HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:
Guest Access | Sign In via User Name/Password

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stanbrook, M. B. Articles by Tullis, D. E. Search for Related Content PubMed PubMed Citation Articles by Stanbrook, M. B. Articles by Tullis, D. E.

The Repeatability of Forced Expiratory Volume Measurements in Adults With Cystic Fibrosis^*

^* From the Adult Cystic Fibrosis Program, St. Michael’s Hospital (Dr. Tullis); the Asthma and Airway Centre, University Health Network (Dr. Stanbrook); the Division of Respirology, University of Toronto (Drs. Stanbrook and Tullis); the Hospital for Sick Children Research Institute (Dr. Corey); the Institute for Clinical Evaluative Sciences in Ontario (Dr. Stanbrook); and the Clinical Epidemiology Program, Department of Health Policy, Management, and Evaluation, University of Toronto (Drs. Stanbrook, Corey, and Tullis), Toronto, ON, Canada.

Correspondence to: D. Elizabeth Tullis, MD, FCCP, Room 6-045, 30 Bond St, St. Michael’s Hospital, Toronto, ON, Canada M5B 1W8; e-mail: tullise{at}smh.toronto.on.ca

	Abstract

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Study objective: To determine the repeatability of measurements of FEV₁ in adults with lung disease due to cystic fibrosis (CF).

Design: Single cohort study nested within a randomized controlled trial.

Setting: Adult CF of a university teaching hospital. Subjects were participants in a randomized trial of an experimental mucolytic drug.

Patients: Twenty-one adults (mean age, 27.5 ± 9.2 years [± SD]) with CF and mild-to-moderate airflow obstruction (FEV₁ 70 ± 15% predicted). Patients were in clinically stable condition prior to and during the study.

Interventions: Repeated FEV₁ measurements were obtained at specific times of the day for 9 consecutive days, for a total of 31 measurements from each subject. Statistical measures of repeatability were calculated. Variation over the course of 1 day and variation from 1 day to the next were examined separately.

Measurements and results: For day-to-day FEV₁ measurements, the within-subject SD was 0.145 L (4.5% of predicted), indicating greater variation compared to values previously established in normal subjects. The coefficient of repeatability indicated that day-to-day measurements could differ by as much as 13% of predicted in the absence of clinical change. For measurements within a single day, variation was not observed to be greater than normal.

Conclusions: In adults with CF, day-to-day variation in FEV₁ measurements is greater than normal and similar to that seen in other obstructive lung diseases. Changes in FEV₁ over time in adults with CF can likely be interpreted using the same criteria that apply to asthma or COPD.

Key Words: adult • cystic fibrosis • forced expiratory volume • reproducibility of results • respiratory function tests

	Introduction

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Pulmonary function tests provide important guidance for the clinical management of patients with cystic fibrosis (CF). In particular, FEV₁ is useful for establishing the severity of CF-related lung disease, measuring the progression of lung disease over time, identifying acute pulmonary exacerbations, evaluating the response to therapy, and determining disease prognosis.^{1 2}

For any given individual, when pulmonary function testing is repeated over time, some degree of variation will exist in the results. Variation may result from any combination of technical reasons, physiologic factors, and changes in disease status.³ The last component of variation, a change in the underlying disease, is usually the one of interest when serial measures of pulmonary function are followed. The repeatability of pulmonary function measurements (the degree to which repeated measurements are similar in the absence of any clinical change) depends on the other components. Knowing the repeatability by determining the magnitude of the physiologic and technical components of variation is necessary in order to be able to interpret whether an observed difference in repeated measurements is likely to represent a true change in clinical status.

FEV₁ variation is greater than normal in patients with asthma or COPD.^{4 5 6} This has important consequences for pulmonary function test interpretation: for a change in FEV₁ to be judged significant in asthma or COPD, it must exceed a larger threshold than would be required for normal subjects.³ Whether other obstructive lung diseases such as CF display similar variation has not been fully evaluated. One study⁷ in children with CF has observed greater- than-normal variation in FEV₁, but also found that variation tended to decrease in older children. Data for adults with lung disease due to CF are lacking. We therefore sought to determine the repeatability of FEV₁ measurements over time in a cohort of clinically stable adults with CF.

