Different experts follow different approaches to interpretation of pulmonary function tests. There is no universally accepted standard for interpretation, but the two most commonly cited standards have been the 1986 American Thoracic Society Disability Standard [1] and the 1991 statement of the American Thoracic Society [2]. In 2005, the American Thoracic Society and the European Respiratory Society updated the pulmonary function interpretation strategies [3].
This chapter describes three approaches. The first uses the flow-volume curve and the normal predicted values. The second uses the test data without the flow-volume curve. The third uses a pulmonary function test ''crib sheet'' developed in the Mayo Clinic Division of Pulmonary and Critical Care Medicine as an instructional tool for residents and fellows.
14A. FLOW-VOLUME CURVE AVAILABLE Step I
Examine the flow-volume curve and compare it with the normal predicted curve (see the Appendix for how to construct the normal curve). Is there any ventilatory limitation (that is, any loss of area)? If not, the test result is most likely normal.
1. Is the forced vital capacity (FVC) normal? If so, any significant restriction is essentially ruled out.
2. Is the FVC reduced? If so, either obstruction or restriction could be the cause (see Fig. 2-3, page 10).
FIG. 14-1. Normal flow-volume curve.
3. Examine the contour of the flow-volume curve.
a. Is it normal-appearing (Fig. 14-1)? If so, and if the FVC is normal, the test result is almost always normal. Proceed to steps V, VI, and VII. If the FVC is reduced and the flow-volume slope and ratio of forced expiratory volume in 1 second to FVC (FEV1/FVC ratio) are normal, restriction, occult asthma, or a nonspecific abnormality may be present (see section 2F, page 12, and section 3E, page 36). The total lung capacity (TLC) will have to be measured to make the differentiation.
b. Is the curve scooped out with reduced flow-volume slope and low flows (Fig. 14-2)? An obstructive defect is most likely. Remember the occasional mixed restrictive- obstructive disorder.
FIG. 14-2. Flow-volume curve in severe chronic obstructive pulmonary disease.
FIG. 14-3. Flow-volume curve in pulmonary fibrosis. Note steep slope and decreased volume.
4. Is the slope of the flow-volume curve increased (Fig. 14-3)? This finding is consistent with a pulmonary parenchymal restrictive process. The FVC, TLC, and diffusing capacity of carbon monoxide (Dlco) must be reduced to be certain. (Grading the degree of restriction is described in section 14C, page 139.)
5. If there is a flow-volume loop, is there any suggestion of a major airway lesion (Fig. 14-4)?
FIG. 14-4. Typical flow-volume curves associated with lesions of the major airway (carina to mouth). A. Typical variable extrathoracic lesion. B. Variable intrathoracic lesion. C. Fixed lesion. Expir, expiratory; Inspir, inspiratory.
Step II
Examine the FEV1 value.
1. Is it normal? If so, all but borderline obstruction or restriction is ruled out. There are exceptions, namely, the rare variable extrathoracic lesion in which the FEV1 can be normal but the maximal voluntary ventilation (MVV) is reduced because of inspiratory obstruction (as in Fig. 14-4A). Also, subjects with respiratory muscle weakness (see section 9D, page 97) can initially present with dyspnea and a normal FEV1.
2. Is the FEV1 reduced below the lower limit of normal? (The equation used to determine the lower limit of normal is in the Appendix.) If FEV1 is reduced, the decrease is most often due to airway obstruction. It could be caused by a restrictive process, however, and thus, the FEV1/FVC ratio needs to be evaluated. Nevertheless, if the TLC value is available, check it first. An increase in TLC by more than 15 to 20% favors obstructive disease. A normal or increased TLC value excludes a pulmonary restrictive process by definition. A normal TLC can occur in the rare mixed obstructive-restrictive disorder. A reduced TLC is expected with a restrictive process.
Step III
Examine the FEV1/FVC ratio.
1. If the absolute ratio is decreased to below the lower limit of normal, an obstructive process is present. (Grading the degree of obstruction is described in section 14C.)
2. If the ratio is normal, an obstructive process is usually excluded. An exception is the case of nonspecific ventilatory limitation, in which the FVC and FEV1 are reduced and the FEV1/FVC ratio, flow-volume slope, and TLC are all normal (see section 3E, page 36). Administration of a bronchodilator often exposes occult asthma (Fig. 14-5), but occasionally a methacholine challenge test is needed. Airway resistance, if available, is often increased and can be helpful in identifying the patient with occult asthma.
