Date: January 15th, 2018

Reference: Sadeghirad B, et al. Corticosteroids for treatment of sore throat: systematic review and meta-analysis of randomised trials. BMJ 2017

Guest Skeptic: Meghan Groth is an Emergency Medicine Pharmacist at the UMass Memorial Medical Center in Worcester, Massachusetts. She has contributed to the Academic Life in Emergency Medicine and EM PharmD blogs, and is a part of the ALiEM Capsules Team. 

Case: 50-year-old man presents with a one day history of sore throat, cough and low-grade fever. He is otherwise healthy with only sports related injuries. The ibuprofen did not help and he is requesting antibiotics so he can get back to work sooner.

Background: Patients present commonly to their primary care providers (PCPs) and to the emergency department (ED) with complaints of a sore throat. In the US, adults accounted for 6.6 million visits annually to PCPs and EDs for sore throat.

Along with this can come unnecessary prescriptions for antibiotics (a topic in itself), even though national guidelines recommend against routine antibiotics (NICE and ESCMID).

As more and more attention is paid to strategies for reducing the overutilization of antibiotics and the subsequent trends in antimicrobial resistance, there is a need for other strategies to provide symptomatic relief for these patients.

Acetaminophen and NSAIDs are often used to treat sore throats. While providing symptomatic relief they can also cause harm. Several studies have looked at corticosteroids and demonstrated some moderate benefit.

Corticosteroids exert an anti-inflammatory effect by inhibiting transcription of pro-inflammatory mediators in airway endothelial cells. However, previous systematic review and meta-analyses have been confounded by co-administration of antibiotics and analgesics, as well as variability in measures of efficacy among studies.

Clinical Question: Are corticosteroids effective and safe as an adjunct treatment for sore throat in addition to standard care compared with standard care alone?

Reference: Sadeghirad B, et al. Corticosteroids for treatment of sore throat: systematic review and meta-analysis of randomised trials. BMJ 2017

  • Population: Adults and or children aged five and over presenting to the emergency department or primary care settings with a clinical syndrome of sore throat (painful throat, odynophagia, or pharyngitis)
    • Exclusions: Participants admitted to the hospital, immunocompromised, those with infectious mononucleosis, sore throat after surgery/intubation, GERD, croup, peritonsillar abscess, or subjects under five years of age
  • Intervention: At least one dose of corticosteroid therapy
  • Comparison: “Standard of care” or placebo
  • Outcomes: 
    • Complete resolution of symptoms at 24 hours and at 48 hours
    • Mean time to onset of pain relief and to complete resolution of pain
    • Absolute reduction of pain at 24 hours
    • Duration of bad/non-tolerable symptoms
    • Recurrence/relapse of symptoms
    • Days missed from school/work
    • Need for antibiotics
    • Rate of adverse events related to treatment

Authors’ Conclusions: “Single low dose corticosteroids can provide pain relief in patients with sore throat, with no increase in serious adverse effects. Included trials did not assess the potential risks of larger cumulative doses in patients with recurrent episodes of acute sore throat.”

checklistQuality Checklist for Therapeutic Systematic Reviews:

  1. The clinical question is sensible and answerable. Yes
  2. The search for studies was detailed and exhaustive. Yes
  3. The primary studies were of high methodological quality. N0
  4. The assessment of studies were reproducible. Yes
  5. The outcomes were clinically relevant. Yes
  6. There was low statistical heterogeneity for the primary outcomes. No
  7. The treatment effect was large enough and precise enough to be clinically significant. Unsure

Key Results: The authors ended up including ten randomized controled trials with 1,426 patients in the analysis. Eight of these enrolled ED patients, and two enrolled patients from the primary care setting. Three studies evaluated children, six focused on adults, and one study enrolled both children and adults. Dexamethasone was the most common corticosteroid used in the trials, however, the doses and routes (both oral and IM) varied.

Corticosteroids provided significant benefit

Table Dex throat

  • Missed Work: One of the included studies reported that 55% of patients treated with corticosteroids and 69% taking placebo took time off work because of sore throat (RR 0.8 [95% CI 0.6 to 1.1]). Another study reported that adult patients treated with corticosteroids missed less days compared to placebo (mean difference −0.3 days [95% CI −0.87 to 0.27]). Both results cross the line of no difference. None of the trials reported duration of bad/ non-tolerable symptoms.
  • Adverse Events: There was no reported increase in the incidence of adverse effects with steroid administration versus placebo.

