The Forced Expiratory Volume Maneuver

Assessments of vital capacity (VC), or forced vital capacity (FVC) and airflow are based on the forced expiratory volume maneuver, in which the subject inhales maximally to total lung capacity (TLC) and then exhales forcefully and completely to residual volume (RV). The expiratory flow rate at any point during this maneuver is determined by the driving pressure for airflow and the airway resistance. During a forceful exhalation, the intrathoracic pressure that surrounds the central airways exceeds the intraluminal pressure, causing dynamic compression of the airway (see Chapter 3, Figure 3-12). As a result, the effective driving pressure becomes the difference between alveolar pressure and the pleural pressure that compresses airways. This pressure difference (PA − Ppl) is equivalent to the elastic recoil pressure of the lung tissue. Thus, even during a forceful effort, the intrinsic elastic properties of the lung are a major determinant of airflow. Airway resistance upstream from the point of compression is determined primarily by airway caliber, which varies directly with lung volume. Throughout exhalation from TLC, both recoil pressure and airway caliber progressively decrease, so that airflow rates, after an early peak, also progressively decrease. Although the peak expiratory flow rate varies with the rapidity and forcefulness of the expiratory effort, once dynamic compression begins, the flow rate during the middle to later portions of the maneuver is limited and independent of further effort beyond the threshold needed to begin compression. These flows also are independent of added resistance downstream from the point of flow limitation. This physiologic arrangement aids in making the basic measurements of spirometry quite reproducible on repeated efforts.

To obtain a satisfactory spirometric tracing, the preceding inspiration must be maximal and the forced expiratory volume maneuver must be continued to cessation of flow or, when emptying is slowed, for at least 6 to10 seconds. The resultant information commonly is displayed in one of two formats. The traditional spirogram (Figure 9-1) plots volume versus time, with flow rate indicated by the steepness of the plot. The orientation of the axes varies with equipment, with time moving to the right and exhaled volume plotted either up or down. In the flow-volume display (Figure 9-2), the instantaneous flow rate is measured continuously and directly plotted on the vertical axis with volume on the horizontal axis. Time is not shown on this plot but may be indicated by tick marks. With this display, the reproducibility of successive efforts and some patterns of abnormality may be more easily seen. It is important to recognize that both the traditional spirogram and the expiratory flow-volume display are obtained from the same maneuver but emphasize different aspects of the information thus obtained.

Figure 9-1 Normal forced expiratory spirogram plotted as exhaled volume versus time. The forced expiratory volume at 1 second (FEV₁) and forced vital capacity (FVC) are indicated by arrows. In this example, FEV₁ = 3.35 L, FVC = 4 L, and FEV₁/FVC = 0.84.

(Modified from Culver BH: Pulmonary function testing. In Kelly WN, editor: Textbook of internal medicine, Philadelphia, 1988, JB Lippincott.)

Figure 9-2 Normal expiratory flow-volume curve. The same forced expiratory volume maneuver shown in Figure 9-1 is plotted here as a flow-volume curve. The airflow rate reaches a peak early in the exhalation and then decreases progressively until flow ceases at residual volume.

(Modified from Culver BH: Pulmonary function testing. In Kelly WN, editor: Textbook of internal medicine, Philadelphia, 1988, JB Lippincott.)

Expiratory Flow Measurements

Basic measurements from the forced expiratory volume maneuver include FVC, the forced expiratory volume in 1 second (FEV₁), and the ratio of FEV₁ to FVC. The FVC or total volume exhaled is equivalent in normal subjects to a so-called slow VC, obtained with a complete but not forceful exhalation. Patients who have advanced obstructive airway disease often manifest exaggerated dynamic compression (narrowing airways with forceful efforts), so that the FVC is smaller than the slow VC. A reduction in VC reflects either a reduction in TLC, an increase in RV, or a combination of both. The FEV₁ is readily obtained from the volume-time spirogram by observing the volume exhaled in the first 1 second of effort. This measurement cannot be seen on the flow-volume display but can be calculated by the microprocessors in the equipment that use this display. Determination of FEV₁/FVC is easily performed using simple equipment; this ratio provides the best index of airflow limitation. When the slow VC also is available, and if it is larger, this may be substituted for FVC, thereby increasing the sensitivity of the ratio for detection of obstruction. This ratio is commonly expressed as a percentage and sometimes is referred to as the “percent FEV₁”; however, this terminology may cause confusion, because the FEV₁ itself is commonly expressed as a percentage of its predicted value. Misunderstanding is lessened if the ratio is reported and discussed as a decimal (e.g., 0.82).

