INTRODUCTION

Top Abstract Introduction References

THE CHEST WALL makes an important contribution to overall respiratory system mechanics, both in terms of their resistive and elastic properties and their influence on lung volume. Previous studies have partitioned total respiratory mechanics into lung and chest wall components and have investigated the effects of frequency and tidal volume on this contribution in humans (2, 9), dogs (3), cats (12, 17, 20), and rats (13). However, there have been no reports that evaluated the effects of lung volume on chest wall impedance (Zcw) and its contribution to respiratory system impedance (Zrs) in the rat. Given the current widespread use of rats in the study of bronchial responsiveness, it is important to characterize their respiratory mechanical properties under as wide a range of normal conditions as possible. It is particularly important to understand the influence of lung volume on respiratory mechanics, because lung volume has been shown to have a dramatic influence on bronchial responsiveness (6). Therefore, the purpose of the present study was to assess the effects of lung volume on lung and chest wall mechanics in anesthetized, paralyzed rats.

	METHODS

Animal preparation. Seven adult male Sprague-Dawley rats (body weight 290-310 g) were studied in the supine position. The rats were anesthetized by the intraperitoneal administration of xylazine (14 mg/kg) and pentobarbital sodium (30 mg/kg). They were then tracheostomized, and the trachea was cannulated by using a 14-gauge metal needle (the length of the cannula inside the trachea was 1 cm). The cannula was connected to a computer-controlled small-animal ventilator (SAV) that we have developed for assessment of respiratory mechanics in small animals (6, 19). The rats were mechanically ventilated at 90 breaths/min, with a tidal volume of 8 ml/kg against a set level of positive end-expiratory pressure (PEEP) established by a water trap. A venous line was established in the right jugular vein for injection of succinylcholine (6 mg) to paralyze the animals.

Measurements. Assessment of respiratory mechanics was made by interrupting mechanical ventilation for 16 s, during which a specially designed volume-oscillation signal (described below) was applied to the tracheal opening by the SAV. Mechanical ventilation was then immediately resumed while the 16-s data records of the piston displacement and cylinder pressure of the SAV were stored on the hard disk of the host computer (a 486-based personal computer). Before experiments with each rat, calibration signals were collected from the SAV by applying the volume oscillations through the trachea cannula, first with the cannula open to atmosphere and then with it completely closed. The piston displacement and cylinder pressure signals collected during the calibration runs were used to correct the subsequent animal data for the mechanical characteristics of the SAV itself, as described previously (6, 19).

Data analysis. The deflation P-V curves were analyzed by using the method reported by Colebatch et al. (8). A fourth-order polynomial

(1)

was fit to the data, where a, b, c, d, and e are constants. This expression fitted the data well for all rats studied (r > 0.981). P-V curves obtained in closed- and open-chest conditions from each rat were then matched on the basis of Pc and Po, as described above. Zero lung volume was defined as FRC with the chest closed and with a PEEP of zero.

(2)

(3)

	RESULTS

Figure 1 shows an example of the real and imaginary parts of Zrs obtained in a representative closed-chest rat and of ZL obtained in an open-chest rat. These examples are typical of all animals studied and show a real part that decreases and an imaginary part that increases monotonically with f. The constant-phase model fitted Z very well in all cases, although there were some small systematic deviations between data and fit as exemplified in Fig. 1 (i.e., at some of the low-f values).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Example of real (top) and imaginary (bottom) parts of impedance of respiratory system (, measurements in closed chest) and lung (, measurements in open chest). Fit provided by constant-phase model (Eq. 2) is also shown (solid lines).

The parameters R, G, and H obtained by fitting the constant-phase model (Eq. 2 without the second term) to Zrs (closed chest) and ZL (open chest) are shown in Fig. 2. R decreased with lung volume in both cases and was not significantly different between the two conditions at any lung volume (P > 0.35). G decreased initially with increasing lung volume above FRC but then essentially plateaued. H achieved its minimum value at ~2.5 ml above FRC for the respiratory system and at 1.5-2 ml above FRC for the lungs alone. The G and H values for the respiratory system were significantly higher than those of the lung over the entire lung volume investigated, but the fractional differences varied with volume. For example, at FRC, the lung contributed about one-half the value of G for the respiratory system, but this contribution decreased in relative terms as lung volume increased. As lung volume changed, the parameter  changed from 0.82 to 0.87 for respiratory system and from 0.87 to 0.93 for the lungs. The value of  varied with lung volume and was significantly higher for the respiratory system than for the lung at every volume investigated except FRC.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Effect of lung volume on impedance of respiratory system () and lung () in terms of parameters of constant-phase model (Eq. 2). R, Newtonian resistance; G, tissue resistance; H, tissue elastance;  = (2/) arctan(H/G); , hysteresivity. Zero lung volume was functional residual capacity (FRC) with chest closed. Statistical significance of differences of each parameter with its value at zero lung volume: * P < 0.05 for lung; # P < 0.05 for respiratory system. Symbols show mean values, and error bars indicate SD for all animals studied.

