Published ahead of print on September 15, 2005, doi:10.1164/rccm.200504-595OC
© 2005 American Thoracic Society doi: 10.1164/rccm.200504-595OC
Lung Mechanics and Dyspnea during Exacerbations of Chronic Obstructive Pulmonary DiseaseDepartment of Medicine, Clinical Sciences Centre, University Hospital Aintree, Liverpool, United Kingdom; and Department of Medicine, Royal College of Surgeons in Ireland, Beaumont Hospital, Dublin, Ireland Correspondence and requests for reprints should be addressed to Peter Calverley, M.D., F.R.C.P., Department of Medicine, Clinical Sciences Centre, University Hospital Aintree, Longmoor Lane, Liverpool L9 7AL, UK. E-mail: pmacal{at}liverpool.ac.uk
Rationale: Exacerbation of chronic obstructive pulmonary disease commonly causes hospitalization. The change in lung mechanics during exacerbation and its relationship to symptoms in spontaneously breathing individuals has not been described. Objective: We hypothesized that changes in both airflow and lung volumes would occur during an exacerbation, but that only volume change would relate to symptomatic improvement. Methods: Lung mechanics and resting dyspnea were recorded in 22 hospitalized patients during recovery from exacerbation. Measurements: Spirometry, inspiratory capacity, respiratory system resistance and reactance, tidal breathing patterns, and expiratory flow limitation were recorded after nebulized bronchodilator therapy on the first 3 d after admission, at discharge, and 6 wk postadmission (Day 42). Prebronchodilator measurements were taken on Day 2, at discharge, and on Day 42. Main Results: Postbronchodilator inspiratory capacity increased 0.23 ± 0.07 L by discharge and 0.42 ± 0.1 L by Day 42, FEV1 rose 0.09 ± 0.04 and 0.2 ± 0.05 L at discharge and Day 42, respectively, and FVC increased 0.21 ± 0.08 and 0.47 ± 0.09 L at discharge and Day 42 (all p < 0.05). Consistent reduction in dyspnea was seen as the exacerbation resolved. Respiratory system resistance, FEV1/FVC, and expiratory flow limitation were unchanged throughout, indicating that changes in lung volume rather than airflow resistance predominated. Conclusions: Improvement in operating lung volumes is the principal change seen as a chronic obstructive pulmonary disease exacerbation resolves and increase in inspiratory capacity is a useful guide to a reduction in dyspnea.
Key Words: breathlessness • inspiratory reserve volume • lung function • lung hyperinflation Periodic exacerbations of symptoms are a major cause of morbidity, mortality, and health care costs in patients with chronic obstructive pulmonary disease (COPD) (1, 2). They are associated with a worse quality of life (3, 4) and a more rapid decline in both health status (4) and FEV1 (5, 6). Most exacerbations are precipitated by either bacterial or viral infections (7, 8), but the resulting symptoms relate mainly to altered lung function. Increased cough, sputum production, and sputum purulence occur during exacerbations, but patients identify the most important symptom as being worsening breathlessness (9, 10). Although changes in lung mechanics are thought to be the major cause of dyspnea in COPD, we have few data about the time course and nature of the change in lung mechanics during the resolution of an exacerbation in normocapnic patients. Studies of patients in intensive care units have shown that the resistive and elastic work of breathing increases significantly during exacerbations, which leads to a marked increase in intrinsic positive end-expiratory pressure (11, 12). How these changes relate to more familiar measurements of flow and volume or to changes in symptom intensity in spontaneously breathing subjects has not been reported, but several mechanisms appear possible. Increased airway resistance, due to the release of mediators or the direct effect of inflammation reducing airway diameter, should lessen as the exacerbation resolves and be reflected by an increase in peak expiratory flow or FEV1. However, the changes reported in these measures during an exacerbation of COPD are relatively small (13). Tidal expiratory flow limitation (EFL) is a better indicator of dyspnea severity than is the FEV1 in stable COPD (14), and thus changes in EFL due to airway narrowing or closure during an exacerbation of COPD might relate to changes in symptom intensity. The relationship of EFL to other indices of lung mechanics during exacerbations has not been reported. However, the factor most likely to explain the change in symptoms during an exacerbation is a change in operating lung volume. An increase in end-expiratory lung volume (EELV) during exercise is the best predictor of symptom intensity and the degree of exercise limitation (15), and both improve after administration of a bronchodilator drug (16). During an exacerbation of COPD closure of small airways may occur because of mucus plugging, airway wall edema, or inflammation of lymph follicles, all of which may increase EELV even under resting conditions (17). We hypothesized that during the resolution of an exacerbation of COPD there would be larger changes in lung volume than in expiratory flow–related measurements. It proved impractical to study the onset of an exacerbation and thus we prospectively studied patients on admission to hospital and monitored their subsequent clinical course. To compare patients we standardized treatment given and measurement timing as well as prior bronchodilator therapy. We recorded pre- and post- bronchodilator values for spirometry, dynamic lung volumes, oscillatory lung mechanics, and breathing pattern and tested for the presence of tidal expiratory flow limitation, relating these to changes in dyspnea and clinical improvement. Some of the results of these studies have previously been reported in the form of abstracts (18–20).
