An Overreach on Overprescribing

A recent article by Jane Brody earlier this week in the New York Times discussed the overuse of an antibiotic group called fluoroquinolones. While she had the right idea, I think her approach and some of her data are at best flawed and at worst downright fear mongering. And while I certainly have a number of specific concerns about this article, they represent a more troubling broader tack being taken by popular press coverage of the phenomenon of antibiotic resistance and antibiotics in general.

Fluoroquinolones were originally created for use against hospital-acquired bacteria, especially antibiotic-resistant gram-negative bacteria. Their use, as Ms. Brody correctly points out, has now spread to more benign locations and situations. The group as a whole is one of the most popular antibiotics for use in community-acquired pneumonia, and it is also extremely popular as a treatment for sinusitis. For my part, as a practicing pediatric infectious disease physician, I use these agents only when a patient is allergic to, or their infection is resistant to, a penicillin-based antibiotic; I also use them when broadly effective oral treatment is required. Indeed, one of the advantages of fluoroquinolones, and why they are used so often in the outpatient setting, is that they are extremely well-absorbed when taken by mouth. They give blood levels nearly equivalent to those seen when administered by IV, which in certain infections can be very useful. All of that said, there is no reason to use them as any sort of routine outpatient or inpatient therapy and most guidelines agree that they are second-line drugs, in essentially all settings.

Ms. Brody recognizes all of this and argues for the decreased use of this class of antibiotics. But then she goes beyond that and attempts to argue that not only shouldn’t fluoroquinolones be used as they currently are, but she implies it is tantamount to malpractice to use them. Now, I grant that she does not use those words, but she points out numerous studies that she feels are strong evidence for problematic fluoroquinolone side effects and argues that physicians should know about these studies and not prescribe fluoroquinolones because of them. She also references the existing black box warning on the drug label (it causes tendinopathy) and says that doctors rarely discuss side effects and black box warnings with their patients.

And so, here is where I will now begin to pick apart and dismantle her article, starting with that near-final point of hers.

First, as a physician, I am offended by that offhand comment about warnings and side effects. I and nearly all of my current and former colleagues always inform patients of pertinent side effects or black box warnings on drugs we prescribe. Always. To offhandedly insinuate otherwise, especially while implying ignorance, is obnoxious and ignorant. Moreover, her handling of her so-called evidence, is shockingly unscientific and inaccurate.

Second, she presents a single patient with bizarre diffuse symptoms that are blamed on fluoroquinolones–as physicians we are (and even as a science journalist she too should have been) trained not to simply believe an isolated case–a study with an “n” of 1, as in only 1 patient participating in the study. This is simply not valid data and to include it is dishonest.

Third, she references one article that claims increased retinal detachments while on fluoroquinolones. The paper is solid and the data is good and is believable. However, the authors themselves state, while they had statistical significance, the clinical impact was very low because while the background rate of retinal detachments was 0.6%, even patients on fluoroquinolones had only a ~3% rate. tendinopathy or the neurologic issues now on the black box warning are both significantly more common. Why Ms. Brody spent time even discussing this eye issue and did not focus more on the things we know to bad about the drugs, is beyond me.

Fourth, the other major article she cites discusses the role of fluoroquinolones in inducing C diff diarrhea. She seems to claim that the article shows that fluoroquinolones are more likely to cause C diff than other antibiotics. This is false. A bit of a diversion on C diff: essentially, this is a secondary bacterial infection that is induced by antibiotics, causing overgrowth of a particular pathogenic bacteria. It is a real entity and is caused by a large number of antibiotics; and it is increasing in incidence. The major issue I take with the paper she cites is that during the time the authors were studying this connection, fluoroquinolone use was indeed rising but was the C diff due to this? Or were both rising for other reasons that could confound the results—an increase in hospitalized patients, for example. This effect by confounding is extremely likely in this case. And because the thrust of the article was that as fluoroquinolone use rose so did C diff, the point they argued was simply that this antibiotic too can cause C diff, so watch out. But there is no reason at the moment to suspect fluoroquinolones are any better or worse at inducing C diff than many, many other antibiotics. This work should not have been cited as it is not about a specific fluoroquinolone problem.

Finally, her tone throughout the article is one of alarm or emergency, as opposed to thoughtfulness, teaching, learning, or any combination thereof. The way to spread news to the NY Times readership is not through fear–it’s through news! I feel like Ms. Brody is peddling her wares differently and as a result ends up (hopefully unintentionally) masking the real truth. Because, the real reason not to use fluoroquinolones is not just that they are too broad but also they are not good , effective drugs! Bacteria can become resistant to almost any fluoroquinolone by a single mutation and this mutation does not seem to confer a survival disadvantage on the bacteria. They are also horrible drugs as a group for treatment of Staphylococcal infections, though many community practitioners do not seem to be aware of this and use them, with numerous clinical failures that we, the ID specialists, need to then fix.

My point is simple: I am not a pharmaceutical apologist and I recognize more than many that antibiotics are overprescribed—it’s my job to tell people when not to use antibiotics as much as it is to advise them on which to use. However there are many good reasons to not use the fluoroquinolones, some of which I have touched on here: tendinopathy, easy resistance, high rate of clinical failure against certain bacteria. For a journalist to cite bad studies or use anecdotal data to support her claim against an antibiotic not only undermines those claims, it undermines even the existing true data about the drug, as some physicians will begin to doubt all the data and may even prescribe the drug more.

But of course, for patients—the true audience for Ms. Brody’s piece—all they will read is the side effects and the personal tragedy and they will start to refuse fluoroquinolones at every turn, out of fear. They may trust their physician less because of her comments regarding proper warnings. They may begin to refuse all antibiotics, even when appropriate. None of these outcomes is a good one. Patients should be informed, but not at the expense of a productive therapeutic relationship with their physicians; not at the expense of appropriate treatment.

The hippocratic oath is often paraphrased as “first, do no harm.” Ms. Brody’s piece has, I fear, done a vast amount of harm; for her, for the reputation of her employer, and most important for the patients we physicians try, desperately, to serve, even in the face of unfortunately careless reporting.