- Email: [email protected]
Peak Expiratory Flow Is Not a Quality Indicator for Spirometry
Original Research PULMONARY FUNCTION TESTING
Peak Expiratory Flow Is Not a Quality Indicator for Spirometry* Peak Expiratory Flow Variability and FEV1 Are Poorly Correlated in an Elderly Population Matthew J. Hegewald, MD, FCCP; Michael J. Lefor, MD; Robert L. Jensen, PhD; Robert O. Crapo, MD, FCCP; Stephen B. Kritchevsky, PhD; Catherine L. Haggerty, PhD, MPH; Douglas C. Bauer, MD; Suzanne Satterfield, MD; and Tamara Harris, MD; for the Health, Aging, and Body Composition Study Investigators
Background: Peak forced expiratory flow (PEF) and FEV1 are spirometry measures used in diagnosing and monitoring lung diseases. We tested the premise that within-test variability in PEF is associated with corresponding variability in FEV1 during a single test session. Methods: A total of 2,464 healthy adults from the Health, Aging, and Body Composition Study whose spirometry results met American Thoracic Society acceptability criteria were screened and analyzed. The three “best” test results (highest sum of FVC and FEV1) were selected for each subject. For those with acceptable spirometry results, two groups were created: group 1, normal FEV1/FVC ratio; group 2, reduced FEV1/FVC ratio. For each subject, the difference between the highest and lowest PEF (⌬PEF) and the associated difference between the highest and lowest FEV1 (⌬FEV1) were calculated. Regression analysis was performed using the largest PEF and best FEV1, and the percentage of ⌬PEF (%⌬PEF) and percentage of ⌬FEV1 (%⌬FEV1) were calculated in both groups. Results: Regression analysis for group 1 and group 2 showed an insignificant association between %⌬PEF and %⌬FEV1 (r2 ⴝ 0.0001, p ⴝ 0.59, and r2 ⴝ 0.040, p ⴝ 0.15, respectively). For both groups, a 29% ⌬PEF was associated with a 1% ⌬FEV1. Conclusion: Within a single spirometry test session, %⌬PEF and %⌬FEV1 contain independent information. PEF has a higher degree of intrinsic variability than FEV1. Changes in PEF do not have a significant effect on FEV1. Spirometry maneuvers should not be excluded based on peak flow variability. (CHEST 2007; 131:1494 –1499) Key words: forced expiratory flow rate; forced expiratory volume; peak expiratory flow; respiratory function tests; spirometry Abbreviations: ATS ⫽ American Thoracic Society; ERS ⫽ European Respiratory Society; ⌬FEV1 ⫽ difference between the highest and lowest FEV1; %⌬FEV1 ⫽ percentage of ⌬FEV1; FEV1-A ⫽ FEV1 associated with the largest peak expiratory flow; FEV1-B ⫽ FEV1 associated with the smallest peak expiratory flow; PEF ⫽ peak expiratory flow; ⌬PEF ⫽ difference between the highest and lowest peak expiratory flow; %⌬PEF ⫽ percentage of difference in peak expiratory flow; PEF-A ⫽ largest peak expiratory flow
expiratory flow (PEF) is a measure of maximal P eak expiratory flow that is used to assess qualitative and quantitative effort in spirometry maneuvers and is clinically utilized independently for asthma monitoring via handheld devices.1–5 FEV1 is a measurement of volume in the first second of a spirometry maneuver that is used for the diagnosis and monitoring of lung disease.1,6 Both of these measurements have played an 1494
important role in the identification and management of lung disease, particularly asthma. Physiologically, flow characteristics influence measurements of both PEF and FEV1. Although the viscosity and density of the gas measured, and the length and caliber of the airways impact change in PEF and FEV1 measurements,7–9 PEF and FEV1 measure different aspects of flow. PEF is thought to Original Research
be a measurement of large-caliber airway function (⬎ 2 mm diameter) and is very effort dependent. FEV1, however, is thought to be a reflection of intermediate and smaller airways. This measurement has both effort-dependent and effort-independent components. Effort during spirometry is, in part, judged by the individual’s PEF. It directly correlates to maximal work and the initial effort during a spirometry maneuver.10 It is also easily quantifiable and can be incorporated in automatic defaults on spirometers that use computer-assisted markers for spirometry acceptability standards. Prior guidelines11 state that individual PEF measurements should be within 10% of the maximal value. Some popular spirometers provide an error code if there are no trials within 10% of the “best” (largest) trial for PEF. As a result, PEF reproducibility has been used as a measure of quality assurance for spirometry. Despite this, the most recent American Thoracic Society (ATS)/European Respiratory Society (ERS) criteria for standardization of spirometry do not use differences in PEF between maneuvers to assess quality within a single session.12 PEF and FEV1 are used to objectively monitor obstructive lung disease and to evaluate occupational asthma, and are often used as primary outcomes in drug studies.1,13–16 FEV1 is commonly assumed to be partly dependent on PEF, based on a high correlation between PEF and FEV1.17 Hence, PEF has been used as a surrogate for FEV1, particularly within an individual over time (ie, change in PEF reflects a similar degree of change in FEV1). There is debate about whether or not changes in PEF truly reflect changes in FEV1 and subsequently correspond to the degree of obstructive disease in an individual.18,19 It has also been suggested that there is a negative effort dependence, also referred to as *From the Division of Pulmonary and Critical Care Medicine (Drs. Hegewald, Lefor, Jensen, and Crapo) LDS Hospital and University of Utah, Salt Lake City, UT; Wake Forest University (Dr. Kritchevsky), Winston Salem, NC; Department of Epidemiology (Dr. Haggerty), University of Pittsburgh, Pittsburgh, PA; University of California San Francisco (Dr. Bauer), San Francisco, CA; University of Tennessee Memphis (Dr. Satterfield), Memphis, TN; and National Institutes of Health (Dr. Harris), Bethesda, MD. This study was supported by contracts N01-AG-6 –2101, N01AG-6 –2103, and N01-AG-6 –2106, and was also supported in part by the Intramural Research program of the National Institutes of Health, National Institute on Aging. The authors have no conflicts of interest to disclose. Manuscript received November 15, 2006; revision accepted January 21, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Matthew Hegewald, MD, FCCP, Pulmonary Division, LDS Hospital, Eighth Ave & C St, Salt Lake City, UT 84143; e-mail: [email protected] DOI: 10.1378/chest.06-2707 www.chestjournal.org
