Prediction of residual disease of spine infection using F-18 FDG PET/CT.

Study Design: Prospective study.
Objective: To determine whether F-18 fluoro-deoxy-glucose positron emission tomography/computed tomography (F-18 FDG PET/CT) follow-up imaging after treatment in patients with spinal infection (SI) could provide useful prognostic information and determine the residual SI.
Summary of Background Data: There are seldom data on the capability of follow-up imaging methods to predict residual disease and treatment efficacy in patients with SI.
Methods: Thirty patients with SI underwent F-18 FDG PET/CT at initial and during follow-up. From F-18 FDG PET/CT, quantitative indexes were obtained. The residual SI was determined by the presence of preoperative symptoms, hematological infection marker, and radiologic findings.
Results: The SUVmax were significantly declined after treatment in both of residual (2.85 +/- 1.17 vs. 2.06 +/- 1.03; P < 0.0001) and nonresidual SI (4.31 +/- 2.07 vs. 1.44 +/- 0.46; P < 0.0001). The SUVmean were also decreased after treatment in both of residual (1.45 +/- 0.45 vs. 1.04 +/- 0.29; P = 0.0014) and nonresidual SI (2.09 +/- 1.03 vs. 0.81 +/- 0.25; P < 0.0001). By lesion-based analysis, when < or =43.01% of %deltaSUVmax was used as threshold value, the area under curve (AUC) was 0.879 (P = 0.0001). The sensitivity and specificity were 85.7% and 82.6%, respectively. When < or =44.12% of %deltaSUVmean was used as threshold value, AUC was 0.828 (P = 0.0001). The sensitivity and specificity were 85.7% and 68%, respectively. In patient-based analysis, when < or =46.14% of %deltaSUVmax was used as threshold value, AUC was 0.904 (P = 0.0001). The sensitivity and specificity were 100% and 76.9%, respectively. When < or =41.78% of %deltaSUVmean was used as threshold value, AUC was 0.923 (P = 0.0001). The sensitivity and specificity were 100%, 76.9%, respectively.
Conclusion: F-18 FDG PET/CT is useful for discrimination of residual and nonresidual SI after treatment. Among the various quantitative indexes, %deltaSUVmax is a potent predictor of residual SI in the current study.