INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Several studies have investigated the effects of acute bronchoconstriction on respiratory muscle recruitment and chest wall mechanics in asthmatic patients (1). In summary, it has been shown that acute airway narrowing was associated with marked hyperinflation ( 50% increase in control end-expiratory lung volume for a 40 to 50% reduction in FEV₁), recruitment of rib cage inspiratory muscles that exceeded that of the diaphragm, and phasic recruitment of abdominal muscles during expiration (1, 3). Because during progressive histamine- induced bronchoconstriction the most positive expiratory pleural pressure during breathing was less than the predicted chest wall relaxation pressure (1), it has been suggested that persistent tonic inspiratory muscle activity during expiration may contribute to hyperinflation (1).

In these studies (1), however, hyperinflation was assessed with the body plethysmographic technique that may substantially overestimate absolute lung volume during bronchoconstriction (4). In fact recent studies in asthmatic patients (5, 6) have shown that the increase in end-expiratory lung volume during acute bronchoconstriction, for a similar degree of airflow obstruction, may be substantially lower than that previously reported (1). Furthermore, the available information on chest wall mechanics during bronchoconstriction (1, 7) is based on the two-compartment chest wall model of Konno and Mead (8) composed of rib cage and abdomen, each behaving with a single degree of freedom, so that changes in volume of each compartment can be measured by a single dimension. This may not be sufficient during acute bronchoconstriction when the chest wall operates with multiple degrees of freedom (7, 9). The large respiratory pressure swings generated during acute bronchoconstriction result in highly nonuniform pressures in the pleural space over the inner surface of the rib cage. The pressure over the pulmonary-apposed part of the rib cage (RC,p) is pleural pressure (Ppl) over the costal surface of the lung, which falls during inspiration, whereas over the diaphragm-apposed part (RC,a), Ppl is approximated by abdominal pressure (10), which normally rises during inspiration. In addition, diaphragm and some of the abdominal muscles act directly only on RC,a, and nondiaphragmatic inspiratory muscles act largely on RC,p (11). The two-compartment rib cage model developed by Ward and coworkers (11) may be more appropriated for the assessment of chest wall mechanics during acute bronchoconstriction.

Recently, it has been shown that an optical reflectance motion analysis system (ELITE) can be used to estimate change in lung volume during breathing by computing the three-dimensional coordinates of passive markers placed on the chest wall (12). Furthermore, unlike magnetometers or respiratory inductance plethysmography, this device allows direct measurements of volume of the three chest wall compartments, RC,p, RC,a, and the abdomen (14). ELITE has been previously used to study the volume distortion between the two compartments of the rib cage, and to infer respiratory muscle action during quiet breathing and exercise in normal subjects (14, 15).

In this study we have assessed the chest wall mechanics in seven asthmatic patients during histamine-induced bronchoconstriction. Our aims were the following: first, to verify that ELITE is able to estimate change in lung volume as accurately during acute bronchoconstriction as during quiet breathing at rest; second, to assess the degree of hyperinflation associated with acute bronchoconstriction and its partitioning into chest wall compartments; third, to assess the relationships between hyperinflation and respiratory muscle recruitment and interaction.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

The study was performed in seven male subjects with mild to moderate asthma (16) (mean age, 40 yr, range, 29 to 57 yr). They were recruited from physicians and respiratory therapists of the Careggi Hospital and were experienced in performance of respiratory maneuvers. Each patient had been clinically stable (defined as an absence of exacerbations of asthma necessitating alterations in medication) for at least 4 wk before the study. All patients gave informed consent to the protocol as approved by the Institution's Ethics Committee.

Measurements

All patients were studied while sitting comfortably in armchairs with forearms supported away from the sides of the body comfortably below shoulder height, allowing visualization by TV cameras of the markers placed on the surface of the chest wall. Each patient breathed through a mouthpiece while wearing a noseclip. Spirometry was performed according to standard technique using a water-sealed spirometer (Godart, Utrecht, The Netherlands); functional residual capacity (FRC) was measured by the helium dilution technique. Predicted values for lung function variables are those proposed by the European Community for Coal and Steel (17).