Subject Recruitment Patients were recruited within 24 h of hospitalization. An acute exacerbation was defined as an increase in at least two major symptoms: dyspnea, sputum purulence, or increased sputum volume, sufficient to require hospitalization (21). Patients were excluded if they had acute pneumonia, pneumothorax, atelectasis, or heart failure; had respiratory acidosis (pH < 7.32); had a coexisting illness that rendered them too ill to participate; or were unwilling to participate. All patients had a diagnosis of COPD defined both clinically and physiologically (22) with an FEV1 less than 80% predicted and an FEV1/FVC ratio less than 0.7 when enrolled (23). Any patient in whom the FEV1 increased by more than 400 ml after administration of nebulized bronchodilator (24) or whose lung function improved to within normal values at any test session was excluded. Written informed consent was obtained from each patient. The local research ethics committee granted study approval (see the online supplement for further details).
Inpatient Management of Exacerbation
Study Design
Measurements
Measurement of total respiratory system resistance and reactance.
Tidal breathing and resting expiratory flow limitation. Negative expiratory pressure during tidal breathing was applied to detect expiratory flow limitation in the seated patient, as described previously (26). Breaths were considered flow limited when negative expiratory pressure did not increase expiratory flow relative to the preceding untested breath. Flow limitation was also expressed as the percentage of expired tidal volume affected (FL % VT).
Assessment of Dyspnea See the online supplement for additional detail on the methods used to make all these measurements.
Statistical Analysis
Clinical Characteristics of the Study Population Of 44 patients identified as being potentially eligible over an 18-mo period, 30 entered the study but 7 withdrew during the early stages because of clinical deterioration (n = 5) or technical difficulty in completing maneuvers (n = 2). One patient was excluded as that patient's lung function subsequently returned to normal. The admission characteristics of the remaining 22 patients are shown in Table 1.
All patients reported symptoms consistent with an acute exacerbation before admission to hospital. Four patients had received antibiotics and/or oral corticosteroids in the week before admission. No patient had been admitted with an exacerbation of COPD in the preceding 6 wk. Three patients received intravenous aminophylline and none required ventilation or died. The median length of hospital stay was 7 d (range, 3–10 d). Two patients were discharged on Day 3 and their measurements on Day 3 were also included in the discharge day data. See the online supplement for sputum microbiological data. Neither the presence of microorganisms in the sputum nor the treatment received before admission appeared to influence the changes in lung mechanics seen during recovery from the exacerbation. Data for 6 of the 22 patients were not available for Day 42, either because of recurrent exacerbations of COPD (n = 5) or because the patient declined to attend for these follow-up studies (n = 1). Admission data for the 16 patients who returned on Day 42 are shown in Table 1. Admission characteristics of the subjects who returned on Day 42 did not differ from those who did not (Table 1), nor did the data at discharge (data not presented).
Changes in Lung Mechanics during the Exacerbation
IC increased from Day 1, when it was 62 ± 4% predicted at admission, to 73 ± 4% at discharge (n = 22, p < 0.05), and to 81 ± 7% on Day 42 (n = 16, p < 0.01). These values correspond to an improvement of 0.23 ± 0.07 L (p < 0.01) by discharge and 0.42 ± 0.1 L (p < 0.01) by Day 42; see also Table 2. The change in IC from admission to discharge was related to the change in FVC (r = 0.47, p < 0.05) but not the change in FEV1.
Respiratory system resistance and reactance.
