Plasma interleukin-6 response is predictive for severity and mortality in canine systemic inflammatory response syndrome and sepsis Stefanie Rau, Barbara Kohn, Constance Richter, Nora Fenske, Helmut Ku ¨ chenhoff, Katrin Hartmann, Stefan Ha ¨ rtle, Bernd Kaspers, Johannes Hirschberger Background: Sepsis is still a major cause of death in both human and veterinary medicine. Early diagnosis is essential for appropriate treatment. Identification of patients at risk for developing sepsis is already possible in human medicine through the measurement of plasma interleukin-6 (IL-6) levels. In veterinary medicine, however, this has been investigated only in canine experimental models. Objectives: The purpose of this study was to measure IL-6 plasma levels in dogs with naturally occurring systemic inflammatory response syndrome (SIRS) and sepsis and to analyze the value of IL-6 as a predictive parameter for severity and mortality. Methods: Included in the study were 79 dogs that had been admitted to the small animal clinics of Munich and Berlin from July 2004 to July 2005 and that satisfied the diagnostic criteria for SIRS and sepsis as defined using established parameters. Measurement of plasma IL-6 levels on days 0, 1, and 2 was performed by the use of a colorimetric bioassay based on IL-6–dependent cell growth. Results: Septic foci were identified in 43 patients (septic group), and 36 patients were enrolled in the SIRS group. The frequency of positive blood cultures was 11%. The overall mortality rate was 48%. Higher plasma IL-6 levels on the day of admission were significantly correlated with a more severe degree of disease, increased mortality rate, and earlier fatality. Conclusions: Plasma IL-6 concentration is predictive of outcome in canine SIRS and sepsis and may be a valuable laboratory parameter for assessing critically ill dogs. (Vet Clin Pathol. 2007;36:253–260) Ó2007 American Society for Veterinary Clinical Pathology Key Words: Dog, IL-6, inflammation, sepsis, SIRS u According to recent reports, the incidence rate for sepsis has been increasing over the past few decades, being now the 10th leading cause of death in human beings in the United States. 1–2 Unfortunately, no such information is available regarding sepsis and systemic inflammatory response syndrome (SIRS) in dogs. Data collected by the University of Pennsylvania Veterinary Teaching Hospital suggest a similar increase in the incidence of sepsis in dogs from 1 per 1000 hospital cases in 1988 to 3.5 in 1998. 3 This increase needs to be further in- vestigated, however, as it may be due to an increased clinical suspicion of sepsis and subsequently increased likelihood of diagnosis, rather than a true increase in the frequency of sepsis. Mortality rates in dogs with sepsis range from 31% to 50% in various veterinary studies. 4–9 Therapeutic intervention in both dogs and people consists mainly of supportive care and antibiotics and, of course, treating the underlying cause, as specific anti-inflammatory strategies have failed in human clinical trials. 10–13 (The only promising exception, that of drotrecogin alfa, is rapidly eliminated, has antigenic poten- tial, 14 and is currently far too expensive to use in companion animals.) Successful treatment depends frequently on early diagnosis. Therefore, identification of animals at risk for sepsis is an urgent current research goal. A presumptive diagnosis of sepsis is based on the criteria for the identification of SIRS, as defined by the American College of Chest Physicians/Society of Critical Care Medicine consensus conference 15 and modified for veterinary use 4,16 and either histologic or microbiologic confirmation of infection. At the time of initial clinical presentation and examination, only classification by the SIRS criteria 4 is available. Bacterial culture takes several days and false-negative results are not uncommon. SIRS and sepsis are characterized by activation of the cytokine network. The proinflammatory cytokine interleukin-6 (IL-6) has a longer plasma half-life than does tumor necrosis factor alpha (TNF-a) or interleukin-1-beta and its concentra- tion has proven to be significantly elevated in the plasma of septic human patients. A correlation between high levels of IL-6 on admission and mortality was found in most studies. 17–32 IL-6 thus appears to be a good marker of the severity of sys- temic bacterial infection when measured at the time of admis- sion. Canine experimental models produced by administering infusions of either live Escherichia coli or lipopolysaccharide From the Clinic for Small Animal Internal Medicine (Rau, Hartmann, Hirschberger), the Consulting Unit, Department of Statistics (Fenske, Ku ¨ chenhoff), and the Institute for Animal Physiology (Ha ¨ rtle, Kaspers), Ludwig-Maximillians University of Munich, Munich, Germany; and the Clinic for Small Animals, Free University of Berlin, Berlin, Germany (Kohn, Richter). This study was presented in part at the 14th Meeting of the Sections of Internal Medicine and Clinical Laboratory Diagnostics (InnLab) of the German Veterinary Society DVG, Munich, Germany, May 13–14, 2006; and at the European Congress of Veterinary Internal Medicine – Companion Animals (ECVIM-CA), Amsterdam, The Netherlands, September 14–16, 2006. Corresponding author: Stefanie Rau (stefanie.rau@gmail.com). ª2007 American Society for Veterinary Clinical Pathology Vol. 36 / No. 3 / 2007 Veterinary Clinical Pathology Page 253