The Value Of Half-dose Contrast Medium Of MRI Dynamic Enhanced Scan In The Diagnosis Of Placental Accrete

PartⅠ Feasibility analysis of half-dose contrast medium of MRI dynamicenhanced scan (HCMMDES) in showing normal placenta and uterusObjective:To investigate the feasibility of half-dose contrast medium of MRI dynamicenhanced scan (HCMMDES) in showing normal placenta and uterus, the purpose ofthis study was to provide a theory basis for further studying the application value ofHCMMDES in the diagnosis of placenta increta.Materials and methods:40cases of normal middle-late pregnant women were randomly divided into2groups (control group and study group), each group with20cases. Conventional-doseof contrast medium（0.1mmol/kg）was administered in control group and half-doseof contrast medium (0.05mmol/kg) in study group. The pelvic MRI dynamicenhanced scan was performed in research object, and the time-signal intensity curve(TIC) was obtained by Mean Cure post-processing software, then the initial–intensity(II), maximum intensity (IMax), the time to peak (TTP) of uterus and placenta weremeasured, as well as the value between placenta and uterus when the uterus reach thepeak value. All statistical analyses were performed with SPSS17.0. Independentexperimental data was expressed as Mean±SD. The difference between two groupswas analyzed by two-sample t test and within group difference by paired-sample t test.P<0.05was considered statistically significant.Results:In the examination of conventional-dose and half-dose contrast medium of MRIdynamic enhanced scan, the TIC index of uterus and placenta were both100percent.There is no significant statistical difference between the two groups in initial signalintensity and time to peak of placenta and uterus (P＞0.05). The maximum intensityof both uterus and placenta in study group is notable higher than in that control group, P<0.05, the difference was statistically significant, and the value between placentaand uterus in control group was obviously lower than that in study group, P<0.05, thedifference was statistically significant. The initial-intensity of placenta was higherthan of uterus in both control group and study group, as well as the maximum-intensity and time to peak, the difference was statistically significant (P<0.05).Conclusion:1. The initial-intensity of placenta was higher than that of uterus, as well as themaximum-intensity, and time to peak of placenta preceded the uterus. All of these didnot change with contrast medium dosage.2. Though the maximum intensity of placenta and uterus in half-dose group waslower than that in conventional-dose group, the value between placenta and uteruswhen the uterus reach the peak value in the half-dose group was significantly higherthan that in the conventional-dose group, that is, when the uterus reach the peak valueof enhancement, the contrast degree of placenta and uterus in half-dose group washigher than that in conventional-dose group, which helps to distinguish from placenta,uterus and anatomical relationship between them. Consequently, HCMMDES ishelpful to medical image workers to improve the diagnosis level of uterus or placentadiseases for middle-late pregnant women.PartⅡ Application value of half-dose contrast medium of MRI dynamicenhanced scan (HCMMDES) in the diagnosis of placental incretaObjective:To investigate the diagnostic value of MRI for placental invasion, and to providea theory basis for clinical prenatal diagnosis and treatment.Materials and methods:1.From January2011to September2013,60hospitalized middle-late pregnantpatients who need pregnancy termination for preoperatively suspected placentaincreta were collected, all of them underwent half-dose contrast medium (0.05mmol/kg) of MRI dynamic enhanced scan. Then films were evaluated and analyzed by at least two experienced high qualification radiologists in consensus to make objectiveMRI imaging diagnosis. In case of disagreement, the final decision was madeconsensus by repeated film reading. After a follow-up, the preoperative imagingdiagnosis and surgical pathologic results were compared to evaluate the efficiencyand suitability of HCMMDES in the diagnosis of placenta increta. Then ROC curvewas drawn to evaluate the accuracy of HCMMDES in diagnosis and hierarchicaldiagnosis of placenta increta.2. Placenta increta was confirmed by surgical pathology, and the signs ofdynamic enhanced MRI in prenatal patients with placenta increta (study group) wereanalyzed and compared with those without placenta increta (control group) to exploreits diagnostic value.Results:1. Among the60patients with suspected placenta increta,43patients wereconfirmed with placental increta by surgical pathology. Placenta accreta was detectedin19patients, placenta increta was detected in16patients, placenta percreta wasdetected in8patients and placenta non-increta was detected in17patients. Of19patients with surgery confirmed with placenta accreta,13cases were correctlydiagnosed,4cases were missed diagnosed and2cases were misdiagnosed as