Down Syndrome: A Novel Risk Factor for Respiratory Syncytial Virus Bronchiolitis— A Prospective Birth-Cohort Study
OBJECTIVES. Respiratory syncytial virus is the single-most important cause of lower respiratory tract infections in children. Preterm birth and congenital heart disease are known risk factors for severe respiratory syncytial virus infections. Although Down syndrome is associated with a high risk of respiratory tract infections, little is known about the incidence of respiratory syncytial virus infections in this group. The aim of our study was to determine the incidence of respiratory syncytial virus lower respiratory tract infection–associated hospitalization among children with Down syndrome.
PATIENTS AND METHODS. We performed a retrospective observational study and a prospective nationwide birth-cohort study of children with Down syndrome. The retrospective cohort comprised 176 children with Down syndrome. A birth cohort of 219 children with Down syndrome was prospectively followed until 2 years of age. All 276 siblings of the birth cohort were used as controls.
RESULTS. Of the 395 patients with Down syndrome, 180 (45.6%) had a known risk factor for severe respiratory syncytial virus infections; 39 (9.9%) of these were hospitalized for respiratory syncytial virus lower respiratory tract infections. Two control children (0.7%) versus 9 term children with Down syndrome without congenital heart disease (7.6%) were hospitalized for respiratory syncytial virus lower respiratory tract infections. The median duration of hospitalization was 10 days; mechanical ventilation was required for 5 children (12.8%).
CONCLUSIONS. This is the first study, to our knowledge, to demonstrate that Down syndrome is a novel independent risk factor for severe respiratory syncytial virus lower respiratory tract infections. These findings should prompt studies to investigate possible mechanisms that underlie severe respiratory syncytial virus lower respiratory tract infections in children with Down syndrome. The effect of respiratory syncytial virus prophylaxis in this specific population needs to be established.
Respiratory syncytial virus (RSV) is the single-most important cause of lower respiratory tract infections (LRTIs) in infants and young children.1 Virtually all children are infected with RSV before the age of 2 years, and 40% of RSV infections progress to LRTI. Approximately 0.5% to 2% of children require hospitalization (and 2.2% of these require mechanical ventilation), which makes RSV LRTI the most common cause of hospitalization among infants during winter.2 Most hospitalized children are younger than 6 months. Premature birth, chronic lung disease, age <6 weeks, and congenital heart disease (CHD) are clinical risk factors for severe disease after RSV infection.3–5 In addition, a decreased lung function at birth is hypothesized to have an important role in the pathogenesis of RSV LRTI.6
Down syndrome (DS) is the most common chromosomal abnormality among live-born infants and is characterized by a variety of dysmorphic features and congenital malformations.7 It is associated with CHD, gastrointestinal disease, various immunologic impairments, and concomitant respiratory pathology.8–11 Because LRTI is the most common cause of acute hospitalization among children with DS,12 we hypothesized that children with DS are at increased risk of severe RSV LRTI and, hence, hospitalization. The primary aim of this study was to determine the incidence of RSV LRTI–associated hospitalization among children with DS with and without known risk factors.
PATIENTS AND METHODS
A retrospective observational study of RSV LRTI–associated hospitalization in a cohort of children with DS was performed. Subsequently, the results of this study were validated in a prospective nationwide birth cohort of children with DS.
Three groups were studied. First, a retrospective study was performed involving a cohort of children with DS who were being monitored by the Down Syndrome Study Group and who attended the outpatient clinic of the Pediatric Department of the VU University Medical Centre. The cohort comprised 206 children with DS, born between 1976 and 2005, with a median age of 6.5 years (range: 0.3–29.6 years). Patient charts were reviewed to determine the number of previous RSV LRTI–associated hospitalizations. The following information was obtained: gestational age, the presence of chronic lung disease of prematurity, congenital heart disease, cardiac surgery and postoperative hemodynamic status, and preexisting airway symptoms. The second study group consisted of a national birth cohort of 241 children with DS, born between 2003 and 2005, and followed until 2 years old. The cohort was prospectively collected by the Down Syndrome Study Group and TNO Quality of Life Leiden, under auspices of the Dutch Pediatric Surveillance Unit. This national registry has been established to facilitate research into the etiology, diagnostics, treatment, prognosis, and incidence of specific disorders and has coverage of 90% to 95%, depending on the disease. Children with DS are registered by their pediatrician. One of the investigators (Dr Bloemers) administered the children's primary caregiver a standardized questionnaire, by telephone, to acquire the same clinical information that was available for the first group. Although the primary aim of this study was to determine the absolute incidence of RSV LRTI–associated hospitalization in children with DS, a third, unmatched, control group was included. This control group was used to estimate the relative risk of RSV LRTI–associated hospitalization in children with DS. All 276 siblings born between 1976 and 2005 (median age: 5.6 years; range: 0–29.8 years) of the prospectively followed birth cohort were used as controls. Exclusion criteria for all 3 study groups were use of palivizumab (a humanized monoclonal antibody against RSV) and death not associated with RSV infection before the age of 2 years. Parental informed consent was obtained for the collection of all data.
