Spirometric Correlates of Improvement in Exercise Performance after Anticholinergic Therapy in Chronic Obstructive Pulmonary Disease

Transcription

1 Spirometric Correlates of Improvement in Exercise Performance after Anticholinergic Therapy in Chronic Obstructive Pulmonary Disease DENIS E. O DONNELL, MIU LAM, and KATHERINE A. WEBB Respiratory Investigation Unit, Departments of Medicine and Community Health and Epidemiology, Queen s University, Kingston, Ontario, Canada We wished to determine which resting spirometric parameters best reflect improvements in exercise tolerance and exertional dyspnea in response to acute high-dose anticholinergic therapy in advanced COPD. We studied 29 patients with stable COPD (FEV % predicted [%pred]; mean SEM) and moderate to severe chronic dyspnea. In a double-blind placebo-controlled cross-over study, patients performed spirometry and symptom-limited constant-load cycle exercise before and 1 h after receiving 500 g of nebulized ipratropium bromide (IB) or saline placebo. There were no significant changes in spirometry, exercise endurance, or exertional dyspnea after receiving placebo. In response to IB (n 58): FEV 1, FVC, and inspiratory capacity (IC) increased by 7 1%pred, 10 1%pred, and 14 2%pred, respectively (p 0.001), with no change in the FEV 1 /FVC ratio. After receiving IB, exercise endurance time (Tlim) increased by 32 9% (p 0.001) and slopes of Borg dyspnea ratings over time decreased by 11 6% (p 0.05). Percent change (% ) in Tlim correlated best with IC%pred (p 0.020) and change in inspiratory reserve volume ( TLC%pred) (p 0.014), but not with FVC%pred, PEFR%pred, or FEV 1 %pred. Change in Borg dyspnea ratings at isotime near end exercise also correlated with IC%pred (p 0.04), but not with any other resting parameter. Changes in spirometric measurements are generally poor predictors of clinical improvement in response to bronchodilators in COPD. Of the available parameters, increased IC, which is an index of reduced resting lung hyperinflation, best reflected the improvements in exercise endurance and dyspnea after IB. IC should be used in conjunction with FEV 1 when evaluating therapeutic responses in COPD. O Donnell DE, Lam M, Webb KA. Spirometric correlates of improvement in exercise performance after anticholinergic therapy in chronic obstructive pulmonary disease. AM J RESPIR CRIT CARE MED 1999;160: In patients with advanced chronic obstructive pulmonary disease (COPD), improvement in lung mechanics, alleviation of dyspnea, and increased activity levels are desirable therapeutic goals. In this population, studies that are designed to evaluate the impact of interventions, such as bronchodilator therapy, increasingly incorporate these important clinical outcome measures. We have shown that measurements of dynamic lung hyperinflation, exercise endurance, and Borg ratings of exertional dyspnea intensity can be used to evaluate therapeutic responses reliably, being both reproducible and responsive (1). Relief of exertional dyspnea following both 2 -agonist and anticholinergic therapy has been shown to correlate well with reduction of dynamic lung hyperinflation, as measured by serial inspiratory capacity (IC) measurements during exercise, in (Received in original form January 12, 1999 and in revised form March 17, 1999) Supported by Boehringer Ingelheim (Canada) Limited and by the Ontario Ministry of Health. Correspondence and requests for reprints should be addressed to Dr. Denis O Donnell, Richardson House, 102 Stuart Street, Kingston General Hospital, Kingston, ON, K7L 2V7 Canada. odonnell@post.queensu.ca Am J Respir Crit Care Med Vol 160. pp , 1999 Internet address: patients with advanced COPD (1, 2). Belman and coworkers (2) showed that reduced dynamic lung hyperinflation was related to improved neuromechanical coupling which, in turn, likely contributed to dyspnea relief (3). Similarly, Chrystyn and colleagues (4) have shown that improved exercise endurance after increasing doses of theophylline was related to reduction of resting plethysmographically determined thoracic gas volume. However, because body plethysmography, cardiopulmonary exercise testing facilities, and the ability to measure dynamic changes in lung hyperinflation may not be readily available to many clinicians managing COPD, we now wish to extend the analysis of our previous study (1) to consider the value of resting spirometric data in evaluating clinical responses. We sought to answer a number of questions with this study. First, are changes in resting spirometric parameters after acute administration of anticholinergic therapy associated with improvements in exertional dyspnea and exercise performance in stable advanced COPD? Second, do resting spirometric indices that indirectly reflect reduced gas trapping, such as increased IC and vital capacity (VC), correlate more strongly with improvements in clinical response (i.e., increased exercise endurance and decreased exertional dyspnea) than do measurements of expiratory flow rate per se? If so, what mag-