An additional flow measurement widely reported from the spirogram is the average forced expiratory flow rate between 25% and 75% of the exhaled VC (FEF_25–75), formerly referred to as the maximum midexpiratory flow rate. This measurement shows wider variability than that typical for FEV₁ or FEV₁/FVC, both within and between individual subjects. When this variability is appropriately accounted for, FEF_25–75 is not more sensitive than FEV₁/FVC for the detection of airflow limitation. Numerous other flow measurements can be obtained from the forced expiratory volume maneuver, but they are highly interdependent with those already described and add little new information. The FEV_0.5 may be used to assess the initiation of effort but adds little diagnostic information in adults. In young children, the FEV_0.5 or FEV_0.75 can be a useful index of flow, because children may fully empty their lungs within the first second.

Because it may be difficult for some subjects to consistently maintain the forced expiratory maneuver to full exhalation, which may take 10 to 15 seconds when airflow obstruction is present, the forced expiratory volume in 6 seconds (FEV₆) may be taken as a surrogate for FVC and used to calculate an FEV₁-to-FEV₆ ratio, which can be compared with appropriate reference values.

The peak expiratory flow rate achieved during the FVC maneuver cannot be accurately calculated from a spirogram display but is readily seen on the flow-volume display and can be calculated by microprocessors. It can show considerable effort-to-effort variability, even when FEV₁ and FVC measurements are nearly identical. A peak flow measurement also can be obtained with simple handheld devices, which are useful for interval follow-up evaluation and for home management of patients who have reactive airway disease but are less accurate and less sensitive than spirometry for screening. Whereas spirographic flow measurements are obtained over a time interval or volume interval, measurements from the flow-volume display or current microprocessors can be reported at specific lung volumes. Maximum flow rates at 50% and 75% of exhaled volume are commonly reported, but nomenclature varies, and the latter may be designated as the flow rate at 25% of remaining VC. Routine use of these measurements is not recommended.

The maximum voluntary ventilation (MVV) is measurable on some office spirometers but is primarily a laboratory measurement. The subject is instructed to breathe deeply and rapidly, typically at 60 to 70 breaths per minute, and the total volume of ventilation over a 12- to 15-second period is extrapolated to liters per minute. Historically, this was the initial dynamic test for obstructive disease; however, it has been supplanted by the forced expiratory maneuver for the diagnosis of airflow limitation. The MVV currently is used as a global assessment of ventilatory capacity in the evaluation of dyspnea, in the interpretation of exercise limitation, in disability assessment, in some preoperative testing, and to evaluate neuromuscular disease of the chest wall and diaphragm.

Reversibility

The usefulness of spirometry in the office or laboratory often is enhanced by the assessment of bronchodilator response. Spirometry is repeated after the administration of an inhaled bronchodilator, waiting 15 minutes after a beta agonist or 30 minutes after ipratropium bromide. An increase of more than 12% in the FEV₁ represents a significant response in a patient who has near-normal baseline spirometry results. With more severe obstructive disease, the magnitude of improvement also should be at least 200 mL to differentiate the pharmacologic response from test-to-test variability. FVC often improves in parallel with FEV₁. An improvement in FVC by more than 12% and 200 mL, in the absence of a significant change in FEV₁, may reflect either an improvement in flow rates after the first second or simply a longer duration of effort.

Although determination of the FEV₁/FVC ratio is the most useful test for the diagnosis of airflow limitation, the value may remain the same or even decrease after administration of a bronchodilator, depending on the relative change in its two components, so this ratio is not a useful index of reversibility. Because of its large intraindividual variability, FEF_25–75 must show an increase of 30% to 40% to represent a significant bronchodilator effect. Occasionally, this parameter changes little or even decreases despite a clear improvement in FEV₁ or, particularly, FVC. This apparent paradox occurs because the bronchodilator has allowed exhalation to continue to a lower RV, so that the 25% to 75% increment is now measured at a lower lung volume, with consequent lower flow rates.

