As bone metastases might be present in lung cancer despite a normal bone scan, we examined various alternatives prospectively. Positron emission tomography using F-18 sodium fluoride (PET) and single photon emission tomography (SPECT) were more sensitive than a planar bone scan. PET was more accurate with a shorter examination time than SPECT but had higher incremental costs.
Introduction: Previous studies have shown that vertebral bone metastases not seen on planar bone scans may be present on F-18 fluoride positron emission tomography (PET) scan or single photon emission computed tomography (SPECT). The purpose of this study was to measure the accuracy, clinical value and cost-effectiveness of tomographic bone imaging.
Materials and methods: A total of 103 patients with initial diagnosis of lung cancer was prospectively examined with planar bone scintigraphy (BS), SPECT of the vertebral column and PET using F-18 sodium fluoride (F-18 PET). Receiver operating characteristic (ROC) curve analysis was used for determination of the diagnostic accuracy. A decision-analysis model and the national charge schedule of the German Hospital Association were used for determination of the cost-effectiveness.
Results: Thirteen of 33 patients with bone metastases were false negative on BS, 4 on SPECT, and 2 on F-18 PET. The area under the ROC curve was 0.771 for BS, 0.875 for SPECT, and 0.989 for F-18 PET (p < 0.05). As a result of SPECT and F-18 PET imaging, clinical management was changed in 8 (7.8%) and 10 (9.7%) patients. Compared with BS, the costs per additional correctly diagnosed patient were 1272 Euro with SPECT and 2861 Euro with F-18 PET. The threshold for the costs of F-18 PET being more cost-effective than SPECT was 345 EUR.
Conclusion: Routine performance of tomographic bone imaging improves the therapeutic strategy because of detection of otherwise missed metastases. F-18 PET is more effective than SPECT but is associated with higher incremental costs.