Objective: To evaluate the usefulness of 99Tc-ciprofloxacin scintigraphy (CFS) in patients with hip or knee arthroplasty and suspected infection.
Methods: Forty patients (26 women, 14 men) with a mean age of 66±10 years and local pain in the hip (21), knee (16), or shoulder (three) prosthesis were recruited. CFS was performed at 1, 4, and 24 h after intravenous injection of 370 MBq. Anterior and posterior views centered on the affected joint were performed in all patients. A routine bone scan, 99Tc-hexamethylpropyleneamine oxime leukocyte scan, and 99Tc-colloid scan [leukocyte scintigraphy-bone marrow scintigraphy (LS-MS)] were performed. Final diagnosis of infection was confirmed by positive microbiological analysis or macroscopic evidence of purulent material.
Results: Diagnosis of arthroplasty infection was established in 16 out 40 cases: coagulase-negative staphylococci (nine), Staphylococcus aureus (three), Enterococcus (one), and macroscopic infection in the remaining three cases. CFS imaging showed the 24-h image to be the best acquisition time-point. The sensitivity, specificity, negative predictive value, and positive predictive value for LS-MS were 75, 92, 86, and 85%, whereas for CFS at 24 h these figures were 88, 71, 67, and 89%. The sensitivity and specificity for LS-MS and for CFS at 24 h for hip were (74, 90, and 88, 85%) and for knee (83, 90 and 100, 50%).
Conclusion: CFS can be useful in the diagnosis of arthroplasty infection of the hip as a substitute for LS-MS. It is recommended that CFS images be obtained 24 h after injection. The lack of specificity of CFS makes this technique inadequate for knee prostheses in this series.