inverse effort dependence, of the FEV1.10,20 This states that maximal effort corresponding to the highest PEF will result in a reduced FEV1 due to thoracic gas compression. In an attempt to clarify these issues, we tested the premise that difference between the highest and lowest PEF (⌬PEF) within an individual during a single session is associated with a parallel difference between the highest and lowest FEV1 (⌬FEV1). Materials and Methods Participants from the Health, Aging, and Body Composition Study were analyzed. All participants were 70 to 79 years old during recruitment, free of disability in activities of daily living, and free of functional limitations. The institutional review boards at both field centers approved the study, and informed consent was obtained. Subjects performed spirometry and were coached to perform maximal efforts. A National Institute for Occupational Safety and Health volume-based spirometer using a digital shaft encoder to measure volume displacement was used. Three-liter syringe calibrations were done daily. Two of the authors (R.L.J. and R.O.C.) from LDS Hospital in Salt Lake City, UT, scored the quality of the spirograms as “A” (best) through “F” (worst) for FEV1 and FVC based on ATS acceptability and reproducibility standards. Spirograms with FEV1 and FVC quality scores of “C” or better were then analyzed. All of these met ATS criteria published in 1995 for reproducibility, with 200 mL between the highest and the next highest FEV1.21 Of those that were acceptable, two groups were formed: group 1, normal FEV1/FVC ratio; group 2, reduced FEV1/FVC ratio, based on the lower limits of normal using prediction equations of Crapo et al.22 For each group, the three best tests (based on the highest sum of FVC and FEV1) were selected for each subject as recommended by ATS spirometry guidelines.21 The largest PEF (PEF-A) and the smallest PEF in a single session were chosen from those three best tests. FEV1 values associated with each PEF were labeled as FEV1-A and FEV1-B, respectively. Equations associated with these values are as follows: Equation 1: ⌬PEF ⫽ PEF-A ⫺ PEF-B; all ⌬PEF values were positive. Equation 2: ⌬FEV1 ⫽ FEV1-A ⫺ FEV1-B;⌬FEV1 ; values could be either positive or negative. Equation 3: %⌬PEF ⫽ (⌬PEF/PEF-A) ⫻ 100. Equation 4: %⌬FEV1 ⫽ (⌬FEV1/largest FEV1) ⫻ 100, where PEF-B is the smallest PEF in a single session. Regression analysis was performed on PEF-A and the largest FEV1, and %⌬PEF and %⌬FEV1 to look for significant relationships between these variables in both normal and obstructed individuals. The frequency of negative effort dependency was determined by calculating the percentage of subjects in which the largest FEV1 was associated with a submaximal PEF. Those subjects with acceptable spirometry results based on ATS acceptability and reproducibility criteria, and a ⬎ 50% ⌬PEF were excluded from analysis to reduce the effect of outliers. This resulted in exclusion of 1.9% of subjects. CHEST / 131 / 5 / MAY, 2007
1495
Results
Discussion
Of the 3,075 participants in the Health, Aging, and Body Composition Study, 2,863 subjects performed the spirometric evaluation; 352 subjects were excluded because they did not meet ATS acceptability and reproducibility criteria (12.3% of total). Fortyseven subjects were excluded based on a 50% ⌬PEF (1.9% of total). Data from 2,464 subjects were analyzed. The mean age of participants was 73.6 ⫾ 2.86 years (⫾ SD). Gender distribution was 49% male and 51% female. Fifty-nine percent of the subjects were white, and the remainder were African American (41%); 10.4% of participants reported being active smokers, 45.7% reported being former smokers, and 43.9% indicated that they never smoked. Group 1 comprised 2,064 subjects with a normal FEV1/FVC ratio. Group 2 had 400 subjects with a reduced FEV1/FVC ratio. Mean ages of group 1 and group 2 subjects were similar: 73.6 ⫾ 2.86 years and 73.5 ⫾ 2.88 years, respectively. Mean values of each group for best PEF, best FEV1, best FVC, ⌬PEF, ⌬FEV1, %⌬PEF, and %⌬FEV1 are contained in Table 1. All data appear to be distributed normally. Correlations between PEF and FEV1 are shown in Figures 1– 4. In both groups, the PEF-A and the largest FEV1 were statistically significantly correlated (group 1: r2 ⫽ 0.70, p ⬍ 0.001 [Fig 1]; group 2: r2 ⫽ 0.79, p ⬍ 0.0001 [Fig 3]). %⌬PEF and %⌬FEV1 were not correlated in either group (group 1: r2 ⫽ 0.0001, p ⫽ 0.59 [Fig 2]; group 2: r2 ⫽ 0.04, p ⫽ 0.15 [Fig 4]). In group 1, the %⌬PEF explains ⬍ 0.3% of the variability in %⌬FEV1 in normal patients with acceptable spirometry results (Fig 2). Among all participants, we calculated that on the average a 29% ⌬PEF was associated with a 1% ⌬FEV1. Mean %⌬PEF was 14.3, and mean %⌬FEV1 was 0.49. In 39.4% of all subjects, the PEF-A was associated with a less than maximal FEV1. This relationship did not change when the %⌬PEF was limited to 10% (those with the most reproducible PEF values).