Reference Equations and Limits of Normality

Unlike many laboratory tests, lung function parameters vary greatly with body size and age, so the expected values must be determined on an individual basis. Numerous prediction equations have been derived from spirometric surveys of normal reference populations. Currently accepted studies exclude all smokers as well as persons who have any thoracic or cardiopulmonary disease. Most studies have found that lung function parameters can be predicted on the basis of gender, age, and height, and that the addition of other body size measurements does not improve the accuracy of the equations. The prediction equations give the midpoint of the normal range, which is unfortunately wide for most spirometry measurements.

The lower limit of normal (LLN) must be established from the variability among individual subjects who have the same prediction parameters. The limits of the normal range are chosen to exclude 5% of a normal population; that is, 5% will be misclassified as having disease. In screening a generally healthy population for a rare disease, a borderline-low result is more likely to reflect this misclassification than to represent true identification of disease. However, when spirometry is done for persons at risk for lung disease, or with suggestive symptoms, the probability that a borderline result reflects a true abnormality is much higher. For the spirometry measurements, only low values are considered of concern, so the LLN is set at the 5th percentile of the reference sample. Because the distribution of values in the reference population is adequately symmetric above and below the mean, the 5th percentile LLN often is taken as the predicted value minus 1.645 times the standard error of the estimate (SEE) of the regression equation. The predicted value and the LLN both are readily calculated from the reference data programmed into the spirometry equipment, and both should be reported for comparison against the actual measured value.

Spirometry reference data from a large, systematic survey of the U.S. population (the Third National Health and Nutrition Examination Survey [NHANES III]) are recommended for use in North America. Equations are provided for Caucasians, African Americans, and Mexican Americans ranging in age from 8 to 80 years. No single reference source currently is recommended for use in Europe, but an international effort is under way to merge the NHANES data with those for many reference populations from throughout Europe and elsewhere, to generate a new, widely applicable reference dataset.

The use of a percentage of the predicted value as a lower limit is convenient but less accurate than the 5th percentile, because it causes the normal range to vary with the magnitude of the predicted value, whereas the true variance is similar around small and large values. A lower limit value equal to 80% of the predicted value has been widely used in spirometric interpretation. Although this is a reasonable approximation of the LLN for FEV₁ and FVC at the midrange of age and height, it creates an overly broad normal range for younger, taller persons (in whom the true LLN is approximately 82% to 83%), and it is overly sensitive for older or smaller subjects (in whom the LLN may be as low as 73% to 75%). An 80% lower limit is quite inappropriate for FEF_25–75, because the normal range extends to 65% of the predicted value in the young and below 50% of predicted in older persons. The normal value for FEV₁/FVC varies little with height but does decline progressively with age (e.g., from 0.87 at age 20 to 0.77 at age 70 in females, and from 0.84 to 0.74 in males). The LNN is approximately 0.10 below the predicted ratio. Because the ratio often is expressed as a percentage, reporting this value as a percent of the predicted value is potentially confusing.

Interpretation of Spirometric Abnormalities

Measurement by Helium Dilution

A spirometer is prepared that contains a known volume and concentration of an inert gas, typically 10% helium (Figure 9-3). While the subject breathes through a mouthpiece with nose clipped, a valve is turned at end-tidal exhalation to connect the airway to this closed system. As normal tidal breathing continues over the course of a few minutes, the gas in the subject’s lung equilibrates with gas in the spirometer, and the helium concentration, which is continuously monitored, falls to a new, lower, steady-state level. Carbon dioxide is removed from the closed system by soda lime absorption, and a low flow of oxygen is added to compensate for the subject’s ongoing oxygen consumption by keeping the mixing chamber or spirometer volume constant. The ratio of the initial to the final concentration of helium allows calculation of the unknown volume (FRC) added to the system. A continuous tracing of the spirogram, including a maximum inspiratory and expiratory effort, allows calculation of the subdivisions of lung volume, and correction for any offset from the relaxed FRC at the moment the valve was opened to start the test.

Figure 9-3 Lung volume measurement by helium dilution. A, The spirometer and circuit are prepared with a known gas volume and concentration of helium. B, The subject breathes through the circuit, as CO₂ is absorbed and O₂ consumption is replaced, until a new, lower helium concentration is established. The unknown lung volume added to the circuit when the valve was turned is calculated from the dilution of the initial helium concentration: [He]_initial × Vol₁ = [He]_final × (Vol₁ + FRC).

