Diffusing Capacity Tests

Published on 12/06/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 6295 times

Chapter 3

Diffusing Capacity Tests

The chapter describes measurement of diffusion in the lungs. Diffusing capacity (also referred to as transfer factor) is usually measured using small concentrations of carbon monoxide (CO) and is referred to as Dlco or Dco. Dlco is used to assess the gas-exchange ability of the lungs, specifically oxygenation of mixed venous blood. Various methods, all of which use CO, have been described. The most commonly used method is the single-breath, or breath-hold technique. The single-breath method is also the most widely standardized. The techniques section focuses on the single-breath method but also describes some of the other methods that are used for specific applications.

Dlco measurements are used in the diagnosis and management of most pulmonary disorders. The importance of standards and guidelines in the performance of Dlco tests and in the overall interpretation of results is highlighted. Interpretive strategies are presented in a format similar to those in previous chapters.

Carbon monoxide diffusing capacity

Description

Dlco measures the transfer of a diffusion-limited gas (CO) across the alveolocapillary membranes. Dlco is reported in milliliters of CO/minute/millimeter of mercury at 0°C, 760 mm Hg, dry (i.e., standard temperature, pressure, dry [STPD]).

Techniques

CO combines with hemoglobin (Hb) approximately 210 times more readily than oxygen (O₂). In the presence of normal amounts of Hb and normal ventilatory function, the primary limiting factor to diffusion of CO is the status of the alveolocapillary membranes. This process of conductance across the membranes can be divided into two components: membrane conductance (Dm) and the chemical reaction between CO and Hb (Figure 3-1). Dm reflects the process of diffusion across the alveolocapillary membrane. Uptake of CO by Hb depends on the reaction rate (θ) and the pulmonary capillary blood volume (Vc). These two components occur in series, so the diffusion conductance can be expressed as:

Figure 3-1 Diffusing capacity is performed to evaluate the alveolar capillary interface where gas exchange occurs.

< ?xml:namespace prefix = "mml" />1DLCO=1Dm+1θVc D

Diffusing capacity can be affected by factors that change the membrane component, as well as by alterations in Hb and the capillary blood volume.

A small amount of CO in inspired gas produces measurable changes in the concentration of inspired versus expired gas. Because little or no CO is normally present in pulmonary capillary blood, the pressure gradient causing diffusion is basically the alveolar pressure (P_Aco). If the partial pressure of CO in the alveoli and the rate of uptake of the gas can be measured, the Dlco of the lung can be determined. There are several methods for determining Dlco (Table 3-1). All methods are based on the following equation:

Table 3-1

Advantages and Disadvantages of Dlco Testing Methods

Method	Technique	Advantages	Disadvantages	Applications
Dlcosb (breath hold)	CO and tracer gas analysis relatively simple; 10-sec breath hold	Easy calculations, simple, fast; highly standardized and automated; minimal COHb back-pressure effect	Sensitive to distribution of ventilation and /matching; “nonphysiologic”; not practical for exercise testing	Screening and clinical applications
Dlcorb (rebreathing)	Rapid analysis of CO and tracer gas required; rebreathing must be controlled	Less sensitive to V_A than Dlcosb; less sensitive to /abnormalities; can be used with NO to measure Dlno	Complex calculations (computerized); rapidly responding analyzers required; sensitive to COHb back-pressure	Clinical and research applications; provides most accurate Dlco
Dlcoib (intrabreath)	Rapid analysis of CO and tracer gas during single controlled exhalation	Breath holding not required; can be used during exercise	Complex calculations (computerized); flow must be controlled; sensitive to uneven /; not standardized	Screening and clinical applications; may be useful in patients who cannot hold their breath
1/Dm + 1/θV_c (membrane diffusing capacity)	Dlcosb repeated at two different levels of alveolar PO₂	Differentiates membrane transfer resistance from red cell uptake	Complex calculations; estimates of alveolar PO₂ are critical	Research with limited clinical applications

Dlco

DLCO=V·COPACO-PCCO

where:

co milliliters of CO transferred/minute (STPD)

P_Aco = mean alveolar partial pressure of CO

P_Cco = mean capillary partial pressure of CO, assumed to be 0

Dlco is expressed in milliliters of gas/minute/unit of driving pressure at STPD conditions.

3-1How To…

Perform a Dlcosb Maneuver

1. Tasks common to all procedures

a. Calibrate and prepare equipment

b. Review order

c. Introduce yourself and identify patient according to institutional policies

d. Describe and demonstrate procedure

2. Review spirometry data and adjust expired volume criteria if appropriate (VC < 2.0 L).

3. Enter recent Hgb and adjust for COHb if available.

4. Have the subject sit upright with feet flat on the floor and adjust the mouthpiece to the proper height.

5. The mouthpiece should rest on top of the tongue. Have the subject gently bit down with his or her teeth and seal lips firmly around the mouthpiece. Emphasize the need for tight lips without a leak. Place the noseclip securely.

6. Have them breathe quietly for several breaths and then empty completely to RV (verbal cues such as “keep blowing,” or “push, push until empty”).

7. Once empty, have subject inhale rapidly as much as possible (verbal cues, “up, up, up”) and hold their breath (verbal cues, “keep holding or hold it”). Counting down the last few seconds can be very helpful.

8. Have subject blow their air out until empty or the system terminates the maneuver.

9. A wait period of 4 minutes is required between maneuvers.

10. Perform until acceptability and repeatability criteria are achieved or a maximum of five maneuvers. (See Criteria for Acceptability 3-1)

11. Note comments related to test quality.

Single Breath-Hold Technique (Modified Krogh’s Technique)

The patient exhales to RV and then inspires a vital capacity breath (referred to as the IVC or V_I) from a system such as that in Figure 3-2. A special diffusion gas mixture is delivered either from a spirometer, a reservoir bag, or by means of a demand valve. The diffusion mixture usually contains 0.3% CO, a “tracer” gas, 21% O₂, and the balance is N₂. The tracer gas is usually an insoluble, inert gas such as helium (He), methane (CH₄), or neon (Ne). The tracer used depends on the type of analyzer implemented to analyze the exhaled gas. The traditional method used He (usually about 10%) as the tracer gas. Rapidly responding infrared analyzers (see Chapter 11) have been used for continuous analysis of the small changes in CO. The same type of infrared analyzer can use CH₄ as the tracer gas, so that a single analyzer can rapidly detect changes in both CO and the tracer. Another method uses gas chromatography (see Chapter 11) to detect changes in CO with neon used as the tracer gas.

Figure 3-2 Dlco apparatus. Basic equipment for performing a Dlco test (specifically the Dlcosb). Gas is delivered to the automated valve from a reservoir bag (not shown), the spirometer, or a demand valve (not shown). The automated valve allows the patient to inhale the test gas rapidly. A computer usually displays a volume-time tracing along with signals from the gas analyzer(s) (see Figures 3-3, 3-4, and 3-5). Timing of the maneuver is usually performed by the computer.

After inspiring the VC breath, the patient holds the breath at total lung capacity (TLC) for approximately 10 seconds. The patient then exhales. After a suitable washout volume (750–1000 mL) has been discarded, a sample of alveolar gas is collected. The alveolar sample may be collected in a small bag (approximately 500 mL or less) or by continually aspirating a sample of the exhaled gas.