3. The ratio is normal or increased with a pure restrictive disorder. Patients with a reduced FVC, reduced FEV1,
FIG. 14-5. Control curve shows mild reduction in forced vital capacity (FVC) and forced expiratory volume in 1 second (FEV1) and a normal FEV1/FVC ratio. After administration of a bronchodilator, the flow-volume curve (dashed line) shows a parallel shift to the right with an increase in FVC and FEV1 but no change in the FEV1/FVC ratio. The patient has occult asthma.
normal to increased FEV1/FVC ratio, and normal response to bronchodilator may have a restrictive process. If there is doubt, have the TLC or Dlco measured; they should be abnormally low. If the TLC test is not available, check the chest radiograph for EV1dence of reduction in TLC, or estimate TLC by the radiographic technique discussed in section 3C (page 31). The alveolar volume (Va) can also be checked, as discussed in the Pearl in section 4B, page 43.
Step IV
Examine the expiratory flow values.
1. The forced expiratory flow rate over the middle 50% of the FVC (FEF25-75) almost always changes in the same direction as the FEV1. This test may be more sensitive for detecting early airway obstruction. The FEF25-75 is occasionally reduced in the face of a normal FVC, FEV1, and MVV. The flow-volume curve has a characteristic appearance. This result tends to occur in elderly persons with minimal symptoms (Fig. 14-6). Also see section 7A (page 75).
2. FEF2550,75. These flows change directionally in concert with the FEV1 and FEF25-75.
FIG. 14-6. The unusual flow-volume curve in which the forced expiratory volume in 1 second is normal but the forced expiratory flow rate over the middle 50% of the forced vital capacity is reduced. Note that the peak flow is normal but the lower 70% is very scooped out.
Step V
Examine the MVV.
1. The MVV will change in most cases in a manner similar to that of the FEV1. With a normal FEV1, a normal MVV should be expected (FEV1 x 40 = predicted MVV). Consider the lower limit to be FEV1 x 30.
2. If the FEV1 is reduced by obstructive disease, the MVV will also be reduced. However, the rule that FEV1 x 40 = MVV is not always true in obstructive disease.
3. If the FEV1 is reduced by a restrictive process, the MVV usually is reduced, but not always as much as suggested by the FEV1 (some subjects with a very steep flow-volume curve can have normal flows high in the vital capacity; see Fig. 2-4D, page 12).
4. If the FEV1 is normal but the MVV is reduced below the lower limit, consider the following possibilities:
a. Poor patient performance due to weakness, lack of coordination, fatigue, coughing induced by the maneuver, or unwillingness to give maximal effort (best judged by the technician).
b. Does the patient have a neuromuscular disorder? The MVV is usually the first routine test to have an
abnormal result. Consider ordering maximal respiratory pressure tests (see Chapter 9).
c. Does the subject have a major airway lesion? The MVV is reduced in all three types of lesions shown in Figure 14-4; to evaluate this, the flow-volume loop needs to be evaluated.
d. Is the subject massively obese? The MVV tends to decrease before the FEV1 does.
Step VI
Examine the response to bronchodilator.
1. Is the response normal (<12% increase in both FEV1 and FVC)?
2. Is the response increased (FEV1 or FVC increased by 12% and 200 mL)? If so, this finding is consistent with hyperreactive airways. A case of occult asthma may be exposed by such a response. (Remember the possible effect of effort on the FEV1, section 5D, page 53, and Fig. 5-4, page 55.) However, the patient with asthma may not always have an increased response. The response can vary with the state of the disease.
Step VII
Examine the Dlco.
1. Is the Dlco normal? This result is consistent with normal lungs. However, the Dlco may also be normal in chronic bronchitis, asthma, major airway lesions, extrapulmonary restriction, neuromuscular disease, and obesity.
2. Is the Dlco reduced? This finding is characteristic of pulmonary parenchymal restrictive disorders. It is also consistent with anatomic emphysema and pulmonary vascular disorders. However, values can also be reduced in chronic bronchitis, asthma, and heart failure.
An isolated reduction in the Dlco (other tests within normal limits) should raise the possibility of pulmonary vascular disorders, such as scleroderma, primary pulmonary hypertension, recurrent emboli, and various vasculitides. Chemotherapeutic agents can also produce this finding.