Measurement by Nitrogen Washout

The nitrogen washout technique also is based on the principle of conservation of mass of an inert gas—in this case, the nitrogen normally resident within the lungs. The subject breathes on a mouthpiece that, at the end of a relaxed tidal exhalation, is connected to an inspiratory source of 100% oxygen (Figure 9-4), while the subsequent exhaled gas is directed by one-way valves into a collection bag, previously flushed with oxygen so that it contains no nitrogen. The resident nitrogen is washed out of the lungs progressively and monitored with continuous analysis at the mouthpiece. When the exhaled nitrogen concentration falls below 2%, the test is terminated, and the volume of nitrogen collected is measured. The FRC can be calculated on the basis that this nitrogen volume represents 80% of the lung gas contained at the beginning of the test. Instead of the collection bag, current microprocessors use a calculation based on instantaneous, breath-by-breath measurement of exhaled volume times nitrogen concentration. Washout can be completed in 3 to 4 minutes in normal subjects but may require longer than 15 minutes with severe obstructive airway disease, so that the gas volume in slowly exchanging spaces can be measured.

Figure 9-4 Lung volume measurement by nitrogen washout. A, Before the test, as the subject breathes air, the nitrogen concentration in the lungs is 80%, and the collection system is flushed free of nitrogen. B, The subject inhales 100% O₂ and exhales into the collection bag until the exhaled N₂ concentration approaches zero. The total volume of the bag and its final N₂ concentration are measured, and the unknown initial lung volume (including valve and mouthpiece dead space) is calculated: [N₂]_bag × Vol_bag = 0.80 × FRC.

Measurement by Body Plethysmography

The volume of gas within the thorax, whether in communication with airways or not, can be measured by the technique of body plethysmography, based on the physical principles of gas compression described by Boyle’s law. The subject sits within a fully enclosed rigid box and breathes through a mouthpiece connected through a shutter to the internal volume of the box (Figure 9-5). Sensitive manometers monitor the pressure at the airway and inside the chamber. The apparatus is calibrated with the subject in place, so that the volume addition required within the chamber to raise the chamber pressure by 1 cm H₂O is known. At the end of a tidal exhalation, the airway shutter is closed, and the subject is asked to make panting efforts with the glottis open. An effort to expand the chest decompresses intrathoracic gas and reciprocally compresses that in the chamber; the opposite occurs during an expiratory effort.

Figure 9-5 Lung volume measurement by body plethysmography. See text for description of the procedure. As the subject makes panting efforts against a closed airway shutter valve, the product of pressure and thoracic gas volume (V_TG) stays constant (Boyle’s law). Thus, P_B × V_TG = (P_B – ΔP) × (V_TG + ΔV). Solving for V_TG yields: V_TG = P_B (ΔV/ΔP) − ΔV. Because ΔV is very small relative to V_TG, it is ignored, so V_TG = P_B (ΔV/ΔP). ΔP is directly obtained from the airway pressure transducer, and ΔV is obtained from the pressure change in the plethysmograph, after calibration by cycling a known volume with a piston pump. V_TG is obtained from the slope of this relationship plotted during the panting maneuver.

Under conditions of constant temperature, which is well maintained by the high blood flow through the lungs, the product of airway pressure and lung gas volume is a constant. The slope of the relationship between the change in airway pressure and the change in thoracic volume, which can be calculated continuously or plotted on an oscilloscope, is inversely related to the intrathoracic volume. Because this technique is sensitive to gas volume not in free communication with the airways, such as that in bullae or even a pneumothorax, the measurement often is called “thoracic gas volume,” which may exceed the FRC measured by gas dilution techniques. Advantages of this method, besides its inclusion of “trapped” gas, are that several measurements can be repeated rapidly, and that airway resistance can be measured with the same apparatus when panting is continued with the shutter open.

Interpretation of Lung Volume Abnormalities

Inspiration is limited at TLC when the maximum inspiratory force that can be applied by the chest muscles and diaphragm is opposed equally by the increasing recoil force of the lungs as they are distended to higher volumes. TLC is limited primarily by the elastic properties of the lungs, because variations in muscle strength have only a small effect on total chest expansion until weakness becomes quite marked. Parenchymal restrictive diseases reduce lung compliance, so greater distending pressure is required to achieve any volume change, and the maximum volume that can be achieved (TLC) is reduced. The displacement of intrathoracic gas volume by effusions, edema, intravascular volume, and inflammatory cells also can contribute to a reduction in measured lung gas volumes. Except for pleural effusions, these quantities are relatively small and outweighed by the frequently associated changes in lung elastic properties.

