Biphasic anaphylaxis

I have written about anaphylaxis before (in the form of a video) and also shared a Why We Do What We Do on Epinephrine. Perhaps appropriately so, here is a “delayed” post on biphasic reactions, which the actual occurrence of, or the potential for lead to further ED observation and admissions for children with anaphylaxis.

Generally, a biphasic reaction means that there has been an asymptomatic period of ≥1 hour with a subsequent resumption of symptoms without further antigen exposure. Though it is hard to precisely estimate the incidence Stark followed a group of 25 patients prospectively and Ellis followed 103, both noting rates of about 20%. A larger retrospective review from Alqurashi noted a rate of 14.7%. These were ED based. However, children receiving oral food challenges according to Järvinen in J Allergy Clin Immunol and Lee in Allergy Asthma Proc saw rates much lower (1.5-2%). So the bottom line is that we don’t really know.

In general the second phase is usually less severe than the first. Urticaria is more common, and is many times the only symptom per Ellis. As is the case with many understudied phenomena there are case reports of fatalities as well. Recurrent symptoms generally appear in the first 4-6 hours, but can occur out to 72 hours. However, most will be within 24-30 hours.

With regard to specific risk factors for biphasic reactions in children I wanted to highlight one study that reviewed nearly 500 patients with anaphylaxis  the aforementioned retrospective review from Alqurashi. Overall in their database from two Canadian Pediatric Emergency Departments 71/484 (14.7%) visits developed biphasic reactions. It is interesting that 49/71 (69%) had respiratory and/or cardiovascular manifestations (different than Ellis) and 35/71 (49%) got epinephrine. Alqurashi also identified five independent predictors for biphasic reactions, and if you take anything home from this post I want this to be it:

  • Age 6 to 9 years – OR 3.60 (95% CI 1.5-8.58)
  • Delay in presentation to the ED >90 min after the onset of the initial reaction – OR 2.58 (95% CI 1.47-4.53)
  • Wide pulse pressure at triage – OR 2.92 (95% CI 1.69-5.04)
  • Treatment of the initial reaction with >1 dose of epi – OR 2.7 (95% CI 1.12-6.55)
  • Treated with inhaled β-agonists in the ED – OR 2.39 (95% CI 1.24-4.62)

What should you do with this info? Well, asymptomatic children at 4 to 6 hours can be discharged home. Any child that has developed biphasic reaction or is still symptomatic should not go home. I would strongly consider admitted/observing children with the above five predictors as well.


Alqurashi W, Stiell I, Chan K, Neto G, Alsadoon A, Wells G. Epidemiology and clinical predictors of biphasic reactions in children with anaphylaxis. Ann Allergy Asthma Immunol. 2015 Sep;115(3):217-223

Ellis AK, Day JH. Incidence and characteristics of biphasic anaphylaxis: a prospective evaluation of 103 patients. Ann Allergy Asthma Immunol 2007; 98:64.

Järvinen KM, Amalanayagam S, Shreffler WG, et al. Epinephrine treatment is infrequent and biphasic reactions are rare in food-induced reactions during oral food challenges in children. J Allergy Clin Immunol 2009; 124:1267.

Lee J, Garrett JP, Brown-Whitehorn T, Spergel JM. Biphasic reactions in children undergoing oral food challenges. Allergy Asthma Proc 2013; 34:220.

Stark BJ, Sullivan TJ. Biphasic and protracted anaphylaxis. J Allergy Clin Immunol 1986; 78:76.


Briefs: Henoch-Schönlein Purpura, belly pain and steroids (Re-post)

You see a kid that you’re pretty sure has Henoch-Schönlein Purpura (HSP). OK, you’re certain ’cause they have the rash and EVERYTHING.

This is HSP. Not something else. From wikimedia commons

They aren’t hypertensive, and their urine is normal. They have belly pain, but no currant jelly stools, and they don’t have intussusception because the ultrasound was normal. He is eating and drinking OK, and can walk to the bathroom , despite some pain. But the kid’s belly still hurts. And mom is wondering if there’s something that can be done. She tried acetaminophen and that didn’t work. After consulting Dr. Google she saw something about steroids – now she is asking whether or not they would help her son.

So watch’cha gonna do?

Take NSAIDs says you

Well first, let’s look at some treatment options. For most kids with HSP, the symptoms are relatively mild. Though no RCTs exist on the use of NSAIDs in HSP, they are the first line in therapy. There does not seem to be an increased risk of GI bleeding in gut vasculitis, and thus agents that modulate COX activity should be safe. If a patient with HSP has abdominal or joint pain, and does not have GI bleeding or glomerulonephritis, then I recommend NSAIDs. Namely Naproxen, 10-20 mg/kg/day divided into 2 doses. It is easier to do 2 doses as opposed to . The max of 1500mg per day is safe in older children. If you are going to need to use it for >5-7days I suggest dropping back to a max of 1000mg/day. Ibuprofen would be fine as well, but it requires more frequent dosing, and thus adherence may be an issue.

OK, so mom hasn’t tried that yet, but what about the therapy she is asking about?