3. Is the Dlco increased? This result occurs in some patients with asthma and in some very obese subjects. Alveolar hemorrhage may also increase the Dlco, as can polycythemia vera, left-to-right intracardiac shunt, or any process that produces pulmonary vascular engorgement.
Step VIII
Examine other test results that you may have available. They should confirm the interpretation at which you have already arrived and fit the patterns in Table 12-1, pages 112-113.
14B. FLOW-VOLUME CURVE NOT AVAILABLE
A spirogram (volume versus time curve) may be available; (see Fig. 2-5, page 15.)
Step I
Examine the FVC.
1. Is it normal? If so, any significant restriction is ruled out.
2. Is it reduced? If so, this finding could be due to either obstruction or restriction (see Fig. 2-3, page 10). If restriction is present, it can be graded (see section 14C, page 139).
Step II
Examine the FEV1.
1. Is it normal? If so, any significant obstruction or restriction is ruled out. There are exceptions, namely, the rare variable extrathoracic lesion in which the FEV1 can be normal but the MVV is reduced because of the inspiratory obstruction. Also, subjects with respiratory muscle weakness can initially present with dyspnea and a normal FEV1.
2. Is it reduced below the lower limit of normal? (The equation used to determine the lower limit of normal is in the Appendix.) If so, this finding is most often due to airway obstruction. However, it could be caused by a restrictive process, and thus the FEV1/FVC ratio needs to be evaluated. First, though, if the TLC is available, it should be checked. A TLC that is increased more than 15 to 20% favors obstruction. By definition, a normal or increased TLC rules out pure parenchymal restriction. The TLC is occasionally normal in a mixed obstructive-restrictive disorder. A reduced TLC is expected in a pure restrictive process.
Step III
Examine the FEV1/FVC.
1. If the absolute ratio is decreased to below the lower limit of normal, an obstructive process is present. (Grading the degree of obstruction is described in section 14C, page 139.)
2. If the ratio is normal, this finding excludes the usual obstructive process. An exception is nonspecific ventilatory limitation, in which the FVC and FEV1 are reduced and the FEV1/FVC ratio and TLC are normal (see section 3E, page 36). Administration of a bronchodilator usually exposes occult asthma, but occasionally a methacholine challenge test is needed.
3. Otherwise, the ratio is normal or increased in a pure restrictive process. A reduced FVC, reduced FEV1, normal to increased FEV1/FVC ratio, and normal response to bronchodilator may indicate a restrictive defect. If there is any doubt, the TLC or Dlco should be measured; they should be low. If the TLC is not available, the chest radiograph can be checked for EV1dence of reduction in the TLC, or the TLC can be estimated by the radiographic technique described in section 3C, page 31. The Va can also be checked, as discussed in the Pearl in section 4B, page 43.
Step IV
Examine the expiratory flow values.
1. The FEF25_75 almost invariably changes in the same direction as the FEV1. This test may be more sensitive than the FEV1 for detecting early airway disease.
2. Rarely, the FEF25-75 is reduced in the face of a normal FVC, FEV1, and MVV. This situation tends to occur in elderly persons who have few symptoms (see section 7A, page 75).
Step V
Examine the MVV.
1. The MVV will, in most cases, change in a manner similar to that of the FEV1. With a normal FEV1, a normal MVV should be expected (that is, FEV1 x 40 = predicted MVV). Consider the lower limit to be FEV1 x 30.
2. If the FEV1 is reduced by obstructive disease, the MVV will also be reduced. However, the rule that FEV1 x 40 = MVV is not always true in obstructive disease.
3. If the FEV1 is reduced by a restrictive process, the MVV usually is reduced. However, the MVV is not always reduced as much as suggested by the reduction in FEV1 because some subjects with a restrictive process have normal flows high in the vital capacity.
4. If the FEV1 is normal but the MVV is reduced, consider the following possibilities:
a. The patient's performance was poor because of weakness, lack of coordination, fatigue, coughing induced by the maneuver, or unwillingness to give a maximal effort (best judged by the technician).
b. Does the patient have a neuromuscular disorder? The MVV test is usually the first routine test to have an abnormal result. Determination of maximal respiratory pressures should be considered (see Chapter 9).
c. Does the patient have a major airway lesion? The MVV is reduced in all three types of lesions (see Fig. 2-7, page 19).
d. Is the subject massively obese? The MVV tends to decrease before the FEV1 does.