The minimum lung volume, or RV, is determined by a combination of two factors. The smallest volume to which the expiratory muscles can squeeze the chest wall and raise the diaphragm is the dominant factor in youth. By middle age, with the normal loss of tissue elastic recoil forces, the lung volume at which small airways close and trap remaining gas behind them increases and becomes the dominant factor in determining RV.

Obstructive Diseases

Airflow obstructive processes lead to airway closure that limits exhalation at higher lung volume because of the combined effects on luminal caliber of airway inflammation and loss of tissue recoil. These changes result in a progressive increase in RV (Figure 9-6), as increasing amounts of gas are trapped behind closed airways. Affected patients breathe at an increased FRC because of the combined effects of a decrease in lung recoil force from emphysema and the need to increase luminal caliber to minimize the resistive work of air flow. The TLC is normal to high, which again reflects the loss of lung recoil forces. Because RV increases to a greater extent than that seen in TLC, the VC decreases with severe airway obstruction.

Figure 9-6 Severe airflow obstruction is associated with an increase in residual volume. Prolonged expiratory airflow may continue until the subsequent inhalation, and alveolar gas is trapped behind narrowed and closed airways. The functional residual capacity at which tidal breathing occurs also is increased, and total lung capacity may be high as well.

(Modified from Culver BH: Pulmonary function testing. In Kelly WN, editor: Textbook of internal medicine, Philadelphia, 1988, JB Lippincott.)

Principles and Technique

The diffusing capacity (DL), also called transfer factor, is a measure of the capacity to transfer gas from alveolar spaces into the alveolar capillary blood. This process occurs by passive diffusion and is a function of the pressure difference that drives gas, the surface area over which exchange takes place, and the resistive properties to gas movement through the membrane and into chemical combination with the blood. The units are milliliters per minute per millimeter of mercury of driving pressure (mL/minute/mm Hg). (In SI units, 1 mole/minute/kPa = 2.896 mL/minute/mm Hg.) Carbon monoxide is used for the clinical test of diffusing capacity (DLCO), because its extreme avidity for hemoglobin allows the back pressure to diffusion to be considered negligible.

In the widely used single-breath method, the subject exhales to RV and then takes a VC inhalation of the test gas, which contains a low level of carbon monoxide (0.3%) and an inert gas not taken up in the blood (e.g., 10% helium). After breath-holding at full inspiration for 8 to 10 seconds, the subject exhales quickly. The initial portion of the expirate, which includes anatomic dead space, is discarded, and a sample of the subsequent alveolar gas is collected or measured by a rapidly responding sensor. The reduction in helium concentration in the alveolar sample allows calculation of the alveolar volume at TLC into which carbon monoxide was distributed, and of the initial carbon monoxide concentration after its dilution by the resident RV. The final concentration of carbon monoxide measured in the exhaled alveolar sample is applied in calculating the volume of carbon monoxide transferred out of alveoli and also allows a calculation, for which an exponential decline is assumed, of the mean carbon monoxide driving pressure during breath-holding. An effective residence time is calculated from the breath-hold period plus a portion of the time of inspiration and sample collection.

A significant problem with the diffusing capacity measurement is that numerous variations in the handling of small correction factors (for gas conditions, apparatus dead space, timing measurement, and so on) can cumulatively cause the calculated value to vary substantially among laboratories. Typically, two measurements are done and the results are averaged. Although reproducibility within a laboratory can be quite acceptable (±2.5 mL/minute/mm Hg), the accuracy of comparisons between laboratories or to reference data is much less consistent, as reflected by published predicted values that vary by 20% or more. It is essential that each laboratory choose prediction equations that are appropriate to the nuances of its equipment and technique.

Interpretation of Diffusion Abnormalities

Although diffusion often is thought of as a function of alveolar membrane thickness, the dominant factor is usually the capillary blood volume, which influences both the surface area available for exchange and the volume of blood and hemoglobin available to accept carbon monoxide. The influence of hemoglobin concentration, [Hb], can be accounted for by theoretical or empirical correction factors. The velocity of blood flow is not important, because carbon monoxide is taken up even by stagnant blood (or extravasated blood in the case of pulmonary hemorrhage), but the recruitment of additional capillary segments during high flow conditions such as exercise or with congenital left-to-right shunt increases the measured diffusing capacity.

