© 2003 American Thoracic Society
Exhaled nitric oxide as a diagnostic toolTo the Editor: Aaron Deykin and colleagues are to be commended for an excellent study establishing that whether collected online (expired flow 42–250 ml/second) or offline (expired flow 50–500 ml/second), exhaled nitric oxide (NO) levels (FENO) discriminate well between non–steroid-treated asthmatic and nonasthmatic subjects (1). As discussed by the authors, this reflects the robust nature of NO in diagnosis. However, although this applies to diagnosis of non–steroid-treated asthma, this does not mean that all techniques at whatever flow rates will have adequate discriminatory power for any situation. In non–steroid-treated subjects with asthma, the difference in FENO levels is maximized compared with controls, as shown in their Figure 1. However, when monitoring subjects with asthma on steroids for lesser changes in FENO, the situation may well be different, as will be the case when trying to detect decreased levels in certain conditions, such as cystic fibrosis.One basis for the contention that lower flow rates will be more discriminatory is as follows: FENO at flow rates < 50 ml/second predominantly reflect that diffusing in the airways, whereas as flows increase above 50 ml/second, the relative contribution of alveolar NO convection to the output rises (2). As most studies suggest that alveolar NO levels are normal in asthma, this means that the airway NO output signal will tend to be obscured at higher flow rates. By way of illustration, in a study presented at the American Thoracic Society meeting in 2000, pairwise comparisons of FENO measured at 40 ml/second but not 100 ml/second indicated a significant difference between the subjects with nocturnal asthma and nonasthmatic controls (p = 0.0075) with the nocturnal asthma group having higher FENO levels (3). Further studies are necessary to see whether the use of high flows rates maintains its discriminatory power in other applications of this marker.
National Jewish Medical and Research Center Denver, Colorado REFERENCES
From the Authors:We appreciate Dr. Silkoff's comments regarding our recent work comparing the diagnostic power of exhaled NO (FENO) for asthma when collected at higher expiratory flow rates to that of FENO when collected at lower flows (1). Dr. Silkoff hypothesizes, on the basis of mathematical modeling and work presented in abstract form, that FENO collected at higher flows may lose discriminatory capacity (relative to FENO collected at lower flow rates) when applied to other conditions such as separating steroid-treated subjects with asthma. Although this was not the focus of our article, several lines of reasoning suggest that his hypothesis may not be correct.The abstract that he cites as evidence studied small groups of nocturnal and non-nocturnal subjects with asthma (NA, n = 8, and NNA, n = 6, respectively) and compared FENO collected at 40 ml/second and 100 ml/second to that in six control subjects without asthma (NC, n = 6) (2). Whereas FENO at 40 ml/second but not FENO at 100 ml/second was significantly higher in NA as compared with NC, no significant differences between NNA and NC were reported for FENO at 40 ml/second or for FENO at 100 ml/second. Should we then conclude that both FENO 40 ml/second and FENO at 100 ml/second are poor discriminators between NNA and NC? Perhaps a more appropriate synthesis is that the small size of this study and the resultant high likelihood of a ß-type error prevents drawing conclusions from these negative findings. Additional evidence supporting the acceptability of higher flow rates has recently been published by Delclaw and colleagues who used FENO at 50 ml/second, FENO at 200 ml/second, and bronchial NO flux (determined by Dr. Silkoff's method) to distinguish between non–steroid-treated subjects with asthma and healthy individuals (3, 4). They reported that the three methods were essentially identical with regard to discriminatory capacity. Since it is very likely that inhaled steroids modify FENO by reducing bronchial NO flux, it appears reasonable to conclude that these three methods would detect these changes equally well. In fact, highly significant reductions in exhaled NO occurring during steroid treatment have been documented using expiratory flows of approximately 100 ml/second (5) Finally, we did not observe even a subtle trend for loss of area-under the ROC curve with increasing flow rate. Thus, although we agree with Dr. Silkoff that further studies are needed—especially with regard to conditions other than asthma—the currently available data suggest that in the setting of asthma, discriminatory capacity is not flow-dependant across the range of flows we examined.
Brigham and Women's Hospital Boston, Massachusetts REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
