Feb 21, 2014 - J Neurosurg: Pediatrics / Volume 13 / April 2014. ©AANS, 2014 ... nearly 7400 deaths, 60,000 hospitalizations, and 600,000 emergency ...
J Neurosurg Pediatrics 13:448–455, 2014 ©AANS, 2014
Venous thromboembolism in the setting of pediatric traumatic brain injury Clinical article Dominic A. Harris, B.A., and Sandi Lam, M.D., M.B.A. Department of Neurosurgery, Baylor College of Medicine, Houston, Texas Object. The risk of venous thromboembolism (VTE) in children with traumatic brain injury (TBI) has not been well characterized given its rarity in the pediatric population. Investigation of risk factors for VTE in this group requires the use of a large sample size. Using nationally representative hospital discharge data for 2009, the authors of this study characterize the incidence and risk factors for VTE in children hospitalized for TBI. Methods. The authors conducted a cross-sectional study using data from the Healthcare Cost and Utilization Project Kids’ Inpatient Database to examine VTE in TBI-associated hospitalizations for patients 20 years of age or younger during the year 2009. Results. There were 58,529 children with TBI-related admissions, including 267 with VTE diagnoses. Venous thromboembolisms occurred in 4.6 per 1000 TBI-associated hospitalizations compared with 1.2 per 1000 pediatric hospitalizations overall. By adjusted logistic regression, patients significantly more likely to be diagnosed with VTE had the following: older age of 15–20 years (adjusted odds ratio [aOR] 3.7, 95% CI 1.8–8.0), venous catheterization (aOR 3.0, 95% CI 2.0–4.6), mechanical ventilation (aOR 1.9, 95% CI 1.2–2.9), tracheostomy (aOR 2.3, 95% CI 1.3–4.0), nonaccidental trauma (aOR 2.8, 95% CI 1.1–6.9), increased length of stay (aOR 1.02, 95% CI 1.01–1.03), orthopedic surgery (aOR 2.4, 95% CI 1.8–3.4), and cranial surgery (aOR 1.8, 95% CI 1.1–2.8). Conclusions. Using the Kids’ Inpatient Database, the authors found that risk factors for VTE in the setting of TBI in the pediatric population include older age, venous catheterization, nonaccidental trauma, increased length of hospital stay, orthopedic surgery, and cranial surgery. (http://thejns.org/doi/abs/10.3171/2014.1.PEDS13479)
Key Words • traumatic brain injury • deep vein thrombosis • trauma • venous thromboembolism • pediatric • children
T
brain injury (TBI) is a major health problem among children and a leading cause of morbidity and mortality in the United States.23 With nearly 7400 deaths, 60,000 hospitalizations, and 600,000 emergency department visits annually, TBIs in children impose a significant burden on our health care system.25 In the adult population, a number of studies have found major trauma to be an independent risk factor for the development of venous thromboembolism (VTE) and pulmonary embolism.5,20,27 Several factors may contribute to this, including extended immobilization, inadequate VTE prophylaxis, or trauma-induced coagulopathy. Moreover, raumatic
Abbreviations used in this paper: aOR = adjusted odds ratio; HCUP = Healthcare Cost and Utilization Project; ICD-9-CM = International Classification of Diseases, Ninth Revision, Clinical Modification; ICU = intensive care unit; KID = Kids’ Inpatient Database; OR = odds ratio; TBI = traumatic brain injury; VTE = venous thromboembolism.
448
patients with TBIs are at high risk for coagulopathy and VTE formation.9,17 It is postulated that changes in the coagulation cascade are due to widespread release of tissue factor after a cerebral insult.9,17 Venous thromboembolism is a rare complication in children, with an incidence roughly one-tenth of that in adults.20,27 Its study thus requires the use of a large sample size. Risk factors for VTE in the pediatric population have been studied predominantly in single-institution settings. More recently, Vu et al.27 were able to use the Healthcare Cost and Utilization Project (HCUP) Kids’ Inpatient Database (KID) for the years 1997, 2000, and 2003 to estimate the incidence and risk factors for VTE in the overall pediatric population. Their analysis found older age (15–17 years); comorbid conditions of obesity, inflammatory bowel disease, and malignancy; and surgery (thoracoabdominal or orthopedic) to be significant risk factors for VTE. Interestingly, trauma was not found to be a significant risk factor despite trauma’s association with VTE in the literature.3,7,15,20 J Neurosurg: Pediatrics / Volume 13 / April 2014
Venous thromboembolism and pediatric TBI Using nationally representative data from the KID for 2009 for a cross-sectional study, we characterize the incidence and risk factors for VTE for children hospitalized for TBI. To our knowledge, this is the first study to assess the risk of VTE specifically in children with TBI.
