Serology testing can allow a no-biopsy approach to coeliac disease diagnosis
The global prevalence of celiac disease is around 1%, although it is likely that many people remain undiagnosed. Historically, diagnosis was usually made in younger children, but the condition can develop at any age and the incidence is increasing in adults. Traditionally, a biopsy has been required to confirm diagnosis, but increased endoscopy waiting times have encouraged the development of a no-biopsy approach to celiac disease diagnosis. CLI chatted to Dr Mohamed Shiha (Department of Gastroenterology, University Hospitals of Leicester NHS Trust, Leicester, England, UK) to find out more about how this can be possible and what the challenges are.
Firstly, what is celiac disease?
Celiac disease is a relatively common autoimmune disorder that affects about one in 100 people worldwide. It is unique among autoimmune conditions because we know exactly what triggers it. For people with celiac disease, consuming gluten triggers an immune reaction. The immune system mistakenly identifies the gluten as foreign rather than as food, which causes an inflammatory response that damages the lining of the small intestine. This damage leads to uncomfortable symptoms and prevents the body from properly absorbing nutrients. The positive news is that because we know the trigger, we can stop the damage. When the patient stops eating gluten, the immune reaction halts, and the bowel heals. This reversibility is what makes celiac disease so unique.
Why can diagnosis be challenging and how is it usually done?
Symptoms
Diagnosis of celiac disease can be challenging for a number reasons. The first reason is how do we know who to test for celiac disease?
The ‘classic’ presentation involves children suffering from wasting, stunted growth, diarrhea and nutrient malabsorption. While we still see this occasionally, it has become quite rare. Today, we encounter much more subtle presentations, which makes the diagnosis significantly more difficult.
There are a lot of possible symptoms, basically any gastrointestinal (GI) symptoms – upper or lower GI tract – could indicate celiac disease. For example, If you have IBS-type symptoms, you have around 6% the chance of having celiac disease.
However, the symptoms are not limited to the bowel, as celiac disease can also present with neurological symptoms, such as ataxia (loss of coordination). It can manifest as skin rashes, arthralgia (joint pain),
or even very subtle symptoms like extreme, long-term fatigue that cannot be explained by standard blood tests. Some experts even suggest that certain chronic headaches, unresponsive to normal treatment, could be linked to celiac disease.
Testing
The next step is testing. Initially, we do serology testing, which is an excellent test for celiac disease. The tests are possibly the most accurate compared to serology testing for all other autoimmune conditions.
However, diagnosis then needs to be confirmed by biopsy via endoscopy, and following the COVID-19 pandemic there is still a massive waiting list for endoscopy. This leaves us in a difficult situation: we have a patient with clear symptoms and positive serology, yet the doctor must tell them to keep eating gluten, which is actively causing them pain and damage, until they can get their endoscopy. If that appointment is months away, you can understand the frustration. Many patients simply say, ‘No thank you, I don’t need a formal diagnosis’, and start a gluten-free diet immediately. This creates new problems. We end up with many people who are not formally diagnosed and, consequently, do not receive proper follow-up care or access to dietitians. Even if they waited and had their endoscopy, unfortunately often the right number of biopsy samples are not taken. To diagnose coeliac disease, at least five to six biopsies need to be taken. Unfortunately, endoscopists often take only two or three, which can result in a missed diagnosis.
On top of that, even with the correct number of samples, the interpretation is subjective. Histopathologists do not always agree; one might look at a sample and confirm celiac disease, while another looks at the same sample and says, ‘No, it is not’.
So, at every step you can easily miss a diagnosis.
How is serological testing for celiac disease usually done?
We rely on two main serology tests. The first, and most commonly used, is for IgA tissue transglutaminase antibodies (IgA-tTG). This is our first-line test because it has excellent sensitivity and good specificity. If this test is negative, it is very unlikely the patient has celiac disease, so it serves as a great initial screen.
