Asymptomatic thrombosis following the use of central venous catheters in children

Abstract

Central venous catheters (CVCs) are the major cause of thrombosis in acutely unwell neonates and children. This thesis reports the outcomes of a prospective study of 189 children in a paediatric intensive care unit (PICU) that aimed to determine the frequency of asymptomatic CVC-related thrombosis during hospital admission, and the incidence of residual CVC-related thrombosis and clinically significant post thrombotic syndrome (PTS) two years following CVC placement. The study also sought to identify risk factors predictive of CVC-related thrombosis and clinically significant PTS and determine the utility of ultrasound screening for asymptomatic CVC-related thrombosis in acutely unwell children. This study is distinct from previous work as children in this cohort identified to have asymptomatic CVC-related thrombosis were not treated and were followed for two years to determine the natural history of asymptomatic thrombosis.

The incidence of asymptomatic CVC-related thrombus during hospital admission was 21.2% (n=31), despite over 80% of the cohort having received thromboprophylaxis. One child had a symptomatic CVC-related thrombosis. Sixteen children had new or residual thrombus identified on the follow-up ultrasound, performed a mean of 26 months post CVC placement. Of 126 PTS assessments completed at follow-up, 10.3% of children were diagnosed with mild PTS as classified by both the Manco-Johnson instrument and Modified Villalta scale. Only two children were identified to have clinically significant PTS (Manco-Johnson instrument only), one of whom had a symptomatic CVC-related thrombosis.

Whilst over one fifth of the cohort had asymptomatic CVC-related thrombosis, only one of these children developed clinically significant PTS. There were no clinical embolic phenomenon or deaths attributable to CVC-related thrombosis. These findings suggest that ultrasound screening for asymptomatic CVC-related thrombosis is not indicated in children in PICU as, despite no treatment (in all but one child), the incidence of any thrombosis-associated morbidity was extremely low. In addition, the study’s findings support previous evidence that thromboprophylaxis provides no protection against CVC-related thrombosis in children.

Neither D-dimer nor Factor VIII were predictive of CVC-related thrombosis. However, this study identified that children with femoral CVCs had a significantly higher incidence of residual thrombosis and thrombosis-associated morbidity. Children with femoral CVCs were also significantly more likely to have a higher PIM2 probability of death, cardiac arrest, hypotension, elevated D-dimer and Factor VIII compared to children with jugular CVCs. Whether the placement of the CVC in the femoral veins or the apparent higher acuity of this sub-group of patients is the mediator of increased risk (residual thrombosis and morbidity) compared to jugular CVC placement remains unclear.

This study contributes novel insights into the natural history of asymptomatic CVC-related thrombosis, demonstrating a very low morbidity associated with jugular CVC placement. This study informs future clinical research targeting children most at risk of long-term CVC-related thrombosis associated morbidity. Specifically, further investigation of similar cohorts of children to those in whom femoral CVC placement occurred in this study is needed to elucidate risk factors for poor outcomes and the potential for a risk stratified approach to the treatment of asymptomatic CVC related thrombosis.