Evaluation of the host cytokine response in dogs with sepsis and noninfectious systemic inflammatory response syndrome

Abstract

Objective: To quantify plasma cytokine concentrations in dogs with sepsis and noninfectious systemic inflammation and to evaluate the association between plasma cytokines and outcome in dogs with sepsis.Design: Prospective, observational cohort study. Setting: University teaching hospital.Animals: Forty-five dogs with sepsis, 10 dogs with noninfectious systemic inflammation (nSIRS), and 15 healthy controls were consecutively enrolled from June 2015 to February 2016 and followed to hospital discharge. Dogs with sepsis satisfied ≥2 SIRS criteria and had a documented or highly suspected bacterial infection. Dogs with nSIRS satisfied ≥2 SIRS criteria but had no evidence of infection. Dogs <3 kg and those with documented coagulopathy were excluded.Interventions: Measurement of inflammatory cytokines and high-mobility group box-1 (HMGB-1) was performed on each group. Measurements and Main Results:High-mobility group box-1 concentrations were analyzed by ELISA. Plasma concentrations of 13 cytokines were measured in singlet using multiplex magnetic bead assays. Kruskal-Wallis with Dunn's multiple comparison tests were used to compare biomarker concentrations between groups. Mann-Whitney U-tests were used to compare biomarker concentrations between survivors and nonsurvivors. Associations between biomarkers were evaluated using Spearman's correlation coefficients. Independent outcome predictors were identified using multivariable logistic regression. Alpha was set at 0.05. Concentrations of interleukin (IL)-6, C-X-C motif chemokine (CXCL)-8, keratinocyte-derived chemokine (KC)-like, C-C motif chemokine ligand 2 (CCL2), and HMGB-1 were significantly greater in dogs with sepsis versus healthy controls (all P ≤ 0.034). In dogs with sepsis, only CCL2 was independently associated with survival (odds ratio [OR] 0.996, 95% CI 0.993-0.999, P = 0.004). A cut-off of 385 pg/mL for CCL2 was 80% sensitive and 91.4% specific for nonsurvival (area under the ROC curve [AUROC] 0.866). Conclusions:Dogs with sepsis have significantly increased concentrations of HMGB-1 and inflammatory cytokines, including IL-6, CXCL8, and KC-like. Increased CCL2 concentration is a negative prognostic indicator in dogs with sepsis. These findings should be confirmed using duplicate analyses in larger, distinct populations of dogs with sepsis before applying them to clinical patients. K E Y W O R D Scanine, CCL2, HMGB-1, outcome, SIRS

High Mobility Group Box-1 and Pro-inflammatory Cytokines Are Increased in Dogs After Trauma but Do Not Predict Survival

“…In this respect our data are consistent with the findings of Gommeren et al ( 30 ) who found no association between the concentrations of IL-6 or TNF-α and survival in their population of dogs with SIRS that included 6 dogs following trauma. The data from the present study are distinct from a recent study we conducted in canine sepsis; however, where the concentration of CCL2 was highly discriminant for non-survival ( 19 ). This may be due to fundamental differences in the pathophysiology of sepsis due to bacterial infection where chemokine concentrations may reflect the degree of pathogen invasion and trauma, where tissue damage is the predominant signal initiating cytokine secretion.…”

“…Concentrations of HMGB-1 are reportedly prognostic in dogs with GDV ( 17 ) and in dogs with systemic inflammation ( 40 , 41 ). Other studies have not confirmed the prognostic value of HMGB-1 in GDV syndrome ( 43 ) or sepsis however ( 19 ), suggesting that the value of this biomarker varies with the population, the illness severity and the nature of the disease process. In the present study, all dogs were assessed following trauma, but the nature, severity and timing of these injuries varied.…”

“…The half-life of HMGB-1 is ~11 h in people ( 16 ). Concentrations of HMGB-1 are increased in dogs with gastric dilation-volvulus (GDV) ( 17 , 18 ), dogs with sepsis ( 19 ), and correlate with plasma nucleosome concentrations in critically ill horses ( 20 ), suggesting it may provide information about tissue injury in dogs with trauma.…”

“…In human medicine, cytokines have been evaluated as potential trauma biomarkers and have been associated with development of disseminated intravascular coagulation, multiple organ dysfunction syndrome and mortality ( 24 – 27 ). The cytokine profiles induced by lipopolysaccharide (LPS) administration ( 28 , 29 ), in naturally-occurring SIRS ( 30 ), in canine sepsis ( 19 ), and in immune-mediated hemolytic anemia (IMHA) ( 22 , 31 ), have been reported, but to date there are no studies focused on the cytokine profiles of traumatized dogs. Evaluating the cytokine response to trauma could establish patterns of release that may enhance survival prediction ( 32 ).…”

Cell-Free DNA, High-Mobility Group Box-1, and Procalcitonin Concentrations in Dogs With Gastric Dilatation–Volvulus Syndrome

“…Although plasma concentrations of both these biomarkers indicate the release of cellular nucleic acids into circulation, their increase might be regulated by different pathways with distinct kinetics. To date, only weak correlations have been reported between cfDNA and DNA-histone-complexes in people after cardiac arrest ( 9 ), and between cfDNA and HMGB1 in dogs with sepsis ( 38 ). The lack of correlation between the two biomarkers and lactate concentrations might also reflect different stimuli for the release of these markers.…”