Methods: We identified relevant clinical studies from a systematic review of English and non-English articles using MEDLINE, EMBASE, and CENTRAL. Articles were screened at the abstract and full text level by two independent reviewers. We included previously published studies which analyzed ULTE4 in human subjects with asthma who had been characterized as having or not having aspirin intolerance on the basis of a specified definition. Receiver operator characteristic (ROC) curves were constructed and area under curve (AUC) calculated for each method used to measure ULTE4 by comparing against the gold standard of a positive aspirin challenge.
Results: The search strategy identified 867 potential articles, of which 86 were reviewed at the full text level and 10 met criteria for inclusion. The sensitivity, specificity, positive predictive value and negative predictive values of ULTE4 to determine aspirin intolerance in asthmatic subjects were 0.55, 0.83, 0.77, 0.65 (Amersham-EIA); 0.76, 0.79, 0.73, 0.82 (Cayman-EIA); 0.73, 0.81, 0.76, 0.79 (mass spectrometry) and 0.81, 0.80, 0.65, 0.90 (radioimmunoassay) at optimal threshold of 192, 510, 165 and 69 pg/mg Cr respectively. The diagnostic odds ratio for each methodology was 6.11; 12.27; 11.70; and 17.33 respectively.
Conclusions: This study defines the diagnostic testing accuracy of ULTE4 in determining aspirin intolerance in asthma.