Opinion

EDITORIAL

Bronchiolitis and Pulse Oximetry

Choosing Wisely With a Technological Pandora’s Box
Lalit Bajaj, MD, MPH; Joseph J. Zorc, MD, MSCE

In this issue of JAMA Pediatrics, Principi et al1 report the sician perceptions about the importance of mild hypoxemia
findings of their study titled, “Effect of Desaturations on are driving hospitalizations that may not be necessary. This con-
Subsequent Medical Visits in Infants Discharged From the cern is not limited to the admission decision. Unger and
Emergency Department With Cunningham5 also found that patients stay in the hospital much
Bronchiolitis.” The investiga- past the resolution of other issues such as hydration status,
Related article tors prospectively enrolled solely due to perceived need for supplemental oxygen. Some
infants with bronchiolitis have used strategies such as home oxygen administration for
deemed suitable for discharge and monitored them with con- otherwise healthy infants with bronchiolitis to decrease ad-
tinuous pulse oximetry at home for the first day. The essen- mission rates and have demonstrated a strong safety profile.6,7
tial element of this study is that the pulse oximeter had deac- Hospital admission is not a benign intervention, with errors
tivated threshold alarms and did not display saturation values. unfortunately being common. Patients hospitalized for hy-
The monitors, in addition to study diaries, were collected and poxemia may also be subjected to a cascade of unwarranted
analyzed, and the patients received follow-up calls at 72 hours. interventions (ie, the chest radiograph that leads to the diag-
The results of this study call into question the assumptions of nosis of “pneumonia” and then leads to antibiotics, etc) as well
current practice around use of pulse oximetry for decision mak- as the costs associated with constant adjustment of oxygen flow
ing in bronchiolitis, but they likely are not surprising to expe- rates as saturations vary. The most recent guidelines from the
rienced clinicians. Principi and colleagues found desatura- American Academy of Pediatrics recommend that patients with
tions to be a common event after discharge from the emergency an oxygen saturation of 90% or greater need not receive oxy-
department, with two-thirds of the infants having at least 1 de- gen supplementation and that most patients do not require
saturation episode at home and many having sustained de- continuous pulse oximetry.8 McCulloh et al9 recently studied
saturations to 70% or less. The primary outcome of rate of un- this in inpatients and found that there was no effect on length
scheduled visits was the same in both groups, at approximately of stay or escalation of care in patients with intermittent moni-
25%, which included unscheduled visits to the primary care toring instead of continuous pulse oximetry.
physician and the emergency department. There was also no There are a number of limitations to the study by Principi
difference in hospitalizations between those who had desatu- and colleagues that are important to consider. This is a rela-
ration episodes and those who did not. The study places the tively small single-center study performed in Canada, where
issue of transient desaturations and their clinical importance universal health care access likely increases reassurance that
at the forefront of the discussion around management of these patients have reliable clinical follow-up. While clinically
patients. In addition, it adds to the dilemma of which pa- meaningful, unscheduled returns at 72 hours after discharge
tients should receive pulse oximetry in their evaluation and may not be the most important clinical outcome measure to
how to interpret the values. assess the effect of desaturations. Some may raise concerns
Bronchiolitis has been a focus for quality improvement and about the effect of transient hypoxemia on long-term cogni-
clinical practice guideline work since Shay et al2 reported in tive development, particularly for children with cardiac and
1999 that bronchiolitis admissions had skyrocketed during the other chronic cardiopulmonary conditions. 10 However,
preceding decades. Many have implicated the use of the pulse chronic hypoxemia is a very different situation from transient
oximeter and the choice of somewhat arbitrary “hypoxemia” desaturations during an acute illness, and those findings
cutoffs as the primary driver of this phenomenon. In 2003, should not be generalized to otherwise healthy children with
Mallory et al3 tested this hypothesis in a survey of pediatric bronchiolitis. As noted in the American Academy of Pediat-
emergency medicine physicians and found that a 2% differ- rics guideline, hypoxemia well below an oxygen saturation of
ence in oxygen saturation in simulated case scenarios was a 90% occurs regularly in healthy infants and is also associated
large driver in the decision to admit. In a recent issue of JAMA, with travel to moderate altitude. The results in the bronchiol-
Schuh et al4 published a trial that randomized patients to either itis study by Principi and colleagues suggest that it is also
display their true oxygen saturation levels to physicians or to common during this infection, which affects more than a
substitute artificially altered saturations (3% higher). The pa- third of infants. Given the apparent ubiquity of intermittent
tients with the altered saturations displayed were less likely hypoxemia during common events of childhood, the long-
to be admitted, and there was no effect on the rate of revisits term impact would be difficult to study and even more chal-
after discharge. These studies add to the hypothesis that phy- lenging to prevent.

jamapediatrics.com (Reprinted) JAMA Pediatrics Published online February 29, 2016 E1

Copyright 2016 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Florida Atlantic University User on 03/01/2016

