The Penicillin Allergy Puzzle: Who’s Really Allergic?

“Most penicillin allergy labels are not indicative of any meaningful penicillin allergic reaction, and the vast majority can be de-labelled with direct oral challenges of amoxicillin."  — Dr. Kimberly Blumenthal

Penicillin allergy is one of the most commonly reported drug allergies—but here’s the twist: most people who carry this label aren’t actually allergic. So how did we get here, and what can be done to fix it?

On this episode, Dr. Mariam Hanna sits down with Dr. Kimberly Blumenthal, an allergist, immunologist, and clinical researcher at Massachusetts General Hospital, and an associate professor at Harvard Medical School. Dr. Blumenthal is internationally recognized for her work on penicillin allergy and the real-world impact of unnecessary labels on patient care, antibiotic resistance, and health equity.

We cover:

• Why penicillin allergy is so commonly (and incorrectly) diagnosed
• How de-labelling can improve antibiotic stewardship and patient outcomes
• Risk stratification: who needs testing, who can go straight to a challenge, and who should avoid penicillin?
• The role of allergists in leading the charge on de-labelling efforts
• Special populations, including pregnant patients, children with serum sickness-like reactions, and marginalized communities with less access to allergy care
• International practices that complicate the picture—like routine penicillin pre-screening in some countries

With the vast majority of penicillin allergy labels being inaccurate, this episode highlights why it's time to stop assuming and start testing.

Visit the Canadian Society of Allergy and Clinical Immunology

Find an allergist using our helpful tool

Find Dr. Hanna on X, previously Twitter, @PedsAllergyDoc or CSACI @CSACI_ca

The Allergist is produced for CSACI by PodCraft Productions

Transcript

 Dr. Mariam Hanna:

Hello, I'm Dr. Mariam Hanna, and this is The Allergist, a show that separates myth from medicine, deciphering allergies and understanding the immune system. 

A pregnant mom is GBS-positive, history of penicillin allergy, advised to defer testing and evaluation till after pregnancy. But what about this baby? 

Penicillin allergy is one of the most commonly reported drug allergies, but it's actually one of the most commonly reported that people are actually not allergic to penicillin. So how's that for irony? Today, we're going to look back at penicillin allergy and why getting tested could change patient management and outcomes. It's my distinct pleasure to introduce today's guest.

Dr. Kimberly Blumenthal is an allergist, immunologist, and clinical researcher at Massachusetts General Hospital, an associate professor of medicine at Harvard Medical School.  She's the co-director of the Clinical Epidemiology Research Center and Director of Research in the Center for Drug and Vaccine Allergy.

Dr. Blumenthal is internationally recognized for identifying the morbidity and mortality associated with unverified penicillin allergies and creating innovative approaches to the evaluation of penicillin and cephalosporin antibiotic allergies in diverse patient populations. Dr. Blumenthal has authored more than 200 peer-reviewed publications in high impact journals, which is why it's my distinct pleasure that she took time out of her day to join us in today's podcast episode. Dr. Blumenthal, welcome to the podcast.

Dr. Kimberly Blumenthal:

Thank you so much for having me.

Dr. Mariam Hanna:

Okay, so we're going to start with a big question right off the bat. Hit you hard with it. What is the public health impact of having a diagnosis of penicillin allergy?

Dr. Kimberly Blumenthal:

Thank you for asking me this specific question because it is so common that we as allergists need to convey the harms associated with thinking that you're penicillin allergic in today's world. And really the harm is not only are penicillins avoided even when they're really strongly indicated, but often even all beta-lactams or at least the cephalosporins are avoided as well, which really leads to less effective care for infections.

So if you're an individual, you might get not the best treatment for you if you have a penicillin allergy label. And this can result in more adverse effects, it can result in more surgical site infections. And for the community at large, a lot of people thinking that they're penicillin allergic when they're not, it leads to broader spectrum antibiotic use. And over time, we've found that this also results in higher levels of antibiotic resistance, colonization, or infection with things like MRSA and VRE.

Dr. Mariam Hanna:

So what about if we flip it? So that's why it's important for the public to know about this diagnosis of penicillin allergy. Why should allergists care? At this point, I have some allergists that are like, none of it is real. It's all a figment in their imagination. I'm not wasting time on these evaluations.

Dr. Kimberly Blumenthal:

Oh, that's so hard to hear. Because really if we don't evaluate and disprove penicillin allergy, who will feel comfortable doing so. We do need to empower people beyond the allergist to take this on to learn about how to de-label penicillin allergies when they're not real or when they're clearly something like a headache or family history or never were exposed. But really, the trust will come from an allergist.

