Venous Thromboembolism in Pediatric Heart Disease Is Associated with Significant Health Care Burden and Expenditures

Background

Most pediatric venous thromboembolism (VTE) occur in children with at least one complex chronic condition (CCC) admitted to a tertiary care pediatric hospital (Setty et al, Pediatr Blood Cancer2012). Reports estimate 14.6%-29% of all inpatient childhood VTE occur in patients with chronic pediatric heart disease (CPHD) (Andrew et al, Blood 1994; Rafinni et al, Pediatrics 2009). The influence of acute VTE risk factors and the impact of non-cardiac CCCs on VTE in CPHD remain unknown. A large cohort of privately-insured children was used to evaluate the odds of VTE, estimate the healthcare burden, and delineate the risk factors associated with VTE diagnoses in CPHD.

Methods

Children <18 years of age between April 1, 2003 and June 30, 2013 at the time of their initial CPHD diagnostic code were identified from the MarketScan Commercial Databases®. CPHD was defined using ICD-9-CM (International Classification of Diseases, 9^th Revision, Clinical Modification) codes for cardiomyopathies, single ventricle physiology congenital heart defects, double ventricle physiology congenital heart defects, conduction disorders and dysrhythmias, and acquired chronic cardiac conditions. Children with CPHD were categorized into the VTE group if an ICD-9-CM for VTE or Current Procedural Terminology (CPT) code for a VTE-associated procedure was found concomitantly with their first CPHD diagnosis or during a 6 month follow-up period. The demographics, mortality, healthcare resource utilization, expenditures, and co-morbid conditions of children with CPHD with and without a VTE diagnosis were compared.

Results

There were 120,884 children 0-17 years of age with a CPHD diagnosis. VTE events occurred in 957 (0.79%) of these children, 802 (83.8%) of which were isolated DVT and 155 (16.2%) were PE with or without concomitant DVT. Male gender was significantly associated with VTE (P<0.001). In-hospital mortality was considerably higher in children with VTE (10.8%) than in those without VTE (1.3%; P<0.001).

All CPHD were associated with a significant VTE risk except those with conduction disorders or dysrhythmias (Table 1). VTE occurred in 2.3% of children with single ventricle physiology, 1.9% of those with acquired CPHD, 1.6% with cardiomyopathy, and 0.5% with double ventricle physiology. Children who experienced VTE were significantly (P<0.001) more likely to have a non-cardiac CCC, acute acquired VTE risk factors (recent trauma, recent non-cardiac surgery, recent cardiac surgery, or an acute acquired cardiovascular condition), recent extracorporeal membranous oxygenation, and heart transplantation (Table 1). Acute acquired VTE risk factors did not appear to affect the VTE risk when controlled for non-cardiac CCC. Stratifying for recent cardiac surgery, (a) all of the non-cardiac CCCs were significantly associated with VTE without a recent cardiac surgery (P<0.001) and (b) all non-cardiac CCCs except for gastrointestinal CCCs and malignancy CCCs were significantly (P≤0.02) associated with VTE in the setting of recent cardiac surgery.

All measures of healthcare utilization (mean number of inpatient admissions (2.19 vs 0.62), mean inpatient length of stays (55.14 vs 8.7 days), mean number of outpatient visits (28.6 vs 11.22), and mean number of pharmaceutical claims (13.27vs 4.74)) were significantly higher (P<0.001) for CPHD patients with VTE compared to those without VTE. VTE diagnosis was associated with significantly higher medical expenditures: a 9-fold increase in total expenditures compared to the non-VTE group ($344,807 vs $37,102; P<0.001); a 10-fold increase in inpatient expenditures ($316,379 vs $29,678; P<0.001); over 3-fold increase in outpatient expenditures ($25,866 vs $6,838; P<0.001); over a 4-fold increase in pharmaceutical claims ($2,561 vs $585; P<0.001).

Conclusions

Although VTE in CPHD patients is rare, it does carry a significantly increased healthcare resource utilization and in-hospital mortality. All of the co-morbid conditions examined were significantly associated with VTE, but when controlling for non-cardiac CCC the odds of VTE was not affected. All of the non-cardiac CCC were significantly associated with VTE except malignancy and GI CCC when accounting for recent cardiac surgery. This suggests that non-cardiac CCCs may be an important determinant of VTE risk for children with CPHD.

View largeDownload slide

View largeDownload slide

Close modal