What is the truth about Zithromax?

It was during my residency that the first indication of heart toxicity of antibiotics affected me personally.  The threat was related to the use of the first of the non-drowsy antihistamines – Seldane – in combination with macrolide antibiotics, such as Erythromycin causing a potentially fatal heart arrhythmia.  I remember the expressions fear from other residents, as we had used this combination of medications often.  Were we killing people when we treated their bronchitis?  We had no idea, but we were consoled by the fact that the people who had gotten our arrhythmia-provoking combo were largely anonymous to us (ER patients).

Fast forward to 2012 and the study (published in the holy writings of the New England Journal of Medicine) that Zithromax is associated with more dead people than no Zithromax.  Here’s the headline-provoking conclusion:

During 5 days of therapy, patients taking azithromycin, as compared with those who took no antibiotics, had an increased risk of cardiovascular death (hazard ratio, 2.88; 95% confidence interval [CI], 1.79 to 4.63; P<0.001) and death from any cause (hazard ratio, 1.85; 95% CI, 1.25 to 2.75; P=0.002).  Patients who took amoxicillin had no increase in the risk of death during this period. Relative to amoxicillin, azithromycin was associated with an increased risk of cardiovascular death (hazard ratio, 2.49; 95% CI, 1.38 to 4.50; P=0.002) and death from any cause (hazard ratio, 2.02; 95% CI, 1.24 to 3.30; P=0.005), with an estimated 47 additional cardiovascular deaths per 1 million courses; patients in the highest decile of risk for cardiovascular disease had an estimated 245 additional cardiovascular deaths per 1 million courses. (Emphasis Mine).

It turns out that they also indicted Levofloxacin, another commonly-used antibiotic as being roughly as risky as Zithromax.

While this is good fodder for the headlines, it hits me right where I live.  I constantly have patients coming into the office with symptoms that make them feel they need an antibiotic, many of whom have gotten Zithromax.  I wrote an early post on the subject of the temptation to give a Z-Pak in the gift basket we give our patients for walking into our office:

Which brings me back to the Z-Pak.  Zithromax (Azithromycin) is truly a great drug, and the friend of many doctors.  It treats strep throat, skin infections, sexually transmitted disease, whooping cough, and certain kinds of, yes, bronchitis.  It is very easy to take, requiring a total of 5 doses over 5 days, and it comes in a handy-dandy pack with a catchy name.  When a patient tells their friends and family, “I got a Z-Pak,” they are much more impressed than if they say, “I got an antibiotic.”

I ended with a warning:

So, when you have a cough and go to the doctor, get the diagnosis of bronchitis, and get a Z-Pak think of me.  You may want to ask if you really need the antibiotic, or if you can wait to see if it will go away without it.  In many, if not most cases, you might just as well meditate with the word “Zithromax” as your mantra, or burn the pills in a sacrifice to the Greek god Z-pacchus.

God bless America, land of the Z.

I even wrote a poem for it:

Six little pills at the patients’ insistence
Six little pills should we now keep our distance?
Six little pills we’ll rue your existence
If Six little pills are paths to resistance.

Oh Zithromax, Zithromax!
You make us desirous
Against our best judgment to cover a virus
Oh Zithromax, Zithromax!
Your pills in a pack
So oft make the best doctor act like a quack.

Yet there are good reasons to use antibiotics like Zithromax, so I am left with the dilemma of how to interpret the results.  Is this a real problem, or is it simply a retrospective study by a bunch of scientists wanting to make a splash?  I have to answer this question because I have to decide whether or not I am going to write a prescription for this medication, risking a “is my doctor trying to kill me?” look from my patients.  I have to prescribe antibiotics, but in doing so do I feed the fortunes of personal injury attorneys who realize the two following things:

  1. Doctors prescribe Zithromax by the bucket
  2. Every one of the patients who get a Zithromax prescription will die.

I give it 2 weeks before we see a commercial soliciting business for people who have loved ones who took Zithromax and then had heart attacks.

