In the past few years it has become apparent that “classic” coeliac disease represents the tip of the iceberg of an overall disease burden
In part 1 we looked at coeliac disease which affects 1% of the population worldwide, but forms only the tip of the gluten iceberg. In this post I will examine the research around the emerging clinical entity known as Non Coeliac Gluten Sensitivity which affects at least six times as many people.
- Non Coeliac Gluten Sensitivity (NCGS) shares certain features with coeliac disease (CD): symptoms are primarily gastrointestinal (GI), but commonly also systemic (mainly neurological), and symptoms improve or fully resolve on a gluten-free diet (GFD).
- However, unlike coeliac disease NCGS has no identified genetic risk factors (e.g. HLA-DQ2/8 which is associated with coeliac), and has no specific serum markers, making it a diagnosis of exclusion. Furthermore, NCGS is not an autoimmune disease and does not show small intestine atrophy.
- Defining and diagnosing NCGS is problematic, but the prevalence of NCGS appears to be six times higher than coeliac disease.
Early evidence for gluten sensitivity outside of coeliac disease
The first confirmation of gluten reactions in non-coeliacs goes back to a paper in Gastroenterology in 1980:
Eight adult female patients suffering from abdominal pain and chronic diarrhea which was often incapacitating and frequently nocturnal, had dramatic relief on a gluten-free diet and return of symptoms after gluten challenge.
These patients are typical of what is now called Non Coelaic Gluten Sensitivity (NCGS) – predominately female, IBS-like symptoms, improving on a gluten free diet (GFD).
By 2011 it was clear that gluten pathology outside coeliac disease was real. A consensus meeting summarised it thus:
A decade ago celiac disease was considered extremely rare outside Europe and, therefore, was almost completely ignored by health care professionals. In only 10 years, key milestones have moved celiac disease from obscurity into the popular spotlight worldwide. Now we are observing another interesting phenomenon that is generating great confusion among health care professionals. The number of individuals embracing a gluten-free diet (GFD) appears much higher than the projected number of celiac disease patients, fueling a global market of gluten-free products approaching $2.5 billion (US) in global sales in 2010. This trend is supported by the notion that, along with celiac disease, other conditions related to the ingestion of gluten have emerged as health care concerns.
Non Coeliac Gluten Sensitivity (NCGS)
The term Non Coeliac Gluten Sensitivity first appeared in the Lancet in 1978 in a paper by Ellis and Linaker entitled “Non coeliac gluten sensitivity?” [ref]. However, its use as a name for real medical condition has only been in accepted use since 2011 when gastroenterologists gathered in Oslo to reach a consensus on what this new gluten sensitivity phenomena should be called.
Prior to this meeting a plethora of terms had arisen to describe coeliac and gluten-related disorders, including: asymptomatic celiac, atypical, classical, latent, non-classical, paediatric classical, potential, refractory, silent, subclinical, symptomatic, typical, CD serology, CD autoimmunity, genetically at risk of CD, gluten ataxia, gluten intolerance, gluten sensitivity and gliadin-specific antibodies. It was a giant mess, with different researchers using different terminology. The Oslo meeting attempted to sort it out once and for all. I have summarised the key terminology they agreed on at at the end of this post, but for now, let’s focus on their definition of Non Coeliac Gluten Sensitivity:
The term NCGS relates to one or more of a variety of immunological, morphological or symptomatic manifestations that are precipitated by the ingestion of gluten in people in whom coeliac disease has been excluded.
Initially this appears to be very reasonable, but the phrase “immunological, morphological or symptomatic manifestations upon gluten exposure” has proven problematic. For example, studies since that meeting in Oslo have shown that exposure of the small intestine cells to gluten elicits morphological changes in everyone [ref]. Taken at face value the Oslo definition would classify the entire population as NCGS! Similarly, innate immune activation and raised wheat protein antibodies are being identified in a large percentage of the apparently healthy population. Consequently, in just three short years the working definition of NCGS has been narrowed to include only the symptomatic manifestations, that is, NCGS now only includes people who have gastrointestinal symptoms which improve on a gluten free diet (GFD).
Problems defining NCGS
NCGS is a very recently identified condition. Consequently, the medical understanding of it is still very much in its infancy, similar, perhaps, to where we were with coeliac disease twenty years ago.