And so people don't know if they're allergic or not, and they were told they were allergic and they were told never to take penicillin. And these are the things that are told to them. And then they carry that with them through their healthcare experiences into adulthood often and avoid penicillin for so long that they might be fearful of taking penicillin again, they might need to talk to a specialist. So if we don't do this and we don't help people evaluate whether or not they are part of the minority with a true allergy or the majority with a bogus allergy, then no one else really will. And we are in the best position to do so and to train a specialized kind of workforce to do this work.

Dr. Mariam Hanna:

Absolutely. Okay. So now when we're doing into the real meat of the true penicillin allergy, are there different types or nuances to this penicillin allergy label that we see in the clinic?

Dr. Kimberly Blumenthal:

Oh, completely. So the most common thing we will all see is this, my mama told me allergy. And I've got nothing against moms, I am one. My mama told me allergy is an allergy that we didn't observe. And it might have been that child, it might've been another child, right>? So it's a remote allergy. It was often cutaneous only. It didn't involve any systemic signs or symptoms like any shortness of breath or any low blood pressure. And it is these cutaneous-only remote reactions that are the lowest risk.

We know that kids, for example, get hives and other types of rashes when they have infections, not even related to antibiotic use. So this is the bulk of the work. And in fact, if we could just take care of the bulk of that, these low-risk patients, we would do a phenomenal good for patients in society.

But there are actually some higher-risk reactions, and these are the ones that the allergists certainly need to weigh in on. The more higher-risk reactions, of course, something that is a systemic reaction or potentially an anaphylactic reaction that can happen with the penicillins more often with the intravenous penicillins and with the oral penicillins. And then some other common ones that we see would be the serum sickness-like reactions. The serum sickness can occur from the oral or the intravenous penicillins.

And then very rarely, but we have to have an eye out for it. The penicillins can also cause delayed severe reactions. Specifically, I've seen a fair number of cases of acute generalized exanthematous pustulosis or AGEP with our penicillins. And then of course Nafcilin can cause an acute interstitial nephritis. So we have to be on the lookout for some rare type phenotypes that could be these delayed severe reactions that as allergists we would want to identify so that patient doesn't just get a drug challenge and false reassurance and de-labeling.

Dr. Mariam Hanna:

Love that. Okay, so that feeds perfectly into how can penicillin allergy be evaluated? Sidebar, I will forever categorize low-risk ones as my mama told me, but how can penicillin allergy best be evaluated?

Dr. Kimberly Blumenthal:

Yeah. So if in a perfect world we would have the risk stratification occur before they even show up at our office because we would know who needs a direct challenge where we're just going to give a moxicillin or another penicillin with observation and then who needs a skin test versus who needs avoidance because of perhaps a scar. Or they might need more sophisticated, delayed-type testing like patch test or delayed intradermal tests.

So having a risk stratification rubric that your clinic uses, or you could use one that drug allergy practice parameters in America, or the ones put out by Europe or Canada, you can use other risk stratification. But ideally we would have some sort of risk stratification that occurs even before they hit the door. Those aren't things that can happen in one visit.

And then all of the higher risk people; systemic reactions, anybody who has a scary acute physiology, so for example, our lug transplant center, and anytime I'm seeing a patient who comes in wheeling their oxygen and they're on the lug transplant list, I'm like, well, you're a low-risk allergy reaction, but you're a high-risk human.

So, therefore, I am going to use a skin test because you're a high-risk human. We have a great skin test for penicillin allergy. It's the only drug we have a good skin test for. So I really feel like when you're either high-risk allergy history, something that sounds like it could have been anaphylaxis or you're a high-risk host, you're waiting a CABG, you're waiting a lung transplant, or sometimes I think a high-risk host might be the person who has multiple drug allergies because they might feel some sort of confidence from a skin test that they wouldn't feel if you just skip right to a challenge. So that's really how I approach risk stratification for things that might be IGE and then the things that I think might be severe and delayed.

Dr. Mariam Hanna:

In what cases would you do patch testing for penicillin or delayed intradermal read and is there good evidence to do that?

Dr. Kimberly Blumenthal:

Oh, gosh, a loaded question. So we just started doing more delayed testing with the idea that we need more data, we need more data points. We've often found that in a case where you have a really clear phenotype that delayed intradermal or patch testing might be helpful for such as dress syndrome or fixed drug eruption, some of these clear phenotypes, the testing might help you narrow down the culprit. So if it's a patient who comes to you with a fixed drug eruption and they were only on vancomycin, that person doesn't need patch testing or delayed intradermal testing to confirm it. You can say the records and the history and the photos clearly indicate it was vancomycin that caused this. We don't need to play with testing.

