SEARCH

SEARCH BY CITATION

Keywords:

  • asthma;
  • bronchiolitis;
  • child;
  • recurrent wheezing;
  • respiratory syncytial virus;
  • rhinovirus

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Background:  Recent studies have suggested that rhinovirus-associated early wheezing is a greater risk factor for development of recurrent wheezing in children than is early wheezing associated with respiratory syncytial virus (RSV). We determined the development of recurrent wheezing in young children within 3 years after hospitalization for RSV or non-RSV bronchiolitis.

Methods:  We identified retrospectively all children <2 years of age who were admitted to Turku University Hospital because of bronchiolitis in the months of August–December during 1988–2001. The primary outcome was recurrent wheezing that required long-term asthma medication. Data on asthma medications of the individual children were derived from the Social Insurance Institution of Finland.

Results:  Within the first year after hospitalization, 36 of 217 (16.6%) children with non-RSV bronchiolitis developed recurrent wheezing, compared with five of 199 (2.5%) children with RSV bronchiolitis [relative risk (RR) 6.6; 95% confidence interval (CI) 2.6–16.5]. The rates of recurrent wheezing were significantly increased in the non-RSV group also within 2 years (RR 2.9; 95% CI 1.7–5.1) and 3 years (RR 3.4; 95% CI 2.0–5.7) after hospitalization. The increased risk of recurrent wheezing in children with non-RSV-associated bronchiolitis was observed both in boys and girls at all time points of the 3-year follow-up, and it was not explained by the age difference between the RSV and non-RSV groups or any confounding seasonal factors.

Conclusion:  Children hospitalized with bronchiolitis caused by other viruses than RSV develop recurrent wheezing at substantially higher rates during a 3-year follow-up period than do children with RSV-induced bronchiolitis.

Abbreviations:
95% CI

95% confidence interval

ICD

International Classification of Diseases

RR

relative risk

RSV

respiratory syncytial virus

Bronchiolitis is a frequent illness and one of the main causes of hospitalization in infants and young children. In the US, approximately 3% of children <1 year of age with no underlying medical conditions are hospitalized for bronchiolitis during respiratory syncytial virus (RSV) epidemics (1). Several viruses are known as causative agents of bronchiolitis. In infants, RSV is clearly the primary cause, although other viruses, e.g. rhinoviruses, enteroviruses, human metapneumovirus and human bocavirus, are also frequently found in infants hospitalized with this condition (2–5). In children with 1 year of age or older, the relative importance of RSV in the etiology of bronchiolitis is substantially lower, and the predominant viruses detected in these children are rhinoviruses and enteroviruses (3).

Many children who experience bronchiolitis in infancy develop recurrent wheezing later in life. RSV infection has been recognized as an important risk factor for recurrent wheezing in several studies (6–10). The pathogenetic mechanisms that lead to recurrent wheezing after RSV bronchiolitis are still unknown but there is evidence to suggest that both genetic and environmental factors contribute to the host immune response to RSV infection and that this response, in turn, may adversely affect the development of lungs and the control mechanisms of the lower airways (11–13).

Recent studies have suggested that children with bronchiolitis associated with other viruses than RSV, especially rhinovirus, have a greater risk of developing recurrent wheezing than do those with RSV-induced wheezing (14–20). In a large prospective study in the US, rhinovirus-associated wheezing in infancy was determined as the strongest predictor of subsequent wheezing during the third year of life and the development of asthma at the age of 6 years (16, 20).

The purpose of our study was to assess the development of recurrent wheezing in young children who had been hospitalized for bronchiolitis caused by RSV or other viruses. In Finland, RSV epidemics follow a distinct 2-year cycle (21), with small outbreaks occurring in the spring and early summer of each odd-numbered year, followed by large outbreaks during the subsequent autumn and winter (Fig. 1). During the autumn of every even-numbered year, RSV is virtually undetectable in Finland, and during this time acute wheezing illnesses in children are caused by other viruses than RSV. This well-established epidemiologic pattern of RSV infection provided us with a unique opportunity to compare the impact of RSV and other viruses on the development of recurrent wheezing.

image

Figure 1.  Monthly occurrence of laboratory-confirmed respiratory syncytial virus (RSV) infections in children seen at Turku University Hospital during 1988–2001. Source of data: Department of Virology, University of Turku, Finland.

