Clinical relevance of adverse pulmonary outcomes following severe bronchiolitis

Acute viral bronchiolitis is among the most common causes of pediatric critical illness, accounting for 12.8–13.5% of all pediatric intensive care unit (PICU) admissions in Europe and North America.^1–3 Nearly 20% of all children hospitalized for bronchiolitis require PICU care.^3,4 While caring for a child with bronchiolitis may be routine for many PICU providers, bronchiolitis hospitalization can have a profound impact on families, with high levels of worry and distress.⁵ Of multiple domains of concern, the area with the greatest impact on parents of children hospitalized with bronchiolitis both at and following hospital discharge is fear for the future.⁵

Anticipatory guidance for families whose children are hospitalized with bronchiolitis is likely to include a discussion of the future risk of pulmonary complications. Many studies have examined wheezing and asthma following bronchiolitis; a recent meta-analysis demonstrated a clear association with unadjusted odds of subsequent wheezing 3.4 times greater and unadjusted odds of asthma 2.6 times greater among children following respiratory syncytial virus bronchiolitis.⁶ It remains unclear, however, whether this represents a causal relationship, with many studies at high risk of bias from unmeasured confounding.⁶

In this month’s issue of Pediatric Critical Care Medicine, de Sonnaville and colleagues examined long-term pulmonary outcomes among children invasively ventilated for acute viral bronchiolitis as infants.⁷ The authors posited that because recurrent wheezing or asthma are more common among children who were hospitalized with bronchiolitis compared to those who were not hospitalized,⁸ adverse pulmonary outcomes may be associated with disease severity. Despite the fact that the proportion of infants hospitalized for bronchiolitis requiring mechanical ventilation has increased over the past several decades,^9,10 there is surprisingly little known about the long-term outcomes of these patients.

In this context, de Sonnaville and colleagues conducted a cross-sectional study of 74 children ages 6–12 years who had previously undergone mechanical ventilation for bronchiolitis as infants at Emma Children’s Hospital in Amsterdam. They assessed pulmonary symptoms via a validated questionnaire, history, and physical exam, used spirometry to evaluate pulmonary function before and after administration of a short-acting beta-2 agonist, evaluated risk factors for abnormal pulmonary function, and assessed the association between adverse pulmonary outcomes with school and sports participation and medication use.

The authors found that 54% of patients had a history of wheezing episodes following their admission for bronchiolitis, and 26% had abnormal pulmonary function on spirometry confirmed by a pediatric pulmonologist. The majority of patients with abnormal pulmonary function were diagnosed with asthma, and the remainder had other obstructive lung pathology not responsive to beta-2 agonist therapy. While the authors did not identify an association between their primary exposure of duration of mechanical ventilation with pulmonary function abnormalities, a secondary analysis just among the patients diagnosed with asthma demonstrated an association with longer duration of ventilation and a family history of atopy. Reassuringly, adverse pulmonary outcomes did not have any clinically significant association with sports participation or school absenteeism, though unsurprisingly patients with pulmonary function abnormalities were found to have had greater use of antibiotics, beta-2 agonists, and corticosteroids.

The authors are to be commended for their rigorous evaluation of pulmonary sequelae with symptom assessment and pulmonary function testing for all included patients, with repeat examination and testing by a pediatric pulmonologist for all patients with recurrent wheezing or pulmonary function abnormalities on initial evaluation to confirm the diagnosis. They also amassed a relatively large sample size for a study of long-term pulmonary outcomes; most studies evaluating spirometry among children with acute respiratory failure have included fewer than 30 patients.^11–13 The five- to 11-year time to follow-up offers important insight into the potential long-term impacts of bronchiolitis on pulmonary function and its effect on patients’ daily lives, though adds the limitations that there was considerable variation in time to follow-up and that potential intervening illnesses or changes in the patient’s environment weaken any attempt to establish a causal relationship between the bronchiolitis episode and later pulmonary function.

The study could have been further strengthened by providing additional information about how the included children differed from the 38% of eligible patients who did not participate, particularly in relation to history of prematurity, socioeconomic status, and demographic information, as these factors are all associated with asthma risk.¹⁴ It would also have been valuable to see a comparison of how the extensive set of patient, environmental, and clinical risk factors assessed differed between patients with and without adverse pulmonary outcomes. While there were too many variables assessed to include most in a multivariable model due to overfitting, the authors only report that duration of mechanical ventilation was not associated with adverse pulmonary outcomes but do not report what other exposure variables were considered. Finally, the hypothesis that illness severity would be associated with long-term pulmonary function may have been more easily examined by comparing children with bronchiolitis who received different levels of respiratory support including both invasive and non-invasive ventilation rather than just evaluating the duration of invasive ventilation.

With 3–8% of all patients hospitalized with bronchiolitis undergoing mechanical ventilation,^3,10 the finding that over half of children mechanically ventilated for bronchiolitis may experience recurrent wheezing and over one-quarter have abnormal pulmonary function years after their admission for bronchiolitis is highly relevant and offers important anticipatory guidance for families. What is important to emphasize, however, is that a causal relationship between bronchiolitis and adverse pulmonary outcomes has yet to be definitively established.⁶ The association between bronchiolitis and wheezing decreases when adjusted for genetic influences,⁶ and bronchial hyperresponsiveness in neonates is associated with later development of severe bronchiolitis.¹⁵ Given the association in this study between a family history of atopy and a diagnosis of asthma, it is difficult to know whether patients with a predisposition to asthma may have been more likely to become severely ill with bronchiolitis and thus be overrepresented in a mechanically ventilated cohort.

Perhaps more relevant than whether the association between bronchiolitis and asthma is causal is that fewer than one-third of the patients with pulmonary function abnormalities in this study had been previously diagnosed with asthma, suggesting that impaired lung function may commonly be underrecognized. It is unclear, however, to what degree this under-recognition is clinically meaningful. The authors appropriately speculate that while most patients with pulmonary function abnormalities participated in sports and missed very few days of school due to respiratory concerns, it is possible that their functional performance could improve with targeted asthma management. One study demonstrated that children with undiagnosed asthma have lower health-related quality of life and more missed school days due to respiratory complaints than healthy children,¹⁶ but it is unknown whether quality of life or school attendance improve following diagnosis given multiple confounding factors.

Interestingly, the 8% of children in this study that had been previously diagnosed with asthma is a similar prevalence to the 8.3% of children in the general population with asthma.¹⁷ Previous population-based studies have estimated that up to 60% of children with asthma diagnosed by pulmonary function testing are undiagnosed and that underdiagnosis is associated with socioeconomic status.¹⁸ This raises the consideration of whether the asthma prevalence among children mechanically ventilated for bronchiolitis is in fact higher than that in the general population, or whether asthma and other obstructive lung disease was just identified at a higher rate in this study given the rigorous screening protocol.

This work by de Sonnaville and colleagues in an important contribution to our understanding of the ongoing burden of severe bronchiolitis on patients and families. Clear next steps in this area of research will be to directly compare long-term pulmonary function between patients with and without a history of severe bronchiolitis and to evaluate whether patient-centered outcomes such as quality of life, participation in school and activities, and healthcare use differ after a diagnosis of previously unrecognized lung disease to better understand the significance of pulmonary function abnormalities that have not been diagnosed clinically. Families should be counseled that while we remain uncertain that their child’s PICU admission for bronchiolitis places them at greater risk for clinically meaningful adverse pulmonary outcomes, the likelihood of future wheezing and obstructive lung disease is high and thus warrants ongoing observation for symptoms that may impact their child’s health and well-being.