Figure 3 shows the mean values (from all animals studied) of R, G, and H for the chest wall and lung. R for chest wall was essentially zero at every lung volume. G for the chest wall was essentially independent of lung volume and was always higher than for the lung. H for the chest wall was at its minimum at a lung volume of 3.0-3.5 ml above FRC. Chest wall  changed markedly with lung volume, achieving a peak value at ~3 ml above FRC, in contrast to the consistently decreasing tendency of  for the lung.

View larger version (35K): [in this window] [in a new window]   Fig. 3.   Effect of lung volume on impedance of chest wall () and lung () in terms of parameters of constant-phase model (Eq. 2), together with .

	DISCUSSION

It is known that the passive mechanical properties of the respiratory system depend on virtually every physical variable relevant to their assessment, including f, tidal volume, and mean lung volume. Numerous studies attest to this in various species (2-4, 9, 12, 13, 17, 20). Consequently, a complete characterization of respiratory mechanics in any species must include an investigation of the influences of these variables over their physiological ranges. In the present study, we have undertaken such a characterization in the rat, because this animal currently enjoys prominence as a model in bronchopharmacological studies. Although the rat has been studied in the past (6, 13, 18), the effects of lung volume on the Newtonian R and tissue Z of the chest wall are not well known. An understanding of the relative contribution of the chest wall to respiratory mechanics is important in any animal that is used for bronchial provocation studies. Without such an understanding, one cannot assess the relative change in lung mechanics that is produced, for example, by a bronchial challenge, from a change in the mechanics of the total respiratory system.

Although our main intention was to separate the effects of the lung and chest wall from overall respiratory Z, we did not measure both Zcw and ZL at the same time. Instead, we assessed lung mechanics independently of the chest wall by opening the chest, as has been done in past studies in other species (12). This makes a number of assumptions that bear consideration. It assumes, for example, that the lung behaves identically at a given volume, regardless of whether the chest wall is present or not. This may not be entirely true, because the removal of the chest wall may cause the lung to be distorted to some degree and there may be some mechanical effects associated with the interaction of the visceral and parietal pleura. Also, loss of blood due to the thoracotomy may affect tissue properties as a result of changes in the degree of vascular congestion. The trauma of the surgery may also release catecholamines and other mediators which might, for example, have an effect on airway smooth muscle tone. We cannot be sure of the magnitudes of these effects, so our results must be interpreted in light of them.

Mechanical Z on its own serves as a model-independent means to characterize a physical system. The variations of the real and imaginary parts of Z can be used as empirical indexes of the state of the system, but they do not lead directly to any meaningful insight into its internal structure. Such insight is only obtained when a model of the system is used to interpret Z; one can hope that the model contains structures and parameters that are readily identified with important components of the real system. In this study, we interpreted our measurements of Z in terms of the constant-phase model (Eq. 1) proposed by Hantos et al. (14, 15) and used by others (21). This model has been shown to provide extremely accurate fits to normal Zrs over the f range we used in our study. Our own results were also fitted very well by this model, although there were some small systematic deviations between model and fit (see the low-f values in Fig. 1) which presumably reflect the fact that the real lung is not quite so simple. The additional attraction of this model is that it is characterized by only four independent parameters (R, I, G, and H), and, in our case, we found we could drop I, thus leaving only three parameters to be estimated. Furthermore, these parameters are readily interpreted in physiological terms, as we discuss below.

The Newtonian Rs of the respiratory system and lung have been studied before in various species. Bates et al. (5) showed R for the chest wall to be roughly equally divided between the lung and chest wall in dogs. Hantos et al. (12) showed in cats that R was almost zero for the chest wall but finite for the lung, which agrees with the present study. This Newtonian R, embodied in the parameter R of the constant-phase model (Figs. 2 and 3), is clearly confined to the lung (R for the chest wall was <3% of G, whereas it was between 25 and 33% of G for the lung). Because previous studies that used the alveolar capsule technique have shown that the Newtonian R of the lung tissue is negligible, this means that the values of R we found in the present study of rats can be ascribed to airway R. This interpretation is supported by our observation that R decreased with increasing lung volume (Fig. 3), as has been previously reported of airway R in humans (7) and rats (18), and as would be expected given that the airways increase in caliber with lung volume.