But if you're a patient who is on three or four medications, specifically antibiotics, and you develop a fixed drug eruption or you develop AGEP or you develop dress, the testing might help us narrow down the culprit because usually, all we have to go on is timing. And the timing can get you pretty far. You have to have been on the medication recently before the reaction started and then maybe got better when those medicines stopped. It has to be a known culprit for that type of reaction. But the testing could say, oh, well if you get a positive, it probably isn't these other antibiotics.

Dr. Mariam Hanna:

And if their test is negative, if I push just one more, would you be confident enough to re-challenge?

Dr. Kimberly Blumenthal:

Great question. So if in a perfect world, one test would be positive, right? We do the patch test and the delayed intradermal and one is screaming, oh, this is the culprit and the other things are okay. So then what we would say is that if necessary to really try one of these again, you can do it and you would do it with the understanding of the patient, the other doctors that with the limitations of all of our current testing, we think that we found the culprit and it was this. And so we think therefore that you could try these other things with close observation and monitoring.

Dr. Mariam Hanna:

Okay. And then these re-challenge back, is this an oral or IV dose or like multi-day dosing? What would happen with these more complicated stories?

Dr. Kimberly Blumenthal:

Yeah, we would wait until it's necessary because that also changes your risk calculation, right? We're not going to electively give it to you again. But if you are in a situation where you have a bacteremia with that same nasty bug and you're in the hospital and we think that we have good evidence, it was vancomycin and it wasn't the zosin, the pipercilentazobactam, then you can try that-

Dr. Mariam Hanna:

In the hospital during an acute need for it.

Dr. Kimberly Blumenthal:

Yes, at the time of acute need.

Dr. Mariam Hanna:

At the time of acute need. Okay. So let's talk about special populations. I love how you're already filling this with lots of different examples as to when you should think about testing versus just avoiding for a period of time. The first special population I think I want to talk about is pregnant patients. We actually had a lot of GBS bacteremia in my region recently, and this had stirred up a lot of referrals with pregnant women. Should those patients be subjected to skin testing or oral challenges for GBS bacteremia or prophylaxis then?

Dr. Kimberly Blumenthal:

If you know that the women are GBS positive and have a penicillin allergy and you have time to fit them in that tiny little window before delivery, you absolutely should be evaluating the women because we know that the vast majority of pregnant women with a penicillin allergy label are not truly allergic. And we know that the penicillin is what they really do need for prophylaxis for GBS.

Now we are in the position of not being able to see all of our pregnant women in that tiny little window. So what we've been doing for the past five years is to see patients in their third trimester before they even get the swab that determines whether or not they have carriage of Group B Strep. Now that is elective because our evaluation is before they have GBS. If you have the evaluation after the GBS, I don't really consider it elective. It's best care, it's best practice.

But before you have GBS, it does seem like it's elective penicillin allergy de-labeling, considering that they may have GBS or they may need a C-section where sephazelin is indicated, or they may have a post-op mastitis where Keflex or Sephalexin is indicated. So because we see our pregnant women electively before the time of the GBS swab, we skin test because they're carrying a human. We want to be risk averse and it's elective.

If you're seeing them at other times, I do know that there are many practices that do direct challenges in the pregnant women, and it's reasonable too, especially if they don't want a skin test or they're in a rush or they're not really concerned about their allergy. So there are many situations where with the patient's buy-in, you could do a direct challenge. We just say because we're doing it as a quality improvement initiative that we would never want to compromise any patient safety.

Dr. Mariam Hanna:

Okay. Would there ever be a case in your mind where you could acutely give a test dose during labor delivery for a patient, given that it is in general low-risk and in general otherwise healthy populations? Or should those patients use alternate antibiotics and differ?

Dr. Kimberly Blumenthal:

That's a really good question. I think that the labor and delivery is an acute period for women with a lot of risks that are hard for me to speak to, as a non-obstetrician. I think if everything is going perfectly well and there's a time period that allows for a test dose to be given, that would be completely appropriate. So giving a 10th of the dose and being able to watch for at least 30 minutes before giving the full dose that was indicated during normal healthy pregnancy and delivery would be, I don't see why that couldn't be an option.

Dr. Mariam Hanna:

Perfect. Okay. When you're the labeler or when you're avoiding while waiting assessment, is there a need to avoid cephalosporins, Dr. Blumenthal?