The sample is analyzed to obtain the fractional CO and tracer gas concentrations in alveolar gas, F_Aco_T (where T is the time of the breath hold), and F_Atracer, respectively. The concentration of CO in the alveoli at the beginning of the breath hold (F_Aco₀) must be determined as well. It is calculated as follows:

FACO0=F1CO×FAtracerFI tracer

where:

F_Aco₀ = fraction of CO in alveolar gas at beginning of breath hold (time = 0)

F_Ico = fraction of CO in reservoir (usually 0.003)

F_Atracer = fraction of tracer in alveolar gas sample

F_Itracer = fraction of tracer in inspired gas (varies with tracer gas used)

The change in tracer gas concentration reflects dilution of inspired gas by the gas remaining in the lungs (i.e., RV). This change is used to determine the CO concentration at the beginning of the breath hold, before diffusion from the alveoli into the pulmonary capillaries. The Dlcosb (single-breath) is then calculated as follows:

DLCOsb = VA × 60(PB−47) × (T) × LnFACO0FACOT

where:

V_A = alveolar volume, mL (STPD)

60 = correction from seconds to minutes

P_B = barometric pressure, mm Hg

47 = water vapor pressure at 37 °C, mm Hg

T = breath-hold interval, seconds

Ln = natural logarithm

F_Aco₀ = fraction of CO in alveolar gas at beginning of breath hold

F_Aco_T = fraction of CO in alveolar gas at end of breath hold

V_A may be calculated from the single-breath dilution of the tracer gas:

VA=(VI−VD)×FItracerFAtracer×STPD correction factor

where:

V_I = volume of test gas inspired, mL (see Figure 3-3)

Figure 3-3 Dlcosb maneuver tracing. Tracing of a single-breath Dlco maneuver proceeding from left to right *(heavy line)*. After exhaling to RV (0 L on the volume axis), the patient rapidly inspires a vital capacity breath (V_I) of the test gas, then holds the breath at TLC for approximately 10 seconds. At the end of the breath hold, the patient exhales rapidly. The dead space washout volume is discarded, an alveolar gas sample is collected, then analyzed. The recommended timing method is the Jones and Meade method.

V_D = dead space volume (anatomic and instrumental), mL

F_Atracer = fraction of tracer in alveolar gas sample

F_Itracer = fraction of tracer in inspired gas (depends on tracer used)

Note that the V_A, usually expressed in BTPS units, must be converted to STPD for the single-breath calculation. The dilution of tracer gas is used twice to determine the CO concentration at the beginning of the breath hold and to determine the lung volume (i.e., V_A) at which the breath hold occurred.

A simplification of the single-breath method described earlier is widely used. The tracer gas and CO analyzers may be calibrated to read full scale (100% or 1.000) when sampling the diffusion mixture, and to read zero when sampling air (no tracer or CO). If the analyzers have a linear response to each other, the fractional concentration of the tracer gas in the alveolar sample is equal to the F_Aco₀. This technique assumes that both the tracer gas and CO are diluted equally during inspiration. Because no tracer gas leaves the lung during the breath hold, its concentration in the alveolar sample approximates that of the CO before any diffusion occurred. The exponential rate of CO diffusion from the alveoli can then be expressed as follows:

Ln(FAtracerFACOT)

where:

F_Atracer = fraction of tracer in alveolar gas, equal to F_Aco₀

F_Aco_T = fraction of CO in alveolar gas at end of breath hold

Ln = natural logarithm of the ratio

This technique avoids the necessity of analyzing the absolute concentrations of the two gases. However, it requires that the analyzers be linear with respect to each other. Analysis of CO is often done using infrared analyzers (see Chapter 11), and their output is nonlinear. Care must be taken to ensure that corrected CO readings are used in the computation. These corrections are easily accomplished either electronically or via software in computerized systems. Systems that use the same detector for both CO and the tracer gas (e.g., infrared, gas chromatography) also need to provide linear output. The linearity of the system should be within 0.5% of full scale. This means that any drift or nonlinearity should cause no more than a 0.5% error when analyzing a known gas concentration (See Chapter 12).

Dlco gas analysis is commonly performed with either a rapidly responding multigas analyzer or gas chromatography. Multigas analysis uses specialized infrared analyzers capable of detecting several gases simultaneously. These systems use methane (CH₄) as a tracer gas. One advantage of multigas analysis is that CO and CH₄ are measured rapidly and continuously (Figure 3-4) so that the calculated Dlco is available as soon as the exhalation has been completed. Gas chromatography (see Chapter 10) can also be used for Dlco gas analysis. Neon (Ne) is used as the tracer gas (Figure 3-5). Helium is used as a “carrier” gas for the chromatograph. Although gas analysis using chromatography is slow (60–90 seconds), it is extremely accurate.

Figure 3-4 Dlcosb maneuver using continuous gas analysis. Changes in gas concentrations are shown *(top)*. Test gases (CO and CH₄) rise rapidly to their initial values of 0.3% during the breath hold. During exhalation, both gas concentrations fall as dead space is washed out. The tracer gas, CH₄, shows a plateau as alveolar gas is exhaled. CO shows a similar pattern, but with a lower concentration because of uptake (diffusion). Gas concentration measurements are made from an alveolar window *(gray lines)* that can be adjusted. Simultaneous changes in lung volumes are shown on the bottom tracing. IVC indicates inspiratory volume (V_I). In this test, the patient failed to inspire at least 85% of the VC *(dashed gray line)*. BHTs and BHTe indicate the breath-hold timing scheme. Calculated Dlco and related measurements are displayed allowing for inspection of changes as the alveolar sample “window” is adjusted. (Courtesy CareFusion Healthcare Critical Care Division, Palm Springs, Calif.)

Figure 3-5 Dlcosb test using gas chromatography. Volume-time tracing with inspiration, breath-hold, and alveolar sampling is shown *(left graph)*. In this scheme, inspiration causes a downward deflection. Output of the gas chromatograph is shown *(right graph)*. Neon (the tracer gas) and CO show distinct peaks on the gas chromatograph tracing. The alveolar sample is superimposed over a pre-analysis (calibration) curve of the test gas. Ne_I is the initial tracer concentration, and Ne_A is the concentration after the breath hold. CO concentrations of the undiluted test gas (CO₀) and end of the breath hold (CO_T) are superimposed. Calculated Dlco and related values for multiple efforts are shown at the top. (Courtesy Medical Graphics, Inc., St. Paul, Minn.)

The resistance of the breathing circuit should be less than 1.5 cm H₂O/L/sec, at a flow of 6 L/sec. This is important in allowing the patient to inspire rapidly from RV to TLC. A demand valve may be used instead of a reservoir bag for the test gas. In a demand-flow system, the maximal inspiratory pressure to maintain a flow of 6 L/sec should be less than 10 cm H₂O. Increased resistance, in either a reservoir or a demand valve system, may cause the patient to produce large subatmospheric pressures during inspiration. This has the effect of increasing pulmonary capillary blood volume and may falsely increase Dlco.

The timing device for the maneuver should be accurate to within 100 msec over a 10-second interval (1%). Most computerized systems time the maneuver automatically. However, a means of verifying the accuracy of the breath-hold time should be available. The Jones and Meade method of timing the breath hold should be used (see Figure 3-5). The Jones and Meade method measures breath-hold time from 0.3 of the inspiratory time to the midpoint of the alveolar sample collection.

Corrections must be made for the patient’s anatomic dead space (V_D) and for dead space in the valve (and sample bag, if used). Anatomic V_D should be calculated as 2.2 mL/kg (1 mL/lb) of ideal body weight. The equipment manufacturer should specify instrument V_D. Instrument V_D should not exceed 350 mL for adult subjects, including mouthpiece and any filters that might be used. Smaller instrument V_D may be necessary for pediatric patients. Anatomic and instrument V_D are subtracted from inspired volume (V_I) before the alveolar volume (V_A) is calculated.

