CONTRIBUTION OF OPENING AND CLOSING OF LUNG UNITS TO LUNG HYSTERESIS

The recruitment and derecruitment of lung units is one explanation of the hysteresis observed in an excised lung during inflation and deflat ion. A simplified model has been proposed in which the recruitment-der ecruitment process is a function of end-expiratory pressure (Frazer, D .G., K.C. Weber and G.N. Franz, Respir. Physiol. 61: 277-288, 1985). T he object of this study was to test this model with three experimental procedures. During the first set of experiments, progressively larger pressure-volume (PL-VL) loops were recorded with end-expiratory press ure held at either -5 cmH(2)O, where all lung units are assumed to be closed, or +5 cmH(2)O, where all recruited lung units are assumed to b e open. In the first case hysteresis is maximal, in the second, minima l. The difference in hysteresis is presumed to arise from the recruitm ent-derecruitment process. In the second set of experiments, excised l ungs are slowly inflated and then deflated at a constant rate while co nstant-amplitude sinusoidal volume oscillations are superimposed. The end-expiratory pressure of the superimposed loops gradually rose as th e lung was inflated and fell as the lung was deflated. Hysteresis was minimal when end-expiratory pressure was above 4+/-1 cmH(2)O even as p eak-to-peak loop pressure greatly varied. This supports the notion of an end-expiratory pressure dependent mechanism of recruitment/derecrui tment. During the third set of experiments lungs were inflated to eith er 50%, 75%, or 100% TLC. Volumes of air were then withdrawn and repla ced so that the initial volume was restored in sinusoidal fashion as t he amplitude of the volume excursions increased. For PL-VL loops with end-expiratory pressures between +4 and -2 cmH(2)O, pressure amplitude s rose and the hysteresis index (loop area/tidal volume) increased, re gardless of the initial lung volume. These results are consistent with the previously described model of Frazer et al. (1985) which assumed that PL-VL curves can be divided into an 'opening' region, an 'open' r egion and a 'closing' region and that the demarcation of these regions depends on transpulmonary pressure, specifically end-expiratory press ure, and to a much lesser degree on lung volume.