Dual Test Gas Pulmonary Diffusing Capacity Measurement During Exercise in Humans Using the Single-Breath Method

Summary

This protocol presents a method to assess pulmonary alveolar-capillary reserve measured by combined single-breath measurement of the diffusing capacity to carbon monoxide (D_L,CO) and nitric oxide (D_L,NO) during exercise. Assumptions and recommendations for using the technique during exercise form the foundation of this article.

Abstract

The combined single-breath measurement of the diffusing capacity of carbon monoxide (D_L,CO) and nitric oxide (D_L,NO) is a useful technique to measure pulmonary alveolar-capillary reserve in both healthy and patient populations. The measurement provides an estimate of the participant's ability to recruit and distend pulmonary capillaries. The method has recently been reported to exhibit a high test-retest reliability in healthy volunteers during exercise of light to moderate intensity. Of note, this technique permits up to 12 repeated maneuvers and only requires a single breath with a relatively short breath-hold time of 5 s. Representative data are provided showing the gradual changes in D_L,NO and D_L,CO from rest to exercise at increasing intensities of up to 60% of maximal workload. The measurement of diffusing capacity and evaluation of alveolar-capillary reserve is a useful tool to evaluate the lung's ability to respond to exercise both in the healthy population as well as in patient populations such as those with chronic lung disease.

Introduction

Exercise leads to a considerable increase in energy demand compared to the resting state. The heart and lungs respond by increasing cardiac output and ventilation resulting in an expansion of the alveolar-capillary bed, mainly the recruitment and distention of pulmonary capillaries¹. This ensures a sufficient pulmonary gas exchange, which can be measured by an increase in pulmonary diffusing capacity (D_L)^2,3,4. The first attempts to measure D_L during exercise date back more than a century^5,6,7. The ability to increase D_L from the resting state is often referred to as the alveolar-capillary reserve^{8, 9}.

Experimentally, the relative contributions of alveolar-capillary membrane diffusing capacity (D_M) and pulmonary capillary blood volume (V_C) to alveolar-capillary reserve can be assessed by different methods, including the classical multiple fractions of inspired oxygen () method¹⁰. An alternative technique that may be useful in this context is the dual-test gas method, in which the D_L to carbon monoxide (CO) and nitric oxide (NO) (D_L,CO/NO) are concurrently measured¹¹. This technique was developed in the 1980s, and takes advantage of the fact that the reaction rate of NO with hemoglobin (Hb) is substantially greater than that of CO, such that the pulmonary diffusion of CO depends more on V_C than does NO. Hence, the main site of resistance (~75%) to CO diffusion is located within the red blood cell, while the main resistance (~60%) to NO diffusion is at the alveolar-capillary membrane and pulmonary plasma¹². The concurrent measurement of D_L,CO and D_L,NO thus permits the assessment of the relative contributions of D_M and V_C to D_L¹², where the change in D_L,NO observed during exercise thus largely reflects the expansion of the alveolar-capillary membrane. An additional advantage of this method when obtaining measurements during exercise is that it involves a relatively short breath-hold time (~5 s) and fewer maneuvers compared to the classical technique, where multiple repeated maneuvers with a standardized 10 s breath-hold are performed at different oxygen levels. Although has recently been applied with a shorter breath-hold time and fewer maneuvers at each intensity¹³. Nevertheless, only permits a total of six D_L,CO maneuvers per session, whereas up to 12 repeated D_L,CO/NO maneuvers can be performed without any measurable effect on the resultant estimates¹⁴. These are important considerations when obtaining measurements during exercise since both a long breath-hold and multiple maneuvers may be difficult to perform at very high intensities or in patient populations who experience dyspnea.

The present paper provides a detailed protocol, including theoretical considerations and practical recommendations on the measurement of D_L,CO/NO during exercise and its use as an index of the alveolar-capillary reserve. This method is easily applicable in the experimental setting and permits the assessment of how diffusion limitation in the lungs may affect oxygen uptake in different populations.

Theory and measurement principles
The D_L,CO/NO method involves a single breath of a gas mixture with the assumption that the gases distribute equally in the ventilated alveolar space after inhalation. The gas mixture consists of several gases including an inert tracer gas. The dilution of the tracer gas in the ventilated alveolar space, as based on its fraction in end-expiratory air, can be used to calculate the alveolar volume (V_A)¹⁵. The gas mixture also includes the test gas CO and NO, both of which are diluted in the ventilated alveolar space and diffuse across the alveolar-capillary membrane. Based on their alveolar fractions, their individual rates of disappearance (k), also termed the diffusing constant, from the alveolar space can be calculated. By convention, the D_L for a test gas measured during a single-breath maneuver, is derived by the following equation¹⁶:

where F_A0 is the alveolar fraction of the test gas (CO or NO) at the onset of the breath-hold of the individual D_L maneuver, while F_A is the alveolar fraction of the test gas at the end of the breath-hold, and tBH is the breath-hold time. D_L is mechanically equivalent to the conductance of the test gas across the alveolar-capillary membrane, through plasma and the red blood cell interior to hemoglobin. It thus depends both on the conductance of D_M and the so-called specific conductance of pulmonary capillary blood (θ), of which the latter depends both on the conductance of the test gas in blood and on its reaction rate with hemoglobin¹⁰. Given that the reciprocal of conductance is resistance, the total resistance to the transfer of a test gas depends on the following resistances in series¹⁰:

