Abstract

Introduction: Cancer is a major risk factor in children with TE. However, information regarding epidemiology of TE in children with cancer is scant. We conducted a retrospective cohort study to define the epidemiology of TE in children with cancer and to identify potential risk factors.

Methods: Records of children (≤18 years of age) with cancer diagnosed and treated at McMaster Children’s Hospital over past 15 years were reviewed for demographics, details of diagnosis and treatment of cancer and of TE, if any. We studied the effect of age (<10 years vs. ≥10 years), gender, type of cancer and presence or absence of intrathoracic disease (defined as mediastinal mass or any primary or metastatic pulmonary disease), type of central venous line (CVL) and CVL dysfunction (defined as persistent or recurrent difficulty of blood draw and/or infusion or documented CVL infection) on the risk of developing TE. Statistical analysis was performed using SPSS version 15.

Results: Overall 49 of 606 children (8.1%) with cancer developed TE (Table 1). Due to very low prevalence of TE in children with brain tumors, regression analyses for risk factors was performed in children with non-CNS cancers. Children with ALL (OR 4.93, 95% CI 1.60, 11.52, p=0.004), lymphoma (OR 4.18, 95% CI 1.37, 12.71, p=0.01), and sarcoma (OR 4.42, 95% CI 1.42, 13.77, p=0.01) had increased risk of TE. Older patients (age ≥ 10 years) were at higher risk of developing TE compared to younger patients (OR 2.2; 95% CI 1.2,3.96; P<0.01). Subgroup analyses showed that patients with CVL-dysfunction (33.3% vs. 9.5%; p<0.0001, 95% CI; 10.1,37.5) and those with intrathoracic disease (17.6% vs.5.7%; p=0.028, 95% CI; 2.2, 21.7) were at significantly higher risk of TE compared to those without CVL dysfunction and intrathoracic disease.

Conclusions: Overall TE is common in children with cancer. We have identified older age and type of cancer are the important risk factors predisposing to TE; children 10 years or older and those with lymphoreticular malignancy and sarcoma are at significantly higher risk of developing TE. In addition presence of mediastinal disease and CVL dysfunction increased the risk of TE. This is one of the largest comprehensive epidemiogical studies of TE in children with cancer identifying different risk factors.

Prevalence of TE according to the tyep of cancer

Type of CancerNPatients with TE% with TE (95% CI)
ALL 185 26 14.5 (9.39, 19.91) 
Brain Tumors 131 0.8 (0.02, 4.20) 
Lymphoma 72 12.5 (5.88, 22.41) 
Sarcoma 70 10 14.3 (7.07, 24.71) 
AML 51 5.9 (1.23, 16.24) 
Neuroblastoma 45 2.2 (0.05, 11.77) 
Wilms’ Tumor 43 1.3 (0.06, 12.3) 
Other 
Total 606 49 8.1 (6.1, 10.5) 
Type of CancerNPatients with TE% with TE (95% CI)
ALL 185 26 14.5 (9.39, 19.91) 
Brain Tumors 131 0.8 (0.02, 4.20) 
Lymphoma 72 12.5 (5.88, 22.41) 
Sarcoma 70 10 14.3 (7.07, 24.71) 
AML 51 5.9 (1.23, 16.24) 
Neuroblastoma 45 2.2 (0.05, 11.77) 
Wilms’ Tumor 43 1.3 (0.06, 12.3) 
Other 
Total 606 49 8.1 (6.1, 10.5) 

Author notes

Disclosure: No relevant conflicts of interest to declare.