Clinical Risk Assessment And Multi-omics Study Of Placenta Previa With Placenta Accreta

Objective:Placenta Previa and Placenta Accreta Spectrum(PAS)have close relationship as high risk factors with each other.Placenta previa with placenta accreta can lead to severe obstetric hemorrhage,hysterectomy,preterm birth,perinatal infant death and even maternal death,which is a catastrophic problem in obstetric fields.PAS is divided into placenta accreta,placenta increta,and placenta percreta based on the depth of that the villus penetrate the myometrium.There are difference on the harm and treatment strategy to the mother with different PAS types and the severity of the disease accordingly.Therefore,accurate prenatal diagnosis,severity estimation,multidisciplinary team management(MDT),and pathogenesis study have become hot research topics in recent years,but its pathogenesis is still lack of in-depth research.Accurate prenatal diagnosis is the prerequisite for effective management of placenta previa with PAS.At present,prenatal diagnosis of PAS mainly relies on clinical features and imaging examinations.Ultrasound and MRI both have high sensitivity and specificity in the diagnosis of PAS.It is of great significance to judge placenta position,shape and degree of implantation.Ultrasound examination is simple and economical,and is the preferred method for PAS prenatal screening.However,there are certain limitations in the accurate assessment of PAS type due to the influence of examination field,placenta position,maternal body size,amniotic fluid volume and operator subjective factors.MRI can make up for these deficiencies in ultrasound.It is particularly of higher value in the evaluation of PAS in the posterior uterine wall,placental implantation depth,and the degree of invision of parametrium and bladder.In addition,MRI can retain more complete and intuitive images,which is of great significance for preoperative evaluation and treatment planning and is an important method for prenatal diagnosis of PAS.But,none of the imaging signs can completely predict the severity of PAS,which requires a comprehensive judgment of multiple indicators.In recent years,scholars have improved the diagnostic accuracy of PAS by using the ultrasonographic features of placenta accreta to develop scoring scales.However,the MRI-based PAS predictive scoring scale has few literature reports,few clinical samples have been included,and the MRI signs are too complex to be popularized and applied in clinical practice.Therefore,a scientific evaluation system based on MRI images combined with clinical features has a positive effect on predicting the degree of placenta accreta,formulating appropriate treatment strategies,improving the prognosis of mother and child,and reducing maternal mortality.The interaction among decidual loss,enhanced trophoblast invasion and abnormal uterine spiral artery remodeling is generally considered to be the important physiological and pathological factors leading to placenta accreta.There are many factors affecting placenta accreta,the molecular regulatory network is complex,and the mechanismis not clear.Through transcriptome,lnc RNAs and mi RNAs and other omics studies,mapping the molecular regulatory network is of great significance for the study on the mechanism of placenta invasion,in-depth understanding of the occurrence and development of PAS,and searching for the markers of PAS with early diagnostic value.The main research objectives of this subject are as follows1.Analysis modeling data set of 568 cases of placenta previa patients through univariate and multivariate analysis,screening the clinical features and MRI core indicators that can predict the grading of PAS.The scoring scale was established,and the validity and accuracy of the diagnosis of the scoring scale were verified in the validation data set of 293 cases of placenta previa patients,so as to guide clinical treatment.2.By using multi-omics study techniques,we analyzed the differences of expression profiles of coding genes,lnc RNAs and mi RNAs between in placenta percreta tissue and normal placental tissue,and mapped the molecular regulatory networks,so as to provide multi-omics experimental evidence for in-depth understanding of the pathogenesis of PAS Materials and Methods:1.Retrospectively analyzed the risk factors of placenta previa with placenta accreta and established a scoring scale: A total of 568 cases of placenta previa delivered in Sichuan Provincial Maternity and Child Health Care Hospital from2013 to 2018 were selected as modling data set.According to clinical-surgical-pathological results,the patients were divided into four groups:non-PAS group,accreta group,increta group and percreta group.