The increase in G with decreasing lung volume at low volume (Fig. 2) is most likely due to progressive airspace closure that would shut off some of the lung, leaving a reduced parenchymal mass to receive the imposed volume oscillations. Similar results have been observed in lung R in dogs (3, 10). At volumes >1 ml above FRC, however, G was essentially independent of volume. This is somewhat at variance with the results of Nagase et al. (18) who showed that total lung R and tissue R increased with increasing lung volume in the rat with the use of alveolar capsules. However, Barnas and Sprung (3) showed in dogs that lung R was independent of lung volume at high lung volume. These differences may have resulted from differences in the methods used to assess mechanics. For example, alveolar capsule measurements of tissue and airway R are influenced by lung inhomogeneity, which may change with lung volume. G for the chest wall alone (Fig. 3) was essentially independent of lung volume, which is also similar to the findings of Barnas and Sprung (3) in dogs. Interestingly, however, our results show G for the lung to be less than G for the chest wall, in contrast to Barnas and Sprung (3) who found lung R to be much greater than chest wall R at low lung volumes in dogs. This discrepancy could be a species difference, or it could be due to the fact that Barnas and Sprung (3) used tidal volume oscillations to make their measurements, whereas we employed relatively smaller amplitude volume perturbations.

H for the lung (Fig. 2) achieved its minimum value at ~2 ml above FRC. The increase in H with decreasing lung volume below this point again most likely reflected progressive airway closure (3, 10). The increase in H at higher lung volumes was probably due to the nonlinear elastic properties of lung tissue (1, 3, 5). Indeed, the quasi-static elastances of the respiratory system and lung calculated from the slopes of open- and closed-chest P-V curves (Fig. 4) have a rather similar volume dependence to the corresponding H values (Fig. 3), especially at the lower lung volumes. This suggests that the elastic properties of the tissues are embodied similarly in both H and the quasi-static elastance. Chest wall H also exhibited a minimum (Fig. 3), but at a somewhat higher lung volume than the lung H. The variation of chest wall H with volume no doubt reflects the nonlinear elastic properties of the various tissues of the chest wall, such as rib cage, diaphragm, and skeletal muscle. However, the volume dependency of H for both lung and respiratory system was almost consistent with the volume change in the elastance calculated from quasi-static P-V curves (Fig. 4). The variations of H with lung volume we found for the lung and chest wall (Fig. 3) also show similarities with the volume dependencies of lung and chest wall elastance found in dogs by Barnas and Sprung (3). Specifically, both lung and chest wall parameters decreased with increasing volume initially, and then the lung parameters started to increase markedly again, whereas the chest wall parameters exhibited only a very minor increase with lung volume. However, as with G, our values of H for the chest wall were relatively greater compared with H for the lung than were the values of chest wall elastance compared with lung elastance found by Barnas and Sprung (3). Again, a species difference or a volume-amplitude effect might account for these differences.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Effect of lung volume on quasi-static elastance of total respiratory system () and lung (). Quasi-static elastance was calculated from quasi-static pressure-volume curves by differentiation of pressure with respect to volume. Statistical significance of differences compared with values at zero lung volume: * P < 0.05 for lung; # P < 0.05 for respiratory system. Error bars indicate SD.

Hantos et al. (12) reported that  for the chest wall was higher than  for the lung in the cat, similar to our present results. In addition, our results show that  depends on lung volume. Although this dependence is rather modest for the lung (Fig. 3), it is marked for the chest wall. Presumably this reflects the multicomponent character of the chest wall, with various components making different contributions to the resistive and elastic properties of the chest wall as volume changes. Interestingly, however, the values of  for lung and chest wall were rather well matched at FRC (Fig. 3). This compels one to speculate that the mechanical properties of the lung and chest wall tissues might be adapted to function in a similar fashion around those lung volumes normally encountered in life.

In summary, we have studied respiratory system and lung mechanics in rats in terms of their mechanical input Zs interpreted via the constant-phase model. We have shown that the chest wall makes significant contributions to the resistive and elastic properties of the respiratory system to a degree that depends on lung volume. In contrast, the Newtonian component of respiratory R is due, virtually exclusively, to the lung and reflects airway R. We conclude that the mechanical properties of the chest wall in the rat are significant and cannot be neglected when studying respiratory mechanics in the intact animal.