Lung mechanics--the inverse problem.

Abstract

The mechanical properties of the lung are important determinants of its efficacy as a gas exchanging organ. These properties are reflected to a precise degree in the relationships between pressure and flow measured at the mouth. Together with oesophageal pressure, which reflects the pressure in the pleural space, these quantities allow one to usefully indulge in inverse modelling of the lung - that is, identify mathematical models of lung mechanics that give insight into its structure and may be diagnostic of certain diseases. The complexity of such models, however, is limited by the number of distinct components that can be unambiguously resolved from the measured signals. The development of more detailed models requires the availability of experimental methods for obtaining additional input-output information from the lungs. One such method is the so-called alveolar capsule technique which allows alveolar pressures at several sites on the lung surface to be measured directly. This technique has been used in animals to show that the mechanical behaviour of normal lungs in the breathing frequency range is well described by a homogeneously ventilated compartment surrounded by viscoelastic tissue. During bronchoconstriction, however, the lungs can become markedly inhomogeneous resulting from differences in regional resistive and elastic properties. Model ambiguity problems again appear as it becomes impossible to distinguish changes in local resistance from changes in elastance using only the information obtained from alveolar capsules. To push the inverse modelling of the lung one step further, we have recently developed a new technique for quantifying changes in local resistance and elastance by applying broad-band oscillations in flow to the lung through a small hole in the pleura, thereby obtaining an alveolar input impedance.