Quantitative assessment of active charcot neuro-osteoarthropathy in diabetes using 18F-sodium fluoride PET/CT.

BACKGROUND

No single imaging modality currently provides quantitative assessment of disease activity in active charcot neuro-osteoarthropathy (CNO) of the foot, either at diagnosis or during follow-up after clinical remission. This study evaluates the role of 18F-sodium fluoride positron emission tomography-computed tomography (18F-NaF PET/CT) in active CNO.

METHODS

In this prospective study, 80 patients with diabetes presenting with asymmetric foot swelling were screened and assessed using an infrared thermometer. 39 symptomatic patients were included (33 active CNO and 6 non-CNO) and underwent magnetic resonance imaging (MRI) and 18F-NaF PET/CT. 20 additional participants (10 diabetic neuropathy and 10 healthy controls) underwent PET/CT for assessing the uptake in diabetic patients and normal individuals. Standardized uptake value maximum (SUVmax) on 18F-Sodium Fluoride PET-CT was measured at the region of interest (ROI) of affected foot at baseline, clinical remission, and three-monthly post-remission. Change in SUVmax from baseline to clinical remission was compared between patients with sustained remission and those with relapse. Exploratory univariate Cox regression analysis was performed to assess associations between SUVmax changes and recurrence.

RESULTS

Median SUVmax (baseline) was 55.1 in active CNO versus 7.5 in non-CNO (p<0.001). An SUVmax value of ≥10.3 yielded 100% sensitivity (95% CI 89.4-100%) and 92.3% specificity (95% CI 74.9-99.1%) against all controls (AUC 0.988, 95% CI 0.965-1.000); specificity was 66.7% against symptomatic non-CNO patients. 29 participants achieved clinical remission, and SUVmax decreased from 58.3 to 19.6 (61.9% reduction; p<0.001). 4 patients with active CNO relapse showed significantly lower SUVmax reduction than those with sustained remission (28.4% vs 64.7%, p=0.006). In exploratory analysis, greater reduction in SUVmax between baseline and clinical remission was associated with lower recurrence risk (HR 0.970 per 1% greater reduction, 95% CI 0.941-0.999). With only four events, this finding is hypothesis-generating.

CONCLUSION

18F-NaF PET/CT shows promise as a quantitative marker of bone remodelling in active CNO in patients with diabetes. These findings are exploratory and require external validation in large cohorts.