When Classification And Diagnosis Diverge Save

Like many rheumatologists who are headed to the 2019 American College of Rheumatology meeting in Atlanta, I am eagerly looking forward to learning about the new criteria in vasculitis, lupus, and autoinflammatory recurrent fevers. Yet conference attendees should remember that these criteria are technically intended for classification, not diagnosis. 

Our lived experience makes it easy to forget that such a distinction exists. Fellows memorize classification criteria to learn about rheumatologic diseases (and pass their boards!). Insurance companies cite them when approving or denying therapies. Few rheumatologic diseases even have diagnostic criteria, so clinicians often rely on classification criteria to diagnose patients in clinic. Because classification criteria were not designed to be used for these purposes, they have important limitations.

For one, classification criteria aim to create a well-defined and homogenous cohort.(2)  This works well for clinical trials, where the inclusion of atypical phenotypes may reduce the likelihood of demonstrating therapeutic benefit.  It does not work well for patients who have atypical phenotypes. The recent EULAR Lupus Guidelines require ANA positivity,(3) for instance, which excludes patients who have lupus but do not test positive for the ANA.(4)  To put this into evidence-based medicine jargon, classification criteria prioritize specificity at the expense of sensitivity, thereby increasing the likelihood of a missed diagnosis.

Classification criteria may also fall short when applied to patients with early disease. It would not make sense to enroll patients with undifferentiated arthritis into a trial for rheumatoid arthritis; such patients may ultimately develop lupus or ankylosing spondylitis. Consequently, classification criteria tend to be based on well-differentiated cohorts of patients. This does not reflect our lived reality, where patients present prior to the development of telltale symptoms, signs, or laboratory findings. In routine clinical practice physicians often have to treat empirically, assigning a diagnosis and offering therapy long before a patient technically meets classification criteria. 

With this in mind, how should you approach the criteria you encounter at the national meeting? For starters, you can be excited! These criteria provide a useful window into disease states and there is value in learning them, sharing them with your fellows, and using them in clinical practice. At the same time, do so while remembering their limitations. Many patients who deserve a diagnosis and may respond to treatment do not meet narrowly defined classification criteria.  

1. Hunder GG, Bloch DA, Michel BA, Stevens MB, Arend WP, Calabrese LH, et al. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990;33:1122–1128.

2. Aggarwal R, Ringold S, Khanna D, Neogi T, Johnson SR, Miller A, et al. Distinctions Between Diagnostic and Classification Criteria?: Diagnostic Criteria in Rheumatology. Arthritis Care & Research 2015;67:891–897.

3. Aringer M, Costenbader K, Daikh D, Brinks R, Mosca M, Ramsey-Goldman R, et al. 2019 European League Against Rheumatism/American College of Rheumatology classification criteria for systemic lupus erythematosus. Ann Rheum Dis 2019;78:1151–1159.

4. Choi MY, Clarke AE, St Pierre Y, Hanly JG, Urowitz MB, Romero-Diaz J, et al. Antinuclear Antibody-Negative Systemic Lupus Erythematosus in an International Inception Cohort. Arthritis Care Res (Hoboken) 2019;71:893–902.