Background/Purpose :  The anti-nuclear antibody (ANA) is the classic biomarker associated with autoimmune connective tissue diseases (CTDs).  ANA testing is often ordered as part of the evaluation of patients with suspected CTD.  Different methods, including indirect immunofluorescence assays (IFAs), enzyme immunoassays (EIAs), and multiplex immunoassays (MIAs) are currently used by clinical labs for general ANA screening.  However, because the methodologies are very different, discordance between results may be observed, leading to challenges in interpretation.  The purpose of this study is to analyze the performance of these methods in a subset of patients referred for rheumatology consultation.

Methods:  A cohort of patients who had an ANA ordered for clinical assessment and a rheumatology consultation at our institution in 2010 – 2011were enrolled.  ANA testing was performed by IFA (Zeus Scientific), EIA (BioRad), and MIA (BioRad).  A titer of ≥1:40 for the IFA and a value >1.0 for the EIA were identified as positive.  Autoantibodies to dsDNA, chromatin, ribosome P, SS-A, SS-B, Sm, Sm/RNP, RNP, Scl-70, Jo-1 and centromere B were detected by MIA; positivity for at least 1 specific autoantibody identified the ANA as positive.  Sensitivities and specificities were computed and were compared using McNemar’s test.

Results:   In the study cohort (n=327; 81% female; mean age 52 [sd: 16] years), a subset had no identifiable autoimmune disease (non-auto; n=87) and the remaining patients had a diagnosed autoimmune CTD (auto CTD; n=240), including systemic lupus erythematosus (SLE; n=51), Sjogren syndrome (SS; n=34), mixed CTD/undifferentiated CTD (MCTD/UCTD; n=51), rheumatoid arthritis (RA; n=33), and various other miscellaneous CTDs (n=71).  Sensitivities and specificities of the IFA, EIA and MIA within these groups are shown in the table.

In the auto CTD group overall, IFA and EIA had similar sensitivities (p=0.83), both of which were significantly higher than MIA (p<0.001).  However, in the non-auto group, the MIA demonstrated significantly improved specificity compared to both the IFA and EIA (p<0.001).  In specific CTD groups, all SS patients were ANA positive by all 3 methods.  In SLE patients, the EIA and IFA had similar sensitivities (p=0.32); the sensitivity of the MIA was significantly lower than the IFA (p=0.046), and marginally lower than the EIA (p=0.08).  In the MCTD/UCTD group, the EIA and IFA were positive in all patients, and the MIA had significantly lower sensitivity (p=0.003).

Conclusion: The sensitivities of IFA, EIA, and MIA vary significantly depending on the specific diagnostic category.  Although the sensitivities of IFA and EIA tend to be improved over the MIA, the specificities of these methods are a significant issue.  Further comparisons of methods used for ANA screening can provide valuable information that could lead to improved interpretation and utilization of lab testing.