January 2000, Volume 9, Issue 1, Pages 33 - 41

Paper

Testing for the antiphospholipid syndrome: importance of IgA anti-beta 2-glycoprotein I TP Greco1,2, MD Amos3, AM Conti-Kelly2, JD Naranjo4 & JW Ijdo1 1Section of Rheumatology, Yale University School of Medicine, New Haven, CT, USA     2Department of Medicine, Section of Rheumatology, St Mary's Hospital, Waterbury, CT, USA     3DiaSorin, Inc., Stillwater, MN, USA     4Department of Mathematics and Statistics, Western Michigan University, Kalamazoo, MI, USA     Correspondence to: ThomasP Greco , 133 Scovill Street, Suite 306-307, Waterbury, CT 06706, USA Tel: (+1) 203 597 3667; fax: (+1) 203 597 3663

Keywords

antiphospholipid antibodies;   anti-beta 2-glycoprotein I;   antiphospholipid syndrome

Abstract

Background: Testing for the antiphospholipid syndrome (APS) using anticardiolipin antibodies (aCL) has been problematic. Titers may fluctuate or even become negative. Anti-beta 2-glycoprotein I assays (a2-GPI) may be more reliable for diagnosis. Methods: In a prospective, blinded study over a nine-month period we retested all patients seen for routine follow-up visits in our clinic who had previously been evaluated for aCL-associated illnesses. Patients were stratified into two groups: group A--patients previously positive for aCL; group B--patients previously negative for aCL. Both groups were further classified according to disease severity. Patients were retested for both aCL and a2-GPI (isotypes G, M, A for each) using uniform testing standards. Results: 118 patients with previously positive aCL (group A) were retested. Repeat aCL were positive in 52/118 (44%), a2-GPI positive in 69/118 (58%) and 82/118 (69.5%) were positive for one or both assays. In patients with serious organ damage (92% with documented APS), 48.6% were aCL positive, 64% positive for a2-GPI, and 75.7% were positive for one or both assays. When only one assay was positive, a2-GPI was most frequent (P=0.0096). Overall, IgA a2-GPI was the most frequent isotype found (60.9%). On retesting of 73 aCL-negative patients (group B), 9/73 (12%) were aCL positive, 27/73 (36%) were a2-GPI positive, with 24/73 (32.9%) having isolated a2-GPI. Of those positive for a2-GPI, IgA a2-GPI was present in 74.1%. Many of these patients had documented APS. Conclusion: Based on our data, a2-GPI assays are superior to aCL assays for diagnosis of APS. The combined use of both assays enhance positive testing results in up to 75% of patients with APS at any stage of illness. ACL negative patients suspected of having APS should be retested for both a2-GPI and aCL. IgA a2-GPI appears to be the most important isotype detected. Lupus (2000) 9, 33-41

Received 10 June 1999; Accepted 1 September 1999

^© Macmillan Publishers Ltd 2000