To the Editor:
As our team analysed in detail the physiological meaning of the transfer factor of the lung for nitric oxide (TLNO)/transfer factor of the lung for carbon monoxide (TLCO) ratio [1], it seems interesting to extend the analysis to patients. This has been done elegantly by Hughes and van der Lee [2]. This ratio, which has the advantage of simplicity, nevertheless has some disadvantages. 1) In a given cohort of homogenous patients, values are widely scattered and a given value cannot, therefore, be used as a strong predictor of a disease. For example, pulmonary hypertension patients have, as a mean, higher TLNO/TLCO than healthy controls; however, a normal value does not exclude the disease. 2) The interpretation of an alteration in this ratio always makes allusions to capillary lung volume (Vc) and membrane conductance for carbon monoxide (DmCO) variables, which stand behind this ratio. So why don’t we used these variables directly? The answer is well known: we do not know the right values of the carbon monoxide and nitric oxide conductance of haemoglobin (specific conductance (θ)), which are necessary to the calculation of Vc and Dm. However, although we thought in 1987 that the conductance for nitric oxide could be taken as infinite [3], we have since changed our minds, as, at that time, we used the recommended carbon monoxide specific conductance of Roughton and Forster [4], which was incorrect, as explained later by Forster [5]. The work of Borland et al. [6] added experimental arguments leading us to consider that the conductance of nitric oxide has a finite value.
What are the right values for θNO and θCO. Following the in vitro work of Carlsen and Comroe [7], θNO is 4.5 mmHg·min−1 and the Forster [5] recommended value for θCO in normoxia is 0.58 mmHg·min−1, the ratio of these conductances (θNO/θCO) would be 7.7. It can be demonstrated using the Roughton and Forster [4] equation for the two gases that the TLNO/TLCO ratio cannot be greater than θNO/θCO, i.e. 7.7 [8]. As the upper normal value of TLNO/TLCO reaches approximately 5.5–6.0 in most reports [2, 9], a θNO value <4.5 or a θCO value >0.58 are unlikely as they would lead to a decrease in θNO/θCO to <7.7. It could be suggested that both conductances might be lower than these in vitro values; however, it seems highly unlikely that both in vitro values were overestimated, and no published θCO value is <0.58 mmHg·min−1 in normoxia. Interestingly, using the aforementioned θNO/θCO leads to increased Dm [8], reaching the morphometric value of Weibel et al. [10]. This reasoning has sharp consequences for the interpretation of both TLCO and TLNO as, if the θNO/θCO value of 7.7 is confirmed, it would lead us to consider that TLCO is mainly dependent on Vc, as TLNO would be equally dependent on Dm and Vc. Thus, TLCO would appear to be a vascular marker and TLNO would be the only marker sensitive to membrane alterations.
TLNO/TLCO ratio appears to be a step in our knowledge of diffusion, not the end: the road is still open.
Footnotes
Conflict of interest: None declared.
- Received January 16, 2014.
- Accepted January 23, 2014.
- ©ERS 2014