放射学实践
放射學實踐
방사학실천
RADIOLOGIC PRACTICE
2015年
6期
660-665
,共6页
陈婷%张晶%张玉东%王小宁%洪汛宁
陳婷%張晶%張玉東%王小寧%洪汛寧
진정%장정%장옥동%왕소저%홍신저
胎盘%磁共振成像%诊断
胎盤%磁共振成像%診斷
태반%자공진성상%진단
Placenta%Magnetic resonance imaging%Diagnosis
目的:研究 MRI 在胎盘侵犯术前诊断中的价值。方法:回顾性分析产前疑似胎盘侵犯的46例孕妇 MRI 图像,两位阅片者分别记录最佳 MRI 序列、方位、图像质量、诊断信心、有无侵犯、侵犯深度以及特异性 MRI 征象;并对两者的结果进行一致性检验。以术后病理或手术记录作为金标准,分别计算 MRI 诊断胎盘侵犯、侵犯深度以及各个 MRI 征象的敏感性、特异性、阳性预测值和阴性预测值。结果:两者术前诊断达到中度至高度的一致性(κ值为0.478~0.739)。单次激发闭气序列(SSFSE)及矢状面是最佳的 MRI 序列和观察方位。MRI 诊断胎盘侵犯的敏感度和特异度为79.4%和66.7%;诊断胎盘粘连、植入和穿透的敏感度分别为28.4%,70.6%和33.3%。胎盘与肌层间低信号带缺失在胎盘侵犯中的出现率明显高于无胎盘侵犯(P =0.027),其胎盘侵犯的阳性预测值为83.95%,敏感度为76.5%。胎盘与肌层间低信号带缺失和 T2 WI 上结节状低信号带同时出现的阳性预测值达到92.3%,但敏感度仅为35.3%。胎盘与肌层间低信号带缺失和子宫下段膨隆同时出现的阳性预测值是100%,但敏感性仅为23.5%。结论:MRI 判断有无胎盘侵犯具有较高的敏感性和特异性。判断侵犯深度(粘连、植入和穿透)时敏感性偏低。MRI 征象判断胎盘植入具有较高的阳性预测值,但是敏感性偏低。
目的:研究 MRI 在胎盤侵犯術前診斷中的價值。方法:迴顧性分析產前疑似胎盤侵犯的46例孕婦 MRI 圖像,兩位閱片者分彆記錄最佳 MRI 序列、方位、圖像質量、診斷信心、有無侵犯、侵犯深度以及特異性 MRI 徵象;併對兩者的結果進行一緻性檢驗。以術後病理或手術記錄作為金標準,分彆計算 MRI 診斷胎盤侵犯、侵犯深度以及各箇 MRI 徵象的敏感性、特異性、暘性預測值和陰性預測值。結果:兩者術前診斷達到中度至高度的一緻性(κ值為0.478~0.739)。單次激髮閉氣序列(SSFSE)及矢狀麵是最佳的 MRI 序列和觀察方位。MRI 診斷胎盤侵犯的敏感度和特異度為79.4%和66.7%;診斷胎盤粘連、植入和穿透的敏感度分彆為28.4%,70.6%和33.3%。胎盤與肌層間低信號帶缺失在胎盤侵犯中的齣現率明顯高于無胎盤侵犯(P =0.027),其胎盤侵犯的暘性預測值為83.95%,敏感度為76.5%。胎盤與肌層間低信號帶缺失和 T2 WI 上結節狀低信號帶同時齣現的暘性預測值達到92.3%,但敏感度僅為35.3%。胎盤與肌層間低信號帶缺失和子宮下段膨隆同時齣現的暘性預測值是100%,但敏感性僅為23.5%。結論:MRI 判斷有無胎盤侵犯具有較高的敏感性和特異性。判斷侵犯深度(粘連、植入和穿透)時敏感性偏低。MRI 徵象判斷胎盤植入具有較高的暘性預測值,但是敏感性偏低。
목적:연구 MRI 재태반침범술전진단중적개치。방법:회고성분석산전의사태반침범적46례잉부 MRI 도상,량위열편자분별기록최가 MRI 서렬、방위、도상질량、진단신심、유무침범、침범심도이급특이성 MRI 정상;병대량자적결과진행일치성검험。이술후병리혹수술기록작위금표준,분별계산 MRI 진단태반침범、침범심도이급각개 MRI 정상적민감성、특이성、양성예측치화음성예측치。결과:량자술전진단체도중도지고도적일치성(κ치위0.478~0.739)。단차격발폐기서렬(SSFSE)급시상면시최가적 MRI 서렬화관찰방위。MRI 진단태반침범적민감도화특이도위79.4%화66.7%;진단태반점련、식입화천투적민감도분별위28.4%,70.6%화33.3%。태반여기층간저신호대결실재태반침범중적출현솔명현고우무태반침범(P =0.027),기태반침범적양성예측치위83.95%,민감도위76.5%。태반여기층간저신호대결실화 T2 WI 상결절상저신호대동시출현적양성예측치체도92.3%,단민감도부위35.3%。태반여기층간저신호대결실화자궁하단팽륭동시출현적양성예측치시100%,단민감성부위23.5%。결론:MRI 판단유무태반침범구유교고적민감성화특이성。판단침범심도(점련、식입화천투)시민감성편저。MRI 정상판단태반식입구유교고적양성예측치,단시민감성편저。
Objective:To evaluate the preoperative diagnostic value of MRI in placenta invasion.Methods:The MRI features of 46 patients suspected to have placenta invasion were analyzed retrospectively.The optimal sequence and scanning plane of MRI,image quality,presence of placental invasion or not,the depth of invasion and the specificity of MRI features were blindly and independently reviewed by two observers,and the results underwent consistent test.The sensitivity,speci-ficity,positive predictive value (PPV)and negative predictive value (NPV)of the placental invasion,the depth of invasion (accreta/increta/percreta)and the specialized MRI features of placenta invasion were calculated respectively according to the pathological or surgical records.Results:The interobserver agreement of two readers achieved a high level of consisten-cy,and K value was from 0.478 to 0.739.The ideal sequence and observation plane is breath hold single shot fast spin echo sequence (SSFSE)and Sagittal plane.The sensitivity and specificity of MRI for diagnosis of placenta invasion was 79.4%and 66.7%.The sensitivity for diagnosis of the depth of placenta invasion was 28.4%,70.6% and 33.3% for accreta,in-creta and percreta respectively.The focally interrupted interface between placenta and myometrium was more observed in patients with placenta invasion than patients without placenta invasion (P =0.027).The PPV and sensitivity of focally in-terrupted interface between placenta and myometrium was 83.95% and 76.5%,and the combination with the low nodular signal intensity bands on T2 WI improved the PPV to 92.3%,however,the sensitivity was only 35.3%.The PPV of combi-nation of focally interupted interface between placenta and myometrium with bulging of the lower uterus segment improved to 100%;however,the sensitivity was only 23.5%.Conclusion:MRI appears to have good sensitivity and specificity for the diagnosis of placenta invasion,nevertheless it shows low sensitivity for the detection of the depth of placenta invasion (ac-creta/increta/percreta).MRI features for the detection of placenta invasion have a high PPV,but low sensitivity.