Washout volume, transit time and DLCO

Recently while reviewing PFT reports I ran across a test from a patient who had been having spirometry, lung volume and DLCO tests performed at regular intervals for the last several years. Compared to the last several set of tests the most recent DLCO had decreased significantly while the FVC, FEV1 and TLC hadn’t changed. I took a closer look at the raw data from the DLCO test and when I did I saw that the washout volume was not correct.

Alveolar_Sample_Unadjusted_Cropped

Or more correctly, even though the washout volume matched the ATS/ERS standard for DLCO testing it was evident the expiratory gas sample was not taken from the alveolar plateau. The CO and CH4 concentrations at this point in the exhalation are higher than they are in the alveolar plateau and this means the reported DLCO was underestimated.

Alveolar_Sample_Adjusted_Cropped

When I re-adjusted the washout so the gas sample was taken from the alveolar plateau, the DLCO went from 18.56 ml/min/mmHg to 22.26 ml/min/mmHg, which is a 20% increase and far more in line with the patient’s prior DLCO test results.

This, however, increased the washout volume from 0.75 L to 1.34 L. Why was the washout volume so high? The answer is it probably wasn’t.

Our DLCO test systems use a real-time gas analyzer. In this kind of test system the gas analyzer signals are delayed because of the length of time it takes a gas sample to transit the sample line and then for the gas analyzer to respond. This time delay is measured during a calibration and is used to match the volume and gas analyzer signals.

Transit_Time_Alignment

The patient exhaled quite forcibly after the breath-holding period and by analyzing the graph I was able to see the expiratory flow rate during the washout period was about 12 L/sec. The entire sample volume was therefore exhaled in approximately 60 milliseconds and the difference between an alveolar sample beginning at 0.75 L and at 1.34 L is about 50 milliseconds.

When I looked at our calibration records I found that depending on the test system in question transit delays ranged from about 300 to 800 milliseconds. I also found that the difference in delay time from one calibration to the next was usually at least 20 milliseconds and often more. This happens even when two calibrations are done back-to-back which means there are limits to the accuracy of the delay time measurement. At the patient’s high exhaled flow-rate even small errors in the delay time determined during calibration can lead to relatively large mismatches between exhaled volume and the exhaled gas concentrations. It also means that the patient’s true washout volume probably wasn’t as high as it was reported.

The ATS/ERS Statement on DLCO testing currently recommends:

Vital Capacity (L): Washout (L): Alveolar Sample (L):
>2.00 L 0.75 to 1.00 0.50 to 1.00
>1.00 and <2.00 0.50 0.50
<1.00 0.50 <0.50

Washout volume, like some of the other aspects of the single-breath DLCO test, was originally a somewhat arbitrary decision made by the authors of the first standardized procedure for DLCO. Their 1957 paper advocated the use of a 0.75 L washout volume and for them this was a balance between being confident that the subject’s dead space was completely washed out while still being able to acquire an adequate sample volume in less than 3 to 4 seconds.

This decision was based in technology that was available at that time. Prior to the 1990’s almost all carbon monoxide gas and tracer gas analyzers analyzers had relatively slow response times and it was necessary for a single, discrete sample of alveolar gas to be collected for analysis. This meant that the washout volume and sample volume had to be set before a single breath DLCO test was performed. The ATS/ERS standards for washout and sample volume originate in the requirements of these kinds of analyzers.

Since that time, real-time gas analyzers for CO and CH4 have become relatively standard. With these analyzers it is possible to directly view the exhaled gas waveforms of a single-breath DLCO test and to be able to determine when washout of dead space gas has occurred. For this reason the need to strictly hew to the ATS/ERS recommendations has become much less necessary.

The ATS-ERS recommendations are just that, recommendations. At least one research study showed that a washout volume of 0.75 liter may be inadequate in up to a half of the patients tested and further, that even a washout of 1.0 liter may be inadequate in approximately a quarter of the patients tested. Although the average difference was small, in approximately 10 percent of the patients increasing the washout volume above 1.0 liter increased the measured DLCO significantly primarily because the alveolar sample contained less dead space gas.

For all these reasons it is important to inspect the exhaled gas waveforms to determine where the alveolar plateau is located and then manually select the correct washout volume even if it doesn’t meet the ATS/ERS standards. This is necessary not only because default settings don’t work for every patient but also because our testing hardware and software isn’t perfect and errors are not uncommon.

That’s not to say that finding the alveolar plateau or selecting a washout volume is always easy. A patient with severe COPD often has a marked ventilation inhomogeneity and this often appears as the lack of a clearly defined alveolar plateau.

