Background: Bone scanning (BS), liver ultrasonography (LUS) and chest radiography (CXR) are commonly used in patients with newly diagnosed breast cancer as part of baseline staging. However, in the absence of symptomatic disease, the usefulness of this routine diagnostic work-up is not evidence-based.
Methods: We selected the study sample from 516 consecutive patients with newly diagnosed invasive breast cancer. For each diagnostic test (BS, LUS, CXR), we analyzed the prevalence defined as the number of patients with diagnosis of metastatic disease after an imaging technique divided by the total number of patients tested. In addition, sensitivity and specificity were calculated. Initial suspicion was confirmed by other independent tests (bone X-ray, computerized tomography scan, magnetic resonance imaging) in order to identify "true" positive diagnoses.
Results: At baseline, BS was carried out in 412 patients, LUS in 412 patients and CXR in 428 patients. Thirty-three patients were correctly diagnosed by the initial staging investigations as having metastatic disease (true positive cases). BS detected skeletal metastases in 6.31% of patients, LUS detected liver metastases in 0.72% of patients and CXR detected lung metastases in 0.93% of patients. Before imaging tests, all patients with either LUS or CXR evidence of metastases were previously classified as having stage III disease. On the other hand, only 26.9% of bone metastases were detected in patients with stage III. Accordingly, the detection rate in stage III patients was 14%, 5.6% and 7.2%, respectively for BS, LUS and CXR.
Conclusions: These findings indicate that a complete diagnostic work-up to detect metastases is unnecessary in the majority of patients with newly diagnosed breast cancer, whereas it may be indicated for specific patient categories such as those with stage III disease.