Controlled reperfusion reduces hemorheological alterations in a porcine infrarenal aortic-clamping ischemia-reperfusion model

1 Introduction

Prolonged acute ischemia leads to cellular death and irreversible damage of any affected organ. Restoration of blood flow to prevent ongoing tissue damage is a cardinal point of treatment. On the other hand, restoration of blood flow and oxygen supply itself may lead to further damage, a phenomenon known as ischemia-reperfusion injury [3, 9].

Acute lower-limb ischemia is the most common reason for emergency admission to a vascular surgery unit. Re-establishment of arterial blood flow to the compromised leg is essential for limb salvage. Despite improvement in revascularization techniques, the results of surgical and interventional treatments have remained unsatisfactory, with high amputation rates and high mortality [6, 29]. The poor results of revascularization therapy alone may be mainly due to the additional reperfusion injury.

During the past decades an increasing amount of evidence was uncovered about molecular and pathophysiological mechanism of reperfusion injury [14]. Though detailed description is beyond the scope of this paper, we must highlight some of the many processes.

Extreme level of ATP depletion and breakdown during ischemia causes hypoxanthine accumulation. During reperfusion hypoxanthine is metabolized into xanthine by xanthine-oxidase that is accompanied by the production of a large amount of reactive oxygen species (ROS). ROS cause lipid peroxidation and membrane injury, damage cytoskeletal and enzymatic proteins, and cause DNA breaks, promoting PARP activation and further energy depletion. ROS also play a role in activation of local and systematic inflammatory response. The sequel increase of vascular permeability, endothelial dysfunction and interstitial edema impairs microcirculation [10].

Impairment of microcirculation may lead to an inability of successful reperfusion, even after the restoration of blood flow, called no-reflow phenomenon. Hemorheological properties of blood and blood cells, like viscosity, erythrocyte deformation and aggregation play an important role among such conditions. Oxidative stress was demonstrated to worsen these parameters by many authors during the pathogenesis of various conditions [1, 20].

Based on pathophysiological rationale and experimental results several strategies have been proposed to reduce reperfusion damage [5, 35]. Nevertheless, their usage in clinical practice is still very limited, the majority of surgeons do not use any of these techniques at all. Evidence supporting benefit of such strategies are essential to advance them into clinical practice. Controlled reperfusion aims to limit reperfusion damage through gradual restoration of blood flow, applying reduced perfusion pressure and using special solutions instead of pure blood to counter unfavorable biochemical changes. Several methods using various composition of reperfusion solution, perfusion pressure and duration were published [5, 7].

We aimed to investigate possible benefits of a simplified reperfusion system [26] in a porcine model of acute limb ischemia. In this paper we assessed the impairment of hemorheological parameters as a consequence of reperfusion damage.

2 Methods

3 Results

Hematocrit showed an increasing tendency in FR group during reperfusion, while a significant decrease was found in case of CR. These changes became attenuated until the 1 week sampling (Fig. 1a).

Plasma viscosity decreased significantly during controlled reperfusion but did not change during full reperfusion. On the other hand, it was found normal at 1 week in CR but elevated in FR (Fig. 1b).

Whole blood viscosity elevated significantly during full reperfusion, but decreased significantly during controlled reperfusion. These changes became attenuated at 1 day, but at 1 week whole blood viscosity in the FR group was higher than in CR (Fig. 1c).

We could not demonstrate a significant change in erythrocyte deformation at any shear stresses in any samples compared to baseline or between CR and FR.

Erythrocyte aggregation index (AI) showed an increasing tendency during full reperfusion, while it was significantly and strongly reduced at R15 and slightly reduced at R60 of controlled reperfusion. At 1 week increased aggregation was visible in FR compared to baseline and to CR (Fig. 2a). Threshold shear rate was found decreased at R15 of controlled reperfusion. At 1 day it was found suspiciously highly elevated in both groups. At 1 week it was similarly and more reasonably elevated in both groups (Fig. 2b).

4 Discussion

Experimental studies on isolated rat hindlimbs have shown that cellular integrity and biochemical function is preserved after 4 hours of warm ischemia [4]. The severe changes occur after the onset of uncontrolled reperfusion. Thus tissues are still savable after the duration of the ischemia in this model, while reperfusion damage should be reduced. Four hours is also a reasonable time window in which an intervention can be done to a patient with acute lower-limb ischemia. In an animal model of skeletal muscle ischemia, a reduction of reperfusion blood flow was shown to reduce edema generation and muscle injury [34]. Modification of the initial reperfusate can also reduce the local consequences of reperfusion injury.

During ischemia aerobic metabolism is suspended and anaerobic metabolism is activated. This leads to a breakdown of high-energy phosphates, an increase in intracellular lactate, and intracellular acidosis. The Krebs cycle loses intermediates. With the start of reperfusion, the washout of lactate and protons from the ischemic tissue leads to an increase in intracellular calcium. Oxidation of accumulated hypoxantin to xantin results in the generation of reactive oxygen intermediates [14].

The crystalloid reperfusion solution is aimed to counter these changes. Glucose is added to provide a substrate for anaerobic metabolism and as a hyperosmolar substance to reduce edema generation. With glutamate and aspartate, amino acid precursors of the Krebs-cycle are added to ensure more effective oxidative metabolism with the onset of reperfusion. Allopurinol is added to inhibit xanthine-oxidase and the generation of ROS. Sodium citrate is added to reduce intracellular calcium [32].