Our study demonstrates that in healthy older subjects within a single test session, there is a poor correlation between PEF variability and FEV1 variability. An average 29% ⌬PEF has an associated average ⌬FEV1 of 1%. FEV1 is a stable measurement even with large changes in PEF. This suggests that FEV1 and PEF measure different aspects of lung function. This is not unexpected because PEF occurs in the first 100 to 200 ms of an expiratory effort and is considered a measure of large airway function, whereas FEV1 measures the entire first second of exhalation and is influenced by both large and small airways function. These data suggest that current ATS/ERS acceptability criteria that emphasize reproducibility of FEV1 and FVC for determining acceptable spirometric results, and do not include PEF reproducibility, are appropriate.12 These findings challenge the utility of using PEF as a measure of quality in spirometry and suggest that spirometry maneuvers should not be excluded based on PEF variability. The primary limitation of this study is that it included only elderly adults and therefore may not be applicable to the general population. The value of using PEF to assess quality in spirometry has been debated. Several studies have tried to address this issue in various ways. Krowka et al10 examined PEF and its effect on FEV1 with maximal and submaximal effort. After demonstrating that maximal work was related to the highest PEF, they discovered that the highest PEF was not associated with the largest FEV1. This concept, called negative effort dependence, was attributed to thoracic gas compression. Krowka et al10 proposed that PEF should be used as an objective, reproducible parameter for individual effort during spirometry maneuvers because this will result in less variability in FEV1. Our results question the importance of negative effort dependence of FEV1. In our data, increasing PEF (and thus increasing effort) resulted in an increase in FEV1 in 60% of subjects. The majority of subjects exhibited positive effort dependence for FEV1. Regardless, ⌬PEF and thus initial effort results in minimal ⌬FEV1. Park23 examined spirometry results from 10 “normal” and 12 “obstructed” patients. The study compared acceptable tracings with the highest PEF (largest effort) vs those with the highest FVC (largest lung volume). In tracings with the largest sum of FEV1 and FVC, the FEV1 was little affected by PEF variability. Park concluded that the ATS guidelines21 for acceptability and reproducibility were appropriate and should not include PEF variability. Our data confirm the conclusions of Park.23 FEV1 has small intraindividual change during a particular session, whereas PEF variability is significantly
Table 1—Demographic and Spirometry Values by Group* Variables
Group 1 (n ⫽ 2,291)
Group 2 (n ⫽ 410)
Age, yr Best PEF, mL/s Best FEV1, mL Best FVC, mL ⌬PEF, mL/s ⌬FEV1, mL %⌬PEF %⌬FEV1
73.6 ⫾ 2.86 5,666 ⫾ 1,850 2,234 ⫾ 615.5 2,905 ⫾ 803.9 889.0 ⫾ 765.1 14.21 ⫾ 104.2 15.82 ⫾ 13.2 0.662 ⫾ 4.92
73.5 ⫾ 2.88 4,183 ⫾ 1,703 1,711 ⫾ 597.1 2,811 ⫾ 828.4 577.9 ⫾ 531.2 13.72 ⫾ 70.28 13.93 ⫾ 10.85 0.899 ⫾ 5.05
*Data are presented as mean ⫾ 1 SD. 1496
Original Research
Figure 1. Regression analysis between best FEV1 and best PEF in normal subjects with acceptable curves.
greater. For all individuals studied, there is an average ⌬PEF of 14.3%. with a ⬍ 0.50% average ⌬FEV1. Other researchers24 have confirmed the significantly greater variability in PEF compared with FEV1 in a single session. Hence, PEF may be unreliable to accurately assess the quality of a spirometry maneuver despite the fact that it is the most appropriate way to assess initial effort. This suggests that FEV1 is not very dependent on the initial effort. Our finding that PEF and FEV1 are significantly
and strongly correlated is in agreement with prior work.17 However, our study uniquely examined the relationship between PEF and FEV1 measurements within a single session, and it was the first to demonstrate that differences in these parameters in repeated measurements during the same testing sessions were not significantly correlated. However, maximal initial efforts should be encouraged because current reference standards were obtained using maximal efforts and all sources of increased variability should be minimized.
Figure 2. Regression analysis of %⌬FEV1 against %⌬PEF for normal subjects with acceptable curves. www.chestjournal.org
CHEST / 131 / 5 / MAY, 2007
1497
Figure 3. Regression analysis between best FEV1 and best PEF in obstructed subjects with acceptable curves.