Download figure to PowerPoint

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Study population

The study population consisted of all children <2 years of age who were admitted to Turku University Hospital for bronchiolitis in the months of August–December during the 14-year period of 1988–2001. Bronchiolitis was defined as the first episode of acute expiratory wheezing in a child <2 years of age. The patients were identified retrospectively by using the patient discharge registry of the hospital. To maximize the probability of finding all children who might fulfill the inclusion criteria, we initially used a wide range of International Classification of Diseases (ICD)-9 and ICD-10 codes to search for the children (ICD-9: 4660A, 4661A, 4801A, 4900A, 4912A, 4939B, 4939X; ICD-10: J20.5, J21.0, J21.8, J21.9). The medical records of all 649 children identified by the initial search were carefully examined. Children with any history of previous wheezing episodes, including those diagnosed in primary health care or at other hospitals, were excluded, leaving 520 children for the analyses.

The children were divided into two groups: RSV and non-RSV group (Fig. 2). All 199 children with laboratory-confirmed RSV infection comprised the RSV group, and all 117 children who were either negative for RSV or who had a confirmed infection by some other virus than RSV were placed in the non-RSV group. Furthermore, 100 children from whom a viral specimen had not been obtained and who were hospitalized in an even-numbered year (when RSV did not circulate) were placed in the non-RSV group. All 104 children without a viral specimen who were hospitalized in an odd-numbered year were excluded from the analyses. The final non-RSV group consisted of 217 children.

image

Figure 2.  Study flow-chart.

Download figure to PowerPoint

Viral diagnosis

Nasopharyngeal aspirates were collected with a disposable catheter that was inserted in the nostril to a depth of 5–7 cm and drawn back while applying gentle suction with an electric suction device. The specimens were transported at room temperature to the laboratory (Department of Virology, University of Turku), where they were processed the same or the following day. Viral antigens in the aspirates were detected by one-incubation, monoclonal time-resolved fluoroimmunoassay for RSV, influenza A and B viruses, parainfluenza types 1, 2, and 3 viruses, and adenovirus (22). In occasional cases, also rhinoviruses and enteroviruses were searched for by using polymerase chain reaction-based methods.

Study outcomes

The primary outcome of the study was development of recurrent wheezing requiring long-term asthma medication that was reimbursed by the Social Insurance Institution. This outcome was analyzed separately for the first 1, 2, and 3 years after the child’s hospitalization for bronchiolitis.

Asthma medications

Data on asthma medications of the children were derived from the Social Insurance Institution that is the sole authority in Finland to keep a register of all sales and reimbursements of prescription medications on an individual level. The Social Insurance Institution supplies reimbursement for asthma medications if certain diagnostic criteria are fulfilled. An essential requirement for granting reimbursement for asthma medications to a child was a medical certificate issued by a physician verifying the presence of recurrent wheezing (≥3 medically attended wheezing episodes) or the diagnosis of asthma. A further requirement was a long-term need for asthma controllers (e.g. corticosteroids); temporary ‘as-needed’ use of bronchodilators alone was not sufficient for granting reimbursement. For the purposes of this study, a child was considered to have recurrent wheezing if reimbursement for asthma medications was granted by the Social Insurance Institution. The time period between the hospital admission and the start of the reimbursement was calculated to 1-month accuracy.

Statistical methods

Comparison of proportions between the groups was performed by using the standard chi-square test. Yates’ correction for continuity was used when the expected value in any cell was less than 5. The Mann–Whitney U-test was used to compare continuous nonparametric data. For the analyses of matched pairs, the relative risk (RR) for the development of recurrent wheezing between the RSV and non-RSV groups was calculated using Poisson regression. Generalized estimating equations were used in the Poisson regression analysis to take into account the matched-pair design. The odds ratio in the matched case–control analysis was calculated using conditional logistic regression. P values <0.05 were considered to indicate statistical significance.

Ethics

The study was approved by the Ethics Committee of the Hospital District of Southwest Finland.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Subjects

The final study population consisted of 416 children (199 children in the RSV group and 217 in the non-RSV group). The median age of the children at hospital admission was 0.34 years in the RSV group and 0.92 years in the non-RSV group (P < 0.001). A total of 180 (90.5%) children in the RSV group and 117 (53.9%) in the non-RSV group were <1 year of age at hospitalization (Fig. 3). Two hundred sixty-eight (64.4%) of all children were boys. The proportion of boys was higher (153/217, 70.5%) in the non-RSV group than in the RSV group (115/199, 57.8%; P = 0.007).

image

Figure 3.  The age distribution of children hospitalized with RSV or non-RSV bronchiolitis.