RSV LRTI–Associated Hospitalization
RSV LRTI–associated hospitalization was defined as hospital admission for lower respiratory tract symptoms (deep or wet chest cough, wheezing, hoarseness, stridor, shortness of breath) and either a positive enzyme immunoassay for RSV, a positive direct immunofluorescence assay for RSV infection of epithelial cells in nasopharyngeal secretions or a positive viral culture for RSV. Age at time of diagnosis, total number of days in hospital, days in the ICU, days on mechanical ventilation, and days on supplemental oxygen were retrieved from hospital charts. RSV infection without hospitalization was not evaluated.
Risk Factors for RSV LRTI–Associated Hospitalization
Known clinical risk factors for RSV LRTI–associated hospitalization are hemodynamically significant CHD and prematurity (gestational age of <37 weeks) with or without chronic lung disease. Because DS is frequently associated with CHD, in the Netherlands all children with DS are routinely evaluated for CHD (and its hemodynamic relevance) by a pediatric cardiologist. For this study, the initial cardiologic evaluation of all children with CHD in the prospective group was reassessed by an independent cardiologist (Dr Strengers) who was unaware whether the children had developed RSV LRTI. Hemodynamically relevant CHD was defined as clinical signs and symptoms of left-right shunting or signs of volume overload because of left-right shunting on echocardiography. The results of the 2 evaluations were in almost perfect agreement. In the combined retrospective and prospective cohort, 63% of children with DS were found to have CHD, and in 35.7% the CHD was hemodynamically significant (Table 1). Of the children with hemodynamically relevant CHD, 18.7% had an atrioventricular septal defect (n = 74), 6.1% a ventricular septal defect (n = 24), 3.8% an atrial septum defect (ASD; n = 15), 2.8% a patent ductus arteriosus (n = 11), and 4.3% other hemodynamically relevant conditions (n = 17). Hemodynamically insignificant CHD, defined as the absence of clinical or echocardiographic signs of increased pulmonary flow, included a patent foramen ovale or ASD type II (9.4% [n = 37]), small to moderate ventricular septal defect (4.8% [n = 19]) and/or ASD (7.3% [n = 29]), small patent ductus arteriosus (3.8% [n = 15]), and others (2.0% [n = 8]). In 64.9% of children with atrioventricular septal defect, the defect was (partially) corrected before 6 months of age. Prematurity, chronic lung disease, significant CHD, and insignificant CHD were predefined risk factors.
The χ2 test was used to compare the proportion of children with RSV LRTI–associated hospitalization between different populations and to calculate odds ratios. Differences in continuous nonparametric variables between populations were assessed using Mann-Whitney U test. Differences in duration of hospitalization for RSV LRTI among children with DS without comorbidity, CHD, and prematurity were analyzed with analysis of variance with Bonferroni correction. We were not able to perform a power analysis for this study, because no estimate of RSV LRTI–associated hospitalization was available. All statistical analyses were performed by using the software program SPSS for Windows 12.0.2 (SPSS Inc, Chicago, IL). A P value of .05 was considered the limit of significance.
Clinical information was available for 176 of 206 children in the retrospective cohort and for 219 of 241 children in the prospective study cohort (Fig 1). Two children were lost to follow-up. A total of 395 children with DS were studied, and 231 (58.5%) were male. All children were followed until the age of 2 years. Baseline characteristics are outlined in Table 1. The third group, the control group, consisted of 276 unmatched siblings; 135 (48.9%) were male.
