妊娠期乳腺病灶超声BI-RADS分级的临床价值及活组织检查方式选择

Clinical significance of breast imaging reporting and data system-ultrasound in diagnosis and biopsy of breast lesions during pregnancy

目的 探讨采用乳腺影像报告与数据系统（BI-RADS）进行超声分级对妊娠期乳腺病灶的诊断价值及其对此期乳腺病灶活组织检查的指导意义,并探索妊娠期乳腺病灶合理的活组织检查方式。方法 回顾性分析2010年1月至2015年1月成都市妇女儿童中心医院收治的73例妊娠期乳腺病灶患者资料,对所有患者的超声结果进行BI-RADS分级,所有病灶均通过活组织检查获得组织学病理诊断,活组织检查方式包括空芯针穿刺活组织检查（44例）、真空辅助旋切活组织检查（23例）及开放手术切除活组织检查（4例）。以术后病理诊断为“金标准”,采用受试者工作特征（ROC）曲线来评价超声BI-RADS的诊断性能,计算超声BI-RADS分级结果的阳性似然比、阴性似然比、敏感度与特异度。采用χ2检验比较不同活组织检查方式并发症发生率。结果 全组73例患者共85个病灶,按照BI-RADS分级,2级病灶7个,3级病灶29个,均为良性病灶;4级病灶45个（癌7个）;5级病灶4个（癌4个）。ROC曲线下面积为0.952,95%置信区间为0.886-1.000。根据诊断结果遴选最优诊断节点为4B级,其诊断敏感度为83.33%（10/12）,特异度为98.63%（72/73）,阳性似然值为59.17,阴性似然值为0.17,所有病例的诊断正确率为96.47%（82/85）。空芯针穿刺活组织检查并发症发生率为4.55%（2/44）,明显低于真空辅助旋切活组织检查的34.78%（8/23）及开放切除活组织检查的3/6（P均〈0.017）。结论 超声BI-RADS分级对妊娠期乳腺病灶的诊断性能好,对活组织检查有一定的指导意义。空芯针穿刺活组织检查创伤较小,并发症较少,应作为妊娠期病灶活组织检查的首选方式。

Objective To explore the diagnostic value of breast imaging reporting and data system- ultrasound （BI-RADS US） for breast lesions in pregnant patients, and provide the guidance for the selection of biopsy method. Methods A total of 73 pregnant women with breast lesions in Chengdu Women ＆ Children＇ s Central Hospital from January 2010 to January 2015 were retrospectively analyzed in this study. All lesions were retrospectively categorized according to BI-RADS US. All patients were histopathologically diagnosed by core needle biopsy（ n = 44）, vaccum assisted biopsy （ n = 23 ） or open biopsy （ n = 4 ）. The gold standard was postoperative pathology, and ROC curve was used to evaluate the performance of BI-RADS US. The positive likelihood ratio, the negative likelihood ratio, the sensitivity and specificity of the ultrasonic BI-RADS category were calculated. Chi-square test was used to compare the incidence of complications using different biopsy methods. Results There were 85 lesions in 73 patients. According to BI-RADS US, there were 7 lesions in grade 2 （benign）, 29 in grade 3 （benign）, 45 in grade 4 （including 7 breast cancer） and 4 in grade 5 （4 breast cancer）. The area under curve（AUC） was 0. 952 （95% CI：O. 886-1. 000）. The selected optimal point was 4B based on diagnosis results, with the sensitivity of 83.33% （ 10/12）, the specificity of 98.63% （72/73）, the positive likelihood ratio of 59. 17 and the negative likelihood ratio of 0. 17. The diagnosticaccuracy of all patients was 96.47% （82/85）. The complication incidence of core needle biopsy was 4. 55% （2/44）, significantly lower than 34. 78% （8/23） of vacuum assisted biopsy and that of open biopsy （3/6） （both P〈0. 017）. Conclusions BI-RADS US has a high sensitivity and specificity for the diagnosis of breast lesions during pregnancy, which can provide the guidance for biopsy. With advantages of less trauma and less complications, the core needle biopsy should be the first choice for the biopsy of breast lesions in pregnant