• Jos F Frencken, Lonneke A van Vught, Linda M Peelen, David S Y Ong, Peter M C Klein Klouwenberg, Janneke Horn, Marc J M Bonten, Tom van der Poll, Olaf L Cremer, and MARS Consortium.
• 1Department of Epidemiology, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, The Netherlands. 2Department of Intensive Care Medicine, University Medical Center Utrecht, Utrecht, The Netherlands. 3Center for Experimental and Molecular Medicine, Academic Medical Center, Amsterdam, The Netherlands. 4Center for Infection and Immunity, Academic Medical Center, Amsterdam, The Netherlands. 5Department of Medical Microbiology, University Medical Center Utrecht, Utrecht, The Netherlands. 6Department of Intensive Care, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands. 7Division of Infectious Diseases, Academic Medical Center, Amsterdam, The Netherlands.
• Crit. Care Med. 2017 May 1; 45 (5): e493-e499.

ObjectiveThe prevailing theory of host response during sepsis states that an excessive production of pro-inflammatory mediators causes early deaths, whereas a predominantly anti-inflammatory response may lead to immunosuppression, secondary infection, and late deaths. We assessed inflammatory (im)balance by measuring pro-inflammatory interleukin-6 and anti-inflammatory interleukin-10 during three distinct time periods after sepsis, and assessed its association with mortality.DesignProspective observational cohort.SettingTwo tertiary mixed ICUs in The Netherlands.PatientsConsecutive patients presenting with severe sepsis or septic shock from 2011 to 2013.InterventionsNone.Measurements And Main ResultsWe repeatedly measured plasma interleukin-6 and interleukin-10 concentrations using cytometric bead array. Poisson regression was used to analyze the relation between inflammatory markers measured on 1) ICU admission and day 4 mortality, 2) day 4 and day 28 mortality, and 3) ICU discharge and 1-year mortality. Secondary outcome was development of ICU-acquired infections. Among 708 patients, 86 (12%) died within 4 days, 140 (20%) died between days 4 and 28, and an additional 155 (22%) died before 1 year. Interleukin-6 and interleukin-10 levels were both independently associated with mortality, but the balance of this response as modelled by an interleukin-6 and interleukin-10 interaction term was not (relative risk, 0.99; 95% CI, 0.95-1.04 on admission; relative risk, 1.02; 95% CI, 0.98-1.06 on day 4; and relative risk, 1.12; 95% CI, 0.98-1.29 at ICU discharge). However, inflammatory imbalance on day 4 was associated with development of ICU-acquired infections (subdistribution hazard ratio, 0.87; 95% CI, 0.77-0.98).ConclusionsAlthough both interleukin-6 and interleukin-10 productions are associated with death, the balance of these inflammatory mediators does not seem to impact either early, intermediate, or late mortality in patients presenting to the ICU with sepsis.

47 keywords...

hide…