Screen Shot 2015-04-25 at 3.11.12 PM

1. Relationship Between “Standard of Care” and Steroids: One study included in this meta-analysis did not report whether or not subjects received adjunct antibiotics and/or analgesics. For a meta-analysis focusing on the clinical question stated, it’s difficult to understand why this study was included when the standard of care is so ill-defined.

For all the discussion on the confounding effects of antibiotics, and the rationale for the prescribing of corticosteroids to potentially decrease the overutilization of antibiotics in pharyngitis, I wish these points could have been addressed in a more rigorous manner in this meta-analysis. The authors mentioned that the impetus for performing this study was the TOAST trial (Hayward et al. Effect of oral dexamethasone without immediate antibiotics vs placebo on acute sore throat in adults. JAMA 2017;317:1535-43), as a few systematic reviews on this topic have already been published, with varying conclusions. The TOAST trial was supposed to address the issue of effectiveness and safety of steroid monotherapy for sore throat, but nearly 40% of all patients in that trial received a prescription for antibiotics. It wasn’t reported how many patients actually ended up taking the treatment course of those antibiotics. What would be really nice is to have a full data set from all of these trials regarding who got antibiotics and who didn’t, and to do a subgroup analysis on the pooled data to see what effects steroids by themselves had. The last piece to mention about this point is the relationship between the decision to prescribe antibiotics and the decision to prescribe corticosteroids. The TOAST trial, included in this meta-analysis, attempted to get at this, but it’s a bit complicated. Do you give a patient a corticosteroid as an alternative to antibiotics because you feel like they’re expecting you to do something more than tell them to stay hydrated and take over the counter analgesics? Or, if you feel really bad for the patient, do you prescribe both? Is your decision to prescribe corticosteroids independent or your decision to prescribe antibiotics? If not, how are they related? I’m interested in your thoughts on this. Please leave comments on the blog or twitter with what you think.

2. Patient Oriented Outcome: What is the most important thing to patients? Is it onset of pain relief, time to complete relief, pain relief at 24hrs, etc? From my perspective, I want to be better sooner, with enough improvement to resume normal activities and have a shortened duration of illness. Other people might value other outcomes. I do not think we should be pushing a complete resolution in 24-48hrs as a goal. We can cause some serious harm in overtreatment in trying to achieve such a goal. In addition, just because it was statistically significant does not mean clinical significance. Does it matter if the onset of pain relief is 5hrs sooner or the time to complete resolution is 11hrs shorter?

3. Steroid Dosing Equivalents: The authors referred to a single, low dose of corticosteroids in their conclusion, but is this really reflective of the studies they included? Two of the ten studies used steroids for a two-day duration, the remainder used a single dose. Most studies used dexamethasone, but some used oral and some used the intramuscular route. It’s helpful to take a step back and compare some of these doses though, because I don’t think they’re all apples and apples. The three studies that focused on children used oral dexamethasone at a dose of 0.6 mg/kg, with a max of 10 mg. Let’s think about that for a second. An 18kg kid in this study would get 10 mg of oral decadron, that’s the same “low dose” that’s referred to for two of the adult studies where “grown-ups” were also given 10 mg of oral dexamethasone. It’s important to note that there’s wide variability in what some practitioners refer to as low dose and high dose steroid regimens, but a publication in the Annals of Rheumatologic Diseases in 2002 from an expert panel of rheumatologists (Ann Rheum Dis 2002;61:718–722) seeking to clear up the nomenclature around steroid regimens defined low dose as 7.5 mg per day of prednisone equivalents (that’s just over 1 mg per day of dexamethasone). The 10 mg of dexamethasone used in many of the trials in this meta-analysis equates to about 67 mg daily prednisone equivalent, or well into the high dose steroid classification. You may say it’s only a single dose, so you could also refer to it as pulse therapy. But it’s hard to figure out how the authors determined these steroid regimens could be called “low-dose.”