2 O Donnell, Lam, and Webb: Improvement in Exercise after Bronchodilation in COPD 543 nitude of change in IC is associated with a significant improvement in exercise endurance and dyspnea? Finally, do combined changes in lung volume and expiratory flow better predict a clinical response than either parameter alone, and how are these parameters interrelated? In this regard, how do changes in IC compare with traditional indices of reversibility (i.e., criteria for significant FEV 1.0 changes recommended by American Thoracic Society [ATS] [5] and European Respiratory Society [ERS] [6] guidelines) in terms of predicting a significant improvement in exercise performance? Using a placebo-controlled study design, we determined the patterns of acute spirometric responses and their relationship to changes in exercise endurance and dyspnea after high-dose anticholinergic therapy in 29 patients with stable advanced COPD. METHODS Subjects The study included 29 patients with stable COPD who satisfied the following criteria: (1) moderate to severe COPD (FEV 1 60% predicted) with a clinical course consistent with chronic bronchitis and/ or emphysema and a long history of cigarette smoking; (2) moderate to severe chronic breathlessness (modified Baseline Dyspnea Index [BDI] 6) (7); (3) age 55 yr or older; and (4) clinically stable as defined by no changes in medication dosage or frequency of administration with no exacerbations or hospital admissions in the preceding 6 wk. Exclusions included a history of asthma, atopy, or nasal polyps; other active lung disease or other significant disease that could contribute to dyspnea or exercise limitation; or oxygen desaturation to less than 80% during exercise on room air. Study Design As described in our previous publication (1), the study had a randomized, double-blind, two-period cross-over design. After hospital/university research ethics approval was obtained, subjects gave informed consent and entered the study on a staggered basis. During an initial screening visit, patients were thoroughly familiarized with all procedures and symptom rating scales, and underwent medical history, pulmonary function testing, dyspnea evaluation, and symptom-limited incremental cycle exercise testing. Subjects were then randomized to receive either nebulized ipratropium bromide (IB, 500 g) or placebo three times a day in unit dose vials for a 3-wk period. Each subject was crossed over to the alternate treatment for an additional 3-wk period, after a 2- to 7-d washout period. Four experimental visits per subject were conducted at the beginning and end of each 3-wk period. During these visits, pulmonary function testing, dyspnea evaluation, and constant-load cycle exercise testing were performed immediately before and 1 h after the treatment corresponding to that being given during the coinciding 3-wk period. Treatments were administered by inhalation via face mask from a Hudson updraft nebulizer unit (model No. 1712; Hudson Respiratory Care, Temecula, CA) at a flow rate of 8 L/min over a period of approximately 15 min. The total volume of nebulized solution was 5 ml for all patients: ipratropium bromide (500 g) was compared with a placebo, which consisted of sterile 0.9% sodium chloride solution. Patients were required to take more than 50% of the study medication during each treatment period. Treatment compliance was assessed at the end of each period by counting all returned unit dose vials (empty and full), with verification against the use recorded in the patient daily diary. Concomitant respiratory medications permitted for the duration of the study included regularly taken inhaled steroids and bronchodilators other than anticholinergic medications. Use of medications remained stable throughout the study treatment periods. Patients requiring additional medication or changes in medication for more than 2 d were withdrawn from the study. Before each visit, patients discontinued use of inhaled 2 -agonists, anticholinergics, and theophyllines for at least 4, 12, and 24 h, respectively, before testing. Subjects avoided caffeine and heavy meals for at least 4 h before testing, and avoided alcohol and major physical exertion entirely on the day of each visit. All visits for each subject were conducted at the same time of day. Procedures Procedures for carrying out pulmonary function testing, exercise testing, and dyspnea evaluation have been described in detail in our previous publication (1). Pulmonary function. Routine spirometry was performed before exercise testing in accordance with recommended standards (8). The predicted normal values for spirometry, maximal expiratory flow rates, and lung volumes were those of Morris and coworkers (9), Knudson and colleagues (10), and Goldman and Becklake (11), respectively. Maximum inspiratory mouth pressure (PI max mo) from the residual volume (RV) was measured with a standard mouthpiece and a direct-reading dial pressure gauge (Magnehelic; Dwyer Instruments, Michigan City, IN), and was compared with the predicted normal values of Black and Hyatt (12). Chronic dyspnea evaluation. The modified baseline dyspnea index (BDI) was used to assess chronic activity-related breathlessness (7, 13). The same unbiased observer, blinded to the treatment received by each patient, was used to assess the BDI at study entry and at the beginning of each 3-wk study period. Exercise testing. In experimental visits for each subject, symptomlimited endurance cycle exercise tests were performed at the same constant work rate equal