Journal of Veterinary Emergency and Critical Care 25(4) 2015, pp 488–494 doi: 10.1111/vec.12324
Original Study
C-reactive protein and high mobility group box 1 in dogs with gastric dilatation and volvulus Ivana Uhrikova, MVDr, PhD; Leona Rauserova-Lexmaulova, MVDr, PhD; Kristina Rehakova, MVDr, PhD; Peter Scheer, MVDr, PhD and Jaroslav Doubek, MVDr, CSc
Abstract
Objective – To (1) measure C-reactive protein (CRP) and high mobility group box 1 (HMGB1) and (2) evaluate their prognostic value and relationship to severity of systemic inflammatory response syndrome, routine hematological and acid-base parameters in dogs with gastric dilatation volvulus (GDV). Design – Prospective observational study from September 2010 to June 2012. Setting – Veterinary teaching hospital. Animals – Forty-one client-owned dogs with GDV. Interventions – None. Measurements and Main Results – Blood was collected before surgery (baseline), postsurgery, 6–10 hours postsurgery, and 18–22 hours postsurgery. CRP and HMGB1 were measured in all samples, and routine hematological, biochemical, and acid-base analyses were performed. Only baseline and postsurgery samples were used from nonsurvivors (n = 10). CRP increased significantly from postsurgery sampling to 18–22 hours postsurgery, while HMGB1 did not change over time. There was a significant difference in HMGB1 between survivors and nonsurvivors over time. Both proteins correlated with systemic inflammatory response syndrome severity, total leukocyte, segmented neutrophils, and band counts. HMGB1 correlated also with acid-base parameters (pH, bicarbonate, base excess). Conclusion – HMGB1 and CRP behaved differently in regards to their kinetic patterns, with HMGB1 appearing to better reflect the severity of tissue injury in dogs with GDV than CRP. (J Vet Emerg Crit Care 2015; 25(4): 488–494) doi: 10.1111/vec.12324 Keywords: canine, mortality, prognosis, systemic inflammatory response syndrome
Abbreviations
CRP GDV HMGB1 OHE ROC
C-reactive protein gastric dilatation and volvulus syndrome high-mobility group box 1 ovariohysterectomy receiver operating curve
From the Department of Physiology, (Uhrikova, Scheer, Doubek), Small Animal Clinic, Department of Surgery and Orthopedics (RauserovaLexmaulova), Small Animal Clinical Laboratory (Rehakova, Doubek), University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic; and Integrated Center of Cellular Therapy (ICCT), St. Annes´s University Hospital - International Clinical Research Center Brno (Scheer), Brno, Czech Republic. The authors declare no conflict of interest. The study was supported by grant no. 24/2011/IGA VFU Brno. Presented, in part, at the ESVCP Congress, Ireland, 2011 and Veterinary and Comparative Clinical Pathology Conference, Slovenia, 2012. Address correspondence and reprint requests to Dr. Ivana Uhrikova, Department of Physiology, University of Veterinary and Pharmaceutical Sciences Brno, Palack´eho tˇr. 1/3, 612 42 Brno, Czech Republic. Email: [email protected] Submitted November 21, 2012; Accepted March 24, 2015.
488
SIRS UDL
systemic inflammatory response syndrome upper detection limit
Introduction Gastric dilatation and volvulus syndrome (GDV) is a severe disease in large and giant breed dogs that usually occurs in otherwise healthy animals. Onset of the clinical signs is acute and dogs may quickly fulfill the criteria of the systemic inflammatory response syndrome (SIRS).1 Surgical treatment with stomach decompression and repositioning can rapidly correct the primary problem and generally reverses accompanying complications. Due to its rapid onset and resolution as a clinical syndrome, we opted to study the development of SIRS in dogs with GDV. SIRS is associated with the increase in circulating concentration of proinflammatory cytokines and acute phase proteins.2 C-reactive protein (CRP) is a major acute phase protein in dogs, and has been widely studied in recent years.3–7 Its synthesis in the liver is stimulated C Veterinary Emergency and Critical Care Society 2015
CRP and HMGB-1 in dogs with GDV
by proinflammatory cytokines and its concentrations increase as early as 4 hours following the initial stimulus.8 Although a single measurement of this protein seems to have no prognostic significance in dogs with SIRS, knowledge of its kinetic pattern may help in the detection of emerging complications or worsening of the disease.4–6 High mobility group box 1 (HMGB1) is a nonhistone chromosomal protein that is highly conserved among various species with 100% amino acid homology between people and dogs.9 HMGB1 is an alarmin, a protein that is released extracellularly during cell necrosis but not apoptosis; it is also a late cytokine with active secretion after a lag phase of 8–18 hours. Outside the cells, it triggers inflammation by the stimulation of proinflammatory cytokine production by monocytes/macrophages, increased expression of adhesion molecules, and increased epithelial permeability.10,11 HMGB1 appears to be a good biomarker that relates to final outcome in some conditions in human medicine,12–15 but studies in veterinary medicine are limited.16,17 To our knowledge, there has been no study focused on HMGB1 in GDV to date. The aims of this study were to describe (1) the kinetic behavior and prognostic value of CRP and HMGB1 in dogs with GDV and (2) relate CRP and HMGB1 concentrations to findings of routine hematological and biochemical parameters.
Material and Methods Animals Dogs with GDV were included in this prospective observations study conducted between September 2010 and June 2012. Inclusion criteria included the diagnosis of GDV based upon compatible clinical signs, positive radiographic examination, and its confirmation during surgical treatment. We obtained signed consent from dog owners for participation in this study. All dogs were sedated by IV administration of midazolam 0.2 mg/kg, butorphanola 0.4 mg/kg, and ketamineb 2 mg/kg, and general anesthesia was induced with propofolc . In addition, buprenorphined 0.005 mg/kg was administered after 20 minutes in most of the patients. General anesthesia was maintained with an inhalation mixture of oxygen (50%) in air and isofluranee with artificial ventilation. Surgical treatment with decompression and derotation of the stomach and gastropexy was performed in all patients according to their individual needs. In cases of operable gastric necrosis, partial gastrectomy was performed. In cases of inoperable gastric necrosis, dogs underwent euthanasia. All dogs received infusion therapy consisting of crystalloidf and colloid (hydroxyethyl starch)g solutions based on results from C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
clinical and acid-base examination. None of the dogs received a blood transfusion. Supplementary data Data related to the time since last feeding and time since onset of clinical signs were obtained from the owners. The surgeon determined the degree of gastric rotation based on a subjective evaluation (eg, 120°, 180°, 270°, 360°). Dogs were assigned into the SIRS group if ࣙ2 of the following criteria were fulfilled18 : tachycardia (>120/min), tachypnea (>20/min), increased (>39.0°C) or decreased (16 × 109 /L [>16000/L]), leukopenia (6.67 ng/mL with sensitivity 88.9% and specificity 77.4%; for CRP the cutoff was C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
Figure 1: Survival prediction fitted receiver operating curve (ROC) for high mobility group box 1 (HMGB1) and C-reactive protein (CRP) from baseline samples. Cut-off values for poor short-term outcome were HMGB1 >6.67 ng/mL (sensitivity 88.9%, specificity 77.4%) and CRP >7.2 mg/L (sensitivity 90.0%, specificity 43.3%). The AUC was 0.849 (95% CI: 0.70–0.94; P < 0.0001) for HMGB1 and 0.696 (95% CI: 0.53–0.83; P < 0.05) for CRP.