Dr. Kimberly Blumenthal:

Oh, that's a good question, too. What a changing target that has been over the years. So the frank answer is that we're still trying to understand the clinical implication of side-chain similarity. We can talk to the cows come home about the pharmacology and the organic chemistry of how our cephalosporins do or do not look like the penicillins and they all share beta-lactam ring. And that caused some massive concerns a long time ago.

And now it does seem to be that it's really more the R-1 side-chain that confers our similarities. So I would say in general, anybody with a penicillinology label, there's no concern with our later-generation cephalosporins because they look completely dissimilar from our penicillins. But there are some cephalosporins that look a little bit like our penicillin. So if you had a severe immediate onset allergic reaction to amoxicillin, for example, it looks pretty similar to cephalexin or cephydroxil or ceprosil. And so maybe you'd be a little bit more cautious with the administration of those specific cephalosporins.

Dr. Mariam Hanna:

Would you test or challenge those specific or those closer ones or just as a precaution, just say avoid those ones?

Dr. Kimberly Blumenthal:

Yeah, if it were relevant, I think I would test dose or do a challenge test for them.

Dr. Mariam Hanna:

Okay.

Dr. Kimberly Blumenthal:

Sorry.

Dr. Mariam Hanna:

The pediatric population that I see has a lot of these serum sickness-like reaction presentations early on in their childhood, and I really struggle with lifelong avoidance for these guys. What is your take on appropriateness for evaluating them?

Dr. Kimberly Blumenthal:

This is also a really interesting question because our data have changed a bit and recommendations have changed. Really the data has come from Canada that suggests that we can consider serum sickness-like reaction re-challenging. And the reason we can consider it is because most of the kids will tolerate an initial dose and even in a course subsequently.

So the data come from children that were re-challenged who mostly had amoxicillin or early-generation serum sickness-like reactions, and they just re-challenged with a direct challenge these kids. And most of them were not allergic with their initial dose. And then they were able to follow a certain portion of those kids and up to 25% of them had a reaction with their full course, if I recall. And so what this means to me and how we've interpreted it in our guidelines in the United States is that we took serum sickness-like reactions off of the contraindication table.

So you can consider it when you need to, but you would have to have some pretty good conversations and counseling in my opinion, because you want to not only provide this direct challenge again, but you want to provide down the road when they have a whole course, if they develop symptoms and signs of reaction that up to 1 in 4 could do this and that you're not going anywhere and you'd still be able to treat and help them through that.

And so it really has to be something that you find might be useful for that particular kiddo. If the kid is struggling a lot with infections, has an immune deficiency, has chronic sinusitis, has chronic strep infections, whatever, it has to be individualized somewhat because your de-labeling in real-time from that one dose is very high. But when they subsequently do a whole course, many of them do get rashes.

Dr. Mariam Hanna:

Excellent, excellent. A great approach to not just do the test dose and adios, but actually follow through with the subsequent courses to see how it goes. So I'm also taking it that you don't do a prolonged oral challenge electively on these patients, is that correct?

Dr. Kimberly Blumenthal:

This is correct. So our guidelines now would advise against it mostly for antibiotic stewardship reasons, and that was another one where they said juice wasn't worth the squeeze because you were just picking up an extra 1% potentially of reactions. I am a big fan of letting the patient, the parents know that there's always a chance that when you do therapeutic courses that you're going to have a reaction. And that doesn't mean that I was wrong or that this test didn't work for you. I think it's really important for everyone to have confidence in our understanding of drug allergy to the limitations we have that you can develop cutaneous reactions with full courses and not have developed one with one dose.

Dr. Mariam Hanna:

Fair. Okay, do I've been actually interested in penicillin allergy over the past year and we're trying to get a provincial program to get penicillin allergy de-labeling happening much more across the province that I'm in. One of the things that I've learned quite a bit through this project is marginalized populations and populations that live in rural settings tend to have the worst access. Well, one, to care, but two to the allergist and get this label more frequently. Is there data to support that?

Dr. Kimberly Blumenthal:

Yeah, it's such a good question. And there's actually conflicting data related to who gets labeled and who is getting de-labeled, and it's something we're really trying to focus on. As far as labeling, at least it does seem like the penicillin allergy label follows a receipt of more healthcare. And who gets more healthcare or advocates more for antibiotic use in kids might be a disproportionate amount of white patients compared to black patients or Hispanic patients. And also seems to carry with the female sex more than male sex.

