Utilidade do FeNO individualizado no diagnóstico da asma em contexto de cuidados primários: Uma análise secundária do EPI-ASTHMA

Abstract

Introduction and objectives: Fractional exhaled nitric oxide (FeNO) has emerged as a non-invasive biomarker for assessing eosinophilic airway inflammation, offering potential support for asthma diagnosis at primary care. However, standard diagnostic cut-offs fail to account for significant inter-individual variation. To overcome this limitation, this study explored the diagnostic utility of individualised FeNO values, derived from previously developed reference equations and compared to fixed cut-offs, for asthma diagnosis in primary care settings. Materials and Methods: This is a secondary analysis of the EPI-ASTHMA study conducted in 38 Portuguese primary care centres, which included adults participating in both stage 1 and stage 2 of the study. Clinical assessments included spirometry, FeNO measurements, and blood eosinophil counts. Individualised FeNO values were calculated using the Lambda-Mu-Sigma method, adjusted for age, sex and height. Diagnostic accuracy of individualised cut-offs was assessed using area under the Receiver Operating Characteristic curve (AUC) and compared against the American Thoracic Society (ATS) recommended absolute cut-offs (>25ppb, >50 ppb). Results: A total of 1,383 participants (median[p25-p75]: 54 [43-65] years) were analised, of whom 60.2% were female, and 11.7% were current smokers. Asthma was diagnosed in 258 (18.7%) participants, who had significantly higher FeNO levels compared to the non-asthma group (23.0 [13.0-43.75] ppb vs 16.0 [11.0-25.0] ppb, p<0.001). The diagnostic utility of FeNO was moderate for both absolute (AUC 0.63; 95% confidence interval (CI) 0.58-0.67) and individualised-predicted (AUC 0.64; 95%CI 0.60-0.69) values. The 202% predicted individualised cut-off demonstrated the highest specificity (84.2%) and accuracy (AUC 0.76; 95%CI 0.74-0.78), while the 100% predicted individualised cut-off yielded the best sensitivity (68.4%) and negative predictive value (86.8%). Thus, 100% and 202% individualised predicted values can be used to rule-out asthma. The 50 ppb cut-off retained the highest specificity (94.3%) and overall diagnostic accuracy (AUC of 0.81; 95%CI 0.79-0.83), although with low sensitivity (22.5%). Conclusion: FeNO demonstrates moderate performance as a stand-alone tool for asthma diagnosis. Individualised FeNO cut-offs improve specificity and may better support clinical decision-making in ruling out asthma in primary care centres and potentially lead to more accurate diagnoses and improved patient management.