Background New consensual classification criteria for primary Sjögren's syndrome (pSS) have been recently developed and endorsed by ACR and EULAR. They differ substantially from previously used AECG criteria in that they consider systemic involvement (defined as ESSDAI score ≥1) as well as sicca symptoms as entry criteria before applying a weighted score. Evaluation of the concordance and differences between the two sets of criteria in independent patient populations is mandatory to establish how future clinical studies using the new criteria will be comparable to previously published studies. Major salivary gland ultrasonography (SGUS) has demonstrated promising diagnostic performance in previous studies, but was not included in these new classification criteria.
Methods This cross-sectional study was conducted in the monocentric Brittany cohort (DIApSS cohort) of patients with suspected pSS (sicca symptoms, parotidomegaly or extraglandular manifestations suggestive of pSS). All patients had standardized clinical examination, basic biology, immunological tests and minor labial salivary gland biopsy. SGUS in mode B was performed by the same experienced operator, who was blinded to the diagnosis. Agreement between the two sets of criteria was assessed using Cohen's κ coefficient and the characteristics of discordant patients were detailed.
Results 290 patients were prospectively included between 2006 and 2016. Mean age was 55.6±13.2 years, 92.1% were female and mean duration of the symptoms 6.0±6.6 years. More patients fulfilled ACR/EULAR criteria (n=125, 43.1%) than AECG criteria (n=114, 39.3%). 114 patients (39.3%) fulfilled both criteria, 11 (3.8%) fulfilled ACR/EULAR only, 0 AECG only and 165 (56.9%) none of the criteria. Concordance between both criteria was good (kappa=0.9). Compared to patients fulfilling both criteria, patients fulfilling ACR/EULAR but not AECG criteria (n=11) had similar age, similar symptom duration, but less frequent sicca symptoms (eye dryness 18.2% versus 96.5%; mouth dryness 54.5% versus 97.4%, p<0.01 for both), and less frequent salivary gland dysfunction (salivary flow ≤0.1 ml/min: 20% versus 70.9%, p<0.01).They had characteristic features of pSS, with frequent systemic involvement at diagnosis (90.9%), positive salivary gland biopsy (90.9%), abnormal SGUS (44.4%) and presence of anti-SSA/SSB autoantibodies (45.4%). 46.5% of them had a diagnosis of pSS according to the physician. Among patients negative for the two sets of criteria, 12% had an abnormal SGUS and 10.3% received a clinical diagnosis of pSS based on physician opinion.
Conclusions Agreement between AECG criteria and new ACR/EULAR criteria is good suggesting that they select quite similar patients. ACR/EULAR criteria display a slightly higher sensitivity and are able to detect more patients with systemic involvement, but some of these patients did not have pSS according to the physician diagnosis. As previously demonstrated for AECG criteria, SGUS inclusion into ACR/EULAR criteria may further enhance their sensitivity.
Disclosure of Interest None declared