Bronchoconstriction of progressively increasing severity was induced by administration of an aerosol of histamine, which was increased in doubling concentrations (initial concentration, 0.031 mg/ml). Solutions of histamine acid phosphate were administered during tidal breathing over 2 min using a nebulizer (Model 646; DeVilbiss, Somerset, PA) that delivered 0.31 ml/min at an airflow of 6 L/min. End-tidal carbon dioxide pressure (PET_CO2) was sampled continuously at the mouth by an infrared carbon dioxide meter (Datex Normocap, Helsinki, Finland), and arterial oxygen saturation (Sa_O2) was monitored by an ear oximeter (Radiometer, Copenhagen, Denmark). Histamine administration was discontinued any time one of the following criteria was met: a fall in FEV₁ > 40% of control; a plateau in FEV₁ (< 5% change in FEV₁ over two or more steps); discomfort or breathlessness sufficient to induce the patient to interrupt the test. The test was also interrupted when FEV₁ approximated 40% of predicted value.

Airflow was measured with a No. 3 Fleisch pneumotachograph and a Validyne pressure transducer (Validyne Corp., Northridge, CA), and the flow signal was integrated to give volume. The dead space of the mouthpiece and flowmeter was 70 ml, and the equipment resistance was 0.92 cm H₂O/L/s.

The volume of the chest wall (VCW) was modeled as the sum of the volumes of the pulmonary-apposed rib cage (VRC,p), diaphragm- apposed rib cage (VRC,a), and abdomen (VAB) (11). The volumes of the chest wall and its compartments were assessed applying a new optical reflectance motion analysis system called ELITE (ELaboratore di Immagini TElevisive, Milan Polytechnic, Milan, Italy) able to compute with high spatial accuracy (± 0.2 mm) (18) the three-dimensional coordinates of 89 reflective markers applied to the chest wall (13). The components of the ELITE have been described in detail previously (13). Briefly, small (6 mm), lightweight hemispherical markers were applied to body landmarks with double-sided adhesive tape. The markers were placed circumferentially in seven horizontal rows between the clavicles and the anterior superior iliac spine (Figure 1). Along the horizontal rows the markers were arranged anteriorly and posteriorly in five vertical rows, and there was an additional bilateral row in the midaxillary line. In agreement with Cala and coworkers (13), the anatomic landmarks for the horizontal rows were: (1) the clavicular line, (2) the manubriosternal joint, (3) the nipples (~ 5 ribs), (4) the xiphoid process, (5) the lower costal margin (tenth rib in the midaxillary line), (6) umbilicus, and (7) anterior superior iliac spine. The landmarks for the vertical rows were: (1) the midlines, (2) both anterior and posterior axillary lines, (3) the midpoint of the interval between the midline and the anterior axillary line, and the midpoint of the interval between the midline and the posterior axillary line, (4) the midaxillary lines. An extra marker was added bilaterally at the midpoint between the xiphoid and the most lateral portion of the tenth rib to provide better detail of the costal margin; two markers were added in the region overlying the lung-apposed rib cage and in the corresponding posterior position. As a result, there were 42 anterior markers, 37 posterior markers, and 10 lateral markers. This marker configuration has previously been validated in normal subjects, along with a sensitivity analysis, which assesses accuracy in estimating change in lung volume as a function of marker number and position (13). With the exception of the midaxillary positions, all markers were attached to the skin with double-sided adhesive tape. Because the midaxillary markers were not visible to the cameras side on, they were mounted on rear-facing right-angle brackets, before being attached to the skin.

View larger version (19K): [in this window] [in a new window]   Figure 1.   Placement of 89 small (6 mm) passive reflective markers on the chest wall (42 anterior, 37 posterior, and 10 laterally). The markers were arranged in seven horizontal rows between the clavicles and the anterior superior iliac spines, five vertical rows both anteriorly and posteriorly, and one vertical row bilaterally in the midaxillary line. The lines indicate the boundaries between the two compartments of the rib cage, and between diaphragm-apposed rib cage and abdomen.

