This article is more than 5 years old. Information may no longer be current.
Low concordance observed between EULAR 2006, other criteria for identifying HSP
Identification of Henoch-Schönlein purpura using European League Against Rheumatism 2006 criteria in patients with cutaneous vasculitis showed discordant results compared with two other criteria sets, researchers found.
A series of 642 patients with primary cutaneous vasculitis were evaluated for Henoch-Schönlein purpura (HSP) using the American College of Rheumatology (ACR) 1990 criteria, a modified version of the ACR 1990 criteria by Michel and colleagues from 1992, and the European League Against Rheumatism (EULAR) 2006 criteria.
The researchers found that of the 642 patients reviewed, 451 (70.2%) met criteria for HSP according to the EULAR 2006 criteria, 405 (63.1%) were classified as having HSP using the ACR 1990 criteria and 392 (61.1%) were classified as having HSP using the Michel 1992 criteria.
Only 336 patients (52.3%) met both the EULAR-2006 and the ACR-1990 criteria, and only 229 patients (35.7%) met both the EULAR-2006 and Michel 1992 criteria. Additionally, only 19% of adults, but 66.6% of children, met criteria using all three methods, according to the researchers. – by Shirley Pulawski
Disclosures: The authors report no relevant financial disclosures.
Perspective
Back to Top
The study by Ortiz-Sanjuán and colleagues has shown poor concordance among three sets of criteria for Henoch-Schönlein purpura, a subset of cutaneous vasculitis. I am not at all surprised. I was a member of the American College of Rheumatology‘s 1990 criteria committee. As much younger and eager investigator, I was given the difficult task of being the principle investigator of the “hypersensitivity vasculitis” category probably because no one wanted this as it was common knowledge it was far from being nosologically defined. The discovery of hepatitis C was occurring at the same time as our work and was not able to be accounted for. A dispassionate examination of all the criteria clearly revealed they make sense (i.e., have face validity), but are non-Bayesian given that, for example, equal weight is given to the 2006 European criteria for concurrent arthritis as much as identifying IgA in tissue.
The authors clearly note these sets of criteria differ in their age selectivity, which is not surprising as we know adults are not infrequently afflicted and can be documented by immunohistology. As with other sets of classification criteria, these criteria can give clinicians a rough outline to assist in clinical diagnosis but are fall far short from serving as rigid diagnostic criteria. That challenge will only be met when we have a far more reliable biomarker than we do now.