The new classification criteria for the hereditary recurrent fever (HRF) syndrome [cryopyrin-associated periodic syndrome (CAPS), TNF-α receptor-associated periodic syndrome (TRAPS), FMF and mevalonate kinase deficiency] have been published recently. These criteria define two core sets of criteria for each HRF: mixed criteria, including genetic and clinical variables, and clinical criteria, relying on clinical variables only. Our aim was to validate the criteria for HRF in an independent cohort, the JIR Cohort database, an international repository of systemic inflammatory diseases.
We enrolled patients with HRF, periodic fever, adenitis, pharyngitis and aphthous stomatitis syndrome (PFAPA) and syndrome of undefined recurrent fever (SURF). A score ranging from zero to two was attributed to their respective genotypes: zero (no mutation), one (non-confirmatory genotype) or two (confirmatory genotype). The criteria were applied to all patients based on genotype scoring. The treating physician's diagnosis served as the gold standard for the determination of specificity.
We included 455 patients. The classification criteria showed excellent specificity for CAPS and TRAPS (98% specificity each), fair specificity for FMF (88%), but poor specificity for mevalonate kinase deficiency (58%). Sub-analysis showed excellent accuracy of the mixed criteria for all four HRFs. Misclassification was mainly attributable to clinical criteria sets, with false-positive patients in all four HRF clinical criteria sets.
This study represents the final validation step of the HRF classification criteria as recommended by the ACR. Genetic data appear to be necessary to classify patients with HRF correctly.