INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Many normal subjects develop plateaus on dose-response curves produced by inhalation challenge tests with bronchoconstricting agonists (1, 2). The plateaus usually occur after only mild degrees of airway narrowing despite the fact that in vitro studies show that maximally activated airway smooth muscle (ASM) is capable of shortening to a sufficient degree that airway closure could occur (3). It has been calculated that in vivo ASM smooth muscle shortening is only 20 to 30% when plateaus develop (4, 5). The development of plateaus on dose-response curves has been interpreted to indicate that maximal ASM activation has occurred but that some mechanism, either neural, humoral, or mechanical, prevents further smooth muscle shortening. The most prevalent hypothesis to explain the plateau is that elastic loads provided by lung parenchymal recoil, tidal transmural pressure swings, and/or mucosal folding are sufficient to balance the shortening ability of maximally activated ASM (6).

However, an alternate explanation for the plateaus has been suggested. Progressive ASM activation could occur throughout the plateau but progressive hyperinflation and/or parenchymal stiffening (7) could increase the muscle load to attenuate further ASM shortening and airway narrowing. If this occurred it would appear that a static equilibrium had been reached between ASM activation and mechanical afterloads when, in fact, a dynamic situation was occurring in which progressive activation was balanced by progressive loading. To test whether the latter mechanism could be responsible for the plateau we measured pulmonary resistance and maximal, minimal, and mean pulmonary elastic recoil pressure (PEL_max, PEL_min, PEL_mean) during tidal breathing throughout methacholine challenge in normal subjects. Our results are consistent with the plateau being related to maximal activation of ASM.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ten nonasthmatic subjects who had no history of cardiopulmonary disease, eight men and two women, volunteered for this study. To make measurements of lung resistance (RL) and to approximate pleural pressure, subjects swallowed an esophageal balloon. The balloon was positioned in the lower third of the esophagus, approximately 10 ml from the gastroesophageal junction, and inflated with 0.6 to 1.0 ml of air (8). Pleural pressure was measured by attaching the esophageal balloon to a differential pressure transducer (Validyne Mp 45-2, ± 100 cm H₂O; Validyne Corp., Northridge, CA), and compared with mouth pressure to obtain transpulmonary pressure (PL). RL was calculated from PL and flow, using a recursive least-squares algorithm that continually fits the data points for flow, volume, and pressure to the equation of motion for the lung (9). In addition, we recorded PL_max, PL_min, PEL_max, PEL_min, and PEL_mean during the last 20 s of the measurement of RL at baseline and after each dose of methacholine. PEL_max was the PL at the end inspiratory zero flow point and PEL_min was the PL at the end-expiratory zero flow point. Mean PEL was (PEL_max + PEL_min)/2.

Prior to challenge, triplicate baseline measurements of RL, PEL, PL, spirometry (FEV₁, FVC), and inspiratory capacity (IC) were made. All studies were done with the subjects seated in a volume-displacement, pressure-compensated body plethysmograph. Because the effect of stretch on airway smooth muscle shortening could be related to the frequency of cyclical stretch, we kept breathing frequency constant during both inhalation of methacholine and during measurements of RL. The respiratory rate was maintained at 30 breaths/min by use of a metronome. Volume was measured using the Krogh water-sealed spirometer coupled to a linear displacement transducer (Type 300HR; Shaevitz, Pennsauken, NJ). Flow was measured using a Fleisch no. 3 pneumotachometer coupled to a differential pressure transducer (MP 45-28, ± 3.5 cm H₂O; Validyne). Data were electronically recorded using a digital data acquisition system (Direc; Raytech Instruments, Vancouver, BC, Canada) at a frequency of 100 Hz. Spirometric values were recorded in liters and expressed as percent predicted, based on the prediction equations of Crapo and coworkers (10).

After baseline values were established, subjects underwent methacholine challenge according to the protocol of Juniper and coworkers (11). Subjects inhaled an aerosol of methacholine beginning with a concentration of 1 mg/ml and increasing by doubling concentrations, to a maximum of 256 mg/ml. The aerosol was generated using a Hudson Bennett Twin Jet nebulizer (Hudson Bennett, Temecula, CA) calibrated to generate an output of 0.13 ml/min. Subjects inhaled the aerosol for 2 min breathing tidally through a mouthpiece with their noses clipped and while seated in the plethysmograph. Tidal volume (VT) and PL pressure were measured and recorded throughout the challenge. RL, PEL_max, PEL_min, PL and VT were measured for 20 s at 30 s and 90 s after each dose. FEV₁, FVC, and IC were repeated in triplicate at the end of the challenge. Spirometry was not performed between doses, and deep inspirations were prohibited during the challenge.

Data Analysis

Dose-response curves for each of the subjects were created by plotting RL and PEL_mean against the log dose of methacholine. A plateau on the dose-response curve was defined as a less than 25% change in RL value over three successive doses. Values for RL, PL, PEL_max, PEL_min, PEL_mean, VT, FEV₁, and IC at the end of challenge were compared with baseline using paired t tests. Pearson's correlation coefficients were used to examine the relationships between changes in IC, FEV₁, and RL.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The baseline pulmonary function and anthropometric data for the subjects as shown in Table 1. Figure 1 shows mean data for RL and PEL_mean before and during the methacholine dose-response curves are shown in Figure 1. There was a significant change in both RL and PEL_mean, from baseline to the final dose of 256 mg/ml. The mean increase in RL was from 1.4 ± 0.13 to 6.0 ± 0.61 cm H₂O/L/s (p < 0.05), and the mean increase in maximal PEL_mean was from 5.4 ± 0.66 to 7.7 ± 0.92 cm H₂O (p < 0.05). There was no change in VT during the challenge. All subjects reached a plateau, at a geometric mean dose of 26 (range, 4 to 64 mg/ml). It is apparent that a plateau develops simultaneously on the dose-response relationships for RL and PEL_mean (Figure 1) and also RL and PL (Table 2).

