Purpose: The aim of this study was to investigate the added value of single-photon emission computed tomography (SPECT)/spiral computed tomography (CT) versus SPECT alone in the differential diagnosis of solitary spinal lesions in patients with extraskeletal malignancies.
Materials and methods: A total of 90 patients who had a solitary spine ‘hot spot’ that could not be definitively diagnosed using planar scintigraphy were enrolled in the study. SPECT/spiral CT was performed on the indeterminate lesions in the spine. Images were independently interpreted by two experienced nuclear medicine physicians. Each spinal lesion was graded on a four-point diagnostic scale (1, benign; 2, likely benign; 3, likely bone metastasis; 4, bone metastasis). The final diagnosis of each lesion was based on pathological confirmation or follow-up. κ scores were used to evaluate inter-reviewer agreement and agreement of the SPECT and SPECT/spiral CT diagnoses with the final diagnosis.
Results: Final diagnoses revealed 25 bone metastases and 65 benign lesions. Forty percent (36/90) of the solitary spinal lesions were diagnosed as equivocal (likely benign or likely bone metastasis) according to SPECT, whereas only 5.6% (5/90) were diagnosed as equivocal according to SPECT/spiral CT. For SPECT/spiral CT-based and SPECT-based diagnoses, the κ scores for inter-reviewer agreement were 0.889 (P<0.001, 95% confidence interval 0.824–0.954) and 0.504 (P<0.001, 95% confidence interval 0.401–0.607). The diagnostic accuracies of SPECT/spiral CT and SPECT images were 91.1% (82/90) and 58.9% (53/90), respectively (χ2=24.919, P<0.001).
Conclusion: Compared with SPECT imaging alone, SPECT/spiral CT imaging was more accurate and valuable in the differential diagnosis of solitary spinal lesions in patients with extraskeletal malignancies.