>7.2 mg/L with sensitivity 90.0% and specificity 43.3%. The AUC was 0.849 (95% CI: 0.70–0.94; P < 0.0001) for HMGB1 and 0.696 (95% CI: 0.53–0.83; P < 0.05) for CRP. Correlations A moderate correlation was observed between the time from last feeding and CRP concentrations in baseline and postsurgery samples (r = 0.53 and r = 0.56, P < 0.001). HMGB1 correlated moderately with time from 491
I. Uhrikova et al.
Table 2: Absolute and differential leukocyte counts over time in dogs with gastric dilatation volvulus stratified by outcome
Sample Absolute leukocyte count (×103 /L or ×109 /L) Segmented neutrophils (×103 /L or ×109 /L) Neutrophil bands (×103 /L or ×109 /L) Monocytes (×103 /L or ×109 /L) Lymphocytes (×103 /L or ×109 /L) Eosinophils (×103 /L or ×10‘ /L)
All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors
Baseline
Postsurgery
9.8 (4.1–26.7) 9.6 (4.1–26.7) 17.0 (7.4–26.6) 8.8 (4.3–24.7) 8.3 (4.3–21.9) 15.1 (5.1–24.7) 0.1 (0.0–3.3) 0.1 (0.0–1.3) 0.5 (0.0–3.3) 0.5 (0.0–2.1) 0.6 (0.0–1.9) 0.5 (0.1–2.1) 0.6 (0.0–2.9) 0.6 (0.1–2.7) 0.4 (0.0–2.9) 0.0 (0.0–0.8) 0.1 (0.0–0.8) 0.0 (0.0–0.2)
7.4 (1.3–21.5) 7.3 (1.3–21.5) 8.7 (1.4–17.3) 5.6 (0.8–15.7) 5.8 (0.8–14.2) 5.4 (0.8–15.7) 0.1 (0.0–1.1) 0.1 (0.0–1.1) 0.3 (0.1–0.7) 0.3 (0.0–4.3) 0.3 (0.0–4.3) 0.3 (0.0–1.0) 0.5 (0.0–2.1) 0.5 (0.0–1.9) 0.2 (0.1–2.1) 0.1 (0.0–1.1) 0.1 (0.0–1.1) 0.0 (0.0–0.2)
6–10 hours postsurgery
18–22 hours postsurgery
19.3 (5.3–35.4) 19.3 (5.3–35.4)
19.8 (6.0–44.2) 19.8 (6.0–44.2)
17.8 (4.6–27.3) 17.8 (4.6–27.3)
16.6 (2.9–37.1) 16.6 (2.9–37.1)
0.5 (0.0–7.8) 0.5 (0.0–7.8)
0.2 (0.0–6.6) 0.2 (0.0–6.6)
0.4 (0.0–2.6) 0.4 (0.0–2.6)
0.5 (0.0–2.1) 0.5 (0.0–2.1)
0.6 (0.0–2.3) 0.6 (0.0–2.3)
1.3 (0.0–4.2) 1.3 (0.0–4.2)
0.0 (0.0–0.2) 0.0 (0.0–0.2)
0.0 (0.0–1.1) 0.0 (0.0–1.1)
Values are shown as median (minimum-maximum).
last feeding in baseline samples (r = 0.42, P < 0.01) and the strongest correlation (r = −0.59, P < 0.01) was seen 6–10 hours postsurgery between HMGB1 and the degree of gastric rotation. There was a mild correlation between the severity of SIRS and CRP concentration as well as HMGB1 at baseline (r = 0.32 and r = 0.45, P < 0.05) and severity of SIRS 6–10 hours after surgery and CRP at baseline (r = 0.46, P = 0.01). Results from the hematologic examination are shown in Table 2. The correlations noted between CRP and hematologic parameters were: a moderate correlation with leukocyte counts at baseline sampling (r = 0.66, P < 0.0001), segmented neutrophils at baseline sampling (r = 0.69, P < 0.0001), band neutrophils at postsurgery and 18–22 hours postsurgery samplings (r = 0.48 and r = 0.57, P < 0.01 and P = 0.01) and eosinophils at postsurgery sampling (r = −0.4, P = 0.01). Similarly, HMGB1 was correlated with leukocyte counts at baseline sampling (r = 0.47, P < 0.01), segmented neutrophils at baseline sampling (r = 0.40, P < 0.05), band neutrophils (r = 0.45, P < 0.01), and eosinophils at postsurgery sampling (r = −0.48, P < 0.01). A moderate correlation was also found with lactate concentration at baseline and postsurgery samplings (r = 0.63 and 0.62, P < 0.001) and selected acid-base parameters (pH, HCO3− , base excess) at baseline, postsurgery, and 18–22 hours postsurgery samplings (r = −0.35 to −0.45, P < 0.05). There was no correlation among CRP or HMGB1 and the remaining biochemical parameters that were measured. Correlation studies among sampling points revealed only moderate
492
correlation (from r = 0.48 to 0.45, P < 0.01) between CRP or HMGB1 (both baseline) and the number of band neutrophils (postsurgery).
