- based on Cochrane review of trials with methodologic limitations
- systematic review of 5 randomized trials comparing oral corticosteroids vs. placebo or standard care (control) in 1,193 patients with acute sinusitis
- 4 trials evaluated addition of oral corticosteroids to antibiotics, 1 trial evaluated oral corticosteroid monotherapy
- all patients in included trials were adults
- all trials evaluating addition of oral corticosteroids to antibiotics had unclear allocation concealment
- addition of oral corticosteroids (prednisone, betamethasone) to antibiotics associated with increased resolution or improvement in symptoms at 3-7 days in analysis of 4 trials with 869 patients
- risk ratio 1.4 (95% CI 1.08-1.81)
- NNT 3-24 with resolved or improved symptoms at 3-7 days in 54% of control group
- similar benefits at 4-14 days
- no significant difference in resolution or improvement of symptoms up to 14 days comparing prednisolone 30 mg/day orally vs. placebo in 1 trial with 185 patients (summarized below)
- Reference – Cochrane Database Syst Rev 2014 Mar 25;(3):CD008115
- fever or pain: APAP and NSAIDs
- symptom severity or duration: OTC combos (in children > 4 yo, adolescents and adults); heated humidified air; liquid solution of Perlargonium sidoides (Umcka, and others); zinc lozenges may decrease duration but zinc has side-effects.
- cough: ACCP recommends brompheniramine/pseudoephedrine and inhaled ipratropium, but against cough suppressants. Honey is harmless (if you’re >1yo).
- rhinorrhea: ipratropium nasal spray improves runny nose but not congestion.
- DO NOT USE: antibiotics, nasal steroids, vitamin C, Echinacea, antihistamines alone.
- Again, Perlargonium sidoides (Umcka ColdCare, et al) seems to shorten duration of illness.
- Do not prescribe cough suppressants for children < 4 yo.
- Duration of cough does not predict pneumonia!
- Do NOT prescribe antibiotics UNLESS: abnormal lung exam, fever, productive cough (i.e., clinical suspicion of pneumonia).
- dextromethorphan has better evidence than codeine (though it’s metabolized into codeine!)
- 2 doses given ≥ 4 weeks apart needed for children during their 1st season of vaccination to optimize immune response
- as vaccine composition has changed this year, children should only have 1 dose if they had ≥ 2 doses of trivalent or quadrivalent vaccination in prior seasons
- children without ≥ 2 doses prior to July 1, 2016, should have 2 doses ≥ 4 weeks apart
- give 1 dose for persons aged > 9 years
Benefits in NNT
- None were helped (preventing an influenza-like illness)
- 1 in 36 were helped (preventing 1 culture-positive influenza case)
- 1 in 83 were helped (preventing pneumonia)
- 1 in 36 were helped (protected from getting influenza when taken prophylactically)
Harms in NNH
- 1 in 20 were harmed (nausea)
- 1 in 17 were harmed (vomiting)
Caveats: Another group published an independent review of the same data with significantly different results. This group was sponsored by the Multiparty Group for Advice on Science (MUGAS) and received pharmaceutical industry funds.6 They also found a 17h reduction in symptom relief; however their interpretation of the industry data concluded a more significant decrease in pneumonia diagnosis as well as a significant reduction in hospitalization. Each team looked at the same clinical trial report data (from original industry documents rather than peer reviewed publications) and came away with different numbers. The MUGAS group notably found several fewer hospitalizations in many studies, mostly small differences (e.g. 6/199 hospitalizations (MUGAS) vs. 9/199 hospitalizations (Cochrane) in the WV15812 trial) that in sum led to a statistically significant treatment effect. This review, however, lacked methodological transparency, failing to report a study protocol and precise outcome definitions (methods commonly used to reduce interpretive bias).
Moreover, a third review of the same data by an independent group without financial conflicts reached the same conclusions as the Cochrane team.7Given the apparently dueling interpretations of data with statistically and clinically meaningful differences, we are forced to make a best-guess judgment: We at TheNNT.com feel the Cochrane review offers a more reliable, transparent interpretation, hence our decision to report their conclusions. We come to this conclusion largely based on their near perfect agreement with another independent, financially unconflicted study group, and based on the lack of methodology transparency of the MUGAS group compared to the Cochrane group.
Finally, an important and unanswered question remains: Do very sick, typically hospitalized patients with influenza benefit from oseltamivir? There is no trial data to help us here. Many including the CDC have cited a recent meta-analysis of observational studies as evidence of mortality benefit.8However the medium to poor quality of these data, their observational nature, the clear evidence of unadjusted biases,9 and the lack of hospitalization benefit in existing outpatient RCT’s all point to an urgent need for high quality RCTs of NI’s in severely ill hospitalized persons. There is certainly adequate substrate for such an effort, as tens of thousands of individuals die in the United States each year due to influenza infection, and it is unclear whether benefits or harms of NI therapy would predominate in such a group.
The long and complex story of oseltamivir, the Cochrane Acute Respiratory Infections Group attempt to study it, multi-billion dollar expenditures to stockpile NI’s, and the ongoing exuberance of public health authorities all continue to merit answers to important questions about the acceptance of new drugs into medical guidelines.10
The lack of an effective alternative option and the decision of influential organizations such as the CDC to conclude a mortality benefit based on unreliable evidence have led to the common perception of oseltamivir as a standard for the treatment of influenza. We have chosen to label NI’s ‘Red’ for symptom relief as we feel the benefit is minimal and not an important enough magnitude when balanced with the harms (also minimal). We have labeled NI’s ‘Yellow’ (further data required) for reduction in death or significant morbidity (pneumonia, hospitalization) and for the prevention of transmission of disease. It seems clear to us that high quality studies, performed transparently by a financially non-conflicted study group, have the potential to answer critical questions of public health and scientific concern. For the meantime the use of NI’s should be left to the discretion of the treating (and hopefully informed) physician and patient.