There are three areas of uncertainty that make the definition of NCGS and its diagnosis difficult:
- The causative factor is still unclear. Multiple components of wheat and other grains are implicated (e.g. gluten, ATIs*, FODMAPS*). There may be different subsets of patients that react to different components.
- There is no definitive blood or histological test. Although NCGS patients typically have blood serum abnormalities e.g. 66% have raised IgG anti-gliadin antibodies (AGA) compared to 10% of the healthy population (see second video below). Diagnosis therefore, is based on symptom improvement on a gluten free diet (GFD)
- Diagnostic criteria are problematic. The current ‘gold standard’ diagnostic test requires worsening of symptom following a double blinded gluten challenge after symptomatic improvement on a gluten free diet. However, standardising this test is proving difficult (see first video below).
These problems make studying the condition difficult. Even the name NCGS is problematic: Non Coeliac Gluten Sensitivity should only strictly apply to patients who react to pure gluten, and would exclude those who react to ATIs*, FODMAPs* or other grain components. Consequently, some researchers use the term Non Coeliac Wheat Sensitivity, but this implies that other gluten containing grains, such as barley and rye are safe. The term I prefer, “Non Coeliac Gluten Containing Grain Sensitivity” is just too much of a mouthful. (Pun intended!)
* POSSIBLE COMPONENTS OTHER THAN GLUTEN CONTRIBUTING TO NCGS
Amylase trypsin inhibitors (ATIs)
These are part of the non-gluten proteins in wheat and related grains. A 2014 study demonstrated that they can cause intestinal inflammation via the innate immune system, providing an alternative mechanism in ‘gluten’ related pathology [ref]
Fermentable Oligo-Di-Monosaccharides and Polyols (FODMAPs)
These are short chain carbohydrates resistant to digestion that find their way into the colon where they can be fermented by microbes. The resulting gas can cause painful bloating and osmotic diarrhea. Wheat contains high levels of FODMAPs called fructans, which may contribute to some of the symptoms seen in NCGS. [ref] FODMAPs are also present in other foods such as onions.
There are many other grain components implicated in disease processes, which are considered potential candidates in NCGS, including wheat globulins, gluten ‘exorphins’ and wheat germ agglutinin.
Homing in on NCGS as a defined clinical entity
So as we have seen, to be classified as NCGS a patient has to have a reduction of symptoms when on a GFD where coeliac disease (CD) and wheat allergy(WA) have been ruled out. This on its own however is insufficient as patients know they are on a GFD and may be subject to the placebo effect. Many researchers are calling for a ‘gold standard’ test where suspected patients are subjected to a double blinded gluten challenge.
Enter the Salerno Experts group, who recently convened to develop a protocol for identifying NCGS (Diagnosis of Non-Celiac Gluten Sensitivity (NCGS): The Salerno Experts’ Criteria Nutrients, June 2015).
In the video below, from the 16th International Coeliac Disease Symposium 2015, one of the Salerno Experts, Dr Luca Elli, unpacks the complexities of studying NCGS.
- (0:00) Gluten and gastrointestinal functional disorders
- (1:15) No biomarkers for GI functional disorders; symptoms and treatment
- (2:25) Up to 50% placebo effect of food exclusion diets
- (3:15) NCGS definition
- (4:17) The need for a blinded gluten challenge
- (5:18) Problems in devising the protocol
- (6:45) Review of NCGS trials, protocols and findings
- (11:00) The GLUTOX trial results
- (13:04) Prevalence of NCGS in GLUTOX trial = 20% of GI Functional patients
- (13:30) Best practice for blinded gluten challenge
- (15:00) Conclusions
You will see from this video that NCGS patients are being identified from among patients with ‘Gastrointestinal functional disorders’ (e.g. IBS). Hence, current studies and prevalence estimates for NCGS do not include people who have gluten induced non-gastrointestinal symptoms, such as neurological or skin manifestations, yet in my clinic I regularly deal with patients that have these symptoms with only minor GI involvement, and who improve on a GFD. They are not coeliac. They are not NCGS. Oslo fails to give me a name for them! This is why the gluten iceberg goes deeper still – and that will be covered in part 3.
