Nasal Versus Conjunctival Allergen Provocation Tests: Diagnostic Performance and Clinical Utility in Respiratory...

Respiratory allergic diseases are highly prevalent and often underdiagnosed, requiring diagnostic approaches that reflect clinically relevant allergic responses. Conventional methods, including skin prick tests (SPTs), serum-specific immunoglobulin E (sIgE), and component-resolved diagnostics (CRD), primarily assess sensitization and may not consistently correlate with clinical reactivity.

Eighty participants (50 patients with respiratory allergy and 30 controls) underwent SPTs, sIgE, CRD, nasal allergen provocation testing (NAPT), and conjunctival allergen provocation testing (CAPT). Diagnostic indices were calculated. Multivariable logistic regression identified predictors of positive provocation tests. Model performance was assessed using receiver operating characteristic (ROC) curves with bootstrap-derived confidence intervals, calibration analysis (locally weighted scatterplot smoothing), and decision curve analysis (DCA).

CAPT demonstrated higher sensitivity than NAPT across all reference standards (91.3-94.9% vs 75.0-86.1%). In contrast, NAPT showed superior discriminative performance, diagnostic accuracy, and model calibration. Polysensitization and asthma were the strongest predictors of positive NAPT responses. ROC analysis showed good discrimination, particularly for Olea europaea (AUC 0.875). DCA indicated greater net clinical benefit for the NAPT-based model across relevant threshold probabilities. Agreement analysis suggested that provocation tests capture aspects of clinical reactivity not fully reflected by sensitization-based diagnostics.

CAPT demonstrates high sensitivity, whereas NAPT provides greater clinical discrimination and decision-making value, particularly in complex allergic phenotypes. NAPT may offer clinically actionable information in patients with discordant or polysensitized profiles, supporting allergen selection and optimization of immunotherapy strategies.