ANTINUCLEAR ANTIBODIES IN PRIMARY CARE SETTING: IS IT WORTH IT?

Background: Antinuclear antibodies (ANA) are the most frequently used screening tests for connective tissue diseases. However, their diagnostic value depends on the pre-test probability of such conditions. Objectives: To evaluate the usefulness, clinical correlates and associated direct costs of ANA testing in the primary care setting in an Early Arthritis Clinic (EAC) referral cohort. Methods: A retrospective study of consecutive patients referred to the EAC between 2011 and 2018 was conducted. Referral is based on the fulfillment of specific criteria: presence of arthritis or clinically suspected arthralgia beginning in the previous 12 months, plus suggestive laboratorial abnormalities (rheumatoid factor, C-reactive protein or erythrocyte sedimentation rate). Many general practitioners also performed ANA testing (ANA-GP) and all patients underwent ANA testing, per protocol, in EAC (ANA-EAC). All patients having these 2 separated ANA results were included in the analysis. ANA-EAC titters and pattern were assessed by indirect immunofluorescence (Hep2, positive=titter≥1:160). Direct associated costs of ANA-GP were calculated, based on the mean charge of 3 different local labs. Positive (PVV) and negative predictive values (NPV) of ANA-GP for the diagnosis of inflammatory rheumatic disease, ANA-related rheumatic disease (ARD) and for the presence of ANA-EAC were determined. Results: 207 patients were referred to the EAC Clinic during this period (64.3% female, aged 53.9 ± 18.2 years-old). Fifty eight percent of these patients (n=120) had their ANA previously determined in primary care setting. Of these, only 9.2% of cases (n=11) were positive, this being one of the main reasons for referral. Only 73% percent of positive (n=8) and 24% of negative ANA-GP were confirmed as such in our lab. Of the 8 patients testing positive in both settings, 2 had no rheumatic disease, 2 had an ARD and 4 had another type of inflammatory rheumatic disease. ANA-GP PPV and NPV were: i) 18.2% and 92.7% (LR 2.44) for ARD,; ii) 63.6% and 27.5% (LR 0.74) for inflammatory rheumatic disease and 72.7% and 23.9% (LR 0.124) for a positive ANA-EAC result. The referral criteria with the highest PPV for the diagnosis of inflammatory rheumatic disease were: positive rheumatoid factor (76.2%), high erythrocyte sedimentation rate (71.6%) and clinical signs of arthritis (70.8%). The direct cost associated with duplicate ANA testing was estimated in 2.160€. Conclusion: ANA testing in the primary care setting had a poor predictive value in this cohort, which can be explained by its application in patients with low pretest probabilities for ARD. Although the direct costs may not seem impressive, we speculate the real cost to be much higher since ANA test rarely is requested solo but, instead, along with a lot of other autoantibodies in a “trawl fishing” attempt to diagnosis. ANA evaluations are not recommended for the study of putative arthritis cases in primary care and local campaigns should be promoted in order to improve referral quality avoiding unnecessary, costly and lengthy lab tests as ANA. Disclosure of Interests: None declared DOI: 10.1136/annrheumdis-2019-eular.4437Citation: Ann Rheum Dis, volume 78, supplement 2, year 2019, page A1378Session: Public health, health services research, and health economics (Scientific Abstracts)