Distinguishing Wheezing Phenotypes from Infancy to Adolescence. A Pooled Analysis of Five Birth Cohorts

• DOI: 10.1513/annalsats.201811-837oc
• Publication Date: 2019-03-19T19:58:29Z
• Authors: Ceyda Oksel; Raquel Granell; Sadia Haider; Sara Fontanella; Angela Simpson; Steve Turner; Graham Devereux; Syed Hasan Arshad; Clare S. Murray; Graham Roberts; John W. Holloway; Paul Cullinan; John Henderson; Adnan Custovic; John Curtin; Silvia Colicino; Ashley Woodcock; Andrew Bush; Sejal Saglani; Clare M. Lloyd; Benjamin Marsland; Jonathan Grigg; Jurgen Schwarze; Mike Shields; Peter Ghazal; Multan Power
• Keywords: Wheeze; Respiratory sounds; Odds

Rationale: Pooling data from multiple cohorts and extending the time frame across childhood should minimize study-specific effects, enabling better characterization of wheezing. Objectives: To analyze wheezing patterns early to adolescence using combined five birth cohorts. Methods: We used latent class analysis derive wheeze phenotypes among 7,719 participants with complete report at periods. tested associations derived late asthma outcomes lung function, investigated uncertainty in phenotype assignment. Results: identified phenotypes: never/infrequent (52.1%), onset preschool remitting (23.9%), midchildhood (9%), persistent (7.9%), late-onset (7.1%). Compared wheeze, all had higher odds lower forced expiratory volume 1 second second/forced vital capacity adolescence. The association was strongest for (adjusted ratio, 56.54; 95% confidence interval, 43.75-73.06). observed considerable within-class heterogeneity individual level, 913 (12%) children having low membership probability (<0.60) any phenotype. Class certainty highest never/infrequent, lowest (with 51% probabilities <0.80). Individual were particularly heterogeneous whereas many assigned reported later points. Conclusions: All significantly diminished function school-age children, suggesting that notion life episodic has a benign prognosis may not be true proportion transient wheezers. within-phenotype patterns.