Gut Check: Benjamin Lebwohl, MD, on Celiac Disease

Dr Brian Lacy hosts celiac disease expert Dr Benjamin Lebwohl to explain the prevalence of the disease, the complexities of accurate diagnosis, and the importance of not relying on symptoms alone.

Brian Lacy, MD, is a professor of medicine at the Mayo Clinic in Jacksonville, Florida. Benjamin Lebwohl, MD, is professor of medicine and director of clinical research at the Celiac Disease Center of Columbia University in New York City.

Clinical Brief: Celiac Disease Diagnosis and Testing—Key Clinical Updates

• Celiac disease: Prevalence is approximately 1% overall and may approach 2% in children <10 years; many cases remain undiagnosed into adulthood. Presentation extends beyond diarrhea to include IBS-like symptoms, iron deficiency anemia, elevated liver enzymes, recurrent miscarriage, migraine, peripheral neuropathy, or asymptomatic disease detected through family screening.

• Diagnosis: Test before initiating a gluten-free diet. Tissue transglutaminase (TTG) IgA plus total IgA is the preferred initial evaluation while patients are consuming gluten. For patients already avoiding gluten, consider HLA-DQ2/DQ8 testing and, when appropriate, a gluten challenge; symptom improvement alone does not confirm celiac disease.

• Management: In the US, diagnosis generally requires duodenal biopsy with 4–6 specimens (including the bulb). Long-term care emphasizes strict gluten avoidance, recognition of hidden gluten exposure, monitoring bone health, and awareness of the small but increased risk of enteropathy-associated T-cell lymphoma while reassuring patients that overall prognosis is generally favorable.

Welcome to Gut Check, a podcast from the Gastroenterology Learning Network. My name is Brian Lacy. I'm a professor of medicine at the Mayo Clinic in Jacksonville, Florida, and I am absolutely delighted to be speaking today with an international expert, Dr. Benjamin Lebwohl, Professor of Medicine and Director of Clinical Research, the Celiac Disease Center at Columbia University in New York City. Our topic today is one that affects healthcare providers of all specialties: celiac disease.

So Dr. Lebwohl, welcome. It's hard not to open a scientific journal or even a major newspaper these days without seeing a headline on celiac disease or gluten sensitivity. Let's start simply. How common is celiac disease?

Dr Lebwohl:

Thanks, Brian. It's a pleasure to be here. Celiac disease is common. The traditional sort of textbook figure that we put out there for the past 20 years has been about 1%. Studies going back to the early 2000s found more precisely 0.8%, so we'd round up. But actually more recent studies, particularly coming out of screening-type studies for children have suggested that that number is inching up. It might be closer to 2% in children under 10, so super common and still not fully recognized. Studies suggest that a solid proportion of people with celiac disease remain undiagnosed for at least a portion of their lives. And if you survey US adults asking when were you diagnosed with celiac disease, the median age is still in the 30s when in fact we know if you follow infants at risk for celiac disease, most people develop it sometime in the first 2 decades of life. So there's still undiagnosed patients out there for us to test for and diagnose.

Dr Lacy:

Something for all of our listeners to be aware of. And we have a lot of really smart people listening in today, but the presentation of celiac disease varies from patient to patient. And what are some of the most common symptoms and signs that healthcare providers and patients should be aware of?

Dr Lebwohl:

Well, in the era before we had the TTG IgA test for celiac disease, and we weren't using that, we were predominantly focused on GI symptoms. It's still the case that a good number of patients, perhaps the majority of patients, will report some gastrointestinal disturbance. It's no longer the case that the majority has diarrhea and clinical malabsorption such as weight loss. But if you add in people who also have constipation-predominant IBS-type symptoms, more than half of all people with celiac disease will report some gastrointestinal disturbance.

But then there's a long list of so-called nonclassical modes of presentation and that includes iron deficiency anemia, other causes of anemia such as B12 deficiency, elevated liver enzymes, recurrent miscarriages, migraine headaches, peripheral neuropathy. It's a real sort of mishmash and grab bag, which makes case finding difficult. And then there are some people who have no symptoms whatsoever, but they get picked up as part of, for example, family screening. If you've got a first-degree relative, you might end up getting screened for celiac disease.

