Secondary Logo

Institutional members access full text with Ovid®

Changes in exhaled 13CO2/12CO2 breath delta value as an early indicator of infection in intensive care unit patients

O'Rourke, Ann P., MD; Buckman, Sara A., MD; Evans, David C., MD; Kerwin, Andrew J., MD; Breunig, Emily A.; Bütz, Daniel E., PhD

Journal of Trauma and Acute Care Surgery: January 2019 - Volume 86 - Issue 1 - p 71–78
doi: 10.1097/TA.0000000000002097

BACKGROUND We have developed a new, noninvasive predictive marker for onset of infection in surgical intensive care unit (ICU) patients. The exhaled 13CO2/12CO2 ratio, or breath delta value (BDV), has been shown to be an early marker for infection in a proof of concept human study and in animal models of bacterial peritonitis. In these studies, the BDV changes during onset and progression of infection, and these changes precede physiological changes associated with infection. Earlier diagnosis and treatment will significantly reduce morbidity, mortality, hospitalization costs, and length of stay. The objective of this prospective, observational, multicenter study was to determine the predictive value of the BDV as an early diagnostic marker of infection.

METHODS Critically ill adults after trauma or acute care surgery with an expected length of stay longer than 5 days were enrolled. The BDV was obtained every 4 hours for 7 days and correlated to clinical infection diagnosis, serum C-reactive protein, and procalcitonin levels. Clinical infection diagnosis was made by an independent endpoint committee. This trial was registered at the US National Institutes of Health ( NCT02327130.

RESULTS Groups were demographically similar (n = 20). Clinical infection diagnosis was confirmed on day 3.9 ± 0.63. Clinical suspicion of infection (defined by SIRS criteria and/or new antibiotic therapy) was on day 2.1 ± 0.5 in all infected patients. However, 5 (56%) of 9 noninfected subjects also met clinical suspicion criteria. The BDV significantly increased by 1‰ to 1.7‰ on day 2.1 after enrollment (p < 0.05) in subjects who developed infections, while it remained at baseline (± 0.5‰) for subjects without infections.

CONCLUSION A BDV greater than 1.4‰ accurately differentiates subjects who develop infections from those who do not and predicts the presence of infection up to 48 hours before clinical confirmation. The BDV may predict the onset of infection and aid in distinguishing SIRS from infection, which could prompt earlier diagnosis, earlier appropriate treatment, and improve outcomes.

LEVEL OF EVIDENCE Diagnostic test, level III.

From the Department of Animal Science, University of Wisconsin–Madison, Madison, Wisconsin (D.E.B.); Department of Surgery, University of Wisconsin–Madison, Madison, Wisconsin (A.P.O.); Department of Surgery, Section of Acute and Critical Care Surgery, Washington University, St. Louis, Missouri (S.A.B.); Department of Surgery, Division of Trauma, Critical Care and Burn, The Ohio State University, Columbia, Ohio (D.C.E.); Department of Surgery, Division of Acute Care Surgery, University of Florida, Jacksonville, Florida (A.J.K.); Isomark, LLC, Madison, Wisconsin (E.A.B., D.E.B.).

Submitted: August 31, 2017, Revised: September 13, 2018, Accepted: October 3, 2018, Published online: October 23, 2018.

76th Annual Meeting of AAST and Clinical Congress of Acute Care Surgery, September 13–16, 2017 in Baltimore, MD, Oral Presentation Friday September 15, 2017.

Address for reprints: Ann P. O'Rourke, MD, MPH, 600 Highland Ave, Madison, WI 53792; email:

© 2019 Lippincott Williams & Wilkins, Inc.