Search  in
Original Articles
NUMBER 3-4 YEAR 2011
An Improved Model of Severe Sepsis in Pigs
1 Institute of Surgical Research, University of Szeged, Hungary,
2 Pius Branzeu Center for Microsurgery and Laparoscopic Surgery, Victor Babes University of Medicine and Pharmacy, Timisoara,
3 Department of Medical Microbiology and Immunobiology, University of Szeged, Hungary

Correspondence to:
Mihály Boros MD, PhD, DSc, Institute of Surgical Research, University of Szeged,
PO Box 427, H-6701 Szeged, Hungary,
Tel: +36-62-545103
Obiectiv: Dezvoltarea unui model de studiu al sepsei pe animalul mare de laborator prin inducerea peritonitei fecale, cu reproducerea caracteristicilor macrohemodinamice, microcirculatorii si inflamatorii, obiectivate in sepsa umana incipienta. Material si metode: Porcii pitici anesteziati au fost supusi peritonitei fecale (n = 9; 0,5 g/kg i.p., autofecale) sau operatiei-control (ser fiziologic i.p., n = 6). Gazimetria a fost monitorizata hemodinamic invaziv la 15-24 ore postchirurgical. Microcirculatia sublinguala a fost exprimata prin modificarile velocitatii eritrocitare (MVE) (prin imagistica spectrala a polarizarii octogonale), in timp ce intervalul PCO2 intestinal a fost masurat prin tonometrie indirecta. Nivelele plasmatice ale proteinei HMGB1 si oxidului nitric/nitrat (NOx) au fost determinate din probe de sange venos. Rezultate: Presiunea arteriala medie a scazut treptat sub 70 mm Hg la animalele cu sepsis, in timp ce frecventa cardiaca si debitul cardiac au crescut constant. In ciuda reactiilor hiperdinamice, MVE a scazut, in timp ce intervalul PCO2 a crescut semnificativ, in comparatie cu grupul control. Concentratiile plasmatice ale NOx si HMGB1 au fost semnificativ crescute la 6-16 ore de la debutul peritonitei. Concluzii: Raportam dezvoltarea unui nou model experimental de studiu al sepsei prin inducerea peritonite fecale la animalul de laborator. Datele obtinute in vivo sugereaza ca acest model experimental este relevant clinic si poate avea un rol inovator in dezvoltarea terapiilor anti-sepsa.

Objective: Our aim was to develop a large animal model of sepsis, induced by fecal peritonitis, which reproduces the characteristic macrohemodynamic, microcirculatory and inflammatory changes seen in early human sepsis. Material and methods: Anesthetized minipigs were subjected to fecal peritonitis (n = 9; 0.5 g/kg i.p. autofeces) or sham-operation (i.p. saline, n = 6). Invasive hemodynamic monitoring was started with regular blood gas analyses between the 15-24 hr of the insult. Sublingual microcirculation was characterized by red blood cell velocity (RBCV) changes (with orthogonal polarization spectral imaging), while the intestinal PCO2 gap was measured by indirect tonometry. The plasma levels of high mobility group box protein 1 (HMGB1) and nitrite/nitrate (NOx) were determined from venous blood samples. Results: The mean arterial pressure gradually decreased below 70 mm Hg in septic animals, while the heart rate and cardiac output increased constantly. In spite of the hyperdynamic reaction the sublingual RBCV decreased, while intestinal PCO2 gap increased significantly as compared with the control group. The NOx and HMGB1 plasma concentrations were significantly elevated between 6-16 hr of peritonitis. Conclusion: We report on the introduction of a new animal model fecal peritonitis-induced sepsis. The in vivo data suggest that this experimental model is of clinical relevance and may play useful roles in the development of novel, sepsis-related therapies.

Sepsis remains a leading cause of mortality in intensive care units (ICUs), without specific therapeutic options. Basic medical research utilizing animal models has provided a greater understanding of its underlying mechanisms, and today the involvement of canonical inflammatory pathways such as the activation of the complement system, leukocytes, lipid mediators and adhesions molecules has been relatively well defined. Nevertheless, the parallel development of clinical treatment strategies did not progress significantly during the last decades. Since similarities between the human and animal responses to septic insults should help to understand the key factors of therapeutic control, the lack of success clearly demonstrates that the currently used in vivo models do not completely mimic human septic conditions.1,2 By definition, a model is a tool via which to understand or describe a system or phenomenon, and the "goodness of fit" through usability testing is its most important characteristic. This suggests that a model with improved clinical relevance may play a more useful role in the development of novel, sepsisrelated therapies.3
Our goal was to narrow the gap between experimental and human sepsis by putting much greater emphasis on the reproduction of ICU scenarios and the consequences of intensive treatment. By this approach, not only the septic insult-induced changes, but other, therapy-caused alterations could be considered, similarly to those seen in clinical conditions. To this aim a standardized porcine model of protracted peritonitis was created using intraperitoneal autologous fecal inoculum and complex resuscitation strategies. This communication reports on the most important specific features of the setup, including reproducible macrohemodynamic and microcirculatory variables, and characteristic biochemical data.