To figure out how to deal with this dilemma, I went to some of the experts among the med blogger community.  Marya Zilberberg is an epidemiologist at the University of Massachusetts and author of the blog, Healthcare, etc.  She even wrote a book about how to properly read medical literature (a book that I need to read, actually).  In short, she’s brainy.  She wrote a post entitled, Why I have the propensity to believe the azythromycin data (I told you she was brainy), in which she states the following:

But there is a second, possibly more important reason that I am inclined to believe the data. The reason is called succinctly “propensity scoring.” This is the technique that the investigators used to adjust away as much as feasible the possibility that factors other than the exposure to the drug caused the observed effect.

She then quotes a part of her book (which I definitely need to read) about propensity scoring.  Tying this to the Zithromax study:

And if you are able to access Table 1 of the paper, you will see that their propensity matching was spectacularly successful. So, although it does not eliminate the possibility that something unobserved or unmeasured is causing this increase in deaths, the meticulous methods used lower the probability of this.

So by this I am led to believe the data have some beef behind them.  I am also much more likely to use the word “propensity,” as it may make me sound as brainy as Marya.

On the counterpoint is Dr. Wes, one of the old guard bloggers (who I’ve drunk beer with), who has been blogging since the internet was run by carrier pigeon.  Dr. Wes is a cardiologist who specializes in heart rhythm problems, the kind of problems that presumably killed the people in the NEJM study.  He wrote an article, How Bad is Azithromycin’s Cardiovascular Risk?  in which he admits the potential risk of this kind of antibiotics, but questions the data methods of the study:

What was far scarier to me, though, was how the authors of this week’s paper reached their estimates of the magnitude of azithromycin’s cardiovascular risk.

Welcome to the underworld of Big Data Medicine.

He minces no words as he continues:

To think that despite all of the confounding factors that the authors had the balls to state that “as compared with amoxacillin that there were 47 additional deaths per 1 million courses of azithromycin therapy; for patients with the highest decile of baseline risk of cardiovascular disease, there were 245 additional cardiovascular deaths per 1 million courses” is ridiculous.  Seriously, after all the manipulation of data, they are capable of defining a magnitude to three significant digits out of a million of anything?

His conclusion is that this study is basically a bunch of sensationalized data meant to get headlines (which it did).  I think he needs a beer.  Call me, Wes.

So I am left to sift through these two opinions of two people I respect, and do so in the backdrop of patients wanting antibiotics and lawyers dreaming of big yachts.  What do I think?  I think we can’t tell what the truth really is.  Yes, the folks who wrote the study are probably gunning for headlines (as is the NEJM), but it is also a fact that antibiotics can be dangerous, and all drugs come with some sort of a price.

I come back to advice I gave in an earlier post: When all else fails, do nothing.  Don’t give an antibiotic unless it’s needed, and don’t ask for one if you don’t need it.

Rob Lamberts is an internal medicine-pediatrics physician who blogs at More Musings (of a Distractible Kind).

Submit a guest post and be heard on social media’s leading physician voice.

Comments are moderated before they are published. Please read the comment policy.

  • http://www.facebook.com/karl.bucus Karl U. Bucus

    I marginally tend to agree with Dr. Wes. We are all so trained to ignore raw numbers that we forget to step back every once and a while and actually look at those numbers, comparing them to our common sense.

  • drwes

    If the authors had wanted to use their retrospective database review study to generate a hypothesis to prospectively study later, I’d be fine with that.  But they went too far and proposed a risk – a very specific cardiovascular risk -  based on disparate databases joined together with little mention of the bias introduced by their technique. 

    Data-mining studies like these are the new “black” in medicine, now that we’ve entered the EMR era, but for the physician actually treats patients based on these flawed studies, caveat emptor.

  • http://distractible.org/ Dr. Rob

    The whole reason this is an issue (aside from trying to grab headlines) is that docs always treat antibiotics (especially Zithromax) as if it’s harmless.  No drug is harmless, and this one is especially coming back to haunt us.  

