Combined effects of type 2 inflammation and lung function on exhaled nitric oxide ability to reflect asthma control

Background: Exhaled nitric oxide (FENO) is a rather reliable marker of asthma control. Its ability is influenced by airway caliber variation (Michils et al. 2017) and should also be impacted by the type of airway inflammation (i.e. type 2- high vs low). Aim: To assess FENO as a marker of control in patients suffering from type 2-high vs type 2-low asthma, defined by FENO level, taking lung function into account. Methods: At each visit at the outpatient clinic, FEV1, asthma control questionnaire score (ACQ) and FENO were assessed. From patients with at least 4 visits, low (L-NO, 29 patients) and high (H-NO, 21 patients) FENO groups were constituted as having a mean FENO < 15 and >60ppb. Between consecutive visits (pairs), an improvement or a worsening of control were defined as ΔACQ<-0.5 or >0.5. An index β was 0 or 1 if ΔFEV1*ΔFENO<0 (opposite directions) or >=0 (same directions). Ability of FENO to reflect control change was assessed by the area under ROC curve (AUC [95% CI]). Results: AUC significantly differed from 0.5 in H-NO for improvement (0.63 [0.55-0.72]) and worsening (0.60 [0.52-0.70]) and never in L-NO. Considering only pairs with β=0 improved AUC in H-NO for improvement (0.78 [0.67-0.87]) and for worsening (0.81 [0.70-0.91]) and not in L-NO. Discussion: In type 2-low asthma patients, FENO never reflects asthma control. In contrast, FENO predicts control changes in type 2-high asthma patients and its ability is significantly improved when FENO and FEV1 change in opposite directions. Conclusion: FENO utility in asthma management appears maximal in a subgroup of type 2-high asthma patients easily identified by lung function in clinical practice.