The study protocol was approved by the EthicalCommittee for the Protection of Animals in ScientificResearch at the University of Szeged.
The experiments were performed on Vietnameseminipigs of both sexes (average weight 23 ± 3kg)which were underwent a 16-hr preoperative fastingwith water ad libitum; the animals were randomlyallocated into control (sham-operated; n = 6) andseptic groups (n = 9). Anesthesia was induced withan intramuscular injection of a mixture of ketamine(20 mg/kg) and xylazine (2 mg/kg) and maintainedwith a continuous infusion of propofol (6 mg/kg/hriv). An endotracheal tube was inserted and the animalswere ventilated mechanically with room air (HarvardApparatus, South Natick, MA, U.S.A.) The tidalvolume was set at 12±2 mL/kg, and the respiratoryrate was adjusted to maintain the end-tidal carbondioxide pressure (controlled by capnometry) and thepartial arterial carbon dioxide pressure in the rangeof 35-45 mm Hg (4.6-5.9 kPa). The adequacy of thedepth of anesthesia was assessed by monitoring thejaw tone regularly. A central venous catheter with threelumina (7 F; Edwards Lifesciences LLC, Irvine, U.S.A)was introduced into the jugular vein using asepticsurgical technique, for blood sampling and for fluidadministration, respectively.
Sepsis was induced with an intraperitonealinjection of autofaeces mixture into the abdominalcavity (0.5 g/kg faeces in 200 ml saline and cultivatedat 38°C through 6 hrs until the induction ofperitonitis). Control animals were treated with 200ml of sterile saline in the same manner. Thereafter,a single nalbuphine (0.5 mg/kg iv) injection was used for postoperative analgesia and the sedated animalswere brought back to their cages.
Invasive haemodynamic monitoring was started15 hr later. A transpulmonary thermodilution catheter(PiCCO, PULSION Medical Systems AG, Munich,Germany) was placed in the femoral artery and apulmonary artery catheter (PV2057 VoLEF Catheter,PULSION Medical Systems AG, Munich, Germany)was introduced via the femoral vein by tracing thepressure signals. A midline laparotomy was performedand a tonometric probe was introduced into the smallintestine through a small incision to record mucosalpCO2 changes. Thereafter the most commonly usedvolume therapies were applied in the septic groupof animals. Infusion of crystalloid-colloid fluidcombinations [lactated Ringer's solution at a rate of10 ml/kg/hr; hydroxyethyl starch in a dose of 5 ml/kg/hr (Voluven 6%; 130 kDa/0.4; Fresenius KabiDeutschland GmbH, Homburg, Germany)] wasstarted for supporting intravascular volume through 4hrs (between 16th and 20th hour of the experiments). Atthe final part of the study period vasopressor therapy(0.015 µg/kg/hr norepinephrine in 20 ml saline iv)was started if the MAP decreased under 65 mmHgto avoid the kidney failure (the pressor treatment wasnecessary in 4 out of 9 animals). The sham-operatedgroup was infused with crystalloid solution (lactatedRinger's solution) during this time at a rate of 10 ml/kg/hr.

Hemodynamic measurements
Mean arterial pressure (MAP) and cardiac outputwere registered by PiCCO monitor, while centralvenous pressure (CVP) and pulmonary artery pressure(PAP) signals were monitored continuously witha computerized data-acquisition system (SPELLHaemosys; Experimetria Ltd., Budapest, Hungary).The systemic vascular resistance (SVR) was calculatedvia the standard formula TPR = (MAP - CVP)/cardiacoutput. pCO2 gap measurements
A difference between local tissue and arterialpCO2 (paCO2) levels is a sensitive parameter withwhich to evaluate the effectiveness of therapy aimedat counteracting a microcirculatory dysfunction inthe gastrointestinal (GI) tract.4 A silastic balloonfreetonometric probe (Tonosoft Medical Technicaland R&D Ltd., Hungary) was introduced through asmall enterotomy into the intestinal lumen to monitorintramucosal pCO2 levels by capnometry.5 Forcalculation of the pCO2 gap values, simultaneouslytaken paCO2 levels were subtracted from thetonometric pCO2 levels. Arterial blood sampleswere taken regularly, and blood-gas parameters were measured with a blood-gas analyzer (Cobas b121, Roche, Austria).