Discussion To date, only a few studies have investigated the changes of CRP concentrations in dogs.4,17 The current study was focused on serial measurement of CRP and the nuclear protein, HMGB1 in dogs with GDV, their prognostic value and relation to SIRS, and selected laboratory parameters. Although CRP has received more attention in veterinary medicine than HMGB1, publications focusing on CRP in dogs with GDV are limited.7 In our study, there was no significant difference in presurgical CRP concentrations between survivors and nonsurvivors, although median values were higher in nonsurvivors (8.0 vs. 12.8 mg/L). These findings are similar to those of Israeli et al,7 although the median CRP concentration in all dogs was far lower in our study (9.1 vs. 23.0 mg/L), and the range of values was wider than that published by Israeli et al. (0.3–167.7 vs. 0.1–111.0 mg/L).7 We found a moderate correlation between CRP and time since last feeding (r = 0.53, P < 0.001) and thus differences between studies could be attributed to the different times of arrival at the clinic. The median value of CRP at admission in survivors was within the reference interval declared by the manufacturern (0–25 mg/L), but higher than that found in 30 healthy dogs sampled in our laboratory (median 2.5, range 0–12.6 mg/L, unpublished).
C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
CRP and HMGB-1 in dogs with GDV
In dogs in the current study CRP gradually increased over time. It is well known that CRP increases after surgery,17,20,21 and thus our findings are consistent with those observations. The magnitude of increase varies across studies. Following ovariohysterectomy, CRP concentrations increased from 4- to 44-fold with a peak at 24–28 hours.20,21 Such a steep increase in CRP concentration following to surgery complicates interpretation of its concentration in early phases of wound healing. A correlation between the invasiveness of surgery and magnitude of CRP increase was described previously,20,r and under the same surgical conditions it seems that CRP values are predictive for complications as late as the third day following surgery.4,22 In our study, there was only a weak association between CRP baseline concentrations and SIRS severity at admission as well as 6–10 hours postsurgery, which may reflect the intensity of tissue damage. Despite the small number of dogs undergoing partial gastrectomy we can conclude, that CRP concentration 6–10 hours postsurgery is higher in these dogs confirming the influence of the invasiveness of surgery on CRP concentrations. A correlation between CRP and hematologic parameters was noted. This is not surprising since the maturation of neutrophils is stimulated by the same proinflammatory cytokines that act on liver to stimulate CRP synthesis.23 The number of neutrophil bands (18–22 hours postsurgery) and segmented neutrophils (6–10 and 18–22 hours postsurgery) was higher in dogs undergoing partial gastrectomy in agreement with CRP concentrations. HMGB1 has been intensively studied in many fields of human medicine.24 This nuclear DNA-binding protein is passively released during cell necrosis and actively as a late cytokine. Its concentration increases within 18 hours following the administration of endotoxin.10,11 Since most of the dogs were sampled less than 18 hours after onset of clinical signs or time since last feeding, we believe that most of the initial HMGB1 originated from its release from damaged cells. The moderate correlation between lactate and HMGB1 (baseline and postsurgery) supports this hypothesis, since lactate in GDV probably originates from hypoperfused tissues. This could also explain the correlation between HMGB1 and the “metabolic” part of the acid-base status (eg, bicarbonate, base excess). Finally, significant differences in HMGB1 between dogs with and without gastric necrosis suggest that all of these findings are more closely associated with the viability of the stomach wall than tissue hypoperfusion due to the shock state. HMGB1 has been the subject of interest in few studies in veterinary medicine, although significant differences between dogs with and without SIRS as well as between survivors and nonsurvivors with SIRS have been described.16,17 Our study C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
also found a significant difference in HMGB1 between survivors and nonsurvivors as well as a correlation with the severity of SIRS. The effect of surgery on HMGB1 release in dogs after ovariohysterectomy was studied in 5 dogs by Ishida et al.17 HMGB1 increased during the first 12 hours, then decreased at 24 hours, and finally gradually increased with a peak at 72 hours. In our study, levels of HMGB1 were not significantly different among different time points. Although, the number of dogs undergoing partial gastrectomy was too low to allow a statistical comparison of HMGB1 difference between dogs undergoing partial gastrectomy with gastropexy and solitary gastropexy, the median HMGB1 was 2-fold higher in dogs with partial gastrectomy compared to nongastrectomized dogs. In the 6–10 and 18–22 hours postsurgery samplings median values were similar in these 2 groups. Of these 2 parameters, initial HMGB1 concentration may be used as an indicator of poor outcome in dogs with GDV since its AUC was higher than for CRP. Additionally, absence of gradual increase of HMGB1 following surgery may be exploited for the early detection of late postsurgical complications in comparison to CRP. However, its measurement is limited by the lack of point-of-care HMGB1 assays. Its advantage over other biomarkers is its homology with human HMGB1,9 thus there is a high probability that future development of point-of-care assays for people will also be useful in dogs. A correlation between HMGB1 and leukocyte count and neutrophils was also observed. This correlation may be due to HMGB1 release from injured cells, which stimulates infiltration of the damaged site with neutrophils. The correlation with band neutrophils and eosinophils may occur for the same reasons as in the case of CRP, since HMGB1 stimulates the release of other proinflammatory cytokines from monocytes/macrophages.10,11,25 We are aware of several limitations of this study. First, the number of dogs included was relatively small. Second, the data related to the time from onset of clinical signs and degree of gastric rotation were based on subjective observations. In addition, the data related to time of last feeding submitted by the owners could contain a degree of error. Although the duration of surgery varied amongst cases, the anesthetic and surgical protocols were standardized in all dogs as much as possible. Postoperative care within the first 24 hours was similar in all dogs, although there may have been small differences in the rate and volume of fluid therapy following surgery. Blood collection as well as sample processing was performed by the same 2 investigators, and the differential leukocyte count was performed by the same investigator for all samples to prevent variations in the preanalytical, analytical, or postanalytical phase. Third, the decision 493
I. Uhrikova et al.
concerning whether gastric necrosis was operable or not (resulting in euthanasia) was made by the attending surgeon; however, whenever possible, partial gastrectomy was performed. In addition, the collection of all samples at the same time rather than at 4 hour intervals would be ideal; but most dogs were admitted during the night, and thus 6–10 and 18–22 hours postsurgery samples were taken in the morning and evening of the next day. Finally, dilution of samples with values beyond the UDL might have allowed us to demonstrate more significant results. Despite all of these limitations, we have identified some important findings.
Conclusion In dogs with GDV, the kinetics of HMGB1 differed from that of CRP and initial HMGB1 values may be reflective of the degree of tissue damage found at surgery. Both parameters correlated only mildly with SIRS severity at the admission. Further research on the uses of HMGB1 assessment in dogs is warranted.