to approximately 50 60% of the maximum work rate achieved during the incremental cycle exercise test performed at screening. During exercise, standard cardiopulmonary measurements were collected and compared with the predicted normal values of Jones (14). Using the Borg Scale (15), subjects rated the intensity of their perceived dyspnea at rest, every minute throughout exercise, and at peak exercise. To assess operational lung volumes, IC maneuvers were performed at rest, every 3 min during exercise, and at peak exercise. Techniques for performing and accepting IC measurements have been previously described (1, 16). Statistical Analysis The study was originally designed to examine both acute and chronic responses to bronchodilator therapy in severe COPD. The focus of this part of the study was to determine the acute responses to IB or placebo; specifically, to examine the physiologic correlates of improvement in exercise endurance time and exertional dyspnea. Baseline demographics, lung function, and exercise-related measurements were summarized using means SEM. Standardized exercise time (STD) was equal to the time of the highest equivalent amount of work or time completed in all experimental exercise tests for each subject, i.e., the time of the shortest exercise test rounded down to the nearest minute. Analysis of variance incorporating the repeated measures cross-over design of the study was applied to assess the responsiveness of exercise endurance, dyspnea, and lung function measurements to treatment. The analysis of variance models (17) included factors and terms to account for subject differences, drug effects, presentation order, carryover, and period effects. Possible carryover and period effects were first considered and assessed in the models; if these effects were not significant, they were excluded from the models, and the significance of the direct treatment effects was tested accordingly. To examine the relationship between acute changes in exercise endurance and changes in resting measurements of lung function in response to therapy, regression analysis was applied. The percent change in exercise endurance time (% Tlim) from pre- to postdose was considered as the dependent variable, and independent variables included changes in resting lung function expressed as percentages of predicted normal values (FEV 1.0, FVC, FEV 1.0 /FVC, peak expiratory flow rate [PEFR], 75% of maximal flow [ V max 75 ], V max 50, V max 25, IC, PI max mo) as well as changes in standardized exercise measurements of lung hyperinflation (IC, inspiratory reserve volume [IRV]), ventilation (minute ventilation [ V E], V E/ V O 2, V E/ V CO 2 ), breathing pattern (respiratory frequency [f], tidal volume [VT], VT/IC), gas exchange ( V CO 2 / V O 2, Sa O2 ), and cardiovascular function (heart rate, blood pressure). The study had a repeated measures cross-over design, which enabled us to investigate both cross-sectional (between subjects) and longitudinal (within subjects) relationships between

3 544 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL each of the independent variables and the dependent variable. These two types of relationships were tested and compared through regression models proposed by Diggle and coworkers (18). If the relationships were similar, they were combined as a common relationship and a regression model was used to describe this common relationship. Owing to the treatment effect, the common relationship between an independent and the dependent variables may be different under the two treatments (IB and placebo). Therefore, different regression models were set up to incorporate the treatment effect and the possible different relationships between the independent and dependent variables under the two treatments. Using these regression models, we attempted to identify the independent variables that best reflected the change in Tlim after high-dose IB. Dependency between repeated measures owing to the design of the study was taken into account in all analyses. Finally, changes in standardized ratings of exertional dyspnea ( Borg std ) were considered as another dependent variable that was similarly analyzed. RESULTS Subjects Of the 36 patients randomized into the study, there were 29 evaluable subjects (Table 1). Seven patients were prematurely withdrawn from the study because of hospital admission with respiratory tract infection (n 2), hospital admission with congestive heart failure (n 1), repeated attaches of respiratory panic (n 1), increased breathlessness requiring a change in treatment for more than 2 d (n 1), facial skin irritation from the nebulizer (n 1), and noncompliance due to lack of time. One of the 29 patients experienced an acute exacerbation of COPD requiring additional medication 7 d before the last visit and did not complete the exercise testing during this visit. The two treatment sequence groups were comparable at study entry for demographics, lung function, and exercise capacity: n 13 were in the treatment sequence that received IB for the first two experimental visits and placebo for the last two visits, n 16 received treatments in the reverse order. Treatment compliance over the two 3-wk periods was excellent, at 95 1%. treatment effects. Cross-sectional and longitudinal relationships