Intravital videomicroscopy of the micro-circulation
The intravital orthogonal polarization spectral (OPS) imaging technique (Cytoscan A/R, Cytometrics, Philadelphia, Pennsylvania, USA) was used for non-invasive visualization of the sublingual microcirculation.6 The OPS method makes use of reflected polarized light, which allows noninvasive imaging of the microcirculation on the surface of solid organs without the need for fluorescence contrast enhancement. In brief, linearly polarized light is scattered in the tissue and serves as a virtual light source. Images are obtained at 548 nm wavelength, which is the isobestic point for oxy- and deoxyhemoglobin. In this way red blood cells in the microcirculation appear in black on the white background of the surrounded tissue. A 10x objective was placed onto the surface of the sublingual area, and microscopic images were recorded with an S-VHS video recorder (Panasonic AG-TL 700, Matsushita Electric Ind. Co. Ltd, Osaka, Japan). Quantitative assessment of the microcirculatory parameters was performed off-line by frame-to-frame analysis of the videotaped images. Red blood cell velocity (RBCV, μm/s) changes in the postcapillary venules were determined in three separate fields by means of a computer-assisted image analysis system (IVM Pictron, Budapest, Hungary). All microcirculatory evaluations were performed by the same investigator.

Plasma nitrite/nitrate level measurements
The levels of plasma nitrite/nitrate (NOx), stable end-products of nitric oxide (NO), were measured by the Griess reaction. The assay depends on the enzymatic reduction of nitrate to nitrite, which is then converted into a colored azo compound detected spectrophotometrically at 540 nm.7

High mobility group box protein 1 measurements in plasma
Two-ml blood samples were drawn from the jugular vein into chilled polypropylene tubes containing EDTA (1 mg ml-1) at baseline, the 6th, 16th hour and at the end of the observation period (24th). The blood samples were centrifuged at 1200g for 10 min at 4oC. The plasma samples were then collected and stored at -70oC until assay. Plasma concentration of high mobility group box protein 1 (HMGB1) was measured by a commercially available HMGB1 ELISA kits (Shino-Test Corporation, Kanagawa, Japan).

Statistical analysis
Data analysis was performed with a statistical software package (SigmaStat for Windows, Jandel Scientific, Erkrath, Germany). Nonparametric methods were used. Friedman repeated measures analysis of variance on ranks was applied within the groups. Time-dependent differences from the baseline (time 0), or from the beginning of the invasive monitoring (the 16th hr of the observation period) for each group were assessed by Dunn's method, and differences between groups were analyzed with Mann-Whitney test. In the Figures, median values and 75th and 25th percentiles are given. P values < 0.05 were considered significant.


In the control group, there were no significant hemodynamic changes as compared with the baseline values, and the plasma mediator levels did not change significantly during the observation period.
There were statistical significant differences in hemodynamics between control and septic groups. The MAP gradually decreased below 70 mmHg in septic animals (Fig. 1A), while the heart rate increased constantly and exceeded the control values significantly (Fig. 1B).

Figure 1.A Changes in the mean arterial pressure (A) and the heart rate (B) in the control and septic groups. The plots demonstrate t [...]
Figure 1.B Changes in the mean arterial pressure (A) and the heart rate (B) in the control and septic groups. The plots demonstrate t [...]

In the septic group, the cardiac index was remarkably elevated and surpassed significantly the control level from the 20th hr. (Fig. 2A) These changes resulted in dramatic decreases of the SVR. In the septic group the SVR started to decrease at the 18th hr of the experiments and reached the deeper point at 20th hr and then the values were kept at this low level until the end of the study. (Fig. 2B)

Figure 2.A Changes in cardiac output (A) and the the systemic vascular resistance (B) in the control and septic groups. The plots dem [...]
Figure 2.B Changes in cardiac output (A) and the the systemic vascular resistance (B) in the control and septic groups. The plots dem [...]

The pCO2 gap is the difference between the local tissue and the arterial pCO2, and a reliable index of local tissue perfusion. The pCO2 gap of the small intestine in the septic group increased significantly at the beginning of the invasive monitoring (from the 16th hr of sepsis), and remained significantly higher than that for the sham-operated control group. (Fig. 3A)
The analysis of the sublingual microcirculation, which represents the peripheral microperfusion status, revealed a gradually decreasing and statistically significantly lower red blood cell velocity in the septic group as compared to both baseline values and the control group, from the 16th hr of the experiment. (Fig. 3B)

Figure 3.A Changes in the intestinal pCO2 gap (A) and the red blood cell velocity of the sublingual microcirculation (B) in the contr [...]
Figure 3.B Changes in the intestinal pCO2 gap (A) and the red blood cell velocity of the sublingual microcirculation (B) in the contr [...]