Risk Factors for RSV LRTI–Associated Hospitalization
In the retrospective cohort, 57 children (32.4%) had hemodynamically significant CHD, and 22 children were born prematurely, 6 of who had hemodynamically significant CHD. The latter children, therefore, had 2 risk factors for RSV LRTI–associated hospitalization. In the prospective cohort, 146 children (66.7%) had a congenital heart abnormality, detected during routine screening during the first weeks of life, and in 84 children (38.4%) the CHD was judged to be hemodynamically significant. Thirty-one children of this cohort had been born prematurely, 9 of who were diagnosed with hemodynamically significant CHD. One child was diagnosed with chronic lung disease. Thus, in the 2 cohorts combined, 180 of a total of 395 children with DS (45.6%) had ≥1 risk factor for severe RSV LRTI. In the control group, 14 (5.1%) of 276 siblings had been born prematurely and 3 children (1.1%) had CHD, which was hemodynamically significant in 1 (0.4%).
Incidence of RSV LRTI–Associated Hospitalization
The incidence of RSV LRTI–associated hospitalization was similar in both the retrospective and prospective cohort, including in children without known risk factors (8.1% vs 7.0%). In the combined cohort, 39 (9.9%) of 395 children with DS were hospitalized for RSV LRTI. (Fig 2) The cases of RSV LRTI–associated hospitalization were equally distributed in different age groups. (Table 2) The rate of hospitalization for RSV LRTI was lower in term children with DS without CHD (7.6%) than in preterm children with DS (9.4%) or in children with DS and significant CHD (11.9%), but these differences did not reach statistical significance. Three children with hemodynamically significant CHD had had complete surgical correction of their heart defect ≥6 weeks before RSV LRTI–associated hospitalization. Two healthy children (0.7%) from the control group were hospitalized for RSV LRTI. Term children with DS without a heart defect were compared with the control group. Because control children did not have routine echocardiography at birth and might have had undiagnosed insignificant CHD, we combined the group of children with DS with hemodynamically insignificant CHD and the group of children with DS without CHD to calculate the odds ratio (OR) for RSV LRTI–associated hospitalization. The OR for RSV LRTI–associated hospitalization was 12.6 (95% confidence interval: 2.9–54.5) among term children with DS without hemodynamically significant CHD and 10.5 (95% confidence interval: 2.2–49.5) among term children with DS without any CHD.
Disease Severity and RSV LRTI–Associated Hospitalization
The severity of RSV LRTI was scored on the basis of the duration of hospitalization, need for supplemental oxygen, and need for mechanical ventilation. The characteristics of RSV LRTI–associated hospitalization are shown in Table 3. Disease severity was not significantly different between children with and without additional risk factors. The 39 children of the combined cohort with DS, who required hospitalization for RSV LRTI, had a median duration of hospitalization of 10 days; 31 children (79.5%) required supplemental oxygen; 5 children (12.8%) required mechanical ventilation.
We found a high incidence of RSV LRTI–associated hospitalization among children with DS. Although pediatricians are keenly aware of the high risk of infections among individuals with DS,10,12 to our knowledge this is the first report of an increased incidence of RSV LRTI–associated hospitalization among children with DS, namely, 9.9% vs 0.7% in control children or 0.5% to 2% among the general pediatric population.1
Different pathophysiologic mechanisms could underlie the high risk of RSV LRTI–associated hospitalization seen among children with DS. For example, ∼50% of children with DS have CHD,7 and hemodynamically significant CHD is likely to be a risk factor for severe RSV LRTI in children with DS. However, no studies have been performed to establish the risk of severe RSV LRTI–associated hospitalization in children with DS-related CHD. In literature, pulmonary hypertension has been mentioned as a possible risk factor for severe RSV infection.13 Pulmonary vascular resistance hypertension occurs more often in children with DS than in the general pediatric population.14,15 Shah et al16 reported that 10% of children with DS had pulmonary hypertension in the absence of CHD. In the current study, pulmonary hypertension was not evaluated in detail. Future studies will have to determine the role of pulmonary hypertension in severe RSV infection in children with DS. Children with DS have an abnormal upper airway physiology, which makes them prone to apnea.17,18 It is possible that apnea is triggered by infections with respiratory viruses, especially RSV. In addition, a decreased lung function at birth may be important in the pathogenesis of RSV LRTI–associated hospitalization.6 Lastly, children with DS seem to have an altered immune response, with thymus development and function being abnormal.19–21 The number of B cells and T cells are low, especially in the first 2 years of life.22 In addition, the defective T-cell ex vivo proliferative responses to nonspecific and antigenic stimuli, cytokine production, and natural killer cell responses detected in individuals with DS are thought to be important to the increased susceptibility of these individuals to infectious pathogens. Taken together, abnormal innate and adaptive immune responses in infants with DS could predispose them to severe disease if they become infected with RSV.