These components may be distinguished by concurrently measuring the D_L to CO and NO, because these have different θ-values, and their respective D_L values thus depend differently on V_C. The pulmonary diffusion of CO depends more heavily on V_C than does NO, with the main site of resistance (~75%) to CO diffusion being located within the red blood cell¹². In contrast, the main resistance (~60%) to NO diffusion is at the alveolar-capillary membrane and pulmonary plasma, because the reaction rate of NO with hemoglobin is substantially greater than that of CO. Hence, by concurrently measuring D_L,CO and D_L,NO, changes in both D_M and V_C will markedly affect the former, while the latter will depend much less on V_C, thus permitting an integrative assessment of the factors that determine D_L.

The reporting of D_L,CO/NO metrics may be done using different units. Hence, the European respiratory society (ERS) uses mmol/min/kPa, whereas the American Thoracic society (ATS) uses mL/min/mmHg. The conversion factor between the units is 2.987 mmol/min/kPa = mL/min/mmHg.

Protocol

The Scientific Ethical Committee for the Capital Region of Denmark has previously approved the measurement of D_L,CO/NO at rest, during exercise, and in the supine position in both healthy volunteers and patients with chronic obstructive pulmonary disease (COPD) at our institution (protocols H-20052659, H-21021723, and H-21060230).

NOTE: Before D_L,CO/NO is measured during exercise, a dynamic spirometry, and a cardiopulmonary exercise test (CPET) must be performed. The dynamic spirometry is used for quality control of the individual D_L,CO/NO maneuvers, while the CPET is used to determine the workload at which D_L,CO/NO is to be measured during exercise. In patients with airflow limitation, notably due to obstructive lung disease, it may be advantageous to supplement the dynamic spirometry with a whole body-plethysmography to obtain a valid measure of vital capacity. A medical health check to rule out any known contraindications before initiating CPET is recommended¹⁷. Importantly, the CPET should be performed at least 48 h prior to the D_L,CO/NO measurement obtained during exercise, as prior vigorous exercise may affect D_L for up to at least 24 h^18,19.

1. Dynamic spirometry

NOTE: Dynamic spirometry should be performed in accordance with the current clinical guidelines from the ERS and ATS²⁰.

1. Measure weight (to the nearest 100 g) and height (to the nearest 1 mm).
2. Ask the participant to sit in an upright chair.
3. Perform a dynamic spirometry during a forced expired maneuver to identify the forced expired volume in 1 s (FEV₁) and forced vital capacity (FVC) of the participant, as described elsewhere²⁰.

2. Cardiopulmonary exercise test (CPET)

NOTE: CPET should be performed in alignment with current clinical recommendations²¹.

1. Adjust the cycle ergometer according to the participant's height and place a heart rate (HR) monitor on the chest.
2. Place the participant on the cycle ergometer. Equip the participant with a mask connected to a metabolic measurement system, to measure ventilation and pulmonary gas exchange throughout the test.
3. Instruct the participant to begin cycling at a self-selected pace ≥60 rounds per min (RPM) and perform a 5 min warm-up period at a submaximal workload based on self-reported activity level, daily fitness and disease status (e.g., 15-150 W).
4. Increase the workload by 5-20 W every min until the participant reaches voluntary exhaustion. The increments should be based on the participant's current fitness level, so that the test is expected to terminate 8-12 min after the commencement of the incremental phase.
5. Instruct the participant to avoid other vigorous exercise for the next 48 h.

3. Calibration of single breath diffusing capacity equipment

NOTE: It is necessary to calibrate flow sensors and gas analyzers to ensure that measurements are both valid and reliable. The exact procedure is manufacturer- and device-specific. The calibration procedure, including biological control, should be completed on each study day, and if less than one study day is executed per week, additional weekly calibrations should be performed. The experimental setup is shown in Figure 1.