The NOx concentration in the plasma gives an estimate of the changes in NO production. Sepsis induction resulted in a statistically significant increase in NOx level as compared with the baseline values and to the control group. (Fig. 4A)
The plasma HMGB1 concentration gradually increased approx. 5-fold by 16 hr after of the induction of sepsis and remained significantly higher than in the control group up to 24 hr in the observation period. (Fig. 4B)

Figure 4.A Changes in the plasma levels of nitrite/nitrate (A) and the high mobility group box protein 1 (B) in the control and septi [...]
Figure 4.B Changes in the plasma levels of nitrite/nitrate (A) and the high mobility group box protein 1 (B) in the control and septi [...]


Several animal models try to replicate the signs and laboratory findings seen in human sepsis, where systemic hemodynamics evolves from an early hyperdynamic ("warm shock") state to a late hypodynamic ("cold shock") state. Such models include intraperitoneally, intravenously or intrapulmonally-administered endotoxin (lipolysaccharide) or live bacteria.8,9 Cecal ligation and perforation or induction of bacterial peritonitis with fecal inoculum are also frequently used.10,11
However, it should be stressed that hemodynamic and microcirculatory measurements together with blood sample collections are difficult to perform in rodents, and short-term, hypodynamic models have limited clinical relevance.2
These are all characterized by initial hypotension and low cardiac output, in contrast to the hyperdynamic circulation commonly seen in patients with septic shock. Furthermore, rodent reactions are markedly differ from human with respect to the immune response to sepsis or the tissue antioxidative capacity and susceptibility to oxidative stress.12,13
Indeed, several therapies proposed based on promising results obtained by these approaches could not be brought to the clinic, and many reviews have shown that the failure to translate basic science results from animals to humans has been mainly attributed to inappropriate animal models that do not fully mimic human sepsis.
In our study we switched to larger animals, similarly to Barth et al., who used a comparable, long-term porcine fecal peritonitis model, in which the hyperdynamic circulatory reaction was present between 12-18 hrs after sepsis induction.14 However, we did not employ 24 hrs anesthesia, and in contrast to the Barth's model, the hemodynamic monitoring started later, 15 hr after the insult.
In this setup the intra-abdominal septic insult could be accurately standardized with the pre-defined amount fecal inoculum, and consequently, the inflammatory-hemodynamic responses were reproducible.
It has been demonstrated in ICU patients that normalization of hemodynamic status in itself is insufficient to prevent the postoperative complications, since existing microcirculatory disturbances can lead to nutritive insufficiency and multiorgan failure in spite of seemingly adequate macrocirculation.15
Non-invasive imaging techniques, such as OPS and its successor Sidestream Dark Field (SDF) imaging are optical techniques allowing assessment of the microperfusion in case of solid organs, covered with thin epithelial layer, as the sublingual mucosal surfaces.6,15 Indeed, it has been demonstrated in humans that improvement in the sublingual microvascular perfusion, as early as 24 hrs after the onset of shock, can be a good predictor of ICU mortality.16
Our study demonstrated different patterns of microvascular alterations between septic and control animals in both intestinal and sublingual regions, and these data clearly indicated the presence of substantial intestinal and peripheral hypoxia in spite of the hyperdynamic macrocirculation.
The microcirculatory reaction could be a consequence of the altered synthesis of NO and proinflammatory cytokines. It is generally accepted that NO produced by constitutive NOS (cNOS), including nNOS and eNOS, are important homeostatic regulators of numerous important physiological functions.
In contrast, inducible NO synthase (iNOS) is produced by inflammatory cells, induced by various stimuli such as inflammatory cytokines and bacterial endotoxin, and plays an important role in inflammation.17 Thus, a long lasting elevation of the NOx level of plasma may be considered a hallmark of inflammation in the present study.
A systemic response to infection is characterised by early-phase inflammatory cytokines, such as tumor necrosis factor (TNF)-alpha or interleukin (IL)-1 beta. In this respect the high-mobility group box-1 protein (HMGB1) has recently been shown to be a late-phase mediator of sepsis with close correlation with the severity of the process.18,19 In our study the plasma HMGB1 levels peaked 16 hr after the start of septic reaction and remained significantly elevated.
HMGB1 is secreted by activated monocytes or macrophages, and is released by necrotic or damaged cells. Extracellular HMGB1 mediates cell-to-cell signalling, provokes the production of inflammatory cytokines and subsequently activated cytokines can induce further release of HMGB1 into the extracellular space.20


The micro- and macrohemodynamic changes and many other signs, which are identical to those observed or usually detected in human septic patients, were present in this porcine model of intraabdominal sepsis. This has lead us to conclude that this large animal model characterises the circulatory failure of human sepsis correctly, and may be used for further research of the disease and to test novel therapeutic opportunities.

