Formation of atelectasis, defined as reversible collapse of aerated lung, often occurs after induction of anaesthesia with mechanical ventilation. As a consequence, there is a risk for hypoxemia, altered hemodynamics and impaired respiratory system mechanics. In certain situations, the risk for atelectasis formation is increased and its consequences may also be more difficult to manage. Anesthesia for bariatric surgery in morbidly obese patients and surgery requiring one-lung ventilation (OLV) with capnothorax are examples of such situations. In Paper I (30 patients with BMI > 40 kg/m2 scheduled for bariatric surgery) a recruit­ment maneuver followed by positive end-expiratory pressure (PEEP) re­duced the amount of atelectasis and improved oxygenation for a prolonged period of time. PEEP or a recruitment maneuver alone did not reduce the amount of atelectasis. In paper II we investigated whether it is possible to predict respiratory function impairment in morbidly obese patients without pulmonary disease from a preoperative lung function test. Patients with mild signs of airway obstruction (reduced end-expiratory flow) in the preoperative spirometry developed less atelectasis during anaesthesia. In paper III we developed an experimental model of sequential OLV with capnothorax using electrical impedance tomography (EIT) that in real-time detected lung separation and dynamic changes in pulmonary ventilation and perfusion distributions. OLV to the left side caused a decrease in cardiac output, arterial oxygenation and mixed venous saturation. In paper IV we used our model of OLV with capnothorax and applied a CO2-insufflation pressure of 16 cm H2O. We demonstrated that a PEEP level of 12-16 cm H2O is needed for optimal oxygenation and lowest possible driving pressure without compromising hemodynamic variables. Thus, the optimal PEEP was closely related to the level of the capnothorax insufflation pressure. With insufficient PEEP, ventilation/perfusion mismatch in the ventilated lung and redistribution of blood flow to the non-ventilated lung occurred.

One-lung ventilation (OLV) as an established procedure during thoracic surgery may be injurious in terms of increased mechanical stress characterised by alveolar cell stretch and overdistension, increased cyclic tidal recruitment of alveolar units, compression of alveolar vessels and increased pulmonary vascular resistance. This may result in ventilation-induced lung injury with pro-inflammatory cytokine production, leukocyte recruitment and neutrophil-dependent tissue destruction. Despite the consequences of delivering the whole tidal volume (VT) to only a single lung, relatively high VT are used during OLV to maintain arterial oxygenation and carbon dioxide elimination. However, this may increase mechanical stress in the dependent lung and may aggravate alveolar injury. There is a lack of data on the alveolar immune consequences of OLV. Therefore, the present studies investigate the epithelial damage and pro-inflammatory response induced by mechanical ventilation and OLV. OLV induced pulmonary injury, but alveolar damage in the ventilated lung decreased by reduction of the tidal volume in patients scheduled for thoracic surgery (study I). The use of the volatile anaesthetic desflurane in OLV patients attenuated the OLV-induced alveolar immune response (study II). Furthermore, an experimental model of thoracic surgery was established to investigate the systemic and pulmonary consequences of OLV and thoracic surgery in comparison with the effects of conventional two-lung ventilation and spontaneous breathing. The experimental data indicate that beside the pulmonary immune response volatile anaesthetics have also modulated the plasma concentrations of cytokines during and after OLV (study III). In contrast, OLV and thoracic surgery increased the expression of pro-inflammatory mRNA in BAL cells and lung tissue samples. General anaesthesia did not affect this response (study 4). The results of the present studies indicate that OLV and thoracic surgery may be injurious to the lung tissue to a similar degree. The recruitment and activation of alveolar granulocytes characterise the alveolar damage. The administration of different anaesthetics modulates the activation of alveolar cells, specified by decreased inflammatory mediator release in subjects that receive desflurane anaesthesia, which does not affect the expression of cytokine mRNA in alveolar cells and lung tissue samples.

Thoracic surgical procedures require partial or complete airway separation and the opportunity to exclude one lung from ventilation (one-lung ventilation, OLV). OLV is commonly associated with profound pathophysiological changes that may affect the postoperative outcome. It is injurious in terms of increased mechanical stress including alveolar cell stretch and overdistension, shear forces secondary to repeated tidal collapse and reopening of alveolar units and compression of alveolar vessels. Ventilation and perfusion distribution may thus be affected during and after OLV. The present studies investigated the influence of OLV on ventilation and perfusion distribution, on the gas/tissue distribution and on the lung histomorphology in a pig model of thoracic surgery. Anaesthetised and mechanically ventilated piglets were examined. The ventilation and perfusion distribution within the lungs was assessed by single photon emission computed tomography. Computed tomography was used to establish the effects of OLV on dependent lung gas/tissue distribution. The pulmonary histopathology of pigs undergoing OLV and thoracic surgery was compared with that of two-lung ventilation (TLV) and spontaneous breathing. OLV induced hyperperfusion and significant V/Q mismatch in the ventilated lung persistent in the postoperative course. It increased cyclic tidal recruitment that was associated with a persistent increase of gas content in the ventilated lung. OLV and thoracic surgery as well resulted in alveolar damage.  In the present model of OLV and thoracic surgery, alveolar recruitment manoeuvre (ARM) and protective ventilation approach using low tidal volume preserved the ventilated lung density distribution and did not aggravate cyclic recruitment of alveoli in the ventilated lung. In conclusion, the present model established significant alveolar damage in response to OLV and thoracic surgery. Lung injury could be related to the profound pathophysiological consequences of OLV including hyperperfusion, ventilation/perfusion mismatch and increased tidal recruitment of lung tissue in the dependent, ventilated lung.  These mechanisms may contribute to the increased susceptibility for respiratory complications in patients undergoing thoracic surgery. A protective approach including sufficient ARM, application of PEEP, and the use of lower tidal volumes may prevent the ventilated lung from deleterious consequences of OLV.