  • http://www.facebook.com/profile.php?id=733625742 Mitchell Ehrenberg

    There were several studies in 1999 and 2000 in which azithromycin was given to patients with CAD in order to prevent secondary events. One competing theory of atherosclerosis was that it was an infectious disease caused by Chlamydia. The studies found not benefit, but more importantly, found no harm from daily use of Zithromax for 3 months. Patients were followed for development of CV events for 2 years. Here’s the citation. Circulation. 2000 Oct 10;102(15):1755-60. The current “study” is a retrospective analysis, not an RCT. In the RCT from 2000, no harm was documented among a high risk population exposed to Zithromax for much longer than 5 days. I fail to see how the new study supplants the rigor of the older one.

  • http://www.facebook.com/adrian.rabe Adrian Paul Rabe

    Likewise, the number needed to treat, or in this case, the number of prescriptions you’d need to make over amoxicillin is 18,000. I think that treatment is worth the risk, and that the risk varies per patient. This is why the authors of that study added analysis for high risk patients, who accounted for majority of those deaths.

  • http://www.facebook.com/jyon1 Jose Yon

    is there any good antibiotic?  Articles like this about zithromax and others on web about cipro makes one leery about taking them.  In researching natural antibiotics, what are people’s take on Manuka Honey.  So far from what i found, it can be better than antibiotics?

    • http://distractible.org/ Dr. Rob

      This does not mean Zithromax is a bad antibiotic, nor does it mean it is dangerous.  All this study suggests is that no antibiotic is without risk.  I will continue to use Zithromax where appropriate.  If your doctor thinks you need to take it, then I would recommend doing so as the risk of problems from infection are probably much, much higher than the risk of heart problems.

  • http://www.caduceusblog.com/ Deep Ramachandran

    Very well written and balanced article, Rob. As a pulmonogist, I also prescribe azithromycin regularly, particularly after an article demonstrating reduced exacerbations of COPD with daily azithromycin therapy that was also published in NEJM, in addition to previously demonstrated benefits in other lung diseases.
    I suppose I agree with Dr. Wes, it’s rare to see an article that has to use that many zeros, after the decimel point. One can only imagine the conversation:
    Steve: Well it looks like our data doesn’t show jack.
    Tom: Add another zero.
    Steve: Ok. Nope, still nothing.
    Tom: Try Again.
    Steve: Nope, I guess we’ll call it a day.
    Tom: Hit it again.
    Steve: But Tom, I don’t think she can handle it, she’ll fly apart!
    Tom: Do it!
    Steve: It worked! My God, we’ve done it!
    Tom: (to secretary) Get me the President of New England Journal on the phone!

    Ultimately I think your conclusion is most apt, no medication is without risk. As physicians we frequently forget that.

    • http://distractible.org/ Dr. Rob

      Well done!  I really don’t think this changes anything but perception.  I have a few more posts coming out about dealing with that perception as a clinician.  People want black and white in a world that is never that way.  The press (and big journals) like to make it look that way because it sells advertising and funds grants for the academic docs.  Patients and docs, unfortunately, are caught in the crossfire (to quote the famous philosopher Stevie Ray Vaughan).

  • https://www.kg-ekgpress.com/ Ken Grauer, MD

    THANK YOU Rob for a well-written and thought-provoking post! I think your BOTTOM LINE says it best: i) We can’t really know the truth from this retrospective study; and ii) Zithromax (or other antibiotic) should not be prescribed unless the patient truly needs it …

    As to risk from provoking long QT and potentially lethal Torsades – my impression was that this risk is virtually negligible IF given to otherwise healthy subjects for short-term and WITHOUT other ongoing medications that are known to lengthen QT/interact with macrolides … Risk is not negigible IF the patient has underlying heart disease and is on multiple potentially-interacting drugs …

    Hard to imagine daily Zithromax for patients with COPD will lead to any good ….