In conclusion, both laboratory and clinical decisions are being made based on spirometry measurements using PEF and FEV1. The changes in these two parameters give different information. Variability in PEF measurements during a single session does not have a significant effect on FEV1. Spirometry data may be discarded erroneously based on PEF variability that has minimal overall effect on other results of the test, specifically FEV1. Using PEF variability as a test quality
criterion could increase the number of tests required in a single testing session, unnecessarily increasing the test burden for the patient and laboratory. These results are in agreement with the most recent ATS/ERS standardization of spirometry guidelines,12 published in 2005, that state that the two largest FEV1 and FVC values must be within 0.150 L of each other and make no mention of PEF reproducibility. Also, given that PEF has a higher degree of intrinsic variability than FEV1,
Figure 4. Regression analysis of %⌬FEV1 against %⌬PEF for obstructed subjects with acceptable curves. 1498
Original Research
one must be more cautious in making clinical changes in management based on PEF. 12
References 1 Expert Panel Report 2: guidelines for the diagnosis and management of asthma. Bethesda, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Heart, Lung, and Blood Institute, 1997 2 Practice parameters for the diagnosis and treatment of asthma. American Academy of Allergy, Asthma and Immunology/American College of Allergy, Asthma and Immunology/Joint Council of Allergy, Asthma, and Immunology. J Allergy Clin Immunol 1995; 96:707– 870 3 Boulet LP, Becker A, Berube D, et al. Canadian asthma consensus report, 1999: Canadian Asthma Consensus Group. Can Med Assoc J 1999; 161:S1–S61 4 British Thoracic Society. The British guidelines on asthma management: 1995 review and position statement. Thorax 1997; 52:S1–21 5 Gibson PG. Monitoring the patient with asthma: an evidencebased approach. J Allergy Clin Immunol 2000; 106:17–26 6 Pauwels RA, Buist AS, Calverley PM, et al. The GOLD Scientific Committee global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease NHLBI/WHO: Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med 2001; 163:1256 –1276Crit Care Med 2001; 163:1256 –1276 7 Robinson DR, Chaudhary BA, Speir WA. Expiratory flow limitation in large and small airways. Arch Intern Med 1984; 144:1457–1460 8 Dolyniuk MV, Fahey PJ. Relationship of tracheal size to maximal expiratory airflow and density dependence. J Appl Physiol 1986; 60:501–505 9 Osmanliev D, Bowley N, Hunter DM, et al. Relation between tracheal size and forced expiratory volume in 1 second in young men. Am Rev Respir Dis 1982; 126:179 –182 10 Krowka MJ, Enright PL, Rodarte JR, et al. Effect of effort on measurement of forced expiratory volume in one second. Am Rev Respir Dis 1987; 136:829 – 833 11 Quanjer P, Tammeling G, Cotes J, et al. Lung volumes and
www.chestjournal.org
13
14
15
16 17
18
19
20 21 22 23 24
forced ventilatory flows: report Working Party Standardization of Lung Function Tests European Community for Steel and Coal. Eur Respir J 1993; 16(suppl):5– 40 Miller M, Hanjinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J 2005; 26:319 –338 McFadden ER, elSanadi N, Strauss L, et al. The influence of parasympatholytics on the resolution of acute attacks of asthma. Am J Med 1997; 102:7–13 Casaburi R, Briggs DD Jr, Donohue JF, et al. The spirometric efficacy of once-daily dosing with tiotropium in stable COPD: a 13-week multicenter trial; the US Tiotropium Study Group. Chest 2000; 118:1294 –1302 Burge PS, Pantin CF, Newton DT, et al. Development of an expert system for the interpretation of serial peak expiratory flow measurements in the diagnosis of occupational asthma: Midlands Thoracic Society Research Group. Occup Environ Med 1999; 56:758 –764 Tan RA, Spector SL. Diagnostic testing in occupational asthma. Ann Allergy Asthma Immunol 1999; 83:587–592 Rosenblatt G, Alkalay I, McCann PD, et al. The correlation of peak flow rate with maximal expiratory flow rate, one-second forced expiratory volume, and maximal breathing capacity. Am Rev Respir Dis 1963; 87:589 –591 Meltzer AA, Smolensky MH, D’Alonzo GE, et al. An assessment of peak expiratory flow as a surrogate measurement of FEV1 in stable asthmatic children. Chest 1989; 96:329 –333 Gautrin D, D’Aquino LC, Gagnon G, et al. Comparison between peak expiratory flow rates (PEFR) and FEV1 in the monitoring of asthmatic subjects at an outpatient clinic. Chest 1994; 106:1419 –1426 Coates A, Desmond K, Demizio D, et al. Sources of variability in FEV1. Am J Respir Crit Care Med 1994; 149:439 – 443 American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med 1995; 152:1107– 1136 Crapo R, Morris A, Gardner R. Reference spirometric values using techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 1981; 123:659 – 664 Park SS. Effect of effort versus volume on forced expiratory flow measurement. Am Rev Respir Dis 1988 138:1002–1005 Enright P, Beck K, Sherrrill D. Repeatability of spirometry in 18,000 adult patients. Am J Respir Crit Care Med 2004; 169:235–238
CHEST / 131 / 5 / MAY, 2007
1499
Peak Expiratory Flow Is Not a Quality Indicator for Spirometry* Peak Expiratory Flow Variability and FEV1 Are Poorly Correlated in an Elderly Population Matthew J. Hegewald, MD, FCCP; Michael J. Lefor, MD; Robert L. Jensen, PhD; Robert O. Crapo, MD, FCCP; Stephen B. Kritchevsky, PhD; Catherine L. Haggerty, PhD, MPH; Douglas C. Bauer, MD; Suzanne Satterfield, MD; and Tamara Harris, MD; for the Health, Aging, and Body Composition Study Investigators