All gas volumes must be corrected from ATPS to STPD for the Dlco calculations. However, when the V_A is used to calculate the ratio of Dlco to lung volume (Dl/V_A), it is normally expressed in BTPS units. Accurate measurement of inspired volume during the maneuver requires that the spirometer have an accuracy of 3.5% (3% + 0.5% for the calibration syringe itself) over a range of 8 L. Volume-based spirometer systems must also be free from leaks.

Gas analyzers that are affected by carbon dioxide (CO₂) or water vapor require appropriate absorbers. Absorption of CO₂ is usually accomplished with a chemical absorber using Ba(OH)₂ (baralyme) or NaOH (soda lyme). Each of these reactions produces water vapor. Therefore, CO₂ absorbers should be placed upstream of an H₂O absorber. Anhydrous CaSO₄ is commonly used to remove water vapor. Selectively permeable tubing (PERMA PURE®) can also be used to establish a known water vapor content. Gas-conditioning devices must be routinely checked to ensure accurate gas analysis. Chemical absorbers typically add an indicator that changes color as the absorber becomes exhausted.

PF Tip 3-1

Watch the patient carefully during the Dlcosb effort to detect either a Valsalva or Müller maneuver. Widely varying results in two or more Dlco measurements are often caused by the patient failing to relax during the breath hold. Some pulmonary function systems provide a pressure monitor to detect large positive or negative pressures during the breath hold.

Dlcosb maneuvers should be performed after the patient has been seated for at least 5 minutes. The patient should refrain from exertion immediately before the test; exercise increases cardiac output, which increases Dlco. The patient should be instructed about the requirements of the maneuver and the technique demonstrated. Expiration to RV should be of a reasonable duration, usually 6 seconds or less. Patients who have airway obstruction may have difficulty exhaling completely in this interval. Inspiration to TLC should be rapid but not forced. Healthy subjects and patients with airway obstruction should be able to inspire at least 85% of their VC within 4 seconds (Criteria for Acceptability 3-1). The single-breath calculation assumes instantaneous filling of the lung. Prolonged inspiratory times will decrease the actual time of breath holding at TLC, typically resulting in a lower Dlco. The breath hold should be relaxed, against either the closed glottis or a closed valve. The patient should avoid excessive positive intrathoracic pressure (Valsalva maneuver) or excessive negative intrathoracic pressure (Müller maneuver). A Valsalva maneuver reduces pulmonary capillary blood volume and may produce a falsely low Dlco. A Müller maneuver increases pulmonary capillary blood volume and may falsely increase Dlco. Expiration after the breath hold should be smooth and uninterrupted. Exhalation should take less than 4 seconds, and alveolar gas sampling should occur in less than 3 seconds (again because the calculations assume instantaneous emptying of the lung). Patients who have moderate or severe airway obstruction may have difficulty achieving these criteria. The breath-hold time, measured using the Jones and Meade method (see Figure 3-3), should be 10 seconds ± 2 seconds. Prolonged inspiratory or expiratory times may result in breath-hold times greater than 12 seconds and should be noted on the report.

Criteria For Acceptability 3-1 Dlco_sb

1. System has passed calibration and quality control procedures.

2. Inspiration from RV to TLC should be rapid and should occur within 4 seconds or less.

3. The volume inspired (V_I or IVC) should be at least 85% of the best previously recorded VC.

4. Breath-hold time should be between 8 and 12 seconds, with no evidence of leaks, or Valsalva or Müller maneuver.

5. Exhalation should be rapid with total exhalation lasting 4 seconds or less with appropriate clearance of VD and proper sampling/analysis of alveolar gas

6. An interval of at least 4 minutes should elapse between repeated tests. No more than five single-breath maneuvers should be performed.

7. The V_A is consistent with the clinical presentation. In a normal subject, the TLC–V_A relationship should be very close, if not the same, and TLC performed by any method in any disease state should be larger than V_A.

8. The average of two or more acceptable tests should be reported. Duplicate determinations should be within 3 mL/min/mm Hg of each other or within 10% of the largest observed value.

PF Tip 3-2

One of the most common problems encountered when performing the Dlcosb maneuver is failure of the subjects to inspire at least 85% of their previously measured VC. Careful instruction or reinstruction of the subject usually helps eliminate the problem, including making sure they expire completely before taking the maximal inhalation of test gas. Patients who have airway obstruction may have difficulty inspiring maximally after exhalation to residual volume (RV).

PF Tip 3-3

Dlco is affected by both Hb and COHb levels in the patient’s blood. A low Hb (anemia) or high COHb reduces the measured Dlco. A high Hb (polycythemia) causes the Dlco to appear elevated. As a “rule of thumb,” a 10% COHb will reduce the measured Dlco by 10%.

To obtain an alveolar sample, dead space gas needs to be washed out (i.e., discarded). A washout of 0.75–1.0 L is usually sufficient to clear the patient and sampling device dead space. For patients with small vital capacities (>2.0 L), the washout volume may be reduced to 0.5 L. A sample volume of 0.5–1.0 L should be collected, but a smaller sample may be necessary in patients whose VC is less than 1 L. In Dlcosb systems that analyze expired gas continuously (see Figure 3-4), inspection of the washout of the tracer gas may be used to select an appropriate alveolar sample. Rapidly responding gas analyzers that measure the CO and tracer gas simultaneously allow adjustment of washout volume (i.e., dead space) and sample volume after completion of the maneuver. These adjustments may be particularly useful in subjects who have very small vital capacities (e.g., pediatric patients or adults with severe restrictive disease). Such adjustments assume that both the CO and tracer gas concentrations reflect changes occurring at the mouth. Alveolar sampling may be adjusted to begin at the point where the tracer gas and CO indicate an “alveolar plateau” (see Figure 3-4). In patients who have uneven mixing or emptying of the lungs, there may not be a clear demarcation between dead space and alveolar gas. Adjustments to either washout (dead space) volume or sample volume should be noted on the report.

Two or more acceptable Dlcosb maneuvers (see Criteria for Acceptability 3-1) should be averaged. Duplicate determinations should be within 3 mL CO/min/mm Hg of each other, or within 10% of the largest value obtained from an acceptable effort. No more than five repeated maneuvers should be performed because of the effect of increasing carboxyhemoglobin (COHb) from inhalation of the test gas. There should be a 4-minute delay between repeated maneuvers to allow for washout of the tracer gas from the lungs.

Corrections for abnormal hemoglobin (Hb) concentrations should be applied with a current Hb value. The predicted Dlco should be corrected so that it reflects the Dlco at an Hb value of 14.6 g% for adult and adolescent males, and to an Hb value of 13.4 g% for women and children of either sex younger than 15 years of age. The Hb-corrected predicted value for males may be calculated as follows:

DLCO (predicted for Hb)=DLCO (predicted)×(1.7×Hb)10.22+Hb)

DLCO (predicted for Hb)=DLCO (predicted)×(1.7×Hb)(9.38+Hb)

Note that this scheme corrects the predicted value rather than the patient’s measured value. For example, in an adult male patient with an Hb of 9.0 g/dL and a predicted Dlco of 25 mL/min/mm Hg, a measured Dlco of 19 mL/min/mm Hg would be reported:

	Actual	Predicted	%
Dlco mL/min/mm Hg	19.0	25.0	76
Dlco (Hb corrected) mL/min/mm Hg	19.0	19.9	95

Both uncorrected and corrected predicted Dlco values and the resulting percentages should be reported, along with the Hb value used for correction. Decreased Hb levels (anemia) will always reduce the predicted value, whereas elevated Hb (polycythemia) will increase the predicted value.

