Analysis of exhaled eicosanoid species for differentiating
asthma from normal controls
In the discovery phase of the study, in addition to those previously
reported11, including LTB4,
LTE4, LXA4 and PGE2, the
level of 15-HETE, but not TXB2, in EBCs of subjects with
asthma (N=60) was significantly lower than that noted in the control
group (N=20) (Data not shown). To confirm the validity of these
eicosanoid species in differentiating asthma patients from normal
subjects, a total of 415 children were included in the validation phase,
which consisted of 318 stable asthmatic and 97 healthy subjects. The
demographics of these asthmatic children and healthy children are
summarized in Table 1. Significant differences were noted for age,
gender, serum total IgE, FEV1 and FeNO between the
subjects in the asthmatic and the control groups (all had
p<0.001 except for age and gender with p<0.05 and
p<0.01, respectively; Table 1). No significant difference was
found between these two groups for BMI. In the expanded case-control
design, the levels of exhaled 15-HETE were significantly lower for
asthmatic subjects than for healthy subjects (p<0.0001; Table
2), while the level of TXB2 was similar between the two
groups. Correlation analysis revealed that in asthmatic children, there
was a significant positive correlation between the levels of
TXB2 and those of LTB4 and
PGE2 (Supplementary Figs. S1A and S1B) in the exhaled
condensate. Moreover, among the asthmatic subjects, negative
correlations were found for TXB2 and
FEV1, and also for 15-HETE and LTB4,
(r=-0.13, p<0.05; r=-0.11, p<0.05, respectively;
Supplemental Figs. S1C and S1D).
When the asthmatic population was stratified into different severity
groups (Table 2), it was noted that in comparison to the mild group, the
moderate group was characterized with lower levels of exhaled 15-HETE,
and the severe group exhibited even lower levels. The difference in
15-HETE levels between healthy subjects and all three asthmatic severity
groups was significant, but no significant difference was found between
groups for TXB2. Further, as 15-HETE is known to exert
inhibitory effect on 5-LOX–derived pro-inflammatory leukotrienes, the
ratios of exhaled 15-HETE/LTB4 were calculated, and the
results showed that the ratio of 15-HETE/LTB4 was
significantly lower in subjects with severe asthma (p<0.01;
Table 2). We then utilized the data of Table 2 to generate the ROC
curves and calculated the AUC values for each eicosanoid species. Figure
1 shows the ROC curves and the AUC values of the analyzed eicosanoids in
differentiating asthma from healthy controls. Results showed a similar
discriminating power for exhaled 15-HETE, FEV1 and FeNO
(Fig. 1).