1. Open the software program on the computer, and an automatic warmup period of 50 min will be initiated to ensure sufficient temperature of the pneumotach.
2. Make sure that the containers with the test gases are open (See Figure 1D).
3. Perform a gas calibration by first connecting the sampling line from the pneumotach to the MS-PFT Analyzer Unit plug-in termed CAL (See Figure 1B).
4. Initiate the gas calibration by selecting Calibration on the Home Page (See Figure 2A) and choose Gas calibration. Start the calibration by pressing Start or F1 (See Figure 2B).
5. Attach the sampling line to the pneumotach when the gas calibration is fulfilled and accepted.
6. Perform a volume calibration using a valid 3 L syringe. Initiate the volume calibration by selecting Calibration on the Home Page (See Figure 2A) and choose Volume calibration. Start the calibration by pressing F1, and follow the instruction provided by the software (See Figure 2C).
7. Make sure that the inspiratory bag is connected to MS-PFT analyzer unit (See Figure 1C).
8. Complete the calibration procedure by performing a biological control measurement at rest in the sitting position. This should be performed by a healthy non-smoker to ensure reliability of the method. If the given subject's week-to-week variation in D_L,CO or D_L,NO varies more than 1.6 and 6.5 mmol/min/kPa (5 and 20 mL/min/mmHg), respectively, the variation can be due to machine error and should be investigated further^{12, 22}.

4. Preparation of the participant

1. Calculate the desired workload from the prior CPET results for the chosen intensity (% of maximal workload (W_max)) at which the D_L,CO/NO will be measured.
2. At least 48 h after the participant has performed the CPET, ask the participant to return to the laboratory to obtain the D_L,CO/NO measurement during exercise.
3. Measure the height (in cm to the nearest mm), weight (in kg to the nearest 100 g) and Hb from capillary blood (in mmol/L to the nearest 0.1 mmol/L) of the patient.
4. On the Home Page of the program choose Patient > New patient (See Figure 2A) and fill in the required data: Identification, Last Name, First Name, Date of Birth, Gender, Height, and Weight of the participant. Continue by selecting OK or F1 (See Figure 2D).

5. D_L,CO/NO measurement during upright rest

NOTE: D_L,CO/NO measurements are performed in accordance with current clinical recommendations from ERS task force¹².

1. On the Home Page choose Measurement > NO Membrane diffusing (See Figure 2E).
2. Start the automatic resetting of the software, to zero the gas analyzer for all tests gases and to initiate the mixing of the test gases in the connected inspiratory bag. Initiate the automatic resetting by pressing F1 (See Figure 2F).
   1. The automatic resetting takes 140-210 s. Observe the instructions provided by the software to recognize when to initiate the measurement. It is important to initiate the measurement immediately when the software instructs to Connect patient.
3. Place the participant in an upright chair equipped with a nose-clip. Instruct the participant on how to perform the maneuver as described below.
   1. Ask the participant to use the nose-clip and begin normal tidal respirations through a mouthpiece connected to the pneumotach. To ensure a closed system for the measurements, make sure that the participant's lips are kept closed around the mouthpiece. 
   2. After three normal respirations, instruct the participant to perform a rapid maximal expiration to reach residual volume (RV).
   3. When RV is reached, immediately instruct the participant to perform a rapid maximal inspiration to total lung capacity (TLC), targeting an inspiratory time of < 4 s. During the maximal inspiration, a valve opens, allowing the participant to inhale the gas mixture mixed with a known concentration of NO (800 ppm NO/N₂) in an inspiratory bag just prior to the inhalation.
   4. Ask the participant to carry out a breath-hold of 5 (4-8) s at TLC. During the inspiration an inspired volume (V_I) ≥90% of the FVC (or plethysmography-based vital capacity) with a 4-8 s breath-hold time is targeted²³ (Table 1).
   5. After the breath-hold, instruct the participant to perform a strong steady maximal expiration with no interruptions.
   6. After the maximal expiration ask the participant to let go of the mouthpiece and nose-clip. The software will then calculate D_L,NO and D_L,CO without any command.
4. Use verbal encouragement throughout the maneuver to ensure that the participant reaches RV and TLC. Assess the acceptability of the maneuver as per Table 1.
5. Perform the maneuver again after at least a 4 min wash out period, and until two maneuvers fulfil the acceptability criteria (Table 1) or until a total of 12 maneuvers (see below) have been performed on the same session.
6. The D_L,NO and D_L,CO are reported according to the criteria outlined in Table 2. We also recommend that breath-hold time, inspired volume, and alveolar volume as reported. Furthermore, the number of acceptable and repeatable maneuvers should be reported, and findings based on maneuvers that either do not fulfill the acceptability or repeatability criteria should be interpreted with caution.

6. D_L,CO/NO measurement during exercise

NOTE: A timeline of D_L,CO/NO measurements during exercise is provided in Figure 3.

