Diagnostic Accuracy of Urinary LTE4 Measurement to Predict Aspirin-Exacerbated Respiratory Disease in Patients with Asthma.

Background: Patients with aspirin-exacerbated respiratory disease (AERD) are distinguished from patients with aspirin-tolerant asthma (ATA) by significantly higher baseline concentrations of urinary leukotriene E (uLTE). However, an overlap between the individual values of the groups exists.

Objective: The objective of this study was to estimate the discriminative value of uLTE concentration in differentiating between patients with AERD and patients with ATA and evaluate the diagnostic accuracy of uLTE measurement alone and added to clinical parameters to predict AERD diagnosis in patients with asthma.

Methods: Clinical data were collected from questionnaires. Spirometry, skin prick tests, total IgE, and blood eosinophilia were evaluated. ULTE concentrations were measured in morning urine samples by enzyme-linked immune assay (ELISA).

Results: Patients with AERD (n = 247) had significantly higher uLTE concentrations than those with ATA (n = 239). The uLTE concentration of 800.0 pg/mg creatinine as measured by ELISA on a spot sample best discriminated the 2 groups (area under the curve 0.7; 95% confidence interval 0.66-0.74, sensitivity 49%, specificity 81%). The positive predictive value and negative predictive value (NPV), after considering the prevalence of AERD in the population of asthmatics, were 16% and 96%, respectively. Nasal polyps, upper airway symptoms, nasal corticosteroid treatment, asthma exacerbations, forced expiratory volume in the 1 second predicted, and age of asthma onset were independent predictors of AERD diagnosis. The addition of elevated uLTE concentration to the set of clinical parameters enhanced slightly the prediction of AERD diagnosis beyond the level predicted by clinical parameters (P = .036).

Conclusions: A set of typical clinical parameters has a superior accuracy in prediction of AERD diagnosis than the measurement of uLTE concentration alone. The addition of uLTE concentration to clinical parameters slightly enhances the prediction of AERD diagnosis, especially due to a high NPV.