• Autoantibodies
• Autoimmune Diseases
• Systemic Lupus Erythematosus

Agmon-Levin et al1 formulated recommendations for the assessment of antinuclear antibodies (ANA). Indirect immunofluorescence (IIF) is considered the reference method for ANA screening, which is in agreement with the American College of Rheumatology (ACR) position statement.2 The recommendations are based on current knowledge and expert experience. However, as recognised by Meroni and Schur,2 no well-planned studies comparing the diagnostic accuracy of the old IIF and the new methods have been undertaken.

In a recent study, IIF was compared to a new automated method for connective disease screening (fluoroenzymeimmunoassay (FEIA) (EliA CTD screen, Thermo Fisher)) using samples obtained at the time of diagnosis from well-characterised patients and controls.3 Diagnostic accuracy data (for systemic lupus erythematosus (SLE), systemic sclerosis (SSc) and Sjögren's syndrome (SS)) from this study is recapitulated in table 1. For comparison, data from a multinational study4 on IIF is included in the Table as well. The results for IIF in the two studies were comparable, thereby validating the study population. A higher specificity was observed for FEIA than for IIF (especially at IIF cutoff dilution 1:80), whereas sensitivity was higher for IIF than for FEIA for SLE and SSc, but not for SS. The diagnostic performance of FEIA was comparable to the performance of a multiplexed bead assay.5

For the combination of IIF and FEIA, the highest likelihood ratios (LR) (35–50) were found for double positivity (FEIA and IIF), the lowest (0.03–0.11) for double negativity. IIF positivity combined with FEIA negativity (22–25% of SLE and SSc patients) had a low LR (<5).

Using the area under the receiver operating characteristic curve (AUC), the diagnostic performance of four diagnostic strategies was compared; (1) IIF on all samples, (2) FEIA on all samples, (3) IIF on all samples and FEIA on IIF-positive samples and (4) both tests (IIF and FEIA) on all samples. The results are summarised in figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Receiver operating characteristics (ROC) curves for different diagnostic strategies; (1) only IIF is performed on all samples, (2) only fluoroenzymeimmunoassay (FEIA) is performed on all samples, (3) IIF is performed on all samples and FEIA on IIF-positive samples, (4) both tests (IIF and FEIA) are performed on all samples. The area under the curve is given, as well as a statistical comparison of the areas under the curves between the different strategies (for 1:80 and 1:160 dilution). The figures and area under the receiver operating characteristic curve (AUC) values are for IIF cutoff dilution at 1:80.

For SLE, the best strategy was performing both tests on all samples. The second best strategy (p=0.08 for IIF cutoff dilution 1:80; p=0.03 for IIF cutoff dilution 1:160) was screening with IIF and FEIA on IIF-positive samples. The AUC of these strategies was significantly higher than the AUC of only IIF or only FEIA.

For SSc, screening with IIF and performing FEIA on IIF-positive samples was comparable to IIF and FEIA on all samples. The AUC of these strategies was significantly higher than the AUC of only IIF or only FEIA.

For SS, the best strategy was performing both tests on all samples. The AUC of this strategy was not significantly higher than the AUC of only FEIA, but was significantly higher than the AUC of screening with IIF and FEIA on IIF-positive samples.

Taken together, our results show that favourable strategies are disease-dependent, and that combining IIF with solid-phase assay can increase the diagnostic information (eg, refined estimation of LR for disease).