Usefulness of sonication in the microbiological diagnosis of cardiovascular implantable electronic device infections: systematic review, meta-analysis and meta-regression.

Abstract

Background: Multiple studies have demonstrated the utility of sonication to improve culture yield in patients with cardiovascular implantable electronic device (CIED) infections.

Objective: To analyze the usefulness of sonication in the microbiological diagnosis of CIED infections in comparison with traditional cultures.

Methods: Systematic database searches were performed to identify studies that provided enough data concerning both sensitivity and specificity of traditional (non-sonicated) and sonicated cultures from CIED samples. The diagnostic accuracy measures were obtained by three different statistical approaches: (i) The univariate model; (ii) The bivariate random; and (iii) The Bayesian bivariate hierarchical model. Heterogeneity was assessed using meta-regression.

Findings: Nine studies met the criteria for inclusion in the meta-analysis (1684 cultures). The summary estimates of sensitivity were higher for sonicated cultures (0.756) in comparison with non-sonicated cultures (0.446). On meta-regression, sonication of CIEDs significantly increased the sensitivity (p = 0.001) as well as the rates of false positive results (p = 0.003). The final model also showed that the studies that used a threshold for positivity were associated with lower rates of false positive results (p < 0.001).

Interpretation: Our results suggest that sonication improves the microbiological diagnosis of CIED infections in comparison with traditional cultures, but a standardization of processes is necessary.