Background: Peak forced expiratory flow (PEF) and FEV1 are spirometry measures used in diagnosing and monitoring lung diseases. We tested the premise that within-test variability in PEF is associated with corresponding variability in FEV1 during a single test session. Methods: A total of 2,464 healthy adults from the Health, Aging, and Body Composition Study whose spirometry results met American Thoracic Society acceptability criteria were screened and analyzed. The three “best” test results (highest sum of FVC and FEV1) were selected for each subject. For those with acceptable spirometry results, two groups were created: group 1, normal FEV1/FVC ratio; group 2, reduced FEV1/FVC ratio. For each subject, the difference between the highest and lowest PEF (⌬PEF) and the associated difference between the highest and lowest FEV1 (⌬FEV1) were calculated. Regression analysis was performed using the largest PEF and best FEV1, and the percentage of ⌬PEF (%⌬PEF) and percentage of ⌬FEV1 (%⌬FEV1) were calculated in both groups. Results: Regression analysis for group 1 and group 2 showed an insignificant association between %⌬PEF and %⌬FEV1 (r2 ⴝ 0.0001, p ⴝ 0.59, and r2 ⴝ 0.040, p ⴝ 0.15, respectively). For both groups, a 29% ⌬PEF was associated with a 1% ⌬FEV1. Conclusion: Within a single spirometry test session, %⌬PEF and %⌬FEV1 contain independent information. PEF has a higher degree of intrinsic variability than FEV1. Changes in PEF do not have a significant effect on FEV1. Spirometry maneuvers should not be excluded based on peak flow variability. (CHEST 2007; 131:1494 –1499) Key words: forced expiratory flow rate; forced expiratory volume; peak expiratory flow; respiratory function tests; spirometry Abbreviations: ATS ⫽ American Thoracic Society; ERS ⫽ European Respiratory Society; ⌬FEV1 ⫽ difference between the highest and lowest FEV1; %⌬FEV1 ⫽ percentage of ⌬FEV1; FEV1-A ⫽ FEV1 associated with the largest peak expiratory flow; FEV1-B ⫽ FEV1 associated with the smallest peak expiratory flow; PEF ⫽ peak expiratory flow; ⌬PEF ⫽ difference between the highest and lowest peak expiratory flow; %⌬PEF ⫽ percentage of difference in peak expiratory flow; PEF-A ⫽ largest peak expiratory flow
expiratory flow (PEF) is a measure of maximal P eak expiratory flow that is used to assess qualitative and quantitative effort in spirometry maneuvers and is clinically utilized independently for asthma monitoring via handheld devices.1–5 FEV1 is a measurement of volume in the first second of a spirometry maneuver that is used for the diagnosis and monitoring of lung disease.1,6 Both of these measurements have played an 1494
important role in the identification and management of lung disease, particularly asthma. Physiologically, flow characteristics influence measurements of both PEF and FEV1. Although the viscosity and density of the gas measured, and the length and caliber of the airways impact change in PEF and FEV1 measurements,7–9 PEF and FEV1 measure different aspects of flow. PEF is thought to Original Research
be a measurement of large-caliber airway function (⬎ 2 mm diameter) and is very effort dependent. FEV1, however, is thought to be a reflection of intermediate and smaller airways. This measurement has both effort-dependent and effort-independent components. Effort during spirometry is, in part, judged by the individual’s PEF. It directly correlates to maximal work and the initial effort during a spirometry maneuver.10 It is also easily quantifiable and can be incorporated in automatic defaults on spirometers that use computer-assisted markers for spirometry acceptability standards. Prior guidelines11 state that individual PEF measurements should be within 10% of the maximal value. Some popular spirometers provide an error code if there are no trials within 10% of the “best” (largest) trial for PEF. As a result, PEF reproducibility has been used as a measure of quality assurance for spirometry. Despite this, the most recent American Thoracic Society (ATS)/European Respiratory Society (ERS) criteria for standardization of spirometry do not use differences in PEF between maneuvers to assess quality within a single session.12 PEF and FEV1 are used to objectively monitor obstructive lung disease and to evaluate occupational asthma, and are often used as primary outcomes in drug studies.1,13–16 FEV1 is commonly assumed to be partly dependent on PEF, based on a high correlation between PEF and FEV1.17 Hence, PEF has been used as a surrogate for FEV1, particularly within an individual over time (ie, change in PEF reflects a similar degree of change in FEV1). There is debate about whether or not changes in PEF truly reflect changes in FEV1 and subsequently correspond to the degree of obstructive disease in an individual.18,19 It has also been suggested that there is a negative effort dependence, also referred to as *From the Division of Pulmonary and Critical Care Medicine (Drs. Hegewald, Lefor, Jensen, and Crapo) LDS Hospital and University of Utah, Salt Lake City, UT; Wake Forest University (Dr. Kritchevsky), Winston Salem, NC; Department of Epidemiology (Dr. Haggerty), University of Pittsburgh, Pittsburgh, PA; University of California San Francisco (Dr. Bauer), San Francisco, CA; University of Tennessee Memphis (Dr. Satterfield), Memphis, TN; and National Institutes of Health (Dr. Harris), Bethesda, MD. This study was supported by contracts N01-AG-6 –2101, N01AG-6 –2103, and N01-AG-6 –2106, and was also supported in part by the Intramural Research program of the National Institutes of Health, National Institute on Aging. The authors have no conflicts of interest to disclose. Manuscript received November 15, 2006; revision accepted January 21, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Matthew Hegewald, MD, FCCP, Pulmonary Division, LDS Hospital, Eighth Ave & C St, Salt Lake City, UT 84143; e-mail: [email protected] DOI: 10.1378/chest.06-2707 www.chestjournal.org