Dr Elli’s own study (the GLUTOX trial) found that just over 80% of the enrolled patients who had gastrointestinal functional disorders reported improvements when on a strict GFD. Of those that improved, only 1/3 of them relapsed when subjected to a randomised gluten challenge. The other 2/3, therefore may have been reacting to other components of the wheat, or were under the influence of a placebo effect. [Details: The GLUTOX Trial: Getting closer to identifying nonceliac gluten sensitivity]. EDIT (28/08/13): a similar proportion of patients responding to a blinded gluten challenge (1/3) has just been found in this study.
Regardless of these difficulties, the term NCGS has become a household name, at least among nutrition journalists and a subset of the public. UK doctors have yet to catch up. However, it is important to realise that even the current usage of the term ‘NCGS’ in the literature and media, by no means covers all research related to the pathology of gluten and grains. Hence, even with NCGS we have not reached the bottom of the gluten iceberg. In post 3 of this series I will look at the effects of gluten beyond coeliac and NCGS.
Case 1 – NCGS Skin Rash
Italy leads the field in the study of pizza pasta gluten related disorders
As the understanding of coeliac disease increased throughout the 2000’s specialist centres emerged all over the world. It is ironic, perhaps, that researchers from the great pasta and pizza eating nation of Italy have risen to prominence in the field of gluten related disorders. Both the videos and many of the references in this article are by Italian authors.
And here is another: a one year survey of suspected NCGS patients identified at 38 Italian centres [ref] found the following which is a typical picture of what other researchers find…
- 486 patients with suspected NCGS were identified in this 1-year period.
- The female/male ratio was 5.4 to 1, and the mean age was 38 years (range 3–81)
- The clinical picture included gastrointestinal and systemic manifestations (see graphs below)
- Reaction to gluten was quick: appearance of symptoms varied from a few hours to 1 day.
- Associated disorders included irritable bowel syndrome (47%), food intolerance (35%) and IgE-mediated allergy (22%) (i.e. classical allergy type manifestation).
- An associated autoimmune disease was detected in 14% of cases.
- 18% of patients had a relative with celiac disease, but no correlation was found between NCGS and HLA-DQ2/-DQ8 genes.
- IgG anti-gliadin antibodies were detected in 25% of the patients tested.
- Of patients who underwent duodenal biopsy; 69% showed normal intestinal mucosa, 31% showed mild increase in intraepithelial lymphocytes.
The two graphs below show the prevalence of symptoms identified in the Italian study. As you can see the range of symptoms associated with NCGS is vast, and like coeliac disease can involve many organ systems.
These gastrointestinal symptoms are similar to those seen in IBS, which has led some to say that NCGS is really just IBS. However, what distinguishes NCGS from IBS is the common occurrence of extra-intestinal (especially neurological) symptoms such as those below. (We will look at NCGS v IBS in more detail later)
If you or anyone you know have any of the above, it is certainly worth trying a strict gluten-free diet for a month to see if things improve. In fact, I would say go as far as to remove all grains, not just the standard gluten containing ones (wheat, rye and barley) as in my clinical experience, some people do not improve unless all grains are excluded.
Case 2 – NCGS symptoms are easy to miss before trying a GFD
The Prevalence of Non Coeliac Gluten Sensitivity
Based on the blinded-gluten challenge outlined above, the prevalence of NCGS appears to be about 6% in the general population, but the Salerno Experts group admit that they do not really know the true prevalence as they are still grappling with defining the condition.
On the other hand, in a recent UK survey the percentage of people self-reporting as gluten sensitive reached 13% [ref]. A similar study in Mexico found a similar number (12%) [ref] whilst a survey in Australia found 10% [ref].
The discrepancy between the diagnosed (confirmed) and self reported figures is often dismissed by the media as simply due to the “gluten-free fad”, as typified by this article in TIME magazine (click on image to read). I love that TIME has decided – ahead of the experts in this field – that anyone outside of the identified 6% NCGS who claims that on a GFD “they feel better, fitter, more energetic, that their withdrawn child has suddenly blossomed and that their man is healthier and happier” is probably just subject to the placebo effect!
“Is gluten sensitivity fad or fact?”
