Venous thromboembolism incidence and risk factors in patients undergoing hematopoietic stem cell transplantation.

Abstract

Background: Malignancy is a well-known risk factor for venous thromboembolism (VTE), and the Khorana risk score is effective for screening solid tumor patients. However, there is a lack of validated screening tools and established risk factors for patients undergoing hematopoietic stem cell transplant (HCT). Current literature reports VTE incidence in HCT patients ranging from 2.5% to 8.5%. Anticoagulation is difficult to manage post-transplant given prolonged thrombocytopenia and the likelihood of bleeding. By identifying risk factors, a predictive model may be developed to prospectively test prophylaxis strategies in patients at the highest thromboembolic event (TE) risk.

Objectives: This study evaluated the cumulative incidence of TE at 6 months following allogeneic and autologous HCTs. This study also aimed to identify risk factors for developing TE, to evaluate time from HCT to TE, and to compare one-year survival following HCT between patients experiencing TE and those who did not.

Study design: This is a retrospective single-center study evaluating the incidence of TE events in adult subjects undergoing HCT between March 2014 and December 2019. ICD-9 and ICD-10 codes were used to determine cancer diagnosis, TE events up 180 days after HCT, and comorbidities of interest. Each subject was reviewed for data accuracy by manual retrospective chart review. The study employed statistical tests such as the cumulative incidence method with competing risks, Gray's test, and univariate and multivariate Cox proportional hazards models to analyze the time to first thromboembolic (TE) event, evaluate risk factors, and assess 1-year survival post-HCT in relation to TE events within 180 days of HCT. Variables examined included age, sex, body mass index (BMI), transplant type, hospital length of stay, history of TE prior to transplant, active infections, graft-versus-host disease (GVHD), veno-occlusive disorder (VOD), cytomegaly virus (CMV) and other factors.

Results: The study included 636 evaluable patients; the majority were male (57.9%), white (68.7%), and underwent an autologous HCT (68.4%). Twenty-nine patients (4.6%) experienced a TE event within 180 days post-transplant. TE events were more common in the allogeneic transplant group (n=13/201, 6.5%) than the autologous transplant group (n=16/435, 3.7%) (p=0.122). The cumulative incidence of TE was higher in patients who developed an active infection than those who did not (7.6% vs 3.1%, P=.011). Hospital LOS [HR 1.03, 95% CI 1.0-1.06, p=0.036] and active infection [HR 2.34, 95% CI 1.1-4.95, p=0.027] were significantly associated with TE in the univariate analysis but were not retained in the final multivariate model. There was no difference in one-year survival among all patients who experienced a TE event and those who did not; however, in the autologous HCT subgroup, one-year survival rate was significantly lower in those with TE compared to those without TE (Figure 3c, 80.4% vs 95.3%, P=.01). Examined variables, including history of TE event and GVHD, were not associated with TE event risk.

Conclusion: While the overall incidence of TE in our study was low, many patients received pharmacologic or mechanical prophylaxis, suggesting such strategies may be effective in mitigating TE risk. Factors such as infection and hospital LOS may further increase TE risk. Providers should continuously monitor for risk factors and signs and symptoms of TE post-transplant. It is also imperative to consider chemical prophylaxis if counts are recovered while hospitalized.