Download figure to PowerPoint

Development of recurrent wheezing

Within the first year after hospital admission for bronchiolitis, 36 of 217 (16.6%) children in the non-RSV group developed recurrent wheezing, compared with five of 199 (2.5%) children in the RSV group [RR = 6.6; 95% confidence interval (CI) 2.6–16.5]. The risk of developing recurrent wheezing was significantly increased in the non-RSV group also within the first 2 years (RR = 2.9; 95% CI 1.7–5.1) and 3 years (RR = 3.4; 95% CI 2.0–5.7) after hospitalization (Table 1). The increased risk of recurrent wheezing in the non-RSV group was observed both in boys and girls at all time points of follow-up.

Table 1.   Development of recurrent wheezing within 1, 2, and 3 years after hospitalization in children with RSV or non-RSV bronchiolitis
 RSV groupNon-RSV groupRR (95% CI)P-value
Total no. of childrenNo. (%) with recurrent wheezingTotal no. of children No. (%) with recurrent wheezing
  1. RR, relative risk; RSV, respiratory syncytial virus; 95% CI, 95% confidence interval.

All children
 1 year1995 (2.5)21736 (16.6)6.6 (2.6–16.5)<0.001
 2 years19915 (7.5)21748 (22.1)2.9 (1.7–5.1)<0.001
 3 years19916 (8.0)21759 (27.2)3.4 (2.0–5.7)<0.001
Girls
 1 year841 (1.2)6410 (15.6)13.1 (1.7–99.9)0.003
 2 years844 (4.8)6413 (20.3)4.3 (1.5–12.5)0.003
 3 years845 (6.0)6415 (23.4)3.9 (1.5–10.3)0.002
Boys
 1 year1154 (3.5)15326 (17.0)4.9 (1.8–13.6)<0.001
 2 years11511 (9.6)15335 (22.9)2.4 (1.3–4.5)0.004
 3 years11511 (9.6)15344 (28.8)3.0 (1.6–5.6)<0.001

When the analysis was restricted to children from whom a viral sample had been obtained (RSV group, n = 199; non-RSV group, n = 117), the RR of recurrent wheezing in the non-RSV group was 8.2 (95% CI 3.2–20.8) within the first year, 3.5 (95% CI 2.0–6.2) within 2 years, and 3.7 (95% CI 2.2–6.4) within 3 years after hospitalization (P < 0.001 for all comparisons).

Overall, 55 of 268 (20.5%) boys and 20 of 148 (13.5%) girls developed recurrent wheezing during the first 3 years after hospitalization. The RR of recurrent wheezing for boys compared with girls was 1.5 (95% CI 0.95–2.4; P = 0.075).

Recurrent wheezing in children hospitalized at 4–9 months of age

Because of substantial differences in the age distribution between the RSV and non-RSV groups (Fig. 3), the development of recurrent wheezing was analyzed separately in an age-balanced subgroup of children 4–9 months of age at the time of hospitalization. This subgroup consisted of 75 children with RSV bronchiolitis (mean age 6.0 months) and 69 children with non-RSV bronchiolitis (mean age 6.6 months). The risk of recurrent wheezing was significantly greater in the non-RSV group than in the RSV group within the first 2 years (RR = 6.9; 95% CI 2.1–22.2; P < 0.001) or 3 years (RR = 3.8; 95% CI 1.6–8.9; P < 0.001) after admission (Fig. 4). The RR for the first year could not be calculated because none of the children in the RSV group developed recurrent wheezing within that period.

image

Figure 4.  Development of recurrent wheezing within 3 years after hospitalization in a subgroup of children aged 4–9 months (75 children with RSV and 69 with non-RSV bronchiolitis).

Download figure to PowerPoint

Matched pair analysis

To further explore the potential impact of children’s age at the time of hospitalization on the results, children with RSV-associated bronchiolitis were matched by age (maximum age difference 7 days in 77% and 20 days in 100% of pairs) and gender with children with non-RSV-associated bronchiolitis. A total of 114 matched pairs were available for analysis. Recurrent wheezing was diagnosed within 3 years after hospitalization in 29 (25.4%) children in the non-RSV group and nine (7.9%) children in the RSV group (RR = 3.2; 95% CI 1.6–6.7; P = 0.002).