(1)Analysis of risk factors for PAS: The accreta group,increta group and percreta group were combined PAS group,and compared with non-PAS group,the clinical features and MRI core indicators were screened for predicting the occurrence and grading of PAS through univariate and multivariate analysis.(2)Analysis of risk factors for invasive placenta accreta: The increta group and percreta group were combined into the severe invasion group,and compared with the adhesion group(accreta group).Through univariate and multivariate analysis,the clinical features and MRI core indicators that can predicted invasive placenta accreta were screened.(3)Based the retrospective analysis results in modling data set.,a scoring scale was established to predict the severity of PAS with MRI characteristics as the main index.2.Clinical validation of the established scoring scale: 293 patients with placenta previa who cesarean delivered in Sichuan Provincial Maternity and Child Health Care Hospital from 2019 to 2020 were selected as validation data set to verify the evaluation effect of the scoring scale.According to clinical-surgical-pathological results,the patients were divided into four groups:non-PAS group,accreta group,increta group and percreta group.(1)According to the scores of PAS by scoring scale for each patient,combining clinical-surgery-pathological diagnosis results,calculate the PAS scoring scale area under the ROC curve(AUC),the cut-off value of PAS of different grades as well as sensitivity and specificity of the diagnosis,and the scoring scale for positive predictive value(PPV),negative predictive value(NPV)and false positive rate among different grades of PAS.(2)The clinical characteristics,intraoperative conditions and maternal and infant outcomes of patients with different grades of PAS were analyzed.(3)Develop the appropriate treatment plan combined the scoring scale and clinical analysis.3.Multi-omics analysis of the molecular mechanism of placenta implantation:Selected the placenta previa patients with PAS risk score≥6 for cesarean delivery between 2019 and 2020 who were diagnosed placenta percreta(6 cases)during operation and patients with cesarean section during the same period with no placenta accreta(6 cases)as the experimental group and control group,respectively,and extracted RNA,quality control and database construction.The multi-omics study and data analysis of coding genes,lnc RNAs and mi RNAs were carried out in 4cases of tissue meeting the requirements of sequencing database construction respectively in two groups.The coding genes and lnc RNAs that may play a key role in placenta accreta were screened out by using KEGG,WGCNA and Venn analysis techniques combined with mi RNA targeting prediction and correlation analysis.Multi-omics expression regulation network of coding genes-lnc RNAs-mi RNAs was constructed.In addition,Q-PCR,immunohistochemistry and fluorescence in situ hybridization staining were used to verify the expression of the screened key coding genes and lnc RNAs in the experimental and control groups of 6 samples.2.Statistical methods: SPSS 25.0 and Graph Pad Prism 5.0 software were used for statistical analysis,including univariate analysis,multivariate analysis and ROC curve analysis,PPV,NPV and false positive rate were calculated.Results:1.We retrospectively analyzed 568 patients with placenta previa and established an MRI-based scoring system for predicting piacenta accrta :(1)The operative time,blood loss and hysterectomy ratio of PAS group were significantly higher than that of N-PAS group.11 variables associated with placenta accreta for univariate and multivariate analysis,determined three MRI imaging signs were significantly associated with placenta accteta,including the T2-dark bands in placenta,loss of well-defined T2-hypointense placental-inner myometrial interface,placenta mainly located in the anterior uterine wall/cervix,the OR values were 2.952,4.032,1.697,the area under the ROC curve(AUC)were 0.785,0.797,0.641,respectively,the sensitivity was 72.3%,73%,59%,and the specificity was 81%,82.4%,67.2%,respectively.(2)The paraty,operation time and blood loss of the invasive placenta accreta group were all higher than those in the placenta accreta group.All the hysterectomy patients were invasive placenta accreta patients.11 indexes related to invasive placenta accreta were analyzed by univariate and multivariate analysis.The five indexes were significantly correlated with invasive placenta accreta,including the frequency of cesarean section ≥ 1,many of the T2-dark bands in placenta,disappearing of the T2 low signal line at the uterine placenta junction,placental /uterine bulge,placenta mainly located in the anterior uterine wall/cervix,The OR value of the five variables was 2.106-8.278,the area under the ROC curve was 0.666-0.815,the sensitivity range was 48.1%-86.1%,and the specificity range was 51.8%-98%.