Step VI
Examine the response to bronchodilator.
1. Is the response normal (<12% increase in both FEV1 and FVC)?
2. Is the response increased (FEV1 or FVC increased by 12% and 200 mL)? If so, this finding is consistent with hyperreactive airways. A case of occult asthma may be exposed by such a response. (The possible effect of effort on the FEV1 needs to be considered; see section 5D, page 53.) However, the patient with asthma may not always have an increased response. It can vary with the state of the disease.
Step VII
Examine the DLCO.
1. Is the Dlco normal? This finding is consistent with normal lungs. However, the Dlco may also be normal in chronic bronchitis, asthma, major airway lesions, extrapulmonary restriction, neuromuscular disease, and obesity.
2. Is the Dlco reduced? This finding is characteristic of pulmonary restrictive disorders. It is also consistent with anatomic emphysema and pulmonary vascular disease. However, values can also be reduced in chronic bronchitis, asthma, chronic obstructive pulmonary disease, and heart failure.
An isolated reduction in the Dlco (other test results are within normal limits) should raise the possibility of pulmonary vascular disorders such as scleroderma, primary pulmonary hypertension, recurrent emboli, and various vasculitides.
3. Is the Dlco increased? This finding occurs in some patients with asthma and in some very obese subjects. Alveolar hemorrhage may also increase the Dlco, as can polycythemia vera, left-to-right intracardiac shunt, or any process that produces pulmonary vascular engorgement.
Step VIII
Examine other test results that may be available. They should confirm the interpretation already arrived at and fit the patterns given in Table 12-1, pages 112-113.
14C. PULMONARY FUNCTION TEST CRIB SHEET
This summary was developed for use by internal medicine residents and pulmonary fellows at Mayo Clinic.
Spirometry Interpretation
Look at the flow-volume curve, the FVC, and the FEV1 /FVC ratio:
1. Does the curve suggest obstruction (scooped out), restriction (shaped like a witch's hat), or a special case (see below)?
2. Is the FEV1 /FVC ratio reduced (below the lower limit of normal), indicating obstruction?
If the FEV1 /FVC ratio is below the lower limit of normal (LLN) → obstruction algorithm
FEV1 > LLN Borderline
<LLN-60% Mild
59-40% Moderate
39-30% Severe
<30% Very severe
REMEMBER: Obstruction, 60/40/30
If the FEV1 /FVC ratio is normal and the TLC is below the lower limit of normal → restriction algorithm
FVC < LLN-60% Mild
59-50% Moderate
49-35% Severe
<35% Very severe
REMEMBER: Restriction, 60/50/35
(Caution: In some cases the FEV1/FVC ratio is normal but obstruction is present. See ''Nonspecific Pattern,'' below. A bron- chodilator response, increased airway resistance, or a positive methacholine challenge test can be helpful in some of these cases.)
Bronchodilator response is positive if either the FEV1 or FVC increases ≥12% and ≥200 mL.
A large bronchodilator response is predictive of:
More rapid decrease in lung function
More severe exacerbations
Increased risk for rapid decline and death
Flow-Volume Curve
1. Gives clues about the presence of obstruction or restriction (see Fig. 2-5, 2-7B and C).
2. Is the best indicator of test quality (see Fig. 2-6):
The curve should be examined for maneuver errors, including the following (letters correspond to the parts of Fig. 2-6): B) slow start, C&D) poor blast, E) early termination, F&G) cough or interruption in the first second.
3. Gives clues about unusual conditions, such as the following:
a. Plateau on curve may indicate a central airway obstructive process (see Fig. 2-7F, page 19)
b. Normal variant curve (tracheal plateau) common in young adults, especially women (see Fig. 2-6H, page 16)
c. Inspiratory obstruction with variable extrathoracic obstruction (see Fig. 2-7D, page 19) (for example, goiter, tracheal tumor, subglottic stenosis, rheumatoid arthritis with cricoarytenoid fusion)
d. Expiratory obstruction with variable intrathoracic (tracheal) obstruction (see Fig. 2-7E, page 19)
e. A convex flow-volume curve (see Fig. 2-6D, page 16) may be found in the following:
Children
Neuromuscular weakness (see below)
Poor performance
Methacholine Challenge
This is positive if there is a 20% decrease in FEV1 after 25 mg/mL (concentration threshold varies among laboratories). Elements needed for asthma diagnosis: (1) EV1dence of airway hyperresponsiveness, (2) obstruction varying over time, (3) EV1dence of airway inflammation.