4. Difference in Pain Scores: In the authors’ conclusion, they state that corticosteroids can provide pain relief in sore throat. It’s interesting to look back at the previous evaluations that have evaluated this intervention and how each has chosen their primary intervention. In 2010 Wing and colleagues chose to focus on the time to “clinically meaningful” pain relief, and found a 4.5 hour difference between steroids and control. We’ll come back to that phrase “clinically meaningful” in a second. The Cochrane review published in 2012 concluded that steroids were beneficial by instead focusing on the time to complete pain relief, which resulted in a 14.4 hour difference between steroids and the control group. The present analysis arrives to their conclusion through a few different measures. Five trials reported complete resolution of symptoms at 24 hours, with the evidence favoring corticosteroids, and four trials reported resolution of pain at 48 hours, which again favored steroids. However, of the eight studies that reported time to onset of pain relief, there was only a 4.8 hour difference between groups. Time to complete resolution in this meta-analysis resulted in an 11 hour advantage of corticosteroids over control. And finally, when the VAS was measured at baseline and 24 hours, those patients who received steroids showed a 1.3 point lower pain score than those treated with placebo at 24 hours – keep in mind that 1.3 points on a VAS ranging from 0 to 10 is the absolute minimum difference you need in order to claim a clinically significant difference or reduction in pain. This seems like a whole bunch of different ways to measure the effectiveness of the same intervention.

5. Comments on Safety: The authors stated in their conclusion that single “low dose” corticosteroids can provide pain relief in sore throat with no increase in serious adverse effects. We already talked about what they’ve referred to as “low dose,” but I’m not confident in this meta-analysis being able to comment as strongly on the safety of the intervention as the authors apparently seem to be. There were two of the ten studies they included that didn’t report evaluating adverse effects, and so they were documented as “none reported.” If they weren’t explicitly evaluated, I’m not sure it’s appropriate to assume the intervention was safe. Furthermore, the studies included had primary outcomes to determine efficacy, and likely weren’t powered to detect differences in what I would surmise are the uncommon occurrences of adverse events.

Comment on Authors’ Conclusion Compared to SGEM Conclusion: It appears that a short course of steroids may help with some measurements of pain relief in patients with sore throat – depending on how you’re evaluating this, though it’s unclear whether these effects are due to steroids or other interventions (e.g. analgesics, antibiotics). Additionally, the studies weren’t rigorous enough to definitively comment on safety. The authors conclusion may have overstepped a bit in their claims that a single, low dose provides pain relief without increasing adverse effects.

SGEM Bottom Line: Steroids appear to provide a modest benefit to patients presenting to the emergency department with a sore throat.

Case Resolution: Using some shared decision making you discuss the evidence for different treatments. This include over the counter analgesics and NSAIDs, not needing antibiotics and considering corticosteroids. He decides to try 10mg of dexamethasone.

Meghan Groth

Meghan Groth

Clinical Application: I am concerned that corticosteroids are going to be used indiscriminately for sore throats. Given the potential side effects, cumulative doses and the use in children, I am going to be cautious. This means I will not be routinely giving corticosteroids for sore throats.

What Do I Tell My Patient? You have a sore throat and it is most likely from a virus. Taking acetaminophen or ibuprofen can help. Antibiotics do not treat viruses and can lead to diarrhea and an allergic reaction. We could try one dose of a steroid medication. There is some evidence it can shorten your illness and get you feeling better quicker.

Keener Kontest: Last weeks’ winner was Dave Lemonick from Pittsburgh, PA. He knew two Swedish chemists, Nils Löfgren and Bengt Lundqvist, discovered the compound we know today as “lidocaine” in 1942. It was clinically introduced by Dr. Torsten Gordh (1907-2010), who was the first physician in Sweden to specialize in anesthesiology.

Listen to the podcast on iTunes for this weeks’ keener question. If you think you know the answer then send an email to TheSGEM@gmail.com with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

Please consider following the SGEM on Twitter and liking the SGEM on Facebook. If you are feeling really appreciative, give the SGEM a five star rating on iTunes.

FOAM logoOther FOAMed Resources:

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.


SkiBEEM 2018