inverse effort dependence, of the FEV1.10,20 This states that maximal effort corresponding to the highest PEF will result in a reduced FEV1 due to thoracic gas compression. In an attempt to clarify these issues, we tested the premise that difference between the highest and lowest PEF (⌬PEF) within an individual during a single session is associated with a parallel difference between the highest and lowest FEV1 (⌬FEV1). Materials and Methods Participants from the Health, Aging, and Body Composition Study were analyzed. All participants were 70 to 79 years old during recruitment, free of disability in activities of daily living, and free of functional limitations. The institutional review boards at both field centers approved the study, and informed consent was obtained. Subjects performed spirometry and were coached to perform maximal efforts. A National Institute for Occupational Safety and Health volume-based spirometer using a digital shaft encoder to measure volume displacement was used. Three-liter syringe calibrations were done daily. Two of the authors (R.L.J. and R.O.C.) from LDS Hospital in Salt Lake City, UT, scored the quality of the spirograms as “A” (best) through “F” (worst) for FEV1 and FVC based on ATS acceptability and reproducibility standards. Spirograms with FEV1 and FVC quality scores of “C” or better were then analyzed. All of these met ATS criteria published in 1995 for reproducibility, with 200 mL between the highest and the next highest FEV1.21 Of those that were acceptable, two groups were formed: group 1, normal FEV1/FVC ratio; group 2, reduced FEV1/FVC ratio, based on the lower limits of normal using prediction equations of Crapo et al.22 For each group, the three best tests (based on the highest sum of FVC and FEV1) were selected for each subject as recommended by ATS spirometry guidelines.21 The largest PEF (PEF-A) and the smallest PEF in a single session were chosen from those three best tests. FEV1 values associated with each PEF were labeled as FEV1-A and FEV1-B, respectively. Equations associated with these values are as follows: Equation 1: ⌬PEF ⫽ PEF-A ⫺ PEF-B; all ⌬PEF values were positive. Equation 2: ⌬FEV1 ⫽ FEV1-A ⫺ FEV1-B;⌬FEV1 ; values could be either positive or negative. Equation 3: %⌬PEF ⫽ (⌬PEF/PEF-A) ⫻ 100. Equation 4: %⌬FEV1 ⫽ (⌬FEV1/largest FEV1) ⫻ 100, where PEF-B is the smallest PEF in a single session. Regression analysis was performed on PEF-A and the largest FEV1, and %⌬PEF and %⌬FEV1 to look for significant relationships between these variables in both normal and obstructed individuals. The frequency of negative effort dependency was determined by calculating the percentage of subjects in which the largest FEV1 was associated with a submaximal PEF. Those subjects with acceptable spirometry results based on ATS acceptability and reproducibility criteria, and a ⬎ 50% ⌬PEF were excluded from analysis to reduce the effect of outliers. This resulted in exclusion of 1.9% of subjects. CHEST / 131 / 5 / MAY, 2007
1495
Results
Discussion
Of the 3,075 participants in the Health, Aging, and Body Composition Study, 2,863 subjects performed the spirometric evaluation; 352 subjects were excluded because they did not meet ATS acceptability and reproducibility criteria (12.3% of total). Fortyseven subjects were excluded based on a 50% ⌬PEF (1.9% of total). Data from 2,464 subjects were analyzed. The mean age of participants was 73.6 ⫾ 2.86 years (⫾ SD). Gender distribution was 49% male and 51% female. Fifty-nine percent of the subjects were white, and the remainder were African American (41%); 10.4% of participants reported being active smokers, 45.7% reported being former smokers, and 43.9% indicated that they never smoked. Group 1 comprised 2,064 subjects with a normal FEV1/FVC ratio. Group 2 had 400 subjects with a reduced FEV1/FVC ratio. Mean ages of group 1 and group 2 subjects were similar: 73.6 ⫾ 2.86 years and 73.5 ⫾ 2.88 years, respectively. Mean values of each group for best PEF, best FEV1, best FVC, ⌬PEF, ⌬FEV1, %⌬PEF, and %⌬FEV1 are contained in Table 1. All data appear to be distributed normally. Correlations between PEF and FEV1 are shown in Figures 1– 4. In both groups, the PEF-A and the largest FEV1 were statistically significantly correlated (group 1: r2 ⫽ 0.70, p ⬍ 0.001 [Fig 1]; group 2: r2 ⫽ 0.79, p ⬍ 0.0001 [Fig 3]). %⌬PEF and %⌬FEV1 were not correlated in either group (group 1: r2 ⫽ 0.0001, p ⫽ 0.59 [Fig 2]; group 2: r2 ⫽ 0.04, p ⫽ 0.15 [Fig 4]). In group 1, the %⌬PEF explains ⬍ 0.3% of the variability in %⌬FEV1 in normal patients with acceptable spirometry results (Fig 2). Among all participants, we calculated that on the average a 29% ⌬PEF was associated with a 1% ⌬FEV1. Mean %⌬PEF was 14.3, and mean %⌬FEV1 was 0.49. In 39.4% of all subjects, the PEF-A was associated with a less than maximal FEV1. This relationship did not change when the %⌬PEF was limited to 10% (those with the most reproducible PEF values).