To measure the volume of chest wall compartments from surface markers, in agreement with Kenyon and coworkers (14), we defined: (1) the boundaries of pulmonary-apposed rib cage (RC,p) as extending from the clavicles to a line extending transversely around the thorax at the level of the xiphoid process (corresponding to the top of the area of the apposition of the diaphragm to the rib cage at end-expiratory lung volume in sitting posture, confirmed by percussion); (2) the boundaries of diaphragm apposed rib cage (RC,a) as extending from this line to the costal margin anteriorly down from the xiphisternum, and to the level of the lowest point of the lower costal margin posteriorly; and (3) the boundaries of abdomen as extending caudally from the lower rib cage to a horizontal line at the level of the anterior superior iliac spine. The anatomic placement of the markers and the boundaries between chest wall compartments are shown in Figure 1.

The co-ordinates of the landmarks were measured with a system configuration of four infrared TV cameras, two placed 4 m behind and two 4 m in front of the subject (Figure 2), at a sampling rate of 50 Hz. Starting from these coordinates the volume of the chest wall was computed by triangulating the surface and then using Gauss's theorem to convert the volume integral to an integral over this surface, as described previously (13). The chosen markers' arrangement and geometrical model allow the computation of the contribution of RC,p, RC,a, and AB to the hyperinflation and tidal volume. End-expiratory and end-inspiratory volume of each compartment was measured at the beginning and the end of inspiratory flow (zero-flow points). The difference between the end-inspiratory and end-expiratory volume of each compartment was calculated as the tidal volume contribution by each compartment. Thus the chest wall volume (VCW) = VRC + VAB, with VRC = VRC,p + VRC,a, and changes in lung volume (VL), can be calculated as:

View larger version (19K): [in this window] [in a new window]   Figure 2.   Experimental set-up of specially designed TV cameras (two cameras were 4 m in front of and two cameras were 4 m behind the subject) and subject position. See text for explanation.

assuming that the only factor causing chest wall volume changes is gas movement.

ELITE calculates absolute volumes and the absolute volume of each compartment at FRC in control conditions was considered as the reference volume. Volumes are reported in either absolute values or as changes from the volume at FRC in control conditions.

Mouth pressure (Pm) was measured through a side port at the mouthpiece using a differential pressure transducer (Validyne). Esophageal (Pes) and gastric (Pga) pressures were measured with conventional balloon-catheter systems connected to Validyne differential pressure transducers, as previously described (19). One balloon, positioned in the midesophagus and containing 0.5 ml of air, measured Pes; the other, positioned in the stomach 65 to 70 cm balloon tip to nares, and containing 2 ml of air, simultaneously measured Pga. Pes was used as an index of pleural pressure (Ppl) and Pga was used as an index of abdominal pressure. Transpulmonary (PL) and transdiaphragmatic (Pdi) pressures were obtained by electrical subtraction of Ppl from Pm, and of Ppl from Pga, respectively. Transdiaphragmatic pressure at end-expiration during quiet breathing in control conditions was assumed to be zero. Total lung resistance was measured during resting breathing using the isovolume method of Frank and coworkers (20). All signals were converted from the analog to the digital mode and recorded on a IBM personal computer.

Protocol

Before the experiments each subject was made well acquainted with the laboratory and equipment.

Lung function tests were performed first. Relaxation characteristics were established for the chest wall and its compartments by having subjects breathe in to TLC and then relax, and breath out to FRC through a high resistance. Relaxation maneuvers were repeated until curves were reproducible, Pm finished at zero, and Pdi was zero throughout. Subsequently, VCW, VRC,p, VRC,a, VAB, along with pressures, and flow and volume signal from the pneumotachograph were recorded during two periods of quiet, resting breathing. Two minutes after each histamine concentration was given, measurements during resting breathing were repeated followed by inspiratory capacity and FEV₁ assessment. In each patient all measurements during resting breathing were averaged over 2 min in control conditions and over 1 to 2 min after each histamine administration.

