基于第五版BI-RADS对乳腺4类钙化恶性风险分层的应用研究

Application of the Fifth Edition BI-RADS in the Risk Stratification of Breast Category 4 Calcification Malignancy

目的探讨乳腺X线摄影中BI-RADS 4类钙化影像特征结合临床特征预测乳腺高危病变恶性风险分层的价值。方法回顾性分析经乳腺X线摄影检查发现的BI-RADS 4类钙化376例,由乳腺诊断医师在未知病理结果的情况下依据第五版BI-RADS对这些钙化的形态和分布进行描述并给出最终分类,以病理结果为金标准,计算各形态和各分布及组合分类的阳性预测值(PPV)。再结合患者的相关临床因素,分析影响BI-RADS 4类钙化恶性风险的因素。结果376例患者中包括193例恶性病灶,183例良性病灶,PPV为51.3%。其中无定形、粗糙不均质、细小多形性和细线样及细分支状钙化的PPV分别为34.0%、38.1%、81.4%和95.2%。区域性、团簇状、线样和段样分布的钙化的PPV分别为26.0%、46.5%、90.0%和89.6%。综合考虑钙化的形态和分布后,区域性分布的无定形钙化的PPV降为13.0%,符合BI-RADS 4B类;线样和段样分布的无定形钙化的PPV为100.0%和75.0%,符合BI-RADS 5类和4C类;区域性分布的粗糙不均质钙化的PPV为0%,符合BI-RADS 2类;段样分布的粗糙不均质钙化的PPV为88.9%,符合BI-RADS 4C类。多因素Logistic回归显示,绝经状态为已绝经、钙化形态为细小多形性或细线样及细分支状、钙化分布为团簇状或段样会增加4类钙化的恶性风险。结论在对BI-RADS 4类钙化的诊断中,钙化形态为细小多形性或细线样及细分支状、钙化分布为团簇状或段样会提高4类钙化的恶性风险,结合患者的临床特征能够更准确的对BI-RADS 4类钙化进行恶性风险分层。

Objective To investigate the value of BI-RADS category 4 calcification image features in mammography combined with clinical features to predict malignant risk stratification of high-risk lesions in the breast.Methods 376 cases of BI-RADS category 4 calcifications detected during mammography examinations were retrospectively analyzed,and the morphology and distribution of these calcifications were described and given a final classification by a breast diagnostician based on the fifth edition of BI-RADS without known pathological findings,using pathological findings as the gold standard.The positive predictive value(PPV)was calculated for each morphology and each distribution and combination.The factors affecting the risk of malignancy of BI-RADS category 4 calcifications were then analyzed in relation to the relevant clinical factors of the patients.Results The PPV was 51.3%in 376 patients including 193 malignant lesions and 183 benign lesions.Among them,the PPV of amorphous,coarse heterogeneous,fine pleomorphic and fine linear or branching calcifications were 34.0%,38.1%,81.4%and 95.2%,respectively.The PPV of regional,grouped,linear and segmental distribution of calcifications was 26.0%,46.5%,90.0%and 89.6%,respectively.After considering the morphology and distribution of calcifications together,the PPV of amorphous calcifications with regional distribution decreased to 13.0%,conforming to category 4B;the PPV of amorphous calcifications with linear and segmental distribution was 100.0%and 75.0%,conforming to categories 5 and 4C,the PPV of coarse heterogeneous calcifications with regional distribution was 0%,conforming to category 2.The PPV of segmental distribution of coarse heterogeneous calcification was 88.9%,which was consistent with category 4C.Multivariate logistic regression showed that menopausal status as menopausal,calcification morphology as fine pleomorphic or fine linear or branching,and calcification distribution as grouped or segmental increased the risk of malignancy for category 4 calcifications.Conclusion In the diagnosis of BI-RADS category 4 calcifications,the calcification morphology is fine pleomorphic or fine linear or branching,and the distribution of calcifications is grouped or segmental can increase the risk of malignancy of category 4 calcifications,and the combination of clinical characteristics of patients can more accurately stratify the risk of malignancy for BI-RADS category 4 calcifications.