Background/Purpose:
There is no single clinical diagnostic test for primary Sjögren’s syndrome (pSS). For research purposes, multiple classification criteria have been proposed over the past decades. These include the 2002 American-European Consensus Group (AECG) criteria as well as the newly proposed (2012) American College of Rheumatology (ACR) criteria. We compared the performance of these two sets of criteria in a large carefully characterized sicca cohort.
Methods:
In a multidisciplinary (eye, mouth, and medicine) clinic for the evaluation of sicca, we determined the classification of subjects under the AECG and new ACR criteria(n=584) and the mRNA expression profile in whole blood of a subset of 201 participants (pSS by both criteria sets n=127, ACR+/AECG-: n =7, ACR-/AECG +: n=29 and controls n=38). The data points recorded for each patient for classification purposes were: the answers to subjective eye and mouth symptoms (according to the AECG questionnaire), Schirmer’s I test, ocular dye score (lissamine green + fluorescein staining of conjunctiva and cornea quantified either by the van Bijsterveld score or the OSS score), salivary whole unstimulated flow, histopathology of minor salivary gland biopsy, serology (antiRo, antiLa, ANA and rheumatoid factor antibodies).
Results:
The initial cohort of participants evaluated at either the Sjogren’s Clinic at Oklahoma Medical Research Foundation and/or the Sjogren’s Clinic at the University of Minnesota consisted of 748 individuals. Of these, a complete data set was available for 584 subjects and these constitute the study cohort. The two cohorts are comparable in terms of age, sex, race and ethnicity. Among the 584 subjects with complete data, 259 were classified as pSS under AECG, and 249 were so classified under ACR criteria while 222 met both sets of criteria and they represent 78% of all pSS cases. The two sets of criteria were not significantly different (p=0.26, McNemar’s test; concordance = 0.77, Kappa statistic, 95% CI=72.5-82.9); the ACR criteria had a sensitivity of 0.86 (95% CI=0.81-0.90) and a specificity of 0.92 (95% CI=0.88-0.94). Thirty seven subjects were classified as pSS by AECG only (ACR-/AECG+), of whom 29 (78%) had a minor salivary gland biopsy focal score >1, while 8 (22%) had positive anti-Ro/La. On the other hand, there were 27 ACR+/AECG- and they met ACR criteria mainly due to differences in the scoring of the corneal staining (OSS ≥3 for ACR; van Bijsterveld ≥4 for AECG). Interestingly, when attempting to correlate the subgroups generated by the classification criteria with their global gene expression profiles, we were unable to identify distinct clustering.
Conclusion:
When considering any subject who would meet classification for pSS by either set of criteria, only 75% would be so classified by both. Twenty five percent would be classified by only one set, leading to heterogeneity by either. This suggests that further refinement of the criteria using molecular data is warranted.
Disclosure:
A. Rasmussen,
None;
J. A. Ice,
None;
H. Li,
None;
K. Grundahl,
None;
J. A. Kelly,
None;
L. Radfar,
None;
K. S. Hefner,
None;
D. U. Stone,
None;
J. M. Anaya,
None;
M. Rohrer,
None;
G. D. Houston,
None;
D. M. Lewis,
None;
J. Chodosh,
None;
J. B. Harley,
None;
P. Hughes,
None;
J. S. Maier-Moore,
None;
C. G. Montgomery,
None;
N. L. Rhodus,
None;
A. D. Farris,
None;
B. M. Segal,
None;
C. J. Lessard,
None;
R. H. Scofield,
None;
K. Moser Sivils,
None.
« Back to 2012 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/effects-of-reclassification-using-the-american-college-of-rheumatology-criteria-on-a-large-cohort-sjogrens-syndrome-patients/