Prenatal Magnetic Resonance Imaging Of Placenta Percreta

Abnormal invasive placenta(AIP)represents a spectrum of disease severity characterized by abnormal,firmly adherent placenta implantation in various depths into the uterus without intervening decidua,which is typically referred to as placenta accreta,increta and percreta.Prior caesarean section and placenta previa are important predisposing factors for AIP.The incidence of AIP has increaseddramatically,as a consequence of the concomitant increasing rate of caesarean section and intrauterine operation.AIP has made a significant contribution to theintrapartum or postpartum massive haemorrhage and perinatal emergency hysterectomy.Placenta percreta(PP)is the most severe invasion type of AIP,where placentalvilliare found to invade the full thickness of the uterine wall or even the surrounding organs.As the most life-threatening type,the management of PP is different from other types.In addition,the topography of invasion may be associated with the methods of surgical operation and hemostasis.Of note,the invasion of some special regions(uterine cervix,urinary bladder,parametriumetc.)would increase the difficulty in the management,which may result in severe complications.The accurate prenatal diagnosis and evaluation(depth and topography)of AIP could facilitatemultidisciplinary management and minimize the potential risks of maternal morbidity.Ultrasonography is the primary modality for antenatal diagnosis of AIP,but it is usually limited by various factors(location of placenta,maternal physique,amniotic fluidetc.)and had the inherent drawbacks of low soft tissue resolution and small field of view,which leads to the difficulty in the evaluation of invasion depth and topography.Placental magnetic resonance imaging(MRI)is playing an important role in the prenatal diagnosis and evaluation of PP.Various MRI features have been shown to be indicative of AIP.However,there is no reliable MRI features for PP.In addition,although MRI has been proved to be superiorin delineating the topography of AIP than ultrasound,the capacity of MRI in evaluating the topography of PP,especially the specific regions(uterine cervix,urinary bladder and parametrium)remains unknown.Therefore,the purposes of our study were to identify specific MRI features for differentiating PP from PA and to evaluate the capability of MRI in delineating the topography of PP.There are two parts in the study:Part 1.MRI features of placenta percretaObjectiveTo characterize and subdivide the MR features of AIP,to identify specific features for differentiating PP from non-placenta percreta(non-PP),and to identify the correlation between MRI features and blood loss during the surgery.Materials and methodsPatients:This study was approved by our Institutional Review Board,and written informed consent was waived.A total of 138 patients with AIP confirmed by the surgical or pathological criteria were recruited in this study:45 pregnant women with PP(mean gestational age,34.3 weeks;mean age,33.4 years)and 93 pregnant women with non-PP(mean gestational age,33.4 weeks;mean age,32.2 years).MRI examination:The MRI examinations were performed on 1.5T MRI scanner(MAGNETOM Sonata,Siemens,Germany).Half-Fourier single-shot turbo spin echo(HASTE),true fast imaging with steady-state precession(True-FISP),T1-weighted imaging(T1WI)and diffusion weighted imaging(DWI)images were obtained.HASTE and True-FISP images were obtained in the axial,coronal,sagittal and oblique sagittalplanes.Additionalimaging planes,perpendiculartothe placenta-uterusinterface,were obtained in theregion ofthe suspectedAIP.T1WI and DWI images were obtained in sagittal planes.Imaging analysis:The MRI features of myometrial thinning,interrupted myometrium,loss of the placental-myometrial interface,marked placental heterogeneity,dark intraplacental band,abnormal intraplacental vascularity,abnormal uterine bulge,placental bulge(type-?,type-IIa and type-?b),uterine serosal hypervascularity,bladder wall nodularity and extrauterine placental extensionwere analyzed.Of note,uterine serosal hypervascularity is preliminarily explored usingMRI.What's more,placental bulge is initially subdivided into type-?,type-?a and type-?b depending on bridge vessels and whether the smooth uterine outline is distorted.All the MR images were evaluated retrospectively by two radiologists,who were blinded to the original radiology reports and surgical and pathological information of the subjects.They evaluated all the MRI features independently for presence or absence.Statistical analysis:All the potential MRI features were compared using Pearson's chi-squared or Fisher's two-sided exact test between women with PP and those with non-PP.Values that reached statistical analyses were then included in a logistic regression analysis to identify independent predictors of PP.The predicted value of the logisticregression model was assessed using the receiver operatingcharacteristic(ROC)curve analysis.The correlation between MRI features and blood loss was tested by one-wayANOVA analysis.ResultsSignificant differences were found in nine depicted MRI features(interrupted myometrium,marked placental heterogeneity,dark intraplacental band,abnormal intraplacental vascularity,abnormal uterine bulge,placental bulge(type-? and type-?),uterine serosal hypervascularity and bladder wall nodularity)between