Bosch Anne M, Kessels Alfons G H, Beets Geerard L, Rupa Jan D, Koster Dick, van Engelshoven Jos M A, von Meyenfeldt Maarten F
Department of Surgery, Maastricht University Hospital, P. Debyelaan 25, Maastricht, Netherlands.
Eur J Radiol. 2003 Dec;48(3):285-92. doi: 10.1016/s0720-048x(03)00081-0.
The clinical breast tumour size can be assessed preoperatively by physical examination, mammography and ultrasound. At present it is not clear which modality correlates best with the histological invasive breast tumour size. This prospective study aims to determine the most accurate clinical method (physical examination, mammography or ultrasound) to predict the histological invasive tumour size preoperatively.
Between October 1999 and August 2000, 96 women with 105 invasive malignant breast tumours were included in this study. All patients underwent excision and the tumour size was measured on histology. Tumour size was measured by all three modalities in 73 cases. Results were evaluated by calculating correlation coefficients. The examination modalities presenting the best estimation of the pathological tumour size were used in a stepwise linear regression analysis to construct a formula predicting the pathological tumour size from the result of the various diagnostic modalities.
The correlation coefficient between ultrasound and pathological size (r=0.68) was significantly better than the correlations between physical examination and pathological size (r=0.42) and mammographic and pathological size (r=0.44). Physical examination overestimates and ultrasound underestimates breast tumour classification. The most accurate prediction formula was: Pathological tumour size (mm) equals sonographic tumour size (mm)+3 mm.
When comparing physical examination, mammography and ultrasound for the prediction of the pathological size of a malignant breast tumour, ultrasound is the best predictor. The ensuing regression formula determines pathological size as tumour size by ultrasound+3 mm. However, with the wide 95% confidence interval of +/-11 mm, it remains difficult to predict the exact pathological size for an individual invasive breast tumour. A small deviation in millimetres of the tumour size could lead to a change in treatment and to another prognostic estimate.
临床乳腺肿瘤大小可通过体格检查、乳房X线摄影及超声检查在术前进行评估。目前尚不清楚哪种检查方式与组织学上浸润性乳腺肿瘤大小的相关性最佳。这项前瞻性研究旨在确定术前预测组织学浸润性肿瘤大小的最准确临床方法（体格检查、乳房X线摄影或超声检查）。
1999年10月至2000年8月期间，本研究纳入了96名患有105个浸润性恶性乳腺肿瘤的女性。所有患者均接受了肿瘤切除，并在组织学上测量了肿瘤大小。73例患者的肿瘤大小通过所有三种检查方式进行了测量。通过计算相关系数来评估结果。将对病理肿瘤大小估计最佳的检查方式用于逐步线性回归分析，以构建一个根据各种诊断方式的结果预测病理肿瘤大小的公式。
超声与病理大小之间的相关系数（r = 0.68）显著优于体格检查与病理大小之间的相关系数（r = 0.42）以及乳房X线摄影与病理大小之间的相关系数（r = 0.44）。体格检查会高估，而超声会低估乳腺肿瘤分级。最准确的预测公式为：病理肿瘤大小（毫米）等于超声肿瘤大小（毫米）+ 3毫米。
在比较体格检查、乳房X线摄影和超声检查对恶性乳腺肿瘤病理大小的预测时，超声是最佳预测指标。由此得出的回归公式将病理大小确定为超声测量的肿瘤大小+ 3毫米。然而，由于95%的置信区间较宽，为±11毫米，仍然难以预测单个浸润性乳腺肿瘤的确切病理大小。肿瘤大小几毫米的微小偏差可能会导致治疗方案的改变以及另一种预后评估。
