A Retrospective Review of Hospital-Acquired Venous Thromboembolism at a Large Pediatric Tertiary Care Center

Introduction:

The incidence of venous thromboembolism (VTE) in hospitalized pediatric patients is increasing secondary to the growing medical complexity of pediatric patients and the increasing use of central venous catheters. Pediatric patients diagnosed with VTE have up to 2% mortality associated directly with their thromboses. While incidence, risk factor identification and preventive strategies are well established in hospitalized adults, this information is limited in the pediatric population. There are currently no standardized VTE risk screening tools or thromboprophylaxis guidelines for children at Duke Children's Hospital. The incidence of hospital acquired VTE (HA-VTE), as well as their associated risk factors were investigated in a retrospective review.

Methods:

Medical records of pediatric patients hospitalized at Duke Children's Hospital during June 2018 through November 2018 were reviewed. The EPIC SlicerDicer tool was used to identify patients with ICD-10 diagnoses codes related to thrombosis or treated with anticoagulants. Included patients were diagnosed with HA-VTE during their hospitalization or within 14 days of discharge. Data collected included demographics, thrombosis characteristics, family history, mobility, and acute or chronic co-morbid conditions. The characteristics of the study population were described by median (with 25th and 75th percentiles) for continuous variables and frequencies (with percentages) for binary or categorical variables.

Results:

Out of 4,176 total pediatric admissions to all units of Duke Children's Hospital (ages 0-18.99 years) during the inclusion timeframe, 33 VTE events were identified. The incidence of VTE events per 1000 patient days was 0.98. The complete patient and VTE event characteristics are listed in Tables 1 and 2. The median age of patients with VTE events was 0.4 years. Of the identified cohort, 73% had an associated central venous line (CVL). Neonates with congenital cardiac disease comprised the majority of the cohort. Other common patient characteristics observed in this cohort included impaired mobility, recent major surgery, and recent mechanical ventilation. Of the 33 VTE diagnoses, 70% received therapeutic anticoagulation with enoxaparin or unfractionated heparin. Only 2 patients (8%) received prophylactic anticoagulation prior to their diagnosis of VTE.

Conclusions:

The retrospective review of HA-VTE events at Duke Children's Hospital identified that the majority of the events occurred in neonates with congenital cardiac disease and the presence of CVLs. It was also noted that there was no standardization among the use of anticoagulation agents that were initiated for treatment of VTE. Furthermore, few patients received VTE prophylaxis during the hospitalization. A limitation of this review was that it was retrospective and the documentation of family history of thrombosis was inconsistent. It is also possible that several VTE events were missed due to inadequate ICD-10 coding. Based on the results of this review, there is a need to implement a risk stratification tool and develop standardized recommendations of VTE prophylaxis and treatments for pediatric patients admitted to Duke Children's Hospital. There is an additional quality improvement phase of this project and the goal is to implement a risk calculator that is based on information learned from the retrospective review. Ultimately, this risk calculator will help to decrease the incidence of VTE events at Duke Children's Hospital.

View largeDownload slide

View largeDownload slide

Close modal