Correction for the presence of COHb in the patient’s blood is also recommended. The predicted Dlco may be adjusted as follows:

DLCO (predicted for COHb)=DLCO (predicted)× (102%−COHb%)

The COHb% is the fraction of carboxyhemoglobin determined by hemoximetry expressed as a percentage. This method assumes that the predicted value already includes 2% COHb in healthy subjects. For subjects who have a COHb% greater than 2%, the predicted value will always be reduced using this method. Patients should be asked to refrain from smoking for 24 hours before the test to reduce the CO back-pressure in the blood. For patients who continue to smoke, the time of last exposure should be recorded.

Dlco varies inversely with changes in alveolar oxygen pressure (P_AO₂). P_AO₂ changes as a function of altitude, as well as with the partial pressure of oxygen in the test gas. Dlco increases approximately 0.35% for each mm Hg decrease in P_AO₂, or about 0.31% for each mm Hg decrease in P_IO₂. When test gas mixtures that produce an inspired O₂ pressure of 150 mm Hg (i.e., 21% at sea level) are used, Dlco values will be equivalent to those measured at sea level. Alternatively, standard test gas (F_IO₂ = 0.21 is typical) can be used and the predicted Dlco corrected by adjusting P_IO₂. For a gas with a P_IO₂ of 150 mm Hg, the equation is as follows:

Dlco predicted for altitude = Dlco predicted/(1.0+0.0031[P_IO₂-150])

where:

PIO2=0.21(PB−47)

and P_B is the local barometric pressure (at altitude). If the patient is breathing supplemental O₂ and the P_AO₂ is measured, the predicted Dlco can be adjusted, assuming a P_AO₂ of 100 mm Hg breathing air at sea level:

DLCO predicted for PAO2=DLCO predicted(1.0+0.0035 [PA−100])

where:

P_AO₂ = measured or estimated alveolar oxygen partial pressure.

Note that corrections for altitude or elevated alveolar oxygen tensions are made to the predicted Dlco values.

Corrections for Hb, COHb, and altitude or elevated P_AO₂ are recommended for all predicted Dlco values when the conditions for the corrections are known. Hb, COHb, and measured P_AO₂ may not be available for all patients; correction for altitude (P_IO₂) is easily performed for laboratories significantly above sea level. Previous guidelines recommended that corrections be applied to the patient’s measured values, rather than to the predicted values. Some laboratories may prefer to use the older method.

In some instances, it may be appropriate to correct the Dlco for the lung volume at which it is measured. A common example would be when the subject inhales a volume that is substantially less than his or her known VC and breath holds at a V_A that is less than expected. The Dlco may be corrected:

DLCO (at VAm)=DLCO (at VAp)×(0.58+0.42(VAmVAp))

where:

V_Am = measured alveolar volume

V_Ap = predicted alveolar volume (i.e., TLC-V_D)

DLVA(at VAm)=DLVA(at VAp)×(0.42+0.58 (VAmVAp))

These corrections are derived from healthy subjects whose Dlco was measured at alveolar volumes less than the expected value. Such corrections may not be applicable in all disease states because Dl and V_A can be altered independently of one another. In addition, some subjects may not exhale completely to RV, with the subsequent inspiration to TLC producing a V_I that is less than 85% of their VC. In such an instance the subject is actually breath holding at TLC, although the reduced V_I suggests otherwise.

Rebreathing Technique

The patient rebreathes from a reservoir containing a mixture of 0.3% CO, tracer gas, and air (or an O₂ mixture) for 30–60 seconds at a rate of approximately 30 breaths/min. The final CO, tracer, and O₂ concentrations in the reservoir are measured after this interval. An equation similar to that used for the single-breath technique is used (see Criteria for Acceptability 3-1):

DLCOrb=VS×60(PB−47)(T2−T1)×Ln(FACOT1FACOT2)

where:

V_S = volume of lung reservoir system (initial volume × F_Itracer/F_Atracer)

60 = correction from seconds to minutes

P_B = barometric pressure, mm Hg

47 = water vapor pressure, mm Hg

T2 -T1 = rebreathing interval, seconds

Ln = natural logarithm

F_Aco_T1 = fraction of CO in alveolar gas at beginning of the rebreathing

F_Aco_T2 = fraction of CO in alveolar gas at the end of the rebreathing

The rebreathing method can also be implemented using a rapidly responding analyzer (for CO and tracer gas) and plotting the slope of the change in CO in relation to the slope of the tracer gas to estimate the rate of CO uptake. The rebreathing method can be used during exercise.

PF Tip 3-4

The V_A calculated from the single-breath Dlco technique is essentially a “single breath” helium dilution measurement of lung volume and should never be larger than TLC calculated by any other method. If the V_A is larger, there is a technical issue with either the measurement of TLC or Dlco, and the results should be questioned.

Slow Exhalation Single-Breath Intrabreath Method

The patient inspires a vital capacity (VC) breath of test gas containing 0.3% CO, 0.3% CH₄ (methane), 21% O₂, and the balance N₂. Then the patient exhales slowly and evenly at approximately 0.5 L/sec from TLC to RV. A rapidly responding infrared analyzer monitors CO and CH₄ gas concentrations. The exponential rate of disappearance of CO can be calculated in a manner similar to the rebreathing method. Change in V_A is calculated from the change in concentration of the CH₄ tracer gas. CH₄ is used as the tracer gas because it can be rapidly measured using an infrared analyzer. Multiple estimates of Dlco can be made during a single exhalation, recording Dlco as a function of lung volume. This is done using an equation similar to that used for the single-breath method. Instead of one estimate of V_A (equal to the lung volume at breath hold), multiple increments of V_A are made, and Dlco is plotted against lung volume. A single estimate of overall Dlco can also be obtained. The intrabreath method can also be used during exercise.

Membrane Diffusion Coefficient and Capillary Blood Volume

The patient performs two Dlcosb tests, each at a different level of alveolar PO₂. The first Dlcosb test is performed as described previously. The patient then breathes an elevated concentration of O₂ (balance N₂) for approximately 5 minutes, exhales to RV, and performs the second Dlcosb maneuver. Dlco values are calculated for both the air- and oxygen-breathing maneuvers. The total resistance caused by the alveolocapillary membrane (Dm) and the resistance caused by the rate of chemical combination with Hb and transfer into the red blood cell (θV_C) is calculated as follows:

1DLCO=1Dm+1θVc

where:

1/Dlco = reciprocal of diffusing capacity, or resistance

1/Dm = alveolocapillary membrane resistance

1/θV_C = resistance of red blood cell membrane and rate of reaction with Hb

θ = transfer rate of CO/milliliter of capillary blood

V_C = capillary blood volume

Because CO and O₂ compete for binding sites on Hb, measurement of diffusion of CO at different levels of alveolar PO₂ can be used to distinguish resistance caused by the alveolocapillary membrane from resistance caused by the red blood cell membrane and Hb reaction rate. V_C is presumed to remain the same for both tests, but θ varies in response to changes in PO₂. Resistance caused by the alveolocapillary membrane (1/Dm) can be calculated by plotting θ at two points against 1/Dlco and extrapolating back to zero (as if no O₂ were present).

The membrane component of resistance to gas transfer can also be estimated by measuring the rate of uptake of nitric oxide (NO). Dlno has been suggested as a direct measure of the conductance of the alveolocapillary membrane. Because NO combines with Hb approximately 280 times faster than CO, the rate of NO uptake by the blood (θ_NO) is very large and 1/θ_NO V_C is negligible compared to 1/Dm for NO. Therefore, Dlno reflects the membrane resistance to gas diffusion in the lungs. Dlno can be measured with either a single-breath or rebreathing technique. A small amount of NO is added to the diffusion mixture and the uptake measured using a chemoluminescence analyzer (see Chapter 11).

