The use of procalcitonin in the diagnosis of bone and joint infection: a systemic review and meta-analysis

Abstract

Only a few studies have investigated the use of PCT in the diagnosis of bone and joint infection, and these studies have had relatively small sample sizes. We performed a systematic review and meta-analysis of the diagnostic performance of serum procalcitonin (PCT) in the identification of osteomyelitis and septic arthritis in patients who present with fever and orthopedic symptoms. EMBASE, MEDLINE, and Cochrane databases and the reference lists of relevant articles were searched, with no language restrictions, through February 2012. All original studies that reported the use of serum PCT alone or in comparison with other biomarkers for diagnosis of osteomyelitis and septic arthritis were included. Seven studies qualified for inclusion. These studies enrolled a total of 583 patients with suspected bone or joint infection, 131 of whom had confirmed osteomyelitis or septic arthritis. Analysis of the PCT data indicated a bivariate pooled sensitivity of 0.67 (95 % CI: 0.37-0.88), specificity of 0.90 (95 % CI: 0.78-0.96), a positive likelihood ratio (LR+) of 6.48 (95 % CI: 2.28-14.6), and a negative likelihood ratio (LR-) of 0.37 (95 % CI: 0.16-0.84). Use of a lower PCT cut-off value (0.2-0.3 ng/mL) improved the LR + to 6.66 and the LR- to 0.15. Analysis of the three studies that also measured serum C-reactive protein (CRP) indicated that CRP had an LR + of 1.39 (95 % CI: 1.17-1.65) and an LR- of 0.40 (95 % CI: 0.12-1.36). Our results indicate that PCT may be more suitable as an aid for rule-in diagnosis rather than for exclusion of septic arthritis or osteomyelitis and that use of a lower cut-off value for serum PCT may improve its diagnostic performance.