As discussed above, it is not clear yet whether it is the gluten, the carbohydrate content of grains, or other proteins in wheat and other grains which are causing symptoms. The likelihood is that different sub-groups are being triggered by different factors. Understanding exactly which aspect of wheat or grains causes which problems is still in its infancy, and researchers are saying:
Inaccurate attribution will be associated with suboptimal therapeutic advice and at least partly underlies the current gluten-free epidemic gripping the Western world
Although the prevalence rate for NCGS quoted by the experts is likely to change as research continues in the coming years, we should bear in mind that the diagnostic criteria of NCGS inevitably excludes certain people who would benefit from a gluten-free diet. Furthermore, there are different kinds of gluten-free diets: At one extreme all normal wheat based foods (breakfast cereals, cakes, bread etc) are replaced with their gluten-free simulacra (gluten free cereals, cakes, biscuits and bread), whilst at the other end additional meat, eggs, fruit and veg make up for the lost calories. It is the latter diet – a paleo or primal diet which excludes all grains, that I consider a true gluten-free diet. I would argue that the prevalence of NCGS would be much higher if this kind of gluten free diet were used in the diagnosis.
Biomarkers for NCGS
As already mentioned, there are currently no biomarkers for NCGS, however, that does not mean there are no serological abnormalities. In the following video Professor Umberto Volta, one of the Salerno Expert group and author of many published papers on matters around gluten, summarises the serum and tissue anomalies in NCGS.
- (3:10) IgG Antigliadin antibodies (AGA) are raised in 66% of NCGS patients. They fall within several months of a GFD, along with symptom resolution, whereas in coeliac disease they remain raised.
- (5:14) NCGS patients tend to have raised antibodoes to gliadin, glutenin, albumins and globulins, at similar levels to coeliac patients, but only coeliac patients have raised auto-antibodies (anti-deamidated gliadin and anti-tissue transglutaminase).
- (6:10) in NCGS and IBS patients on a GFD, peripheral blood mononucleated cells (PBMCs) responded to gluten exposure with increased levels of IL-10 and TNFα
- (6:40) in NCGS PBMCs responded to wheat extract with increased excretion of CXCL10 (interestingly, more so with modern wheat cultivars than ancient wheat varieties)
- (7:15) Compared to controls and IBS sufferers, NCGS patients had raised zonulin levels (an indicator of gut permeability) similar to coeliac disease patients.
- (8:00) NCGS patients had raised sub-mucosal mast cell density, again, similar to coeliacs, compared to healthy controls.
- (8:58) Intestinal deposits of anti-tissue transglutaminase found in some NCGS patients suggesting they may be at higher risk of developing coeliac disease.
- (10:03) No difference between NCGS and general population in HLA-DQ2/8. (i.e. no link to coeliac genetic markers)
- (10:48) Possibly NCGS show unique pattern of raised T-lymphocyte infiltration around villi
- (11:54) Summary: Although none of the above biomarkers is sufficently specific or sensitive for NCGS, they may contribute as part of a diagnosis.
Links between NCGS and irritable bowel syndrome (IBS)
As already mentioned NCGS and IBS share many symptoms. The key features of IBS are chronic, recurrent abdominal pain or discomfort, associated with disturbed bowel habit, in the absence of any structural abnormality to account for these symptoms. According to a review in BMJ Clinical Evidence (July 2015), the prevalence of irritable bowel syndrome (IBS) varies depending on the criteria used to diagnose it, but it ranges from about 5% to 20% of the US population. [ref]. Incredibly, as many as 42% of IBS sufferers claim to improve on a gluten free diet [ref].
Some authors go so far as to claim that all NCGS is really IBS, and that the problem is really with other wheat components such as FODMAPS (which are present in other foods not just grains) not gluten per se. [ref]. However, the Salerno Experts group, point out that “NCGS patients often report symptoms outside of the intestinal tract, e.g., headache and/or foggy mind, which cannot be accounted for by lactose, and/or FODMAPs intolerance.” [ref]
Four lines of evidence showing differences between IBS and NCGS
The following four lines of evidence not only show how NCGS is distinct from IBS in many ways, but it also gives insight into the range of damaging effects that gluten/wheat/grains can cause.
1. Bone loss and NCWS
Here is further evidence that IBS and NCWS* are distinct.
[*note: In two of the following studies authors use Non-Coeliac Wheat Sensitivity (NCWS or WS) as they employed a blinded wheat challenge to identify subjects]
This 2014 study looked at bone mineral density (BMD) in three groups of patients: 60 coeliacs, 75 NCWS and 65 IBS controls. They examined the BMD in the femoral neck and lumber spine of each patient.