Significance and Pathophysiology

See Interpretive Strategies 3-1. The average Dlco value for resting adult patients by the single-breath method is approximately 25 mL CO/min/mm Hg (STPD) with significant variability. Women have slightly lower normal values, presumably because of smaller normal lung volumes. The expected Dlco value in a healthy patient varies directly with the patient’s lung volume. In some instances, it may be appropriate to adjust the predicted Dlco to account for decreased lung volume (see the section about techniques in the chapter). Dlco values can increase two to three times in healthy individuals during exercise in response to increased pulmonary capillary blood flow.

Interpretive Strategies 3-1 Dlco

1. Were the test maneuvers performed acceptably? Were the tests repeatable within 3 mL/min/mm Hg of each other or within 10% of the largest observed value?

2. Were all appropriate corrections made to the predicted values? Hb? COHb? Altitude? V_A?

3. Are the reference values appropriate? Age? Height? Sex? Race?

4. Is Dlco less than the lower limit of normal (LLN) after appropriate corrections? If so, a gas exchange abnormality likely exists. Evaluate Dl/V_A.

5. Is the Dl/V_A ratio within normal limits? If so, the reduced diffusing capacity is likely related to decreased lung volumes, parenchymal changes, pulmonary vascular disease, or pulmonary hypertension. Consider clinical correlation (e.g., patient’s history, presentation, symptoms).

6. Is the Dl/V_A ratio reduced? If so, reduced diffusing capacity is likely related to airway obstruction (uneven distribution of ventilation, /mismatch) or increased dead space. Compare V_A and TLC; a large difference suggests uneven distribution of ventilation. Look for a clinical correlation.

7. Is the Dlco increased after correction for Hb or altitude? If so, consider possible causes of increased pulmonary blood volume, hemorrhage, obesity, or left-to-right shunts. Also consider undiagnosed asthma. Look for clinical correlation.

8. Is Dlco less than 60% of predicted? If so, consider additional tests (blood gases, exercise desaturation study).

Most reference equations use age, height, sex, and race to predict Dlco. Some reference equations use body surface area (BSA) to calculate expected values. If the patient’s weight is used (i.e., to calculate BSA), the ideal body weight is recommended. Using the actual body weight in obese patients can result in erroneously large predicted values unless similar subjects were included in the reference population. Significant differences exist among reference equations. These discrepancies result, in part, from different methods used to measure Dlco in various laboratories. Laboratories should check the appropriateness of their reference equations by comparing the results obtained from healthy subjects. They should measure Dlco in a sample of healthy patients of each sex and compare the results using several reference equations. If the reference equations used are appropriate, the differences between the measured and expected values (i.e., residuals) for the healthy patients should be minimal. Predicted values for Dlco and for Dl/V_A should be taken from the same reference set. Regression equations for calculation of expected Dlco values are included in Chapter 13.

Dlco is often decreased in restrictive lung diseases, particularly those associated with pulmonary fibrosis. Fibrotic changes in the lung parenchyma are associated with asbestosis, berylliosis, and silicosis. Many other diseases caused by inhalation of dusts also result in fibrotic changes in lung tissue. Idiopathic pulmonary fibrosis, sarcoidosis, systemic lupus erythematosus, and scleroderma are also commonly associated with a reduction in Dlco. Inhalation of toxic gases or organic agents may cause inflammation of the alveoli (alveolitis) and decrease Dlco. These disease states are sometimes categorized as diffusion defects. The decrease in Dlco is probably more closely related to the loss of lung volume, alveolar surface area, or capillary bed than to thickening of the alveolocapillary membranes. Dlco also decreases when there is loss of lung tissue or replacement of normal parenchyma by space-occupying lesions such as tumors.

Dlco may also be reduced in the presence of pulmonary edema. Disruption of alveolar ventilation and reduction of lung volume as well as congestion of the alveoli cause the reduction in Dlco in edema. In the early stages of congestive heart failure (CHF), Dlco may be normal or slightly increased. As the left ventricle decompensates, pulmonary vessels become engorged. The increased blood volume can cause Dlco to increase, until the congestion becomes advanced. In most patients with heart failure, Dlco is decreased because of the restrictive ventilatory pattern. Dlco in patients who receive a heart transplant for chronic heart failure does not return to normal, as might be expected.

Dlco may also be decreased as a result of medical or surgical intervention for cardiopulmonary disease. Lung resection for cancer or other reasons typically results in decreased Dlco. The extent of reduction is usually directly proportional to the volume of lung removed. An exception to this pattern occurs in lung volume reduction surgery (LVRS) and in bullectomy. These surgical procedures typically resect areas of the lung that have little or no blood flow. Lack of perfusion is documented by a lung scan. Excision of tissue in such areas reduces lung volume without necessarily reducing the surface area available for diffusion. Improved ventilation-perfusion matching in the remaining lung often results in an improvement in Dlco.

Radiation therapy that involves the lungs usually causes a decrease in Dlco. Radiation causes pneumonitis that commonly results in fibrotic changes. Drugs used in chemotherapy (e.g., bleomycin) and those used to suppress rejection in organ transplantation may cause reductions in Dlco. These drugs appear to directly affect the alveolocapillary membranes. Some drugs used in the treatment of cardiac arrhythmias (e.g., amiodarone) have been shown to decrease Dlco. For this reason, Dlco is commonly used to monitor drug toxicity.

Dlco may also be helpful in evaluating disorders such as hepatopulmonary syndrome, in which gas exchange and pulmonary vascular defects coexist. Diseases that affect the pulmonary vascular bed also typically result in decreased Dlco. These include pulmonary vasculitis and pulmonary hypertension. Pulmonary vascular disease often manifests itself as a reduced diffusing capacity with otherwise normal pulmonary function.

Dlco may be decreased in both acute and chronic obstructive lung disease. Dlco is decreased in emphysema for several reasons. Emphysematous lungs have a reduced surface area for gas exchange, with the loss of both alveolar walls and their associated capillary beds. As a result of the decreased surface area, less gas can be transferred per minute even if the remaining gas exchange units are structurally normal. In addition to loss of surface area for gas exchange, the distance from the terminal bronchiole to the alveolocapillary membrane increases in emphysema. As alveoli break down, terminal lung units become larger. Gas must diffuse farther just to reach the alveolocapillary surface. There is also a mismatching of ventilation and pulmonary capillary blood flow in emphysema. Disruption of alveolar structures causes loss of support for terminal airways. Airway collapse and gas trapping result in ventilation-perfusion (/) abnormalities (Figure 3-6).

Figure 3-6 Dlco and disease. Some of the more common disease disorders which affect gas exchange. **(A)** Depicts the normal alveolar capillary interface where gas exchange occurs; **(B)** Emphysema–destruction of the alveolar capillary interface and hyperinflation; **(C)** Anemia – reduction in oxygen–carrying capacity; **(D)** Pulmonary emboli and/or A-V malformations affect oxygen transport; **(E)** Interstitial lung disease such as pulmonary fibrosis scars the alveolar membrane surface; **(F)** Alveolar volume loss from resection, alveolar collapse, or filling processes.

Other obstructive diseases (e.g., chronic bronchitis, asthma) may not reduce Dlco unless they result in markedly abnormal /patterns. Dlco is sometimes used to differentiate among these obstructive patterns. Low Dlco in the presence of obstruction is sometimes assumed to be evidence of emphysema. However, /mismatching can cause Dlco to appear to be decreased in asthma, chronic bronchitis, or emphysema. Some asthmatic patients may have an increased Dlco, but the cause is not completely understood.

Dlco measurements at rest have been suggested to estimate the probability of O₂ desaturation during exercise. A large retrospective study demonstrated a Dlco less than 60% correlated with O₂ desaturation during exercise (75% sensitivity and specificity). Patients with restrictive lung disease and a low resting Dlco are at risk of O₂ desaturation, even with low levels of exercise. Low resting Dlco may indicate the need for assessment of oxygenation during exercise.

Dlco is directly related to lung volume (V_A) in healthy individuals. The Dl/V_A (also termed K_CO) may be multiplied by the lung volume at which the measurement was obtained to express Dlco. This calculation is simple because V_A must be measured to derive Dlco (see Figure 3-5). In healthy subjects, and even in those with mild to moderate restriction, V_A approximates the TLC minus the assumed dead space (V_D). Analysis of this relationship can be useful to differentiate whether decreased Dlco is the result of loss of lung volume (as in restriction) or from some other cause. In healthy individuals, alveolar volume and Dlco are proportional to body size (height). Two patients of different heights will have different Dlco and V_A values, but their Dl/V_A ratios will be similar. In healthy adults, Dl/V_A is approximately 4–5 mL CO transferred/minute/liter of lung volume.

V_A is measured by the dilution of the tracer gas used (in the Dlcosb) and reflects the same volume into which CO is distributed and diffuses across the pulmonary capillary membranes. Mismatching of ventilation and blood flow (as in obstructive disease) can cause a significant portion of the lung to not participate in gas exchange. This is usually characterized by a difference between the V_A measured during the Dlco maneuver and the TLC measured by plethysmography or multiple-breath gas dilution. When the Dlco is reduced but the Dl/V_A is normal or near normal, the decrement in gas exchange can be assumed to be caused by an uneven distribution of ventilation, rather than loss of lung volume.

In the presence of pulmonary disease, both Dlco and V_A may be affected. Dlco goes down as the lung empties but does so in a nonlinear fashion. In obstruction, low Dlco without reduction in V_A results in a low ratio. In a purely restrictive process, a decrease in Dlco reflects a loss of V_A and the Dl/V_A ratio is preserved. For example, a patient who has a Dlco of 12 mL CO/min/mm Hg (50% of predicted) and a V_A of 3.0 L would have a Dl/V_A ratio of 4. This reduction in Dlco is roughly proportional to a loss of lung volume. Some pulmonary conditions (such as pneumonectomy) may result in an increased Dl/V_A, where gas exchange is preserved and lung volume is decreased.

Dlco and Dl/V_A may also be affected if the patient performs the breath-hold maneuver at a lung volume less than TLC. The predicted Dlco may be corrected for the reduced V_A, as described previously. There are important implications for interpretation of diffusing capacity in the complex relationship between Dlco and Dl/V_A. Patients who fail to inspire fully during the maneuver will have a decreased Dlco, but the Dl/V_A may appear to increase. Correcting for V_A does not correct for poor inspiratory effort (less than 85% of VC). In patients who have a low Dlco, a “normal” Dl/V_A should not be confused with normal gas exchange.

The Dlcosb is the most widely used method because it is relatively simple to perform, is noninvasive, and is more standardized than other methods. The rapidity with which repeated maneuvers can be performed also lends to its popularity. Most automated systems use the Dlcosb, contributing a certain degree of standardization to the methodology. Large differences in reported Dlco values exist between laboratories. This variability has been attributed to different testing techniques, problems in the gas analysis involved in the test, and differences in computations. Breath holding at TLC is not a physiologic maneuver. This and the fact that Dlco varies with lung volume cause some concerns about Dlcosb as an accurate description of diffusing capacity. Dlcosb is not practical for use during exercise. Some patients have difficulty expiring fully, inspiring fully, or holding their breath. The American Thoracic Society-European Respiratory Society (ATS-ERS) have provided guidelines to improve the standardization of the single-breath maneuver (Box 3-1).

Box 3-1 Dlcosb Recommendations

A Equipment

1. Volume accuracy same as for spirometry: ±3.5% over 8-L range, all gases and flow conditions (includes syringe accuracy of 0.5%); checked daily.

2. Documented analyzer linearity from 0 to full span ±0.5%, with minimal drift over test interval; checked quarterly.

3. Circuit resistance less than 1.5 cm H₂O at 6 L/sec.

4. Demand valve sensitivity less than 10 cm H₂O to generate 6 L/sec flow.

5. Timing mechanism accurate to ±1% over 10 sec; checked quarterly.

6. Instrument dead space (e.g., valves, filters) less than 0.35 L.

7. Perform a 3-L syringe Dlco weekly in patient testing mode (see Chapter 12).

8. Validate system by testing healthy nonsmokers (biologic controls) or Dlco simulator weekly.

B Patient preparation

1. The patient should refrain from smoking on the day of the test; time of last cigarette should be noted.

2. No alcohol ingestion on the day of the test.

3. No exercise immediately before test; patient should be seated for 5 minutes before Dlco maneuver.

4. Patients on supplemental O₂ should be switched to breathing air for 10 minutes before testing, if clinically acceptable.

5. The patient should be carefully instructed and the procedure demonstrated.

C Technique (see How-To Box 3-1)

D Calculations

1. Average at least two acceptable tests; duplicate determinations should be within 3 mL/min/mm Hg of each other, or within 10% of the largest observed value.

2. Use Jones and Meade method of timing the breath hold.

3. Alveolar volume should be determined by single-breath dilution of tracer gas.

4. Adjust for V_D volumes (instrument and patient).

5. Determine inspired gas conditions (ATPS or ATPD).

6. Correct for CO₂ and H₂O absorption.

7. Report Dlco/V_A in mL CO (STPD)/min/mm Hg per L (BTPS).

8. Correct predicted value for Hb, COHb, and altitude (P_IO₂) as appropriate (see text).

9. Consider adjusting predicted value for submaximal inspiration.

10. Use reference equations appropriate to the laboratory method and patient population.

Compiled from MacIntyre NM, Crapo RO, Viegi G, et al. Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur Respir J. 2005; 26:720-735.

The rebreathing method (Dlcorb) requires somewhat complicated calculations but offers the advantages of a normal breathing pattern without arterial puncture. Dlcorb is less sensitive to /abnormalities and uneven ventilation distribution than the Dlcosb. The rebreathing method is sensitive to the accumulation of COHb in the capillary blood and the resultant back-pressure. Capillary Pco is routinely assumed to be zero. The actual alveolocapillary CO gradient at the time of testing can be estimated, although with some difficulty. The rebreathing method can be used during exercise, provided the rebreathing interval is carefully controlled. An additional advantage is that acetylene can be added to the diffusion mixture to measure cardiac output simultaneously. Adding a small amount of nitric oxide (NO) to the diffusion mixture allows the Dlno to be measured during rebreathing as well.

Measurement of Dlco by the intrabreath method (Dlcoib) offers the advantage of not requiring a breath hold at TLC. However, the patient must inspire a large enough volume of test gas so that the subsequent exhalation will clear the instrument and anatomic V_D. In addition, the single-breath exhalation must be slow and even. In some systems, a flow restrictor may be necessary to limit expiratory flow. The single-breath, slow-exhalation method produces values similar to those obtained by the breath-hold method in healthy patients when flow is maintained at 0.5 L/sec. Uneven distribution of ventilation may produce intrabreath Dlco values that are artificially elevated. Because the evenness of ventilation distribution can be assessed from the washout of CH₄, unacceptable Dlco values can be detected. Table 3-1 compares some advantages and disadvantages of Dlco testing methods.

Measurement of membrane (Dm) and red blood cell components (θV_c ) of diffusion resistance in healthy patients reveals that each factor accounts for approximately half of the total resistance to gas exchange across the alveolocapillary membranes. Difficulty in quantifying the partial pressure of O₂ in the lungs (pulmonary capillaries) restricts the use of the membrane-diffusing capacity determination. Because uptake of NO is limited almost entirely by the pulmonary capillary membranes, Dlno can be measured to assess membrane resistance.