Matched case–control analysis

Children with recurrent wheezing within the first 3 years after hospitalization were individually matched by age (maximum age difference 7 days in 89% and 14 days in 100% of pairs) and gender with children who did not develop recurrent wheezing. An appropriate control was available for 74 of all 75 (99%) cases with recurrent wheezing. In 58 of 74 (78.4%) children with recurrent wheezing, the initial bronchiolitis had been caused by non-RSV pathogens, and 16 (21.6%) children had had RSV-associated disease. Among the controls, 36 (48.6%) had had non-RSV-associated bronchiolitis and 38 (51.4%) RSV-associated bronchiolitis (odds ratio = 6.5; 95% CI 2.3–18.6; P = 0.0005).

Assessment of potential confounding seasonal factors

Further analyses were performed to rule out the potential impact of any unknown confounding seasonal factors (other than RSV circulation) between the odd-numbered and even-numbered years. First, we included only children who were hospitalized and sampled in odd-numbered years, which practically ensured that the children in the RSV group (n = 198) and non-RSV group (n = 60) were exposed similarly to any confounding environmental factors. In this analysis, the RR of recurrent wheezing was significantly increased in the non-RSV group (1 year, RR = 5.3, 95% CI 1.8–15.5, P = 0.003; 2 years, RR = 2.2, 95% CI 1.0–4.6, P = 0.037; 3 years, RR = 2.7, 95% CI 1.4–5.3, P = 0.004).

Secondly, to assess the potential confounding effect of children from whom viral samples were not obtained, we included all 520 children hospitalized for bronchiolitis during the 14-year study period (odd-numbered years, n = 362; even-numbered years, n = 158). The development of recurrent wheezing was significantly more frequent among children who were admitted in even-numbered years (when RSV did not circulate) than among those admitted in odd-numbered years with RSV circulation (1 year, RR = 2.6, 95% CI 1.5–4.3, P < 0.001; 2 years, RR = 2.0, 95% CI 1.3–2.9, P < 0.001; 3 years, RR = 2.0, 95% CI 1.4–2.8, P < 0.001).

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References

Our results indicate that bronchiolitis caused by other viruses than RSV is associated with a substantially increased risk of recurrent wheezing when compared with an RSV-associated primary wheezing episode. This effect was sustained during a follow-up period of 3 years after hospitalization for bronchiolitis. Our main finding was further strengthened by several subgroup analyses that demonstrated that the observed differences in the development of recurrent wheezing were not explained by age differences between the groups or any potential confounding seasonal factors between the odd- and even-numbered years.

Our findings provide strong, albeit indirect, support for recent studies suggesting that rhinovirus-induced early wheezing is a major risk factor for recurrent wheezing later in life (14–20). In one of the first studies challenging the primary role of RSV in the development of asthma, Reijonen et al. (14) followed infants hospitalized for wheezing and reported that the identification of RSV as the etiologic agent was surprisingly associated with a decreased occurrence of asthma during a 3-year period after the hospitalization when compared with RSV-negative children. In subsequent reports of that cohort of children, hospitalization for rhinovirus-induced wheezing was demonstrated as an important risk factor for the development of asthma, and the increased risk persisted at least until late teenage years (15, 17, 19). In a prospective study of children from birth to 6 years of age in the US, Lemanske et al. (16) and Jackson et al. (20) concluded that the most significant risk factor for the development of preschool childhood wheezing and asthma was the occurrence of rhinovirus-associated wheezing during infancy. In contrast, children who wheezed only with RSV during their first 2 years of life were not at increased risk for asthma at the age of 6 years (20). Recently, Dunder et al. (23) reported that exposure to RSV in early infancy did not increase the consumption of asthma medicines at the population level.

Because rhinoviruses and enteroviruses were not routinely searched for at our hospital during the period of this study, we were unable to determine the viral etiology of most children who had a non-RSV illness. It is most likely, however, that the majority of non-RSV-associated cases were caused by rhinoviruses or enteroviruses. This assumption is based on a recent 2-year study of the viral etiology of acute expiratory wheezing at our hospital, in which the causative viral agent could be detected in 95% of the cases (3, 5). In infants younger than 1 year of age, RSV accounted for 54% and picornaviruses for 42% of all wheezing illnesses, which left only a minor role for any other viruses in the etiology of this illness. In children aged 12–35 months, picornaviruses were detected in 65% and RSV in 22% of the cases (3).

