INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Peak expiratory flow (PEF) is a useful measure of pulmonary

function that is commonly used to monitor the course of airway obstruction and its response to therapy. Recently, studies in both normal volunteers (1, 2) and patients (3) have reported that measurements of PEF may show substantial variability that depends upon the pattern of inspiratory maneuver preceding forceful expiration. Accordingly, PEFs are greater when inspiration to TLC is performed rapidly and followed immediately by forced expiration; in contrast, slow inspiration with a pause of few seconds at TLC invariably results in lower PEF values (1).

The mechanism underlying this time dependency of PEF remains elusive. D'Angelo and coworkers (1, 3, 4) postulated that decrease in the effective lung elastic recoil caused by stress relaxation of respiratory tissues during breathholding primarily accounts for the PEF variability in normal subjects and that time-constant inequality within the lungs may be an additional factor in patients with chronic obstructive pulmonary disease (COPD). However, in two recent studies the lung elastic recoil did not entirely account for differences in PEF in normal subjects (2), and time constant inhomogeneity was not a contributing factor in patients with COPD (4).

In this study we reexamined the issue of PEF variability in relation to inspiratory maneuver preceding forceful expiration in a group of normal volunteers. To the extent that PEF is determined by effort (6), we reasoned that differences in expiratory muscle activation and pressure generation may contribute considerably to differences in maximal expiratory flows induced by the two inspiratory maneuvers.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ten healthy laboratory volunteers (mean age, 34 ± 6 yr) participated in the study. The subjects were familiar with respiratory maneuvers, and all but three were naive to the purpose of the study. No subject had a history of pulmonary disease; there were four smokers, but they all had normal spirometry. The study was approved by the Ethics Review Board.

All experiments were performed with the subjects in the seated position. Esophageal pressure was measured with an esophageal balloon inserted into the lower part of esophagus according to standard techniques (7). The catheter was connected to a pressure transducer (MP-45, ± 250 mm Hg; Validyne Corp., Northridge, CA) calibrated to 300 cm H₂O. To prevent balloon collapse during forced expiration, the balloon was filled with 2 ml of air (8). Inspiratory flow was measured with a heated screen pneumotachograph (range, 0 to 20 L/s; E. Jaeger, Würzburg, Germany) and a differential pressure transducer (MP-45, ± 2 cm H₂O; Validyne). Inspired volume was obtained by integrating (Gould integrator; Gould Instruments, Cleveland, OH) the flow signal. The electromyogram (EMG) of the abdominal and rib cage muscles was recorded by means of two pairs of surface electrodes. The abdominal electrodes were placed over the right side of the abdomen, along the right anterior axillary line, midway between the costal arch and the ileac crest, and the rib cage electrodes were placed at the level of the eight to tenth right intercostal space along the midaxillary line. EMGs were amplified (Nihon-Kohden Co., Tokyo, Japan) and filtered using a band-pass between 20 Hz and 1 KHz. Raw EMG signals were rectified and integrated by a moving averager (MA-821 RSP; CWE, Inc., Ardmore, PA) using 200-ms averaging time. Esophageal pressure, flow, volume, and EMG signals were displayed on a monitor (Gould V1000 video display). All signals were digitized in real time, recorded on a strip chart recorder (Gould ES 1000), and stored on computer (Wyse 486) for later analysis.

Protocol

Initially, all subjects practiced both maneuvers. For the fast (F) maneuver, the subjects inspired rapidly to TLC and immediately performed a forceful expiration to residual volume (RV). For the slow (S) maneuver, the subjects breathed slowly to TLC, held their breaths at TLC for 4 to 5 s, and then performed a forceful expiration to RV. During maximal efforts care was taken to keep the neck at a neutral position in order to decrease flow variability related to tracheal collapsibility (9). Both maneuvers were initiated from relaxation volume and performed randomly. Each subject performed approximately 18 to 20 runs per maneuver over an interval of about 3 h. Approximately 2 to 3 min of rest were provided between each maneuver.