1. Place the cycle ergometer at a distance that enables the participant to breathe through the mouthpiece without having to change the cycling position. Increase the height of the equipment so that the measurements can be carried out with a correct working position on the bicycle (See Figure 2).
2. Place the participant on the cycle ergometer and place a HR monitor on the chest. Instruct the participant to perform each maneuver as outlined in step 5.3.
3. Instruct the participant to begin cycling for 5 min at a submaximal workload, as a warm-up prior to the measurement.
4. Increase the workload to the target intensity while simultaneously starting the automatic resetting of the device by pressing F1 (see step 5.2). The automatic resetting takes 140-210 s, which is sufficient to ensure that the participant has reached steady state.
5. When the automatic resetting is finished, turn the mouthpiece to the participant and perform a maneuver as described below while the participant continues cycling at the target intensity.
   1. Follow the steps in steps 5.4 to 5.5. Assess acceptability and repeatability criteria (Table 1) at each workload, and report as for measurements during rest (see step 5.6 and Table 2).
6. After completion of the maneuver, remove the mouthpiece and decrease the workload to 15-40 W. Perform the active recovery phase for 2 min after which repeat steps 6.4 and 6.5. The 2 min of active recovery and the 140-210 s during the automatic resetting provides a sufficient washout period of 4-5 min.

Results

The protocol was implemented in 2021 and at the time of writing a total of 124 measurements during exercise (i.e., 51 in healthy volunteers and 73 in patients with COPD of various severities) had been performed. The maneuvers, as well as data on fulfilled acceptability and repeatability criteria, and the failure rate are all provided in Table 3.

Calculations
As an example, calculations from a single D_L,CO/NO maneuver are provided here based on data from the first maneuver at 20% of W_max in the healthy group as a case study described below. Based on the measured values provided in Table 4, the following is calculated:

(BTPS)

where F_I is the inspired fraction, V_I is the inspired volume, and D_D,inst and V_D,anat are instrumental and anatomical dead space, respectively.

where F_I is the inspired fraction, P_B is barometric pressure and P_H2O is saturated water vapor pressure, and where

Interpretation of D_L,CO/NO results obtained during exercise
The primary outcome measure of interest is D_L,NO, as the change in D_L,NO from rest to a specific workload is interpreted to provide an overall measure of alveolar-capillary reserve. In healthy individuals, D_L,NO increases linearly with increasing exercise intensity, which is attributed to the enhanced recruitment of blood to the pulmonary capillary bed, facilitated by a rise in cardiac output¹². This leads to capillary recruitment owing to the augmented blood flow or pressure and recruitment of the alveolar-capillary membrane surface area, thereby resulting in a more homogeneous distribution of red blood cells and improved alignment between tissue and red blood cell membrane surfaces¹². In contrast, D_L,CO is considered a secondary measure in this context, primarily utilized to deduce whether concurrent changes in V_C take place. For interpretation at the individual level, differences between two measurements larger than measurement error are considered physiological²⁴, i.e., 2.7 mmol/min/kPa for D_L,NO and 1.6 mmol/min/kPa for D_L,CO.

Case studies
A healthy 25-year-old female with a O_2max of 2696 mL O₂/min (47.3 mL O₂/min/kg) performed eight D_L,CO/NO maneuvers, starting with measurements during upright rest in the seated position, followed by measurements during exercise on a bicycle ergometer (W_max = 208) with increasing intensity up to 60% of W_max (Table 5). All maneuvers fulfilled both the acceptability and repeatability criteria.

A 68-year-old male with moderate COPD (FEV₁= 56% of predicted) with a O₂p_eak of 1852 mL O₂/min (22.8 mL O₂/min/kg) performed eight D_L,CO/NO maneuvers, starting with measurements during upright rest in the seated position, followed by measurements during exercise on a bicycle ergometer (W_max = 125 W) with increasing intensity up to 60% of W_max (Table 6). All maneuvers fulfilled both the acceptability and repeatability criteria.

The reported results for each workload from the two cases outlined above are presented in Figure 4. Furthermore, D_L,NO and D_L,CO as a function of O₂ (calculated from expired air measurements) is presented in Figure 5. In the healthy individual, a near-linear increase in D_L,NO is observed as expected with the exception of a plateau from 20% to 40% of W_max, while a slight gradual increase in D_L,CO occurs across all workloads. This suggests that D_M initially increases with unaltered V_C at the onset of exercise reflecting a redistribution of pulmonary blood flow to recruit previously unperfused capillaries, but with a concomitant gradual increase in V_C at higher workloads, showing that alternating capillary recruitment and distension together function to optimize pulmonary gas exchange during incremental exercise. In the COPD case, D_L,NO increases at the first workload, and then plateaus to remain at the same level during the remaining workloads, indicating that the entire alveolar-capillary reserve is already attained at 20% of W_max. Overall, the extent of pulmonary capillary recruitment and distension, i.e. the alveolar-capillary reserve, is lower in the COPD case than in the healthy individual.