As you can see from the graphs above the incidence of bone loss – either osteopenia or osteoporosis – is significantly higher in NGWS compared to IBS sufferers (as expected, coeliac patients had the highest incidence of the three groups). What the graphs don’t show, is that nearly half (46.6%) of the NCWS patients had low bone mineral density in at least one of the two sites checked, whereas half of coeliac patients had bone loss from both.
Compared to the IBS controls, the NCWS group also had higher frequencies of weight loss, anemia (90% had iron deficiency anemia), coexistent atopic diseases (which include asthma, eczema and hayfever) and a family history of coeliac disease than IBS controls.
2. Multiple food allergies in NCWS patients
As part of the initial screening for the above study, a double blind wheat challenge was used to seperate NCGS patients from IBS. However, the researchers went further, and conducted double blind challenges for a range of other foods.
30 of the 75 NCWS patients were found not only to react to wheat, but to cows milk proteins (casein). Moreover, eight of these thirty patients experienced IBS-like symptoms after open challenges with egg (four cases), tomato (three cases) or chocolate (two cases).
3. High levels of autoimmune disease found among NCWS
Hot off the press, a new study matched patients with diagnosed non coeliac wheat sensitivity(NCWS) with a similar number of IBS cases (who were not NCWS) and a similar number of coeliac cases. What they found is that among those with NCWS approximately 25% had autoimmune disease (mainly Hashimoto’s thyroiditis). This was similar to the number of autoimmune cases found in the coeliac group, whereas among the IBS group the rate was only 4% [ref], which is similar to the rate in the general population (i.e. 3%) [ref]
Another study found autoimmune diseases were present in 14% of NCGS patients, two thirds of whom had autoimmune thyroiditis. Psoriasis and Graves’ disease were frequently seen, whereas myasthenia gravis, atrophic autoimmune gastritis, scleroderma, type 1 diabetes mellitus, Crohn’s disease and IgA deficiency were infrequent. [ref]
4. Neurological symptoms in NCGS – e.g. depression
As already mentioned, one of the distinguishing features of NCGS is the common occurrence of neurological symptoms. These include headaches, anxiety, foggy mind, tiredness or simple lack of wellbeing.
Interestingly, a subset of patients excluded from a NCGS study category – for example if their IBS-like symptoms did not improve on a gluten-free diet – choose to stay on a GFD after the study, explaining that they simply “felt better”.
This prompted a study, published in 2014 [ref], looking at the effects of gluten on feelings of depression. They concluded “Short-term exposure to gluten specifically induced current feelings of depression with no effect on other indices or on emotional disposition. Such findings might explain why patients with non-coeliac gluten sensitivity feel better on a gluten-free diet despite the continuation of gastrointestinal symptoms.”
The current gastrointestinal focus of NCGS research may be missing more widespread effects of gluten and grains, making it likely that the quoted NCGS prevalence of 6% is likely to be an underestimate.
When Carroccio et al, compared 276 NCWS to IBS controls they found the following significant differences [ref]:
- a personal history of food allergy in the pediatric age
- coexistent atopic diseases
- positive serum anti-gliadin and anti-betalactoglobulin antibodies
- positive cytofluorimetric assay revealing in vitro basophil activation by food antigens
- a presence of eosinophils in the intestinal mucosa biopsies
These differences led them to suggest that NCWS patients “might be suffering from non-IgE-mediated food allergy”.
When considering research about conditions associated with NCGS, it is important to realise that the reverse is also true. In other words, people suffering with IBS, osteopoenia, food intolerance, depression or an autoimmune disease are at higher risk of having NCGS (and coeliac disease).
The Gluten-free diet
Anyone diagnosed with conditions linked to NCGS if not NCGS itself, should try a proper gluten-free diet to see if it will benefit them, but they should do the proper paleo grain-free version, not the half-baked gluten-free copy-cat version which continues to beguile the coeliac charities.
