In this study we evaluated the significance of false positive screening bone scintigraphy (BS) in primary invasive breast cancer patients. Lymphatic vessel invasion (LVI), estrogen receptor (ER), progesterone receptor (PR), nuclear grade, histology grade, epidermal growth factor receptor (EGFR) and C-erb-B2 values were examined in terms of their abilities to predict the accuracy of abnormal B5. We also examined the incidence of bone metastasis in primary invasive breast cancer patients according to the 1988 and 2003 AJCC classifications.

A retrospective review was performed on 2,044 primary invasive breast cancer patients that had received BS screening, and who were treated by mastectomy or breast conserving surgery at the 5eoul National University Hospital between Jan 1995 and Jul 2003. Abnormal screening B5 results were divided into "less suspicious" and "highly suspicious" groups. Patient's stages according to the 1988 AJCC classification were reclassified according to the 2003 AJCC classification. Bone metastasis was confirmed by further radiological examination or follow-up B5. All statistical analyses were two-tailed.

The incidences of bone metastasis and an abnormal screening BS result were 1.7% (35/2,044) and 13.8% (283/2,044), respectively. The false positive rate of screening B5 was 87.6% (248/283), LVI was the only significant predictive factor of bone metastasis in 283 of the abnormal BS patients (p < .001). c-erb-B2 showed no sig nificance to predict bone metastasis in the "less suspicious" group, but was Bone is the most common site of distant metastasis in invasive breast cancer at the time of primary diagnosis. The vertebrae are the most common sites of bone metastasis and the ribs, skull, sternum and proximal long bones are also frequently involved. Bone metastases affect 8% of patientsmarginally significant in the "highly suspicious" group (p = .046). ER, PR, nuclear grade, histology grade, and EGFR showed no significance in terms of predicting the accuracy of an abnormal B5 result. The incidences of bone metastasis were 0.6, 1.3 and 7.6% in stages I, II and III, respectively, according to the 1988 AJCC classification, while these incidences were 0.6, 0.7 and 5.8% according to the 2003 AJCC classification.

The use of screening bone scintigraphy as a routine screening test is hard to justify due to its high false positive rate. LVI may be a useful factor in that it predicts the accuracy of an abnormal BS result. The incidences of bone metastasis in stages II and III were lower for the 2003 AJCC staging system.