Figure 1: Overview of the study setup. (A) Study set-up for measurement performed during exercise. (B) Gas calibration with a connected sampling line to the MS-PFT Analyzer Unit plug-in termed CAL. (C) A connected inspiratory bag to the MS-PFT Analyzer Unit. (D) Containers containing the test gases. Please click here to view a larger version of this figure.

Figure 2: Guide to the program. (A) On the Home Page select Calibration. (B) Select Gas calibration. (C) Select Volume calibration. (D) Select New Patient. (E) Select New patient and fill out the required information. (F) Select measurements and choose NO diff Membrane. (G) Start the automatic resetting by pressing F1. Please click here to view a larger version of this figure.

Figure 3: Timeline of a diffusing capacity measurement during exercise. Created using BioRender. Please click here to view a larger version of this figure.

Figure 4: Pulmonary diffusing capacity. Comparison of pulmonary diffusing capacity to carbon monoxide (D_L,CO) and nitric oxide (D_L,NO) during incremental exercise as a function of % of maximal workload (W_max) in a healthy individual and an individual with chronic obstructive pulmonary disease (COPD). Please click here to view a larger version of this figure.

Figure 5: Pulmonary diffusing capacity. Comparison of pulmonary diffusing capacity to carbon monoxide (D_L,CO) and nitric oxide (D_L,NO) during incremental exercise as a function of oxygen uptake (O₂) in a healthy individual and an individual with chronic obstructive pulmonary disease (COPD). Please click here to view a larger version of this figure.

Acceptability criteria
1.	≥ 90% of FVC or VC
	OR ≥ 85% of the FVC or VC
		AND VA within 200 ml of the largest VA from other acceptable maneuvers
	OR ≥ 85% of the FVC or VC
		AND VA within 5% of the largest VA from other acceptable maneuvers
2.	A stable 4-8 sec breath-hold with no evidence of leaks or Valsalva/Müller maneuvers
Repeatability criteria
Two acceptable maneuvers with values within
< 5.8 mmol·min-1·kPa-1 for DL,NO
< 1 mmol·min-1·kPa-1 for DL,CO

Table 1: Acceptability and repeatability criteria. Abbreviations: D_L,CO: Pulmonary diffusing capacity to carbon monoxide, D_L,NO: Pulmonary diffusing capacity to nitric oxide, FVC: Forced vital capacity, V_A: Alveolar volume; VC: Vital capacity.

No. of acceptable maneuvers	Repeatability critiera fulfilled	Action
≥2	Yes	Report mean DL,NO and mean DL,CO of two acceptable and repeatable maneuvers
≥2	No	Report values from the maneuver with the highest DL,NO
1	Yes	Report values from the acceptable maneuver
1	No	Report values from the acceptable maneuver
0	Yes	Report mean DL,NO and mean DL,CO of all arepeatable maneuvers
0	No	Failed measurement

Table 2: Reporting of data. Abbreviations: D_L,CO: Pulmonary diffusing capacity to carbon monoxide, D_L,NO: Pulmonary diffusing capacity to nitric oxide.

Group	Measurements (n)	Maneuvers pr. measurement (median [IQR])	Acceptability criteria fulfilled, n (%)	Repeatability criteria fulfilled, n (%)	Failed measurement, n (%)
Healthy	51	2 (2-2)	50 (98)	51 (100)	0 (0)
Mild COPD	24	3 (2-3)	22 (92)	22 (92)	0 (0)
Moderate COPD	39	2 (2-3)	26 (67)	32 (82)	3 (8)
Severe COPD	10	2 (2-3)	1 (10)	4 (40)	6 (60)
All	124	2 (2-3)	99 (80)	109 (88)	9 (7)

Table 3: Completed D_L,CO/NO measurements during exercise at our institution between July 2021 and December 2023. Abbreviations: COPD, chronic obstructive pulmonary disease.

Fractions
FI,CO	0.238
FI,NO	48.75 x 10-6
FI,He	0.08
FA,CO	0.12
FA,NO	6.18 x 10-6
FA,He	0.0603
Volumes (BTPS)
VI (L)	4.13
VD,anat (L)	0.132
VD,inst (L)	0.220
tBH (sec)	5.65

Table 4: Measured test and inert tracer gas fractions in inspired (F_I) and alveolar (F_A) air during a single-breath maneuver. Abbreviations: V_I: inspired volume; V_D,anat: anatomical dead space; V_D,inst: instrument dead space; tBH: breath-hold time.

