compared maternal outcomes in cases diagnosed antenatal versus that one's diagnosed at delivery. RESULTS: There were 20/198 cases of placenta accreta spectrum disorders. The two-criteria system identified 12 cases of placenta accreta, providing a 60.0% of sensitivity, 98.8% of specificity, 85.7% of positive and 95.7% of negative predictive value. Maternal outcomes were better in women with prenatal diagnosis of placenta accreta spectrum disorders, although not statistical significant. CONCLUSIONS: Our data showed that grey-scale and Color-Doppler ultrasound evaluation for detecting placenta accreta spectrum disorders on posterior placenta previa have high specificity, positive and negative predictive value, but a low sensitivity. Nevertheless, an antenatal diagnosis of placenta accreta spectrum disorders for posterior placenta previa should be encouraged.

Machine learning analysis of MRI-derived texture features to predict placenta accreta spectrum in patients with placenta previa.

PURPOSE: To evaluate whether a machine learning (ML) analysis employing MRI-derived texture analysis (TA) features could be useful in assessing the presence of placenta accreta spectrum (PAS) in patients with placenta previa (PP). The hypothesis is that TA features may reflect histological abnormalities underlying PAS in patients with PP thus helping in differentiating positive from negative cases. MATERIALS AND METHODS: Pre-operative MRI examinations of 64 patients with PP of which 20 positive (12 accreta, 7 increta and 1 percreta) and 44 negative for PAS were retrospectively selected. Multiple (n = 3) rounded regions of interest (ROIs) were manually positioned on sagittal or coronal T2-weighted images over homogeneous placental tissue close to the placental-myometrial interface for each patient to extract TA features. After balancing the dataset with the Synthetic Minority Over-sampling Technique, training and testing sets were obtained using Hold-out with a 75/25% split. Different algorithms were applied on the training set using the wrapper method, which looks for the best combination of features based on the optimization of a heuristic function in order to get the highest accuracy, and a 10-fold Cross-validation. The accuracy of the best models was also assessed on the test set. Histology was used as the standard of reference. RESULTS: A total of 192 ROIs were positioned and a ROI-based analysis was then conducted. Among the different algorithms, k-nearest neighbors obtained the highest accuracy (98.1%), precision (98.7%), sensitivity (97.5%) and specificity (98.7%) while exploiting the lowest number of features (n = 26); conversely, the Naïve Bayes algorithm got the lowest scores showing an accuracy of 80.5%. CONCLUSION: ML analysis using MRI-derived TA features could be a feasible tool in the identification of placental tissue abnormalities underlying PAS in patients with PP. This approach might represent an additional tool in the clinical practice, thus expanding the application field of artificial intelligence to medical images.