Diagnosis of osteomyelitis. Accuracy and limitations of antigranulocyte antibody imaging compared to three-phase bone scan.

Thirty-seven patients with suspected osteomyelitis in conjunction with diabetic gangrene (N = 14, group 1), arthroplasty (N = 8, group 2), and various diseases (N = 15, group 3) were examined. Three-phase bone scans, followed by granulocyte imaging using I-123 labeled anti-NCA 95 monoclonal antibodies, were performed to evaluate and compare the diagnostic accuracy of both procedures. Final diagnosis was established histologically, bacteriologically, or by both methods either through the clinical course or by long-term follow-up in patients in group 1. Osteomyelitis was proven in 17 out of the 37 patients. Bone and antigranulocyte imaging demonstrated positive results in all patients with osteomyelitis (sensitivity 100% for each method). No signs of skeletal infection were found in 20 patients. Fifteen of these patients had no antigranulocyte antibody accumulation, resulting in 75% specificity. Ten patients without infection had normal three-phase bone imaging results (specificity 50%). Antigranulocyte imaging results were negative in 6 out of 10 patients without osteomyelitis in groups 2 and 3 whose bone imaging results were questionable. However, because of identical bone and granulocyte imaging results, no increase of diagnostic accuracy could be obtained by additional granulocyte imaging in patients with diabetic gangrene. Final diagnoses of false-positive antigranulocyte studies were aseptic osteonecrosis (N = 2), loosening of prostheses, gouty arthritis, and pain after arthrolysis. In summary, antigranulocyte antibody imaging offered high sensitivity and acceptable specificity for the diagnosis of osteomyelitis. Diagnostic accuracy can be improved through the adjuvant use of antigranulocyte imaging for patients with suspected osteomyelitis, especially when radiographic and scintigraphic results are questionable or unreliable.