Host response signature trends in persistent bacteraemia and metastatic infection due to Staphylococcus aureus and Gram-negative bacilli: a prospective multicentre observational study

Abstract

Background

A prompt diagnosis of bacteraemia and sepsis is essential. Markers to predict the risk of persistent bacteraemia and metastatic infection are lacking. SeptiCyte RAPID is a host response assay stratifying patients according to the risk of infectious vs sterile inflammation through a scoring system (SeptiScore). In this study we explore the association between SeptiScore and persistent bacteraemia as well as metastatic and persistent infection in the context of a proven bacteraemia episode.

Methods

This is a prospective multicentre observational 14-month study on patients with proven bacteraemia caused by Staphylococcus aureus or Gram-negative bacilli. Samples for assessment by SeptiCyte were collected with paired blood cultures for 4 consecutive days after the index blood culture.

Results

We included 86 patients in the study, 40 with S. aureus and 46 with Gram-negative bacilli bacteraemia. SeptiScores over the follow-up were higher in patients with Gram-negative compared to S. aureus bacteraemia (median 6.4, IQR 5.5–7.4 vs 5.6 IQR 5.1–6.2, p = 0.002). Higher SeptiScores were found to be associated with positive blood cultures at follow-up (AUC = 0.86, 95%CI 0.68–1.00) and with a diagnosis of metastatic infection at day 1 and 2 of follow-up (AUC = 0.79, 95%CI 0.57–1.00 and AUC = 0.82, 95%CI 0.63–1.00 respectively) in the context of Gram-negative bacteraemia while no association between SeptiScore and the outcomes of interest was observed in S. aureus bacteraemia. Mixed models confirmed the association of SeptiScore with positive blood cultures at follow-up (p = 0.04) and metastatic infection (p = 0.03) in the context of Gram-negative bacteraemia but not S. aureus bacteraemia after adjusting for confounders.

Conclusions

SeptiScores differ in the follow-up of S. aureus and Gram-negative bacteraemia. In the setting of Gram-negative bacteraemia SeptiScore demonstrated a good negative predictive value for the outcomes of interest and might help rule out the persistence of infection defined as metastatic spread, lack of source control or persistent bacteraemia.

Keywords:

• Bloodstream infection
• host response
• blood culture
• biomarkers
• clinical score

Acknowledgements

We acknowledge Ms Tiffany Au for the assistance with the study submission to the HREC and with obtaining site-specific approval; Ms Megan Ratcliffe, Ms Maree Duroux, Ms Samantha Shone, Ms Kylie Jacobs and Ms Julia Affleck for the assistance with the study coordination as well as data collection at Redcliffe Hospital and Caboolture Hospital; Ms Michelle Bauer for the laboratory support.

Authors’ contributions

DLP and AMP designed the study. AMP, NR, KOC, AB, BG, HS, ME, KL, JPS, AS, ND, AT and PNAH contributed to screening and enrolling eligible patients. AMP performed data analysis and wrote the main version of the manuscript. HB run the SeptiCyte RAPID tests. MC supervised statistical analysis and data interpretation. DLP supervised data interpretation and the manuscript drafting. All authors substantially contributed to revising the manuscript and approved its final version.

Disclosure statement

DLP has research funding from Shionogi, Merck, bioMerieux, BioVersys and Pfizer and has received consulting fees from the AMR Action Fund, CARB-X, Aurobac, Pfizer, Merck, Cepheid, bioMerieux and Spero. PNAH reports research grants from Gilead, has served on advisory boards for OpGen, Merck and Sandoz, and has received honoraria from OpGen, Sandoz, Pfizer and bioMerieux. The other authors declare no conflict of interest.

Additional information

Funding

The study was supported by The University of Queensland. The manufacturer of SeptiCyte RAPID (ImmunExpress, Seattle, WA) did not provide any funding to the study, nor was involved in study design, samples’ testing or samples’ supply, data analysis or manuscript drafting. AMP is receiving a scholarship from the University of Queensland in support of her PhD candidature. PNAH was supported by an Early Career Fellowship from the National Health and Medical Research Council (GNT1157530).