Footnotes a b c d e f
g h i j k l m n o p q r
Butomidor, Richter Pharma, Wels, Austria. Narkamon 5% a.u.v. inj., Spofa a.s., Czech Republic. Norofol 1%, Norbrook Laboratories Limited, Germany. Vetergesic, Reckitt Beckiser Healthcare, Berkshire, UK. Isoflurane, Nicholas Piramal India Ltd., UK. 0.9% NaCl, Sodium chloride, B. Braun Medical, Prague, Czech Republic; Hartmann’s Viaflo, Plasmalyte solution, Baxter Czech spol. s.r.o., Prague, Czech Republic, Ringer’s injection “fresenius,” Fresenius Kabi Italia S.r.l., Verona, Italy. Voluven, Baxter Czech spol. s.r.o. Dispolab s.r.o., Brno, Czech Republic. Sarstedt s.r.o., Brno, Czech Republic. Celltac Alpha, Nihon Kohden, Tokyo, Japan. Hemacolor, Biovendor, a.s., Brno, Czech Republic. Konelab DCP20i, Thermo Scientific, Vantaa, Finland. EasyBloodGas, Medica a.s., Czech Republic. CRP, Biovendor a.s. HMGB1, IBL International, GmbH, Hamburg, Germany. STATISTICA 6.0, StatSoft, Inc., Tulsa, OK. MedCalc Software bvba, Ostend, Belgium. Kjelgaard-Hansen M, Strom H, Mikkelsen LF, et al. Grading of surgical trauma by means of canine C-reactive protein measurements. In Proceedings of the 2008 ASVCP Annual Meeting.
References 1. Silverstein DC, Montealegre C, Shofer FS, et al. The association between vascular endothelial growth factor levels and clinically evident peripheral edema in dogs with systemic inflammatory response syndrome. J Vet Emerg Crit Care 2009; 19(5):459– 466. ¨ PO. The systemic inflammatory response syndrome: def2. Nystrom initions and aetiology. J Antimicrob Chemother 1998; 41(Suppl A): 1–7. 3. Chan DL, Rozanski EA, Freeman LM. Relationship among plasma amino acids, C-reactive protein, illness severity, and outcome in critically ill dogs. J Vet Intern Med 2009; 23(3):559–563.
494
4. Gebhardt C, Hirschberger J, Rau S, et al. Use of C-reactive protein to predict outcome in dogs with systemic inflammatory response syndrome or sepsis. J Vet Emerg Crit Care 2009; 19(5): 450–458. 5. Galezowski AM, Snead EC, Kidney BA, et al. C-reactive protein as a prognostic indicator in dogs with acute abdomen syndrome. J Vet Diagn Invest 2010; 22(3):395–401. 6. Whittemore JC, Marcum BA, Mawby DI, et al. Associations among albuminuria, C-reactive protein concentrations, survival predictor index scores, and survival in 78 critically ill dogs. J Vet Intern Med 2011; 25(4):818–824. 7. Israeli I, Steiner J, Segev G, et al. Serum pepsinogen-A, canine pancreatic lipase immunoreactivity, and C-reactive protein as prognostic markers in dogs with gastric dilatation-volvulus. J Vet Intern Med 2012; 26(4):920–928. 8. Yamashita K, Fujinaga T, Myiamoto T, et al. Canine acute phase response: relationship between serum cytokine activity and acute phase protein in dogs. J Vet Med Sci 1994; 56(3):487–492. 9. Murua Escobar H, Meyer B, Richter A, et al. Molecular characterization of the canine HMGB1. Cytogenet Genome Res 1993; 101(1):33– 38. 10. Andersson J, Erlandsson HH, Yang H, et al. HMGB1 as a DNAbinding cytokine. J Leukoc Biol 2002; 72(6):1084–1091. 11. Yang H, Wang H, Czura CJ, et al. The cytokine activity of HMGB1. J Leukoc Biol 2005; 78(1):1–8. 12. Huang LF, Yao YM, Dong N, et al. Association of high mobility group box-1 protein levels with sepsis and outcome of severely burned patients. Cytokine 2011; 53(1):29–34. 13. Wang KY, Yu GF, Zhang ZY, et al. Plasma high-mobility group box 1 levels and predtiction of outcome in patients with traumatic brain injury. Clin Chem Acta 2012; 413(21–22):1737–1741. 14. Huang JM, Hu J, Chen N, et al. Relationship between plasma highmobility group box-1 levels and clinical outcomes of ischemic stroke. J Crit Care 2013; 28:729–797. 15. Fahmueller YN, Nagel D, Hoffmann RT, et al. Immunogenic cell death biomarkers HMGB1, RAGE, and DNAse indicate response to radioembolization therapy and prognosis in colorectal cancer patients. Int J Cancer 2013; 132(10):2349–2358. 16. Yu DH, Nho DH, Song RH, et al. High-mobility group box 1 as a surrogate prognostic marker in dogs with systemic inflammatory response syndrome. J Vet Emerg Crit Care 2010; 20(3):298– 302. 17. Ishida A, Ohno K, Fukushima K, et al. Plasma high-mobility group box 1 (HMGB1) in dogs with various diseases: comparison with C-reactive protein. J Vet Med Sci 2011; 73(9):1–26. 18. Hauptman JG, Walshaw R, Olivier NB. Evaluation of the sensitivity and specificity of diagnostic criteria for sepsis in dogs. Vet Surg 1997; 26(5):393–397. 19. Veiga AP, Price CA, de Oliviera, ST, et al. Association of canine obesity with reduced serum levels of C-reactive protein. J Vet Diagn Invest 2008; 20(2):224–228. 20. Caspi D, Snel FW, Batt RM, et al. C-reactive protein in dogs. Am J Vet Res 1987; 48(6):919–921. 21. Burton SA, Honor DJ, Mackenzie AL, et al. C-reactive protein concentration in dogs with inflammatory leukograms. Am J Vet Res 1994; 55(5):613–618. 22. Fischer CL, Gill C, Forrester MG, et al. Quantitation of “acute-phase proteins” postoperatively. Value in detection and monitoring of complications. Am J Clin Pathol 1976; 66(5):840–846. 23. Le JM, Vilcek J. Interleukin 6: a multifunctional cytokine regulating immune reactions and the acute phase protein response. Lab Invest 1989; 61(6):588–602. 24. Abidi K, Khoudri I, Belayachi J, et al. Eosinopenia is a reliable marker of sepsis on admission to medical intensive care units. Crit Care 2008; 12(2):R59. 25. Park JS, Arcaroli J, Yum HK, et al. Activation of gene expression in human neutrophils by high mobility group box 1 protein. Am J Physiol Cell Physiol 2003; 284(4):C870–C879.
