Pii: S0168-8278(01)00224-0

The interruption of tissue oxygenation – ischemia – is probably the most important problem associated with liver surgery [1,2]. This is especially true for patients with impaired liver function, such as liver cirrhosis or steatosis, where the effects of ischemic injury may be irreversible even if the duration of the interruption of blood flow is short. Paradoxically, the generation of reactive oxygen species (ROS) seems to be a key factor in the development of liver damage in irreversible ischemia. The answer to this apparent paradox lies in that ROS are generated not during the initial period of ischemia but upon restoration of blood flow or reperfusion. During ischemia, a variety of cellular and molecular events take place in the liver that, upon reoxygenation, result in the massive production of ROS, which inflict direct tissue damage and initiate a cascade of deleterious cellular responses leading to inflammation, cell death, and organ failure. The term ischemia-reperfusion (IR) injury has been coined to indicate this condition. Therapeutic strategies aimed to prevent IR injury are focused on controlling ROS generation at the time of restoration of blood flow or reperfusion, and intervening in the signal cascade initiated by reoxygenation. Liver injury induced by hepatic IR is characterized by activation of the transcription factor NF-kB, activation of Kupffer cells, increased production of proinflammatory cytokines, such as tumour necrosis factor-a (TNF-a) and interleukin-1 (IL-1), liver polymorphonuclear (PMN) leukocyte activation and infiltration, and hepatocellular damage [2]. TNF-a is an important mediator in PMN leukocyte recruitment and activation which in turn are a potent source of ROS. In view of this critical role played by TNF-a in liver injury, it has been long proposed that suppression of TNF-a production may attenuate liver injury after IR. Thus, anti-TNF-a antibodies [3] and suppression of cytokine generation with FR167653 [4] decrease hepatic IR injury in experimental models. Recent studies indicate that Kupffer cells play a key role in the generation of TNF-a during IR liver injury [5]. For instance, the inhibition of TNF-a release from Kupffer cells with gadolinium chloride (GdCl3), previous to the IR treatment, maintained low levels of TNF-a attenuating liver injury, an effect that could be reverted by the administration of exogenous TNF-a [6,7]. In addition to GdCl3, and other inhibitors of Kupffer cell activation, a variety of different pharmacological approaches, such as the administration of adenosine [8], nitric oxide donors [9], and antioxidants [10,11], have been successfully applied in animal models to attenuate liver injury after IR. Identifying the mechanisms of action by which these and other agents provide ischemic tolerance will provide potential targets for the development of new treatments for IR liver injury. It is known for many years that cellular injury, independently of its nature (infectious, oxygen deprivation, inflammatory, heat, etc.), induces an adaptive response. Accordingly, another therapeutic strategy aimed to ameliorate hepatic IR injury consists in the exposure of the liver to brief periods of blood flow interruption followed by reperfusion prior to the prolonged IR treatment, a procedure that is know as ischemic preconditioning. The protective effects of ischemic preconditioning were first described in the myocardium and have been also shown experimentally in the liver. It has been recently demonstrated that ischemic preconditioning protects sinusoidal endothelial cells and suppresses Kupffer cell activation after storage and reperfusion [12]. Heat shock proteins (HSPs) are well known as cytoprotective proteins. Accordingly, different groups have investigated whether a period of hyperthermia followed by recovery from this heat treatment is another mechanism for acquiring ischemic tolerance. Heat shock pretreatment has been shown to prevent hepatic ischemic injury caused by Pringle’s manoeuvre [13] and in an isolated rat liver perfusion model [14]. Heat shock preconditioning has been also shown to reduce liver injury induced by carbon tetrachloride administration [15] and to improve liver transJournal of Hepatology 35 (2001) 663–665