A Prospective Validation of the NUT CRACKER Diagnostic Algorithm for Walnut and Pecan Allergy with Prediction of Severity.

Abstract

We previously devised an algorithm using skin prick tests (SPT), basophil activation tests (BAT), and co-allergy status to reduce the need for oral food challenges (OFCs) in 63 patients with suspected walnut/pecan allergies. To validate prospectively the NUT CRACKER (Nut Co-Reactivity-Acquiring Knowledge for Elimination Recommendations) diagnostic algorithm in a new cohort of patients (n = 120) and to study the utility of SPT and BAT in predicting walnut-pecan allergy severity in both groups (n = 183). Patients (n = 183) aged 8 (6-11) years, median (interquartile range), with suspected tree nut allergy were studied. SPT used ground nut extract (10 mg/mL), and BAT assessed allergen-induced basophil CD63 expression. Allergy was determined based on a recent reaction or a positive OFC, and tolerance was defined by regular consumption or a negative OFC. Walnut BAT was significantly higher in walnut/pecan dual compared with single walnut allergy (P = .003) and predicted the need for epinephrine during positive walnut OFCs (P = .009). Results of walnut and pecan BAT correlated with the corresponding nut eliciting dose (P = .014 and P < .001, respectively). Receiver operating curves for walnut and pecan allergy in the prospective cohort yielded area under the curves ranging from 0.87 to 0.93 for SPT and BAT. Concordant with the original study, pecan-allergic patients were all co-allergic for walnut. The algorithm decreased the need for OFCs by 78.8 % with 6 of 240 (2.5%) falsely positives and no false negatives. The algorithm was validated prospectively and awaits broader testing across other populations. The use of BAT further associates with severity in walnut/pecan allergy.