were found to be similar for all spirometric variables, thus they were combined as a common relationship. Pre- to postdose changes in variables at the beginning of each 3-wk period were not significantly different from those at the end of each period. Significant drug effects were found as changes in resting lung function measurements after IB compared with placebo (Figure 1). Assuming that total lung capacity (TLC) and breathing pattern remained constant (19, 20), maximal expiratory flows increased significantly at any given volume after IB (Figure 2), while FVC, IC, and IRV increased significantly (p each) by a mean of , , and L, respectively (Figure 2). If TLC was reduced after IB, this would lead to an underestimation of the improvement in volume-matched flows over the operating range. The FEV 1.0 /FVC ratio did not change in response to therapy and FEV 1.0 correlated strongly with FVC (p ) and IC (p ), suggesting that changes in flow were directly related to lung volume changes. Symptom-limited exercise endurance time (Tlim), a highly reliable measurement (1), improved significantly after IB ( min or 32 9%) compared with placebo ( min Treatment Effects Stability of baseline spirometry (i.e., FEV 1.0 and FVC) and the BDI, and the absence of various carryover and period effects, were verified before establishing the significance of direct TABLE 1 SUBJECT CHARACTERISTICS* Parameter Value Male:female 22:7 Age, yr 67 1 Height, cm Weight, kg Body mass index, kg/m Smoking history, pack-years 54 7 Modified Baseline Dyspnea Index ( severe ) Pulmonary function (% predicted) FEV 1, L (40 2) FVC, L (59 3) FEV 1 /FVC, % 47 2 (69 3) PI max mo at RV, cm H 2 O 60 9 (65 9) Symptom-limited maximal exercise (% predicted) Work rate, W 54 4 (39 3) Heart rate, beats/min (64 2) V O 2, L/min (61 4) V E, L/min (72 4) Sa O2, % 93 1 * Values presented are means SEM. Figure 1. Changes in resting lung function in response to IB and placebo are shown, expressed as a percentage of the predicted value (top) and as the percent change from the pretreatment value (bottom). See text for definition of abbreviations. *p 0.01, # p 0.05, significant change pre- to postdose. MIP maximum inspiratory mouth pressure (PI max mo).

4 O Donnell, Lam, and Webb: Improvement in Exercise after Bronchodilation in COPD 545 or 0 3%, p ). Exertional dyspnea also improved significantly after IB compared with placebo (Figure 3): Borg time slopes fell by 11 6% (p 0.05) and Borg std fell by units (p 0.01) after IB administration in the laboratory. The only measured physiological parameters that changed significantly after IB were those related to operational lung volumes and breathing pattern. Significant drug effects were noted as changes in IC and IRV evaluated both at rest and during exercise (Figure 4). Although breathing frequency (f) and VT did not change at rest, there was a shift toward a slower, deeper breathing pattern during exercise after IB (Figure 4): f decreased (p 0.05) and VT increased (p 0.05) significantly at a standardized time during exercise. Figure 2. Maximal expiratory flows increased from pre-ib (dotted line) to post-ib (solid line) in association with a decrease in endexpiratory lung volume, as reflected by a significant increase in IC (p 0.001). After IB, tidal volume (VT) was positioned at a lower operational lung volume and inspiratory reserve volume (IRV) was significantly increased (p 0.001). Pre- and postdose expiratory curves were constructed from measurements of PEFR, Vmax 75, Vmax 50, and Vmax 25 at their respective lung volumes: each of these flow measurements increased significantly (p 0.001) after IB. Hatched areas represent the flows available for the given VT. Correlates of Improvement In response to IB, the percent change in exercise endurance time (% Tlim) correlated significantly with changes in resting measurements of IC and IRV (Table 2); % Tlim also correlated strongly with the reduction in standardized Borg ratings of exertional dyspnea intensity (p 0.01). To increase the power of this analysis, the two repeated tests for each patient were included and this was accounted in the analysis of each variable. After accounting for drug and period effects, the relationship between % Tlim and IC remained significant (p 0.05). Finally, the strength of the relationship between % Tlim and IC was not improved by adding FEV 1.0 ; conversely, IC contributed an additional 7% to the variance in % Tlim after accounting for FEV 1.0. Relationships between % Tlim and changes in FEV 1 and IC (changes expressed as the percentage of the predicted value and as the percent change) are shown in Figure 5. Although Borg std correlated significantly with the resting IC in response to IB (Table 2), it correlated more strongly with concurrent exercise measurements of IC%pred (p 0.01) and VT/IC (p 0.01), and to some extent with f (p 0.07). Criteria for Reversibility and a Positive Clinical Response We categorized patients as bronchodilator responders and nonresponders according to an arbitrary change in IC of at least Figure 3. Borg ratings of exertional dyspnea during endurance exercise are shown before and after IB (left) and placebo (right). Slopes of dyspnea intensity over time during cycle exercise fell significantly after IB (*p 0.05) but not placebo.