Data Analysis

The digitized data were analyzed using the Labdat software program. For a given maneuver, only efforts that were within 5% for vital capacity (VC) and 10% for PEF were accepted for analysis. Negative esophageal pressure measured just prior to forceful expiration was taken to reflect elastic recoil pressure of the lung (PesL), and peak positive expiratory pressure (Pexp) was taken to represent muscular effort (Figure 1). Rate of rise of Pexp (dPexp/dt) was assessed by differentiating the pressure waveform. Integrated EMG signals were analyzed for peak EMG activity and rate of rise of EMG activity, obtained by dividing peak EMG activity by the time-to-peak EMG activity. Data are reported as means ± SE. All statistical analyses were performed with the SPSS statistical package (Norusis/SPSS, Inc., Chicago, IL) on a power Macintosh. Within- and between- subject intermaneuver differences in a given parameter were assessed by the t test. A repeated-measures analysis of variance (ANOVA) was used to compare intermaneuver differences in mean expiratory pressure or flow measured over a range of lung volumes. In addition, analysis of covariance (ANCOVA) was used to assess whether the maneuver effect on PEF was influenced by the lung elastic recoil and/or by specific effort indices. A probability value of 0.05 or smaller was considered significant.

View larger version (15K): [in this window] [in a new window]   Figure 1.   A representative tracing obtained during fast (A) and slow (B) maneuvers in one subject. Vertical dotted line passes through beginning of expiration. Arrow shows EMG activity of abdominal muscles during inspiration. Pes = esophageal pressure; PesL = lung elastic recoil pressure; EMGrc and EMGabd = integrated EMG activity of rib cage and abdominal muscles, respectively; au = arbitrary units.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Complete EMG data were available in eight subjects. Typical maneuvers from which the variables were obtained are illustrated in Figure 1. A total of 210 runs (104 F, 100 S) met our acceptance criteria and were analyzed for differences between the two maneuvers (Table 1). Similar results were obtained when, instead of all accepted runs, individual data were averaged and mean values of each variable were compared between the two maneuvers (data not shown). Overall, PEF values were greater with F than with S maneuver (p < 0.05). There were no differences in FVC, FEV₁, FEV_0.25, FEV_0.50, or FEV_0.75 between the two maneuvers (Table 1). Analysis of individual data showed that F yielded higher PEF values in six of the 10 subjects (p < 0.05), whereas in the remaining four the two maneuvers produced comparable PEF (p > 0.05).

The mean PesL was lower (more negative) with F than with S; both mean Pexp and mean dPexp/dt were similarly greater with F than with S (p < 0.001) (Table 1). Analysis of individual data showed that Pexp and dPexp/dt were significantly greater with F in eight and nine of the 10 subjects, respectively. Mean expiratory esophageal pressures measured at 90, 80, 70, 60, and 50% of VC were greater with F than with S (p < 0.05) (Figure 2). There were no differences in mean expiratory flows at 90, 80, 70, 60, and 50% of VC between the two maneuvers (Figure 2).

View larger version (11K): [in this window] [in a new window]   Figure 2.   Pes-volume (A) and flow-volume (B) curves obtained with slow (closed circles) or with fast (open circles) maneuvers in all subjects. Data are means ± SE. *p < 0.05 by repeated-measures ANOVA for pressure data.

The peak expiratory EMG activity of abdominal muscles and rate of rise of abdominal EMG were greater with F than with S (Table 1); there was no difference in mean peak expiratory EMG activity of rib cage muscles between the two maneuvers. The aggregate expiratory activity of abdominal and rib cage muscles was also greater with F than with S maneuvers (Table 1). In addition, mean peak inspiratory EMG activity of abdominal muscles (EMGabd_insp) was greater with F than with S.

To examine what determines PEF differences between the two maneuvers, the data of each of the six subjects with PEF changes were analyzed separately using an ANCOVA model. This analysis showed that PesL, dPexp/dt, or Pexp independently accounted for the maneuver effect on PEF in four, five, and four of six subjects, respectively (p increased to > 0.05 in all). Overall, PesL combined with Pexp or dPexp/dt explained the maneuver effects on PEF in five of the six subjects; in one subject none of theses indices, alone or in combination, reduced the maneuver effect on PEF.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main findings of the study are as follows. (1) The fast inspiratory maneuver produced greater peak expiratory flows than did the slow maneuver in some, but not in all, subjects. (2) The fast maneuver was associated with high lung elastic recoil and maximal expiratory pressure in most of the subjects. (3) Differences in peak expiratory flows between the two maneuvers appear to be related to differences in both the effective lung elastic recoil and the expiratory muscle force.

