A Retrospective Multicenter Study from the Nordic Society of Pediatric Hematology and Oncology (NOPHO) on Cerebral Sinus Venous Thromboses in Children with Acute Lymphoblastic Leukemia

Introduction

Cerebral sinus venous thrombosis (CSVT) is potentially life-threatening thrombosis with mortality around 10%. Venous thromboembolism (VTE) is a common complication in children with cancer. These children have several thrombotic risk factors such as the malignancy itself, severe infections, prothrombotic medication and immobilization. The treatment of acute lymphoblastic leukemia (ALL) includes steroids and asparaginase (ASP), raising the VTE risk. In children with ALL the central nervous system (CNS) is a common localization for VTE. However, retrospective studies on small numbers of patients, larger studies and population-based data in children are scarce. The five Nordic countries, Estonia and Lithuania have a common treatment protocol for children with ALL between 1 and 18 years of age with prospective registration of toxicities, including CSVT offering a unique opportunity to study CSVT in this patient group. This is to our knowledge the largest report of children with ALL and CSVT describing the incidence, symptoms, treatment and the effect of CSVT on ALL treatment.

Methods

We assessed the symptoms, treatment, clinical risk factors and outcome of all children between ages 1 and 17 years at diagnosis of B-cell precursor or T-cell ALL between June 2008 and July 2013 and with CSVT. Data were collected from the patients’ medical records and the NOPHO leukemia registry.

Results

In total, 20 (1.9%) of the 1038 children with ALL treated according to the NOPHO ALL 2008 protocol developed CSVT. The cumulative incidence of CSVT was 2.0%. All the thromboses occurred within the first 5 months of treatment. The most common symptoms at the diagnosis of CSVT were headache, convulsions, weakness/fatigue and cerebral nerve palsy/hemiparesis/hemiplegia. The most frequent localizations for CSVT were sinus sagittalis (n=16) and sinus transversus (n=10). However, in most cases multiple cerebral veins were involved ( 70%). Median D-dimer at time of the CSVT diagnosis was 0.85 mg/L (range 0.19-4.7 mg/L) with 5 patients having normal D-dimer. We could not identify any clinical risk factors for CSVTs. CSVT was associated with steroids (treatment within 2 weeks before the diagnosis of CSVT) in 16/20 and with Pegylated asparaginase in 16/20. Fifteen patients were later screened for the inherited thrombophilic factors; one child had heterozygous prothrombin G20110A mutation and another heterozygous factor V (R506Q) Leiden mutation.

Most patients (19/20) were treated with anticoagulants: mostly low molecular weight heparin (LMWH). The median treatment with LMWH was 26 weeks (range 14-119 weeks). No bleeding complications were observed in connection with LMWH.

Two deaths were directly related to CSVT. Asparaginase was omitted from the treatment in 7 and delayed or reduced in 5 of the cases raising the risk for subsequent suboptimal leukaemia treatment. Of the surviving 18 patients, follow-up imaging revealed complete recanalization in 7 and partial recanalization in 7 cases. No imaging was available for the remaining 4 patients.

Conclusions

The incidence of CSVT in children with ALL was approximately 2%. No statistically significant clinical predictors for CSVT were identified. The mortality related to CSVT was 10%. Anticoagulation with LMWH was the treatment of choice in most cased and was well tolerated.