Dr Lacy:

Wonderful. Or that occasional patient with elevated liver chemistries, and that's their one sign, but they're asymptomatic. So wow, we really need to keep our blinders off, don't we? So a lot of patients or many patients with symptoms that they believe are related to ingesting wheat, stop eating those products and they feel better. Is that sufficient proof to make a diagnosis of celiac disease? Do we need to do other testing?

Dr Lebwohl:

It is not sufficient proof. Many people find that when they go low on wheat, gluten-related food ingredients or components, they feel better, particularly with regard to bloating. This is what we call a nonspecific response. For a lot of people, that is because a gluten-free or low-gluten diet correlates a lot with a low FODMAP diet and that has been shown to be effective for a number of different disorders of gut-brain interaction, and bloating particularly often gets better with a low FODMAP diet. So it could be due to that. There have been a number of double-blind challenges where people who report a response to a gluten-free diet but who don't have celiac disease are given either gluten or a nongluten placebo and it turns out that the so-called nocebo effect is very high. When people think that they're being exposed to gluten or wheat, they often report symptoms.

This has been shown in many different contexts with many different so-called active comparators or placebos, whether it's rice or some other compound. And so expectation appears to be a big part of this. So we can't rely on symptoms alone. And what we really try to impart upon our patient community and the physician community is that you should test first before you go gluten-free or before you encourage a patient to go gluten-free because once you start the gluten-free diet, our ability to detect celiac disease really diminishes and the sensitivity of the celiac disease blood tests starts to go down. It can take about a year on average for it to normalize in people with celiac disease. But in some people that can become normal more quickly, especially people with more mild elevations. The same goes for histology. It will eventually normalize if you go gluten-free. And if we have someone who's been on a gluten-free diet for years before getting tested for celiac disease, it's a more challenging situation.

Dr Lacy:

So that's a great teaching point that actually, although we try to be efficient in terms of healthcare costs, we don't want to over-test in this situation because symptoms are so nonspecific, testing early on may be the best way to do this. And so let's think about the general population and we have a healthcare provider who wants to test a patient for celiac disease. What's the most effective and efficient means of testing somebody for celiac disease who is still ingesting wheat?

Dr Lebwohl:

The best test we have remains the tissue transglutaminase IgA. We call it the TTG, but be careful when you're ordering it that you choose the TTG IgA because that is the one that has the highest sensitivity and specificity, overall high accuracy. But typically when we test it, we also add a total IgA and we do that because a subset of people in a population have selective IgA deficiency and people who don't make IgA will not make a TTG IgA even if they have celiac disease. And so we might miss such people. And so if a TTG IgA is checked and it's negative and the total IgA is normal, then that is a negative test result. It makes celiac disease very unlikely. If, however, the total IgA is undetectable, then we're dealing with a patient who appears to have selective IgA deficiency and then we jump to second line tests.

There are IgG forms of TTG and there's an IgG gliadin and peptide antibody that can be useful. But what I tell trainees in terms of if you think about celiac disease and you have a desert island test you want to lean on, it's the TTG IgA. It's the best one we have.

Dr Lacy:

Simple, easy, sensitive, specific, not expensive, high yield.

So we kind of discuss the fact that a lot of patients stop eating wheat products and they seem to improve a variety of symptoms, including a lot of GI symptoms. So let's say you see that patient in that office and they've been off wheat products for about a week, but they are very concerned or convinced that they might have celiac disease. So how do we test those patients? Do we ask them to challenge themselves with gluten and then do the simple TTG antibody test with a serum IGA or do we do genetic testing or endoscopy? What do you think is the best approach?

Dr Lebwohl:

Sure. It depends a little bit on the details of the clinical case. For example, if someone's only been off weed or gluten for about a week, that's okay. You can test right now today in the clinic with a TTG IgA because it's pretty unlikely that that would normalize so quickly in someone with celiac disease. But after a few weeks or a month, then I get nervous that this is a patient that might have celiac disease, but they're not keeping enough gluten in their diet for it to be detected. And so when that's the case, there are a couple things we do. One is we offer genetic testing for dq2 and dq8. These are components of our HLA immune system and these are very common haplotypes; 40% of us are positive for dq2 or dq8. And the great majority of people with that result, it doesn't really have any significance.