COPD_Washout

The two different sample windows both meet ATS-ERS criteria but each produces different CH4, CO and BHT times. Because calculated alveolar volume (VA) depends on the average tracer gas (CH4) value in the sample volume, it can be seen that a leftward shift in the sample window increases the average CH4 whereas a rightwards shift decreases it. Since the calculated VA is inversely related to the change in CH4, a leftward shift decreases the calculated VA, a rightwards shift increases it. By the same token, a leftward shift in the sample window decreases breath-hold time (BHT) and a rightwards shift increases it. In this example there was a 7% increase in BHT and a 7% increase in alveolar volume from the left to the right sample windows. Although VA and BHT to some extent cancel each other out there is still a 5% difference between DLCO calculated with the different sample windows.

Which washout volume is correct? The answer is both and neither, and this highlights one of the limitations of the single-breath DLCO test. The problem in this example is that the inhaled gas mixture is maldistributed and a relatively small sample of exhaled gas is unable to represent the lung as a whole.

The need to match gas analyzer and volume signals that are out of synch with each other is necessary in real-time DLCO test systems (and most CPET systems). In order to match these signals the time delay caused by the transit of gas through a sampling line and the response time of gas analyzers must be determined with a high level of accuracy. It’s not clear to me however, what level of accuracy is possible when delay times are measured.

In most systems a test gas is switched off and on with a solenoid valve and the time it takes the gas analyzers to respond is used to determine the transit time. There are at least two factors that affect this measurement, the most important of which is the rise time of the gas analyzer. No gas analyzer responds instantaneously and this is due to the internal volume of the sensor and the electronic amplifiers of the gas analyzer. Gas analyzers are usually characterized by the length of time it takes to reach 90% of the maximum signal and the delay time measurement must somehow take this response curve into consideration.

Another factor is the smearing that gas samples undergo as they travel through a sampling line. This is because gas flow has a parabolic cross-section (i.e. gas flow is slowest near the wall of the sample line and fastest in the center) and means that an abrupt change in gas concentration at one end of a sampling line becomes a more gradual change as it passes through the line.

Both of these factors cause a certain level of indeterminacy when the transit delay time is measured. When signals are changing rapidly even minor discrepancies in the measured transit time can cause a mismatch between different signals. Whether or not this mismatch is significant depends a lot on the specific test. In this case the only apparent problem was a washout volume that was likely mis-calculated.

Everybody that performs and reviews DLCO tests should be aware of the issues surrounding the washout volume. Manufacturers usually default to the ATS/ERS standards, but these standards are recommendations, not requirements. Since washout volume is not factor in the DLCO calculation what matters is that the exhaled gas sample comes from the alveolar plateau and staff should feel able to adjust the washout volume in whatever way is necessary to achieve this goal.

References:

Brusasco V, Crapo R, Viegi G editors. ATS/ERS Task Force: Standardization of Lung Function Testing. Standardization of the single-breath determination of carbon monoxide uptake in the lung. Eur Resp J 2005; 26:720-735

Graham BL, Mink JT, Cotton DJ. Overestimation of the Single-Breath Carbon Monoxide Diffusing Capacity in Patients with Air-Flow Obstruction. Am Rev Resp Dis 1984; 129:403-408

Huang YCT, MacIntyre NR. Real-Time Gas Analysis Improves the Measurement of Single-breath Diffusing Capacity. Am Rev Resp Dis 1992; 146: 946-950

Ogilvie CM, Forster RE, Blakemore WS, Morton JW. A standardized breath holding technique for the clinical measurement of the diffusing capacity of the lung for carbon monoxide. J Clin Invest 1957; 36: 1-17.

Prediletto R, Fornai E, Catapano G, Carli C. Assessment of the alveolar volume when sampling exhaled gas at different expired volumes in the single breath diffusion test. BMC Pulmonary Medicine 2007; 7:18

Creative Commons License
PFT Blog by Richard Johnston is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

2 thoughts on “Washout volume, transit time and DLCO

  1. I have a question: In assessing risk for PAH in patients (%FVC/%DLCO > 1.6), should the DLCO corrected to alveolar volume (DLCO/VA) be used or the uncorrected DLCO?
    Thanks!

    • Tal –

      It’s DLCO not DLCO/VA. When the articles are referring to corrected DLCO they mean corrected for hemoglobin, not alveolar volume. To be honest I had not run across the FVC%predicted/DLCO%predicted ratio before although I am not overly surprised at this since all of the articles that include this ratio appear to be concerned primarily with PAH in scleroderma and systemic sclerosis, not the PAH population at large. I will also note that none of the articles I found using the FVC%/DLCO% specify which predicteds they are using for FVC and for DLCO. This aspect is particularly troubling for anybody that is attempting to compare ratio FVC%/DLCO% values from different studies.

      – Richard

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.