5 546 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL Figure 4. Exercise responses. Operational lung volumes improved significantly after ipratropium bromide (IB) compared with placebo, i.e., IC and IRV were increased at rest and at any given exercise time. Bottom: Breathing frequency (f) is expressed relative to tidal volume (VT) during constant-load exercise: at a standardized exercise time after IB, f decreased and VT increased significantly (*p 0.05). Values shown are means SEM and include all predose (dashed lines) and postdose tests (solid lines) for IB (left, n 58) and placebo (right, n 57). 10% of the predicted value (based on the lower 95% confidence interval for the response to IB in this study); measurements were expressed as the percentage of predicted values to avoid the unwanted bias for a positive response that can occur when looking at percent change in patients with low baseline values. Using these IC reversibility criteria, 18 of 29 patients TABLE 2 REGRESSION COEFFICIENTS* BETWEEN CHANGES IN EXERCISE ENDURANCE TIME, EXERTIONAL DYSPNEA, AND RESTING SPIROMETRIC MEASUREMENTS IN RESPONSE TO HIGH- DOSE IPRATROPIUM BROMIDE, AFTER ACCOUNTING FOR REPEATED TESTS WITHIN PATIENTS % Tlim (n 58) Borg isotime (n 58) FEV 1, %pred 1.33 (p 0.275) 0.02 (p 0.446) FVC, %pred 1.65 (p 0.070) 0.02 (p 0.349) FEV 1 /FVC, % 0.26 (p 0.838) 0.01 (p 0.579) PEFR, %pred 2.08 (p 0.096) 0.02 (p 0.294) Vmax 75, %pred 0.28 (p 0.808) 0.01 (p 0.829) Vmax 50, %pred 0.14 (p 0.947) 0.03 (p 0.442) Vmax 25, %pred 0.28 (p 0.760) 0.01 (p 0.661) IC, %pred 1.18 (p 0.020) 0.02 (p 0.033) VT/IC, % 1.65 (p 0.088) 0.03 (p 0.079) IRV, %pred TLC 2.88 (p 0.014) 0.03 (p 0.123) Definition of abbreviations: IRV inspiratory reserve volume; PEFR peak expiratory flow rate. * p Values. p (62%) were classified as bronchodilator responders. ATS criteria, which require an increase in FEV 1 of 12% and 200 ml, classified 16 of 29 patients (55%) as responders. ERS criteria, which require a change in FEV 1 of at least 10% of predicted, selected only 13 of 29 patients (45%) as responders. In looking at bronchodilator response patterns, we found a significant increase (i.e., 10% of predicted) in both FEV 1 and IC in 9 of 29 patients (31%), an increase in IC but not FEV 1 in 9 of 29 patients (31%), an increase in FEV 1 but not IC in 4 of 29 patients (14%), and no response in 7 of 29 patients (24%). On the basis of these results and previous experience in our laboratory, we defined a positive clinical response in this COPD population as an increase in Tlim of at least 25% (i.e., approximately 2 min) and a decrease in Borg std of at least 10% (i.e., approximately 0.5 Borg units). By categorizing patients according to their IC response, we accurately predicted the clinical response in Tlim in 59% of cases. ATS and ERS criteria accurately predicted a clinical response for Tlim in 52 and 48% of patients, respectively. DISCUSSION The main findings of our study are as follows. First, changes in spirometric measurements are generally poor predictors of improved exercise performance and symptom relief after anticholinergic therapy in COPD. Second, of the available spirometric measurements, those that indirectly measure reduced lung hyperinflation (i.e., IC, IRV, and FVC) correlate better with reduced dyspnea and improved exercise endurance than

6 O Donnell, Lam, and Webb: Improvement in Exercise after Bronchodilation in COPD 547 Figure 5. Correlations between percent change in exercise endurance (Tlim) and changes in FEV 1 and IC (changes expressed as a percentage of the predicted value and as the percent change from pretreatment) in response to IB (n 58). Expressed either way, change in IC correlated significantly, albeit modestly, with percent change in Tlim (p 0.05), whereas change in FEV 1 did not correlate significantly. For presentation purposes, one outlier was removed from the graphs at a percent change in Tlim of 170%. do expiratory flow measurements. Third, an increase in IC of 10% predicted (i.e., approximately 0.3 L in this group) was associated with a significant ( 25%) improvement in exercise endurance time (Tlim). Fourth, in a proportion of patients, improvements in IC and exercise endurance occurred despite minimal changes in FEV 1.0. There is increasing evidence that dynamic lung hyperinflation during exercise, a consequence of expiratory flow limitation, contributes importantly to the intensity and quality of dyspnea and exercise intolerance in patients with COPD (3, 16). Thus, interventions that reduce dynamic lung hyperinflation, either pharmacologically (1 3) or surgically (21 23), have been shown to contribute importantly