Recognizing Sjögren's Disease in Established Rheumatic Disease

Key Takeaways:

• In a US cohort of primary disease, physician-diagnosed prevalence was 5 times the criteria-confirmed figure, because confirmatory testing was rarely performed.
• Sjögren's often follows an existing autoimmune diagnosis, so the patient is frequently already under rheumatologic care under a different label.
• The 2016 criteria were derived in primary disease. In associated disease, sensitivity falls to 46% against a broad gold standard while specificity stays at 100%.

The patient most likely to have undiagnosed Sjögren's disease may already be on a rheumatology schedule. The distance between how the disease is diagnosed in practice and how it is formally classified is where most of the missed patients sit.

Where the recognition gap sits

A population-based US cohort of primary Sjögren's disease found an age- and sex-adjusted prevalence of 10.3 per 10,000 by physician diagnosis against 2.2 per 10,000 by formal classification criteria.¹

The cause of that 5-fold gap is the useful part: Of 106 physician-diagnosed patients, only 23 – 22% – met either the American-European Consensus Group or 2012 American College of Rheumatology criteria. The authors state plainly that the main explanation was that the required tests were not performed rather than performed and negative. Minor salivary gland biopsy was done in 16 patients. That cohort excluded patients with an associated systemic autoimmune disease, so it does not describe the overlap population.¹

The inference still travels: If confirmatory testing is this rarely completed when Sjögren's is the working diagnosis, it is unlikely to fare better when the presenting label is rheumatoid arthritis.

When the second diagnosis comes first

Sjögren's frequently arrives after another autoimmune diagnosis is established. In a French claims analysis of 14,809 patients with Sjögren's plus an associated autoimmune disease, that disease was diagnosed first in 38% of rheumatoid arthritis cases, 42% of systemic lupus erythematosus, 52% of systemic sclerosis, 66% of juvenile idiopathic arthritis, and 81% of other overlap syndromes.²

Rheumatoid arthritis accounted for 53% of the associated conditions. Cases were identified by a claims algorithm rather than clinical confirmation, and the authors note that the low prevalence they observed may partly reflect underdiagnosis.²

Recognition changes management here. Sjögren's Syndrome Foundation guidelines state that where tumor necrosis factor (TNF) inhibition is used for rheumatoid arthritis or another overlap condition in a patient with Sjögren's, providers should monitor for:³

• Lymphoma and other malignancies
• Serious infections including tuberculosis
• Invasive fungal infections
• Hepatitis B reactivation
• Hepatotoxicity
• Heart failure
• Cytopenias
• Hypersensitivity and infusion reactions
• Demyelinating disease

The same guideline is explicit that this is not a reason to avoid the class: It states there is no evidence that patients with rheumatoid arthritis and Sjögren's treated with anti-TNF agents have an increased incidence of lymphoma, and that the recommendation should not be read as discouraging TNF inhibition where it is indicated for inflammatory arthritis. What recognition changes is the monitoring attached to a therapy that continues unchanged, informed by the elevated non-Hodgkin's lymphoma risk the disease itself carries.³

What the criteria do and do not cover

The 2016 American College of Rheumatology/European League Against Rheumatism criteria are weighted rather than a checklist. Labial salivary gland focal lymphocytic sialadenitis with a focus score of at least 1 and anti-SSA/Ro positivity each carry a weight of 3.⁴

Three objective measures carry 1 each: an ocular staining score of at least 5 or a van Bijsterveld score of at least 4, a Schirmer's test of 5 mm or less per 5 minutes, and unstimulated whole salivary flow of 0.1 mL per minute or less.⁴

Classification requires a total of at least 4. Symptoms of ocular or oral dryness, or systemic features producing at least 1 positive disease activity index domain, determine eligibility to apply the criteria at all – they are not scored. One footnote is easy to miss: Patients on anticholinergic medications should be assessed for objective dryness only after a sufficient interval off those agents.⁴

The limitation that matters here is scope. The criteria were derived and validated in primary disease. The authors suggested they may also apply to Sjögren's associated with other autoimmune diseases, but left confirmation to future research.⁴

One cohort has since supplied it. Among 300 patients with rheumatoid arthritis, lupus, or scleroderma – 100 each, drawn from a 350-patient cohort that also included 50 with primary disease – the 2016 criteria showed sensitivity of 46% with specificity of 100% against a gold standard of definitive-or-probable clinical diagnosis, and sensitivity of 75% with specificity of 91% against definitive diagnosis alone. It was the best-performing of the 3 criteria sets tested, and the authors conclude the criteria are applicable in this setting – a conclusion driven by that specificity.⁵

The practical reading is that a positive result is trustworthy and a negative one is not reassuring. In a patient with established rheumatoid arthritis, lupus, or scleroderma and unexamined sicca symptoms, criteria non-fulfillment should not close the question – particularly where the 3 objective measures were never obtained.