		Upright		0.2		0.4		0.6
		rest		of Wmax		of Wmax		of Wmax
Workload (watt)		0		40		80		125
Maneuver		1	2	1	2	1	2	1	2
DL,NO (mmol/min/kPa)		35.0	34.7	37.0	38.9	37.4	38.4	42.2	43.4
DL,CO (mmol/min/kPa)		8.0	7.8	8.4	8.4	9.2	9.1	9.8	9.9
Breath-hold time (s)		5.8	5.6	5.7	5.8	5.8	5.7	5.7	5.5
VI (L)		4.1	4.1	4.1	4.1	4.0	4.0	3.8	4.0
VA (L)		4.9	4.8	5.0	5.0	5.1	5.1	5.2	5.3

Table 5: Data from a healthy individual. Abbreviations: D_L,NO: Pulmonary diffusing capacity to nitric oxide, D_L,CO: Pulmonary diffusing capacity to carbon monoxide, V_I: Inspired volume, V_A: Alveolar volume.

		Upright		0.2		0.4		0.6
		rest		of Wmax		of Wmax		of Wmax
Workload (watt)		0		25		50		75
Maneuver		1	2	1	2	1	2	1	2
DL,NO (mmol/min/kPa)		17.9	21.6	23.35	24.35	24.9	24.2	21.8	23.6
DL,CO (mmol/min/kPa)		4.7	5.3	5.0	5.2	5.1	4.9	3.3	4.1
Breath-hold time (s)		6.6	6.1	6.1	5.8	5.8	5.8	5.8	6.0
VI (L)		4.3	4.4	4.2	4.3	4.1	4.0	3.8	3.9
VA (L)		6.7	6.6	6.7	6.7	6.7	6.7	6.7	6.8

Table 6: Data from an individual with chronic obstructive pulmonary disease. Abbreviations: D_L,NO: Pulmonary diffusing capacity to nitric oxide, D_L,CO: Pulmonary diffusing capacity to carbon monoxide, V_I: Inspired volume, V_A: Alveolar volume.

Discussion

The protocol provides a standardized approach to the measurement of D_L,CO/NO during exercise using the dual test gas single-breath technique. Since the obtained D_L,CO/NO-metrics increase due to pulmonary capillary recruitment and distension, the method provides a physiologically meaningful measure of the alveolar-capillary reserve.

Critical steps in the protocol
The method requires an exhalation to residual volume followed by an inspiration to total lung capacity at which a 5 s breath-hold is carried out and terminated with an expiration to RV. This is a critical step, as it can be complicated to perform during exercise and especially during exercise at high intensities. The increasing exercise intensity can lead to a decrease in V_I, and if it decreases below 85% of vital capacity, the maneuver is not acceptable (see Table 1). Thus, it is important that the instructor of the test notes whether the participant inhales sufficiently and confirms a sufficient breath hold-time of four to eight seconds, immediately after each maneuver¹². Furthermore, it may in some cases be difficult to achieve repeatability criteria; in such cases, data from the maneuver with the highest D_L,NO is reported, and we recommend that it is explicitly stated in how many cases this was necessary when presenting data. In some cases, it may not be possible to obtain acceptable or repeatable measurements during exercise at all, for example in studies of patients experiencing a severe dyspnea so that they are unable to achieve a sufficient breath-hold and/or those with dynamic hyperinflation with a concomitant decrease in inspiratory capacity during exercise. In such cases it may be more suitable to use D_L,CO/NO measurements obtained in the supine position, which also leads to pulmonary capillary recruitment and distension, albeit less pronounced than during submaximal exercise^24,25.

Modifications and troubleshooting of the method
It is important that a resting measurement always precedes any measurement performed during exercise, as D_L,CO can be reduced for up to 6-20 h after high intensity exercise performed until exhaustion^18,19,26. Furthermore, it is important to record HR and/or other indices of metabolic load to ensure that the measurements obtained in different subjects have been made at steady state and at similar metabolic workloads.

The method might not be sensitive for detecting small changes in either D_L,NO or D_L,CO, as the test-to-test variability within the same session has been reported up to 7% depending on the specific metric¹². Consequently, it is important to choose an exercise intensity which is sufficient to induce an increase larger than the measurement error, while also keeping in mind that the participant must be able to perform at least two acceptable maneuvers at the given intensity. Among previous studies that used the dual test gas method, various intensities from mild to moderate have been used. Most studies have used a relative intensity related to % of ventilatory threshold^{24, 27}, % of age-predicted maximal HR²⁸, or to % of maximum oxygen reserve²⁹, while only one study has applied an absolute intensity at a fixed workload of 80 W³⁰. Across the studies, these workloads correspond to relative intensities of ranging from 20% to 86% of W_max^{24, 27, 29}. To ease the comparison of measurements between studies, it is recommended to implement a relative intensity i.e., % of W_max, % of maximal HR (HR_max) or % of O_2max (or O_2peak), and to both report W_max and the workload at which the measurement was obtained.