    And as Mitchell reminds us – there was a time when the “theory” was an infectious cause of CAD ….

  • http://pulse.yahoo.com/_GR2F7ZHJRA2AVVT7CXGNBNXLY4 Dorris P

    They are comming out with the information that CPacks are connected to effecting the heart, what is the TRUTH behind this clam?

  • http://www.facebook.com/profile.php?id=120501600 Milinda Houlette

    This is going to scare some patients senseless. Many patients are allergic to penicillin, and therefore in combination with the medications that have to be taken for chronic illnessses the Z-pac is the only antibiotic that is given. For a study to now state that the Z-pac will cause the patient to have a heart attack will cause many to not seek medical attention until the illness has progressed to in-patient status.  

  • Marya Zilberberg

    Rob, this is indeed a great post for many reasons,one of which is your conclusion — nothing is without a risk. we just need to examine honestly whether the benefit outweighs the risk. This is true for all antibiotics, including azithro. At the same time I also have to mention that when it comes to safety studies, it is unrealistic to expect that a cohort study will generate a hypothesis that will then result in a RCT to test this hypothesis. This is not feasible for many reasons, including the fact that, once on the market, it is hard to randomize people to a drug like azithro vs. something else. Another important obstacle is that the resources it would take to run a practical RCT like this would be exorbitant, and the time to establish a signal like this would be prolonged. The study was done well and within the methods of pharmacovigilance. The old refrain that observational studies cannot establish causality is getting a little old, since it can also be said of interventional data. In short, the best we can do is infer and estimate causality, and when it comes to harm, we need to be that much more humble. Thanks, Rob, for this provocative piece.     

  • StephenModesto

    …very nice article; I like your perspective approach and writing style. The pardox of irony is the dilemma for providing customer service…if it is not the patient then it is the indignant family member of an `elder’ demanding that something be done. There was a time, in the not so distant past, that the term meta-analysis was just a twinkling glimmer in a statistican’s eye. It would be good to know that there will be those with a self-honesty and tenacity of spirit to ferret the facts behind the story and not merely spin the yarn for the fabric of the pharmaceutical monied interests.

  • dacooke

    The bigger question which the above discussion misses is why doctors have been prescribing azithromycin for bronchitis at all?  Azithromycin, as have all other antibiotics, have been repeatedly shown to be ineffective for bronchitis, yet they are prescribed for this indication by the bucketful.  The CDC, ACP, AAFP, AAP, and IDSA released guidelines recommending against antibiotics for bronchitis nearly a decade ago, but they have been widely ignored, as have similar guidelines by the American Academy of Chest Physicians.  Yes, I am aware that azithromycin has a FDA approval for acute bacterial exacerbations of chronic bronchitis, but only a small proportion of the azithromycin scripts written for “bronchitis” are actually for ABECB.

    The harms of unnecessary antibiotic prescriptions go far beyond the issues of whether azithromycin is actually pro-arrhythmic.  C. difficile colitis, yeast vaginitis, serious allergic reactions, and progressive development of widespread antibiotic resistance are huge adverse effects of inappropriate prescribing, and that is without even considering the dollar cost of all these useless prescriptions.

    Azithromycin has its uses, as do other antibiotics, but doctors (and yes, I am one) need to face up to the fact that we are prescribing them in large quantities for indications that they don’t help, and causing widespread harm in the process.

  • http://www.facebook.com/carol.gino Carol Gino

    There’s another known side effect of Zithromax that’s hidden in the fine print and that is a great danger.  That is in some patients it causes hemorrhage and denuding of the bowel.  Several years ago that happened to me and it took 7 days to stop the bleeding.  The doctors thought they would have to do a colostomy until it finally stopped.  Still, though it may be idiosyncratic, it still is in the fine print, so it has happened to others.  Doctors should clearly check that those who take it don’t have sensitive GI tracts or they might wind up with no bowel to speak of….Just a caution.