In this study, the effectiveness of offline FeNO measurement after collection was evaluated in terms of its application to epidemiologic research. A number of previous studies reported that FeNO levels were valuable for the quantitative and noninvasive assessment of airway inflammation
[3, 6, 9]. Because the offline method of FeNO measurement has an advantage in that exhaled air can be collected anywhere and from multiple subjects at the same time
[20, 22], it is considered to be suitable for large-scale field research
[25–27]. However, the storage conditions may affect the FeNO levels because of the lag-time from sampling to measurement
[14, 21]. Therefore, the accuracy of FeNO levels in the offline method should be evaluated after consideration of storage conditions, including temperature and time. The ATS/ERS recommend that offline samples be measured within 12 hours from collection
. If the storage time from sampling to measurement can be prolonged, it will be possible to measure FeNO levels for many more people. This raises the possibility that the offline method of FeNO measurement may be used to evaluate the degree of airway inflammation in large-scale epidemiologic research. Previous studies reported that refrigerated storage prolonged the stability of FeNO levels in the bags
[21, 23, 28]. However, most of the previous studies investigated only small numbers of subjects. Few studies have evaluated the time course of the changes in offline FeNO levels among enough subjects including healthy subjects.
The present study demonstrated that FeNO levels remained stable in the bags stored at room temperature for the first 9 hours after collection, but increased starting at 24 hours. When the bags were stored at 4°C, the stability of the FeNO levels was prolonged for a comparatively long time. These findings are consistent with those previously reported by Shimizu et al.
, who investigated asthma patients. The present study showed that the offline FeNO levels changed in relation to time and temperature among healthy adults, as well as among asthma patients.
Shimizu et al.
 found that the time course of FeNO levels after collection differed in relation to the initial FeNO levels. In the present study, the time courses of FeNO levels after collection were compared among four groups divided by the quartiles of the initial FeNO levels. In the bags stored at room temperature, FeNO levels increased with time regardless of the initial values. When the bags were stored at 4°C, FeNO levels decreased gradually with time in the group with an initial FeNO level of 51 ppb or above, although the changes in FeNO levels with time were small in the other groups. These findings differ from the results of Shimizu et al.
 The difference may be due to the fact that almost all of the present subjects were healthy adults.
The subjects with wheezing showed significantly higher FeNO levels immediately after collection than those without, in accordance with many previous studies
[18, 29, 30]. When stored at 4°C, FeNO levels among the subjects with wheezing decreased slightly starting at 24 hours, while the levels among those without wheezing remained stable for a long time. Because the initial FeNO levels among the subjects with wheezing had been quite high, the FeNO levels still remained sufficiently high, even when the levels decreased during storage. Thus, the difference in FeNO levels between the subjects with and without wheezing remained significant. Therefore, it appears that the measurement of FeNO levels after storage at 4°C for several days does not affect the qualitative differentiation of airway inflammation. However, it may be difficult to assess the degree of airway inflammation quantitatively, because FeNO levels decreased gradually among the subjects with wheezing starting at 24 hours. Elevated FeNO levels in subjects with rhinitis have been reported
[30–32]. However, in the present study, there were no differences in FeNO levels in relation to a history of allergic rhinitis or pollinosis, though only a history of these diseases, and not present symptoms, was evaluated. The effects of upper airway symptoms, including rhinitis and pollinosis, on FeNO levels have yet to be evaluated.
In this study, FeNO levels immediately after collection were lower among smokers than among non-smokers, despite smoking cessation for 2 hours before the collection of exhaled air. The non-significance of differences in initial FeNO in relation to smoking habits might be due to the small number of current smokers in the study. FeNO levels have been reported to be chronically reduced in cigarette smokers, in addition to the acute effects immediately after smoking
[33–35]. The high NO concentration in cigarette smoke may chronically inhibit the activity of NO synthase (NOS) in the respiratory tract
, resulting in a reduction in FeNO levels among smokers. However, even in the bags stored at 4°C after collection, FeNO levels among smokers started to increase at a relatively early stage and reached 107.4% ± 15.9% relative to the baseline values at 9 hours. On the other hand, FeNO levels among ex-smokers decreased with time, and those among nonsmokers changed very little. Even when the subjects were limited to non-smokers and ex-smokers whose initial FeNO levels were similar to those among current smokers (<30 ppb), the changes in FeNO levels with time were small compared with those among current smokers, although the differences were not significant [see Additional file
1. The mechanisms of these time courses changes in relation to cigarette smoking are unclear. The activity of NOS inhibitors that reduce NO levels in exhaled air of smokers may be diminished during storage. This finding suggests that FeNO levels measured by the offline method may be decreased among smokers even within 12 hours after collection, which is recommended by the ATS/ERS. Various factors, including cigarette smoking, that affect the changes in FeNO levels with time should be further evaluated.
The results of this study should be interpreted cautiously for several reasons. First, the effects of storage conditions other than temperature and time, such as humidity, were not considered. Paredi et al.
 reported that FeNO levels remained stable for 24 hours in Mylar bags containing silica gel, though controls without silica gel were not investigated. Bondini et al.
 found that the addition of silica gel increased the variability of FeNO levels during storage. Second, the time courses of FeNO levels in the bags stored at 4°C or room temperature after collection were evaluated; storage at other temperatures was not evaluated. Although the effects of warming or freezing have also been evaluated
[21, 23], the previous studies showed that FeNO levels in the bags were stable for the longest time under refrigerated storage. In addition, a simple storage condition is desired for epidemiologic research. Third, most of the subjects of this study were healthy subjects, and only one subject had received treatment for asthma. However, eight subjects, including an asthmatic patient, had wheezing, and their FeNO levels were considerably higher than those without wheezing, in agreement with the previous reports
[29, 30]. Therefore, this method appears to assess airway inflammation adequately. The effect of infections
 was not evaluated, because none of the subjects had any symptoms associated with respiratory tract infections. Fourth, this study was performed under identical conditions. That is, exhaled air was collected in the morning from the subjects who were requested to refrain from eating, drinking, and smoking for 2 hours before the collection. These conditions should be considered when the offline measurement of FeNO is applied to epidemiologic research.
In summary, this study demonstrated that the stability of FeNO levels after collection was prolonged in the bags stored at 4°C. It is desirable that offline FeNO levels should be measured as soon as possible after collection, as the ATS/ERS recommended measurement within 12 hours. However, the samples collected in field research might not often be measured within this time. The present results suggest that valid measurement can be done within several days after collection if the samples are handled appropriately. However, the time course of the changes in FeNO levels after collection differed in relation to the initial FeNO values and cigarette smoking, not just the storage temperature. Despite these problems, this study shows that the offline method of FeNO measurement can be highly effective for large-scale epidemiologic research under conditions in which the bags including exhaled air are stored at 4°C immediately after collection and are delivered to the laboratory under refrigeration.