Lung Volumes
Gas-dilution techniques (He dilution or N2 washout) underestimate lung volumes in obstructive disorders compared with plethysmography:
Obstructive disorders have a TLC that is high (hyperinflation) or normal
An increased residual volume (RV) (air trapping) and an increased RV/TLC ratio
Restrictive disorders have a reduced TLC
RV may be high (muscular restriction, chest wall limitation, superimposed obstruction)
Neuromuscular Restriction
This looks like pulmonary restriction in spirometry, but:
Lung volumes usually show decreased TLC but increased RV FVC is disproportionately reduced relative to TLC (quantify severity based on FVC, not TLC)
RV/TLC is increased (obstruction is not the only cause of high RV/TLC)
Maximal respiratory pressures are reduced Flow-volume curve looks like poor performance or a child's curve (see Fig. 2-6D, page 16)
Early in the course of disorders causing muscular weakness (for example, amyotrophic lateral sclerosis), maximal respiratory pressures may be reduced, but lung volume, FVC, FEV1, and MVV are still normal (see Table 12-1, page 112-113 and section 9D, page 97).
Nonspecific Pattern
A nonspecific pattern is sometimes termed a ''spirometric restriction.'' These patients have a low FEV1 and FVC, normal FEV1 /FVC ratio, and normal TLC. If possible, airway resistance should be measured. If it is increased, we consider it an obstructive disorder and grade severity based on FEV1. Otherwise, we call it a nonspecific pattern (see section 2F, pages 12-14 and page 38). The most common associated clinical conditions are asthma and obesity. Airway hyperreactivity can be documented in more than half the cases.
Dlco
This is reduced in patients with a gas exchange abnormality (for example, emphysema, idiopathic pulmonary fibrosis, other parenchymal or vascular processes).
A low Dlco is characteristic of emphysema (not as sensitive or specific as high-resolution computed tomography), whereas in asthma and some cases of obstructive chronic bronchitis Dlco is normal.
Dlco may be reduced in pulmonary hypertension, but it is insensitive for detecting cases.
Dlco is often used to monitor for an adverse pulmonary effect of chemotherapy.
Dlco should be adjusted for low hemoglobin for anemic patients.
Dlco maybe increased in (1) asthma, (2) obesity, (3) left-to-right shunt, (4) polycythemia, (5) hyperdynamic states, postexercise, (6) pulmonary hemorrhage, and (7) supine position.
Maximal Respiratory Pressures ("Bugles")
These are used to assess respiratory muscle strength. If low, they indicate muscle weakness or poor performance. Inspiratory pressure is mostly a function of diaphragmatic strength. Tetraplegics show reduced expiratory pressures with inspiratory pressures (diaphragm) relatively preserved. Diaphragmatic paralysis is the opposite.
Obesity
Obesity has a small but sometimes considerable effect on pulmonary function. The increased chest wall impedance causes a restrictive pattern in some obese patients. On average, a person with a body mass index of 35 will have a 5 to 10% reduction in FVC. TLC is usually not reduced to the same degree as FVC. Obese people may wheeze when they breathe near residual volume, sometimes called pseudo-asthma. Dlco is normal or increased. Additional effects of obesity on pulmonary function are discussed in section 12I (page 117) and Table 12-1 (page 112-113).
14D. NEW STANDARD FOR PULMONARY FUNCTION TESTING AND INTERPRETATION
The American Thoracic Society (ATS) has provided standards for the performance of pulmonary function testing [1,2]. The standards have been periodically updated. British and European societies published standards for their respective countries. The most recent update, including lung volume measurements, was developed in conjunction with the European Respiratory Society
Identify presence or absence of obstruction (check curve)
Confirm restriction with lung volumes
Grade severity of obstruction or restriction
Assess gas exchange abnormality
Consider other factors:
• Performance
• Chest wall limitation/obesity
• Neuromuscular weakness
• Central airways
FIG. 14-7. Algorithm for interpretation of results of pulmonary function tests.