C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
Original Study
C-reactive protein and high mobility group box 1 in dogs with gastric dilatation and volvulus Ivana Uhrikova, MVDr, PhD; Leona Rauserova-Lexmaulova, MVDr, PhD; Kristina Rehakova, MVDr, PhD; Peter Scheer, MVDr, PhD and Jaroslav Doubek, MVDr, CSc
Abstract
Objective – To (1) measure C-reactive protein (CRP) and high mobility group box 1 (HMGB1) and (2) evaluate their prognostic value and relationship to severity of systemic inflammatory response syndrome, routine hematological and acid-base parameters in dogs with gastric dilatation volvulus (GDV). Design – Prospective observational study from September 2010 to June 2012. Setting – Veterinary teaching hospital. Animals – Forty-one client-owned dogs with GDV. Interventions – None. Measurements and Main Results – Blood was collected before surgery (baseline), postsurgery, 6–10 hours postsurgery, and 18–22 hours postsurgery. CRP and HMGB1 were measured in all samples, and routine hematological, biochemical, and acid-base analyses were performed. Only baseline and postsurgery samples were used from nonsurvivors (n = 10). CRP increased significantly from postsurgery sampling to 18–22 hours postsurgery, while HMGB1 did not change over time. There was a significant difference in HMGB1 between survivors and nonsurvivors over time. Both proteins correlated with systemic inflammatory response syndrome severity, total leukocyte, segmented neutrophils, and band counts. HMGB1 correlated also with acid-base parameters (pH, bicarbonate, base excess). Conclusion – HMGB1 and CRP behaved differently in regards to their kinetic patterns, with HMGB1 appearing to better reflect the severity of tissue injury in dogs with GDV than CRP. (J Vet Emerg Crit Care 2015; 25(4): 488–494) doi: 10.1111/vec.12324 Keywords: canine, mortality, prognosis, systemic inflammatory response syndrome
Abbreviations
CRP GDV HMGB1 OHE ROC
C-reactive protein gastric dilatation and volvulus syndrome high-mobility group box 1 ovariohysterectomy receiver operating curve
From the Department of Physiology, (Uhrikova, Scheer, Doubek), Small Animal Clinic, Department of Surgery and Orthopedics (RauserovaLexmaulova), Small Animal Clinical Laboratory (Rehakova, Doubek), University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic; and Integrated Center of Cellular Therapy (ICCT), St. Annes´s University Hospital - International Clinical Research Center Brno (Scheer), Brno, Czech Republic. The authors declare no conflict of interest. The study was supported by grant no. 24/2011/IGA VFU Brno. Presented, in part, at the ESVCP Congress, Ireland, 2011 and Veterinary and Comparative Clinical Pathology Conference, Slovenia, 2012. Address correspondence and reprint requests to Dr. Ivana Uhrikova, Department of Physiology, University of Veterinary and Pharmaceutical Sciences Brno, Palack´eho tˇr. 1/3, 612 42 Brno, Czech Republic. Email: [email protected] Submitted November 21, 2012; Accepted March 24, 2015.
488
SIRS UDL
systemic inflammatory response syndrome upper detection limit
Introduction Gastric dilatation and volvulus syndrome (GDV) is a severe disease in large and giant breed dogs that usually occurs in otherwise healthy animals. Onset of the clinical signs is acute and dogs may quickly fulfill the criteria of the systemic inflammatory response syndrome (SIRS).1 Surgical treatment with stomach decompression and repositioning can rapidly correct the primary problem and generally reverses accompanying complications. Due to its rapid onset and resolution as a clinical syndrome, we opted to study the development of SIRS in dogs with GDV. SIRS is associated with the increase in circulating concentration of proinflammatory cytokines and acute phase proteins.2 C-reactive protein (CRP) is a major acute phase protein in dogs, and has been widely studied in recent years.3–7 Its synthesis in the liver is stimulated C Veterinary Emergency and Critical Care Society 2015
CRP and HMGB-1 in dogs with GDV
by proinflammatory cytokines and its concentrations increase as early as 4 hours following the initial stimulus.8 Although a single measurement of this protein seems to have no prognostic significance in dogs with SIRS, knowledge of its kinetic pattern may help in the detection of emerging complications or worsening of the disease.4–6 High mobility group box 1 (HMGB1) is a nonhistone chromosomal protein that is highly conserved among various species with 100% amino acid homology between people and dogs.9 HMGB1 is an alarmin, a protein that is released extracellularly during cell necrosis but not apoptosis; it is also a late cytokine with active secretion after a lag phase of 8–18 hours. Outside the cells, it triggers inflammation by the stimulation of proinflammatory cytokine production by monocytes/macrophages, increased expression of adhesion molecules, and increased epithelial permeability.10,11 HMGB1 appears to be a good biomarker that relates to final outcome in some conditions in human medicine,12–15 but studies in veterinary medicine are limited.16,17 To our knowledge, there has been no study focused on HMGB1 in GDV to date. The aims of this study were to describe (1) the kinetic behavior and prognostic value of CRP and HMGB1 in dogs with GDV and (2) relate CRP and HMGB1 concentrations to findings of routine hematological and biochemical parameters.
Material and Methods Animals Dogs with GDV were included in this prospective observations study conducted between September 2010 and June 2012. Inclusion criteria included the diagnosis of GDV based upon compatible clinical signs, positive radiographic examination, and its confirmation during surgical treatment. We obtained signed consent from dog owners for participation in this study. All dogs were sedated by IV administration of midazolam 0.2 mg/kg, butorphanola 0.4 mg/kg, and ketamineb 2 mg/kg, and general anesthesia was induced with propofolc . In addition, buprenorphined 0.005 mg/kg was administered after 20 minutes in most of the patients. General anesthesia was maintained with an inhalation mixture of oxygen (50%) in air and isofluranee with artificial ventilation. Surgical treatment with decompression and derotation of the stomach and gastropexy was performed in all patients according to their individual needs. In cases of operable gastric necrosis, partial gastrectomy was performed. In cases of inoperable gastric necrosis, dogs underwent euthanasia. All dogs received infusion therapy consisting of crystalloidf and colloid (hydroxyethyl starch)g solutions based on results from C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
clinical and acid-base examination. None of the dogs received a blood transfusion. Supplementary data Data related to the time since last feeding and time since onset of clinical signs were obtained from the owners. The surgeon determined the degree of gastric rotation based on a subjective evaluation (eg, 120°, 180°, 270°, 360°). Dogs were assigned into the SIRS group if ࣙ2 of the following criteria were fulfilled18 : tachycardia (>120/min), tachypnea (>20/min), increased (>39.0°C) or decreased (16 × 109 /L [>16000/L]), leukopenia (6.67 ng/mL with sensitivity 88.9% and specificity 77.4%; for CRP the cutoff was C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
Figure 1: Survival prediction fitted receiver operating curve (ROC) for high mobility group box 1 (HMGB1) and C-reactive protein (CRP) from baseline samples. Cut-off values for poor short-term outcome were HMGB1 >6.67 ng/mL (sensitivity 88.9%, specificity 77.4%) and CRP >7.2 mg/L (sensitivity 90.0%, specificity 43.3%). The AUC was 0.849 (95% CI: 0.70–0.94; P < 0.0001) for HMGB1 and 0.696 (95% CI: 0.53–0.83; P < 0.05) for CRP.
