Labelled White Cell Scanning for Infection

I would recommend combined labelled white cell scan and bone marrow (sulphur colloid) scan, which is the gold standard nuclear medicine technique for prosthetic joint infection with greater than 90% accuracy. The technique involves: (1) Labelling the patient's own WBCs with In-111 or Tc-99m HMPAO ex vivo, (2) Imaging at 4 and 24 hours to see where WBCs accumulate, (3) Injecting Tc-99m sulphur colloid to map bone marrow distribution, (4) Comparing the two scans spatially. Interpretation: If WBC uptake occurs WITHOUT matching marrow uptake at the same site (mismatch), this indicates infection. If both WBC and marrow uptake are present and matched, this indicates marrow hyperplasia (normal post-arthroplasty change) rather than infection. Plain bone scan is not helpful as it remains positive for years after arthroplasty.

In diabetic foot infection, the three-phase bone scan is sensitive (greater than 90%) but has low specificity because Charcot neuroarthropathy, fractures, and other diabetic changes also cause positive scans. WBC scanning can improve specificity in this setting. However, there are important limitations: (1) Complex foot anatomy makes spatial correlation difficult, (2) Charcot joints may show some WBC uptake due to inflammation, (3) Combined WBC/marrow technique is less validated in the foot compared to PJI. Despite these limitations, WBC scan can help differentiate osteomyelitis from Charcot by showing more focal, intense WBC uptake with osteomyelitis. However, MRI remains the preferred modality for diabetic foot osteomyelitis as it provides better anatomic detail and can identify abscess, extent of infection, and differentiate from Charcot based on pattern.

I would NOT recommend WBC scanning for suspected vertebral osteomyelitis. WBC scans have poor sensitivity in the spine due to: (1) Normal vertebral bone marrow shows variable, unpredictable WBC uptake, (2) Paradoxically, vertebral infection may appear 'cold' or photopenic rather than hot, (3) Interpretation is unreliable. For suspected spine infection, I would recommend: (1) MRI as first choice - highly sensitive (greater than 90%), excellent anatomic detail, can show disc involvement (key distinguishing feature from tumour), epidural abscess, and paraspinal extension. (2) If MRI contraindicated, FDG-PET/CT is an alternative - better performance than WBC scan in spine, can assess entire spine, and helps differentiate infection from degenerative changes. (3) CT-guided biopsy for microbiological diagnosis if imaging suggests infection.