(3)According to the results of the retrospective study,five indexes,including the number of cesarean section,T2 low signal zone in placenta,T2 low signal line at utero-placenta interface,placental/uterine bulge,and placenta position,were used to establish a MRI-based scoring scale for predicting PAS.2.Clinical verification was performed on the established scoring scale: We also included 293 patients with placenta previa for clinical verification of the diagnostic efficacy of the scoring scale.The ROC curves of the scoring scale for the diagnosis of different grade of PAS were mapped.The areas under the curves for the diagnosis of PAS,invasive placenta accreta and placenta percreta were 0.908,0.905 and 0.894 respectively,and the corresponding cut off values were 1.5,3.5 and 5.5 points respectively.According to the cut-off value,the grades of PAS in the scoring scale was determined: 0-1 was normal,2-3 was accreta,4-5 was increta,and 6-8 was percreta.The PPV of PAS was 83.4%,the NPV was 100%,and the false positive rate was 16.6%.The PPV of increta was 100%,the NPV was 98%,and the false positive rate was 0%.The PPV of percreta was 65.5% and the NPV was 100%.Clinical analysis of the four groups showed that the percreta group was higher than the other three groups in terms of operation duration,blood loss and blood transfusion rate,with statistical differences.The operative duration,blood loss and transfusion rate of the increta group were statistically different from those of the no accreta group and accreta group.The amount of erythrocyte suspension and plasma input in implantation group was significantly higher than that in accreta group and no accreta group.Twenty-nine cases(9.9%)were treated with interventional procedures such as abdominal aortic balloon or internal iliac artery balloon occlusion,of which 26 were treated with increta and percreta,which used more interventions for hemostasis than the no accreta and accreta group.All the 17 percreta patients underwent hysterectomy,and there was no maternal or neonatal death.3.The molecular mechanism of PAS was analyzed by multi-omics: Multi-omics study and data analysis of coding genes,lnc RNAs and mi RNAs in 4 cases of penetrating placenta and 4 cases of normal placenta were performed.It was found that 790 coding genes,382 mi RNAs and 541 lnc RNAs were differentially expressed between the two groups(differential multiple ≥ 2,P <0.05).KEGG was used to analyze the differentially expressed coding genes related signaling pathways between the two groups.The differentially expressed coding genes in the penetrating placenta may regulate the pathological process of the invasive placenta through the regulation of several important signaling pathways such as cell proliferation,apoptosis and differentiation.WGCNA analysis confirmed that the coding genes involved in the regulation of placenta invasion were located in the Me-Black and Me-Turboise modules,and quantitative PCR and immunohistochemical experiments verified that the screened Wnt5 a and MAPK13 genes were down-regulated in the placenta implantation tissues.The top 20 lnc RNAs related to the expression of Wnt5 a and MAPK13 were further screened by Venn analysis,and quantitative PCR and FISH experiments showed that the expression levels of Wnt5 a and MAPK13 were positively correlated with the expression levels of lnc RNA-PTCHD1-AS and lnc RNA-PAPPA-AS1.Finally,by building a multi-omics regulation network for the expression of the coding gene-lnc RNAs-mi RNAs,lnc RNA-PTCHD1-As and lnc RNA-PAPPA-AS1 can co-regulate the expression of Wnt5 a and MAPK13 by interacting with multiple specific mi RNAs,which provides multi-omics experimental evidence for in-depth understanding of the pathogenesis of placenta implantation.Conclusion:We established the PAS scoring system predicting severity based on MRI signs,through the analysis of placenta previa patients,and through the clinical case further verified it’s effectiveness and accuracy.It can effectively predict the severity of PAS,effectively guided perioperative intervention,achieved a better clinical prognosis,and improved the clinical treatment strategy.Multi-omics analysis of the molecular regulatory network and multiple potentially important regulatory molecules involved in PAS to provides potential biomarkers for the prediction of PAS.