Numerous other physiologic factors can influence the observed Dlco:

1. Hemoglobin and hematocrit (Hct). Decreased Hb or Hct reduces Dlco, whereas increased Hb and Hct elevate Dlco. Dlco may be corrected if the patient’s Hb is known. CO uptake varies approximately 7% for each gram of Hb. The predicted Dlco may be corrected so that the value reported is compared to a standardized Hb level of 14.6 g% for men and 13.4 g% for women and children younger than 15 years. When this correction is applied, the predicted Dlco will be reduced (and the percentage of predicted increased) if the patient’s Hb is less than the standard value (14.6 g% or 13.4 g%, respectively). Conversely, the predicted Dlco increases (and the percentage of predicted decreases) if the Hb is greater than the standard value. Both corrected and uncorrected Dlco predicted values should be reported, along with the Hb value used for adjustment. Care should be taken to use Hb values that are representative of the patient’s actual Hb level at the time of the Dlco test.

2. COHB. Increased COHb levels, often found in smokers, reduce Dlco. Smokers may have COHb levels of 10% or greater, causing significant CO back-pressure. The diffusion gradient for CO across alveolocapillary membranes is assumed to equal the alveolar pressure of CO. In healthy nonsmoking patients, very little CO (usually less than 2% COHb) is present in pulmonary capillary blood. When there is carboxyhemoglobinemia, diffusion of CO is reduced because the gradient across the membrane is reduced. COHb also shifts the oxyhemoglobin dissociation curve, further altering gas transfer. Each 1% increase in COHb causes an approximate 1% decrease in the measured Dlco. CO back-pressure corrections can also be made by estimating the partial pressure of CO in the pulmonary capillaries. This pressure can be used to correct the F_Aco₀ and the F_Aco_T. More commonly, however, the predicted Dlco is corrected for the presence of COHb in excess of 2% (see the section, Techniques).

3. Alveolar Pco₂. Increased Pco₂ elevates Dlco because the alveolar Po₂ is necessarily decreased. Significant increases in alveolar Pco₂ (i.e., moderate to severe hypoventilation) reduce the alveolar Po₂.

4. Pulmonary capillary blood volume. Increased blood volume in the lungs (V_C) causes increased Dlco. Increases in pulmonary capillary blood volume may result from increased cardiac output as occurs during exercise. Patients should be seated and resting for several minutes before Dlco testing is performed. Pulmonary hemorrhage and left-to-right shunts may also cause an increase in blood volume in the lungs. In each of these cases, the increase in Dlco is related to the increased volume of Hb available for gas transfer. Excessive negative intrathoracic pressure during breath holding can increase pulmonary capillary volume and elevate the Dlco. Conversely, excessive positive intrathoracic pressure (Valsalva maneuver) can reduce pulmonary blood flow and decrease Dlco.

5. Body position. The supine position increases Dlco. Changes in body position affect the distribution of capillary blood flow.

6. Altitude above sea level. Dlco varies inversely with changes in alveolar oxygen pressure (P_AO₂). At altitudes significantly greater than sea level, Dlco increases unless corrections are made (see the section, Techniques).

7. Asthma and obesity. Asthma and obesity have been associated with an elevated Dlco in some studies. Increased pulmonary capillary blood volume may explain these observations, but the exact physiology is unclear.

Several additional technical considerations may affect the measurement of Dlco (particularly Dlcosb). V_A is calculated from the dilution of a tracer gas during the single-breath maneuver. This technique typically underestimates lung volume in patients who have moderate or severe obstruction. Low estimated V_A results in low Dlco values. Some clinicians prefer to use a separately determined lung volume to estimate V_A. RV, measured independently by plethysmography or one of the gas dilution techniques, can be added to the inspired volume (V_I) to derive V_A. The V_A calculated by this method is usually larger in patients with airway obstruction than V_A calculated from the single-breath dilution method. The resulting estimate of Dlco is larger. The Dlco calculation, however, is based on the volume of gas into which both the inspired CO and tracer gas are distributed, but not the lung volume in which gas mixing is minimal. Some laboratories report Dlco values calculated by both methods. The ATS-ERS recommends the calculation of V_A using the single-breath tracer gas method. Comparison of the single-breath V_A measured during the Dlco maneuver with lung volumes (i.e., TLC) measured by the standard methods is often helpful in elucidating the cause for a low Dlco.

A simple method of evaluating the quality of the test is to compare the TLC–V_A relationship, whereas the V_A should never be larger than the measured TLC regardless of the TLC testing methodology (see Chapter 12).

The method of timing of breath hold also influences the calculation of Dlco (see Figure 3-3). Most systems measure breath-hold time by one of three methods:

1. Jones and Meade method: includes 0.7 of the inspiratory time to the midpoint of the alveolar sample

2. Epidemiology Standardization Project (ESP) method: from the midpoint of inspiration (half of the V_I) to the beginning of alveolar sampling

3. Ogilvie method: from the beginning of inspiration (V_I ) to the beginning of alveolar sampling

Theoretically, breath-hold time is considered the time during which diffusion occurs. That is, the Dlco equation assumes that the entire breath hold is at TLC. However, because some gas transfer takes place during inspiration and expiration, Dlco will be influenced by the proportion of time spent in these two phases. A three-equation method has been proposed that uses separate equations for the three phases of the maneuver (i.e., inspiration, breath hold, expiration). This method has not been widely used, however. The timing method may become significant if the reference values used for comparison were generated by one of the other methods. The Jones and Meade method is the recommended method (see Box 3-1). Rapid inspiration from RV to TLC and rapid expiration to the alveolar sampling phase reduces differences resulting from the timing methods (see Box 3-1). Conversely, patients who display prolonged inspiratory or expiratory times may have reduced Dlco values.

The volume of gas discarded before collecting the alveolar sample may affect the measured Dlco. Most automated systems allow the washout volume to be adjusted, with 0.75–1.0 L most commonly used. Washout volume may need to be reduced to 0.5 L if the patient’s VC is less than 2.0 L. In patients who have obstructive disease, reducing the washout volume may result in increased dead-space gas being added to the alveolar sample. Because dead-space gas resembles the diffusion mixture, Dlco may be underestimated.

Alveolar sampling technique also affects Dlco measurement. Alveolar samples should be collected within 4 seconds, including washout and alveolar sampling. A sample volume of 0.5–1.0 L is recommended. Patients with a small VC (i.e., less than 2.0 L) may require a smaller volume, just as with the washout. When only a small sample is obtained, the gas may not accurately reflect alveolar concentrations of CO and tracer gas, particularly in the presence of /abnormalities. Continuous analysis of the expirate using rapidly responding analyzers allows identification of alveolar gas. Infrared analyzers that can simultaneously analyze the tracer gas and CO will allow the entire breath to be analyzed. These instruments permit adjustment of the alveolar sampling window so that a representative gas sample can be obtained (see Figure 3-4).

Summary

The chapter addresses the measurement of diffusing capacity (Dlco), also referred to as transfer factor:

• Dlco can be measured by various techniques, including the single-breath, rebreathing, and intrabreath methods.

• The single-breath method, or Dlcosb, is the most commonly used. Dlcosb is noninvasive and can be repeated easily to obtain multiple measurements. Many automated Dlcosb systems are available.

• The ATS-ERS and others have published standardization guidelines for Dlcosb. Careful attention to standards and clinical practice guidelines can reduce the variability in Dlcosb measurements in different laboratories.

• Dlco measurements are used diagnostically for a variety of diseases. Because Dlco assesses gas exchange, it is useful in both obstructive and restrictive disease patterns.

• Dlco measurements may be affected by a variety of factors, such as Hb level, COHb, or altitude. Techniques for correcting Dlco for these factors include adjusting the predicted values used for interpretation.

