Early-Life Wheezing and Atopic Burden Are Associated with Treatment-Defined Asthma Severity and Control in School-Age Children: A Two-Center Cross-Sectional Study

Abstract

Background: Childhood asthma is clinically heterogeneous, yet simple clinical approaches for describing concurrent disease burden are limited. We tested whether a two-axis phenotype matrix—built from early-onset and/or high-burden wheeze and atopic load—is associated cross-sectionally with treatment-defined severity and control in school-aged children with asthma. This exploratory, unvalidated classification requires prospective external validation before clinical use. Methods: In a two-center retrospective cross-sectional study, 337 children aged 6–12 years with physician-confirmed asthma were classified into four clinical strata (low-burden, familial/atopic-dominant, wheeze-dominant, atopic-wheezing). The primary outcome was moderate-to-severe asthma, defined by controller treatment step and deliberately independent of acute events. Variables independently associated with the outcomes were identified using multivariable logistic regression with prespecified sensitivity analyses; trends across ordered wheezing-onset categories were assessed using the Cochran–Armitage trend test, adjusted associations were confirmed by likelihood-ratio tests, and secondary outcomes were false-discovery-rate corrected. Results: The atopic-wheezing stratum was the largest group (140/337, 41.5%) and carried the highest burden: moderate-to-severe asthma 49.3% vs. 17.4% in low-burden and uncontrolled asthma 42.1% vs. 13.0% (both q < 0.01). Adverse outcomes were generally more frequent with earlier wheezing onset (moderate-to-severe asthma 26.9% to 57.9% from no early wheeze to onset ≤ 12 months; p for trend <0.001). After adjustment, the atopic-wheezing stratum had the largest adjusted odds ratios for both moderate-to-severe asthma (aOR 4.49, 95% CI 1.91–10.54) and uncontrolled asthma (aOR 4.78, 95% CI 1.85–12.35), and remained significant across all sensitivity analyses. Passive smoke exposure was independently associated with uncontrolled asthma (aOR 1.86, 95% CI 1.13–3.05) but not severity. The FEV1/FVC z-score differed significantly across strata and was lowest in the atopic-wheezing group (median z-score −1.28 vs. +0.10 in low-burden; p < 0.001). Conclusions: The co-occurrence of high atopic load with early-onset or high-burden wheezing defined the clinical stratum with the highest observed treatment requirement and poorest control, using only routine history and standard allergy work-up. Passive smoke exposure was independently associated with uncontrolled asthma.