Support literature from Coeliac UK, for example, is jam packed with adverts for ‘gluten free’ products, products which have no place in a diet intended to actually nourish us. Indeed these so-called gluten-free products are nutritionally sub-standard and can contribute to deficiencies. A 2013 study into nutritional status of coeliac patients on a typical gluten-free diet found pre-diagnosis deficiencies in thiamin, folate, vitamin A, magnesium, calcium, iron and zinc were not resolved after 2 years on a standard gluten-free diet, in fact thiamin and vitamin A intakes declined. They concluded: “dietary inadequacies are common and may relate to habitual poor food choices in addition to inherent deficiencies in the gluten free diet” (my emphasis)
A gluten-free diet should be seen as an opportunity to move one’s diet away from highly processed foods, towards real, natural produce, as illustrated below. If the NHS and Coeliac UK embraced this ethos they would be doing a world of good, instead of attracting headlines like “NHS handed out £116m of gluten-free junk food on prescription in the past year“.
What this means for you – and why I say no one should eat grains
Using Professor Alessio Fasano’s figures (1% of the population is coeliac, 6% are NCGS) for each of us the chance of currently suffering from a gluten/wheat induced health problems of about 7%.
For an entirely avoidable health risk this is a huge percentage and represents a significant degree of misery, burden on the health service and risk of serious diseases in the future.
Although some people will consider a 7% personal risk to be small it starts to look different if you extend your consideration to the circle of people closest to you.
If you consider the ten most important people in your life – you and your nine nearest and dearest (spouse, parents, children and best friends), then the chance of at least one of you having a wheat/gluten induced health problem right now is greater than 50%. The chance of one of you having full blown autoimmune coeliac disease is 10%.
[For the maths geeks: 0.93^10=0.484, and 0.99^10=0.904]
Remember, you cannot tell in advance who will develop these conditions or when. There are no screening tests, and these conditions often go undiagnosed for many years, meaning that you or a loved one is likely to be suffering, and probably receiving inappropriate treatment, for a long time before the true cause is identified.
Whilst the whole extended family is complacently eating a normal wheat-based diet it will be almost impossible for anyone suffering with undiagnosed CD or NCGS to stumble upon the solution accidentally. Even if they start to suspect that they might be better off on a GFD social habits and pressure are likely to make it difficult for them to remain gluten free for long. Being the only one on a GFD is isolating and is one of the main reasons for poor adherence and relapses.
The only preventative measure you can take is for you and your loved ones to adopt a proper grain-free/gluten-free diet now. Done correctly it will actually improve your nutrition and health. Done together it will dispel the social isolation and stigma. What’s to lose?
The final point I will leave you with reflects researchers thoughts about why so many people are affected by such a common food as wheat and related grains:
Possibly, the introduction of gluten-containing grains, which occurred about 10,000 years ago with the advent of agriculture, represented a “mistake of evolution” that created the conditions for human diseases related to gluten exposure – Sapone, BMC medicine, 2013
Which is why a pre-agricultural, or paleolithic diet probably offers the best template for a healthy diet, which I would argue is a diet free of all grains – not just wheat, rye, barley and oats!
Finally, a case illustrating how quickly symptoms can improve with proper dietary advice:
Case 3 – A triathlete improves rapidly on a proper GFD
Table 1: The OSLO accepted definitions for gluten related disorders
|Non-classical CD||Presents without signs and symptoms of malabsorption.|
|Subclinical CD||Below the threshold of detection without signs or symptoms sufficient to trigger CD testing in routine practice.|
|Symptomatic CD||Clinically evident gastrointestinal and/or extraintestinal symptoms attributable to gluten intake.|
|Refractory CD||Persistent or recurrent malabsorptive symptoms and signs with villous atrophy (VA) despite a strict GFD for more than 12 months|
|Potential CD||Normal small intestinal mucosa who are at increased risk of developing CD as indicated by positive CD serology.|
|CD autoimmunity||Increased TTG or EMA on at least two occasions when status of the biopsy is not yet known.|
|Genetically at risk of CD||Family members of patients with CD that test positive for HLA DQ2 and/or DQ8|
|Gluten-related disorders||Used to describe all conditions related to gluten. May include gluten ataxia, DH, non-coeliac gluten sensitivity (NCGS) & CD|
|Non-coeliac gluten sensitivity||Is a condition in which gluten ingestion leads to morphological or symptomatic manifestations despite the absence of CD|
|Gluten ataxia||Idiopathic sporadic ataxia and positive serum antigliadin antibodies even in the absence of duodenal enteropathy|
|Dermatitis herpetiformis||Cutaneous manifestation of small intestinal immunemediated enteropathy precipitated by exposure to dietary gluten|