1. Kazarian KK, Perdue PW, Lynch W, et al. Porcine peritoneal sepsis: modeling for clinical relevance. Shock 1994;1(3):201-12.
2. Dejager L, Pinheiro I, Dejonckheere E, et al. Cecal ligation and puncture: the gold standard model for polymicrobial sepsis? Trends Microbiol 2011;19(4):198-208.
3. Sadowitz B, Roy S, Gatto LA, et al. Lung injury induced by sepsis: lessons learned from large animal models and future directions for treatment. Expert Rev Anti Infect Ther 2011;9(12):1169-78.
4. Creteur J, De Backer D, Sakr Y, et al: Sublingual capnometry tracks microcirculatory changes in septic patients. Intensive Care Med 2006;32:516-23.
5. Boda D, Kaszaki J, Tálosi G: A new simple tool for tonometric determination of the pCO2 in the gastrointestinal tract. In vitro and in vivo validation studies. Eur J Anaesthesiol 2006;23:680-5.
6. Groner W, Winkelman JW, Harris AG, et al. Orthogonal polarization spectral imaging: A new method for study of the microcirculation. Nature Med 1999;5(10):1209-13.
7. Moshage H, Kok B, Huizenga JR, et al: Nitrite and nitrate determinations in plasma: a critical evaluation. Clin Chem 1995;41:892-6.
8. Wolfárd A, Kaszaki J, Szabó C, et al. Prevention of early myocardial depression in hyperdynamic endotoxemia in dogs. Shock 2000;13:46-51.
9. Matejovic M, Krouzecky A, Rokyta R Jr, et al. Effects of combining inducible nitric oxide synthase inhibitor and radical scavenger during porcine bacteremia. Shock 2007;27:61-8.
10. Garrido AG, Figueiredo, LFP, Silva MR. Experimental models of sepsis and septic shock: an overview. Acta Cir Bras 2004;19(2):82-8.
11. Buras JA, Holzmann B, Sitkovsky M. Animal Models of sepsis: setting the stage. Nature Reviews Drug Discovery 2005;4:854-65.
12. Bauer M, Reinhart K. From mice and MOF: Rodent models, immune modulation, and outcome in the critically ill. Crit Care Med 2006;34:921-3.
13. Godin DV, Garnett ME. Species-related variations in tissue antioxidant status I-II: Differences in antioxidant enzyme profiles. Comp Biochem Physiol B 1992;103:737-48.
14. Barth E, Bassi G, Maybauer DM, et al. Effects of ventilation with 100% oxygen during early hyperdynamic porcine fecal peritonitis. Crit Care Med 2008;36:495-503.
15. Ince C. The microcirculation is the motor of sepsis. Critical Care 2005;9(S4):S13-9.
16. Sakr Y, Dubois MJ, De Backer D, et al. Persistent microcirculatory alterations are associated with organ failure and death in patients with septic shock. Crit Care Med 2004;32(9):1825-31.
17. Parratt JR. Nitric oxide in sepsis and endotoxaemia. J Antimicrob Chemother 1998;41(Suppl A):31-9.
18. Wang H, Yang H, Czura CJ, et al: HMGB1 as a late mediator of lethal systemic inflammation. Am J Respir Crit Care Med 2001;164:1768-73.
19. Yang H, Ochani M, Li J, et al: Reversing established sepsis with antagonists of endogenous high-mobility group box 1. Proc Natl Acad Sci USA 2004;101:296-301.
20. Scaffidi P, Misteli T, Bianchi ME. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature 2002;418:191-95.

"Victor Babes" Publishing House "Victor Babes" University of Medicine and Pharmacy Romanian Academy of Medical Sciences National Council of Scientific Research in Higher Education (B+) Index Copernicus
Journals Master List
Embase Journal List Timisoara Medical Scientific Association
ISSN: 1583-526X (Online Edition)
Copyright © 2002-2017 Timisoara Medical Journal. All rights reserved.