Inter-observer variability between readers in detection of bone metastases comparing different modalities; PET/CT, SPECT/CT and planar bone scintigraphy.

Abstract

To assess the inter-observer variability between readers in the detection of bone metastases for different modalities [planar bone scintigraphy (BS), single photon emission computed tomography/computed tomography (SPECT/CT), and fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (PET/CT) (F-18 FDG PET/CT)]. Patients with known primary tumors referred for metastatic workup either by F-18 FDG PET/CT or conventional planar BS and SPECT/CT were enrolled in this prospective study. The three modalities (BS, SPECT/CT, and PET/CT) were acquired for each patient. Interpretation was done by two independent nuclear medicine physicians separately and blindly; reader 1 (R1) and reader 2 (R2). A three-point subjective scale was used (1 = negative for bone metastases, 2 = equivocal, and 3 = positive). The findings were compared to the final patient status based on clinical and radiological follow-up for at least 6 months. The Kappa test was used to assess the degree of agreement between readers in their interpretation of each modality. A total of 54 patients (39 females and 15 males, ages 26-76; mean: 54.7 ± 12) were eligible for this study. Fair agreement 0.372 between R1 and R2 in interpretation of BS was noticed that improved to 0.847 after addition of SPECT/CT. Perfect agreement was achieved between R1 and R2 while interpreting PET/CT images (kappa = 0.964, P < 0.001). F-18 FDG PET/CT significantly improved the inter-observer variability between readers, allowing for better diagnosis of bone metastases in known cancer patients. Also, it was superior to BS and SPECT/CT in the detection of bone metastases.