At the end of experiments albuterol was administered by nebulization. A physician not involved in the study checked the clinical conditions of the patients throughout the experiments until FEV₁ returned to baseline value.

In two patients a complementary study was performed to compare changes in end-expiratory lung volume recorded by water-sealed spirometer with changes in VCW recorded by ELITE. Subjects had a visual feedback of change in lung volume recorded by spirometer and were requested to voluntary increase their end-expiratory lung volume by 0.3, 0.6, 0.8, and 1 L and to maintain apnea for 10 s after each step. Each patient repeated these maneuvers twice.

Data Analysis

The slopes of VCW/Ppl and VRC,p/Ppl relaxation lines were calculated in each subject by single regression analysis. The plot of VRC,p versus VRC,a during relaxation defined the undistorted rib cage configuration. Because the plot of VAB versus Pga during relaxation was curvilinear, the VAB/Pga curve was manually fitted to the pressure-volume data.

Data were averaged for the group of subjects, and they are presented as means (SE). Data obtained under control conditions and during maximal bronchoconstriction were compared by using Student's t test for paired samples. Changes from baseline values in measured variables during maximal bronchoconstriction are presented as means with 95% confidence intervals (95% CI). Single regression analysis was performed to assess relationships between tidal volume estimated by ELITE (VTCW) and pneumotachograph (VTPT). The absolute error of ELITE was calculated as |VTCW-VTPT|/VTPT × 100. Similarly, in the complementary study, the error of ELITE in estimating the voluntary increase in end-expiratory lung volume (EELV) was assessed by comparison with measurements obtained from the spirometer. A p value of  0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The anthropometric and baseline pulmonary function data of the seven patients are shown in Table 1. The geometric mean of the largest concentration of histamine administered was 4.0 mg/ml. FEV₁ fell by mean of 42.8% of the control value (95% CI, 35.1 to 50.5) at maximal bronchoconstriction (Table 1), and this was accompanied by a rise in RL from a control value of 3.0 (0.5) to 15.8 (3.5) cm H₂O/L/s (mean change, 12.8 cm H₂O/L/s; 95% CI, 5.2 to 20.4). During the experiments, however, FEV₁ remained  40% of predicted value in all the patients. The parameters of linear regressions relating VTCW to VTPT, and the absolute errors of ELITE in estimating VT, are shown in Table 2. In all patients the relationship of VTCW to VTPT both in control conditions and during maximal bronchoconstriction was highly significant (r > 0.96), the slope and the intercept being not significantly different from 1 and 0, respectively. The absolute error of ELITE in estimating VT was never greater than 4%.

The slopes of VCW/Ppl and VRC,p/Ppl plots during relaxation are given in Table 3. In each patient the correlation coefficient for both relationships was greater than 0.94. 