The significance of the method with respect to existing/alternative methods
As for , D_M and V_C may be mathematically derived by D_L,CO/NO^12,31, and while this should be done with caution (see 'Limitations of the method' below), it does permit a more direct mechanistic assessment of how expansion of the alveolar-capillary surface area through pulmonary capillary recruitment (assessed by D_M) and distension (an increase in V_C that exceeds that of D_M) contribute to the exercise-associated changes in pulmonary gas exchange. However, to our knowledge, the single-breath D_L,CO/NO method has only been validated against during upright resting conditions¹¹. The two methods have been used during exercise in several previous studies and show similar physiological changes in D_M and V_C in healthy young individuals^3,24. However, a different number of maneuvers is possible with each method, with permitting a maximum of six and D_L,CO/NO permitting up to 12 maneuvers in the same session¹². This is because despite having the same CO fraction (~0.30), the shorter breath-hold time (5 s vs. 10 s) of D_L,CO/NO results in less CO accumulation in the blood and subsequently less CO backpressure¹⁴. Additionally, up to 22 D_L,CO/NO maneuvers can be conducted without impacting D_L,NO, because the levels of endogenous exhaled NO, ranging between 11 and 66 ppb, are a 1000-fold lower than the NO measurements, which are in the ppm range¹⁴. Hence, given that uses 10 s D_L,CO, and at least two maneuvers are required to assess repeatability at each , corresponding to a minimum of four maneuvers at each exercise intensity, when a double termination is carried out, this might not be feasible during exercise. Thus, previous based methods have used a single maneuver at each , resulting in a minimum of three maneuvers at each exercise intensity³², with the notable drawback that it cannot be assessed to which extent the maneuvers are indeed repeatable. Still, the D_L,CO/NO method only requires two measurements if they fulfill the repeatability criteria and are considered acceptable at each exercise intensity. However, it has been shown that provides acceptable repeatability comparable to that of D_L,CO/NO during exercise, even when breath-hold time is shortened. Hence, during moderate exercise, we previously found a between-day coefficient of variance (CV) of 2% to 6% for the different D_L,CO/NO metrics at breath-hold time of ~ 6 s²⁴, while only slightly higher CVs of 7%, 8%, and 15% for D_L,CO, V_C and D_M, respectively, have been reported using  at a similar breath-hold time³².

On a related note, D_L,CO measured in the context of D_L,CO/NO is known to be consistently lower than the more widely used D_L,CO based on a 10 s breath-hold^{12, 33}. According to previous studies, this is not due to the difference in breath-hold time, as a shorter breath-hold time would increase D_L,CO³⁴. Rather, it might stem from various other factors including inhaled gas composition and disparate CO vs. NO kinetics³³. Firstly, D_L,CO/NO employs helium, while the classical 10 s D_L,CO utilizes methane as the inert tracer gas; owing to their distinct physical properties, these gases exhibit different distributions and solubilities in the lungs and tissues. This might result in a lower V_A with helium than with methane. Lastly, the reactivity of the test gases means differences in the kinetics of NO and CO when binding with hemoglobin could play a part. Although speculative, the presence of NO in D_L,CO/NO may, therefore, influence the binding of CO to hemoglobin³³.

The rate of diffusion of CO across the alveolar-capillary membrane depends on the binding of CO to hemoglobin in blood, and apart from being used to calculate θCO, hemoglobin correction of the D_L,CO-value may be appropriate depending on the specific context³⁵. This is prevalent in a clinical setting, but is less crucial in healthy individuals where the impact on D_L,CO is often negligeable. Such corrections may also be used for appraising D_L,CO/NO during exercise, but are less relevant when specific rest-to-exercise changes are assessed, where (acute) changes in hemoglobin are of minor importance. They should in any event be done with caution, as these equations presuppose a ratio of 0.7 between the D_M and θ∙V_c for CO³⁵, a presumption which might not hold true during exercise.

Limitations of the method
The intensity-dependent increase in D_L,NO and D_L,CO during exercise in healthy individuals reflects pulmonary capillary recruitment and distension. A direct measure of alveolar-capillary reserve can probably only be obtained at submaximal intensity, as the approach would not be practically feasible neither in the experimental or a clinical setting at maximal intensity where maximal recruitment and distension may be evident. The pragmatic choice is thus to target a prespecified (absolute or relative) workload sufficient to trigger pulmonary capillary recruitment and distension in a systematic fashion, while also being feasible for all participants. In the present protocol, the intensity was based on % of W_max as this is easily transferable to other studies. Traditionally, exercise has been prescribed according to % of O_2max or HR_max, but this requires that all participants reach their true max. If not, participants could potentially perform the measurement at different relative intensities³⁶, which may particularly pose a problem and complicate physiological interpretation in populations with severe exertional dyspnea, such as patients with chronic lung or heart disease.