The group mean respiratory system reactance measured at 5 Hz, X5, improved from admission to discharge by 0.11 ± 0.02 kPa/L/s (p < 0.001) and was not significantly different on Day 42 (Table 2 and Figure 2B). From admission to discharge postbronchodilator X5 became less negative in 16 subjects (–0.45 ± 0.04 to –0.29 ± 0.04 kPa/L/s) and remained constant in the other 6 patients. There were no significant differences on admission between patients whose X5 improved over time and those for whom it did not. The "X5 improvers" had significant improvements in FEV1 (mean change, 130 ml; p < 0.05) and IC (mean change, 270 ml; p < 0.01) whereas the "nonimprovers" did not show significant improvement (see Table E1 in the online supplement).
Breathing pattern.
Tidal flow limitation during exacerbation. On admission, when seated, 9 of the 22 patients showed EFL, whereas 13 were not tidal flow limited. EFL resolved in four patients and appeared for the first time by discharge in two patients; the remaining 16 patients were unchanged. In EFL, subjects' mean percentage of FL as a percentage of VT (FL % VT) was 44 ± 3% at baseline. There was no clear pattern of change or improvement in FL % VT during recovery.
Bronchodilator Response during Exacerbation
Breathing pattern was little affected by the bronchodilator, with a small but significant increase in VT occurring on Day 2 (+80 ml) and Day 42 (+70 ml). Expressing the IC as a percentage of the FVC showed no acute change with bronchodilator administration, although the postbronchodilator IC/FVC ratio value rose from 64% on Day 2 to 68% at discharge. The acute change in IRV was small on Day 2 but increased by a mean of 120 ml (p < 0.05) on discharge and by 240 ml (p < 0.05) on Day 42.
Relative Change in Resting Breathlessness and Lung Mechanics Only 16 of 22 patients reported any reduction in resting dyspnea by discharge. These patients had a significantly lower IC on admission compared with patients not reporting a reduction in dyspnea (p < 0.01). Both FEV1 and IRV were also lower in this group on admission (both p = 0.05). The degree of resting dyspnea these patients reported on discharge was similar to that on admission among the patients whose dyspnea did not improve over time (see Table E2). In patients whose Borg score improved during hospitalization, mean FEV1 increased by 0.20 ± 0.07 L (p < 0.05), mean IC by 0.54 ± 0.16 L (p < 0.01), and mean X5 by 0.15 ± 0.04 kPa/L/s (p < 0.01) by Day 42 (n = 11). In contrast, in those whose dyspnea was unchanged, neither FEV1 nor IC increased significantly (see Table E2). Similar changes were seen in each of these variables at the time of discharge although these were generally of a smaller magnitude. By discharge, there was no significant difference in IC between patients who felt an improvement in breathlessness and those who did not. The magnitude of the change in dyspnea from admission to discharge or follow-up was unrelated to the change in IC or other measured variables.
Although exacerbations of COPD are a frequent cause of hospitalization, surprisingly little is known about the physiologic changes that accompany their resolution and what relationship these bear to the patient's symptoms. Hypoxia and hypercapnia have been carefully studied, with the latter relating directly to lung mechanics (28) and showing a variable resolution over time (29) and between episodes (30). Our data demonstrate that from admission to discharge and through the 6 wk after the exacerbation there was a small increase in FEV1, no significant change in airway resistance or tidal FL, but a significant increase in both FVC and IC together with an improvement in reactance. In nonhypercapnic exacerbations, changes in operating lung volumes rather than either tidal or forced expiratory flow related to both the resolution of the exacerbation and the change in breathlessness reported by the patient.