During maximal bronchoconstriction, compared with control conditions, the ratio of the inspiratory time to total time of the respiratory cycle (TI/Ttot) decreased significantly, whereas the other time and volume components of the breathing cycle did not exhibit any significant change (Table 4). End-expiratory Pga but not PplEE increased significantly (Table 4). Plots of changes in VCW versus Ppl, VRC,p versus Ppl, VRC,p versus VRC,a, and VAB versus Pga in a representative patient under control conditions and during maximal bronchoconstriction are shown in Figure 3. In this patient, the hyperinflation associated with bronchoconstriction was due to an increase in end-expiratory volume of the rib cage, the dimension of VAB at end-expiration being essentially unchanged (Figure 3, lower right panel ). Pulmonary-apposed rib cage and diaphragm- apposed rib cage shared proportionally the hyperinflation of the rib cage, as shown by the fact that end-expiratory VRC,p and VRC,a move along the rib cage relaxation configuration (Figure 3, lower left panel ). The plots of VCW versus Ppl (Figure 3, upper left panel ) and VRC,p versus Ppl (Figure 3, upper right panel ) during bronchoconstriction were completely displaced leftward of the relaxation lines, such that the maximal expiratory pleural pressure during breathing was less than chest wall and pulmonary-apposed rib cage relaxation pressures. This provides evidence that hyperinflation in this patient was associated with persistent activity of the rib cage inspiratory muscles throughout expiration. The plot of VCW versus Pdi (Figure 4) shows that Pdi at end-expiration was slightly increased during induced bronchoconstriction compared with control conditions. Transdiaphragmatic pressure at end-expiration was unchanged in two patients and increased in the other five (mean change, 2.1 cm H₂O; 95% CI, 0.5 to 3.6; p < 0.05). To quantify the tonic activity of inspiratory muscles throughout expiration during bronchoconstriction, we measured the difference (PplEE-R) between the maximal expiratory pleural pressure and both chest wall and RC,p relaxation pressures (Figure 3, upper left and right panels).

View larger version (21K): [in this window] [in a new window]   Figure 3.   Plots of changes in chest wall volume (VCW) versus Ppl, pulmonary-apposed rib cage (VRC,p) versus Ppl, pulmonary-apposed rib cage (VRC,p) versus diaphragm-apposed rib cage (VRC,a), and abdominal volume (VAB) versus Pga in a representative patient (Patient 5 in Table 1) under control conditions (continuous line) and during bronchoconstriction (dashed line). The dotted line in each panel is the relaxation curve; closed and open circles indicate end-expiration and end-inspiration, respectively. The arrow in the plot of VCW versus Ppl, and VRC,p versus Ppl indicates the difference (PplEE-R) between the maximum (less negative) expiratory pleural pressure during bronchoconstriction and chest wall or RC,p relaxation pressure.

View larger version (13K): [in this window] [in a new window]   Figure 4.   Plot of changes in chest wall volume (VCW) versus transdiaphragmatic pressure (Pdi) in a representative patient (Patient 5 in Table 1) under control conditions (continuous line) and during bronchoconstriction (dashed line). Closed and open circles indicate end-expiration and end-inspiration, respectively.

The VAB/Pga plot (Figure 3, lower right panel ) shows that in this patient under control conditions inspiration was associated with a progressive rise in Pga and a progressive increase in VAB, indicating a shortening and a descent of the diaphragm, whereas during expiration the fall in Pga occurred with a decrease in VAB close to the relaxation line. In all patients bronchoconstriction was associated with considerable recruitment of abdominal muscles during expiration as indicated by the end-expiratory displacement of VAB/Pga plot rightward of the relaxation line.

In all patients bronchoconstriction was associated with hyperinflation (Figure 5), the mean increase in end-expiratory VCW being 0.72 L (95% CI, 0.43 to 1.0; p < 0.001) (23.6% of FRC under control conditions). Changes in end-expiratory VCW during bronchoconstriction related neither to changes in FEV₁, and RL, nor to changes in time and volume components of the breathing cycle. On the contrary, there was a significant inverse relationship between changes in end-expiratory VCW and PplEE-R (r = 0.75, p = 0.05) (Figure 6), such that hyperinflation was greatest in patients with the highest tonic inspiratory muscle activity throughout expiration. A similar inverse relationship was found between changes in end-expiratory VRC,p and PplEE-R (r = 0.78, p < 0.05). Although end-expiratory Pdi increased during bronchoconstriction in five of the seven patients (see above), changes in end-expiratory VCW did not relate to changes in Pdi at end-expiration (r = 0.60, p = 0.15).

View larger version (28K): [in this window] [in a new window]   Figure 5.   Individual (open bars) and mean (hatched bar) changes in end-expiratory volume of the chest wall (VCW) associated with maximal bronchoconstriction in seven patients with asthma.

