The Burden of Severe Respiratory Syncytial Virus Disease Among Children Younger than 1 Year in Central and Eastern Europe

Globally, respiratory syncytial virus (RSV) is a major cause of hospital admission in infants <1 year of age [1]. Mortality data demonstrate that RSV kills more infants in this age group than any other single pathogen with the exception of malaria [2]. Environmental factors and individual characteristics carrying increased risk of RSV-associated acute respiratory tract infections include low birth weight, being a member of a multiple birth, smoking during pregnancy, having siblings, day care attendance (or sibling attendance), living in a crowded household, low parental education, second-hand smoke exposure, history of atopy and lack of breastfeeding [3‐10]. Notably, being male carries a risk ratio of 1.425:1 compared with being female [11]. Certain high-risk groups including premature infants, infants with bronchopulmonary dysplasia (more commonly referred to as chronic lung disease of prematurity) and those with congenital heart disease are particularly vulnerable to severe RSV disease [12, 13]. RSV is more likely to progress to the lower respiratory tract and cause more serious disease in these high-risk children compared with otherwise healthy term children. Furthermore, the course of illness tends to be more severe in these children versus otherwise healthy term children, and is associated with higher rates of hospitalization, intensive care unit admission, mechanical ventilation and death [14‐18].

The RSV burden of disease is relatively well characterized in Western Europe; however, there are limited epidemiologic data available for Central and Eastern Europe (CEE) and the developing world. Mortality data are especially difficult to ascertain as the majority of children in developing countries die outside of the hospital setting [19]. Characterization of children with severe RSV disease is critically important to inform health care authorities to target interventions that maximize population health. The present study aimed to characterize RSV disease incidence and severity among premature and term children in CEE. We evaluated the severity of RSV-associated lower respiratory tract infection (LRTI) disease in children <1 year of age during the 2009–2010 and 2010–2011 RSV seasons and investigated the influence of gestational age (GA) on the treatment, complications and outcomes of RSV LRTI hospitalization.

During the study period of October 2009–April 2010 and October 2010–April 2011, overall, 3841 records were collected from 48 sites across 12 CEE countries. Among them, 3474 children (90.4%) fulfilled the case definition of a documented hospitalized LRTI and were enrolled in the study (Fig. 1). Of the 3474 evaluable children, 757 children (21.8%) were premature (≤36 weeks GA) and 2679 (77.1%) were term. The majority of the population was male (58.3%) and younger than 6 months (69.1%) at the time of admission. RSV prophylaxis had been administered in 2% (n = 68) of the entire cohort. Demographic and clinical characteristics of the study cohort are summarized in Table 1.

A rapid RSV test was performed in 3354 (96.5%) cases and RSV tests were positive in 1423 (42.4%). Of the RSV-positive cases, 266 (18.7%) were premature children and 1034 (72.7%) were term children with no risk factors or chronic diseases. RSV prophylaxis had been administered in 11 RSV-positive cases, among them 9 children with GA <33 weeks and 2 term children. A comparison of the baseline admission demographics, clinical characteristics and outcomes of RSV LRTI hospitalization between premature and term children is illustrated in Table 2.

Significant differences in the diagnosis at admission were observed between premature and term children. Premature children were more frequently hospitalized due to pneumonia (37.6% vs 20.0%) and term children were more frequently hospitalized due to bronchiolitis (48.5% vs 71.2%; P < 0.001). Hospital stays were significantly longer for premature children compared with term children (17 vs 8 days; P < 0.001; P value adjusted for admission diagnosis: <0.001). Premature children had a 3.3-fold (95% CI: 2.66, 4.09) increased risk for ICU hospitalization compared with term children (P < 0.001); after adjustment for admission diagnosis (including only those admitted with a diagnosis of either pneumonia or bronchiolitis), risk was increased 2.8-fold (95% CI: 2.23, 3.47; P < 0.001). ICU hospitalizations were more frequent for premature children (41.4% vs 12.6%; P value not determined), and ICU stays were longer (13 vs 6 days; P < 0.001) compared with term children (P value adjusted for admission diagnosis: <0.001).

Premature children had a 1.6-fold (95% CI: 1.43, 1.74) increased risk of receiving supplemental oxygen compared with term children (P < 0.001); after adjustment for admission diagnosis, risk was 1.6-fold greater (95% CI: 1.44, 1.78; P < 0.001). Compared with term children, oxygen was administered in premature children markedly more frequently (72.9% vs 46.3%; P value not determined) and for a significantly longer mean duration (9 vs 4 days; P < 0.001; P value adjusted for admission diagnosis: <0.001).