But the reason we do it here is if you don't have dq2 or dq8, that essentially rules out celiac disease. They are necessary but insufficient steps for having celiac disease and they don't change. You're born with these genes. So it's very helpful for someone who's already on a gluten-free diet. The other thing that we consider is this involves some shared decision-making. What is the patient's willingness to start eating gluten again? So if they get a genetic test and they have a compatible gene that doesn't make the diagnosis, it just keeps them in the game. It's still possible that celiac disease is present. And so then the most effective way to determine someone's celiac disease status is a gluten challenge. That involves eating gluten and then getting tested. Now the details for how to do a gluten challenge will vary a little bit between guidelines. And frankly, if you go to 3 different celiac disease specialty centers, you're not going to get identical instructions in part because the evidence is, I would say, sort of sporadic that different groups have tried different approaches.

The general principle is the more gluten you give someone and the longer you give it to them for, the greater the ability to bring out celiac disease. But we have to balance that with what will patients tolerate? How long can patients continue to eat gluten every day and potentially suffer symptoms? Not everyone's willing to go through a gluten challenge, but for those who do, we've actually studied this. We find that about 15% of patients who go through the gluten challenge so as to figure out if they have celiac disease, end up having celiac disease on that full set of testing, including an intestinal biopsy. Some people are more likely to have celiac disease than others and the biggest predictors of having celiac disease at the end of that are one, having any positive serology in your history. If you go through your chart and there was something that showed positive, regardless of what you were eating and when, those people were more likely to end up having celiac disease.

And the other predictor of having celiac disease is an absence of bloating as part of your symptomatology. So if someone reports that when I eat gluten I bloat and that's their primary symptom, that person is actually less likely to have celiac disease at the end of the day. So this was a single center study we published a couple of years ago and it really is, I would say, useful information we can impart to patients when you're trying to consider whether to go through a gluten challenge.

Dr Lacy:

That last teaching point is so important because bloating is such a common but nonspecific symptoms. So that's really fascinating. And that's kind of a perfect segue because I want to go back to something you said earlier. Let's take that patient who had some symptom, maybe not bloating in this case, but they had some symptom. You did the blood work, their TTG antibody is elevated, they have appropriate IgA levels. So now what do we do? Is an upper endoscopy with duodenal biopsies required or is that symptom or symptoms and a positive TTG antibody or a positive genetic test?

Dr Lebwohl:

The answer is right now it depends on who you are and where you are. If you are in the United States, the guidelines all state, everyone really should have a duodenal biopsy to confirm the diagnosis of celiac disease, but children in Europe can follow European guidelines that in some cases allow for a biopsy sparing approach, which requires highly elevated TTG, 10 times elevation or greater, plus a second positive blood test they call for an endomysial antibody—not really because that's so much more specific, but because frankly, you want to make sure that there wasn't a laboratory mix up. It has to be done on a separate occasion and before you put someone on a lifelong gluten-free diet and you also want to make sure that it wasn't a transient phenomenon. There are these cases where someone will spike a TTG perhaps in the setting of a recent infection and then it actually over time normalizes and that person might not need a lifelong gluten-free diet.

In adults, this has been slower to catch on in part because the study suggests that the positive predictive value of a highly elevated TTG is not quite as high as it is in children. Studies on both sides of the Atlantic suggest it's about 95%. That sounds high, but that means that if you do this for everyone, then 5% of the time or 1 in 20 people, you are potentially telling them to stay away from gluten for the rest of their life when maybe it wasn't necessary and that's a burdensome thing to prescribe if it's not necessary. So it hasn't made its way to the US though some guidelines a few years ago suggested that if you have someone who meets these criteria than 10 times greater elevation of TTG IgA or more, and they have a secondary test that again shows an elevation and they're unwilling or unable to start a gluten challenge, you could consider them as likely having celiac disease as a sort of after the fact diagnosis.

The rationale there is that there are people in this situation and we want to discern between those who really likely have celiac disease and those whose diagnosis is on much shakier grounds. One day we hope there will be a medication approved for celiac disease and we don't want people like this to be left out. So we want to have some sort of way for people who didn't have the biopsy to still be considered to have likely celiac disease.

Dr Lacy:

Wonderful. So we've got a lot of listeners who may be performing upper endoscopies and for our patients too who may be undergoing upper endoscopy, this seems like a simple question, but a duodenal biopsy seems easy, but there's a real nuance to this. How do you like to take duodenal biopsies to confirm or make an accurate diagnosis of celiac?