 Opinion Editorial

Pulse oximetry has undoubtedly contributed to im- with apparent negative outcomes. Incorporating this infor-
proved quality and safety of pediatric care, as these boxes have mation into clinical practice will require health care profes-
become a fixture at virtually every hospital bedside during re- sionals to take a more judicious approach to the use of pulse
cent decades. For bronchiolitis, however, some may view the oximetry in the evaluation of the patient. The American Acad-
oximeter as a Pandora’s box that was opened before the re- emy of Pediatrics guidelines serve as a good start, but more
search had been done to appropriately interpret this stream work is needed to rationalize the use of this important but over-
of data. This has led to arbitrary thresholds for oxygen imple- emphasized technology.
mentation and widespread use of continuous pulse oxim- Health care professionals and parents are obviously frus-
etry. As with many areas where data are unclear, use of pulse trated in caring for patients with bronchiolitis. The sheer vol-
oximetry may reinforce tendencies of clinicians. For those look- ume strains the system, and there are no proven therapies, no
ing for a reason to admit, continuous pulse oximetry can pro- helpful predictive models, and no easy objective measures of
vide an excuse when an inevitable desaturation occurs. For oth- severity. The evidence points to a clinical evaluation that in-
ers, oximetry can be used as an indication for outpatient corporates oxygen saturation into the decision making but does
interventions such as home oxygen use. With this study, we not absolutely determine disposition. The time has come to
now have a clearer view that the transient hypoxemia events stop focusing on the numbers on Pandora’s box and to de-
these interventions are intended to prevent are likely occur- velop strategies to thoughtfully use the data it provides us in
ring in many infants with bronchiolitis and are not associated the overall clinical care of the patient.

ARTICLE INFORMATION from the emergency department with bronchiolitis department for acute bronchiolitis. Pediatrics.
Author Affiliations: Department of Pediatrics, [published online February 29, 2016]. JAMA Pediatr. 2006;117(3):633-640.
Section of Emergency Medicine, University of doi:10.1001/jamapediatrics.2016.0114. 7. Halstead S, Roosevelt G, Deakyne S, Bajaj L.
Colorado School of Medicine/Children’s Hospital 2. Shay DK, Holman RC, Newman RD, Liu LL, Stout Discharged on supplemental oxygen from an
Colorado, Aurora (Bajaj); Department of Pediatrics, JW, Anderson LJ. Bronchiolitis-associated emergency department in patients with
The Children’s Hospital of Philadelphia, hospitalizations among US children, 1980-1996. bronchiolitis. Pediatrics. 2012;129(3):e605-e610.
Philadelphia, Pennsylvania (Zorc); University of JAMA. 1999;282(15):1440-1446. 8. Ralston SL, Lieberthal AS, Meissner HC, et al;
Pennsylvania, Philadelphia (Zorc). 3. Mallory MD, Shay DK, Garrett J, Bordley WC. American Academy of Pediatrics. Clinical practice
Corresponding Author: Lalit Bajaj, MD, MPH, Bronchiolitis management preferences and the guideline: the diagnosis, management, and
Department of Pediatrics, Section of Emergency influence of pulse oximetry and respiratory rate on prevention of bronchiolitis. Pediatrics. 2014;134(5):
Medicine, University of Colorado/Children’s the decision to admit. Pediatrics. 2003;111(1):e45-e51. e1474-e1502.
Hospital Colorado, 13123 E 16th Ave, B 251, Aurora, 4. Schuh S, Freedman S, Coates A, et al. Effect 9. McCulloh R, Koster M, Ralston S, et al. Use of
CO 80045 (lalit.bajaj@childrenscolorado.org). of oximetry on hospitalization in bronchiolitis: intermittent vs continuous pulse oximetry for
Published Online: February 29, 2016. a randomized clinical trial. JAMA. 2014;312(7): nonhypoxemic infants and young children
doi:10.1001/jamapediatrics.2016.0090. 712-718. hospitalized for bronchiolitis: a randomized clinical
Conflict of Interest Disclosures: None reported. 5. Unger S, Cunningham S. Effect of oxygen trial. JAMA Pediatr. 2015;169(10):898-904.
supplementation on length of stay for infants 10. Bass JL, Corwin M, Gozal D, et al. The effect of
REFERENCES hospitalized with acute viral bronchiolitis. Pediatrics. chronic or intermittent hypoxia on cognition in
2008;121(3):470-475. childhood: a review of the evidence. Pediatrics.
1. Principi T, Coates AL, Parkin PC, Stephens D, 2004;114(3):805-816.
DaSilva Z, Schuh S. Effect of oxygen desaturations 6. Bajaj L, Turner CG, Bothner J. A randomized trial
on subsequent medical visits in infants discharged of home oxygen therapy from the emergency

E2 JAMA Pediatrics Published online February 29, 2016 (Reprinted) jamapediatrics.com

Copyright 2016 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Florida Atlantic University User on 03/01/2016