Assessment of thrombotic risk and pathophysiological mechanisms of splenectomy-associated hypercoagulability in patients with immune thrombocytopenia

Abstract Immune thrombocytopenia (ITP) is an autoimmune disease characterized by isolated thrombocytopenia in the absence of an alternative diagnosis. One of the second-line treatments after failure or resistance to corticosteroids is splenectomy. Although this therapy has high rates of durable response, there is concern about long-term complications, such as thrombosis and infection. The risk of venous thromboembolism appears to increase after splenectomy in patients with ITP; however, the reason this occurs is not well understood. The von Willebrand factor (VWF) cycle may be altered in patients who have undergone splenectomy, as the spleen is one of the organs responsible for clearing this factor. This study's primary aim is to evaluate whether patients with ITP undergoing splenectomy have an increased thrombotic risk. Subsequently, we will analyze the reasons for hypercoagulability in these patients, especially with regard to the VWF cycle. We conducted a retrospective, multicenter cohort study with patients diagnosed with primary ITP that were evaluated between 2005 and 2021. First, we compared the frequency of thrombosis in ITP in splenectomized patients to non-splenectomized patients. Subsequently, we measured factors involved in the VWF production and clearance cycle from splenectomized and non-splenectomized. We evaluated the association between splenectomy and thrombosis by Cox regression and splenectomy with factor VIII (FVIII) and VWF antigen by linear regression. Of the 320 patients included in the cohort, 119 underwent splenectomy, while 201 did not. The majority of the patients in both groups were female (76% in each group) and white (77% and 85%, respectively). The median age at diagnosis was 29 years (interquartile range [IQR] 21 – 45) in the splenectomized group and 35 years (IQR 24 – 49) in the non-splenectomized group (p=0.022). Antiphospholipid (aPL) antibody positivity occurred in 22% of the splenectomized group versus 12% of the non-splenectomized group (p = 0.037). There were no differences regarding comorbidities or antinuclear antibodies between the two groups. Finally, the splenectomized group had a lower platelet count at diagnosis than the non-splenectomized group (17x10⁹/L vs. 27x10⁹/L; p < 0.001). The median follow-up duration was 74.2 months and the median time from diagnosis to splenectomy was 15.4 months. The frequency of thrombosis was 12% in the splenectomized group, compared to 3% in the non-splenectomized group. The median time from splenectomy to thrombosis was 20.5 months. Cox proportional hazard analysis revealed that splenectomy was associated with a higher risk of thrombosis (hazard ratio [HR], 7.91; 95% confidence interval [CI]: 2.45–25.4; p < 0.001). This increased risk persisted after adjusting for sex, age, aPL antibodies, and Charlson comorbidity index. We also compared laboratory factors related to blood count and VWF cycle. These factors included FVIII, VWF antigen and ristocetin cofactor. We examined 47 patients with ITP (21 who had undergone splenectomy and 26 who had not). In a linear regression analysis adjusted for sex, age, and blood type, splenectomized patients demonstrated significantly higher levels of FVIII activity and VWF antigen compared to non-splenectomized patients. The absolute difference in FVIII activity was 34.7% (95% CI: 14.4–55; p = 0.006), and the absolute difference in VWF antigen was 43.4% (95% CI: 19.3–67.6; p = 0.004). Finally, we divided the patients into four groups: 1) no splenectomy, blood type non-O (n = 14); 2) no splenectomy, blood type O (n = 12); 3) splenectomy, blood type non-O (n = 10); and 4) splenectomy, blood type O (n = 11). Then, we analyzed factor VIII and VWF antigen levels between the groups. In the linear regression analysis, patients who underwent splenectomy with a non-O blood type had increased VWF antigen levels in 86.3% (95% CI: 44.8%-127.7%; p<0.001) and increased FVIII activity in 72.2% (95% CI: 37.2%-107.1%; p<0.001) compared to non-splenectomized patients with blood type O.This study confirms previous results that splenectomy is associated with a higher risk of thrombosis in patients with ITP. We also showed that splenectomized patients with ITP have higher levels of VWF antigen and FVIII, especially those patients who have blood type non-O. This increase in VWF and FVIII activity could explain why thrombotic events occur even many years after splenectomy, possibly due to a persistent hypercoagulable state.