View larger version (12K): [in this window] [in a new window]   Figure 6.   Relationship between changes in end-expiratory volume of the chest wall (VCW) and differences (PplEE-R) between the maximal expiratory Ppl during bronchoconstriction and the correspondent chest wall relaxation pressure. The solid line is the regression line, and the closed circles are individual patients.

The partitioning of hyperinflation in the rib cage and abdomen compartments of the chest wall during bronchoconstriction is shown in Figure 7. End-expiratory VRC increased in all patients (mean increase, 0.63 L; 95% CI, 0.41 to 0.83; p < 0.001), whereas changes in VAB were inconsistent (mean increase, 0.09 L; 95% CI, 0.09 to 0.27, NS). The RC compartment of the chest wall accounted for the 89.9% (8.6) of the volume of hyperinflation during bronchoconstriction.

View larger version (19K): [in this window] [in a new window]   Figure 7.   Individual (open bars) and mean (hatched bar) changes in end-expiratory volume of the rib cage (VRC) (left panel ) and abdomen (VAB) (right panel ) associated with maximal bronchoconstriction in seven patients with asthma.

As shown in Figure 8, changes in end-expiratory VRC,p and VRC,a were close to the relaxation configuration in all patients, indicating that both compartments shared proportionally the hyperinflation of the rib cage.

View larger version (17K): [in this window] [in a new window]   Figure 8.   Changes in end-expiratory volume of pulmonary-apposed rib cage (VRC,p) versus diaphragm-apposed rib cage (VRC,a) during bronchoconstriction in seven patients with asthma. The solid line is the mean relaxation line of VRC,p versus VRC,a, and the closed circles are individual patients.

In the two subjects in whom we compared during voluntary hyperinflation changes in end-expiratory VCW with changes in EELV measured by water-sealed spirometer, there was a good agreement between these two measurements. The absolute error of ELITE in assessing change in EELV in the two subjects was 3.0 (0.6) and 1.7 (0.6)%, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main findings of the present study can be summarized as follows: (1) In asthmatic patients there was a good agreement between VTCW and VTPT during acute bronchoconstriction and between changes in end-expiratory VCW and EELV during voluntary hyperinflation, indicating that ELITE system allows accurate estimation of changes in lung volume during bronchoconstriction as in control conditions. (2) The rib cage largely accounted for the volume of hyperinflation associated with acute bronchoconstriction, whereas phasic abdominal muscle recruitment during expiration minimized the increase in end-expiratory volume of the abdomen in the majority of patients. (3) Changes in end-expiratory VCW during acute bronchoconstriction related significantly to PplEE-R but not to changes in time and volume components of the breathing cycle, and indices of bronchial obstruction (FEV₁ and RL), suggesting that hyperinflation is influenced by persistent tonic activity of the inspiratory muscles throughout expiration. (4) Pulmonary-apposed rib cage and diaphragm-apposed rib cage contributed proportionally to the hyperinflation, such that volume distortion of the rib cage at end expiration was negligible.

Critique of Methods

In the present study changes in lung volume during acute bronchoconstriction was inferred by measuring change in VCW with ELITE system. This assumed that blood shifts to and from the trunk and gas compression were negligible. Previous studies (13) have shown that the ELITE system with the marker configuration we used allows very accurate estimation of changes in lung volume during both breathing at rest and during exercise in normal subjects. During acute bronchoconstriction, however, the increase in respiratory pleural pressure swings could increase the flux of the blood to and from the trunk compared with control conditions. Furthermore, it has been shown that during acute bronchoconstriction the chest wall moves with multiple degrees of freedom (9), such that estimates of lung volume changes from measurements of cross- sectional areas or diameters of the chest wall may introduce substantial errors (7, 21). In the present study we found a close agreement between VT calculated from surface markers using the ELITE system and VT measured by pneumotachograph during acute bronchoconstriction as in control conditions. Furthermore, we found a good relationship between changes in EELV estimated by the ELITE system and those measured by a water seal spirometer. These findings indicate that, unlike magnetometers and inductance plethysmography, the ELITE system with the marker configuration we used allows accurate estimates of tidal volume and change in EELV during acute bronchoconstriction in patients with asthma.

