Widespread Adoption of Low-Value Therapy: The Case of Bronchiolitis and High-Flow Oxygen

Bronchiolitis is one of the most common reasons for pediatric hospitalization in the United States, incurring costs of $734 million each year.¹ Supportive care is the mainstay of treatment, with numerous studies demonstrating the futility of adjunct therapies including corticosteroids, β-agonists, epinephrine, and hypertonic saline. In this issue of Pediatrics, Kalburgi and Halley² describe patterns of adoption of high-flow nasal cannula (HFNC) therapy in general pediatric units in the United States and Canada, illustrating wide uptake of another intervention for bronchiolitis that is not supported by evidence.

Surveying 114 hospitals that participate in the Pediatric Research in Inpatient Settings Network, Kalburgi and Halley² found that 48% of hospitals use HFNC therapy for bronchiolitis on the general pediatric ward. At three-quarters of hospitals using HFNC therapy, locally developed protocols guided its use. These guidelines are heterogeneous and vary in their scope, with only 68% including criteria for HFNC initiation, 57% including criteria for HFNC discontinuation, and less than one-quarter including guidelines for feeding and/or nutritional support concurrent with HFNC administration. A salient concept not addressed in this survey is hospitals’ criteria for HFNC therapy initiation, which would allow us to differentiate between hospitals using this therapy as a first-line treatment versus a rescue therapy after treatment failure with standard oxygen therapy. This is an important distinction and area for future work.

Kalburgi and Halley² demonstrate that HFNC therapy for bronchiolitis has proliferated over the last decade despite a conspicuous lack of high-quality evidence regarding its use outside of intensive care settings. However, the last 2 years have seen the publication of 3 randomized trials in which authors compare HFNC to standard therapy, as well as 3 systematic reviews.^3–8 The first 2 trials, conducted in Australia and New Zealand, pose some challenges in interpretation because they allowed for crossover from the standard therapy group to HFNC therapy.^3,4 Yet the results of both trials were remarkably consistent, with no differences between treatment groups in the duration of oxygen therapy, length of hospital stay, or risk of transfer to an ICU. Although the authors of these studies did find that patients on standard therapy met study-defined treatment failure criteria at a higher rate than those on HFNC, “treatment failure” in the standard therapy group resulted in initiation of HFNC therapy. Thus, a reasonable interpretation of both studies’ findings would be that the authors actually evaluated early versus late initiation of HFNC and found no significant differences between these 2 strategies. Most recently, in a French multicenter trial evaluating HFNC versus standard therapy using a strategy that did not allow crossover, researchers also found no meaningful differences between treatment groups.⁵ Finally, authors of 3 recent systematic reviews on this topic all reached similar conclusions: HFNC is a safe rescue therapy for infants who have failed standard oxygen therapy, but its use as a first-line therapy is not supported by evidence.^6–8 Beyond clinical utility, equipment costs for HFNC therapy have been estimated to be 16 times higher than those of standard oxygen therapy, making early use of this therapy even more challenging to justify.⁴

How did we get here, with almost 50% of pediatric inpatient services that participate in the Pediatric Research in Inpatient Settings Network using this low-value therapy? The most recent US national clinical practice guideline (CPG) on bronchiolitis management, published in 2014, states that “the absence of any completed randomized trial of the efficacy of high-flow nasal cannula in bronchiolitis precludes specific recommendations on it use…”⁹ Despite this, the work by Kalburgi and Halley² suggests that hospital-specific clinical protocols may play a significant role in HFNC’s adoption, particularly at children’s hospitals. CPGs, as defined by the National Academy of Medicine, include recommendations informed by a systematic review of evidence and an assessment of the relative benefits and harms of alternative treatment options.¹⁰ Clinical protocols, in turn, are endorsed by the National Academy of Medicine as a mechanism to translate CPGs into practice. The widespread implementation of clinical protocols in the absence of CPGs is not unusual in pediatrics given a paucity of evidence or CPGs for many clinical problems. Yet the proliferation of HFNC therapy protocols in the absence of CPGs or risk/benefit assessment may not be bending the value curve in the right direction.

In the quest to “do something” for infants with bronchiolitis, the use of albuterol, epinephrine, corticosteroids, and hypertonic saline therapies has followed similar patterns: widespread adoption, accumulation of evidence indicating ineffectiveness, CPGs recommending against their use, and therapy deimplementation. Kalburgi and Halley² have shown us that HFNC therapy has boomed, supported by local clinical protocols. In light of the growing evidence of this therapy’s low value as a first-line intervention, the time has come to revise these protocols and deimplement this costly yet common therapy.

• Copyright © 2020 by the American Academy of Pediatrics