Just Added!

New Videos with Amal Mattu, MD

Watch NowGo

900% Greater Odds of Going Home – Nasal Fentanyl for Pediatric Sickle Cell Crisis

November 17, 2023

Written by Amanda Mathews

Spoon Feed
In this large, multi-center retrospective study of academic pediatric emergency departments across the US and Canada, researchers found that children presenting with sickle cell disease vaso-occlusive events had nine times (900%) greater adjusted odds of discharge home if given intranasal fentanyl (INF).

Why wait when you can go IN?
This retrospective study was conducted across 20 academic pediatric emergency departments in the US and Canada. Patients aged 3-21 who were treated for sickle cell pain were included even if they had a fever upon presentation. Patients diagnosed with acute chest syndrome were excluded. 400 patients were ultimately included, and the primary outcome of interest was discharge home from the ED.

Children who received INF received higher overall total parenteral opioid morphine equivalents; however, there was no difference in the mean total IV opioid morphine equivalents administered to those who received INF and those who did not (0.2 mg/kg in both groups). Children who received INF were more likely to receive their first parenteral opioid within 30 minutes of presentation and within 60 minutes of presentation compared to children who did not.

Children who received INF had a nearly nine-fold greater adjusted odds of discharge from the ED. There was also three-fold greater adjusted odds of discharge for children who received oral opioids.  Median length of stay was three hours in both groups. Additionally, children who presented with lower initial pain scores, who had larger reduction in their pain scores during their ED visits, and those who received a lower overall morphine equivalent dosage of opioids had higher odds of discharge home.

How will this change my practice?
I will be advocating for the use of intranasal fentanyl as a first line analgesia agent for pediatric patients presenting with sickle cell pain crisis in my institution. Not only does this paper make the case that this improves dispositions to home, but it’s also the right thing to do for a population of patients who are experiencing significant pain.

A note from the lead author Chris Rees:
“The magnitude of the benefit of intranasal fentanyl on discharge to home, even when controlling for pain scores and other potentially contributory factors, was astounding. In medicine, we are often impressed by 50% to 60% improvements, but here we saw more than a 900% greater likelihood of being discharged to home with intranasal fentanyl.”

An additional note from senior author Claudia Morris:
“One caution for physicians and nurses. INF was meant as a bridge to IV medications and was not meant to replace IV analgesics in moderate-to-severe sickle cell pain.  Since INF is short acting, the clinical team should utilize INF to provide rapid pain relief but follow it up with IV opioids to avoid a rapid return of pain when it wears off.”

Editor’s note: Many thanks to Dr. Jason Woods for pointing out this important article and connecting us with the authors for these quotes! ~Clay Smith

Editor’s note 2: Within sites that routinely give INF, children who received it were 4,000% more likely to be discharged home from the ED. ~Nick Zelt

Intranasal fentanyl and discharge from the emergency department among children with sickle cell disease and vaso-occlusive pain: A multicenter pediatric emergency medicine perspective. Am J Hematol. 2023 Apr;98(4):620-627. doi: 10.1002/ajh.26837. Epub 2023 Feb 6.

3 thoughts on “900% Greater Odds of Going Home – Nasal Fentanyl for Pediatric Sickle Cell Crisis

  • Are we not worried about the juxtaposition about what this study actually posits (an association between IN fentanyl and discharge) and the conclusions drawn by the authors/summary (IN fentanyl leads to discharge)? There appears to be a misunderstanding of the limitations of retrospective cohort based research in addition to blatant ignorance of confounding factors in groups that are both limited in size and clearly subject to confounding (lower pain scores at presentation, etc).

    I think the conclusions, both those of the authors and of the reviewer/editors, should be changed to reflect what this study actually ‘proves’ and what it does not. A better conclusion would be, “Future randomized trials to determine the effectiveness of IN fentanyl on discharge versus admission rates are warranted based on this association”

    • No misunderstanding. And also no assertion of proof. So, the take home for you is to wait for a RCT? Of course, it could be confounded. Adjustments were made for known confounders, but there is no way to prove this works given the study design. Yet, it’s compelling with such a large effect. It would have to be a whopper of a confounder! You’re welcome to wait for a RCT. But is there a downside to initiating rapid pain control IN? I mean, why would we not?

  • I don’t disagree fundamentally with the conception/implementation of early parental analgesia for children (or adults) presenting with acute severe pain of any etiology and not to a particular disease process.

    What I do disagree with fundamentally are the logic concessions needed to state that this article, which is merely a retrospective crunch of numbers from a database, does anything more that suggest their could be an association between the intervention and the outcome. Two glaring points would be the effect size without a described mechanism of action (there is no difference between admission/discharge in patients receiving all ‘parenteral’ opioids within 60 minutes of presentation – what about the administration of IN fentanyl vs IV fentanyl/morphine would result in this discrepancy if the timing of medication administration is fixed?) and the small sample size included in the analysis (only 38 patients in the discharge group received IN fentanyl). If these same patients were given a turkey sandwich as a po challenge prior to discharge, we could, by the same analytical methods find that turkey sandwiches are associated with a significant increase in hospital discharge. Also drawing population level conclusions from such a small sample seems cavalier at best.

    Because many may not statistically inclined, we won’t even get into the fact that there should have been a correction for level of significance based on the number of factors examined to determine association with discharge (even I think the Bonferroni correction is a bit conservative).

    What is clear is that patients who have greater pain reduction and require lower total morphine equivalent units are more likely to be discharged home (though this doesn’t make for a very interesting paper).

    And then I would ask, to your first points, if the assertions by JF editors or the authors seem to drive home the context of association vs cause and effect:

    A note from the lead author Chris Rees:
    “The magnitude of the benefit of intranasal fentanyl on discharge to home…”

    How will this change my practice?
    Not only does this paper make the case that this improves dispositions to home…

    I appreciate what your service provides for clinicians who may lack the time to review a large number of relevant manuscripts in full detail, but I would caution oversights in interpretation and methodology and strongly worded recommendations for those studies that do not reach a certain level of scientific rigor and are not designed to determine cause and effect. RCTs are not the only answer to our questions, but may shed additional light on this topic.

    Until then, please treat all of your patients’ pain appropriately, whether or not it leads to hospital discharge.

What are your thoughts?