Our study demonstrates that in healthy older subjects within a single test session, there is a poor correlation between PEF variability and FEV1 variability. An average 29% ⌬PEF has an associated average ⌬FEV1 of 1%. FEV1 is a stable measurement even with large changes in PEF. This suggests that FEV1 and PEF measure different aspects of lung function. This is not unexpected because PEF occurs in the first 100 to 200 ms of an expiratory effort and is considered a measure of large airway function, whereas FEV1 measures the entire first second of exhalation and is influenced by both large and small airways function. These data suggest that current ATS/ERS acceptability criteria that emphasize reproducibility of FEV1 and FVC for determining acceptable spirometric results, and do not include PEF reproducibility, are appropriate.12 These findings challenge the utility of using PEF as a measure of quality in spirometry and suggest that spirometry maneuvers should not be excluded based on PEF variability. The primary limitation of this study is that it included only elderly adults and therefore may not be applicable to the general population. The value of using PEF to assess quality in spirometry has been debated. Several studies have tried to address this issue in various ways. Krowka et al10 examined PEF and its effect on FEV1 with maximal and submaximal effort. After demonstrating that maximal work was related to the highest PEF, they discovered that the highest PEF was not associated with the largest FEV1. This concept, called negative effort dependence, was attributed to thoracic gas compression. Krowka et al10 proposed that PEF should be used as an objective, reproducible parameter for individual effort during spirometry maneuvers because this will result in less variability in FEV1. Our results question the importance of negative effort dependence of FEV1. In our data, increasing PEF (and thus increasing effort) resulted in an increase in FEV1 in 60% of subjects. The majority of subjects exhibited positive effort dependence for FEV1. Regardless, ⌬PEF and thus initial effort results in minimal ⌬FEV1. Park23 examined spirometry results from 10 “normal” and 12 “obstructed” patients. The study compared acceptable tracings with the highest PEF (largest effort) vs those with the highest FVC (largest lung volume). In tracings with the largest sum of FEV1 and FVC, the FEV1 was little affected by PEF variability. Park concluded that the ATS guidelines21 for acceptability and reproducibility were appropriate and should not include PEF variability. Our data confirm the conclusions of Park.23 FEV1 has small intraindividual change during a particular session, whereas PEF variability is significantly
Table 1—Demographic and Spirometry Values by Group* Variables
Group 1 (n ⫽ 2,291)
Group 2 (n ⫽ 410)
Age, yr Best PEF, mL/s Best FEV1, mL Best FVC, mL ⌬PEF, mL/s ⌬FEV1, mL %⌬PEF %⌬FEV1
73.6 ⫾ 2.86 5,666 ⫾ 1,850 2,234 ⫾ 615.5 2,905 ⫾ 803.9 889.0 ⫾ 765.1 14.21 ⫾ 104.2 15.82 ⫾ 13.2 0.662 ⫾ 4.92
73.5 ⫾ 2.88 4,183 ⫾ 1,703 1,711 ⫾ 597.1 2,811 ⫾ 828.4 577.9 ⫾ 531.2 13.72 ⫾ 70.28 13.93 ⫾ 10.85 0.899 ⫾ 5.05
*Data are presented as mean ⫾ 1 SD. 1496
Original Research
Figure 1. Regression analysis between best FEV1 and best PEF in normal subjects with acceptable curves.
greater. For all individuals studied, there is an average ⌬PEF of 14.3%. with a ⬍ 0.50% average ⌬FEV1. Other researchers24 have confirmed the significantly greater variability in PEF compared with FEV1 in a single session. Hence, PEF may be unreliable to accurately assess the quality of a spirometry maneuver despite the fact that it is the most appropriate way to assess initial effort. This suggests that FEV1 is not very dependent on the initial effort. Our finding that PEF and FEV1 are significantly
and strongly correlated is in agreement with prior work.17 However, our study uniquely examined the relationship between PEF and FEV1 measurements within a single session, and it was the first to demonstrate that differences in these parameters in repeated measurements during the same testing sessions were not significantly correlated. However, maximal initial efforts should be encouraged because current reference standards were obtained using maximal efforts and all sources of increased variability should be minimized.
Figure 2. Regression analysis of %⌬FEV1 against %⌬PEF for normal subjects with acceptable curves. www.chestjournal.org
CHEST / 131 / 5 / MAY, 2007
1497
Figure 3. Regression analysis between best FEV1 and best PEF in obstructed subjects with acceptable curves.
