C-reactive protein misdiagnoses delayed postoperative spinal implant infections in patients with low-virulent microorganisms

Abstract

Purpose

C-reactive protein (CRP) has been shown to be a powerful parameter for detecting acute postoperative spinal implant infections (PSII) with a high sensitivity and specificity. However, little data are available on the performance of CRP in the diagnosis of delayed PSII. The aim of the current study was therefore to establish cutoff values for diagnosing delayed infection based on serum CRP.

Methods

All patients who underwent a revision surgery after instrumented spinal fusion from January 2013 through January 2016 were included. Demographic data, laboratory values, type of infection (including microbiological and pathological results), comorbidities and clinical manifestation were collected. The European Bone and Joint Infection Society criteria, proposed to diagnose periprosthetic joint infection, were used to diagnose PSII.

Results

A total of 257 patients were included. PSII was diagnosed in 61 patients, representing 24% of the study cohort. There was a significant difference in serum CRP levels between septic and aseptic cohorts (19.3 vs. 4.8 mg/l, p < 0.001). However, 26 patients (43%) from the PSII group had a normal (< 5 mg/l) serum CRP level prior to revision surgery. According to the ROC curve, a serum CRP threshold of 4.05 mg/l had a sensitivity of 64% and specificity of 68%. The most common isolated microorganism was Propionibacterium spp. followed by coagulase-negative staphylococci.

Conclusion

Serum CRP showed low sensitivity and specificity for diagnosis of delayed PSII, even after applying cutoffs optimized by using receiver operating curve analysis, because of the high incidence of low-virulent pathogens.

Graphical abstract

These slides can be retrieved under Electronic Supplementary Material.

http://bit.ly/2HNDBVU