This study has potential methodologic limitations. CHD is a known risk factor for RSV LRTI–associated hospitalization.4 However, although 63% of our population had CHD, only 35.7% had hemodynamically significant CHD. Moreover, the power of this study was inadequate to show differences in the incidence of RSV LRTI–associated hospitalization or disease severity among subgroups of children with DS. We consider it plausible that hemodynamically significant CHD and prematurity are independent risk factors for RSV LRTI–associated hospitalization in children with DS. We did not have a matched-control group, but instead collected information from siblings. The rate of RSV LRTI–associated hospitalization of 0.7% in this large control group was consistent with estimates reported in the literature, which supports the estimated OR of 12.6 for RSV LRTI–associated hospitalization in children with DS without known risk factors for RSV LRTI.1
It could be argued that hospitalization bias explains our study results. However, we do not think that the hospitalization threshold for RSV LRTI was lower for the children with DS. In fact, the children with DS may have had more severe RSV LRTI than the general pediatric population. The duration of hospitalization for RSV LRTI varies among countries, but in the Netherlands the median length of stay is 8 days.2 It was 10 days in the children with DS. Moreover, 79.5% of the hospitalized children with DS required supplemental oxygen, compared with 38.7% to 68.2%, and 12.8% of the children with DS required mechanical ventilation, compared with 2.2% in the general pediatric population.2 These results indicate that the children with DS had more severe RSV LRTI and that the increased incidence of RSV LRTI–associated hospitalization was not because of hospitalization bias.
The results may have clinical implications. We found a substantial proportion of children with DS to have RSV LRTI. Although vaccination against RSV LRTI is not yet possible, targeted RSV prophylaxis in high-risk populations is currently possible with palivizumab, a monoclonal antibody against the F-protein of RSV.23 We also found DS to be associated with a similar rate of RSV LRTI–associated hospitalization as that previously reported for prematurity, chronic lung disease, and CHD.13,23,24 Our findings therefore support the possibility of a new indication for RSV prophylaxis in children with DS up to 2 years of age, although the safety and efficacy of such an approach remains to be determined.
Our study demonstrates a very high incidence of RSV LRTI–associated hospitalization in children with DS and suggests that these children have more severe disease when hospitalized. These findings warrant additional study of the mechanisms underlying severe RSV LRTI in children with DS.
- Accepted May 25, 2007.
- Address correspondence to Louis J. Bont, MD, PhD, University Medical Centre Utrecht, Department of Pediatrics, Room KE 04.133.1, PO Box 85090, 3508 AB, Utrecht, Netherlands. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Figueras-Aloy J, Carbonell-Estrany X, Quero J. Case-control study of the risk factors linked to respiratory syncytial virus infection requiring hospitalization in premature infants born at a gestational age of 33–35 weeks in Spain. Pediatr Infect Dis J.2004;23 :815– 820
- ↵Welliver RC. Review of epidemiology and clinical risk factors for severe respiratory syncytial virus (RSV) infection. J Pediatr.2003;143(5 suppl) :S112– S117
- ↵Marcus CL, Keens TG, Bautista DB, Vonpechmann WS, Ward SLD. Obstructive sleep-apnea in children with Down syndrome. Pediatrics.1991;88 :132– 139
- Murphy M, Hyun W, Hunte B, Levine AD, Epstein LB. A role for tumor necrosis factor-alpha and interferon-gamma in the regulation of interleukin-4-induced human thymocyte proliferation in vitro. Heightened sensitivity in the Down syndrome (trisomy 21) thymus. Pediatr Res.1992;32 :269– 726
- Copyright © 2007 by the American Academy of Pediatrics