Critique of Methods

Several methodologic issues related to the present study require comment. Lung volume was measured by means of a pneumotachograph rather than a body plethysmograph, and thus our lung volume measurements were underestimated because of gas compression (10). This error was probably greater for the fast maneuver, which produced greater expiratory pressures than the slow maneuver. No effort was made to correct for gas compression. We estimate that lung volume differences related to gas compression were small (1, 4, 11), and they most likely did not affect our results.

Measurement of inspiratory and expiratory pressures was done by means of a single esophageal balloon filled with a relatively large volume of air. Although this accurately measured expiratory pressure (8), it may have underestimated the lung elastic recoil pressure (7). This underestimation, however, influenced both maneuvers in a systematic fashion (7) and most likely did not bias our data.

Our Pexp measurements included not only pressure related to expiratory muscle contraction but also pressure produced by the elastic recoil of the chest wall, which was not measured. Because of the viscoelastic properties of the respiratory system, the fast manuever probably caused a greater augmentation of the expiratory muscle pressure than did the slow maneuver. However, such differences in elastic recoil of the chest wall were small (3, 12) and cannot account for differences in Pexp produced by the two maneuvers.

Inspiratory Maneuver Effects on Peak Expiratory Flow

In comparison with the slow maneuver, the fast maneuver produced PEF values that were, on average, 7% greater. This figure is identical to that reported by Wanger and coworkers (2) in normal subjects and considerably smaller than that reported by D'Angelo and coworkers (1, 3) in normal subjects (16%) and in patients with COPD (30%) or by Braggion and coworkers (5) in patients with cystic fibrosis (26%). This may be due in part to the fact that not all of our subjects produced higher flows with the fast maneuver. In this respect, our findings are in agreement with those of Wellman and coworkers (13) who similarly found that the fast inspiration produces higher flows in some but not in all subjects.

Our study confirms previous observations (1, 13) and further extends them by showing that differences in PEF between the two maneuvers appear to be related to differences in lung elastic recoil and expiratory muscle activation. On the basis of the viscoelastic behavior of the respiratory system, D'Angelo and coworkers (1, 3) postulated that changes in the effective elastic recoil of the lung and chest wall could readily explain PEF differences between the two maneuvers. Thus, fast inspiration, by increasing the elastic recoil more so than slow inspiration, is expected to increase PEF; in contrast, interposition of a breathhold between inspiration and expiration would have an opposite effect on the effective elastic recoil of the lung and chest wall and hence diminish PEF. However, in a recent study in normal subjects (2) lung elastic recoil did not entirely account for the effect of different inspiratory maneuvers on PEF. Furthermore, the extent to which differences in effort actually contribute to differences in PEF has not been fully studied. In normal subjects, D'Angelo and coworkers (1) provided evidence that the fast inspiratory maneuver was associated with a greater rate of rise of expiratory muscle EMG activity than the slow inspiratory maneuver. In patients with COPD, however, no such difference in EMG activity was found (3). In the absence of expiratory pressure measurements, D'Angelo and coworkers (1, 3, 4) could not assess the contribution of muscular effort to PEF nor explain differences in the rate of rise of EMG with the fast maneuver.

In our study, the fast maneuver produced a greater activation of the expiratory muscles and higher expiratory pressures than did the slow maneuver. Indeed, this augmentation in muscular effort was the most consistent intermaneuver difference, being significant in all but one subject. Our analysis, however, suggests that differences in PEF between the two maneuvers were largely related to differences in both lung elastic recoil and muscular effort, with neither variable actually being more influential than the other. A different analysis is shown in Figure 3. Here, average intermaneuver differences in PesL, Pexp, or dPexp/dt were calculated for each subject and then compared between subjects with (Group 1, n = 6) or without PEF changes with F (Group 2, n = 4). On average, subjects in Group 1 had greater intermaneuver differences in PesL, dPexp/dt, and Pexp values. Again this analysis suggests that intermaneuver differences in PesL and effort indices are actually what differentiate subjects with PEF changes from those without.