Surface markers were used to define the boundary between chest wall compartments. This may cause some errors in compartmental volume calculation for two reasons. First, there could be movements of skin relative to the underlying bone structure during increased ventilation. It has been previously shown (14) that in normal subjects during exercise, because of hyperventilation, the marker displacements relative to the lower costal margin may introduce errors relative to the absolute volume of RC,a and AB of about 10 and 2.5%, respectively. In the present study, however, tidal volume and minute ventilation did not change significantly during acute bronchoconstriction compared with control conditions. Second, the boundary between RC,p and RC,a was assumed to be at the level of the xiphoid, which corresponds to the top of the area of the apposition of the diaphragm to the rib cage at end-expiration in sitting posture, as confirmed by percussion. During acute bronchoconstriction end-expiratory lung volume increased, and as a consequence, the upper boundary of the diaphragm-apposed rib cage could change. To the extent that abdominal displacement is the principal determinant of the upper boundary of the area of apposition (22, 23), our observation that changes in end-expiratory VAB were generally minimized during bronchoconstriction suggests that the error caused by a changing boundary of the upper limit of the area of apposition on the dynamics of RC,p and RC,a was small. The recruitment of different groups of respiratory muscles was inferred by comparing volume-pressure plots during breathing with the relaxation curves of the chest wall compartments, which are notoriously difficult to achieve. To minimize this problem, we studied only patients experienced in physiologic studies and in performance of respiratory maneuvers, and relaxation maneuvers were repeated in each subject until curves were reproducible, Pm finished at zero, and Pdi was zero throughout.

Chest Wall Hyperinflation

In line with previous studies (1, 5, 24) the present findings indicate that end-expiratory lung volume increases during induced bronchoconstriction in patients with asthma. As a mean the increase in EELV we found corresponded to 23.6% of FRC under control conditions, for a 43% FEV₁ decrease. Although there was some variability among patients, this degree of hyperinflation is substantially lower than that (40 to 50% of control FRC) reported previously in studies devoted to investigate respiratory muscle recruitment and chest wall mechanics in asthma (1, 7). The body plethysmographic technique used in these studies may, however, overestimate absolute lung volume during bronchoconstriction (4). The effects of acute bronchoconstriction on chest wall mechanics have been previously analyzed by comparing volume-pressure plots with the predicted chest wall relaxation curves (1); thus, a marked overestimation of the degree of hyperinflation imposes caution about the conclusions of these reports on respiratory muscle recruitment and work of breathing. The present findings allow us to confirm clearly that hyperinflation during acute bronchoconstriction in asthmatic patients is associated with the complete displacement of VCW/Ppl plot leftward of the chest wall relaxation line, indicating persistent tonic activity of inspiratory muscle throughout expiration (1). Our results showing that during bronchoconstriction VRC,p/Ppl plot was consistently displaced leftward of the RC,p relaxation curve, and that changes in end-expiratory VCW related significantly to PplEE-R, but not to changes in Pdi at end-expiration, suggest that hyperinflation was mainly influenced by tonic activity of inspiratory muscles acting on RC,p, i. e., the scalenes and inspiratory intercostal muscles. The increase in end-expiratory Pdi we observed in five of the seven patients during bronchoconstriction, however, could play a role in the increase in VRC,a at end-expiration (see next paragraph).

Recently it has been shown that hyperinflation during acute bronchoconstriction in asthmatic patients occurs when it would be impossible for them to keep tidal volume at the control FRC without expiring under conditions of flow limitation (24). On the basis of this observation it has been hypothesized that the increase in EELV during acute bronchoconstriction may represent a strategy of breathing adopted to prevent flow limitation (25). Our results suggest that sustained postinspiratory activity of inspiratory muscles contributes significantly to this breathing strategy in asthmatic patients. The mechanisms by which acute bronchoconstriction induces sustained postinspiratory activity of the inspiratory muscles may only be hypothesized. Afferent information from mechanoreceptors of the airway (26) could modulate the activity of both inspiratory and expiratory muscles during expiration in order to avoid breathing under conditions of flow limitation.

