FDG PET/CT is useful for the interim evaluation of response to therapy in patients affected by haematogenous spondylodiscitis.

Antibiotic therapy in patients affected by discitis is often empirical. Therefore, early evaluation of response to therapy is important. In many patients inflammatory indexes are low during all the phases of the diseases or are altered by concomitant diseases. The aim of the study was to assess the possible role of FDG PET/CT for the early evaluation of response to therapy in patients affected by infective discitis, in comparison to C-reactive protein (CRP) serum levels.Enrolled in the study were 38 patients diagnosed with haematogenous infective discitis. Of the 38 patients, 7 had tubercular infection, 1 fungal infection and 30 pyogenic discitis. Four patients were excluded because the second PET/CT scan was not performed. Thus 34 patients (18 women, mean age 64 years) were analysed. All the patients included underwent a FDG PET/CT scan and determination of CRP level at baseline and again 2 to 4 weeks after the start of therapy. The PET results in terms of SUV of the first and second scans (SUV1 and SUV2) and delta-SUVmax were compared to the inflammatory indexes and clinical status during therapy.The mean SUVmax at diagnosis was 8.6 ± 3.7. The mean CRP level at diagnosis was 3.8 ± 3.8 mg/dl. A progressive clinical response was seen in 26 patients and 8 patients showed no response. SUV1 was not correlated with the baseline CRP level (CRP1, p = 0.7) and SUV2 was not correlated with the CRP level at the time of the second scan (CRP2, p = 0.4). In responders, SUV2 and CRP2 were significantly lower than SUV1 and CRP1 (p < 0.0001 and p = 0.001, respectively). ROC curves for delta-SUVmax showed a sensitivity of 82 \% and a specificity of 82 \% with a cut-off of 34 \%. ROC curves for SUV2 showed a sensitivity of 83 \% and a specificity of 46 \% with a cut-off of 6.4. ROC curves for delta-CRP showed a sensitivity of 67 \% and a specificity of 89 \% with a cut-off of 74 \%. ROC curves for CRP2 showed a sensitivity of 65 \% and a specificity of 70 \% with a cut-off of 0.7 mg/dl. No statistically significant difference was found between delta-SUVmax AUC and delta-CRP AUC (p = 0.5).Delta-SUVmax provided a higher sensitivity and specificity for identifying responders. SUV2 provided comparable sensitivity, but significantly lower specificity. CRP level performed less well for identifying responders. There was no significant difference in the global performance of the two tests (delta-SUVmax AUC and delta-CRP AUC). However, the higher sensitivity of delta-SUVmax for the early identification of responders may have an important clinical impact in guiding antibiotic therapy especially in patients with a noninformative CRP test at diagnosis.