But there are conflicting data, especially with children, data in children showing differences that even have that swapped. So I think that we don't have enough great population-level data. I know that Canada has great population data, so maybe somebody will be listening and want to study it in Canada generally or in your province to identify the label. Does the label follow healthcare use in sickness or does it follow advocacy for antibiotic use in children, which is a different population?

Some of what we're looking at in the Boston area has been very interesting. So de-labeling access in a place that's really allergist rich and what we found so far is that de-labeling follows who's getting referred as the sort of more midlife healthier people, well-to-do people, educated people advocating for de-labeling because they've maybe heard about it or their doctor mentioned it because they're not going over their 10 health problems at their primary care visit. The primary care doctor is then able to say, huh, you have a penicillin allergy label and why don't we just go get that evaluated?

So it's really almost a privilege at this point. To be able to get to the allergist for de-labeling, you have to have either a great indication because you're about to have a bone marrow transplant or a lung transplant or nothing much going on. And so we're trying to study these patterns to be able to more equitably evaluate penicillin allergy in our city.

Dr. Mariam Hanna:

I also learned that there are specific populations coming from other areas in the world where they've now gone one step further about worrying about amoxicillin to pre-screening patients is what I gather. And pre-screening with like skin testing or an intradermal test dose before they'll even give a dose. These patients now have immigrated to Canada and they're now on this we're trying to create is what do you do with them?

Dr. Kimberly Blumenthal:

Oh gosh, this is a great area of research too, if anybody listening wants to write up their patient cases. So we have seen that manufacturing of penicillins and other antibiotics can be taken on by certain countries and that they're equivalent to my FDA label can say to do a test on the skin before administration. So specifically China has had practices for years and years where on their label it says, before you give this penicillin, do a skin test.

I have yet to see what that skin test entails. Is it just with the diluted amount? Is it just a skin prick of the amount? But we have a lot of patients that we see who come to us and they say, I've never had penicillin, but I'm skin test positive from China. And well, at first I was thinking, well, if you were skin test positive, we should just skin test you again, right? Because we have a skin test. You'll see if you're still skin test positive.

And then I realized that they never had an exposure or reaction. So we shouldn't be skin testing people who never had an exposure or a reaction. It doesn't make any sense. They haven't developed IGE. So we have now largely realized that these are lower-risk individuals that could just get direct challenges if they're amenable if they were scared by the skin test in their childhood, that was positive. We can always offer them a skin test first. But my goodness, this is not in our literature. So somebody needs to get on that.

Dr. Mariam Hanna:

Absolutely. And another plug for unnecessary screening for allergies when there isn't a good story to match up to it. So I love that.

Dr. Kimberly Blumenthal:

Right. We can't just screen kids before they've eaten things. We can't screen patients before they're exposed to penicillin.

Dr. Mariam Hanna:

Absolutely. Absolutely. Okay, we've made it to the end. Now, time to wrap up today's episode and ask today's allergist, Dr. Kimberly Blumenthal, for her top three key messages to impart to patients and physicians on today's topic, penicillin labeling/de-labeling/some advocacy in there. All right, Dr. Blumenthal, over to you.

Dr. Kimberly Blumenthal:

Great. Number one, penicillin allergy labels are harmful to individuals and to the public at large, including increased antibiotic resistance. Number two, most penicillin allergy labels are not indicative of any meaningful penicillin allergic reaction, and the vast majority can be de-labeled with direct oral challenges of amoxicillin. And number three, it is our job as allergists to identify the severe or higher-risk individuals, those who might have systemic reactions to penicillins or severe delayed phenotypes. Because if we don't identify those people, they are going to be unsafe by all of the massive de-labeling that we are advocating for.

Dr. Mariam Hanna:

Perfect. Thank you, Dr. Blumenthal, for joining us on today's episode of The Allergist.

Dr. Kimberly Blumenthal:

Thank you so much for having me.

Dr. Mariam Hanna:

This podcast is produced by the Canadian Society of Allergy and Clinical Immunology. The Allergist is produced for CSACI by Podcraft Productions. The views expressed by our guests are theirs alone and do not necessarily reflect the views of The Canadian Society. This podcast is not intended to provide any individual medical advice To our listeners.

Please visit www.csaci.ca for show notes and any pertinent links from today's conversation. The Find an Allergist app on the website is a useful tool to locate an allergist in your area. If you liked the show, please give us a five-star rating and leave a comment wherever you download your podcasts and share it with your networks because first, do no harm and then get to work. Thank you for listening. Sincerely, The Allergist.