Children with Down Syndrome Are High-Risk for Severe Respiratory Syncytial Virus Disease David R. Stagliano, MD1,2, Cade M. Nylund, MD2, Matilda B. Eide, MPH2, and Matthew D. Eberly, MD2 Objective To assess Down syndrome as an independent risk factor for respiratory syncytial virus (RSV) hospitalization in children younger than 3 years of age and to evaluate illness severity.

Study design A retrospective cohort study of children enrolled in the military health system database was conducted. The effect of Down syndrome on RSV hospitalization was assessed by Cox proportional hazards model, while we controlled for risk factors. Disease severity was assessed by length of hospital stay, need for respiratory support, and age at hospitalization. Results The study included 633 200 children and 3 209 378 person-years. Children with Down syndrome had a hospitalization rate of 9.6% vs 2.8% in children without Down syndrome. Down syndrome had a greater adjusted hazard ratio (HR) for RSV hospitalization than most risk factors, 3.46 (95% CI 2.75-4.37). A sensitivity analysis demonstrated HR 3.21 (95% CI 2.51-4.10) for patients with Down syndrome ages 0-23 months and HR 5.07 (95% CI 2.21-11.59) ages 24-36 months. The median (IQR) length of stay of children with and without Down syndrome was 4 days (2-7) and 2 days (1-4) (P < .001). Patients with Down syndrome had a greater risk of requiring respiratory support (relative risk 5.5; 95% CI, 2.5-12.3). The median (IQR) ages at admission for children with and without Down syndrome were 9.8 months (5.5-17.7) and 3.5 months (1.7-8.7) (P < .001). Conclusions Down syndrome is independently associated with an increased risk for RSV hospitalization. Children with Down syndrome are older at time of RSV hospitalization and have more severe RSV illness than children without Down syndrome. This increased risk for hospitalization continues beyond 24 months. (J Pediatr 2015;166:703-9).

R

espiratory syncytial virus (RSV) is a leading cause of lower respiratory tract infection and bronchiolitis in children, infecting nearly all children by age 2.1,2 An estimated 1%-2% of healthy infants require hospitalization for bronchiolitis, resulting in an estimated 75 000 to 125 000 RSV-related admissions annually in the US.3,4 Recent studies demonstrate the average annual rates of RSV hospitalizations are 17 per 1000 for infants younger than 6 months of age and 5.2 per 1000 for children younger than 24 months old.2,4,5 A humanized monoclonal antibody, palivizumab, is used for prophylaxis against RSV and reduces hospitalizations in highrisk children younger than 2 years of age.6,7 Children at increased risk for severe RSV disease include infants with chronic lung disease (CLD), hemodynamically significant heart disease (HSHD), neuromuscular disease, immunodeficiency, and/or premature birth.8-12 Currently, the American Academy of Pediatrics does not recommend the routine use of palivizumab in patients with Down syndrome who do not qualify for other reasons.12,13 Down syndrome is the most common identifiable chromosomal abnormality in children. After maternal age and race is adjusted for, the national prevalence of Down syndrome is 1 in 691 live births, with approximately 6000 new cases annually.14 Advanced maternal age remains the leading risk factor for Down syndrome, and although social factors may influence decisions about termination in certain populations, the overall live birth prevalence appears to be stable across varying surveillance systems even when outcomes include terminations.14-16 Various comorbidities and chronic disease, such as the presence of HSHD, may predispose children with Down syndrome to severe RSV lower respiratory tract infection and require hospitalization for RSV.17

CAA CF CHD CLD HR HSHD ICD-9-CM IDRSV MHS RR RSV

Congenital airway anomaly Cystic fibrosis Congenital heart disease Chronic lung disease Hazard ratio Hemodynamically significant heart disease International Classification of Diseases, 9th revision, Clinical Modification Incidence density for respiratory syncytial virus hospitalization Military health system Relative risk Respiratory syncytial virus

From the 1Department of Pediatrics, Walter Reed National Military Medical Center; and 2Department of Pediatrics, Uniformed Services University of the Health Sciences, Bethesda, MD The views expressed in this article are those of the authors and do not reflect the official policy or position of the United States Army, United States Air Force, the Department of Defense, or the U.S. Government. The authors declare no conflicts of interest. Portions of the study were presented at the Pediatric Academic Societies’ meeting, May 4-7, 2013, Washington, DC. 0022-3476/$ - see front matter. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jpeds.2014.11.058

703

THE JOURNAL OF PEDIATRICS



www.jpeds.com

A previous study found the rate of hospitalization for RSV lower respiratory tract infection to be greater in term children with Down syndrome without congenital heart disease (CHD; 7.6%, 9 of 119) than in healthy term control children (0.7%, 2 of 276).18 Further studies, however, are warranted to validate these findings and evaluate Down syndrome in relation to other risk factors. Our objective was to assess Down syndrome as an independent risk factor for RSV hospitalization for children younger than 3 years of age by using military dependents as a demographically and socioeconomically diverse study population.19 We hypothesized that children with Down syndrome have an increased risk for RSV hospitalization and have a more severe disease course than children without Down syndrome.

