Conference Coverage

Dueling SLE classification criteria: And the winner is...



– Newer isn’t necessarily better – especially when it comes to the plethora of SLE classification criteria, according to Michelle A. Petri, MD, professor of medicine and director of the Hopkins Lupus Center at Johns Hopkins University, Baltimore.

Dr. Michelle A. Petri, professor of medicine and director of the Hopkins Lupus Center at Johns Hopkins University, Baltimore Bruce Jancin/MDedge News

Dr. Michelle A. Petri

“We have an embarrassment of criteria for lupus right now, and everyone wants to know if one is better than the others,” the rheumatologist said at the annual meeting of the American College of Rheumatology.

She and several coinvestigators who are members of the Systemic Lupus International Collaborating Clinics (SLICC) set out to learn the answer. They developed a new, modified, weighted version of the 2012 SLICC criteria and compared its sensitivity and specificity for SLE diagnosis by 690 physicians with three other major classification systems: the 1997 update to ACR-11 criteria, the nonweighted SLICC 2012 criteria, and the proposed EULAR/ACR criteria, which uses a differentially weighted approach in which the various possible disease manifestations are each assigned a different point score. In contrast, the revised ACR-11 and SLICC 2012 criteria count each SLE manifestation equally.

Long story short: “The two newly derived weighted classification rules did not perform better than the existing list-based rules in terms of overall agreement,” according to Dr. Petri.

How the various SLE classification rules stack up

“We don’t think that weighting all the criteria, which is what the EULAR/ACR and weighted SLICC 2012 rules do, adds to the performance of the criteria set, and in fact it makes it much more difficult for clinicians to use when there’s a complicated weighting system, unless it’s web-based or there’s an app for it. And to be honest, clinicians are so busy that they’re probably not going to take time out in a clinic visit to go use the web or an app. Our criteria need to be user friendly,” she continued.

So which of the four classification systems is most user friendly? The EULAR/ACR criteria can be dismissed on that score because they are supposed to be used only for research, according to the rheumatologist.

“I think the SLICC 2012 criteria are very useful for clinicians because they have the highest sensitivity. And what a clinician wants is not to miss a diagnosis and to start treatment early,” Dr. Petri said.

To develop the weighted SLICC criteria, whose future at this point doesn’t look bright, she and her coinvestigators redeployed the same physician-rated patient scenarios used to develop the nonweighted 2012 SLICC classification criteria and assigned each of the potential manifestations of SLE a specific point score. For example, acute cutaneous manifestations received 16 points, serositis 9, oral ulcers 16, thrombocytopenia 15, and so forth. Under this system, a patient with a score of at least 56 points, or lupus nephritis, or least one clinical and one immunologic component of SLE was classified as having the disease.

Dr. Petri reported having no financial conflicts regarding her study, supported by the National Institutes of Health.

SOURCE: Petri MA et al. Arthritis Rheumatol. 2018;70(Suppl 10): Abstract 1712.

Next Article: