Lu Tao, Zhang Ju, Wang Li, Li Mou, Chen Yazheng, Huang Yufu
Department of Radiology, Sichuan Provincial People's Hospital, University of Electronic Science and Technology of China, Chengdu, China.
Quant Imaging Med Surg. 2025 Jun 6;15(6):4995-5006. doi: 10.21037/qims-24-2450. Epub 2025 May 22.
Placenta accreta spectrum (PAS) is usually difficult to diagnose in posterior placenta. This study aimed to determine whether certain magnetic resonance imaging (MRI) features and diffusion-weighted imaging (DWI) parameters could help diagnose PAS in different placental locations.
We retrospectively enrolled 19 posterior PAS and 68 anterior PAS patients who had been diagnosed with PAS based on intraoperative findings and pathology. All the patients underwent DWI. DWI parameters calculated from different DWI models, and MRI features analyzed from conventional MRI images were compared. The discriminative ability of the DWI parameters and MRI features in diagnosing PAS in different placental locations were measured by receiver operating characteristic (ROC) curve and logistic regression analyses.
Diffusion coefficient (D) was selected from DWI to predict posterior PAS, and had a sensitivity of 84% and a specificity of 68%, while pseudo-diffusion coefficient (D*) was selected to predict anterior PAS, and had a sensitivity of 65% and a specificity of 86%. Of the MRI features, T2 dark bands and abnormal intraplacental vascularity were found to predict both anterior and posterior PAS. When the MRI features and DWI parameters were combined, the model had an area under the curve (AUC) of 0.903 [95% confidence interval (CI): 0.791-1], a sensitivity of 90%, and a specificity of 89% for posterior PAS, and an AUC of 0.858 (95% CI: 0.767-0.948), a sensitivity of 77%, and a specificity of 90.9% for anterior PAS.
D and D* were associated with posterior PAS and anterior PAS, respectively. A combination of morphological and functional markers may improve the diagnostic confidence of PAS in different placental locations.
胎盘植入谱系疾病(PAS)在胎盘后壁时通常难以诊断。本研究旨在确定某些磁共振成像(MRI)特征和扩散加权成像(DWI)参数是否有助于诊断不同胎盘位置的PAS。
我们回顾性纳入了19例后壁PAS患者和68例前壁PAS患者,这些患者均根据术中发现和病理诊断为PAS。所有患者均接受了DWI检查。比较了从不同DWI模型计算得出的DWI参数以及从常规MRI图像分析得出的MRI特征。通过受试者操作特征(ROC)曲线和逻辑回归分析来衡量DWI参数和MRI特征在诊断不同胎盘位置PAS中的鉴别能力。
从DWI中选择扩散系数(D)来预测后壁PAS,其灵敏度为84%,特异度为68%,而选择伪扩散系数(D*)来预测前壁PAS,其灵敏度为65%,特异度为86%。在MRI特征中,发现T2暗带和胎盘内血管异常可预测前壁和后壁PAS。当将MRI特征和DWI参数相结合时,该模型对于后壁PAS的曲线下面积(AUC)为0.903 [95%置信区间(CI):0.791 - 1],灵敏度为90%,特异度为89%;对于前壁PAS的AUC为0.858(95% CI:0.767 - 0.948),灵敏度为77%,特异度为90.9%。
D和D*分别与后壁PAS和前壁PAS相关。形态学和功能标志物的组合可能会提高不同胎盘位置PAS的诊断可信度。