Dr Lebwohl:

When I'm doing a duodenal biopsy and celiac disease is on the differential, I take 4 to 6 specimens from multiple portions of the duodenum, the duodenal bulb and the second portion of the duodenum. This is in accordance with what guidelines have been saying in recent years. The rationale behind this is that celiac disease can present in a patchy way histologically and you might have people who have normal villi adjacent to subtotal or totally atrophic villi and you don't want to miss those pieces. And so I typically look for 4 in D2 and I look for 2 in the bulb and guidelines typically say 4 to 6, including 1 to 2 from the bulb. People who only have celiac disease in the bulb are so- called ultra-short celiac patients. That refers not to their stature, but to where in a small bowel is the celiac disease found? And there are some people, maybe up to 10% of people with celiac disease, they only have villus atrophy in a duodenal bulb. It sort of makes sense because if you think about where's gluten most highly concentrated in a small bowel, it's going to be where it first hits the duodenum and then it dissipates. So why some people only have it in the bulb and others don't, it could be because of that. Whether this matters clinically, I would say right now as a matter of debate. Studies of ultra-short celiac disease have found that they tend to have more mild symptoms. They tend to have less elevation of their TTG IgA. So instead of 10 or 20 times upper limits of normal, it might be just be slightly above the upper limit of normal or double the upper limit of normal. I would say a hot topic and matter of debate that you sometimes hear chatter about in the celiac disease conferences is, is this really the same as regular celiac disease, particularly with regard to its permanence?  Right now the traditional teaching is people with this, they need to be treated for a lifelong gluten-free diet.

I think there are some who are interested in knowing, well, maybe the cat's not yet out of the bag. Maybe this could still be transient and reversible. No guideline says that such people can continue to be watched and monitored on a regular gluten-containing diet. The traditional teaching is that ultrashort celiac disease, like any other celiac disease should be treated with a permanent gluten-free diet. But I would say that we're at the stage now where this is now being referred to as dogma, and you know dogma has a little bit of a pejorative valence. And people are starting to question it. I hope that we do so in the context of trials and studies.

Dr Lacy:

For what seems like an easy topic, celiac disease and just avoid wheat, there are so many nuances to this and I think I like what you said a lot, meaning that we should be constantly challenging thought. This is how we advance science and medicine.

So you mentioned earlier so many adult Americans are on a gluten-free diet for any number of reasons, maybe just to improve bloating. Can you help our listeners with the distinction between celiac disease and nongluten wheat sensitivity? Really, what does that term mean? And is that still the best term, nonceliac wheat sensitivity for patients who don't have celiac disease, but have symptoms while ingesting wheat?

Dr Lebwohl:

So this is still a difficult topic to really expound upon with confidence. I would argue that we should approach this topic with humility because there's so much we don't know and understand. In terms of this entity, it still has a few different names or guises when you look at the medical literature. I think the most common one remains nonceliac gluten sensitivity. Others call it nonceliac wheat sensitivity. There was on acronym that was used for a short period, PWAG,,people who avoid gluten or people who avoid wheat or gluten PWOG. I think that this is a reflection of our uncertainty about what exactly this entity is. We understand celiac disease much better than we understand this other entity. Celiac disease has known biomarkers. They make TTG IgA or in certain cases like people with IgA deficiency, these IgG homologs and they have duodenal villus atrophy and their natural history has been very well studied.

There have been multiple large scale epidemiologic studies that link celiac disease to certain cancer risks such as lymphoma as well as small bowel adenocarcinoma and possibly other upper GI cancers. And it is also associated with metabolic bone disease and osteoporotic fracture risk. And so because of these associations that are known, we have an approach to diagnose- monitor-surveill.

With people who don't have celiac disease but report a symptomatic response to gluten exposure, we don't have that kind of high quality epidemiologic data because we lack a biomarker and we lack a clear way to follow such patients longitudinally. We're limited to smaller scale types of studies in terms of numbers of centers, which are typically specialized centers. We often have more trouble following people longitudinally over many years because it's unclear how to actually code for such patients in terms of medical coding. And we can't just go to Sweden and look at biopsy reports as our ground truth.

