Pentoxifylline and tumour necrosis factor-induced

Acute Gram-negative bacterial infections evoke characteristic pathophysiological responses, including changes in white blood cell counts, fever, haemodynamic disorders, and various coagulatory disturbances, which may result in respiratory distress syndrome, multiorgan failure and irreversible shock. It is generally accepted that endotoxins, the lipopolysaccharides (LPS) of Gramnegative bacteria, are the agents causing these pathophysiological events. Endotoxins stimulate macrophages to elaborate various biologically active mediators, which can induce phenomena typical of endotoxaemia and bacterial sepsis. Tumour necrosis factor-alpha (TNF) has been identified as an important mediator of lethal shock and cachexia in animals [1]. Passive immunization against TNF attenuates the lethal effect of endotoxin in mice and protects against septic shock during Gram-negative bacteraemic in primates [2]. In addition, TNF increases the adherence of polymorphonuclear leucocytes to endothelial cells. Increased adherence of activated granulocytes in the microvasculature in vivo is one of the major causes of pulmonary vascular injury in adult respiratory distress syndrome (ARDS), which is one of the most severe consequences of Gram-negative sepsis in man [3]. Intrapulmonary activation of leucocytes and release of cellular mediators and enzymes in ARDS is reflected by high bronchoalveolar levels of TNF and its soluble inhibitors (sTNF-Rc I + II) as well as high levels of granulocyte elastase in the bronchoalveolar fluid in the early stages of severe ARDS [4]. TNF plasma levels measured in parallel in patients developing ARDS after trauma, sepsis or shock were found to be in the normal range, indicating a local release of TNF, possibly by pulmonary macrophages or other cells [4]. This is consistent with the concept that stimulated lung macrophages can produce much more TNF than peripheral blood monocytes [5]. TNF, possibly together with interleukin-1β, may favour the adherence and accumulation of granulocytes in lung capillaries and contribute to transendothelial passage of these cells, protein and fluid. TNF may stimulate granulocyte degranulation, and, thereby, release of elastase, damaging lung capillaries, interstitium and alveolar structures. Furthermore, TNF is one of the most potent inducers of oxygen radicals in neutrophils. A major role of activated neutrophils and neutrophilderived oxidant production has been demonstrated in animal models of lung injury, including the isolated perfused lung [6]. Lung ischaemia/reperfusion in rabbits results in transient generation of TNF, which is probably responsible for the lung injury following reperfusion [7]. Recently, a direct mediation of TNF in the decrease of surfactant production by human type II pneumocytes has been demonstrated [8]. Although a great number of circulating factors, such as other cytokines, prostaglandins and free radicals, have been implicated in the aetiology of ARDS, it is generally accepted that TNF is at least one of the major pathophysiological factors. Injection of TNF alone into experimental animals can induce all typical signs of endotoxicity, including the development of acute lung injury. Furthermore, immunotherapeutic interventions with either the murine monoclonal antibody OKT3 in transplant recipients [9], or recombinant interleukin-2 in cancer patients, results in an acute cytokine release syndrome, including severe side-effects, such as capillary leakage syndrome and ARDS, which are related to the endogenous TNF formation. These data strongly suggest that the overwhelming production of TNF plays a pivotal role in ARDS. Therefore, drugs interfering with the production of TNF or the adherence and activation of granulocytes may have beneficial effects in septic shock and ARDS. Pentoxifylline (POF; oxpentifylline, 3,7-dimethyl-1-(5oxohexyl)-xanthine) and related xanthines are drugs of known haemorrheological activity, and are used clinically for the treatment of patients with various types of vascular insufficiency. Its effects were supposed to be based on its ability to increase erythrocyte flexibility, to reduce blood viscosity and filterability, and to increase capillary flow. Recently, new pharmacological aspects of these established drugs could be demonstrated. Evidence was produced that POF is able to inhibit the synthesis of TNF in vitro and in vivo. In studies addressing the mechanism by which POF exerts its TNF inhibitory activity, it was found that this pharmaceutical drug reduced, in macrophage cultures in vitro the formation of TNFspecific messenger ribonucleic acid (mRNA) [10], and in vivo (in mice) the production of endotoxin-induced circulating TNF [11]. It was also found that POF selectively inhibits the formation of TNF, but not interleukin-6 (IL6), probably by inhibiting phosphodiesterase activity and, thus, causing accumulation of intracellular cyclic adenosine monophosphate (cAMP) [12]. In addition to blocking TNF production, POF also interferes with TNF EDITORIAL