The second test looks for anti-endomysial IgA antibodies (EMAs). This is used less frequently but is highly specific, around 99%. However, it is not as sensitive as the tTG test, meaning a negative result doesn’t necessarily rule out the disease. Therefore, we typically use it as a secondary, confirmatory test.
What we have found with the IgA-tTG test is that the concentration of antibodies matters. The higher the level, the more likely the patient is to have celiac disease. If the levels reach around 10 times the upper limit of normal (ULN), the chance of having the disease is almost 100%.
Pediatricians recognized this back in 2012 and updated their guidelines: if levels are that high, it can’t be anything else, so there is no point in doing a biopsy. However, adult medicine lagged behind. It was only after COVID-19, faced with long endoscopy waiting lists, that we really began to question our approach. Endoscopy is invasive and unpleasant. Why keep doing biopsies if we know the diagnosis is practically confirmed by high IgA-tTG levels? Despite some initial pushback from clinicians who felt the biopsy was indispensable, our research addressed these concerns. First, we assessed accuracy: we found that for patients with IgA-tTG levels at least 10 times the ULN, the blood test is extremely accurate, with a positive predictive value of around 98% and specificity of 100%. Second, using a discrete choice experiment, we asked patients for their preference. As you might expect, the majority preferred a diagnosis by serology alone, avoiding the need for an endoscopy.
Consequently, for patients with IgA-tTG levels more than 10 times the ULN, we are shifting toward a serology-based diagnosis. The latest European guidelines, published just a few months ago, now support this pathway. This marks the first time a non-biopsy approach has been formally applied to adults.
Celiac disease is an autoimmune disease triggered by gluten that results in damage to the lining of the small intestine (Adobe Stock)
Is serology testing currently good enough for a no-biopsy approach?
The consistency of serology testing was a major concern regarding the no-biopsy approach. Since different hospitals use different tests, we needed to know if the results were comparable.
To map out the landscape, we contacted every lab in the UK. We collected data on the availability of tests, assay types, reporting methods, and crucially, their cut-off thresholds (the ULN). This knowledge was vital: if the new guideline requires a result of ’10 times the ULN’, we had to ensure that ‘ULN’ meant the same thing everywhere.
We discovered that 12 different kits are being used across the UK, with ULNs ranging wildly from 3 to 30 IU/mL. We even found variation between identical assays: one lab might use a specific kit and set the limit at 7 IU/mL, while another lab using the exact same kitsets it at 15 IU/mL.
This inconsistency means we cannot use a single, absolute antibody number as a nationwide reference. For example, a specific test result in one hospital might qualify for a no-biopsy diagnosis, whereas that same number in another hospital could be interpreted as a negative result. While the tests are accurate within their own specific labs, the raw numbers are not directly comparable between them.
Therefore, we can move forward with the no-biopsy approach, but with one condition: we must use a relative threshold – specifically 10 times the local ULN for each lab – rather than attempting to set a fixed absolute number for the whole country.
Histopathology showing normal small intestine villi (top) compared to the damaged small intestine of celiac disease (bottom) In patients with IgA-tTG levels at least 10 times the upper limit of normal, endoscopy is not necessary to confirm diagnosis. (Adobe Stock)
What improvements could be made to allow serology testing to allow a no-biopsy diagnosis of coeliac disease?
Firstly, the main issue is standardization. Ideally, we would have a universal standard, a specific number, such as an IgA-tTG titre of 300 IU/mL, that confirms celiac disease nationwide. This would be clear for both patients and GPs. However, currently, we cannot do this. We are forced to use the relative measure of ’10 times the local ULN’ specific to each lab.
Secondly, we must address the logistics of the patient care pathway. A diagnosis in primary care should not mean the patient remains limited to primary care. We must ensure patients are still referred to gastroenterology for proper follow-up and, crucially, a review by a dietitian. We want to avoid the ‘diagnose and dismiss’ scenario, where a GP simply tells a patient they have celiac disease and sends them on their way. We saw this happen in Finland, which has the highest prevalence of celiac disease in the world (over 2%). In 2018, Finland became the first country to implement serology-based diagnosis. When they reviewed the data later, they found a significant disparity. Patients diagnosed without a biopsy often just received a phone call telling them to stop eating gluten, without proper nutritional guidance. Consequently, these patients suffered from more residual symptoms over time compared to those who went through the traditional biopsy route and received structured follow-up. We must ensure that avoiding a biopsy does not mean receiving inferior care.
