Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Assessing the ANA Test as a Screening Tool: Utility or Futility?

Abeles,  Aryeh M., Abeles,  Micha


Current ANA testing may be overly sensitive, resulting in frequent false positives as well as unnecessary follow-up testing and misdiagnosis. Several studies have focused on ANA prevalence and predictive value, but the purpose of this investigation was to evaluate the clinical utility of a positive ANA test in a real-world setting, by reviewing the outcomes of patients referred to a tertiary rheumatology clinic for evaluation of a positive ANA test. No prior study has directly addressed this question.


We reviewed records for all consultation patients presenting to the authors at the UCHC rheumatology clinic between July 2007 and July 2009. Patients were included in the evaluation if they had been referred for a positive ANA. Patients presenting with an already-diagnosed ANA-associated rheumatic disease (AARD) were excluded. All ANA tests and all referrals had been ordered by non-rheumatologists. A complete review of systems and physical examination was performed on each patient. 91 out of 101 patients with ANA titers >=1:320 had the following labs drawn: antibody (Ab) panel (Anti-dsDNA, SSA/SSB, Sm, RNP), C3, C4, CH50, CBC, and ESR. Of 67 patients with ANA >=1:640, 64 had lab evaluations performed as above. Referring physician reasons for ordering ANA testing were also collated.


Of 1,306 initial consultation visits to our clinic, 232 were referrals for a positive ANA, with titers ranging from 1:40 to greater than 1:5120. Initial ANA testing had been performed at several different laboratories (the majority at UCHC). 98% of all initial ANA tests had been run using immunofluoresence assay with Hep-2 cells.

9.1% of patients evaluated (21/232) had an AARD (5 SLE, 12 Sjogren's syndrome, 3 scleroderma, 1 MCTD). The remainder of patients had no AARD, of whom twenty-one (9.1%) had isolated autoimmune thyroid Ab present. No patient with an ANA <=1:80, and only 1 patient with ANA<=1:160, had a rheumatic disease. There was no correlation between the highest ANA titers with a specific rheumatic illness. The most common reason for ordering ANA testing was widespread pain and tenderness (54/232, 23.2%) but also included, among others, chronic lower back pain, episodic chronic rhinitis, palpitations, thinning hair, unilateral heel pain, and lateral buttock pain. The positive predictive value of an ANA test in our cohort of patients was 2.2% for lupus and 9.1% for any AARD.


In this retrospective study, over 90% of patients referred to a tertiary rheumatology clinic for positive ANA testing had no evidence of an AARD. The poor predictive value of a positive ANA test may in part be attributable to poor patient selection with concomitant low pretest probabilities. The extremely low positive predictive value of ANA tests at titers below or equal to 1:160 suggests that the cutoff of a "positive" ANA may need to be re-evaluated by testing laboratories. Even at high titers, however, many patients referred for a positive ANA had no discernible AARD, corroborating previously noted lack of predictive value of ANA testing.

To cite this abstract, please use the following information:
Abeles, Aryeh M., Abeles, Micha; Assessing the ANA Test as a Screening Tool: Utility or Futility? [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1845
DOI: 10.1002/art.29610

Abstract Supplement

Meeting Menu