Many laboratories also report the diffusing capacity as a ratio of such capacity to the alveolar volume (DL/VA). This also is called the transfer coefficient (K_CO). The implication is that loss of lung volume secondary to mechanical abnormalities is accompanied by a parallel loss of diffusion capacity. This, however, is not the case with a voluntary limitation of inspiration, in which capillaries remain perfused and DL/VA rises, or with pneumonectomy, in which capillaries are recruited in the remaining lung and DL/VA is again high. Diffusing capacity is commonly reduced in parenchymal inflammatory diseases, primarily because of the loss of available capillaries. The most common pattern in diseases such as sarcoidosis and interstitial fibrosis is for DL to be reduced and DL/VA to be slightly low or “normal,” because volume also is lost. Both DL and DL/VA are low with the loss of capillary surface area and blood volume in emphysema and in diseases that are primarily vascular, such as vasculitis, recurrent emboli, and pulmonary hypertension. Clinical interpretation of diffusion abnormality should be based primarily on the DL, not on the DL/VA, with small correction factors available for [Hb] and the lower oxygen levels of altitude.

Closing Volume

As lung volume decreases, the smaller, intraparenchymal airways decrease in caliber until they close at low lung volume and ventilation to or from alveoli beyond them ceases. Because there is a vertical gradient in the pleural pressure that surrounds the lungs, the lung tissue is less distended in dependent regions than at higher levels in the thorax. In late exhalation, dependent airways close (i.e., these areas reach their regional RV), whereas air continues to flow from the upper portions of the lung until they too close, and overall RV is reached. The beginning of this wave of ascending airway closure can be detected by physiologic tests and is termed closing volume. Closing volume usually is expressed as a percentage of VC. That is, a closing volume of 20% means that airway closure can be detected during a slow exhalation when 20% of the VC remains before RV is attained (Figure 9-7). Alternatively, when RV is measured, this can be added to closing volume, and the sum, termed closing capacity, is expressed as a percentage of TLC.

Figure 9-7 Effects of age on the normal spirogram. Residual volume progressively increases, with an associated small reduction in vital capacity. The functional residual capacity (FRC) increases slightly, but the closing volume, at which dependent airways cease to ventilate, increases more steeply and exceeds the normal FRC in older ages.

(Modified from Culver BH, Butler J: Alterations in pulmonary function. In Andres R, Bierman E, Hazzard W, editors: Principles of geriatric medicine, New York, 1985, McGraw-Hill.)

Both of these measures have been used as tests of early airway dysfunction in the natural history of COPD. Abnormalities can be detected in a high percentage of smokers, but the prognostic usefulness is limited, because this group includes many who do not go on to develop progressive airflow limitation. On an individual patient basis, the closing volume is most helpful in its relationship to the lung volume at which tidal breathing occurs. When airway closure occurs at a volume below FRC, the airways are open throughout the lungs during tidal breathing, but when airway closure occurs above FRC, the affected alveoli are underventilated. Because the dependent regions are well perfused, premature airway closure creates a low ventilation-perfusion region, which contributes to hypoxemia. This abnormality occurs when the closing volume is increased by normal aging or COPD, and by the effect of peribronchial edema in left ventricular failure. Similar consequences are observed when FRC is reduced by recumbent posture or by obesity.

Upper Airway Obstruction

Obstruction in the central airways (e.g., tracheal tumor or stenosis) affects the expiratory flow-volume relationship in a different way than does the more common peripheral airway obstruction of COPD. The latter has its predominant effect late in expiration, with slowing of terminal flow rates, so that peak flow tends to be relatively maintained while the remaining flow-volume curve becomes progressively convex toward the horizontal axis (Figure 9-8). Central obstructions have their primary effect early, which results in a truncated, flat-topped flow-volume curve (Figure 9-9), reflecting a steady effort against a constant resistance. In the latter portion of the expiration, the decreasing lung volume and airway caliber shift the site of major resistance to the more peripheral airways, so that the latter portion of the flow-volume curve is normal.

Figure 9-8 Expiratory flow-volume curve of chronic airflow obstruction. Airflow rates are markedly reduced at mid- to lower lung volumes, with a curve that is convex to the horizontal axis. In this example, just under 50% of the vital capacity has been exhaled in 1 second.

