Patients are advised not to smoke prior to DLCO testing, primarily because it increases carboxyhemoglobin. The effect of COHb on DLCO has been well studied, but COHb is not often measured before DLCO testing. Alveolar carbon monoxide (PACO) can be measured however, and there is a good correlation between PACO and COHb.

The single-breath DLCO calculation assumes that PACO is zero. At first glance this is a reasonably good approximation since a non-smoker will normally have a PACO of less than 5 ppm. This is no more than 0.17 percent of the 0.3% CO concentration (3000 ppm) used for testing which is negligible. Even so, smokers can have significantly elevated COHb levels and COHb increases during testing.

Elevated PACO and COHb levels will decrease DLCO. PACO back-pressure is estimated to be responsible for about 40% of the decrease and the anemia effect of COHb about 60%. Since PACO and COHb are usually in equilibrium in the lung, the effects are combined and DLCO is corrected according to COHb.

COHb has been estimated to increase by 0.5% to 0.7% from a single-breath DLCO test. The ATS-ERS statement on DLCO testing acknowledges this, but also does not set an upper limit on the number of tests performed in a single session. My PFT Lab has set an arbitrary upper limit of four attempts but since DLCO decreases approximately by 1% for every 1% increase in COHb, by the fourth attempt DLCO may have declined by over 2%. This is well within the reproducibility requirements for the DLCO test however, and DLCO inter-test variability is likely more related to factors such as inspired volume and breath-holding time.

A larger concern is that smokers can have significantly elevated levels of COHb. I have seen at least one study that estimated that half of all smoker have a COHb level greater than 5%. COHb levels tend to correlate with smoking status but smokers are often poor reporters of their smoking habits and ex-smokers often relapse. Exhaled CO has been and can be used to determine whether patients are being compliant with smoking reduction programs (6 ppm is the usual cutoff for normalacy).

There is an oxygen re-breathing technique for estimating COHb, but this requires a three-minute period of breathing 100% oxygen followed by a three-minute period of rebreathing into an anesthesia bag equipped with a soda lime filter. This was considered a rapid technique for the time it was developed (1960) but it was tested using a small number of young, healthy subjects (6) and has not been verified.

Highly sensitive solid-state CO analyzers were developed in the 1980’s and have been used primarily to monitor a patient’s smoking status. Using these devices a number of investigators have obtained an alveolar sample from a 20-second room air breath-holding period to measure PACO. They showed that COHb is approximately equal to PACO (in PPM) x 0.20 with a range of 0.16 to 0.25. The range in factors is due to several reasons, one of which is that these devices are sensitive to exhaled hydrogen which is elevated in some subjects for metabolic reasons. It’s also been shown that airway obstruction can cause the COHb to be calculated from PACO to be underestimated by as much as 3% when airway obstruction is severe.

Although their overall contribution is not great in most patients, elevated levels of COHb and PACO can reduce the accuracy of DLCO measurements. Although the error bar is larger than I would like, COHb can be estimated from a room-air measurement of PACO. I would like to suggest that pulmonary function test system manufacturers add the option of measuring PACO prior to DLCO testing. Alternatively, a PFT Lab could acquire a CO analyzer used for monitoring smokers and estimate COHb through a manual calculation. In either case this would improve the accuracy of DLCO testing and would also help monitor and document a patient’s smoking status. 

References:

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

Henderson M, Apthorp GH. Rapid method for estimation of carbon monoxide in blood. Brit Med J. 1960 2: 1853-1854.

Jarvis MJ, Belcher M, Vesey V, Hutchison DCS. Low cost carbon monoxide monitors in smoking assessment. Thorax 1986; 41: 886-887.

Kleerup EC, Zeider MR, Fedor BC, Kim HJG, Tashkin DP. Adjustment of diffusing capacity (DLCO) using exhaled carbon monoxide (ECO). Amer J Respir Crit Care Med 2011; 183: A6292.

Leech JA, Martz L, Liben A, Becklake MA. Diffusing capacity for carbon monoxide: The effects of different derivations of breathhold time and alveolar volume and of carbon monoxide back pressure on calculated results. Amer Rev Resp Dis 1985; 132: 1127-1129.

McNeil AD, Owen LA, Belcher M, Sutherland G, Fleming S. Abstinence from smoking and expired-air carbon monoxide levels: lactose intolerance as a possible source of error. Amer J Public Health 1990; 80: 1114-1115.

Togores B, Bosch M, Agusti AGN. The measurement of exhaled carbon monoxide is influenced by airflow obstruction. Eur Resp J 2000; 15: 177-180.

Wald NJ, Idle M, Boreham J, Bailey A. Carbon monoxide in relation to smoking and carboxyhaemoglobin levels. Thorax 1981; 36: 366-369 

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