It must be noted that within the individual D_L,CO/NO maneuver, the test gases may not be distributed to relatively poorly ventilated areas of the lungs. This poses a minor problem in individuals without lung disease, but in the presence of substantial ventilation inhomogeneity, including overt air trapping, the true D_L of the participant may be overestimated, because the measurement only reflects the conditions in the best-ventilated regions of the lungs, an effect that is accentuated by shorter breath-holds³⁷. In principle, this may lead to an apparently paradoxical reduction in alveolar-capillary reserve if a participant with lung disease is exposed to an intervention that reduces ventilation inhomogeneity.

The exercise-associated decrease in D_L,CO that exceeds that of D_L,NO at the highest intensity (60% of W_max) in the COPD case reported here must be interpreted with caution, as it is not easily interpreted from a physiological point of view. A similar pattern has been noted in majority of the 73 COPD patients we have studied at our institution so far, and the contribution of merely methodical limitations must be considered. Hence, apart from CO possibly being more susceptible than NO to the impact ventilation inhomogeneity outlined above, the fact that that NO reacts almost 300 times faster with hemoglobin and also diffuses through tissues and plasma twice as fast than CO may also play a part³¹. Hence, while both NO and CO normally undergo diffusion limited gas exchange, the uptake of CO may become perfusion limited when perfusion in individual lung units decrease ~100 fold³¹, thus leading to a reduction of the measured D_L,CO without affecting D_L,NO. Given that COPD is associated with alveolar destruction and a progressive loss of capillaries with a concomitantly inhomogeneous ventilation-perfusion distribution throughout the lungs³⁹, lung units with a 100-fold reduction in perfusion are not uncommon⁴⁰, and they do indeed represent areas in which the transit time of red blood cells may become critically reduced to impair both oxygen and CO uptake during exercise. An additional complementary factor that may be at play is an uneven distribution of the red blood cells within the capillary network of the individual lung units⁴¹, which may also have a much more profound effect on D_L,CO than on D_L,NO.

It is possible to derive D_M and V_C from  measurements¹², but nevertheless not widely used because systematic errors are introduced as their derivation involves several assumptions and empirical constants³¹. For example, the prevailing scientific consensus acknowledge the diffusivity ratio α as 1.97, representing the ratio of physical solubilities of NO and CO in tissue⁴². Several studies have challenged this value, with some proposing higher α values to reconcile discrepancies between different measurement methods. However, these propositions are predominantly dismissed as they deviate from the physical diffusivity ratio, leading to inconsistent α values¹². Furthermore, θNO is assumed to have a finite value, but was historically presumed infinite due to its rapid reaction rate with free hemoglobin. However, comprehensive debates and recent studies have contested this assumption, establishing θNO as finite, with 1.51 mL_blood/min/kPa/mmol_CO providing the best current estimate, as it aligns well theoretical predictions as well as extensive in vitro and in vivo experimentation¹². Similarly, the equations for θCO are based on empirical constants obtained at pH 7.4, rejecting earlier values that were based on less accurate and non-physiological pH measurements⁴³. However, of the different metrics that can be obtained by this method, D_L,NO is in any event based on fewest assumptions and appears to provide the most reproducible estimates of alveolar-capillary reserve²⁴, and therefore remains the main outcome measure of interest in the context of alveolar-capillary reserve.

Importance and potential applications of the method in specific research areas
D_L,CO/NO measurements may provide a comprehensive account of pulmonary gas exchange during exercise. The method may potentially be easier to implement during exercise than in clinical studies on populations with exertional dyspnea, such as patients with heart failure and chronic lung disease, because of the shorter breath-holds and fewer maneuvers required at each workload. Furthermore, D_L,CO/NO specifically provides D_L,NO which probably provides the most unbiased estimate of alveolar-capillary reserve at a given exercise intensity, thus rendering it a suitable outcome measure in many instances.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
HemoCue Hb 201+	HemoCue, Brønshøj, Denmark	Unkown	For measurements of hemoglobin
Jaeger MasterScreen PFT pro (Lung Function Equipment)	CareFusion, Höchberg, Germany	Unkown	For measurements of DLCO/NO
Mouthpiece	SpiroBac, Henrotech, Aartselaar, Belgium	Unkown	Used together with the Lung Fuction Equipment. (dead space 56 ml, resistance to flow at 12 L s−1 0.9 cmH2O)
Nose-clip	IntraMedic, Gentofte, Denmark	JAE-892895
Phenumotach	IntraMedic, Gentofte, Denmark	JAE-705048	Used together with the Lung Fuction Equipment
SentrySuite Software Solution	Vyaire's Medical GmbH, Leibnizstr. 7, D-97204 Hoechberg Germany	Unkown
Test gasses	IntraMedic, Gentofte, Denmark	Unkown	Concentrations: 0.28% CO, 20.9% O2, 69.52% N2 and 9.3% He