Diagnostic performance of 18F-FDG PET/CT in patients with spinal infection: a systematic review and a bivariate meta-analysis

Abstract

Diagnosis of spinal infection (SI) is challenging and usually requires multiple tests. We aimed to perform a systematic review and a bivariate meta-analysis on the diagnostic role of 18F-FDG PET/CT in patients with SI. A comprehensive literature search of studies published through February 2019 in PubMed/MEDLINE and Cochrane library databases was carried out. Studies investigating the diagnostic performance of 18F-FDG PET/CT in patients with SI were eligible for inclusion in the qualitative analysis. For the quantitative analysis, pooled sensitivity, specificity, positive and negative likelihood ratio (LR+ and LR−) and diagnostic odds ratio (DOR) of 18F-FDG PET/CT in patients with suspected SI were calculated on a per examination-based analysis. Pooled data were presented with 95% confidence intervals (95% CI). Twenty-six articles (833 patients) using 18F-FDG PET/CT were eligible for the qualitative analysis. Twelve studies (396 patients) were selected for the meta-analysis. Overall, 18F-FDG PET/CT demonstrated a very good diagnostic performance in patients with SI and several studies underlined the value of 18F-FDG PET/CT in assessing the response to treatment. The bivariate meta-analysis on 18F-FDG PET/CT in patients with suspected SI provided the following results: sensitivity 94.8% (95% CI 88.9–97.6%) and specificity 91.4% (95% CI 78.2–96.9%). The pooled LR+, LR− and DOR were 4.7 (95% CI 2.9–7.7), 0.11 (95% CI 0.07–0.16) and 63.4 (95% CI 28.9–139), respectively. No significant heterogeneity or publication bias was found. 18F-FDG PET/CT demonstrated a very good diagnostic performance in patients with SI and can be used in patients in which MRI cannot be performed or is non-diagnostic or inconclusive. Several studies underlined the value of 18F-FDG PET/CT in assessing the response to treatment in patients with SI. Overall, larger multicentre and prospective studies and cost-effectiveness analyses are warranted.