(Modified from Culver BH: Pulmonary function testing. In Kelly WN, editor: Textbook of internal medicine, Philadelphia, 1988, JB Lippincott.)

Figure 9-9 Expiratory flow-volume curve shows the pattern typical of central airway obstruction. Peak flow is markedly truncated, but the flow rates at low lung volume are unaffected. Despite the dramatic effect on the flow-volume curve, the ratio of FEV₁ to FVC is only modestly affected, with a value of 0.71 in this example.

(Modified from Culver BH: Pulmonary function testing. In Kelly WN, editor: Textbook of internal medicine, Philadelphia, 1988, JB Lippincott.)

When a central obstruction is in the extrathoracic airway and has some flexibility (e.g., vocal cord paralysis), its effect is much greater during inspiratory flow than expiratory flow. The negative intraluminal pressure generated during inspiration narrows the airway, which exacerbates the obstruction, whereas during expiration, the positive airway pressure below the site of obstruction tends to distend the airway, which reduces the abnormality. These lesions are assessed by recording on the flow-volume display the maximum effort inspiratory flow curve, as well as that during expiration, to complete a flow-volume “loop.” The normal inspiratory flow pattern has a semicircular shape with peak inspiratory flow at midvolume that consistently exceeds midvolume expiratory flow (Figure 9-10).

Figure 9-10 Flow-volume loop showing both forced expiratory and inspiratory airflows. The expiratory peak flow is somewhat truncated, consistent with a mild central obstruction during exhalation; inspiratory flow is markedly reduced from that for the normal curve. This pattern is typical of a flexible, extrathoracic obstruction, such as that secondary to paralysis of the vocal cords.

(Modified from Culver BH: Pulmonary function testing. In Kelly WN, editor: Textbook of internal medicine, Philadelphia, 1988, JB Lippincott.)

Bronchoprovocation Testing

Patients who have suspected reactive airway disease frequently demonstrate normal spirometry values when asymptomatic. When the diagnosis is unclear, provocation testing may be performed in an attempt to explain symptoms or predict future risks. A common clinical provocation of asthma is the airway mucosal cooling and drying effect of exercise hyperpnea or cold air inhalation. Provocation testing with exercise can utilize free running, treadmill running, or bicycle pedaling, with the relative yield of abnormal tests in that order. To provide a sufficient ventilatory stimulus, exercise level is increased over 2 to 3 minutes to 70% to 90% of maximal capability, as estimated from heart rate, and then sustained for an additional 4 to 6 minutes. Spirometric testing is done before exercise and repeated after exercise, with the most marked decrease in flow rates noted at 5 to 10 minutes after exercise. A reduction in FEV₁ of 10% is considered significant, because normal subjects typically show a small increase shortly after exercise. Cold air challenge, in which spirometric testing is done before and after isocapneic hyperventilation of air that has been dehumidified and cooled to 4° C, is less widely available as a provocation maneuver.

Methacholine responsiveness is a nonspecific indicator of airway reactivity. Starting with a single inhalation at a very low concentration, patients are tested after progressively increasing inhaled doses until either a predetermined maximum dose has been achieved or the FEV₁ has been observed to fall by 20%. Normal subjects do not respond to the maximum dose, whereas patients with asthma usually respond to very low to intermediate doses. A negative response at the maximum dose in a patient suspected to have asthma and recently symptomatic makes that diagnosis very unlikely, but positive responses are less specific. Persons who have a family history of asthma or hay fever symptoms may show intermediate responses, as may patients with COPD or cystic fibrosis, as well as some normal persons in the recovery period after viral respiratory infections.

Whereas methacholine directly stimulates smooth muscle contraction, indirect stimuli such as exercise or the recently developed mannitol challenge test act through changes in mucosal osmolarity and release of endogenous mediators. These stimuli are more specific for asthma and more responsive to variations in airway inflammation, but they tend to be less sensitive than methacholine challenge.

In selected circumstances, provocation testing may be carried out with suspected specific allergens or occupational exposures. Dose preparation needs to be very careful, to avoid an excessive and dangerous response. Studies may be designed to mimic the circumstances of the patient’s clinical or occupational exposure. Testing may need to be continued for several hours to identify a late-phase reaction.