The main objective of this multinational study was to characterize the severity of RSV-associated LRTI disease among premature and term CEE children younger than 1 year across two RSV seasons. Our analysis of 1423 RSV-positive children demonstrates substantial morbidity in the first year of life among children hospitalized for RSV LRTI in CEE. Consistent with other geographic regions, premature children with chronological age <1 year are at greater risk for a more severe course of RSV disease compared with their term-delivered counterparts. These results provide further evidence that, like exposure to second-hand smoke, being male, and living with siblings, prematurity is a near-universal risk factor for severe RSV-associated infection [3‐9, 11].

Premature children exhibited longer stays in the hospital, more frequent and longer stays in the ICU and more frequent and longer administration of supplemental oxygen than term children; even after adjusting for the admission diagnosis, the differences remain significant (P < 0.001). In fact, the risk of an ICU stay was 2.8 times higher in premature than term children after adjusting for admission diagnosis. Furthermore, decreasing GA at birth was found to increase risk for ICU hospitalization, prolonged duration of hospitalization, supplemental oxygen administration and death. Even among late premature children (33–36 weeks gestation), the course of RSV-positive LRTI disease was more severe than that of full-term children. Compared with term children, a greater proportion of late premature children (33–36 weeks gestation) with RSV LRTI were hospitalized in the ICU (33.8% vs 14.1%), spent ≥15 days in the hospital (1.4% vs 1.0%) and required supplemental oxygen administration (68.3% vs 47.6%). These findings are consistent with previous studies in Western Europe and the United States that reported a high rate of healthcare utilization for RSV-positive LRTI among premature children 33–35 weeks gestation [15, 21].

The strength of this study is that it investigates a large multinational cohort of children from a geographical area that has not been previously well characterized. However, this study has several key limitations. We acknowledge that the study data from the 2009 to 2010 and 2010 to 2011 RSV seasons may not appear current. However, to our knowledge, no substantial developments in the medical practice of prevention or care of these children for RSV LRTI, including the use of prophylaxis with palivizumab, have since occurred in CEE that would change study results if conducted today. As such, the clinical relevance and severity of RSV-associated LRTI disease likely has remained the same. RSV reinfection is a common occurrence [22] and we were not able to account for possible reinfections in this study. In addition, the number of collected cases in our study was affected by factors associated with the recruitment process, such as time resources of the investigators, availability of diagnostic tests, strategies for identifying cases and obtaining consent for study participation. Given that data were collected on the day of hospital discharge or during the first checkup visit after LRTI hospitalization, bias may have been introduced. The RSV rapid diagnostic tests that utilize antigen capture technology have a sensitivity and specificity of approximately 90%. Data regarding timing of the RSV rapid test during the course of the disease were not captured. As a result, the true proportion of RSV infections may have been underestimated. Use of univariate analyses, instead of multivariate analyses, may have increased the likelihood of an analysis reaching statistical significance by chance. Furthermore, the lack of adjustment for potential confounders or other potential risk factors limits the value of the univariate assessment of potential risk factors. However, the primary endpoint was to evaluate the severity of RSV-associated LRTI disease in premature and term children. Primary outcomes included duration of hospitalization, and the incidence, duration and risk of ICU hospitalization. Because investigation of the potential risk factors was an additional, exploratory data analysis in the statistical analysis plan and the findings would not have impacted the primary endpoint of this study, we did not pursue a multivariate analysis of these factors. Because diagnosis of LRTI was based on individual physician/hospital practice, local or regional differences may also have introduced bias. National and regional differences in treatment practice and healthcare systems, especially the criteria for hospitalization, may have affected our overall risk factor findings and may also affect assessments of congenital heart disease, bronchopulmonary dysplasia, Down syndrome and neuromuscular disease, ICU admittance and use of supplemental oxygen and bronchodilators.

Overall, the results of this study demonstrate the importance of RSV as a significant contributor to pediatric morbidity in the first year of life and confirm results from studies outside of CEE that show that prematurely born children hospitalized for RSV-related infections have a more severe course of disease than full-term children, as demonstrated by longer stays in the hospital and more frequent and longer stays in the ICU than term children without risk factors.