So because of that, much less is known about long-term natural history. I will tell you as a clinician who sees a lot of such patients, it seems like it's very heterogeneous and people without celiac disease who nevertheless avoid gluten, some of them are happy to do so for the long-term and others seem to regain the ability to tolerate gluten or wheat. It's very common that if you listen carefully to these patients and you do additional evaluation, you'll find something else or something more. You might find that they meet criteria for irritable bowel syndrome. You might find that they have gastroparesis or other food intolerance on testing. And then there's some people for whom it really was a transit phenomenon and they are now able to eat wheat or gluten without any difficulty. That's often the time that I say, okay, great, let's make sure you don't have celiac disease because you're now back on a regular diet and they almost always do not have celiac disease.

So I anticipate that as we learn more about this condition, we're going to find that it's heterogeneous. It's likely more than one condition and likely other conditions under the guise or in some cases mistaken culpability of gluten, but maybe we'll find a biomarker and separate out some others from it really is gluten or another component of wheat.

Dr Lacy:

Wonderful. So we mentioned before that if you have true celiac disease, you just go on a gluten-free diet and boy, that sounds so easy. Just avoid your pasta and bread and rolls and crackers, but we all know it's really complicated. You've done some really interesting research in this area. Tell us a little bit about some of those foods commonly considered to be gluten-free but maybe are really contaminated.

Dr Lebwohl:

So gluten can be hiding in various guises and this includes food items that you might not immediately associate with gluten. A common one is soy sauce, actually. So that's one where components of wheat are included in the process of creating the soy sauce, generating soy sauce. And gluten, which is a protein component of wheat, might be present and that has been a culprit. Another common area is cross-contact, sometimes referred to as cross-contamination, where a food item might not intrinsically have any gluten, but it's prepared in such a way that it's in close contact with gluten-containing items. A shared fryer is one that is a common concern. So you can have French fries that are gluten-free in terms of their ingredients, but they might be in a fryer that was just used to fry gluten-containing items and a restaurant might not be changing that fryer even for days.

And so that's another potential area. When we have our patients see our expert dieticians, they really want to take a full inventory of the kind of things people are eating in the home but also when eating out. Gluten-free pizza is another example. It can be entirely gluten-free in terms of its ingredients, but if it is being prepared in a milieu like an oven right next to a gluten-containing pizza and the handler isn't washing their hands when they touch one and then touch the other, you can imagine the potential for cross-contact. So hidden sources really abound. Gluten is everywhere and this can make it really difficult to navigate this diet.

Dr Lacy:

It's so tricky, isn't it? And that's a perfect segue too in terms of cooking and eating out and being at restaurants. And we know that for many people, they cook less at home now than they used to. There's a lot more going out to eat or ordering in or picking up food. When restaurants say it's gluten-free, you've kind of touched on this a little bit, but they say it's gluten-free. Are we confident about that? Do we need to use a device? What do we do?

Dr Lebwohl:

Well, I would say that the device that's most important to use is your own brain in terms of your judgment because ultimately this will vary tremendously and it's a big challenge for our patients to try to figure out is this place safe? We can sometimes rely on the wisdom of the crowd. There are communities, including app-based approaches to understanding certain restaurants and other eateries with regard to how trustworthy they are and the kind of care they take. There are restaurants that are entirely gluten-free. Those tend to be congregated in just a small number of cities. They tend to be more pricey and cater to really only a relatively small proportion of the population.

There are ways to try to understand if the food is safe beyond just asking the food handler, food server, looking them in the eye to get a sense of how serious they're taking your concern. There is gluten detection technology. There are now devices out there where you can test food before you eat it. There's something called the NIMA sensor that's commercially available. There's another one called LEO that's being marketed that patients can purchase. They appear to detect gluten with some important caveats. They might not detect every type of gluten, so fermented gluten like soy sauce that's not going to work. And ultimately they're just testing a small proportion or sample of what's on your plate. And so it is not a guarantee. Now, whether to recommend the use of this technology to our patients remains an open question. We have real clinical equipoise about this and we don't know whether to recommend gluten detection technology.

