Ischemia-reperfusion injury (IRI) is tissue damage caused by returning blood to organs temporarily deprived of circulation. IRI arises in coronary and cerebral infarctions, trauma, surgery and transplantation. Global IRI have been studied by clamping the aorta in various rat, rabbit and dog models. The translational value to humans are limited from small animal models and could advantageously be studied in large animals, where pig is a highly relevant animal model.

Several porcine models of thoracic aortic clamping have been described. However, these models have focused on the effect of single organs and not taken into account the global IRI from all organs affected by the clamping. Studies are described for the effects on cardiac (1, 2), renal (3, 4), hepatic (5), limb (6), spinal cord (7), and pulmonary IRI (8).

We aimed to establish a new porcine model for thoracic aortic cross-clamping using light weight piglets to allow for a minimal use of costly experimental therapeutics, while preserving translational value of the model. Such a model is highly relevant since thoracic aortic aneurysm surgery often require clamping of aorta. We have shown that the inflammatory reaction in such patients is largely mediated through the lectin complement pathway (9).

In IRI, reactive oxygen species and innate inflammation exacerbate ischemic injury through e.g. increased complement activation (10). Complement 1 inhibitor (C1-INH) controls the physiological activity of both the classical and lectin pathway of complement, and several studies show that C1-INH reduces IRI in animal models and in patients (10). In addition, C1-INH is by far the most important suppressor of bradykinin production from the kallikrein-kinin system, and would inhibit bradykinin-mediated inflammation in pigs (11), capillary leakage and pain (12).

The aim of the present study was to establish a novel and clinically relevant IRI pig model of aortic clamping. Secondly, to investigate the effect of IRI on cardiovascular, hematologic, inflammatory and organ damage markers during ischemia and for a five hour observation period, and thirdly, how these effects were modified by supraphysiological amounts of human C1-INH administered before clamping.

A broad panel of markers changed considerably from before cross-clamp onset and showed a uniform pattern in both C1-INH-treated and untreated animals. The variation in time course of each marker is shown. The only marker that was significantly reduced by C1-INH treatment was TCC. To show the magnitude of the total response, we present the fold change (FC) of the markers in treated and untreated animals combined, excluding sham animals. The change is from 50 minutes before the start of IRI (approximately 5 minutes before cross-clamp onset) until maximal change of marker value during the following observation period. FC for each marker is mean of maximal values after clamping in all animals divided by mean value before clamping. Each marker´s name in the Figure panels is succeeded by the FC in parenthesis when meaningful.

The developed novel porcine model of ischemia-reperfusion injury after thoracic aorta cross-clamping produced robust results and proved to be a forceful and reliable model for the assessment of large scale IRI. The animals receiving thoracic aorta cross-clamping showed signs of severe large-scale ischemia with substantial lactic acidosis and, typical for porcine models, pulmonary hypertension. Keeping animals stable was challenging due to the severity of ischemia, but it was manageable. Despite severe hemodynamic changes, the animals in general responded in a uniform way. The effects of the IRI on blood and tissue markers also showed uniformity, supporting the value of the model and its potential for future testing of drugs aimed to restrict systemic IRI.

We did not see any significant effects of treatment with C1-INH on the damage from IRI in our model. Other animal studies have previously shown effect of C1-INH on IRI (19–22). These studies may not be fully comparable as few of them are in pigs, and most importantly do not involve a larger part of the body in a near systemic way. Our model is global and one of the most forceful models described, where all sub-diaphragmatic organs are virtually devoid of blood supply for 45 minutes. Therefore, the IRI in our model could be so severe that some inhibitors, including C1-INH, are inefficient.

Furthermore, blood in intercostal arteries and their branches could bypass the aortic clamping in some species. Also the aortic trifurcation in pigs could result in variable aortic retrograde flow compared to humans (23).Thus, interspecific differences could explain why human C1-INH did not improve outcome in our study. Finally, minor differences in the glycosylation of the C1-INH-protein could be of importance. An IRI study in mice found plasma-derived C1-INH, as we have used in the present study, inferior to recombinant human C1-INH (24). On the other hand, half-life of a recombinant human C1-Inhibitor produced in rabbits with a distinctive glycosylation (Ruconest, Pharming) had a shorter half-life in pigs than humans (14). However, the half-life of plasma-derived C1-inhibitor as we used in our study, administered both i.v. and subcutaneously to pigs had half-lives of more than 24 hours (25). Therefore C1-inhibitor levels are expected to remain high during our study period.

