The balancing act after pediatric venous thromboembolism Open Access

In this issue of Blood Advances, Lenahan et al¹ present findings from their study in which they evaluated recurrent thrombosis and major bleeding after pediatric venous thromboembolism (VTE). This single-center, retrospective cohort study included 632 consecutive patients, aged ≤21 years, who were treated with anticoagulation for VTE and who were seen in their outpatient anticoagulation clinic. The incidence rate of recurrent VTE was relatively high at 13.7 per 100 person-years, whereas the incidence rate of major bleeding was relatively low at 2.21 per 100 person-years. A multivariable analysis identified increased odds of recurrent VTE in patients with central venous catheters (CVCs), anatomic venous abnormalities, and inflammatory bowel disease (IBD). Identifying children who have an increased risk of recurrence may guide decisions around secondary thromboprophylaxis if bleeding rates are acceptable.

The recurrence rate in this study was higher than previous studies of unprovoked VTE in pediatrics (10.4 per 100 person-years)² and adults (10.3 per 100 person-years).³ The higher risk in this cohort may be because of the inclusion of patients with strong provoking risk factors who may benefit from intermittent thromboprophylaxis.

This study further supports that CVCs are the most common risk factor for pediatric VTE in that CVCs were the most prevalent risk factor for the index VTE (40.7%), and 14.8% of patients with an index CVC-associated VTE had a recurrent CVC-provoked VTE. This aligns with previous work that demonstrated a high risk for recurrence with a CVC-provoked VTE.^4,5 Although recent pediatric trials of direct oral anticoagulants (DOACs) vs standard anticoagulants reported low rates of recurrent thrombosis, these studies may not have included the patients with the highest risk for recurrence.^6,7 Notably, the proportion of children with a CVC-associated VTE in the DOAC trials was relatively low (18%-25%), which may partly explain their lower risk of recurrence.^6,7 

Interestingly, 10% of patients with recurrent thrombosis also had major bleeding, and 56% of patients with major bleeding had recurrent thrombosis. This highlights what pediatric hematologists see in clinical practice; patients with a high risk of bleeding are also at risk for thrombosis. Patients with a high bleeding risk were excluded from the DOAC trials, providing another potential explanation for the lower recurrence rates in those studies.

Although there are extension studies from the pediatric DOAC trials,^8,9 we have limited data on the long-term follow-up of pediatric patients off anticoagulation or those who were only treated during high-risk periods. The median time to recurrent VTE in this study was 295 days, and the median follow-up time was 357 days, suggesting that longer observational cohort studies are important. Notably, 26.7% of recurrent events in this study occurred while therapeutic anticoagulation was prescribed. Other work has demonstrated high rates of recurrence in pediatric patients with unprovoked VTE despite being prescribed anticoagulation.² Potential explanations for recurrence despite being on anticoagulation therapy include poor adherence, breakthrough clots in the setting of strong prothrombotic risk factors, or thrombosis development during anticoagulant holds for bleeding or procedures. These factors have not been well explored in pediatric populations and should be the focus of prospective studies that evaluate extended thromboprophylaxis.

This study excluded patients who were never treated or who were only treated in the hospital and those with cardiac disease who were managed separately. Despite these limitations, the authors identified groups to focus on for thromboprophylaxis both in research studies and in clinical practice. However, questions remain about the optimal dose, duration, and timing of anticoagulation therapy.

Reassuringly, the authors reported low rates of major bleeding. Similar to previous studies⁴ and our clinical experience, most major bleeding (76.5%) occurred in the hospital, highlighting that these patients likely had other bleeding risk factors in addition to receiving anticoagulation therapy. Capturing bleeding in a retrospective study is limited by variable documentation and requires thorough chart review. The primary bleeding outcome was major bleeding, as defined by the International Society on Thrombosis and Haemostasis,¹⁰ which requires an intervention, a significant hemoglobin drop, or an occurrence in a critical site. We would expect this to be captured in clinical documentation and have a lower likelihood of missing data or bias. The authors did not evaluate less severe bleeding that may still be problematic for a patient and may impact clinical decision-making around the anticoagulant dose and duration. This type of bleeding is more difficult to identify during retrospective review and would be best collected prospectively. Similar to the recurrence rates, the bleeding rates in pediatric DOAC trials were low,^6,7 but the trials excluded patients with a perceived bleeding risk who may receive anticoagulation therapy in clinical practice. This further highlights the importance of prospective collection of real-world bleeding outcomes using standardized definitions. Although DOACs have significantly decreased the burden of anticoagulation by offering oral agents with fewer interactions and limited monitoring, there remain quality-of-life implications with long-term anticoagulation therapy that have not been explored in pediatrics. These include activity restrictions, which are particularly important for young athletes, and the psychological burden of anticoagulation therapy. As we identify those with a long-term increased VTE recurrence risk, we must also investigate the implications of long-term anticoagulant use.

Identifying cohorts of children with VTE who have a low or high risk of recurrence will help to optimize treatment and secondary prevention strategies. The Kids-DOTT trial identified a subgroup of children with provoked VTE who had a low risk of recurrence and who could be safely treated with a short (6 weeks) duration of anticoagulation.¹¹ Currently, there are very limited data to guide which pediatric patients with VTE are most likely to benefit from extended-duration or targeted thromboprophylaxis. Lenahan et al provided important preliminary data to support consideration of anticoagulation for prophylaxis during high-risk times, such as with a CVC or IBD flare. Some questions still remain, including those related to the efficacy of thromboprophylaxis, adherence, the appropriate dose intensity, rates of less severe bleeding, and quality-of-life implications of extended-duration anticoagulation. These should be the focus of future pediatric studies that investigate thrombosis prevention. We have come a long way in understanding pediatric thrombosis through research like this, but we have more work to do to improve outcomes in children and adolescents with VTE.

Conflict-of-interest disclosure: The authors declare no competing financial interests.