icu

Force, Strain, and Stress of Ventilator-Induced Lung Injury

　　Ventilator-induced lung injury (VILI) is usually addressed in terms of force, strain, and stress. It is helpful to define these entities: force is equal to pressure times area, strain is the relative change in shape or size of an object due to externally applied forces, and stress is the internal force (per unit area) associated with the strain. In determining the forces that act on the lung and that might potentially cause injury, it is important to understand what the forces are that are actually distending the lung. This is often addressed by measuring the pressures at the airway opening. During controlled mechanical ventilation, the plateau pressure (Pplat: the pressure measured at end-inspiration when there is no flow) represents the pressure required to inflate the entire respiratory system (lung plus chest wall); but in terms of stresses on the lung, the important pressure is the distending pressure of the lung -- the transpulmonary pressure (Ptp). This issue becomes very important in patients who have a relatively stiff chest wall (eg, massive ascites). From a respiratory mechanics point of view, the chest wall is largely made up of the rib cage and the abdomen. For example, if the Pplat is 30 cm H2O and the patient has a stiff chest wall with an elastance (inverse of compliance) equal to 15 cm H2O, then the Ptp is only 15 cm H2O. However, if the chest wall has a relatively low elastance (eg, 5 cm H2O), then the Ptp is 25 cm H2O, and the lung is possibly being overdistended. Consideration of chest wall mechanics is thus important in assessing the factors leading to a high airway pressure; and most importantly, this has an impact on how one ventilates patients with ARDS. In a patient with a compliant chest wall, it may be reasonable to ventilate the patient with a Pplat of 30 cm H2O; but in a patient with a very stiff chest wall, maintaining the Pplat at less than 30 cmH2O could lead to underventilation.
In assessing the propensity to lung injury, it is helpful to understand the supporting structures of the lung as well as the stresses and strains put on the lung during ventilation. In a homogeneously expanded lung, the stress distribution is also relatively homogeneous. However, in areas of the lung surrounding regions of the lung that are collapsed, the stresses can be extremely high, as pointed out by Mead and colleagues[5] more than 30 years ago. The targets of injury in the lung are the fiber system of the extracellular matrix, the alveolar cells, and the endothelial cells. The fiber system consists of a number of elements, including collagen and elastin, that are prone to stress failure, especially in relation to regions of inhomogeneity. One can think of the putative mechanisms leading to excessive stress or strain as leading to either stress failure of the membrane of the matrix and/or cell activation of resident or itinerant cells. These stimuli lead to inflammation, repair, and remodeling of the lung.
　　There have been a number of studies that have examined what happens when one overstretches alveolar cells. At relatively low levels of stretch, there is no activation; at midlevels, there can be release of various cytokines and chemokines from cells; and at very high levels of stretch, cell death can occur. Stretching of alveolar cells in vitro leads to release of interleukin-8 and metalloproteinase activity, which can lead to neutrophil attraction and activation, as well as alterations in extracellular matrix. These changes also occur ex vivo and in vivo in various animal models and in patients with ARDS. A recent study using antibodies to CXC-chemokines and CXC-receptors has demonstrated the importance of this mechanism in VILI.[6] Further confirmation was obtained in knock-out mice in which the gene coding for the CXC receptors was knocked out. It is important to point out, however, that one has to be careful in extrapolating the size of the specific tidal volume (Vt) that can lead to injury among animal models and humans since the relationship between Vt normalized to body weight (Vt/kg) varies greatly among species. For example, at a Vt/kg of 10, the average alveolar diameter in a mouse is roughly 3 times that of a human.