There's another kind of technology out there that measures gluten after you may have eaten it. So this is for people who are developing symptoms and they want to know, did I get glutened? And you can test for gluten in stool and gluten in urine. That latter test picks up a potential gluten exposure in the last 12 or 24 hours or so. And we are actually subjecting that technology to a randomized trial where we're enrolling newly diagnosed patients with celiac disease and randomly assigning them either the standard management, which involves regular dietician follow-up or all of that. Plus we give them these urine kits and tell them to test themselves regularly—weekly if they have no symptoms plus on demand if they have symptoms, because we don't know whether testing and getting this additional knowledge will make people better off. We're looking at intestinal healing, that's our primary endpoint, but we're also looking at symptoms, quality of life, anxiety, et cetera. So we're hoping that the results of this trial will lead to actionable advice that we can give our patients and we can learn something from this.

Dr Lacy:

Wow, very exciting. Looking forward to hearing those results.

So as we wind down here, maybe you could educate our listeners a little bit about some of these downstream consequences. Let's take somebody with documented celiac disease, they're on a gluten-free diet, knowing that sometimes it sneaks in despite everybody's best efforts. What are some of the worst things we can think about in terms of long-term consequences that maybe they should bring to the attention of their healthcare provider to check? You mentioned iron deficiency anemia and bone disease. Anything else we should be aware of?

Dr Lebwohl:

So the first thing I impart upon patients who are asking about this is that the great majority of people with celiac disease live long lives full of good health and the celiac disease is unlikely to shorten their lifespan considerably. That is important to start at because there are a lot of studies out there that are measuring associations with outcomes that can be concerning, even frightening. These outcomes are real. These associations are real, but they are small in magnitude. They speak to the importance of celiac disease and the need to treat it, but still the great majority of people do not have a significantly shortened lifespan because of their celiac disease.

That said, the things we worry about I would say are bone health, cancer. Those are the 2 top ones that we talk about. In terms of bone health, there is an increased risk of osteoporosis and fractures. This tends to actually be mitigated by the gluten-free diet. If you look at bone density, particularly in people who are still around the age where they're putting on bone mass, bone density often improves after starting a gluten-free diet. For that reason, I often don't check bone density immediately at diagnosis. I recommend waiting a little bit, up to a year because if they're on a borderline low bone density and they start the gluten-free diet, I want them to have a good result. And so monitoring their bone health is important.

And the other is cancer risk. Now the cancers that are most increased in people with celiac disease are something called enteropathy-associated T-cell lymphoma and then other kinds of non-Hodgkin's lymphoma. Thankfully, these still remain rare. Someone with celiac disease is much more likely to have that kind of lymphoma than someone without celiac disease, but 99 plus percent of people with celiac disease never get this lymphoma.

And so people with celiac disease, they're more likely one day to get colon cancer just like everyone else. It's a super-common cancer. There's actually most studies show no increased risk of colon cancer in people with celiac disease, but because it's so common, they should be more concerned about that and make sure that they get their colon cancer screening. And in fact, there's no additional screening for cancers that we do in celiac disease compared to someone without. We just might be more vigilant. For example, if someone's presenting with symptoms of, like a lymphoma—unintentional weight loss, night sweats—we're more likely to initiate a workup for it. But we tell people with celiac disease first to be reassured that these risks while documented are not high in absolute magnitude. And second, that they should look at absolute risks and not just what special diseases their celiac disease puts them at risk for because heart disease, cancer, the common things that we should be concerned about, monitor for, and prevent, that counts for celiac patients too.

Dr Lacy:

Yeah, that's a great way to put things in perspective because it's easy for patients to become overly concerned about some of these rare things. Ben, this has been an amazing discussion. Thank you so very much. Any last comments for our listeners?

Dr Lebwohl:

So I just appreciate the opportunity to be here and I appreciate the listeners for taking celiac disease seriously, thinking about it, maybe testing for it more than you ordinarily might, and then treating people with the diet that they need and deserve.

Dr Lacy:

And as you mentioned, maybe testing earlier.

So Dr. Lebwohl, once again, thank you for lending your expertise on this important topic to our listeners on Apple, Spotify, and other streaming networks. I'm Brian Lacy, professor of medicine at the Mayo Clinic in Jacksonville, Florida. You have been listening to Gut Check, a podcast from the Gastroenterology Learning Network. Our guest today was this international expert, Dr. Benjamin Lebwohl, professor of medicine at Columbia University in New York City. I hope you found this just as interesting as I did and just as enjoyable and I look forward to having you on future Gut Check podcasts. Stay well.