Background: Thrombotic thrombocytopenic purpura (TTP) presents with microangiopathic hemolytic anemia, thrombocytopenia and microvascular thrombosis. Arterial thromboembolic events are relatively common and well-described in these patients. However, the literature describing venous thromboembolism (VTE) in this patient population is scarce and outdated. We hypothesize that patients with TTP are at greater risk for VTE due to the use of central venous catheters and use of solvent-detergent plasma (SDP) for plasma exchange (PLEX).

Methods: Eligible patients from a single tertiary care centre were identified through apheresis records, and a retrospective chart review was conducted. The criteria for TTP diagnosis included the presence of accepted clinical/laboratory criteria and ADAMTS13 activity <10%. We considered the diagnosis of VTE if any of the following were present: pulmonary embolism (PE), deep vein thrombosis (DVT) or superficial vein thrombosis (SVT). Data were analyzed using simple descriptive statistics. Institutional research ethics board approval was obtained.

Results: We identified 77 consecutive patients with 123 episodes of TTP between January 1, 2008 and December 31, 2018. 51 patients (66%) were female, and 74 (96%) had immune TTP. Standard of care was daily PLEX and high-dose steroids, while rituximab was used for relapsed/refractory disease. In addition, 6 patients received caplacizumab. 13 of the 77 patients (17%) experienced a VTE (14 events: 6 PE, 5 DVT, 3 SVT), all of which were acute and associated with admission for either first presentation or relapse of TTP. None of the VTEs were catheter related. 8/13 patients were female, and all had immune TTP. Median age at diagnosis of VTE was 45 years (IQR: 36.0-56.5 years). 10/14 events were symptomatic; diagnosis was confirmed a median of 1.5 days after VTE symptom onset (IQR: 1-3 days). All patients were treated with PLEX, receiving a median of 17 exchanges (IQR: 10-23 exchanges). In addition, 13/14 TTP episodes were treated with corticosteroids, 9 with rituximab, and no patients received caplacizumab. 8 patients experienced a VTE while receiving daily PLEX; the majority (6/8) were being exchanged with SDP. VTE was diagnosed a median of 13.5 days after initiation of PLEX (IQR: 9.2-23.5 days). At TTP presentation, median platelet count was 11 x 109/L (IQR: 7-16 x 109/L), and median LDH was 894 U/L (IQR: 508-1272 U/L). At VTE diagnosis, the median platelet and LDH levels were 170 x 109/L (IQR: 126-248 x 109/L), and 232 U/L (IQR: 176-254 U/L) respectively. No patients had D-dimer testing at VTE symptom onset or diagnosis. Of the 6 patients with PE, 3 had elevated troponin levels (>0.040 ug/L) at VTE diagnosis. In terms of VTE risk factors, median BMI was 30.0 (IQR: 28.7-32.1); the majority of patients (7/13) were obese (BMI>30). 4/13 patients were active cigarette smokers. No patients had a history of lupus or antiphospholipid antibody syndrome. 2 patients had a past history of VTE, one was associated with a prior TTP episode. Most VTEs (8/14) occurred while patients were not on any pharmacologic thromboprophylaxis. For the remaining patients, dalteparin (3), ASA (2) and rivaroxaban (1) were used. VTE events were treated with direct oral anticoagulants (DOACs) including rivaroxaban and apixaban, in 10 cases. 7 of these patients were started on a DOAC after a brief initial period of low molecular weight or unfractionated heparin, 3 patients were given DOAC as upfront therapy. 4 cases were treated with warfarin (3 bridged with heparin and 1 following rivaroxaban). One distal DVT was not treated. None of the VTEs were fatal.

Discussion: The point prevalence of VTE was 17% over one decade amongst patients with TTP; this is higher than previously reported in the literature. Our data also suggests that TTP patients may be at greater risk for VTE compared to general hospitalized medical patients. The majority of affected patients on PLEX were receiving SDP at VTE diagnosis. SDP contains reduced levels of proteins C and S and may have a pro-thrombotic effect. Almost all VTEs occurred after platelet recovery and normalization of hemolytic markers, suggesting a different pathophysiology than arterial thrombosis in this setting. VTE thromboprophylaxis was uncommon since it was often held due to thrombocytopenia and not resumed upon platelet recovery. Our findings suggest the need to implement VTE thromboprophylaxis earlier in patients admitted with TTP.


Sholzberg:Novartis: Honoraria; Amgen: Honoraria, Research Funding. Pavenski:Ablynx: Honoraria, Research Funding; Alexion: Honoraria, Research Funding; Shire: Honoraria; Octapharma: Research Funding; Bioverativ: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.