pregnancies with PP and those with non-PP(P<0.05).These nine significant MRI features were then introduced into a logistic regression analysis,which was used for selecting the optimal variables for distinguishing PP from non-PP.Placental bulge type-? and uterine serosal hypervascularity were independently associated with PP(odds ratio =48.618,P<0.001;odds ratio = 4.165,P = 0.018 respectively),and the combination of the two MRI features to distinguish PP from non-PP yielded an area under the ROC curve(AUC)of 0.92for its predictive performance.PP patients with the feature of placental bulge type-?a(with bridging vessels)experienced greater blood loss during caesarean section than type-?b(without bridging vessels)(3,829.6 ± 2,826.5.0 vs.2,192.3±1,583.0ml,P = 0.016).Similar findings were noted when comparing the blood loss between women with and without the feature of uterine serosal hypervascularity(2,044.0 ± 1,224.8 vs.1,286.8 ±1,360.7,P = 0.010).In the non-PP cohort,women with the feature of abnormal intraplacental vascularity experienced greater blood loss during caesarean section than women without this feature(1,730.2 ± 1,463.8 vs.1,092.9 ± 1,078.1 ml,P = 0.030).ConclusionPlacental bulge type-?(with distorted uterineoutline)and uterine serosal hypervascularity are specific MRI features for differentiating PP from non-PP.Profoundly abnormal vessels appears to be risk factors for massive blood loss during caesarean section.These results would contribute to an accurate prenatal diagnosis of PP and the evaluation of the risk of massive haemorrhage.Part 2.MRI of the topography of placenta percretaObjectiveTodetermine the capability of MRI fordelineating the topography of PP,and to identify the correlation between the topography of invasion and surgical procedures.Materials and methodsPatients:This study was approved by our Institutional Review Board,and written informed consent was waived.Thirty-nine pregnant women with PP(mean gestational age,33.9 weeks;mean age,34.4 years)and 50 women with non-PP(mean gestational age,33.4 weeks;mean age,32.9 years)with complete topography information in the surgical records selected from the part1 were recruited in this study.PP and non-PP were confirmed based on the surgical or pathological criteria.The topography was identified according to the surgical criteria:lower uterine segment,upper uterine segment and special regions(uterine cervix,urinary bladder and parametrium).MRI examination:All the MRI examinations were performed on 1.5T MR1 scanner(MAGNETOM Sonata,Siemens,Germany),and HASTE,True-FISP,TIWI and DWlimages were obtained.The scan azimuth was consistent with the part 1.Imaging analysis:All the MR images were evaluated retrospectively by two experienced radiologists,who were blinded to the clinical,surgical and pathological information of the subjects.AIP was diagnosed with PP or non-PP according the invasion depth.The MRI criteria of invasion topography:In the sagittal MRI images,a plane perpendicular to the center of the upper bladder axis allows the division of placenta invasion into two sectors--the upper and lower uterine segments.When uterine cervix distorted,bladder wall nodularity or extrauterine placental extension were demonstrated on the images,uterine cervix,urinary bladder or parametrium invasion was diagnosed respectively.Statistical analysis:The sensitivity,specificity,positive predictive value(PPV),negative value(NPV)of MRI in diagnosing PP were calculated.Fisher's two-sided exact test was used to compare the capability of MRI for evaluating the topography between PP and non-PP.Chi-squared was used to analyze the correlation between the topography and surgical procedure.ResultsMRI had good sensitivity,specificity,PPVand NPV(87.5%,86.0%,82.9%and 89.6%,respectively)in diagnosing PP.The total accuracy of MRI in evaluating the topography of PP was higher than that in non-PP(95.5%(37/39)vs.80.0%(40/50),P=0.039).The accuracy of MRI for evaluating the invasion of uterine cervix,urinary bladder and parametrium was 100%,83.3%and 100%respectively.In the PP cohort,the rate of hysterectomy inlower uterine segment invasion was higher than that in upper uterine segment invasion(75.0%vs.39.1%,P=0.029).In non-PP cohort,the rate of hysterectomy in lower uterine segment invasion and upper uterine segment was 21.4%and 36.4%respectively,and the difference was not significant(P=0.197).Five cases of PP with uterine cervix invasion were performed total hysterectomy(100%).Four cases of PP with parametrum invasion were placed bilateral ureteral stentspreoperatively.In the PP cohort,the rate of placenta-uterus local excision in upperuterine segment was significantly higher than that in lower uterine segment(52.2%vs.18.8%,P=0.049).But in non-PP cohort,the difference was not significant(P=0.254).The difference between the rate of abdominal aortic balloon occlusion in lower uterine segment and that in upper uterine segment was not significant(PP:69.6%vs.75.0%,P=0.500;non-PP:64.3%vs.59.1%.P=0.466).ConclusionMRI was reliable for evaluating the topography of PP,especially the urinary bladder,uterine cervix and parametrium.In addition,the topography of PP effected the selection of surgical procedure,and the lower uterine segment invasion had a higher risk of hysterectomy.Accurate MRI prenatal evaluation for PP facilitates surgical plan and decreases the risk of severe surgical complications.