It is vital that patients see a specialist who can explain exactly what celiac disease is. Non-specialists may provide inaccurate advice or misconceptions about celiac disease and the strictness of the diet, which can lead to ongoing symptoms. Alternatively, patients might start a gluten-free diet before a formal diagnosis. This creates a trap: once they are on the diet, we cannot test them accurately. Trying to convince a patient to reintroduce gluten, knowing it will make them ill, just to get a blood test or biopsy is incredibly difficult. A clear care pathway prevents this situation.
Finally, simply stopping gluten and feeling better is not a valid diagnosis. While only 1% of the population has celiac disease, about 10% report symptoms when eating gluten. Most often, this is not celiac disease but non-celiac gluten sensitivity, or a functional disorder like IBS. This is why a concrete diagnosis is essential. Without it, we risk labeling thousands of people as ‘celiac’ when they actually need treatment for something else – perhaps a reaction to fructans, or underlying psychological comorbidities like anxiety or depression. A referral to a dietitian helps pinpoint the actual cause, which is often not gluten at all, ensuring the patient gets the specific treatment they need to get better.
The interviewee
Dr Mohamed Shiha MBBCh, PhD
Specialist Registrar in Gastroenterology
Department of Gastroenterology,
University Hospitals of Leicester NHS Trust, Leicester, England, UK
Email address: mohamed.shiha1@nhs.net
Bibliography
1. Shiha MG, Nandi N, Raju SA, Wild G, Cross SS et al. Accuracy of the no-biopsy approach for the diagnosis of celiac disease in adults: a systematic review and meta-analysis. Gastroenterology 2024; 166(4):620–630 (https://doi.org/10.1053/j.gastro.2023.12.023).
2. Shiha MG, Wickramasekera N, Raju SA, Penny HA, Sanders DS. Patient preferences for the diagnosis of coeliac disease: a discrete choice experiment. United European Gastroenterol J 2025; 13(3):330–337 (https://doi.org/10.1002/ueg2.12651).
3. Shiha MG, Schiepatti A, Manza F, Maimaris S, Aziz I, Sanders DS. Global prevalence of celiac disease in patients with Rome III and Rome IV irritable bowel syndrome: a systematic review and meta-analysis. Am J Gastroenterol 2025;120(12):2776–2787 (https://doi.org/10.14309/ajg.0000000000003586).
4. Al-Toma A, Zingone F, Branchi F, Schiepatti A, Malamut G et al. European Society for the Study of Coeliac Disease 2025 updated guidelines on the diagnosis and management of coeliac disease in adults. Part 1: diagnostic approach. United European Gastroenterol J 2025;13(10):1855–1886 (https://doi.org/10.1002/ueg2.70119).
5. Alex A, Hong AS, Dimmock M, Zaman MI, Adam M et al. Nationwide survey of coeliac disease serology testing in the UK. BMJ Open Gastroenterol 2025;12(1):e001900 (https://doi.org/10.1136/bmjgast-2025-001900).
6. Shiha MG, Manza F, Figueroa-Salcido OG, Ontiveros N, Caio G et al. Global prevalence of self-reported non-coeliac gluten and wheat sensitivity: a systematic review and meta-analysis. Gut 2025;gutjnl-2025-336304 (https://doi.org/10.1136/gutjnl-2025-336304).
7. Sareila H, Kurppa K, Huhtala H, Laurikka P, Arnala S, Koskela T et al. Patient perceptions of the Finnish guidelines enabling coeliac disease diagnosis without biopsies in adults. Scand J Gastroenterol 2025;60(1):20–27
(https://doi.org/10.1080/00365521.2024.2431628).