It is important to emphasize that the increasing recognition of rhinovirus-associated primary wheezing as a major risk factor for the development of asthma does not necessarily imply that primary wheezing caused by RSV infection would not run an increased risk of asthma. In an 11-year follow-up of infants hospitalized for wheezing, the risk of teenage asthma was 10-fold after rhinovirus-induced wheezing but also fivefold after RSV-induced wheezing (17).

In our study, 8% of children hospitalized with RSV-associated bronchiolitis developed recurrent wheezing within the following 3 years. This rate is much lower than the rates observed for RSV-infected children in many previous studies with varying durations of follow-up (8, 9, 20). While the threshold of admitting children to hospital may differ between different settings and countries, the most likely explanation for the apparent discrepancy between our results and those of the previous studies is the stringent criteria that we used in defining our primary outcome. In the present study, children were classified as recurrent wheezers only if they were granted reimbursement for their long-term asthma medication by the Social Insurance Institution, which practically excluded all children with less severe wheezing who might have used asthma medications on a temporary basis. Although our rates of recurrent wheezing are thus not comparable with those observed in other studies, there is no reason to assume that the criteria for granting reimbursement would have been different for children with previous RSV or non-RSV bronchiolitis. It should be noted, however, that even recurrent wheezing during the first years of life does not automatically lead to persistent asthma later in life.

Recent studies assessing the efficacy of systemic corticosteroids in the management of wheezing have provided additional evidence for the concept that there are distinct differences between infants hospitalized with rhinovirus and RSV-associated wheezing (18, 24, 25). As observed also in the present study, infants with rhinovirus illnesses are generally older than those hospitalized with RSV infection (24, 26). Moreover, infants hospitalized with rhinovirus-induced wheezing seem to have a different response to anti-inflammatory treatment when compared with RSV-infected infants. In studies where children were followed up to 1 year after hospitalization for wheezing, prednisolone treatment was associated with a significantly decreased risk of recurrent wheezing in infants hospitalized with rhinovirus infection, but not in infants hospitalized with RSV-induced wheezing (18, 24).

Male gender is a well-established risk factor for recurrent wheezing (27). In our study, two-thirds of children initially admitted for bronchiolitis were boys. In the development of recurrent wheezing, however, it appears that the viral etiology of the primary wheezing episode is a more important factor than gender. The risk of recurrent wheezing was significantly increased after non-RSV bronchiolitis in both girls and boys during all time periods analyzed. Although there was a tendency for boys to develop recurrent wheezing at a relatively higher rate than in girls, the difference did not reach statistical significance in a group of 416 children.

Our study has some limitations. Most important, we could not determine the viral etiology of bronchiolitis in most children in the non-RSV group because rhinoviruses and enteroviruses were not routinely searched for during the study period. Because of the retrospective nature of the study, we were unable to retrieve comprehensive information on various environmental and genetic factors, such as the presence of atopy or maternal asthma, that are known as strong predictors of persistent wheezing in children. We also studied only hospitalized children. On the other hand, a major strength of our study was the meticulous examination of all medical records to rule out any previous wheezing episodes, which largely excluded the possibility that children hospitalized for non-RSV illnesses had previously had an RSV illness that could have predisposed them to the non-RSV illness. Other strengths of the study included the utilization of a reliable and comprehensive register of reimbursement for asthma medications, a study period of 14 consecutive years to diminish the potential effect of any season-to-season variation, and the well-established seasonal pattern of RSV epidemics in Finland.

In conclusion, our results demonstrate that bronchiolitis caused by other viruses than RSV confers a substantially greater risk of recurrent wheezing during a 3-year period after hospitalization than in RSV-induced bronchiolitis. These data provide indirect support for the recent recognition of rhinovirus as a major risk factor for the development of recurrent wheezing in children. Further studies are needed to determine whether primary rhinovirus-associated wheezing during a critical period of life induces a long-lasting ‘innate imprint’ on the child’s immune response (28), or whether it just serves as a sensitive indicator of children who are at increased risk of developing recurrent wheezing and asthma later in life.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References