Alerjik rinitte astım ilerlemesini tahmin etmek için klinik bir nomogram: bir karar destek aracı

Özet (abstract)

Allergic rhinitis (AR) often precedes allergic asthma (AA), yet clinicians lack practical decision support tools to estimate individual progression risk at the time of AR diagnosis. Nasal nitric oxide (nNO) is a non-invasive marker of upper airway type 2 inflammation, and component-resolved diagnostics (CRD) enables precise molecular sensitization profiling that may identify patients at higher risk for lower airway involvement. We aimed to develop and internally validate a prognostic nomogram—a clinical decision support tool—for AR-to-AA progression using nNO, CRD, and readily available clinical variables. In a single-center retrospective longitudinal cohort, 842 adults with newly diagnosed isolated AR (house dust mite sensitized) were followed until AA diagnosis or December 2025 (median 62 months). Baseline predictors included nNO, peripheral blood eosinophils, CRD for Der p 1, Der p 2, and Der p 23, and clinical characteristics. LASSO regression selected candidate predictors for a multivariable Cox model. A survival nomogram was constructed and internally validated (7:3 split) using Harrell’s C-index, calibration plots, and decision curve analysis (DCA). AA developed in 186 patients (22.1%). Four independent predictors were retained: higher nNO (per 100 ppb: HR 1.24; 95% CI 1.15–1.34), Der p 23 sensitization (HR 2.18; 95% CI 1.62–2.93), peripheral blood eosinophil count (per 100 cells/µL: HR 1.15; 95% CI 1.08–1.22), and family history of atopy (HR 1.55; 95% CI 1.12–2.14). The nomogram showed good discrimination (C-index 0.812 [95% CI 0.774–0.850] in training and 0.789 [95% CI 0.741–0.837] in validation) and good calibration for 3- and 5-year asthma-free survival. DCA confirmed net clinical benefit. A clinical nomogram combining nNO, Der p 23 sensitization, blood eosinophils, and family history provides individualized 3- and 5-year estimates of asthma progression risk in patients with isolated AR. As a practical decision support tool, it may assist clinicians in risk-stratified surveillance and earlier preventive intervention. Not applicable.

Yazarların özeti; kaynağından alınmıştır. BMC Medical Informatics and Decision Making, 2026 · DOI ↗