>7.2 mg/L with sensitivity 90.0% and specificity 43.3%. The AUC was 0.849 (95% CI: 0.70–0.94; P < 0.0001) for HMGB1 and 0.696 (95% CI: 0.53–0.83; P < 0.05) for CRP. Correlations A moderate correlation was observed between the time from last feeding and CRP concentrations in baseline and postsurgery samples (r = 0.53 and r = 0.56, P < 0.001). HMGB1 correlated moderately with time from 491
I. Uhrikova et al.
Table 2: Absolute and differential leukocyte counts over time in dogs with gastric dilatation volvulus stratified by outcome
Sample Absolute leukocyte count (×103 /L or ×109 /L) Segmented neutrophils (×103 /L or ×109 /L) Neutrophil bands (×103 /L or ×109 /L) Monocytes (×103 /L or ×109 /L) Lymphocytes (×103 /L or ×109 /L) Eosinophils (×103 /L or ×10‘ /L)
All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors All dogs Survivors Nonsurvivors
Baseline
Postsurgery
9.8 (4.1–26.7) 9.6 (4.1–26.7) 17.0 (7.4–26.6) 8.8 (4.3–24.7) 8.3 (4.3–21.9) 15.1 (5.1–24.7) 0.1 (0.0–3.3) 0.1 (0.0–1.3) 0.5 (0.0–3.3) 0.5 (0.0–2.1) 0.6 (0.0–1.9) 0.5 (0.1–2.1) 0.6 (0.0–2.9) 0.6 (0.1–2.7) 0.4 (0.0–2.9) 0.0 (0.0–0.8) 0.1 (0.0–0.8) 0.0 (0.0–0.2)
7.4 (1.3–21.5) 7.3 (1.3–21.5) 8.7 (1.4–17.3) 5.6 (0.8–15.7) 5.8 (0.8–14.2) 5.4 (0.8–15.7) 0.1 (0.0–1.1) 0.1 (0.0–1.1) 0.3 (0.1–0.7) 0.3 (0.0–4.3) 0.3 (0.0–4.3) 0.3 (0.0–1.0) 0.5 (0.0–2.1) 0.5 (0.0–1.9) 0.2 (0.1–2.1) 0.1 (0.0–1.1) 0.1 (0.0–1.1) 0.0 (0.0–0.2)
6–10 hours postsurgery
18–22 hours postsurgery
19.3 (5.3–35.4) 19.3 (5.3–35.4)
19.8 (6.0–44.2) 19.8 (6.0–44.2)
17.8 (4.6–27.3) 17.8 (4.6–27.3)
16.6 (2.9–37.1) 16.6 (2.9–37.1)
0.5 (0.0–7.8) 0.5 (0.0–7.8)
0.2 (0.0–6.6) 0.2 (0.0–6.6)
0.4 (0.0–2.6) 0.4 (0.0–2.6)
0.5 (0.0–2.1) 0.5 (0.0–2.1)
0.6 (0.0–2.3) 0.6 (0.0–2.3)
1.3 (0.0–4.2) 1.3 (0.0–4.2)
0.0 (0.0–0.2) 0.0 (0.0–0.2)
0.0 (0.0–1.1) 0.0 (0.0–1.1)
Values are shown as median (minimum-maximum).
last feeding in baseline samples (r = 0.42, P < 0.01) and the strongest correlation (r = −0.59, P < 0.01) was seen 6–10 hours postsurgery between HMGB1 and the degree of gastric rotation. There was a mild correlation between the severity of SIRS and CRP concentration as well as HMGB1 at baseline (r = 0.32 and r = 0.45, P < 0.05) and severity of SIRS 6–10 hours after surgery and CRP at baseline (r = 0.46, P = 0.01). Results from the hematologic examination are shown in Table 2. The correlations noted between CRP and hematologic parameters were: a moderate correlation with leukocyte counts at baseline sampling (r = 0.66, P < 0.0001), segmented neutrophils at baseline sampling (r = 0.69, P < 0.0001), band neutrophils at postsurgery and 18–22 hours postsurgery samplings (r = 0.48 and r = 0.57, P < 0.01 and P = 0.01) and eosinophils at postsurgery sampling (r = −0.4, P = 0.01). Similarly, HMGB1 was correlated with leukocyte counts at baseline sampling (r = 0.47, P < 0.01), segmented neutrophils at baseline sampling (r = 0.40, P < 0.05), band neutrophils (r = 0.45, P < 0.01), and eosinophils at postsurgery sampling (r = −0.48, P < 0.01). A moderate correlation was also found with lactate concentration at baseline and postsurgery samplings (r = 0.63 and 0.62, P < 0.001) and selected acid-base parameters (pH, HCO3− , base excess) at baseline, postsurgery, and 18–22 hours postsurgery samplings (r = −0.35 to −0.45, P < 0.05). There was no correlation among CRP or HMGB1 and the remaining biochemical parameters that were measured. Correlation studies among sampling points revealed only moderate
492
correlation (from r = 0.48 to 0.45, P < 0.01) between CRP or HMGB1 (both baseline) and the number of band neutrophils (postsurgery).
