对比真空辅助乳腺活检（VABB）与空芯针穿刺活检（CNB）在评估乳腺癌诊断方面的综合效能。

采用回顾性多中心设计，收集2023年1月至2024年8月阿克苏地区第一人民医院与浙江省肿瘤医院收治的241例乳腺癌患者的临床资料。根据术前活检方式分为VABB组（119例）与CNB组（122例）。以术后病理检查结果为金标准，采用Kappa检验比较两种方法与病理诊断结果一致性。采用曲线下面积（AUC）和Delong检验评价两种方法诊断性能的差异。正态分布的计量资料组间比较采用独立样本t检验，非配对的计数资料组间比较采用χ²检验或Fisher确切概率法，配对的计数资料组间比较采用McNemar检验。

对三阴型，VABB与CNB的诊断结果与病理金标准相比，差异均无统计学意义（VABB：P=0.754；CNB：P=0.804），且 Kappa 值分别为 0.711和0.674（P均<0.001），一致性良好。对 HER-2阳性型，VABB与CNB的诊断结果与病理金标准相比，差异均无统计学意义（VABB：P=0.774；CNB：P=0.424），且Kappa 值分别为0.749和0.698（P均<0.001），一致性较高。对于 Luminal A 型和 Luminal B 型，两种活检方法的诊断结果与病理金标准相比，差异均有统计学意义（P均<0.05），且其中 VABB 的 Kappa 值分别为 0.232、0.162，CNB 的 Kappa 值分别为 0.425、0.374，一致性较低。在4种乳腺癌分子亚型中，CNB组与VABB组的AUC比较，差异均无统计学意义（P均>0.05）。Luminal B型与三阴型患者的AUC较高（Luminal B型：CNB组AUC=0.836，VABB组AUC=0.882；三阴型：CNB组AUC=0.831，VABB组AUC=0.843），诊断准确性较好。HER-2阳性型与Luminal A型的AUC相对偏低（HER-2阳性型：CNB组AUC=0.684，VABB组AUC=0.604；Luminal A型：CNB组AUC=0.757，VABB组AUC=0.624），诊断准确性有限。VABB组的样本完整性均为良好，在样本体积方面显著优于CNB组［（1.04±0.30） cm³比（0.10±0.02） cm³， t=-34.532，P<0.001］，活检成功率为100%，无重复活检。VABB组的肿瘤空间异质性阳性检出率显著高于CNB组（30.3%比18.9%，χ²=4.234，P=0.040）。与CNB组比较，VABB组术中出血量更高［（1.84±0.89） ml比（2.24±0.72） ml，t=3.749， P<0.001］，但术后血肿发生率更低（9.9%比0.8%，χ²=9.553， P=0.002），患者满意度评分更高［（6.16±0.80）分比（8.59±0.56）分，t=27.352，P<0.001］。

VABB在获取高质量组织样本、提高肿瘤空间异质性阳性检出率及患者满意度方面优于CNB，且整体安全性良好。

To compare the comprehensive efficacy of vacuum-assisted breast biopsy (VABB) and core needle biopsy (CNB) in the diagnostic evaluation of breast cancer.

A retrospective multicenter design was adopted. Clinical data of 241 breast cancer patients admitted to the First People's Hospital of Aksu Prefecture and Zhejiang Cancer Hospital from January 2023 to August 2024 were enrolled. According to the preoperative biopsy method, patients were divided into VABB group (n=119) and CNB group (n=122). Taking the postoperative pathological examination results as the gold standard, the Kappa test was used to compare the consistency of the two methods with the pathological diagnosis. Receiver operating characteristic (ROC) curve analysis and the Delong test were employed to evaluate the difference in diagnostic performance between the two methods. Continuous data with normal distribution were compared by independent sample t test;unpaired categorical data were compared by χ² test or Fisher's exact test;paired categorical data were compared by McNemar test.

No significant statistical differences were found between the diagnostic outcomes of VABB/CNB and pathological results (gold standard) for triple negative subtype (VABB: P=0.754; CNB: P=0.804). The corresponding Kappa values were 0.711 and 0.674 (both P<0.001), demonstrating good consistency with pathological diagnosis. For HER-2-positive subtype , no statistically significant differences were observed as well (VABB: P=0.774; CNB: P=0.424). The Kappa values were 0.749 and 0.698 (both P<0.001), reflecting favorable consistency. Statistically significant differences existed in Luminal A and Luminal B subtypes between the diagnostic outcomes of VABB/CNB and pathological results (both P<0.05). The Kappa values of VABB were 0.232 and 0.162, while those of CNB were 0.425 and 0.374, presenting poor consistency. In the four molecular subtypes of breast cancer, there was no statistically significant difference in AUC between CNB and VABB (Delong test, all P>0.05) . The AUC values for Luminal B and triple negative subtypes were relatively high (Luminal B: CNB group AUC=0.836, VABB group AUC=0.882; triple negative: CNB group AUC=0.831, VABB group AUC=0.843), indicating good diagnostic value. The AUC values for HER-2 positive and Luminal A subtypes suggested limited diagnostic accuracy (HER-2 positive: CNB group AUC=0.684, VABB group AUC=0.604; Luminal A: CNB group AUC=0.757, VABB group AUC=0.624) . Sample integrity in the VABB group was consistently good. The sample volume in the VABB group was significantly larger than that in the CNB group [(1.04±0.30) cm³vs (0.10±0.02) cm³, t=-34.532, P<0.001]. The biopsy success rate of 100% and no repeated biopsies. The positive detection rate of tumor spatial heterogeneity in the VABB group was higher than that in the CNB group (30.3% vs 18.9%, χ²=4.234, P=0.040). Intraoperative blood loss in the VABB group was significantly higher than that in the CNB group [(2.24±0.72) ml vs (1.84±0.89) ml, t=-3.749, P<0.001] , but the incidence of postoperative hematoma was lower (0.8% vs 9.9%, χ²=9.553, P=0.002) , and the patient satisfaction score was higher (8.59±0.56 vs 6.16±0.80, t=–27.352, P<0.001) .

VABB is superior to CNB in obtaining high-quality tissue samples, improving the positive detection rate of tumor spatial heterogeneity and patient satisfaction, with good overall safety.