Novel Magnetic Resonance Imaging (MRI) Features Predicting Relapse After Clinical Remission in Pyogenic Spondylodiscitis: A Study Using Simultaneous ¹⁸F-Fluorodeoxyglucose Positron Emission Tomography/MRI Data
Seongmin Jeon, Heesung Yang, Eunjung Kong, Dongwoo Yu, Atman Desai, Vivek Sanker, You Seon Song, Ikchan JeonObjective: This study aimed to identify magnetic resonance (MR) imaging features that can predict relapse after clinical remission in pyogenic spondylodiscitis (PSD) based on simultaneous ¹⁸F-fluorodeoxyglucose positron emission tomography/MR (¹⁸F-FDG PET/MR) imaging data.Methods: This retrospective study included 101 patients (67 men and 34 women) with thoracic, lumbar, and/or sacral PSD who received at least 4 weeks of parenteral antibiotic therapy. All patients underwent ¹⁸F-FDG PET/MR imaging and blood inflammatory markers at the time of clinical remission and antibiotic discontinuation, and were followed for at least 12 months to detect relapse. Clinical variables and ¹⁸F-FDG PET/MR imaging features associated with relapse were analyzed.Results: Relapse occurred in 21 patients (20.8%) and was significantly associated with PSD lesions confined to the thoracic spine (p=0.016), a higher maximum standardized ¹⁸F-fluorodeoxyglucose uptake value (p=0.048), and higher mean signal intensity of edema at the PSD lesion on MR imaging (p=0.005). Regarding Contrast-Edema Mismatch (relative reduction in the edematous area compared with contrast-enhanced area, grades I–III), grade I (<25% difference) was significantly more frequent than grade III (>50% difference) in patients with relapse (p=0.016). Additionally, the mean signal intensity of the contrast area remained similar across grades (p=0.320), whereas the mean signal intensity of edema decreased progressively with increasing grade (p<0.001).Conclusion: Persistent and extensive edematous change on MR imaging was significantly associated with relapse after clinical remission in PSD, and may serve as a potential imaging marker for monitoring treatment response and predicting relapse.