to exertional dyspnea relief in this population. It has also become increasingly clear that exclusive reliance on change in FEV 1.0 as the primary end point for bronchodilator therapy in COPD may underestimate significant clinical benefits in some patients (2, 24, 25). For this reason, we and others have advocated direct measurements of exercise endurance and symptom intensity to assess accurately therapeutic responses to bronchodilators. A reasonable question is, however, whether conventional simple spirometric measurements, which indirectly reflect a reduction in resting thoracic gas entrapment, provide complementary information to expiratory flow measurements in assessing clinical improvement after bronchodilator therapy. Our results support the contention that additional assessment of spirometric lung volumes may have important clinical utility with respect to bronchodilator assessment. Thus, we found a significant correlation between change in IC (expressed as a percentage of the predicted value) after acute IB therapy and improvement in exercise endurance time ( Tlim). Changes in IRV and FVC were also more strongly correlated with changes in dyspnea and Tlim than expiratory flow measurements. Resting IC (percent predicted) provides useful information with respect to the prevailing ventilatory mechanics in COPD. First, the IC provides an indirect measure of the elastic load on the inspiratory muscles; i.e., the smaller the IC, the more end-expiratory lung volume exceeds static FRC and, therefore, the greater the inspiratory threshold load and work of breathing (3, 26). Second, the IC (and not the VC) represents the operational limits for tidal volume expansion during exercise in COPD (i.e., the degree of mechanical volume restriction) (3, 26). Thus, in advanced COPD, VT/IC ratios are increased and IRV is reduced both at rest and during low levels of exercise when compared with those of age-matched controls (3, 16, 26). It follows that an increase in IC and IRV after bronchodilator therapy indirectly reflects reductions in mechanical restriction, elastic/threshold loading, and inspiratory pleural pressure swings, which should collectively translate into reduced dyspnea (3, 26). As a result of improved airway function in response to IB, patients could maintain the same resting ventilation at lower operational lung volumes than during placebo treatment. Increased resting IC and IRV and improved inspiratory muscle function (noted by the 12% increase in PI max mo at RV) meant that patients could maintain the same (or greater) exercise ventilation for a longer duration with a more efficient breathing pattern (increased VT, reduced f) and lower dynamic operational lung volumes (Figure 4). This, together with the prevention of further dynamic hyperinflation during exercise, must mean that muscle force-generating requirements are diminished relative to maximum and that neuromechanical coupling of the respiratory system is enhanced, with consequent reduction of dyspnea (3, 26). In this study, a mean improvement in IC of 14% predicted was associated with an improvement of exercise Tlim of 32%. It is noteworthy, however, that changes in resting IC measurements did not correlate as strongly with the changes in exertional dyspnea as dynamic IC measurements during exercise (1); the latter reflects the increase in dynamic end-expiratory lung volume and is the more relevant physiological measurement with respect to respiratory sensation and exercise intolerance. Change in FEV 1.0 (percent predicted) did not correlate significantly with improved exercise endurance and dyspnea relief after IB treatment. It is noteworthy that, on average, there was no significant change in the FEV 1.0 /FVC ratio after IB treatment, indicating that in many instances, improved expiratory flow rates reflected increased lung volume recruitment as a result of reduced air trapping. Thus, change in FEV 1.0 correlated strongly (p ) with the change in FVC and IC after IB treatment. In a multiple regression analysis, change in FEV 1.0 did not add to the change in IC in explaining improved Tlim after IB treatment; however, change in IC added 7% to the predictive value of FEV 1.0 in explaining this improvement. The lack of correlation between change in FEV 1.0 and change in Tlim or Borg dyspnea ratings was not unexpected and corroborates the results of other studies (24, 25). FEV 1.0 gives no information about the level of expiratory flow limitation, the shape of the expiratory flow volume curve (i.e., the maximal expiratory flows available) over the tidal volume operating range, or the extent of lung hyperinflation required to maximize tidal expiratory flows. Each of these parameters can vary greatly for a given FEV 1.0 (27). FEV 1.0 has been shown to correlate poorly with a variety of measures of exercise performance and dyspnea (13, 28). There was considerable variability in the spirometric response patterns after IB treatment among study subjects. It is noteworthy that 31% of the overall sample showed little or no change in FEV 1.0 ( 10% predicted) but had an improved IC

