Respiratory symptom scores to capture acute respiratory events in children with cystic fibrosis

Abstract

<b>Background:</b> Acute respiratory events (ARE) contribute to the progression of cystic fibrosis (CF) lung disease [Stanojevic et al, ERJ, 2021]. Symptom scores could improve the management of these events in individual patients. This study describes the accuracy of the Cystic Fibrosis Questionnaire-Revised Respiratory score (CFQ-RResp) and the Chronic Respiratory Infection Symptom Score (CRISS) in detecting AREs in children with CF. <b>Methods:</b> In this prospective observational study, the parent-proxy (6 -13 years) and self-reported (6 - 18 years) CFQ-R_Resp and CRISS (6-18 years) were captured quarterly over 2 years. The lung clearance index (LCI) and FEV1 were also measured. We calculated the accuracy of each PRO in distinguishing AREs (treated and untreated with antibiotics) and clinically stable visits, using the minimal important difference of each PRO as the diagnostic threshold. <b>Results:</b> Ninety-eight children (mean age 9.7 years) with CF were included. A ≥8.5-point worsening in the parent-proxy CFQ-R_Resp was more sensitive (75.6%; 90/119) than the self-reported score (32.8%; 39/119) in distinguishing ARE from stable visits; but the specificity of the parent-proxy (84.2%; 203/241) and self-reported (88.8%; 214/241) scores were similar. The sensitivity of an ≥11-point worsening in the CRISS in distinguishing ARE from stable visits was poor (56.9%; 78/137). Combining the symptom scores with LCI and/or FEV1 improved their accuracy for AREs. <b>Conclusion:</b> The accuracy of respiratory symptoms scores to detect AREs is improved when combined with lung function outcomes, suggesting that combining patient-reported and physiologic outcome measures could best inform the management of these clinical events.