View larger version (30K): [in this window] [in a new window]   Figure 3.   Intermaneuver differences in PesL, dPexp/dt, and Pexp in subjects with (group 1, n = 6) or without (group 2, n = 4) PEF changes with the fast maneuver. Note that the intermaneuver differences in PesL, dPexp/dt or Pexp are in general greater in group 1 than in group 2. Data are in mean ± SE. PesL = lung elastic recoil; Pexp = maximal expiratory esophageal pressure; dPexp/dt = maximal rate of rise of Pexp; F = fast maneuver; S = slow maneuver.

Inspiratory Maneuvers and Expiratory Muscle Activation

Our pressure and EMG data clearly show that the fast maneuver was associated with greater activation and force production by the expiratory muscles. The exact mechanism underlying this effect is unknown. We speculate that this is probably related to a specific inspiratory-expiratory muscle interaction analogous to those described in skeletal muscles for a system of agonists-antagonists. Therefore, on the basis of similar interactions in peripheral muscles we offer two potential explanations.

The first (14) refers to the force potentiation of an agonist muscle (expiratory muscles in this instance) when it is immediately preceded by a conditioning contraction of the antagonist muscle (inspiratory muscles in our case). Such conditioning activity may serve to maximize the level of segmental excitability and enhance the neural drive to the agonist muscle (14). Although some contradictory evidence has been presented (14, 15), force potentiation related to antagonist conditioning contraction appears to be an accepted principle in skeletal muscles (16, 17). In a recent study involving knee muscles, Grabiner (14) suggested that there may be a direct relationship between the intensity of the antagonist conditioning contraction and the extent of agonist activation enhancement and force potentiation. In examining this interaction in the knee muscles, Grabiner (14) found that a conditioning contraction of the antagonist muscles generally increased agonist muscle activation as shown by greater EMG activity and greater rate of rise of force for the agonist muscles.

A second possibility may be that the enhanced contraction of the expiratory muscles during fast maneuvers was related to their preceding eccentric contraction during inspiration. Studies (18) have indicated that the concentric force output of a skeletal muscle is generally greater when it is immediately preceded by an eccentric muscle action than in a pure concentric action. This synergistic behavior is known as "rebound or countermovement" effect and has been used in skeletal muscle training (22). In our study, the EMG data clearly showed that the expiratory muscles, particularly the abdominal muscles, were more active during inspiration with the fast than with the slow maneuver. Thus, the greater force production with the fast maneuver can be attributed to the fact that the fast maneuver elicited a greater eccentric contraction of the abdominal muscles, which was immediately followed by a concentric contraction of the same muscles.

Abdominal muscle contraction during inspiratory efforts, especially at the end of a full inspiration, has previously been described (23). Early studies (23, 24) raised the possibility that this abdominal muscle activation could be regarded as antagonistic in the sense that the abdominal muscles antagonize the action of the inspiratory muscles and therefore contribute to set the upper lung volume extreme. In this study, the greater abdominal muscle activation with fast inspiration can be readily explained in the context of the coactivation described for antagonistic muscles during rapid-goal-directed movements (27, 28). Indeed, as in skeletal muscles, abdominal muscle coactivation increased with fast inspiration, suggesting a similar dependence on muscle velocity (27). Although either mechanism outlined above provides an attractive explanation of our data, further studies are required to fully explain differences in Pexp induced by the two maneuvers.

Implications

The need to develop guidelines to standardize the inspiratory maneuver in measurements of expiratory flow has already been emphasized (1). Apart from this, our data may have implications for training respiratory muscles. In this context, combined respiratory maneuvers may be included in training regimens in order to achieve greater activation of the respiratory muscles. In fact, in a study (31) in which respiratory muscle training was carried out at different lung volumes, an analogous agonist-antagonist interaction might have accounted for the greater gain of respiratory muscle strength with training at RV than at relaxation volume.

In conclusion, our study has confirmed recent observations that the pattern of inspiration preceding forceful expiration influences the magnitude of peak expiratory flows and that fast inspirations without pause produce higher flows than do slow inspirations with a few seconds pause at TLC. The greater expiratory flows appear to be related to both greater effective lung elastic recoil and greater activation of the expiratory muscles induced by fast inspirations. Although the former is related to presence of stress relaxation in the respiratory tissues, the latter suggests a unique agonist-antagonist interaction analogous to those described for skeletal muscles.