7 548 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL ( 10% predicted). Thirty-one percent of the group showed changes in both IC and FEV 1.0 ( 10% predicted each), whereas only 14% of subjects showed a significant change in FEV 1.0 ( 10% predicted) without a concomitant improvement in IC. Bronchodilator reversibility criteria proposed by the ERS and ATS vary considerably. Using ERS criteria for a significant change in FEV 1.0 (changes expressed as a percentage of the predicted value are shown in Figure 1, top), which attempts to correct for improvement bias based on low prebronchodilator FEV 1.0 values, 55% of our sample would be classified as nonresponders. Using ATS criteria (percent changes are shown in Figure 1, bottom), 45% were nonresponders. On the other hand, using arbitrary criteria for the change in IC ( 10% predicted), 38% of the sample would be classified as nonresponders. It is also of interest that 24% of study subjects did not show significant changes in either IC or FEV 1.0 after IB treatment. It is possible that these poor responders did not have significant expiratory flow limitation at rest. This suggestion has been offered by Pellegrino and Brusasco (19) and by Tantucci and colleagues (20) to explain the lack of bronchodilator responses (i.e., increase in IC) in a subpopulation of COPD patients after acute salbutamol therapy. We did not measure expiratory flow limitation directly in our study; however, our finding that nonresponders had significantly higher FEV 1.0 (i.e., a mean of 41% predicted) and less dynamic lung hyperinflation during exercise (i.e., end-expiratory lung volume increased by 0.27 L) when compared with responders who demonstrated a lower FEV 1.0 (i.e., 31% predicted) and greater levels of dynamic hyperinflation (i.e., endexpiratory lung volume increased by 0.41 L) suggested that nonresponders had less expiratory flow limitation, but this possibility needs to be verified in further studies. On the basis of our experience with a number of controlled studies designed to evaluate a variety of therapeutic interventions (23, 29 31), together with information about the reliability of exercise endurance and symptom measurements (1), we arbitrarily selected an improvement in Tlim of 25% and a decrease in standardized Borg ratings by 10% as the minimal change required for a positive acute clinical response. Using these criteria, change in IC alone (i.e., 10% predicted) accurately predicted a clinical response in 59% of our study patients. This IC reversibility criterion was as good as, if not superior to, traditional reversibility criteria based on change in FEV 1.0 proposed by the ATS and ERS, which was accurate in predicting a clinical response in 52 and 48% of our study subjects, respectively. In summary, resting spirometric increases in IC correlated significantly with improvements in exercise endurance and dyspnea after acute anticholinergic therapy in advanced COPD. In almost one-third of our study sample, improvements in IC occurred despite minimal change in FEV 1.0. Reduced lung hyperinflation, as indirectly reflected by increased IC, IRV, or VC, provides a reasonable mechanistic basis for observed clinical improvements and should be considered in conjunction with the change in FEV 1.0 when evaluating acute therapeutic responses in more advanced COPD. In evaluating therapeutic efficacy, resting spirometric measurements, however, do not obviate the need for direct pre- and postbronchodilator assessments of exercise endurance and symptom alleviation. References 1. O Donnell, D. E., M. Lam, and K. A. Webb Measurement of symptoms, lung hyperinflation and endurance during exercise in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 158: Belman, M. J., W. C. Botnick, and J. W. Shin Inhaled bronchodilators reduce dynamic hyperinflation during exercise in patients with chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 153: O Donnell, D. E., J. C. Bertley, L. K. L. Chau, and K. A. Webb Qualitative aspects of exertional breathlessness in chronic airflow limitation: pathophysiological mechanisms. Am. Rev. Respir. Crit. Care Med. 155: Chrystyn, H., B. A. Mulley, and M. D. Peake Dose response relation