Abstract

OBJECTIVE. The purpose of this article is to assess the feasibility and utility of PET/CT in distinguishing benign from malignant pulmonary nodules in patients with solid childhood malignancies. SUBJECTS AND METHODS. This prospective study was conducted between March 2008 and August 2010. We enrolled 25 subjects 21 years old or younger with solid childhood malignancies and at least one pulmonary nodule measuring 0.5-3.0 cm. PET/CT was performed within 3 weeks of diagnostic chest CT. Three panels of three reviewers each reviewed diagnostic CT only (panel 1), PET/CT only (panel 2), or diagnostic CT and PET/CT concurrently (panel 3) and predicted each nodule's histologic diagnosis as benign, malignant, or indeterminate. Interreviewer agreement was assessed with the kappa statistic. Using nodule biopsy or clinical follow-up as reference standards, the sensitivity, specificity, and accuracy for each panel was assessed. Logistic regression was used to assess the nodule's maximum standardized uptake value (SUVmax) association with its histologic diagnosis. RESULTS. There were 75 nodules with a median size of 0.74 cm (range, 0.18-2.38 cm); 48 nodules were malignant. Sensitivity was 85% (41/48) for panel 1, 60% (29/48) for panel 2, and 67% (32/48) for panel 3. All panels had poor specificities. Interreviewer agreement was moderate for panel 1 (0.43) and poor for panels 2 (0.22) and 3 (0.33). SUVmax was a significant predictor of histologic diagnosis (p = 0.004). CONCLUSION. PET/CT assessment of pulmonary nodules is feasible in children with solid malignancies but may not reliably improve our ability to predict a nodule's histologic diagnosis. The SUVmax may improve the performance of PET/CT in this setting.

View details for DOI 10.2214/AJR.12.10205

View details for Web of Science ID 000327501500013

View details for PubMedID 24261397

View details for PubMedCentralID PMC4276039