CB, chronic bronchitis; CW, chest wall; Dlco, diffusing capacity of carbon monoxide; FEV1, forced expiratory volume in 1 second; ILD, interstitial lung disease; LLN, lower limit of normal; NM, neuromuscular; PV, pulmonary vascular; TLC, total lung capacity; VC, vital capacity.
(ERS) and included a strategy for interpreting the results of lung function tests [3]. The basic sequence for interpretation (Fig. 14-7) is not changed from the methods described in this chapter. However, several aspects of the new standard are troublesome.
The three points of concern are (1) the recommendation to use the FEV1/VC (vital capacity) ratio rather than FEV1/FVC; (2) the "nonspecific abnormality'' with normal FEV1/FVC ratio, low FVC, and normal TLC; and (3) the isolated reduction in Dlco with normal spirometry and lung volumes.
The recommendation for use of the FEV1 /VC ratio rather than FEV1 /FVC ratio for diagnosis of obstruction is based on European reference equations. The VC reported is the largest of any VC maneuver regardless of how it was obtained. Hence, the VC is always equal to or more than the FVC, and the FEV1 /VC is always equal to or less than the FEV1/FVC. The reference equation used is from a U.S. population and calculates an expected value for FEV1/FVC, not FEV1/VC [4]. Arbitrarily substituting FEV1/VC for FEV1/FVC will cause a systematic bias toward overdiagnosis of obstruction.
With use of ATS-ERS flow diagram, the ''nonspecific abnormality'' (normal FEV1/FVC ratio, low FVC and normal TLC) is interpreted as obstruction (that is, asthma or chronic bronchitis). At Mayo Clinic, this common pattern accounts for nearly 10% of all pulmonary function tests. As noted in section 3E, page 36, most cases are associated with asthma, airway hyperreactivity, obesity, and chronic lung disease. However, 25% of the cases studied had no EV1dence of airway obstruction.
The isolated reduction in Dlco with normal spirometry and lung volumes is said to be due to pulmonary vascular disorders. It is true that pulmonary vascular disorders, such as pulmonary hypertension, can cause this pattern; however, they are relatively rare. The most common cause of an isolated reduction in Dlco with normal lung mechanics is emphysema, and it is not always mild when examined with computed tomography. In addition, relatively mild cases of interstitial lung disease often present with low Dlco with lung volumes in the low-normal range [5].
The new ATS-ERS interpretation strategy changes the thresholds defining the severity of obstruction or restriction without providing any rationale for doing so. However, we use the classification of severity shown in Table 14-1.
Meanwhile, guidelines for chronic obstructive pulmonary disease have been established by numerous groups, most notably the Global Initiative for Chronic Obstructive Lung Disease and the ATS. These standards have shifted the grading of severity compared with older standards. This change is said to be motivated by a desire to improve early recognition of disease, but it may prompt prescription of expensive medications with associated adverse effects for asymptomatic patients who may not benefit as clearly as more severely affected patients. These concerns are discussed in references [6] and [7].
TABLE 14-1. Impairment and severity stratifications
‘Numbers in parentheses are forced expiratory volume in 1 second. +Numbers in parentheses are forced expiratory vital capacity.
FEV1, forced expiratory volume in 1 second; FVC, forced expiratory vital capacity; LLN, lower limit of normal.
REFERENCES
1. American Thoracic Society. Evaluation of impairment/disability secondary to respiratory disorders. Am Rev Respir Dis 133:1205-1209, 1986.
2. American Thoracic Society. Lung function testing: Selection of reference values and interpretative strategies. Am Rev Respir Dis 144:1202-1218,1991.
3. Pellegrino R, Viegi G, Brusasco V, et al. Interpretative strategies for lung function tests. Eur Respir J 26:948-968, 2005.
4. Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 159:179-187,1999.
5. Aduen JF, Zisman DA, Mobin SI, et al. Retrospective study of pulmonary function tests in patients presenting with isolated reduction in single-breath diffusion capacity: implications for the diagnosis of combined obstructive and restrictive lung disease. Mayo Clin Proc 82:48-54, 2007.
6. Enright P. Flawed interpretative strategies for lung function tests harm patients [editorial]. Eur Respir J 27:1322-1323, 2006.
7. Pellegrino R, Brusasco V, Crapo RO, et al. From the authors [editorial]. Eur Respir J 27:1323-1324, 2006.