Stering [sic] down the Corticosteroid option

Some investigator suggest that NSAIDs may fail to work in cases of HSP and belly pain because of bowel edema and compromised absorption. Their claims would not support the use of steroids either since they, you know, require the gut for absorption too. Nevertheless, there is some evidence to suggest that corticosteroids help with;

  • Length of pain
  • Decreased risk of intussusception
  • Decreased risk of recurrence of pain
  • Decreased risk of kidney involvement
  • Decreased risk of invasive GI interventions (surgery)

This is all far from iron clad however. If you’re going to take a look at one paper check out Weiss et al, Pediatrics, 2007 – a meta-analysis that suggested that corticosteroids reduced the mean time to resolution of pain, and the odds of development of renal disease. On the other hand, a prospective study out of Finland showed no benefit in the risk of onset of renal disease or the subsequent clinical course over 6 months. If you are convinced that steroids would be of benefit the patients most likely to reap said benefits are those with pain severe enough to compromise PO intake, limit their ability to walk, and/or that necessitates admission. You can then give 1/2mg/kg per day up to 60mg/day max of PO prednisone, or 0.8 to 1.6mg/kg/day (max 64mg/day) methylprednisolone in patients that can’t tolerate PO. Depending on how long you continue dosing (there is no solid information on this of course) you may need a protracted taper.

So what about the patient in question?

Well, since he hadn’t received Naproxen  that’s what I would try first. I would also spend extra time discussing the usual clinical course, and the expectations that mom has about her son’s symptoms and their eventual resolution. I would assure that PMD follow up is scheduled, and make a plan to discuss steroids if the trial of NSAIDs fails and/or his symptoms worsen. I would not give steroids ostensibly to prevent renal disease, since the evidence is lacking.

How useful is the pediatric appendicitis score in adolescent females?

You will undoubtedly see a patient with right lower quadrant abdominal pain in the near future. One of the most challenging patient populations to evaluate are adolescent females with RLQ pain. Why? Because of the ovaries of course! Their presence adds to the differential diagnosis and thus, female patients often require more testing to make the diagnosis.

The Pediatric Appendicitis Score (seen in the separate panel) is a prospectively validated tool for use in the assessment of patients with suspected appendicitis.

In a recent edition of Academic Emergency Medicine Scheller et al. attempted to answer the above question. This was a retrospective observational cohort of the female adolescents (13-21 years) present within the data of a prior prospective evaluation of 1,228 Pediatric Emergency Department patients with suspected appendicitis. The aim was to calculate the sensitivity, specificity, PPV and NPV of the PAS for acute appendicitis in the subset of female teens.

Overall the original cohort had a high rate of appendicitis for patients who had all elements of the PAS present – 73% (n=901/1,228) of the cohort had all elements of the PAS with 26.7% (n=249) of the 901 ultimately diagnosed with appendicitis. 272 of the 901 were adolescent females, making up the population for this secondary analysis. Salient take home points from the results are presented in the following table.

PAS ≤2 Low risk – Minimize unnecessary testing

PAS 3-6 Equivocal – Efficient use of diagnostic tests while minimizing unnecessary CT scans

PAS ≥7 High suspicion – Prompt surgical consult and minimize imaging via a consistent team based approach


Goldman et al. Prospective validation of the pediatric appendicitis score. J Pediatr. 2008 Aug;153(2):278-82.

Scheller et al. Utility of Pediatric Appendicitis Score in Female Adolescent Patients. Acad Emerg Med. 2016 Jan 29.

Take Home Points

  • If the PAS is ≥8, which is really high, it is very sensitive for appendicitis in female adolescents. That is a very high PAS generally picks up on teenage females with appendicitis.
  • The PPV is really quite poor no matter the score cutoff
  • This population of teens was derived from another previous study. Thus a dedicated study of teens – ideally a large population from multiple centers is warranted.
  • I will push for the diagnosis of appy (over other pathology) in female teens with a PAS of 8 or higher however. This may not result in decreased use of imaging studies, but I do think, and many agree, that even teenage girls can have a “clinical” diagnosis of appendicitis. So, yes, it is OK to diagnose appendicitis in a teenage girl without doing an ultrasound or CT.

Talking to patients and families about chest pain in the Pediatric Emergency Department

I just came across this fantastic video from the Cincinnati Children’s Hospital Medical Center Blog and Cardiologist Nicolas Madsen. Chest pain is a very common presenting symptom in the Pediatric Emergency Department, and families are often very worried about the heart. To quote many a Cardiologist – with some paraphrasing, “It’s not (usually) the heart.”

The video is only five minutes long, and well worth a view. After watching it I suspect you will be able to take several new things back to the bedside to help make your history taking more robust, and enhance your ability to reassure the family, who is often worried about the worst case scenario.

And remember, red flags for chest pain in otherwise health kids include:

  • Chest pain with activity
  • Chest pain with syncope

In general, short brief episodes that are reproduced by pressing on the chest wall (i.e. reproducible chest pain) are generally reassuring.

You can follow Dr. Madsen on Twitter @MadsenNicolas