View larger version (16K): [in this window] [in a new window]   Figure 1.   Dose-response curves for lung resistance and pulmonary elastic recoil pressure (mean ± SEM). The group mean data reflect similar results in all subjects, and there was a plateau on both curves.

There was a significant decrease in IC during the challenge (3.36 ± 0.23 L before and 2.92 ± 0.22 L after challenge), indicating hyperinflation; however, the decrease in IC did not correlate with the increase in RL or the decrease in FEV₁. There was also no correlation between the increase in RL and the decrease in FEV₁. There was a significant fall in FEV₁ postchallenge (mean, 29 ± 14%).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of this study show that in healthy subjects, a plateau develops on the methacholine dose versus RL curve that is paralleled by a plateau on the dose versus PEL_mean curve. The increase in RL did not correlate with the decrease in FEV₁ nor the degree of hyperinflation measured by the change in IC at the end of the challenge.

Plateaus on dose-response curves were first described for FEV₁ by Woolcock and coworkers (2) in 1984 and for flows on partial flow-volume curves by Sterk and coworkers (12) in 1985. A plateau is characterized by mild and limited airway narrowing, despite progressively higher doses of agonist. Woolcock and coworkers found that nonasthmatic subjects tended to have plateaus, whereas they were absent in most asthmatic subjects, especially those who had more severe disease. The observation that plateaus develop on the dose-response curve in normal subjects has stimulated a whole new paradigm in our understanding of airway hyperresponsiveness in airway disease. Prior to these observations, the concept was that airway hyperresponsiveness was related to factors that increased the sensitivity of airway smooth muscle (ASM) to contractile stimuli, i.e., there would be more airway narrowing at lower concentrations of agonist in hyperresponsive persons as compared with normal subjects. The assumed corollary of this paradigm was that normal subjects would develop a similar degree of airway narrowing as subjects who had hyperresponsiveness, if given a sufficiently high dose of agonist. The observation that there are major differences in the maximal achievable airway narrowing between normal and hyperresponsive subjects raised the possibility of different pathophysiologic mechanisms and has led researchers to hypothesize that factors determining maximal achievable narrowing are more important than those determining sensitivity (13, 14).

The assumption has been that the plateau on the dose- response curve is related to maximal ASM activation (3). Thus, the important difference between asthmatic and normal airway function must be a failure of the factors that normally limit or counterbalance maximally activated ASM shortening. However, it is conceivable that the plateau on dose-response curves to contractile agents in normal subjects may not be a static balance between maximal muscle activation and counterbalancing factors, of which the load on ASM appears to be the most important (6, 15). It is possible that the plateau represents a dynamic balance between increasing ASM activation and an increasing load applied to it by increased lung elastic recoil secondary to hyperinflation and/or parenchymal stiffening.

We used RL to measure airway narrowing. Although in animals the tissue component of pulmonary resistance (RTISS) can be significant; in normal humans, RTISS is substantially less and the proportion of RL that is related to tissue resistance becomes even less during bronchoconstriction (16). However, it is likely that a part of the increase in RL we observed is due to a stiffening of the lung parenchyma, especially since increasing lung volume increases RTISS. In fact, it has been suggested that increasing RTISS during bronchoconstriction is one of the factors that limit airway narrowing (7).

We hypothesized that if the plateau was due to a dynamic balance between increasing ASM activation and increasing load, we would have seen a dissociation between the dose- response curves for RL and PEL_mean at the plateau. Increasing ASM activation would be effectively counterbalanced by increasing PEL_mean, which would prevent the airways from narrowing further. This would mean that RL remained unchanged but PEL_mean would continue to increase. However, the results of this study show that PEL_mean changes in parallel with the changes in RL, suggesting that RL is not being maintained by increasing load caused by hyperinflation or tissue stiffening.

In addition to the static load reflected by mean PEL, tidal breathing imparts a dynamic load to ASM, which could be partly responsible for the dose-response plateau and its height. These oscillating forces reflected in the swings in PL during tidal breathing are able to reduce the force generation by ASM, thereby attenuating airway narrowing. The degree of attenuation is dependent on the magnitude and the frequency of the oscillating force (17).

Our results also showed a substantial decrease (mean, 29 ± 14%) in FEV₁ after challenge, a change that is large for normal subjects. This large decrease could be explained by the fact that throughout the challenge RL was measured, but the subjects did not take a deep breath until the forced expiratory maneuvers at the end of the challenge. It has been shown that inhibition of deep inspiration enhances the decrease in FEV₁ in response to methacholine in normal subjects (18, 19). In a previous study we found a 27 ± 15% decrease in FEV₁ at the end of a methacholine dose-response curve during which deep inspiration was prohibited. In those same subjects the fall in FEV₁ when forced expiratory maneuvers were performed throughout the challenge was only 14 ± 8%.

Changes in RL did not correlate with any changes in FEV₁. Similar disparities have been found in previous studies (20) in which normal subjects underwent methacholine challenge. The use of RL in our study, rather than FEV₁, was advantageous in that greater airway narrowing could occur and the effects of differences in relative hysteresis between the airways and parenchyma were avoided.

In summary, the results of this study have shown that during induced bronchoconstriction in normal subjects, there is a static balance between pulmonary resistance and pulmonary elastic recoil pressure consistent with the theory that there is maximal ASM activation at the plateau.