Objective: Bone scintigraphy has been advocated as a useful diagnostic tool for patients with suspected osteonecrosis or in screening for multifocal disease. We evaluated the sensitivity of bone scanning relative to magnetic resonance imaging (MRI) in the diagnosis of osteonecrosis.
Methods: Forty-eight patients presented with suspected osteonecrosis of the shoulder, hip, knee, or ankle. All patients underwent simultaneous (< 3 months apart) bone scans and MRI studies as part of diagnostic investigations. Histological confirmation of osteonecrosis was obtained for all suspected lesions. The diagnostic result for each imaging modality was then assessed and compared.
Results: All 163 (100%) histologically confirmed lesions were identified by MRI, while only 91 lesions (56%) were identified by bone scan. There was complete congruency of bone scans with MR images in only 38% of patients (18/48). Bone scanning identified 72% of lesions (47/65) in oligofocal patients (< or = 2 joints involved) compared with 45% of the lesions (44/98) in multifocal patients (> or = 3 joints involved). Sensitivity of lesions was highest for the knee and hip and lower for the shoulder and ankle. Larger and later-stage lesions had a higher bone scan sensitivity.
Conclusion: Our results demonstrated the low sensitivity of bone scintigraphy for diagnosing symptomatic osteonecrosis. It is least sensitive for early-stage lesions where it might be most useful to diagnose the disease. Our study also confirms that this test is less sensitive for joints other than the hip and is also not useful as a screening tool. Our study does not support the use of bone scans as a diagnostic or screening tool for osteonecrosis.