Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.

Anti-Phosphatidylserine/Prothrombin Antibody Titers Are Strongly Correlated with Lupus Anticoagulant Assays in Patients with Antiphospholipid Antibodies.

Miyara1,  Makoto, Arnaud1,  Laurent, Dufat1,  Laurent, Diemert1,  Marie-Claude, Ankri1,  Annick, Mathian1,  Alexis, Haroche1,  Julien

CHU Pitié-Salpêtrière, Paris, France
Instrumentation Laboratory Werfen Group, Lexington, MA


Biological criteria for the antiphospholipid syndrome (APS) diagnosis are the detection of either anticardiolipin antibodies (aCL), antib2GPI antibodies (ab2GPI) or lupus anticoagulant (LA). LA is strongly associated with thrombotic events in APS. Detection of LA is complex as it requires several confirmatory steps, which may lead to delayed results. Anti-phosphatidylserine/prothrombin antibodies (aPS/PT) have been recently associated with the presence of LA. However, it is still unclear whether aPS/PT titers correlate with LA assays.


504 sera from 101 patients with either definite APS syndrome (n=82) or stable antiphospholipid antibodies (APL, either aCL, ab2GPI or LA; ACL levels >80 UGPL) without clinical APS manifestations (n=19) primarily collected for aCL, b2gPI and/or LA detection, were tested on a aPS/PT (IgG+IgM) ELISA and on individual isotype aPS/PT IgG and aPS/PT IgM ELISA assays (INOVA Diagnostics, San Diego, CA). Correlations between titers of aPS/PT and those of aCL, ab2GPI, with Rosner index, dilute tissue thromboplastin inhibition ratio (dTTI) and dilute Russell's viper venom time ratio (dRVVT) were assessed using Spearman's non-parametric test. According to manufacturer instructions, aPS/PT titers were considered positive when >30 U/mL.


aPS/PT were present in 12 of the 19 patients (63 %) with APL without thrombotic or obstetric events (4 with IgG isotype, 4 with IgM isotype and 4 with both) and in 69 of the 82 patients (84 %) with definite APS (19 with IgG, 11 with IgM and 39 with both isotypes).

Among the 81 patients with positive LA, 74 had aPS/PT (91 %) while 7 did not (9 %). Among the 20 patients without LA, 16 did not have aPS/PT (80 %) while 4 did (20 %). Presence of aPS/PT was significantly associated to the presence of LA (p<0.0001 using a chi-square test). 10 of 19 patients without APS had LA among whom 9 had aPS/PT (90%). Among the 9 other patients without LA, 3 had aPS/PT (33%). In the 82 patients with APS, 68 patients had LA, among whom 65 had aPS/PT (96%). Among the 14 other patients without LA, 4 had aPS/PT (28%).

aPS/PT titers strongly correlated with Rosner index (r=0.72, p<0.0001), dTTI (r=0.801, p<0.0001) and dRVVT (r=0.67, p<0.0001). The ROC Curve yielded an excellent sensitivity of 84%, specificity of 98%, and area under the ROC Curve of 0.92 for detecting the presence of LA using an aPS/PT cut-off titer of 53 U/mL.


We confirmed that the presence of aPS/PT antibodies (either IgG isotype, IgM isotype or both) is strongly correlated with the presence of LA. Furthermore, titers of aPS/PT were strongly correlated with Rosner index, dTTI and dRVVT. Our data suggest that aPS/PT (IgG+IgM) ELISA assay could be a good alternative test for the detection of LA.

To cite this abstract, please use the following information:
Miyara, Makoto, Arnaud, Laurent, Dufat, Laurent, Diemert, Marie-Claude, Ankri, Annick, Mathian, Alexis, et al; Anti-Phosphatidylserine/Prothrombin Antibody Titers Are Strongly Correlated with Lupus Anticoagulant Assays in Patients with Antiphospholipid Antibodies. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :19

Abstract Supplement

Meeting Menu