The aim of the present study was to investigate if 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) outperforms conventional imaging techniques for excluding distant metastases prior to neoadjuvant chemotherapy (NAC) treatment in patients with stage II and III breast cancer. Second, we assessed the clinical importance of false positive findings. One hundred and fifty four patients with stage II or III breast cancer, scheduled to receive NAC, underwent an 18F-FDG PET/CT scan and conventional imaging, consisting of bone scintigraphy, ultrasound of the liver, and chest radiography. Suspect additional lesions at staging examination were confirmed by biopsy and histopathology and/or additional imaging. Metastases that were detected within 6 months after the PET/CT scan were considered evidence of occult metastasis, missed by staging examination. Forty-two additional distant lesions were seen in 25 patients with PET/CT and could be confirmed in 20 (13%) of 154 patients. PET/CT was false positive for 8 additional lesions (19%) and misclassified the presence of metastatic disease in 5 (3%) of 154 patients. In 16 (80%) of 20 patients, additional lesions were exclusively seen with PET/CT, leading to a change in treatment in 13 (8%) of 154 patients. In 129 patients with a negative staging PET/CT, no metastases developed during the follow-up of 9.0 months. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of PET/CT in the detection of additional distant lesions in patients with stage II or III breast cancer are 100, 96, 80, 100, and 97%, respectively. FDG PET/CT is superior to conventional imaging techniques in the detection of distant metastases in patients with untreated stage II or III breast cancer and is associated with a low false positive rate. PET/CT may be of additional value in the staging of breast cancer prior to NAC.