Discussion To date, only a few studies have investigated the changes of CRP concentrations in dogs.4,17 The current study was focused on serial measurement of CRP and the nuclear protein, HMGB1 in dogs with GDV, their prognostic value and relation to SIRS, and selected laboratory parameters. Although CRP has received more attention in veterinary medicine than HMGB1, publications focusing on CRP in dogs with GDV are limited.7 In our study, there was no significant difference in presurgical CRP concentrations between survivors and nonsurvivors, although median values were higher in nonsurvivors (8.0 vs. 12.8 mg/L). These findings are similar to those of Israeli et al,7 although the median CRP concentration in all dogs was far lower in our study (9.1 vs. 23.0 mg/L), and the range of values was wider than that published by Israeli et al. (0.3–167.7 vs. 0.1–111.0 mg/L).7 We found a moderate correlation between CRP and time since last feeding (r = 0.53, P < 0.001) and thus differences between studies could be attributed to the different times of arrival at the clinic. The median value of CRP at admission in survivors was within the reference interval declared by the manufacturern (0–25 mg/L), but higher than that found in 30 healthy dogs sampled in our laboratory (median 2.5, range 0–12.6 mg/L, unpublished).
C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
CRP and HMGB-1 in dogs with GDV
In dogs in the current study CRP gradually increased over time. It is well known that CRP increases after surgery,17,20,21 and thus our findings are consistent with those observations. The magnitude of increase varies across studies. Following ovariohysterectomy, CRP concentrations increased from 4- to 44-fold with a peak at 24–28 hours.20,21 Such a steep increase in CRP concentration following to surgery complicates interpretation of its concentration in early phases of wound healing. A correlation between the invasiveness of surgery and magnitude of CRP increase was described previously,20,r and under the same surgical conditions it seems that CRP values are predictive for complications as late as the third day following surgery.4,22 In our study, there was only a weak association between CRP baseline concentrations and SIRS severity at admission as well as 6–10 hours postsurgery, which may reflect the intensity of tissue damage. Despite the small number of dogs undergoing partial gastrectomy we can conclude, that CRP concentration 6–10 hours postsurgery is higher in these dogs confirming the influence of the invasiveness of surgery on CRP concentrations. A correlation between CRP and hematologic parameters was noted. This is not surprising since the maturation of neutrophils is stimulated by the same proinflammatory cytokines that act on liver to stimulate CRP synthesis.23 The number of neutrophil bands (18–22 hours postsurgery) and segmented neutrophils (6–10 and 18–22 hours postsurgery) was higher in dogs undergoing partial gastrectomy in agreement with CRP concentrations. HMGB1 has been intensively studied in many fields of human medicine.24 This nuclear DNA-binding protein is passively released during cell necrosis and actively as a late cytokine. Its concentration increases within 18 hours following the administration of endotoxin.10,11 Since most of the dogs were sampled less than 18 hours after onset of clinical signs or time since last feeding, we believe that most of the initial HMGB1 originated from its release from damaged cells. The moderate correlation between lactate and HMGB1 (baseline and postsurgery) supports this hypothesis, since lactate in GDV probably originates from hypoperfused tissues. This could also explain the correlation between HMGB1 and the “metabolic” part of the acid-base status (eg, bicarbonate, base excess). Finally, significant differences in HMGB1 between dogs with and without gastric necrosis suggest that all of these findings are more closely associated with the viability of the stomach wall than tissue hypoperfusion due to the shock state. HMGB1 has been the subject of interest in few studies in veterinary medicine, although significant differences between dogs with and without SIRS as well as between survivors and nonsurvivors with SIRS have been described.16,17 Our study C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
also found a significant difference in HMGB1 between survivors and nonsurvivors as well as a correlation with the severity of SIRS. The effect of surgery on HMGB1 release in dogs after ovariohysterectomy was studied in 5 dogs by Ishida et al.17 HMGB1 increased during the first 12 hours, then decreased at 24 hours, and finally gradually increased with a peak at 72 hours. In our study, levels of HMGB1 were not significantly different among different time points. Although, the number of dogs undergoing partial gastrectomy was too low to allow a statistical comparison of HMGB1 difference between dogs undergoing partial gastrectomy with gastropexy and solitary gastropexy, the median HMGB1 was 2-fold higher in dogs with partial gastrectomy compared to nongastrectomized dogs. In the 6–10 and 18–22 hours postsurgery samplings median values were similar in these 2 groups. Of these 2 parameters, initial HMGB1 concentration may be used as an indicator of poor outcome in dogs with GDV since its AUC was higher than for CRP. Additionally, absence of gradual increase of HMGB1 following surgery may be exploited for the early detection of late postsurgical complications in comparison to CRP. However, its measurement is limited by the lack of point-of-care HMGB1 assays. Its advantage over other biomarkers is its homology with human HMGB1,9 thus there is a high probability that future development of point-of-care assays for people will also be useful in dogs. A correlation between HMGB1 and leukocyte count and neutrophils was also observed. This correlation may be due to HMGB1 release from injured cells, which stimulates infiltration of the damaged site with neutrophils. The correlation with band neutrophils and eosinophils may occur for the same reasons as in the case of CRP, since HMGB1 stimulates the release of other proinflammatory cytokines from monocytes/macrophages.10,11,25 We are aware of several limitations of this study. First, the number of dogs included was relatively small. Second, the data related to the time from onset of clinical signs and degree of gastric rotation were based on subjective observations. In addition, the data related to time of last feeding submitted by the owners could contain a degree of error. Although the duration of surgery varied amongst cases, the anesthetic and surgical protocols were standardized in all dogs as much as possible. Postoperative care within the first 24 hours was similar in all dogs, although there may have been small differences in the rate and volume of fluid therapy following surgery. Blood collection as well as sample processing was performed by the same 2 investigators, and the differential leukocyte count was performed by the same investigator for all samples to prevent variations in the preanalytical, analytical, or postanalytical phase. Third, the decision 493
I. Uhrikova et al.
concerning whether gastric necrosis was operable or not (resulting in euthanasia) was made by the attending surgeon; however, whenever possible, partial gastrectomy was performed. In addition, the collection of all samples at the same time rather than at 4 hour intervals would be ideal; but most dogs were admitted during the night, and thus 6–10 and 18–22 hours postsurgery samples were taken in the morning and evening of the next day. Finally, dilution of samples with values beyond the UDL might have allowed us to demonstrate more significant results. Despite all of these limitations, we have identified some important findings.
Conclusion In dogs with GDV, the kinetics of HMGB1 differed from that of CRP and initial HMGB1 values may be reflective of the degree of tissue damage found at surgery. Both parameters correlated only mildly with SIRS severity at the admission. Further research on the uses of HMGB1 assessment in dogs is warranted.