	Materials and Methods

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Subjects
Subjects were outpatients of the Toronto Adult Cystic Fibrosis Program. Eligible subjects were 16 years old, had a diagnosis of CF as defined by a positive sweat chloride test result, had an FEV₁ between 40% and 80% of predicted, and had an oxygen saturation of at least 92% on room air. To be enrolled, subjects had to be able to perform pulmonary function testing satisfactorily. Written informed consent was obtained from all enrolled subjects. The study protocol received official ethics approval from the hospital institutional review board.

Subjects were excluded if they had experienced a pulmonary exacerbation within the previous 4 weeks; if a new steroid, antibiotic, or bronchodilator medication was initiated within the previous 2 weeks; if they had experienced a symptomatic viral or bacterial infection within the previous 2 weeks; or if they had had a fever > 38°C (100.4°F) within the previous 5 days before the first FEV₁ measurement was obtained.

Measurement of FEV₁
All FEV₁ measurements were obtained by experienced pulmonary technologists following the usual protocols of our pulmonary function laboratory, conforming to established standards for reproducibility and quality control.⁸ The largest FEV₁ obtained from three properly performed maneuvers was recorded as the FEV₁ value. All subjects were able to achieve a reproducible test at all test sessions. Measurements were made on a KoKo spirometer (Roxane; Toronto, ON, Canada) that was calibrated daily over multiple flow rates using a 3-L syringe and additionally calibrated weekly against biological controls.

FEV₁ was measured at 10 AM, 10:30 AM, and 4 PM daily for 9 consecutive days. On day 2 and day 8, measurements were also obtained at 12 PM and 2 PM.

Co-interventions
The FEV₁ measurements were obtained as part of a randomized, double-blind, placebo-controlled trial of inhaled gelsolin (Biogen Inc., Cambridge, MA), an experimental mucolytic agent. By design, 75% of subjects enrolled received the study drug. Gelsolin or placebo (saline solution diluent) was administered via a Hudson T Up-draft II Neb-U-Mist nebulizer (Hudson RCI; Temecula, CA) once daily immediately following the 10 AM FEV₁ measurement. A drug volume of 5 mL was nebulized with compressed air at a rate of 5 L/min until the mist stopped. All subjects randomized to gelsolin received a dose of 25 mg after completing an initial dose escalation phase (3 mg for 2 days, then 10 mg for 2 days). The results of this study of the efficacy of gelsolin, which have been described elsewhere,⁹ indicated that gelsolin as compared to placebo had no significant effect on lung function. The sponsor of this prior study was not involved in any way with the approval, design, analysis, or reporting of the present study.

During the study, 12 subjects received inhaled tobramycin as part of their usual therapy. All subjects received albuterol, either with inhaled tobramycin (1 mL of 5 mg/mL albuterol via nebulizer) or at the time of morning physiotherapy (via nebulizer, or two puffs via metered-dose inhaler). Tobramycin treatment and physiotherapy usually occurred approximately 2 h prior to the first FEV₁ measurement; subjects receiving tobramycin received a second nebulized dose in the evening, after all measurements for the day had been completed. No other nebulized agents, such as dornase alfa, ipratropium, or hypertonic saline solution, were used. ß-Agonist administration was not permitted within 1 h prior to the start of testing or at any time during the testing period itself. Doses of nebulized and other medications were required to be kept constant during the study period; apart from the study drug, no medications were introduced or discontinued during this period.

Statistical Analysis
To express the repeatability of FEV₁ measurements, the within-subject SD and the coefficient of repeatability were calculated.^{10 11} In order to allow comparison with previous studies of FEV₁ variability, the coefficient of variation was also determined.¹² The pooled within-subject SD was calculated using one-way analysis of variance. The coefficient of repeatability was derived as 2 x 2 x pooled within-subject SD.^{10 11} Prior to these calculations, the appropriate scale for FEV₁ was determined by plotting SD vs mean and by calculating the index of heterogeneity.¹²

The coefficient of repeatability is the value that is 95% of the differences between any two repeated measurements, ie, it represents the 95% range for such differences.^{10 11} For greater clarity, the coefficient of repeatability can also be expressed as the 95% range for change; an observed change outside this range is unlikely to result from the inherent variation in a measurement over time and thus can be attributed to external factors.

Variation over the course of a single day ("diurnal variation") and variation over several days ("day-to-day variation") were analyzed separately. The effects of gender, study drug assignment, and day and time of measurement on FEV₁ values were assessed using analysis of covariance. The patterns of diurnal and day-to-day variation were analyzed using simple linear regression with linear and quadratic terms; patterns of change between prespecified pairs of time points were analyzed with the signed-rank test. Statistical analysis was performed using SAS version 8.02 (SAS Institute; Cary, NC).

