* It does not decrease hospitalizations in patients with influenza.
* It does not decrease complications of influenza.
* It makes patients vomit. (NNTH=22 in adults, 19 in kids.)
* It messes with the mind. (NNTH=94 for neuropsychiatric adverse events.)
* It hurts the head. (NNTH=32 for headaches.)
* It can kick the kidneys. (NNTH=150 for adverse renal events.) (Cochrane Database Syst Rev 2014 Apr 10;:CD008965)
For those who heard that oseltamivir decreases the incidence of “unverified pneumonia,” this “benefit” disappeared in trials where pneumonia was more rigorously defined and diagnosed. And just in case you weren't clear about what exactly constitutes a “neuropsychiatric adverse event,” it represents a wide but universally unpleasant constellation of symptoms: depression, suicidal ideation, suicide attempt, suicide, insomnia, sleep disorders, abnormal dreams, somnolence, fatigue, and anxiety.
And the benefit? We'll get to that in a minute, but first it's worth looking back at the pitch for this drug. When Roche initially rolled out this blockbuster drug (by blockbuster, I mean that it made them billions of dollar, not that it benefited any actual patients), physicians were told that it was going to save lives (it doesn't), prevent admissions (it won't), and protect children, pregnant woman, and the elderly (again that's a big nein, non, and nee for the German, Creole, and Dutch speakers out there).
It turns out that the reason they were able to make those claims was that drug companies were withholding huge chunks of data from a large number of trials. How much data were missing? The most recent Cochrane Review found that “60% (3145/5267) of patient data from randomised, placebo-controlled, phase III treatment trials of oseltamivir have never been published.” (Emphasis added.) You might reasonably ask, why would they withhold so much data? Maybe it's because all this information made Tamiflu look even more effective and wonderful, and the drug companies didn't want to look like they were bragging about all the benefits of the medication they had already convinced the United States to stockpile in massive quantities. I suppose the cynics among us might wonder if it's actually because they knew their drug didn't do anything for patients.
Thankfully, we no longer have to wonder about that missing data. The full story of how independent researchers gained access to the withheld trials is one that's already been the subject of multiple articles and editorials, and is well worth reading. (An excellent introduction to the saga appeared in The New York Times: http://nyti.ms/2lzDUE9.) In the end, all of the data were finally released, allowing for a true analysis of the benefits and harms of neuraminidase inhibitors for treating influenza.
What do we get for rolling the dice when prescribing our patients this expensive drug with a panoply of nasty side effects? A grand total of 0.7-day (6.3 vs 7 days for Tamiflu vs. placebo, 16.8 hours, to be exact) reduction in time to the first alleviation of symptoms. (Emphasis added.) That's it. The best evidence suggests patients experience relief from symptoms just over half-a-day sooner in an illness that otherwise takes seven days to start to improve for the average patient. We don't save any lives, we don't prevent anyone from getting admitted, and we expose our patients to a wide range of harms, ranging from vomiting to acute kidney injury to clinical depression. Of course, shared decision-making is important, but we providers need to be extremely candid about the limited upside, high cost, and significant downside of a medication that, if we are being honest, would likely not have become part of common practice had it not been for some ethically dubious behavior.
Nerd Corner: As if all of this weren't enough to put Tamiflu permanently in the dust bin of clinical practice, the Cochrane authors also caledl into question whether Tamiflu even works as purported: “The influenza virus-specific mechanism of action proposed by the producers does not fit the clinical evidence,” they wrote. (Emphasis added.) Instead, they suggested that it may have more general anti-inflammatory properties that have nothing at all to do with inhibiting neuraminidase. If this is indeed the case, might I humbly suggest popping a couple of ibuprofen instead (1000 of which are currently available at Costco for $9.99), and calling it a day?
One final note: I realize that there have already been numerous excellent articles and summaries on the dubious benefits of Tamiflu. It's always nice to feel like you're at the front edge of new developments in evidence-based clinical practice, but sometimes it's important to take a moment and highlight issues that have already been discussed often and well. We do this to underline important points and hopefully to change practice for the benefit of our patients, which is why I chose to beat this drum one more time.
Share this article on Twitter and Facebook.
Access the links in EMN by reading this on our website or in our free iPad app, both available at www.EM-News.com.
Comments? Write to us at firstname.lastname@example.org.Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.