ab21409

Relative value of commercial kits for ANA testing

A.S. Russell¹, C. Johnston²

¹Department of Medicine and 2Rheumatic Disease Unit Laboratory, University of Alberta, Edmonton, Alberta, Canada.

ABSTRACT
Aims
We have tested the relative performance of 20 commercial ANA test kits along with that of our own laboratory to assess whether one was clearly superior.

Methods
The sera were drawn from 3 pools that had all been pre-tested in our laboratory: patients with definite SLE; patients with non-connective tissue diseases (CTD), but where a positive FANA had been found; and normal blood donors. The tests were used in accordance with the recommendations of the specific supplier but in a routine serology laboratory.

Results
Sensitivity and specificity ranged between 38 and 100%. While the negative predictive value of 4 ELISA kits was 100%, and most others were close, the HEp-2 kits were 100% in only 1 case. A positive predictive value of 100% was also seen with 1 kit.

Conclusion
Some of the tests are clearly better than others, but the choice may differ depending on the clinical needs, e.g. preference for a good positive or negative predictive value. However, the ELISA kits offered better results than the immunofluorescent technique. Two of them had sensitivity/specificity of > 90%.

Key words
ANA kits, predictive value, sensitivity, specificity.

Please address correspondence to: Dr. A.S. Russell, 562 Heritage Medical Research Centre, University of Alberta, Edmonton, Alberta, T6G 2S2 Canada.