	Results

TOP Abstract Introduction Materials and Methods Results Discussion References

 
A total of 21 subjects were enrolled in the study (7 men and 14 women). Baseline demographic variables are shown in Table 1 . With the exception of body mass index (BMI), all baseline variables were normally distributed across the study population. Subjects were adequately nourished, with only three subjects having a BMI < 20. The severity of airflow obstruction³ at baseline was mild in eight subjects (38%), moderate in eight subjects (38%), and moderately severe or severe in five subjects (24%). As expected, male subjects had larger baseline FEV₁ measurements than female subjects (2.55 L vs 2.05 L, p = 0.05). However, when comparison was made to predicted values, the severity of airflow obstruction was not significantly different for male and female subjects (67% vs 71% predicted, p = 0.55). Baseline FEV₁ did not differ significantly between subjects who did or did not receive nebulized tobramycin (2.16 L vs 2.29 L, p = 0.60) or between subjects who received the study drug and those who received placebo (2.22 L vs 2.19 L, p = 0.93). Logarithmic transformation of the FEV₁ values reduced variance heterogeneity marginally for the analysis of diurnal variation, but not for day-to-day variation. For ease of interpretation and comparison, all results are presented in the original scale.

View larger version (25K): [in this window] [in a new window] [Download PPT slide]   Figure 1.. Pattern of FEV1 measurements over the course of a single day in 21 clinically stable adults with CF. Each symbol represents an individual patient. Measurements shown were obtained on study day 2.

Day-to-Day Variation in FEV₁
Figure 2 illustrates the pattern of FEV₁ measurements obtained daily for 9 days for each of the 21 subjects. No consistent pattern was observed among subjects during this time period, and FEV₁ values showed no tendency either to increase or to decrease from day 1 to day 9 (p = 0.15 by the signed-rank test).

View larger version (30K): [in this window] [in a new window] [Download PPT slide]   Figure 2.. Pattern of FEV1 measurements over the course of 9 consecutive days in 21 clinically stable adults with CF. Each symbol represents an individual patient. Measurements shown were obtained at 10 AM on each study day.

Four subjects had a clinical diagnosis of asthma in addition to CF. Exclusion of these subjects did not significantly affect measurements of diurnal variation (coefficient of repeatability, 0.292 [8.8% of predicted] on day 2, 0.354 [10.9% of predicted] on day 8) or day-to-day variation (coefficient of repeatability, 0.419 [12.8% of predicted] at 10 AM, 0.463 [13.7% of predicted] at 4 PM).

The study drug had no consistent effect on diurnal variation, but day-to-day variation appeared to be lower among subjects receiving placebo than among subjects receiving the study drug (pooled within-subject SD, 0.104 L [3.1% of predicted] vs 0.155 L [4.8% of predicted], respectively).

	Discussion

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Using a unique data set of FEV₁ measurements made repeatedly for several days, this study was able to provide estimates of the variation of FEV₁ in adults with lung disease due to CF. Technical and disease-related variation were minimized through rigorous attention to established laboratory standards for quality control⁸ and by ensuring that participants were clinically stable during the study period, respectively. The variation measured should therefore reflect the physiologic variation that is inherent in adults with CF. Since it is difficult or impossible to control for physiologic variation, the results of this study represent important information that is necessary for proper interpretation of spirometry in this patient population. In addition, these data may be valuable for use in estimating effect size when planning clinical trials of therapeutic interventions in adult CF patients.

The day-to-day variation in FEV₁ observed in this study of adults with CF is greater than values previously described in normal subjects,^{13 14 15} and is similar to values reported in patients with other obstructive lung diseases.^{16 17 18 19} One study of 40 normal subjects tested for 6 days¹³ observed a within-subject SD in FEV₁ of 0.089 L, about 40% less than in our adult CF patients. In 5,887 smokers with mild COPD enrolled in the Lung Health Study, the coefficient of repeatability for FEV₁ was 0.330 in men and 0.240 in women.¹⁸ While these values are also smaller then those in our patients (which were 0.498 and 0.356, respectively), the results of the Lung Health Study¹⁸ were derived from only two measurements in each subject and the number of days in between measurements was not standardized.