Case Studies

Case 3-1

History

A 55-year-old woman referred to the pulmonary function laboratory because of shortness of breath on exertion. She has a 38-pack-year smoking history but stopped smoking 6 months ago. She still coughs each morning, but her sputum volume has decreased since she stopped smoking. She has no significant environmental or family history of pulmonary disease. She had been using an inhaled β₂ agonist but withheld it for 12 hours before the test.

Pulmonary function testing

Personal Data

Age:	55
Height:	65 in. (165 cm)
Weight:	137 lb (62.3 kg)
Race:	African-American

Spirometry

	Prebronchodilator				Postbronchodilator
	Predicted	LLN*	Actual	% Predicted	Actual	% Predicted	% Change
FVC (L)	2.89	2.16	2.77	96	2.82	98	2
FEV₁ (L)	2.30	1.67	1.91	83	2.01	87	5
FEV_1% (%)	80	70	69		71
FEF_25%–75% (L/sec)	2.33	0.92	1.44	62	1.51	65	5
PEF (L/min)	6.08	4.01	4.01	66	5.13	84	28

^*Lower limit of normal.

Lung Volumes

	Predicted	LLN*	Actual	% Predicted
TLC (L)	4.39	3.42	4.97	113
FRC (L)	2.45		3.10	127
RV (L)	1.58		2.20	139
VC (L)	2.89	2.16	2.77	96
IC (L)	1.94		1.87	96
ERV (L)	0.87		0.90	103
RV/TLC (%)	36		44

^*Lower limit of normal.

Diffusing Capacity

	Predicted	Actual	% Predicted
Dlcosb (mL CO/min/mm Hg)	19.7	10.0	51
Dlcosb corr (mL CO/min/mm Hg)	18.1	10.0	55
V_A (L)	4.39	3.99	91
Dl/V_A	4.49	2.51	46
Hb (g/dL)		11.0

Technologist’s Comments

Spirometry maneuvers met ATS-ERS criteria. Lung volumes by body plethysmography were performed acceptably and met ATS-ERS criteria. All Dlco maneuvers met ATS-ERS criteria (two tests were averaged). Dlco was corrected for an Hb of 11 g/dL.

Questions

1. What is the interpretation of:

a. Prebronchodilator and postbronchodilator spirometry?

b. Lung volumes?

c. Dlco?

2. How are the pulmonary function test results related to the subject’s symptoms?

3. What other tests might be indicated?

4. What treatment might be recommended based on these findings?

Discussion

Interpretation

All maneuvers were performed acceptably. FVC and FEV₁ are within normal limits, but the FEV_1% is below the lower limit of normal, consistent with airway obstruction. After inhaled bronchodilator, there is only a minimal change in FEV₁ and FVC. Lung volumes reveal a slightly increased functional residual capacity and moderately increased RV. The RV/TLC ratio is increased, consistent with air trapping. Dlco is markedly decreased, even after correction for Hb. Dl/V_A is decreased, consistent with an obstructive process.

Impression: Mild obstruction with minimal response to bronchodilator; this should not preclude a therapeutic trial if clinically indicated. Lung volumes suggest air trapping. Dlco is severely decreased even when corrected for Hb.

Cause of Symptoms

This subject has symptoms characteristic of airway obstruction that has progressed to the point where dyspnea on exertion prompted a visit to the physician. Her obstruction appears mild. Her FEV₁ is still above the lower limit of normal (LLN) for her age and height. Her response to bronchodilator therapy seems to indicate obstruction caused by inflammation rather than reversible bronchospasm. Lung volume testing confirms that the obstruction appears to have caused some air trapping. This pattern is not unusual for subjects with chronic bronchitis and emphysema.

Her gas exchange, as measured by Dlco, is markedly impaired. Emphysema reduces Dlco by reducing the alveolocapillary surface area available for diffusion. Chronic bronchitis can reduce Dlco by causing a ventilation-perfusion mismatch. The subject may have both of these disease processes disrupting gas transfer. Her V_A measured by the single-breath inhalation of tracer gas during the Dlco maneuver is about a liter less than her TLC measured in the body plethysmograph. This discrepancy suggests poor distribution of inspired gas and is typical in subjects with some degree of airway obstruction.

Subjects who have reduced Dlco values seldom have normal blood gases. Exertion or exercise often aggravates the gas exchange impairment. Many subjects with markedly reduced Dlco (less than 50%–60% of predicted) display exercise desaturation. That is, their Pao₂ falls to levels of 55 mm Hg or less with exercise. The decrease in Dlco does not, however, accurately predict the degree of desaturation that will occur.

Other Tests

An obvious additional test for this subject would be measuring resting arterial blood gases while she breathes room air. Resting hypoxemia would explain dyspnea on exertion. The subject returned to have a blood gas sample drawn. Her Pao₂ while breathing air was 63 mm Hg, with a saturation of 91%. Because this value did not qualify her for supplemental O₂, an exercise test was performed with an arterial catheter in place (see Chapter 7). At a low workload on the treadmill, her Pao₂ decreased to 51 mm Hg. She was then retested while breathing O₂ via nasal cannula at 1 L/min. She then tolerated more exercise, and her Pao₂ never decreased below 68 mm Hg.

Treatment

On the basis of the results of the exercise evaluation, supplemental O₂ was prescribed for the subject to use during exertion. Because she showed little response to bronchodilators, her inhaled β₂ agonist was discontinued. However, a trial of inhaled steroids (beclomethasone) resulted in a noticeable improvement of symptoms.

Case 3-2

History

A 63-year-old woman with a history of cardiomyopathy, hypertension, and pernicious anemia. She has had episodes of ventricular tachycardia that have been managed by means of an automatic implantable cardiac defibrillator (AICD). She has never smoked and denies cough or sputum production. She experiences shortness of breath with exertion. Her family history includes a sister who had asthma and chronic bronchitis. She has no history of environmental toxin exposure. To manage her arrhythmias, her physician prescribed amiodarone. To monitor the effects of this medication, she was referred for tests before starting the drug and again after 3 months of therapy.

Pulmonary function testing

Personal Data

Sex:	Female
Age:	63
Height:	62 in. (157 cm)
Weight:	131 lb (59.5 kg)
Race:	White

Spirometry

	Pretreatment				3 Months
	Predicted	LLN	Actual	% Predicted	Actual	% Predicted
FVC (L)	2.96	2.31	2.50	84	1.97	67
FEV₁ (L)	2.27	1.72	2.09	92	1.81	80
FEV_1% (%)	77	68	84		92
FEF_25%–75% (L/sec)	2.09	0.94	2.82	135	2.77	133

Lung Volumes (He Dilution)

	Pretreatment				3 Months
	Predicted	LLN	Actual	% Predicted	Actual	% Predicted
TLC (L)	4.88	3.66	3.92	87	3.50	78
FRC (L)	2.54		2.26	89	2.23	88
RV (L)	1.71		1.50	87	1.52	89
VC (L)	2.96	2.31	2.42	82	1.97	67
IC (L)	1.94		1.66	85	1.26	65
ERV (L)	0.83		0.76	91	0.71	85
RV/TLC (%)	38		38		44

Diffusing Capacity (Dlcosb)

	Pretreatment			3 Months
	Predicted	Actual	% Predicted	Actual	% Predicted
Dlco (mL/min/mm Hg)	18.0	12.6	69	9.7	54
Dlcosb corr (mL/min/mm Hg)	16.7	12.6	75	9.7	58
V_A (L)	4.48	3.26	72	3.31	95
Dl/V_A	4.0	3.9		2.9
Hb (g/dL)	13.4	11.3		11.4