Partitioning of Hyperinflation in Chest Wall Compartments

Because during acute bronchoconstriction the chest wall operates with multiple degrees of freedom, the partitioning of hyperinflation in the chest wall compartments cannot be accurately estimated by using magnetometers or respiratory inductive plethysmograph (7, 9). Unlike these devices, the ELITE system allows direct measurements of volume of the chest wall compartments, such that one of the most important findings of the present study was that the rib cage compartment accounted for 89.9% of the volume of hyperinflation during acute bronchoconstriction, whereas changes in end-expiratory abdominal volume were inconsistent. Furthermore, our findings suggest that even though the two compartments of the rib cage are exposed to the action of different respiratory muscles (see opening statements), RC,p and RC,a shared proportionally the hyperinflation minimizing the volume distortion of the rib cage at end-expiration. This finding contrasts with previous investigations that have found significant cross-sectional shape changes of the rib cage during acute bronchoconstriction in patients with asthma (9). The discrepancy is probably due to the fact that the model of chest wall we used did not allow us to assess intracompartmental distortion but only volume distortion between compartments.

The lack of volume distortion of the rib cage at end-expiration was probably the result of a coordinated action of respiratory muscles. The sustained postinspiratory activity of rib cage inspiratory muscles contributed to the expansion of RC,p at end-expiration. The proportional expansion of end-expiratory VRC,a could be explained by the action of abdominal muscles and diaphragm. The net effect of abdominal muscle contraction on RC,a depends on the balance between their insertional action, which tends to deflate the RC,a, and the rise in abdominal pressure, which, operating by way of the area of apposition, tends to inflate RC,a (27). We can speculate that preferential expiratory recruitment of abdominal muscles that do not have an insertional action on the lower rib cage (i.e., the transversus abdominis) contributed to the increase in end-expiratory VRC,a during bronchoconstriction. Second, compared with control conditions, Pdi at end-expiration during bronchoconstriction was increased in five of the seven patients studied. This increase could be due to passive stretching of diaphragm muscle fibers because of abdominal muscle recruitment during expiration (28), or to persistent activity of diaphragm throughout the expiration (29). Whatever the mechanism, diaphragm tension tends to raise the lower ribs (28) and could contribute to the increase in end-expiratory VRC,a during acute bronchoconstriction.

As a consequence of changes in the end-expiratory configuration of the chest wall we found in asthmatic patients during bronchoconstriction, diaphragmatic length and shape, and force-generating ability of the diaphragm tended to be preserved at the onset of inspiration despite lung hyperinflation. On the other hand, the predominant contribution of the rib cage to the hyperinflation reduces the operational length of the rib cage inspiratory muscles. Although the force-generating ability of the parasternal intercostal muscles appears to be preserved at high lung volumes (30), recent animal studies (31) show that their ability to expand the rib cage and inflate the lung is decreased during hyperinflation due to increased impedance of the ribs to cranial motion (32). The pattern of respiratory muscle recruitment during acute bronchoconstriction in asthmatic patients may therefore have a protective effect on the diaphragm, but it may increase the load on the rib cage inspiratory muscles.

In conclusion, we have shown that during acute bronchoconstriction in patients with asthma the rib cage accounts largely for the volume of hyperinflation, whereas abdominal muscle recruitment during expiration minimizes increases in abdominal volume. Hyperinflation appears to be significantly influenced by sustained postinspiratory activity of the inspiratory muscles. This pattern of respiratory muscle recruitment seems to minimize volume distortion of the rib cage at end-expiration and to preserve diaphragm length despite hyperinflation, but it may increase the load on the inspiratory muscles of the rib cage. Finally, the ELITE system can be used to estimate accurately changes in lung volume during acute bronchoconstriction as in control conditions.