A particular strength of this study is the assessment of both flow- and volume-related measurements at standardized times and with standardized therapy as the exacerbation resolved. Treatment before or during admission, including theophylline use, did not influence lung mechanics, in keeping with another report (31). All measurements were made at rest and so we cannot assess the effect of dynamic changes in lung volume that are likely to occur on exercise in these patients (32). We did not measure TLC by body plethysmography as this was not possible in these sick patients. Nonetheless, we believe it unlikely that TLC increased during recovery and instead either remained constant or fell. Hence we believe it likely that the change in postbronchodilator FVC and IC reflects a fall in residual volume over time. Likewise, the calculated Noninvasive measurements of total respiratory system resistance and reactance were easy to make, were well tolerated, and provided information complementary to that obtained from the change in lung volumes. The accuracy of these data may be influenced in COPD by the presence of upper airway shunt compliance (33). This factor does not change acutely, and our data reported relative to the admission values are likely to be valid. We have reported data only at 5 Hz as this best reflects total respiratory system resistance rather than considering frequency dependence of resistance, a field in which the interpretation of oscillatory mechanics in COPD remains controversial. Tidal EFL was determined by the negative expiratory pressure technique in seated subjects. More information might have been obtained had the subjects been able to lie supine (34), although analysis of the percentage of each breath showing FL, a potentially more sensitive descriptor, did not change our results. The small increase we observed in postbronchodilator FEV1 from admission to discharge is similar to that reported previously (35). The changes in spirometry were due to an increase in volume rather than flow as judged by the static FEV1/FVC ratio throughout the recovery period. This suggests that as the exacerbation resolved there was an opening of lung units with mechanical properties similar to the lung at admission, as the unchanged FEV1/FVC ratio reflects the mechanical time constant of the respiratory system. The IC increase in proportion to the FVC during the course of the exacerbation is in keeping with this, as is the relatively constant R5 despite the fall in EELV. Respiratory frequency was also constant across the study days and did not seem to be an important determinant of the volume change at rest. In our patients, we saw no consistent relationship between the presence of tidal EFL and the change in either IC or FVC over time. This might be due to breath-to-breath variation in the operating lung volume, as has been reported after administration of nebulized bronchodilators in stable COPD (26), but is more likely to reflect the change in residual volume over time. This contribution from a reduced static lung volume helps explain why the improvement in postbronchodilator inspiratory reserve volume occurred without any change in breathing pattern or respiratory timing. Total respiratory system reactance became significantly less negative during recovery. This may in part result from the lower operating lung volume, but the magnitude of this change is larger than would be expected if this were the only operative factor. Taken together with the change in EELV and constant R5, the data suggest that the "specific conductance" of the respiratory system was increasing as the patient improved but that individuals opt to reduce operating lung volume rather than to maintain a higher EELV with lower respiratory system resistance. This is the first study of the response of patients with COPD to nebulized bronchodilators during and after an exacerbation. The large doses used are on the flat part of the dose–response curve for spirometry (36) and significantly improved FVC and Borg score at all time points support a relationship between operating lung volumes and breathlessness. In contrast, significant changes in FEV1, IC, and reactance became apparent only on discharge, becoming larger in those monitored to 6 wk postadmission. These help explain why spirometric reversibility testing is unreliable soon after an exacerbation and why tests looking at FEV1 change do not relate well to symptomatic response (37). Increased breathlessness at rest is a common but not invariable accompaniment of COPD exacerbations (13). As in studies during exercise, Borg dyspnea scores were not normally distributed, but for comparison with other data in the literature we have reported them parametrically (15, 38, 39). Although these data can be indicative only of the direction of change, not its magnitude, we found that those patients reporting less breathlessness at the time of discharge and follow-up were the ones in whom IC and IRV improved significantly over this time, a change not seen in the smaller number of subjects in whom dyspnea remained constant. Patients reporting worse breathlessness at admission had significantly worse resting lung mechanics and larger studies will be needed to properly test the relationship between symptom change during exacerbation resolution and different measures of lung volume. Mechanically, the situation at rest during an exacerbation was comparable to that at maximal tolerated exercise when clinically stable, the change in postbronchodilator inspiratory capacity of 420 ml from Day 1 to Day 42 being comparable to the volume change reported during exercise (32, 40). In summary, in normocapnic patients hospitalized with an exacerbation of COPD, improvement in operating lung volumes was related to reduction in dyspnea rather than any index of expiratory flow; however, it was measured. The impairment in resting lung mechanics resolved slowly and was not complete at the time of discharge as determined on the basis of clinical criteria. The relationship of resting lung mechanics on discharge to the dynamically regulated lung volume that usually determines exercise performance will require further study, as it may explain the variability in the subsequent clinical course postexacerbation.
Supported in part by an EU 5th Framework CARED program grant. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200504-595OC on September 15, 2005 Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form April 17, 2005; accepted in final form September 14, 2005
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
E were calculated. 
p < 0.01; 
p < 0.001, discharge and Day 42 compared with Day 1. Note that data for Day 42 apply to 16 patients only (see 
