Study of the references listed leaves no doubt that essential specific and unspecific immune functions are more or less strongly influenced by a wide variety of anaesthetics. In clinically employed concentrations these negative effects are rapidly and completely reversible. No clinically relevant adverse actions on the immune system have ever been identified in short-term anaesthesia with any substance, not even with the thiobarbiturates although these are known to be highly inhibitory. However, matters are entirely different in respect of highly dosed long-term sedation. In such cases, it may occur with e.g. thiobarbiturates or diazepam that in the course of days or weeks concentrations are attained in the tissue which may lead us to expect in vivo quite considerable suppression of both unspecific and antigen-specific cellular immune mechanisms. This expectation, which is first of all based on in-vitro results, has been indirectly supported by a clinical study in patients suffering from craniocerebral trauma who received artificial respiration and in whom an association was seen between thiopental administration and the incidence of bacterial pneumonias. Over and above this, however, the preoperative condition of the patient and of course mainly the operation as such with its associated stress reaction will cause direct and indirect changes in the immune system (Fig. 4). Trauma and anaesthesia exert not only a direct action, but act as inhibitors or stimulators of important immune functions chiefly by modulating the stress response. We are still unaware as to whether and to what extent the prognosis of the individual patient is actually adversely affected by the combined action of these factors. Many findings, however, seem to point to an essential role played by both the surgical trauma and the stress response it induces, as well as by the pre-operative immune status of the patient, in the occurrence of subsequent infectious complications (Fig. 4).