The procedure of reperfusion was modified from the method of Wilhelm and Beyersdorf [32]. We used a 1 : 1 ratio mixture of blood and crystalloid reperfusion solution instead of 6 : 1 proposed by the authors, based on the following assumption: The more diluted reperfusion suspension had a hematocrit of 15–20%, which according to clinical experience is still tolerable for skeletal muscle if the perfusion rate is fair. On the other hand the reduced hemoglobin, and thus reduced oxygen content may result in less oxidative damage in this first phase of reperfusion, while the higher supply of protective metabolites may also help to limit reperfusion damage. Diluted blood may cause less clogging in the damaged capillaries resulting in better microcirculation.

As there is evidence that most additional tissue injury caused by uncontrolled reperfusion occurs during the first 20 to 30 minutes [8], a 30-minute interval was chosen for controlled reperfusion.

Many previous reperfusion protocols used heart-lung machine or roller pumps. In this set-up these expensive instruments are substituted with a simple, easily accessible and safe blood bag reperfusion system [26].

No significant change of the measured parameters was found during ischemia. After clamping of the aorta lower limbs receive minimal perfusion through collaterals. Venous drain of the region is minimal as well, and may not cause major changes in samples.

Opposite trends could be seen for hematocrit, plasma and blood viscosity during full and controlled reperfusion, changes being more pronounced at 15 minutes. Reperfusion injury induces cellular swelling and interstitial edema that draw water out of the intravascular compartment, and may lead to hemoconcentration, increasing hematocrit [10, 19]. Controlled reperfusion on the other hand involves infusion of considerable amount of low viscosity crystalloid solution. This hemodilution outweighs aforementioned changes, and results in reduction of plasma viscosity, hematocrit and hence whole blood viscosity.

As R15 samples were taken from the inferior caval vein, blood from the reperfused region is overrepresented compared to true mixed venous blood. In the CR group, effect of hemodilution at R15 is thus more pronounced than at R60 when the limbs are already perfused with own blood.

Higher plasma and whole blood viscosity could be seen 1 week after FR. Reperfusion washing out debris from the ischemic tissue results in poisoning of the body, systemic inflammatory response syndrome (SIRS) and in most severe cases multiorgan failure and death [21]. Second-phase changes in parameters may refer to this systemic inflammatory response. It is hypothesized that full reperfusion is associated with more severe reperfusion injury, SIRS and production of acute phase proteins, including fibrinogen [27]. Fibrinogen, a major determinant of plasma and blood viscosity may explain the changes [25], and the higher erythrocyte aggregation at 1 week as well.

Several studies described the detrimental effect of oxidative stress on erythrocyte deformability both in experimental models [1, 24], and hypothesized its role in the pathophysiology of ischemia-reperfusion injury [15, 28]. In this study, however, no significant change was found in elongation indices measured with ektacytometry. A possible source of error may be that the osmolarity of the standard PVP solution we used is a bit lower than the optimal osmolarity found for pigs with osmotic gradient ektacytometry [22].

Aggregation index represents the extent of aggregate formation. Threshold shear rate means the lowest shear rate needed to achieve complete disaggregation, and it is usually well correlated to AI. The fall in erythrocyte aggregation during controlled reperfusion is explained by hemodilution, as both hematocrit and local fibrinogen level is reduced. However, at 1 day, AI and threshold shear rate results were inconsistent, and extraordinarily high threshold shear rates were found. Threshold shear rate is determined using an iteration process, finding the maximal laser light backscattering [12]. At lower shear rates intensity decreases due to erythrocyte aggregation. At higher shear rates no aggregation is present, but erythrocytes elongate and cause smaller backscattering. Threshold shear rate is an identifiable extreme of this shear rate-intensity curve. In this case, however, several samples displayed a flattened high shear arm with ill-defined extreme, which resulted in high mean value and large standard deviation. The biological meaning of the phenomenon is unclear. Decreased elongation is not suspected after finding no change in deformability. Nevertheless, we do not suppose an extremely strong aggregation, and suggest treating this result as a curiosity.

It was also reported that in pigs infrarenal aortic-clamping does not develop as serious lower limb ischemia as expected due to a highly developed collateral circulation through the epigastric artery [11]. This may partially be reasonable for the minor changes in some hemorheological parameters. Nevertheless, this model still resulted in ischemia-reperfusion injury according to change in oxidative stress markers and histology, which was attenuated by controlled reperfusion [13].

Our results were sometimes only partially consistent with the ones of other authors. However, change in hemorheological parameters are often different, depending on the species and the ischemia-reperfusion method used. Changes were found different in a supra- and infrarenal aortic clamping model in rats, and the hemodynamic and acid base changes were even larger than the microrheological ones [23]. Other studies reflected more expressed microrheological alterations mostly during the first 1–5 postoperative days together with acute phase reactions [2, 17].

5 Conclusion

Our results demonstrate the impairment of hemorheological parameters after full reperfusion. Controlled reperfusion with a simple blood bag reperfusion system could reduce the changes in these markers, and may be beneficial to limit ischemia-reperfusion damage, though studies with clinical endpoints are needed to validate this hypothesis.