Methods We conducted a retrospective cohort database study of children born between October 1, 2005 and April 30, 2011 within TRICARE, and with minimum continued enrollment beyond 90 days of life (Figure 1; available at www.jpeds. com). The study was approved by our Institutional Review Board and the Human Research Protection Office for the Office of the Assistant Secretary of Defense for Health Affairs/Defense Health Agency. All patient data were deidentified. TRICARE is the Department of Defense’s health care program for members of the uniformed services and their families. The military health system (MHS) database records TRICARE billing data for outpatient and inpatient encounters of beneficiaries from both civilian and military treatment facilities. These beneficiaries include children within a broad spectrum of socioeconomic classes whose parents are stationed at military locations throughout the US.19 We queried the MHS database for all children born between October 1, 2005 and April 30, 2011. We classified subjects as having Down syndrome, a hospitalization for RSV, and/or a risk factor for RSV hospitalization by any primary or nonprimary diagnosis by using their International Classification of Diseases, 9th revision, Clinical Modification (ICD-9CM) discharge diagnostic codes (Table I; available at www. jpeds.com). Patients discharged with RSV lower respiratory tract infection were identified by the ICD-9-CM codes 079.6 (RSV infection), 466.11 (RSV bronchiolitis), or 480.1 (RSV pneumonia). Patients discharged with Down syndrome were identified using 758.0 (Down syndrome). Table I summarizes the ICD-9-CM codes for the following risk factors: CLD, HSHD, other CHD, neuromuscular disease, immunodeficiency, cystic fibrosis (CF), congenital airway anomalies (CAAs), and premature birth less than 37 weeks’ and 0 days’ gestation. We subdivided the subjects with heart disease into 2 separate groups. HSHD was defined only by subjects with pulmonary hypertension (ICD-9-CM codes 416 and 417) and heart failure (ICD-9CM code 428). CHD of unknown significance, or other CHD, was defined by patients with CHD who could 704

Vol. 166, No. 3 potentially be hemodynamically significant but who did not have a diagnosis of pulmonary hypertension or heart failure. For other CHD, we also excluded subcodes 746.86 (Congenital heart block), 747.0 (Patent ductus arteriosus), 747.5 (Absence or hypoplasia of umbilical artery), 747.6747.69 (Other anomalies of peripheral vascular system), and 747.8-747.89 (Other specified anomalies of circulatory system) from the study. These subcodes represent congenital heart conditions that are unlikely to cause any long-lasting, HSHD. We subdivided the subjects with pulmonary diagnoses into the following 3 separate groups: CLD, CF, and CAA. Additional data obtained for analysis included date of hospital admission, age at admission, sex, length of stay, and type of treatment facility (military or civilian). Procedure codes identifying the use of an airway adjunct, need for respiratory intubation, or mechanical ventilation are summarized in Table I, along with diagnostic, procedure, and drug codes for palivizumab use. Patients who received palivizumab as either outpatients or inpatients were excluded from the study. The control group included all of the patients without Down syndrome within the MHS database who were eligible for care, enrolled for more than 90 days of life, who were within the same age range (between 0 and 36 months), and during the same time frame (October 1, 2005 to April 30, 2011) as the children with Down syndrome. An event was defined as the first hospitalization of an individual patient with any discharge diagnosis of RSV lower respiratory tract infection. Microbiologic data were unavailable for diagnostic confirmation of RSV disease. Respiratory support was defined by the use of an airway adjunct, respiratory intubation, and/or mechanical ventilation. Each study year was defined as July of the initial year through June of the following year. Study year one (October 2005 to June 2006) and study year 6 (July 1, 2010 to April 30, 2011) were exceptions because the data obtained from the MHS database started in October 2005 and ended in April 2011. Subjects were censored at either health care disenrollment or 36 months of age, whichever occurred earlier. Statistical Analyses Absolute hospitalization rates were calculated by dividing the number of patients admitted for RSV lower respiratory tract infection by the total number of patients in each group. The incidence density for RSV hospitalization (IDRSV) was calculated by dividing the number of patients hospitalized for RSV by the person-years at risk. The incidence density rate ratios including 95% CI were calculated by Poisson regression. As non-normally distributed variables, length of stay and age were expressed as median days and months, respectively, with IQRs. Mann-Whitney U test was used to compare length of stay and age at the time of RSV hospitalization for the presence vs absence of Down syndrome, as well as the time in the study and time to disenrollment. The c2 test and relative risk (RR) with 95% CI were used to compare categorical variables. Stagliano et al

