Abstract

AIMS: We investigated the prevalence and clinical relevance and significance of the various immunoglobulin (Ig) isotypes of the different aPLs and lupus anticoagulant (LAC).

METHOD: Our study registry for the year 2003 included 472 patients who had a complete antiphospholipid antibody laboratory work up and clinical evaluation. 137 had the presence of LAC, detected by the dilute Russell’s viper venom time (Dade Behring, Marburg, Germany), PTT-LA (Diagnostica Stago, Asnieres, France), or hexagonal phospholipid neutralization test (Diagnostica Stago, Asnieres, France); 211 had elevated titers of antibodies of IgG, IgM or IgA isotypes against cardiolipin (aCL), phosphatidylserine (aPS), or β2-glycoprotein-I (aβ2GPI), detected by ELISA (Corgenix, Inc, Westminister, CO, USA). 204 had no evidence of either LAC or aPL.

RESULTS: Of the patients with aPL, 67 (32%) had elevated titers of IgG, 126 (60%) had IgM, and 118 (56%) had IgA. Thrombotic events occurred in 61% (41 of 67), 52% (65 of 126), and 61% (72 of 118) of the patients with IgG, IgM, and IgA respectively. Patients with IgG and IgA isotypes of aCL, aPS and aβ2GPI had higher rates (58–67%) of thrombotic events (see table 1) than patients with IgM isotypes (46–50%). Stepwise logistic regression analysis identified elevated titer of IgA of any aPL as an independent risk factor for thrombosis (see table 2), even in the absence of LAC. Thrombotic events observed include deep venous thrombosis (25%), pulmonary embolism (12%), cerebrovascular accident (50%), myocardial infarction (8%), and peripheral arterial occlusion (3%); pregnancy complication, retinal vascular thrombosis, and dialysis access thrombosis were also seen.

CONCLUSION: Our study demonstrates that aPL of IgA isotype is prevalent and significantly associated with thrombotic events.

Proportions of Patients with Thrombosis amongst various aPLs and Ig Isotypes

aCLaPSaβ2GPITotal
* Two-sided P < 0.05 by McNemar’s test for equality of two dependent proportions 
IgG 28/43 (65%) 23/36 (64%) 13/21 (62%)* 41/67 (61%) 
IgM 17/37 (46%)* 28/58 (48%) 45/90 (50%)* 65/126 (52%) 
IgG 15/26 (58%)* 22/33 (67%) 54/91 (59%)* 72/118 (61%) 
Total 44/79 (56%) 58/105 (55%) 85/159 (53%) 118/211 (56%) 
aCLaPSaβ2GPITotal
* Two-sided P < 0.05 by McNemar’s test for equality of two dependent proportions 
IgG 28/43 (65%) 23/36 (64%) 13/21 (62%)* 41/67 (61%) 
IgM 17/37 (46%)* 28/58 (48%) 45/90 (50%)* 65/126 (52%) 
IgG 15/26 (58%)* 22/33 (67%) 54/91 (59%)* 72/118 (61%) 
Total 44/79 (56%) 58/105 (55%) 85/159 (53%) 118/211 (56%) 

Risk Factors for Thrombosis

AntibodyOdds Ratio95% Wald Confidence Limit
* p < 0.05 by the Wald chi-square test. Hosmer and Lemeshow goodness-of-fit test was used to assess the model fit. N=472 for the 1st four analyses, N=335 for the last three analyses. 
LAC 1.627 1.073–2.466* 
IgG regardless of LAC 1.364 0.781–2.384 
IgM regardless of LAC 0.874 0.569–1.344 
IgA regardless of LAC 1.587 1.012–2.490* 
IgG LAC negative 1.790 0.863–3.714 
IgM LAC negative 0.878 0.515–1.499 
IgA LAC negative 1.812 1.035–3.172* 
AntibodyOdds Ratio95% Wald Confidence Limit
* p < 0.05 by the Wald chi-square test. Hosmer and Lemeshow goodness-of-fit test was used to assess the model fit. N=472 for the 1st four analyses, N=335 for the last three analyses. 
LAC 1.627 1.073–2.466* 
IgG regardless of LAC 1.364 0.781–2.384 
IgM regardless of LAC 0.874 0.569–1.344 
IgA regardless of LAC 1.587 1.012–2.490* 
IgG LAC negative 1.790 0.863–3.714 
IgM LAC negative 0.878 0.515–1.499 
IgA LAC negative 1.812 1.035–3.172* 

Author notes

Corresponding author