Footnotes a b c d e f
g h i j k l m n o p q r
Butomidor, Richter Pharma, Wels, Austria. Narkamon 5% a.u.v. inj., Spofa a.s., Czech Republic. Norofol 1%, Norbrook Laboratories Limited, Germany. Vetergesic, Reckitt Beckiser Healthcare, Berkshire, UK. Isoflurane, Nicholas Piramal India Ltd., UK. 0.9% NaCl, Sodium chloride, B. Braun Medical, Prague, Czech Republic; Hartmann’s Viaflo, Plasmalyte solution, Baxter Czech spol. s.r.o., Prague, Czech Republic, Ringer’s injection “fresenius,” Fresenius Kabi Italia S.r.l., Verona, Italy. Voluven, Baxter Czech spol. s.r.o. Dispolab s.r.o., Brno, Czech Republic. Sarstedt s.r.o., Brno, Czech Republic. Celltac Alpha, Nihon Kohden, Tokyo, Japan. Hemacolor, Biovendor, a.s., Brno, Czech Republic. Konelab DCP20i, Thermo Scientific, Vantaa, Finland. EasyBloodGas, Medica a.s., Czech Republic. CRP, Biovendor a.s. HMGB1, IBL International, GmbH, Hamburg, Germany. STATISTICA 6.0, StatSoft, Inc., Tulsa, OK. MedCalc Software bvba, Ostend, Belgium. Kjelgaard-Hansen M, Strom H, Mikkelsen LF, et al. Grading of surgical trauma by means of canine C-reactive protein measurements. In Proceedings of the 2008 ASVCP Annual Meeting.
References 1. Silverstein DC, Montealegre C, Shofer FS, et al. The association between vascular endothelial growth factor levels and clinically evident peripheral edema in dogs with systemic inflammatory response syndrome. J Vet Emerg Crit Care 2009; 19(5):459– 466. ¨ PO. The systemic inflammatory response syndrome: def2. Nystrom initions and aetiology. J Antimicrob Chemother 1998; 41(Suppl A): 1–7. 3. Chan DL, Rozanski EA, Freeman LM. Relationship among plasma amino acids, C-reactive protein, illness severity, and outcome in critically ill dogs. J Vet Intern Med 2009; 23(3):559–563.
494
4. Gebhardt C, Hirschberger J, Rau S, et al. Use of C-reactive protein to predict outcome in dogs with systemic inflammatory response syndrome or sepsis. J Vet Emerg Crit Care 2009; 19(5): 450–458. 5. Galezowski AM, Snead EC, Kidney BA, et al. C-reactive protein as a prognostic indicator in dogs with acute abdomen syndrome. J Vet Diagn Invest 2010; 22(3):395–401. 6. Whittemore JC, Marcum BA, Mawby DI, et al. Associations among albuminuria, C-reactive protein concentrations, survival predictor index scores, and survival in 78 critically ill dogs. J Vet Intern Med 2011; 25(4):818–824. 7. Israeli I, Steiner J, Segev G, et al. Serum pepsinogen-A, canine pancreatic lipase immunoreactivity, and C-reactive protein as prognostic markers in dogs with gastric dilatation-volvulus. J Vet Intern Med 2012; 26(4):920–928. 8. Yamashita K, Fujinaga T, Myiamoto T, et al. Canine acute phase response: relationship between serum cytokine activity and acute phase protein in dogs. J Vet Med Sci 1994; 56(3):487–492. 9. Murua Escobar H, Meyer B, Richter A, et al. Molecular characterization of the canine HMGB1. Cytogenet Genome Res 1993; 101(1):33– 38. 10. Andersson J, Erlandsson HH, Yang H, et al. HMGB1 as a DNAbinding cytokine. J Leukoc Biol 2002; 72(6):1084–1091. 11. Yang H, Wang H, Czura CJ, et al. The cytokine activity of HMGB1. J Leukoc Biol 2005; 78(1):1–8. 12. Huang LF, Yao YM, Dong N, et al. Association of high mobility group box-1 protein levels with sepsis and outcome of severely burned patients. Cytokine 2011; 53(1):29–34. 13. Wang KY, Yu GF, Zhang ZY, et al. Plasma high-mobility group box 1 levels and predtiction of outcome in patients with traumatic brain injury. Clin Chem Acta 2012; 413(21–22):1737–1741. 14. Huang JM, Hu J, Chen N, et al. Relationship between plasma highmobility group box-1 levels and clinical outcomes of ischemic stroke. J Crit Care 2013; 28:729–797. 15. Fahmueller YN, Nagel D, Hoffmann RT, et al. Immunogenic cell death biomarkers HMGB1, RAGE, and DNAse indicate response to radioembolization therapy and prognosis in colorectal cancer patients. Int J Cancer 2013; 132(10):2349–2358. 16. Yu DH, Nho DH, Song RH, et al. High-mobility group box 1 as a surrogate prognostic marker in dogs with systemic inflammatory response syndrome. J Vet Emerg Crit Care 2010; 20(3):298– 302. 17. Ishida A, Ohno K, Fukushima K, et al. Plasma high-mobility group box 1 (HMGB1) in dogs with various diseases: comparison with C-reactive protein. J Vet Med Sci 2011; 73(9):1–26. 18. Hauptman JG, Walshaw R, Olivier NB. Evaluation of the sensitivity and specificity of diagnostic criteria for sepsis in dogs. Vet Surg 1997; 26(5):393–397. 19. Veiga AP, Price CA, de Oliviera, ST, et al. Association of canine obesity with reduced serum levels of C-reactive protein. J Vet Diagn Invest 2008; 20(2):224–228. 20. Caspi D, Snel FW, Batt RM, et al. C-reactive protein in dogs. Am J Vet Res 1987; 48(6):919–921. 21. Burton SA, Honor DJ, Mackenzie AL, et al. C-reactive protein concentration in dogs with inflammatory leukograms. Am J Vet Res 1994; 55(5):613–618. 22. Fischer CL, Gill C, Forrester MG, et al. Quantitation of “acute-phase proteins” postoperatively. Value in detection and monitoring of complications. Am J Clin Pathol 1976; 66(5):840–846. 23. Le JM, Vilcek J. Interleukin 6: a multifunctional cytokine regulating immune reactions and the acute phase protein response. Lab Invest 1989; 61(6):588–602. 24. Abidi K, Khoudri I, Belayachi J, et al. Eosinopenia is a reliable marker of sepsis on admission to medical intensive care units. Crit Care 2008; 12(2):R59. 25. Park JS, Arcaroli J, Yum HK, et al. Activation of gene expression in human neutrophils by high mobility group box 1 protein. Am J Physiol Cell Physiol 2003; 284(4):C870–C879.
C Veterinary Emergency and Critical Care Society 2015, doi: 10.1111/vec.12324
