Antiphospholipid Syndrome (APS) is characterized by thrombotic events mediated by antiphospholipid antibodies (aPL). Most patients have primary APS (PAPS), while a significant minority has SLE (Ann. Rheum. Dis. 74, 1011). Although PAPS affects 0.5% of the population (Thrombosis Res. 151, S43), it may be underdiagnosed in the absence of SLE. Thus, we evaluated aPL testing in non-SLE patients with deep venous thrombosis (DVT), pulmonary embolism (P), and stroke or transient ischemic attack (S/TIA).

Methods:

1835 patients were evaluated for PAPS at our Institution between 2010 and 2018: 513 were diagnosed with PE, 583 with DVT, and 739 with S/TIA. Lupus anticoagulants were assessed by hexagonal phase phospholipid neutralization assay (HPPNA) and diluted Russell viper venom test (dRVVT) using Stago protocols (Parsippany, NJ, USA). Platelet neutralization procedure (PNP) was performed, as earlier described (Am. J. Clin. Pathol. 79, 678). IgG and IgM antibodies to β2-glycoprotein 1 (aβ2-IgG, aβ2-IgM) and cardiolipin (aCL-IgG, aCL-IgM) were measured in house while IgA isotypes (aβ2-IgA, aCL-IgA) were tested by LabCorp (Burlington, NC). Sensitivities, specificities, positive (PPV) and negative predictive values (NPV) were assessed by 2-tailed χ² tests using GraphPad software.

Results:

In patients with PE, HPPNA was the most sensitive individual test for detecting APS (from 349/513 tested, 144/349 resulted in positive HPPNA (41%). 50/513 patients were assessed with all 9 tests; 30/50 had at least one positive result (sensitivity: 60%; p<0.0001 relative to HPPNA). In patients with DVT, HPPNA was also most sensitive for detecting APS (from 485/583, 216/485 resulted in positive HPPNA (45%), p<0.01 relative to all other individual tests). 47/583 patients were assessed with all 9 tests; 36/47 had a least one positive test (sensitivity: 76.6%; p<0.0001 relative to HPPNA). In 739 patients with S/TIA, PNP was the most sensitive individual test for detecting APS (from 81/739, 26/81 resulted in positive PNP (32%). 54/739 patients were evaluated with all 9 tests; 39/54 had a least one positive test (sensitivity: 72%; p<0.0001 relative to PNP).When combining patients with DVT, PE, and S/TIA (Table 1), the complete 9-test panel yielded the greatest sensitivity for detecting APS (105/151, 69.5%) relative to the most sensitive individual test, HPPNA (382/1060, 36%; p<0.0001). The 9-test panel had similar specificity and PPV but markedly higher NPV at 46% over any individual test (p<0.0001).

Conclusion:

This study indicates that a 9-test aPL panel has superior sensitivity for detecting PAPS in patients with DVT, PE, or Stroke/TIA. However, the complete panel was only performed in 151 of 1835 patients with thrombotic events. These findings have major implications for diagnosis of PAPS patients who need lifelong anticoagulation for preventing potentially fatal thrombotic events.