High Incidence of Intra-Cardiac Supra-Centimetric Thrombosis and Embolic Complications in Sickle Cell Disease Patients with Dual Lumen Ports

In Sickle Cell Disease (SCD) patients with cerebral vasculopathy and/or other severe complications, a transfusion program is indicated. A sustainable venous access is vital for patients engaged in these programs (5-10% of patients with SCD). Dual lumen ports (Vortextechnology) have emerged has a way to bypass the frequent local complications associated with arteriovenous fistulae.

Between July 2015 and December 2020, a dual lumen port was implanted in 27 SCD patients in transfusion program at Necker Hospital. Over a 30 months period, we observed 20 venous thromboembolic events (VTE) in 12 patients (44%): 18 were voluminous intracardiac thrombosis (Fig 1), asymptomatic (n=15) or complicated of pulmonary embolism (n=3), one jugular thrombosis and one superior cellar syndrome, leading to long-term curative anticoagulation, except in cases of contraindication such as Moya-Moya. VTE were diagnosed on average 18 months after implantation (range 1 to 29 months). This observation has considerably limited the use of the device, even though no complications were observed in 56% of patients, suggesting that a subgroup of patients may be at lower risk of thrombosis and could still benefit from a Vortex.

We conducted a clinical study aiming to identify potential biological markers related to Vortex thrombosis in SCD patients. Among the biological markers, alterations of thrombin generation and fibrino-formation, as well as potentially procoagulants items altered in SCD, such as microvesicules (MV), red blood cell (RBC) deformability and the intensity of splenic dysfunction were explored. Between January and June 2021, 11 healthy donors and 58 patients with SCD treated by erythrapheresis in the biotherapy department of Necker Hospital were included, 23 patients with a Vortex with (n=12) or without (n=11) intra-cardiac and/or other VTE and 36 patients without a Vortex (including 8 with a history of VTE). Thrombin generation Assay (TGA) in plasma and standard coagulation assays were performed on pre-transfusion sample, including Prothrombin Time (PT), activated partial thromboplastin time (APTT) and fibrinogen, dosage of FII, FV, FVII, FX and FXII, Protein C (PC), Protein C (PS), Antithrombin (AT) dosage. All samples were pretreated by activated carbon for removal of Direct Oral Anticoagulants from plasma.

Quantification and characterization of plasma MV, determination of erythrocyte deformability and point of sickling (ektacytometry with oxygenscan) and assessment of splenic function (quantification of pocked RBC and Howell Jolly bodies) were also performed.

We found a significant decrease in the mean levels of PC (69.7 vs 109.1%, p=0.0002) and PS (58.4 vs 99.4, p<0.0001) in SCD patients when compared to healthy donors while AT levels were only slightly decreased (91 vs 98,6%, p=0,04). We also observed a significant decrease in vitamin-K dependent coagulation factors: FII (77.5 vs 100.6%, p<0,0001), FVII (61.9 vs 84.5% p=0.005) and FX (71.8 vs 96.8%, p=0.0001) while FV (80.1 vs 93.8% p=0.02) was only slightly decreased and fibrinogen levels were comparable (2.62 vs 2.57g/L p=0.86). No significant differences were noted between SCD patient groups (with or without thrombosis).

Thrombin generation Assay (TGA) in plasma showed an increase in endogenous thrombin potential (ETP; 714 vs 451nM.min, p=0.04) and Peak (161 vs 95 nM.min, p=0.03) in SCD patients compared to healthy donors but no differences were found between patient groups.

Assessment of splenic function, point of sickling in oxygenscan and deformability of RBC measured by ektacytometry showed no differences between patient groups.

Patients with SCD had a significant increase of circulating CD235+ MV (93268 vs 15 207/µL, p<0.0001), Annexin 5 MV (473553 vs 248404 /µL) and CD235+ erythrocyte ghosts (1757 vs 61/µL) compared to healthy donors. Again, no differences were seen between patient study groups.

In conclusion, no biological predictive factors for intra cardiac Vortex thrombosis could be demonstrated by our study despite an extensive study of biological factors known to date to favor thrombosis. Given the high incidence of intracardiac thrombosis and the impossibility to date of predicting the risk of occurrence, we recommend that Vortex technology be reserved for patients who are already receiving effective anticoagulant therapy or for whom no other solution can be found, while maintaining a close surveillance of echocardiography.

No relevant conflicts of interest to declare.