We have experienced in several in vitro studies that pharmacological inhibition of classical complement activation using C1-INH require extensively high doses, up to 20 times the physiological concentration (13). To the extent that the classical complement pathway plays an important role in IRI, this is in line with Ziccardi who 40 years ago found C1-INH to poorly inhibit when immunoglobulins were involved (26). Therefore, the dose of C1-INH per kg could have been insufficient to inhibit complement activation in severe IRI. Although some studies show effect of a lower (27) and some with higher doses up to 500 U/kg C1-INH (28), the reperfused organs were smaller, varying from skin flaps to kidneys and part of the heart. This makes a comparison of the studies difficult. In some models, researchers infuse C1-INH intra-arterially, directly into the targeted organs, increasing the dose pr kg tissue considerably. But even then, C1-INH can be inefficient. In a porcine infarction model Schreiber et al. found no effect of intracoronary infusion of 500 units C1-INH (29). However, the model by Schreiber et al. differed importantly from earlier models in that C1-INH was given after ischemia and during reperfusion. In our study C1-INH was infused prior to ischemia. Taken together, lack of effect in our global and forceful model due to too low C1-INH concentration cannot be totally excluded.

Supraphysiological doses of C1-INH have previously been reported to induce a procoagulant state which could have masked a positive effect by C1-INH (30). However, an in vitro study by our group found the opposite showing a dose-dependent anticoagulant effect of C1-INH in human whole blood (31). Our in vivo data from the present study also suggests a tendency towards an anticoagulatory effect, showing a non-significant tendency towards a reduction in TAT by C1-INH treatment. This could be explained by inactivation of FXII resulting in inhibition of the contact system by C1-INH, which could play a more significant role in IRI in this model. Cacci et al. suggested that thrombin inhibition by C1-INH might play a more important role in cases of severe inflammation as other major anticoagulatory systems are reduced in this state (32). Taken together, procoagulant effects of the C1-INH-infusion in the present study is an unlikely explanation for the observed lack of effect.

Publication biases could lead to overestimates of beneficial effects of C1-INH in IRI and underestimate negative results. According to Fanelli the proportion of positive results in the scientific literature in general increased to 85.9% in 2007 with a yearly increase of 6%, constant across most of the disciplines and countries (33). Therefore, our finding of limited effect of C1-INH in global ischemia is in our opinion an important contribution to the current literature. C1-INH is often placed on the list of complement inhibitors. Our study supports a general notion that C1-INH is not an efficient therapeutic, except for the reconstitution of patients with C1-INH deficiency. In our opinion, C1-INH should not be placed on the list of specific inhibitors for complement inhibition in the future, as we also have reviewed previously (34).

The present study raises the question which therapeutic targets could potentially reduce IRI.in vivo According to the danger model introduced by Matzinger, an inflammatory response involving innate immunity and complement can be expected in IRI (35). Therefore we anticipate a possible beneficial effect of combined inhibition of CD14 and complement C5 in IRI. In pediatric kidney transplant recipients, C5 inhibition with eculizumab prior to graft reperfusion was shown to improve early graft function, graft morphology and early graft survival indicating a reduction in IRI (36). There are also several other therapies available including highly specific complement inhibitors, as well as drugs targeting key upstream innate immune recognition molecules (34).

Our aim to establish a new porcine model for thoracic aortic cross-clamping using light weight piglets to allow for a minimal use of costly experimental therapeutics, while preserving translational value of the model is also a limitation of the study. The model is focused on the acute response and do not cover late complications.

This study describes a novel, forceful model of porcine IRI that did not respond significantly to adequate C1-INH treatment but should allow for future testing of other drugs that could attenuate global IRI.

The original contributions presented in the study are included in the article/supplementary files. Further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by Norwegian National Animal Research Authority (FOTS-ID-8197).

EN, YM, and TM designed the study. O-LB, JG, AD, HF, JL, KP, LR, RS and HJ planned and/or performed the experiments and analyses. EN and TM drafted the paper. EN and TM edited the draft. All authors critically revised the paper and approved the final draft.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.