incretaby MR dynamic contrast enhanced scan. Of16cases with surgery confirmed withplacenta increta,13cases were correctly diagnosed,1case was misdiagnosed asplacenta accreta and2cases misdiagnosed as placenta percreta by MR dynamiccontrast enhancement. Of8cases with surgery confirmed with placenta percreta,7cases were correctly diagnosed,1case was misdiagnosed as placenta increta by MRdynamic contrast enhanced scan; Of17cases with surgery confirmed with placentanon-increta,2cases were misdiagnosed as placenta accreta by MR dynamic contrastenhanced scan.2. Half-dose MRI dynamic contrast enhanced scan of placenta increta revealed90.7%sensitivity,88.2%specificity,11.8%misdiagnosis rate,9.3%rate of misseddiagnosis,90.0%the overall conformity rate,0.79Youden index,95.1%positivepredictive value,78.9%negative predictive value, Kappa=0.762(P<0.001). The areaunder the ROC curve was0.895. The difference was statistically significant compared with0.5(P<0.001). MRI diagnosis of placenta accreta revealed68.4%sensitivity,92.7%specificity,7.3%misdiagnosis rate,31.6%rate of missed diagnosis,85.0%theoverall conformity rate,0.61Youden index,68.4%positive predictive value,86.4%negative predictive value, Kappa=0.638(P<0.001). The area under the ROC curvewas0.806. The difference was statistically significant compared with0.5(P<0.001).MRI diagnosis of placenta increta revealed81.2%sensitivity,93.6%specificity,6.4%misdiagnosis rate,18.8%rate of missed diagnosis,95.0%the overall conformity rate,0.75Youden index,81.3%positive predictive value,93.6%negative predictive value,Kappa=0.749(P<0.001). The area under the ROC curve was0.874. The differencewas statistically significant compared with0.5(P<0.001). MRI diagnosis of placentapercreta revealed87.5%sensitivity,96.2%specificity,3.8%misdiagnosis rate,12.5%rate of missed diagnosis,95.0%the overall conformity rate,0.84Youden index,77.8%positive predictive value,98.0%negative predictive value, Kappa=0.795(P<0.001). The area under the ROC curve was0.918. The difference was statisticallysignificant compared with0.5(P<0.001).3.There were some direct signs when using dynamic enhanced MRI scan in thediagnosis of placenta increta:①The decidual layer between uterus and placenta was fuzzy or discontinuous,and the placenta outer contour on the interface between placenta and uterusdisappeared, and margin between the placenta and adjacent myometrium wasill-defined during time domain of interested, while the muscular continuity existed.②In early period of dynamic enhancement, time to peak of myometrium cameearly, replaced by relative high signal placental tissue, and high signal placentatissue invaded relative low signal muscle layer, presented as the "gear" or "roots",uterine serosa layer edges presented obvious enhancement; enhancement signal ofnormal myometrium disappeared or interrupted in the middle-late dynamic enhancedscan period.③The early period of dynamic enhanced scan showed notable enhancedplacental tissue went through myometrium and involved serosa or adjacentsurrounding structures such as bladder, bladder serosa layer was irregular interruptedor with wall nodules, the middle-late period showed that signal of normal muscle layer at the penetration area disappeared.The display rate were32.56%、37.2%、20.93%respectively. Comparing thedirect signs①of control group with study group respectively for statistics analysis: P<0.05, no statistical difference; comparing the direct signs②,③of the two groups, P<0.05,with statistical difference.4. There were some indirect signs in diagnosing placenta increta by MRIdynamic enhanced scan:(1) morphological changed:①The uterus lost its normal inverted pear-shape, with the lower segmentswelling or broadening②Uterine serous surface was uneven, normal smooth arc curve was disappeared③The placenta was irregular with local thickening④Placenta outer edge: continuous smooth arc curve edge structure wasdisappeared, local outward bulging as a wavy structure(2) Changes of the placenta parenchyma signal①Parenchyma enhancement signal gathered into a big shape, and contrastmedium leaked out as plenty of placental blood sinus opened②Non-enhanced low-signal-intensity bands being in placenta parenchyma③Placental vascularity increased, became bulky, distorted and extended to theedge of the placental maternal surface④Enhancement signal of placenta parenchyma was heterogeneous(3)The display rate of vermis arteriolar under serosa surface were79.07%、20.93%、62.79%、81.40%、20.93%、100%、23.26%、88.37%、65.12%,respectively. Comparing MRI signs of study group with control group forstatistics analysis respectively:(1)-②、(1)-③、(2)-①、(2)-③and3, P<0.05, withstatistical difference.(4) The volume of low-signal-intensity in placenta accreta, increta, pecreta inMRI enhancement scan indirect signs were:12.41±3.27cm3、19.50±4.25cm3、21.49±2.58cm3, P<0.05, the difference was statistically significant.Conclusion:half-dose contrast medium of MRI dynamic enhanced scan (HCMMDES) is helpful for prenatal diagnosis and hierarchical diagnosis of placental increta.