Most previous studies have used the coefficient of variation to represent FEV₁ repeatability, with results ranging from 2 to 3% in normal subjects^{13 14 15} and 4 to 8% in those with obstructive disease.^{17 18} For day-to-day measurements, the coefficient of variation that can be calculated from our data is 6.7%, similar to the latter group. However, as was the case in a previous study of children with CF,⁷ we determined that the coefficient of variation for day-to-day measurements was not the most valid measure of repeatability in our population, for statistical reasons (the SD was not proportional to the mean). This implies that it is preferable to evaluate FEV₁ in terms of absolute change, rather than relative change, when spirometry is used as a measure of clinical status over time.

For diurnal FEV₁ measurements, the coefficient of variation was 5.1% in our patients, only marginally larger than the values of 2 to 5% previously reported in normal individuals^{5 14 15} and less than the values of 7 to 8% described in patients with obstructive disease,^{4 5 6} in contrast to the findings for day-to-day variation. This may indicate that adults with CF do not exhibit the increased diurnal variation in FEV₁ that is characteristic of asthma.²⁰ However, our subjects received a nebulized treatment each morning during the study period that was observed to decrease the FEV₁ measurement made 30 min later. In addition, all subjects received albuterol, and half received nebulized tobramycin, 2 h prior to the first spirometry measurement. The possibility that these co-interventions altered the inherent diurnal FEV₁ variation in our patients cannot be ruled out; in fact, tobramycin use appeared to decrease variation. Nevertheless, since inhaled ß-agonists and nebulized tobramycin are widely used daily therapies in adult CF patients, our measurements are likely a valid representation of the variability that will be measured in clinical practice.

Our findings in adults are similar to results previously described in children with CF. Cooper and colleagues⁷ performed pulmonary function testing in 28 such children at various times on 5 different days. The within-subject SD for FEV₁ was approximately twice as large as in 23 normal children matched for height. However, while their study suggested that variation decreased in older children, our observations suggest that adults with CF continue to manifest increased variation in FEV₁. An earlier study of 15 children and adults with CF compared to 15 normal control subjects found increased variation in repeated FEV₁ measurements obtained 2 to 3 min apart at a single testing session, but variation over time was not evaluated.²¹

Strengths of this study include the large number of repeated measurements made and the fact that the times of FEV₁ measurement throughout the day were standardized, thus allowing separate determination of the diurnal and day-to-day variation. An important limitation of the study is the fact that measurements were made during a period when subjects were receiving an experimental drug. Although this drug was found to have no significant effect on the level of FEV₁ achieved by the end of the study, we cannot exclude the possibility that the study drug affected FEV₁ variation. Indeed, day-to-day variation was observed to be greater among those who received active drug than among placebo recipients, although in the latter group variation was still greater than normal. Another limitation of the study is the absence of a concurrent control group of normal subjects, necessitating a comparison to observations from previous studies in order to interpret whether the observed variation was greater than normal. We would point out, however, that such a comparison is similar in validity to pulmonary function test interpretation itself, for which the classification of results as normal or abnormal relies on comparison to predicted values that are based on historical control populations. Studies of variation in pulmonary function tests require a large commitment of time and resources and recruitment of subjects willing to comply with repeated measurements can be very difficult. A randomized controlled trial of therapy provided us with a unique opportunity to obtain a complete set of data for analysis of repeatability; only 2 of 651 scheduled measurements were missed.

The increased variation in FEV₁ measurements over time must be borne in mind when interpreting changes in the results of spirometry in adults with CF. Based on our observations, a new FEV₁ measurement that changes by at least 13% of predicted relative to a measurement made several days previously is likely to represent a true change in clinical status. Similarly, measurements made within the same day than differ by an absolute amount of at least 10% of predicted likely indicate a true clinical change. Since these values are similar to existing criteria by which changes in spirometry are interpreted in patients with asthma or COPD,³ it would seem both appropriate and convenient for clinicians to apply these same familiar criteria when interpreting spirometry in adults with CF.

	Footnotes

 
Abbreviations: BMI = body mass index; CF = cystic fibrosis

Data for the present study were derived from a study originally funded by a grant from Biogen Inc., but Biogen had no financial or other involvement with the present study itself.

Dr. Stanbrook is supported by a fellowship from the Canadian Institutes of Health Research and by the Clinician-Scientist Program of the Department of Medicine, University of Toronto.