In conclusion, both laboratory and clinical decisions are being made based on spirometry measurements using PEF and FEV1. The changes in these two parameters give different information. Variability in PEF measurements during a single session does not have a significant effect on FEV1. Spirometry data may be discarded erroneously based on PEF variability that has minimal overall effect on other results of the test, specifically FEV1. Using PEF variability as a test quality
criterion could increase the number of tests required in a single testing session, unnecessarily increasing the test burden for the patient and laboratory. These results are in agreement with the most recent ATS/ERS standardization of spirometry guidelines,12 published in 2005, that state that the two largest FEV1 and FVC values must be within 0.150 L of each other and make no mention of PEF reproducibility. Also, given that PEF has a higher degree of intrinsic variability than FEV1,
Figure 4. Regression analysis of %⌬FEV1 against %⌬PEF for obstructed subjects with acceptable curves. 1498
Original Research
one must be more cautious in making clinical changes in management based on PEF. 12
References 1 Expert Panel Report 2: guidelines for the diagnosis and management of asthma. Bethesda, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Heart, Lung, and Blood Institute, 1997 2 Practice parameters for the diagnosis and treatment of asthma. American Academy of Allergy, Asthma and Immunology/American College of Allergy, Asthma and Immunology/Joint Council of Allergy, Asthma, and Immunology. J Allergy Clin Immunol 1995; 96:707– 870 3 Boulet LP, Becker A, Berube D, et al. Canadian asthma consensus report, 1999: Canadian Asthma Consensus Group. Can Med Assoc J 1999; 161:S1–S61 4 British Thoracic Society. The British guidelines on asthma management: 1995 review and position statement. Thorax 1997; 52:S1–21 5 Gibson PG. Monitoring the patient with asthma: an evidencebased approach. J Allergy Clin Immunol 2000; 106:17–26 6 Pauwels RA, Buist AS, Calverley PM, et al. The GOLD Scientific Committee global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease NHLBI/WHO: Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med 2001; 163:1256 –1276Crit Care Med 2001; 163:1256 –1276 7 Robinson DR, Chaudhary BA, Speir WA. Expiratory flow limitation in large and small airways. Arch Intern Med 1984; 144:1457–1460 8 Dolyniuk MV, Fahey PJ. Relationship of tracheal size to maximal expiratory airflow and density dependence. J Appl Physiol 1986; 60:501–505 9 Osmanliev D, Bowley N, Hunter DM, et al. Relation between tracheal size and forced expiratory volume in 1 second in young men. Am Rev Respir Dis 1982; 126:179 –182 10 Krowka MJ, Enright PL, Rodarte JR, et al. Effect of effort on measurement of forced expiratory volume in one second. Am Rev Respir Dis 1987; 136:829 – 833 11 Quanjer P, Tammeling G, Cotes J, et al. Lung volumes and
www.chestjournal.org
13
14
15
16 17
18
19
20 21 22 23 24
forced ventilatory flows: report Working Party Standardization of Lung Function Tests European Community for Steel and Coal. Eur Respir J 1993; 16(suppl):5– 40 Miller M, Hanjinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J 2005; 26:319 –338 McFadden ER, elSanadi N, Strauss L, et al. The influence of parasympatholytics on the resolution of acute attacks of asthma. Am J Med 1997; 102:7–13 Casaburi R, Briggs DD Jr, Donohue JF, et al. The spirometric efficacy of once-daily dosing with tiotropium in stable COPD: a 13-week multicenter trial; the US Tiotropium Study Group. Chest 2000; 118:1294 –1302 Burge PS, Pantin CF, Newton DT, et al. Development of an expert system for the interpretation of serial peak expiratory flow measurements in the diagnosis of occupational asthma: Midlands Thoracic Society Research Group. Occup Environ Med 1999; 56:758 –764 Tan RA, Spector SL. Diagnostic testing in occupational asthma. Ann Allergy Asthma Immunol 1999; 83:587–592 Rosenblatt G, Alkalay I, McCann PD, et al. The correlation of peak flow rate with maximal expiratory flow rate, one-second forced expiratory volume, and maximal breathing capacity. Am Rev Respir Dis 1963; 87:589 –591 Meltzer AA, Smolensky MH, D’Alonzo GE, et al. An assessment of peak expiratory flow as a surrogate measurement of FEV1 in stable asthmatic children. Chest 1989; 96:329 –333 Gautrin D, D’Aquino LC, Gagnon G, et al. Comparison between peak expiratory flow rates (PEFR) and FEV1 in the monitoring of asthmatic subjects at an outpatient clinic. Chest 1994; 106:1419 –1426 Coates A, Desmond K, Demizio D, et al. Sources of variability in FEV1. Am J Respir Crit Care Med 1994; 149:439 – 443 American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med 1995; 152:1107– 1136 Crapo R, Morris A, Gardner R. Reference spirometric values using techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 1981; 123:659 – 664 Park SS. Effect of effort versus volume on forced expiratory flow measurement. Am Rev Respir Dis 1988 138:1002–1005 Enright P, Beck K, Sherrrill D. Repeatability of spirometry in 18,000 adult patients. Am J Respir Crit Care Med 2004; 169:235–238
CHEST / 131 / 5 / MAY, 2007
1499