Methods Data Source
Data were obtained from the 2009 Kids’ Inpatient Database (KID), one of a family of administrative databases developed by the Healthcare Cost and Utilization Project (HCUP) sponsored by the Agency for Healthcare Research and Quality.11 The 2009 KID contains discharge-level data from over 3.4 million pediatric hospitalizations from 4121 nonfederal community hospitals in 44 states.11 The KID allows for more precise national and regional estimates for pediatric conditions, clinical outcomes, and hospital services, which are otherwise difficult to analyze given that children make up a relatively small proportion of hospital stays. The KID contains randomly selected pediatric discharges stratified by non–birth related discharges and complicated or uncomplicated births.11 To obtain national estimates, patient records were provided with weights using the American Hospital Association universe of nonfederal community hospitals as the standard. Hospital data were stratified by the following 6 characteristics: ownership/control, bed size, teaching status, rural/urban location, US region, and status as a freestanding children’s hospital.11
Patient Selection
We selected records associated with TBI for patients up to 20 years of age or younger. These records were identified using discharge diagnosis codes defined by the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). The following ICD9-CM codes representing TBI were used: fracture of the vault or base of the skull (800.0–801.9); other unqualified and multiple fractures of the skull (803.0–804.9); intracranial injury, including concussion, contusion, laceration, and hemorrhage (850.0–854.1); injury to the optic chiasm, optic pathways, or visual cortex (950.1–950.3); unspecified head injury (959.01); and shaken infant syndrome (995.55).23,24 Discharge records associated with venous thromboembolism (VTE) were identified using the ICD-9-CM codes: deep venous thrombosis (453.2, 453.40, 453.41, 453.42, 453.82–89, and 453.9), thrombophlebitis (451.11, 451.19, 451.2, 451.81, 451.83, 451.84, 451.89, and 451.9) of the upper and lower extremities, and pulmonary embolism (415.11, 415.13, and 415.19).22,26,28
Patient Characteristics
Patient characteristics obtained from the database include age (in years), sex, race (white, black, Hispanic, Asian or Pacific Islander, Native American, and other), hospital death, length of hospitalization (days), and primary payer (Medicaid, private, self-pay, or other).
J Neurosurg: Pediatrics / Volume 13 / April 2014
Nonaccidental traumatic injuries were identified with the following ICD-9-CM codes: child abuse (995.50, 995.54, 995.55, 995.59) and E-codes (external causes) for inflicted injury (E-960–969). We created variables for surgical procedures (cranial, thoracic or abdominal, orthopedic, and other) using the following ICD-9-CM procedure codes: cranial surgery (01xx–02xx), orthopedic surgery (77xx–84xx), and thoracoabdominal surgery including surgeries related to the gastrointestinal tract (42xx–54xx), cardiovascular excluding procedures on vessels (35xx–37xx), and pulmonary (32xx–34xx). All remaining procedure codes for surgeries were included in the “other” category. Procedure codes not associated with surgeries were included in the “none” category. We used ICD-9-CM procedure codes to create variables for the following inpatient procedures: venous catheterization (38.93 and 38.97), intubation and mechanical ventilation (96.04, 96.05, and 96.7×), tracheostomy (31.1, 31.21, and 31.29), and gastrostomy tube insertion (43.2, 43.11, 43.19, 43.2, and 44.32).2,28 Statistical Analysis
Descriptive statistics with weighted national estimates were conducted to evaluate the distribution of patient and hospital characteristics for those with or without VTE complications during their TBI-associated hospitalization. As KID is a sampled database, our results are reported as estimated values such as means and frequencies with 95% confidence intervals. These estimates represent national estimates for the associated year. Because of the sampling methodology, national medians cannot be obtained. Hence continuous data are expressed as estimated means. Standard errors were adjusted for stratification and clustering of the KID sampling design as described in the 2009 KID documentation published by the Agency for Healthcare Research and Quality.11 It is important to note that because the data in KID is a composite of deidentified state-level information, our unit of analysis is a patient discharge and not a patient. In order to evaluate the effect of patient and hospital characteristics on the incidence of VTE in this group, univariate and multivariate logistic regression were used to calculate unadjusted and adjusted odds ratios and 95% confidence intervals. Twosided tests were used, with p values < 0.05 considered to be statistically significant. All statistical analyses were performed with Stata 12 (StataCorp).
Results
Of the 3,407,146 encounters from January 1, 2009, through December 31, 2009, 40,143 cases involved TBI. Among children up to 20 years of age, there were an estimated 58,529 hospitalizations in the US that were associated with TBI, or 7.9 per 1000 hospital discharges (95% CI 7.4–8.5). Among this subpopulation, an estimated 267 hospitalizations were associated with VTE, or 4.5 per 1000 TBI-associated hospitalizations (95% CI 3.8–5.3). The overall incidence of VTE in the pediatric population is very low, with an estimated prevalence of 1.2 cases per 1000 pediatric hospitalizations in 2009 (95% CI 1.1–1.3). Patient demographic data are summarized in Table 1. 449
D. A. Harris and S. Lam TABLE 1: Year 2009 national estimates of TBI-associated hospitalizations and VTE* Variable total age