Received for publication May 8, 2003. Accepted for publication August 21, 2003.

	References

TOP Abstract Introduction Materials and Methods Results Discussion References

 

1. Davis, PB (1999) Clinical pathophysiology and manifestations of lung disease. Yankaskas, JR Knowles, MR eds. Cystic fibrosis in adults. ,45-67 Lippincott-Raven Publishers. Philadelphia, PA:
2. Kerem, E, Reisman, J, Corey, M, et al Prediction of mortality in patients with cystic fibrosis. N Engl J Med 1992;326,1187-1191[Abstract]
3. American Thoracic Society.. Lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis 1991;144,1202-1218[ISI][Medline]
4. Hruby, J, Butler, J Variability of routine pulmonary function tests. Thorax 1975;30,548-553[Abstract/Free Full Text]
5. Pennock, BE, Rogers, RM, McCaffree, DR Changes in measured spirometric indices: what is significant? Chest 1981;80,97-99[Free Full Text]
6. Lewinsohn, HC, Capel, LH, Smart, J Changes in forced expiratory volumes throughout the day. BMJ 1960;1,462-464[Medline]
7. Cooper, PJ, Robertson, CF, Hudson, IL, et al Variability of pulmonary function tests in cystic fibrosis. Pediatr Pulmonol 1990;8,16-22[ISI][Medline]
8. American Thoracic Society.. Standardization of spirometry: 1994 update. Am J Respir Crit Care Med 1995;152,1107-1136[ISI][Medline]
9. Tullis, DE, Corey, M, Dabora, R, et al A double blind, placebo-controlled dose-escalating, tolerability study of inhaled recombinant gelsolin in cystic fibrosis [abstract]. Pediatr Pulmonol 1998;S15,A265[CrossRef]
10. Bland, JM, Altman, DG Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1,307-310[CrossRef][ISI][Medline]
11. Chinn, S Repeatability and method comparison. Thorax 1991;46,454-456[Free Full Text]
12. Chinn, S The assessment of methods of measurement. Stat Med 1990;9,351-362[ISI][Medline]
13. Oldham, PD, Cole, TJ Estimation of the FEV₁. Thorax 1983;38,662-667[Abstract/Free Full Text]
14. McCarthy, DS, Craig, DB, Cherniack, RM Intraindividual variability in maximal expiratory flow-volume and closing volume in asymptomatic subjects. Am Rev Respir Dis 1975;112,407-411[ISI][Medline]
15. Cochrane, GM, Prieto, F, Clark, TJH Intrasubject variability of maximal expiratory flow volume curve. Thorax 1977;32,171-176[Abstract/Free Full Text]
16. Chinn, S, Britton, JR, Burney, PGJ, et al Estimation and repeatability of the response to inhaled histamine in a community survey. Thorax 1987;42,45-52[Abstract/Free Full Text]
17. Spencer, DA, Marshall, T, Weller, PH, et al A superior method for assessing the reproducibility of pulmonary function measurements [abstract]. Am Rev Respir Dis 1992;145,A253
18. Enright, P Reporting the reproducibility of spirometry results. Am Rev Respir Dis 1992;146,1367-1368[Medline]
19. Enright, PL, Johnson, LR, Connett, JE, et al Spirometry in the Lung Health Study: methods and quality control. Am Rev Respir Dis 1991;143,1215-1223[ISI][Medline]
20. Troyanov, S, Ghezzo, H, Cartier, A, et al Comparison of circadian variations using FEV₁ and peak expiratory flow rates among normal and asthmatic subjects. Thorax 1994;49,775-780[Abstract/Free Full Text]
21. Nickerson, BG, Lemen, RJ, Gerdes, CB, et al Within-subject variability and per cent change for significance of spirometry in normal subjects and in patients with cystic fibrosis. Am Rev Respir Dis 1980;122,859-866[ISI][Medline]

This article has been cited by other articles:

				M Burkhardt, E Mayordomo, K-J Winzer, F Fritzsche, T Gansukh, S Pahl, W Weichert, C Denkert, H Guski, M Dietel, et al. Cytoplasmic overexpression of ALCAM is prognostic of disease progression in breast cancer J. Clin. Pathol., April 1, 2006; 59(4): 403 - 409. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stanbrook, M. B. Articles by Tullis, D. E. Search for Related Content PubMed PubMed Citation Articles by Stanbrook, M. B. Articles by Tullis, D. E.