Thrombotic Events in Lupus Anticoagulant Positive Patients Prospectively Correlate with Clinical Risk Factors - the Vienna Lupus Anticoagulant and Thrombosis Study (LATS)

INTRODUCTION: Patients with the lupus anticoagulant (LA) are at an increased risk of thrombotic events (TE), which in turn increase the risk of death (Gebhart J et al. Blood. 2015. 125:3477). Understanding the determinants of thrombotic risk in LA patients may pave the way towards targeted thromboprophylaxis. To date, several retrospective studies have investigated the association between anamnestic thrombosis and certain factors, such as antibodies against cardiolipin and beta2-glycoprotein I. However, robust prospective evidence is still limited. We aimed to investigate clinical and laboratory risk factors for development of TE in patients with a persistently positive LA.

PATIENTS & METHODS: In this prospective, observational cohort study with a baseline biobank we followed 150 patients (median age: 41.3 years, interquartile range (IQR): 32.3-60.2, female gender: n=122 (81.3%)), who tested repeatedly positive for the LA until the development of TE, death, or censoring. The primary endpoint was the time-to-TE during the observation period, defined as a composite of arterial or venous, independently-adjudicated thrombotic complications. Ninety-eight (65.3%) of the 150 patients had a history of at least one TE event (arterial TE: n=21, venous TE: n=84, both: n=7), and 70 (46.7%) were on oral anticoagulation at baseline. Sixty-five (43.3%) patients also had IgM and/or IgG isotype antibodies against cardiolipin and beta2-glycoprotein I (“Triple positivity”). For investigation of the LA, lupus-sensitive aPTT reagents were used (aPTT-LA, Diagnostica Stago, Asnieres, France). Prospective associations were analyzed using competing risk analysis treating death-from-any-cause as the competing event. Evaluated risk factors included (1) the lupus-sensitive activated partial thromboplastin time (aPTT-LA), (2) antibodies against cardiolipin, beta2-glycoprotein I, domain 1 of beta2-glycoprotein I, prothrombin, and “triple positivity”, (3) general cardiovascular risk factors (diabetes, hypertension, smoking, and hypertriglyceridemia, body mass index), and (4) AnnexinA5 resistance (A5R).

RESULTS: During a median follow-up period of 8.7 years (range: 12 days – 12.4 years), 30 TE events occurred (arterial TE: n=14, venous TE: n=16), and 20 patients died. The cumulative incidence of TE at 1, 5, and 10 years of follow-up were 4.0% (95% Confidence Interval (CI): 1.7-8.1), 12.8% (95% CI: 7.9-18.9), and 24.4% (95% CI: 16.8-32.8), respectively. In univariable analysis, a prolonged  aPTT-LA (Subhazard ratio (SHR) for aPTT≥118 seconds (i.e. 75^th percentile of the aPTT-LA distribution)=2.59, 95%CI: 1.27-5.31, p=0.009), diabetes (SHR=4.36, 95%CI: 1.44-13.19, p=0.009), active smoking (SHR=2.35, 95%CI: 1.16-4.77, p=0.018), and elevated triglycerides (SHR per 50mg/dL increase=1.14, 95%CI: 1.09-1.18, p<0.001) were associated with a higher risk of TE events. In multivariable analysis, the only independent predictors of a higher thrombotic risk were a prolonged aPTT-LA (adjusted SHR=2.33, 95%CI: 1.06-5.13, p=0.035), diabetes (adjusted SHR=3.79, 95%CI: 1.12-12.89, p=0.033), and active smoking (adjusted SHR=2.62, 95%CI: 1.24-5.52, p=0.012). These results prevailed after adjusting for anticoagulation at baseline. Using these three parameters allowed identification of clinical subgroups with a very high and a low risk of TE events (cumulative risk of TE after 5 years 43.3% in the high risk group and 5.6% in the low risk group, respectively, Figure 1). The other risk factors did neither in univariable nor in multivariable analysis emerge as predictors of thrombotic risk in this large cohort. 

CONCLUSION: Diabetes and smoking, which are established risk factors for vascular events in the general population, turned out to be relevant also in patients with the LA. Moreover, a very long lupus sensitive aPTT was as well predictive for occurrence of TE in these patients. This effect was independent of anticoagulation. Interestingly, disease defining antibodies, such as those against cardiolipin or beta2-glycoprotein I (including those against domain I) were not associated with future occurrence of TE in this LA positive patient population. These data suggest that above standard anticoagulation, interventions to control and improve metabolic status and smoking habits might influence the rates of future TE in patients with known persistent LA.