March 20th, 2016

“Choosing Wisely” in HIV Medicine — Should We Stop Giving MAC Prophylaxis?

mac and cheese(Disclosure:  The following post represents personal opinion, and is in conflict with treatment guidelines. Proceed at your own risk.)

E-mail recently from one of our outstanding first-year fellows:

Hi Paul,
I’ve heard you recommended against the use of MAC [M. avium complex] prophylaxis in most settings in the modern HAART era. We admitted a 21yo F patient, non-compliant with meds, with a CD4 of 2. She has poor tolerance of pills in general which has been a major obstacle to her care. Would you recommend against the use of MAC ppx in this patient? Even with a CD4 of 2?

It’s a great question, notwithstanding the use of the term “HAART”. (I think she did that just to bug me).

In general, I’m not recommending MAC prophylaxis anymore, and here’s why:

  1. It’s never been proven necessary in the modern ART era. For the record, this is the most relevant placebo-controlled trial, and it enrolled patients from 1992-1994—eons ago, ART-wise (only available HIV therapies were NRTIs). There’s certainly no controlled clinical trial since 1996, when effective ART became widely available. This observational study from the HIV Outpatient Study (too bad it wasn’t the MACs cohort, ha ha) didn’t find any benefit, and this one from the early ART era was conducted in 1996-1997, a time when virologic suppression rates in clinical cohorts was 20-30%. Please tell me if I’m wrong and there are other more relevant studies.
  2. Effective HIV therapy reduces MAC risk much more than azithromycin ever did. Disseminated MAC was, along with CMV retinitis, one of those dreaded “final common pathway” OIs that occurred late in HIV disease, almost exclusively in patients with longstanding severe immunosuppression. The incidence of both of these conditions—especially if you limit diagnoses to non-IRIS cases—has dropped dramatically.
  3. Some of these advanced disease patients may have undiagnosed MAC. It could be clinical or subclinical, and you wouldn’t want to give low-dose “monotherapy” to them anyway. They might develop MAC IRIS, but do we have any evidence that MAC prophylaxis prevents MAC IRIS?
  4. High-dose, weekly azithromycin has side effects. These are mostly GI side effects, and I’d argue anything that might make it tougher for someone with a CD4 < 50 to take ART is a bad thing.
  5. Any antibiotic administered chronically selects for resistance, will alter the “microbiome.” Although clinical studies of MAC prophylaxis didn’t find much macrolide resistance in M avium, resistance to macrolides among other bacteria is widespread. And one dose of an antibiotic can alter the microbiome.
  6. It costs money. Not a lot, but something. And if it’s not doing any good, or even worse, distracting someone from taking ART, it’s bad value.

So could there be anyone in whom I’d still recommend MAC prophylaxis? I suppose—but without much enthusiasm, and only if all the following criteria were met:

  • Newly diagnosed with HIV, or newly ready to take ART.
  • CD4 < 50.
  • No clinical symptoms or signs consistent with MAC.
  • Expresses strong commitment to taking ART now that he/she has this serious diagnosis (AIDS).

In these cases, one could argue that it’s possible that giving it would provide some benefit during this very vulnerable period. If the patient takes ART, furthermore, the CD4 will soon be above the threshold allowing discontinuation of primary MAC prophylaxis,

But I’d have a very low threshold for stopping it (any GI toxicity in particular) even before the CD4 cell count is > 100, and would make it abundantly clear that this patient’s true lifeline to health is HIV therapy.

So what about a controlled clinical trial? It’s never going to happen—the incidence of MAC is too low these days, so the sample size would need to be untenably large. Additionally, finding and enrolling patients who meet the CD4 entry criteria would be an enormous challenge. These are not the easiest patients to get into clinical trials.

So what say you, ID/HIV specialists—should we still be prescribing MAC prophylaxis?

A patient is hospitalized with AIDS and has a CD4 cell count < 50. Would you recommend prophylaxis for M. avium complex?

View Results

8 Responses to ““Choosing Wisely” in HIV Medicine — Should We Stop Giving MAC Prophylaxis?”

  1. Loretta S says:

    I didn’t vote because I am not an ID and/or HIV specialist. But I thank you, Paul, for point #5. We need to keep stressing the effects of antibiotics on the microbiome (esp. the gut microbiome), and we need to educate our patients about the importance of that microbiome. And thank you for mentioning the GI effects of azithromycin. In primary care, I hardly ever hear patients complain about the GI effects. In fact, they frequently demand that I prescribe azithromycin for them! But I had the misfortune of taking it myself about a year ago. Oh, boy, I won’t get into what happened next. But I am now painfully aware of how bad the GI effects can be. For an HIV+ patient who has a poor tolerance of pills, causing those GI effects would definitely NOT improve her more general drug adherence.

    Sorry for the sidetrack!

  2. Adam Lake says:

    For my patients who are poorly or intermittently adherent I have been using azithro 250mg 5x/wk (“daily but you can forget twice a week”). It seems to be much better tolerated, and while there are no RCTs to back me up, our ID pharmacist agreed that pharmacokinetically it makes sense as the concentrated big dose isn’t the key feature to its success. I absolutely agree that I have a low threshold to drop it completely, but I wanted to share this and get some feedback on it.

  3. David says:

    I prefer to call it just “fairly active anti-retroviral therapy” aka FAART…

  4. JTB says:

    I agree with you about avoiding MAC proph. in these pts. The ART’s are clearly providing the benefit here. I have also been avoiding giving Bactrim proph. as well as the incidence of hypersensitivity reactions is 30-40 % , when I have a clearly compliant pt. I recently saw a man with a CD4 of zero with newrly diagnosed KS who did great with ART’s alone . without any prophylaxis , for MAC or PCP.

  5. Brenda Crabtree says:

    Hi, great reflexion. I think it might also depend on your prevalece… 20% of our mycobacteria in advanced hiv patients are MAC, which is pretty high! Therefore we still give azitro with ART.

    • Paul Sax says:


      Thanks for your comment.

      Almost all the mycobacteria in our advanced HIV disease patients is MAC, but it’s become so rare with effective ART (except for IRIS) that I’m not sure we still need the prophylaxis.


  6. Susan Jacobson says:

    The only MAC prophylaxis I give is the long term non-compliant who are willing to take a weekly medication, but can’t commit to daily ARVs.

  7. Ajay says:

    It’s rare in india and we don’t give Mac prophylaxis here .

HIV Information: Author Paul Sax, M.D.

Paul E. Sax, MD

Contributing Editor

NEJM Journal Watch
Infectious Diseases

Biography | Disclosures | Summaries

Learn more about HIV and ID Observations.