ORIGINAL ARTICLES

March 2015 Single event analysis was performed using Cox proportional hazards regression including Down syndrome, the other risk factors for RSV hospitalization, and sex as independent variables. All of these variables were entered into the analysis as binary time-fixed variables. The final model included all of the individual RSV study years as timedependent, stratification variables. For purposes of model adjustment and optimization, 2-way interaction terms were included in the model if significant. The unadjusted and adjusted hazard ratios (HRs) were reported with 95% CI. A subgroup analysis was performed to evaluate the incidence, the age at diagnosis, the length of stay, and the need for respiratory support in children with Down syndrome without additional risk factors, because a majority of patients with Down syndrome had concomitant risk factors.18 Patients with Down syndrome without CLD, HSHD, other CHD, neuromuscular disease, immunodeficiency, CF, CAA, and premature birth were compared with healthy children. Healthy children, the control group for the subanalysis, were defined as children without Down syndrome and without any of these other studied risk factors (Figure 1). For visualization purposes, the estimated percent survival without RSV hospitalization was plotted by the presence or absence of Down syndrome with and without the studied risk factors vs time in months (Figure 2).20 A sensitivity analysis also was performed to confirm the increased risk of RSV hospitalization in children with

Down syndrome at and beyond 2 years of age. We compared 2 Cox proportional hazards regression models that were generated by using the same methods as the overall study model; one model included subjects for ages 0-23 months and the other model included subjects for ages 24-36 months. The adjusted HRs were reported with 95% CIs. An alpha of 0.05 was used to determine significance. All analyses were performed using SAS 9.3 (SAS Institute, Cary, North Carolina).

Results A total of 633 200 children younger than 3 years of age and 3 209 378 person-years were studied. With the exception of male sex, the percentages of the individual known risk factors, ie, CLD, HSHD, etc., for RSV hospitalization were different between the cohort of patients with Down syndrome and the cohort of patients without Down syndrome (Table II). In addition, children with Down syndrome were more likely to have one or more concomitant risk factors than the patients without Down syndrome (P < .001). The time in the study and the time to disenrollment for the patients with Down syndrome and without Down syndrome, however, were no different (P = .11). Altogether, 9048 patients (1.4%) were hospitalized with a discharge diagnosis of RSV lower respiratory tract infection

Figure 2. The estimated percent without RSV hospitalization (percent survival) stratified by the presence or absence of Down syndrome and any known risk factors. The estimated percent survival curves were generated from a multivariable Cox proportional hazards model. Risk groups: DS RF, patients without Down syndrome and without risk factors for RSV lower respiratory tract infection hospitalization; DS +RF, patients without Down syndrome with risk factors for RSV lower respiratory tract infection hospitalization; +DS RF, patients with Down syndrome and without risk factors for RSV lower respiratory tract infection hospitalization; +DS +RF, patients with Down syndrome with risk factors for RSV lower respiratory tract infection hospitalization. DS, Down syndrome; RF, risk factor. Children with Down Syndrome Are High-Risk for Severe Respiratory Syncytial Virus Disease

705

THE JOURNAL OF PEDIATRICS



www.jpeds.com

Table II. Descriptive characteristics of the patient cohorts*

Characteristics

Patients with Down syndrome (n = 842)

Patients without Down syndrome (n = 632 358)

P value†

Male HSHD CHD, other Prematurity CLD CF CAA Immunodeficiency Neuromuscular disease No risk factors One risk factor $2 risk factors

418 (49.6) 97 (23.4) 422 (50.1) 239 (28.4) 18 (2.1) 2 (0.2) 75 (0.1) 15 (1.8) 18 (2.1) 290 (34.4) 314 (37.3) 238 (28.3)

322 649 (51.0) 1008 (0.2) 13 297 (2.1) 42 683 (6.7) 538 (0.1) 194 (

Children with Down syndrome are high-risk for severe respiratory syncytial virus disease.

To assess Down syndrome as an independent risk factor for respiratory syncytial virus (RSV) hospitalization in children younger than 3 years of age an...
460KB Sizes 0 Downloads 7 Views