|
|
|
| Diagnostic value of MRI on placental invasion and clinical value of the topography classification |
| ZENG Si-hui, LIU Hong-sheng, QIN Huan-di, CHEN Xi-wen, CHEN Yuan-kai, MIAO Ding-qiang, DING Xiao-fei |
| Guangzhou Women and Children’s Medical Center, Guangzhou 510000, China |
|
|
|
|
Abstract Objective: To analyze MRI features of placental invasion and evaluate the clinical value of MR topography classification. Method: Records from December 2009 to May 2014 were reviewed to identify the accuracy of MRI in the diagnosis of placental invasion in the patients with a diagnosis of placenta previa and suspected placental invasion by ultrasound. Sensitivity, specificity, PPV and NPV were calculated on each MRI feature of placental invasion including myometrial attenuation, loss of placental-myometrial interface, T2WI dark bands and outward protruding of the uterus. All final diagnosis was based on surgical and pathological findings. MR topography classification was carried out on confirmed cases according to the topography of placental invasion. Over 50%, no more than 50% of uterine axial circumference and parametrial invasion were categorized as invasion of category A, B and C. Success rates of conservative treatment were compared by χ2 test between each category. Result: Of the 69 suspected cases, 46 had confirmed placenta invasion. MRI accurately predicted placenta invasion in 43 cases(sensitivity 93.48%) and 21 cases were correctly ruled out(specificity 91.30%). Loss of placental-myometrial interface was of the highest sensitivity and NPV(95.56%, 90.0%), followed by myometrial attenuation. T2WI dark bands was of high PPV(90.01%), and outward protruding of the uterus was of the highest specificity and PPV(100%). Success rates of conservative treatment were 100%(category A), 23.08%(category B) and 12.5%(category C) with statistical difference(P<0.05). Conclusion: Diagnostic accuracy of MRI is high in evaluating placental invasion. Loss of placental-myometrial interface is the prime feature in assessment. MR topography classification helps to determine conservative treatment.
|
|
Received: 13 October 2014
|
|
|
|
|
|
|
|
|
2015, Vol. 26 
