Venous thromboembolism related to warm autoimmune hemolytic anemia: A case–control study

Abstract Background The risk of venous thromboembolism (VTE) during warm autoimmune hemolytic anemia (wAIHA) is apparent in several published series. Unlike proximate disorders (autoimmune thrombocytopenia, non-immune hemolytic diseases) little is known about the presentation and risk factors for VTE in this setting. Objective To determine the frequency, presentation and risk factors for VTE associated with wAIHA. Methods We performed a single center retrospective study of adult patients (> 18 years) followed for wAIHA between 2009 and 2013. VTE risk factors were systematically assessed. The characteristics of patients with or without VTE were compared. VTE presentation and precipitating factors were analyzed. The Padua VTE risk score was calculated in each case. Results Forty patients were included. wAIHA was idiopathic in 24 patients (60%). Twelve patients (30%) had Evans syndrome. Mean lowest hemoglobin level was 6.6 g/dl [3.7–11.5]. Eight patients (20%) presented VTE after the appearance of wAIHA, at a mean age of 52.5 years. All patients had pulmonary embolus, associated with a deep venous thrombosis in 4 cases. At the time of VTE 7/8 patients had frank hemolysis (median hemoglobin level: 7 g/dL) and 6/8 were outpatients with a low Padua VTE risk score. The frequency of usual VTE risk factor was similar in cases and controls. By contrast, lowest hemoglobin level was significantly lower in patients that experienced VTE (5.3 vs 7.2 g/dL, p = 0.016). During the first episode of wAIHA, patients with concurrent VTE had a more pronounced anemia (5.3 vs 7.4 g/dL, p = 0.026). At the time of VTE, anemia was more severe when no other precipitating factor was present (6 vs 8.9 g.dL, p = 0.04). Conclusion In our cohort, 20% of patients with wAIHA presented VTE. The vast majority of VTE occurred during severe hemolytic flares and were not attributable to usual VTE risk factors. VTE prophylaxis is advisable in any patient admitted for wAIHA, irrespective of Padua VTE risk score. Prophylaxis also seems reasonable for outpatients with marked hemolysis.