to oral theophylline in severe chronic obstructive airways disease. Br. Med. J. 297: American Thoracic Society Lung function testing: selection of reference values and interpretative strategies. Am. Rev. Respir. Dis. 144: Siafakas, N. M., P. Vermeire, N. B. Pride, P. Paoletti, J. Gibson, P. Howard, J. C. Yernault, M. Decramer, T. Higenbottam, D. S. Postma, and J. Rees, on behalf of the Task Force ERS consensus statement: optimal assessment and management of chronic obstructive pulmonary disease (COPD). Eur. Respir. J. 8: Stoller, J. K., R. Ferranti, and A. R. Feinstein Further specification and evaluation of a new clinical index for dyspnea. Am. Rev. Respir. Dis. 135: Gardner, R. M., J. L. Hankinson, J. L. Clausen, R. O. Crapo, R. O. Johnson, Jr., and G. R. Epler American Thoracic Society standardization of spirometry 1987 update. Am. Rev. Respir. Dis. 136: Morris, J. F., A. Koski, and L. C. Johnson Spirometric standards for healthy non-smoking adults. Am. Rev. Respir. Dis. 103: Knudson, R. J., R. C. Slatin, M. D. Lebowitz, and B. Burrows The maximal expiratory flow volume curve. Am. Rev. Respir. Dis. 113: Goldman, H. I., and M. R. Becklake Respiratory function tests: normal values at median altitudes and the prediction of normal results. Am. Rev. Tuberc. Pulm. Dis. 79: Black, L. F., and R. E. Hyatt Maximal respiratory pressures: normal values, and relationship to age and sex. Am. Rev. Respir. Dis. 99: Mahler, D. A., D. H. Weinberg, C. K. Wells, and A. R. Feinstein The measurement of dyspnea: contents, interobserver agreement, and physiologic correlations of two new clinical indexes. Chest 85: Jones, N. L Clinical Exercise Testing, 3rd ed. W.B. Saunders, Philadelphia. 15. Borg, G. A. V Psychophysical basis of perceived exertion. Med. Sci. Sports Exercise 14: O Donnell, D. E., and K. A. Webb Exertional breathlessness in patients with chronic airflow limitation: the role of lung hyperinflation. Am. Rev. Respir. Dis. 148: Jones, B., and M. G. Kenward Design and Analysis of Cross-Over Trials. Chapman & Hall, London. 18. Diggle, P. J., K. Y. Leung, and S. L. Zeger Analysis of Longitudinal Data. Clarendon Press, Oxford. 19. Pellegrino, R., and V. Brusasco Lung hyperinflation and flow limitation in chronic airway obstruction. Eur. Respir. J. 10: Tantucci, C., A. Duguet, T. Similowski, M. Zelter, J.-P. Derenne, and J. Milic-Emili Effect of salbutamol on dynamic hyperinflation in chronic obstructive pulmonary disease patients. Eur. Respir. J. 12: Martinez, F. J., M. Montes de Oca, R. I. Whyte, J. Stetz, S. E. Gay, and B. R. Celli Lung-volume reduction improves dyspnea, dynamic hyperinflation and respiratory muscle function. Am. J. Respir. Crit. Care Med. 155: Laghi, F., A. Jurban, A. Topeli, P. J. Fahey, E. Garrity, Jr., J. M. Arcidi, D. J. DePinto, L. C. Edwards, and M. J. Tobin Effect of lung volume reduction surgery on neuromechanical coupling of the diaphragm. Am. J. Respir. Crit. Care Med. 157: O Donnell, D. E., K. A. Webb, J. C. Bertley, L. K. L. Chau, and A. A. Conlan Mechanisms of relief of exertional breathlessness following unilateral bullectomy and lung volume reduction surgery in emphysema. Chest 110: Wolkove, N., E. Dajczman, A. Colacone, and H. Kreisman The relationship between pulmonary function and dyspnea in obstructive lung disease. Chest 96: Hay, J. G., P. Stone, J. Carter, S. Church, A. Eyre-Brook, M. Pearson, A. Woodcock, and P. Calverley Bronchodilator reversibility, exercise performance and breathlessness in stable chronic obstructive pulmonary disease. Eur. Respir. J. 5: O Donnell, D. E Exertional breathlessness in chronic respiratory disease. In D. A. Mahler, editor. Lung Biology in Health and Disease, Vol. 111: Dyspnea. Marcel Dekker, New York

8 O Donnell, Lam, and Webb: Improvement in Exercise after Bronchodilation in COPD Bauerle, O., C. A. Chrusch, and M. Younes Mechanisms by which COPD affects exercise tolerance. Am. J. Respir. Crit. Care Med. 157: O Donnell, D. E., and K. A. Webb Breathlessness in patients with severe chronic airflow limitation: physiologic correlates. Chest 102: O Donnell, D. E., M. McGuire, L. Samis, and K. A. Webb The impact of exercise reconditioning on breathlessness in severe chronic airflow limitation. Am. J. Respir. Crit. Care Med. 152: O Donnell, D. E., D. J. Bain, and K. A. Webb Factors contributing to relief of exertional breathlessness during hyperoxia in chronic airflow limitation. Am. J. Respir. Crit. Care Med. 155: O Donnell, D. E., M. McGuire, L. Samis, and K. A. Webb Effects of general exercise training on ventilatory and peripheral muscle strength and endurance in chronic air flow limitation. Am. Rev. Respir. Crit. Care Med. 157: