26 October 2015
Non coeliac gluten sensitivity: A grain of truth?
Why are millions of Australians adopting a gluten-free diet when only a fraction have coeliac disease?
Patients reporting mild diarrhoea, a bloated or sensitive abdomen, fatigue and headaches are par for the course in general practice. But over the past decade the wellness industry has been waging a large and successful campaign against a relatively new potential culprit: gluten. And doctors are being forced to take notice as more and more of their patients ask if a gluten-free diet is right for them.
Gluten is the tough protein that gives bread its elasticity and it’s present in wheat, barley, rye and oats. There is no doubt that it triggers severe autoimmune reactions in patients with coeliac disease and left untreated this leads to malabsorption, anaemia, and osteoporosis.
The prevalence of coeliac disease is around one percent and a gluten-free diet is the only way to avoid symptoms. But eighteen percent of Australians are trying to reduce their gluten intake and half of those avoid it entirely. The worldwide gluten-free industry is expected to exceed $20 billion this year.
So what can explain the popularity of the gluten-free diet among non-coeliacs? While some are simply convinced that it is healthier, six percent of Australians report gastrointestinal effects from wheat consumption in the absence of any formally diagnosed allergy or intolerance.
Is gluten sensitivity much more widespread than traditionally understood or are people somaticizing these symptoms in response to media hype? Since the 1980s the term non-coeliac gluten sensitivity (NCGS) has gained traction.
NCGS is a diagnosis of exclusion where the villous atrophy that identifies coeliac disease has not been detected in duodenal biopsies (see box-out ‘Coeliac disease in brief). There is currently no accepted biomedical explanation of how gluten is causing this reaction in patients, and in some cases there is overlap with a diagnosis of irritable bowel syndrome.
The hard data
Professor Peter Gibson and Dr Jessica Biesiekierski had been quietly working away leading the Department of Gastroenterology at Monash University when a 2011 publication shot them into this controversial area.
The double-blind, randomised, placebo-controlled trial showed that NCGS patients re-challenged with bread and muffins for two weeks experienced worse bowel symptoms and tiredness compared to those receiving a placebo. Similar outcomes were reported the following year from an Italian study.
But Professor Gibson wanted to make sure. He was worried baseline differences between the groups might have compromised that statistical analysis.“That’s been the big problem with the attribution of the effects of a gluten free diet…. If you want to show real statistical effect the gold standard is the crossover design,” he says.
In 2013 the team published a more thorough study that refuted their earlier findings and found no specific effect of gluten over placebo on intestinal inflammation, immune activation or fatigue1.
In that study, 37 participants were cycled between high-gluten and zero-gluten diets in a random crossover design. When subjects act as their own controls like this, the individual variation between them is ironed out. And the subjects were given 16g/day of purified gluten in order to exclude other proteins found in bread, with a washout period of two weeks between each diet phase to avoid lingering effects.
This compelling negative result was used by some media to slam the gluten-free foodies. But the September 2015 issue of Clinical Gastroenterology and Hepatology has rocked the boat once more. A study from the Italian universities of Pavia and Bologna used an equally fastidious design in 59 NCGS patients and reported that gluten capsules specifically caused increases in abdominal bloating and pain as compared to placebo2.
A mixed group of patients?
This conflict in findings might be explained by inconsistent classification of patients. HLA-DQ2/DQ8 is a genetic marker found in most coeliacs. But the marker is found in 50 percent of those identified as having NCGS which is almost twice the population average.3
Similarly, the serological markers that typify coeliac disease (see box ‘coeliac disease in brief’) are also enriched in the NCGS population and up to a quarter of patients diagnosed with NCGS have a family history of coeliac disease.4
But according to the Oslo consensus of 2012, those who are positive for genetic and antibody markers but don’t have gut pathology might better be classified as ‘potential’ coeliacs. Could underdiagnosis of potential coeliac disease explain why many people report the benefits of a gluten-free diet?
Dr Jason Tye-Din heads an immunology lab at the Walter and Eliza Hall research institute in Melbourne. He and his colleagues randomly selected 2500 members of the public from the Geelong electoral roll and tested them for circulating antibodies to tissue transglutaminase 2, which is a target of the autoimmune reaction in coeliac disease.
Around four percent of the sample was positive for the both the immune marker and the HLA-DQ2/DQ8 haplotype though Dr Tye-Din is cautious about interpreting these results given the high rate of false positives in the antibody assays. Nevertheless this cohort is worthy of concerted follow up as “something like 1/3 over several years may eventually convert into full blown coeliac disease,” says Dr Tye-Din.
If potential coeliacs make up a fraction of those currently presenting with non-coeliac gluten sensitivity then research into the prevalence and aetiology of NCGS will be confounded. “How studies have excluded patients has not been very stringent,” says Professor Peter Gibson. “All you have to do is put a few coeliacs in your study and you would get positive results.”
In his 2013 study Professor Gibson’s team also tested patient serum for T-cell cytokine release specifically in response to gluten. This is the most sensitive and specific marker for immune reactivity and can produce a signal within days of gluten challenge rather than months. The conflicting Italian study was not as restrictive and the effect of gluten was driven by a large reaction in just a few patients, two of whom were positive for anti-gliadin antibodies and the HLA markers of coeliac disease.
The same research team published another study showing that anti-gliadin antibodies disappeared from blood serum in non-coeliac patients after a six-month gluten-free diet. In half of the coeliac patients tested, however, anti-gliadin levels remained high. This supports the concept that there are degrees of immunoreactivity to gluten on the coeliac spectrum.
Coeliac disease is also associated with increased permeability of the intestinal epithelia, and this has also been observed in patients diagnosed with irritable bowel syndrome positive for the HLA-DQ2/DQ8 genotype. By contrast in biopsies from NCGS patients there was no change in permeability compared to normal tissue.3 It remains to be demonstrated how whether gluten can irritate the bowel by pathways in true non-coeliacs by pathways other than the adaptive immune response.
Gluten on the brain
Despite the patchy evidence in regards to bowel disturbance in NCGS, extra-intestinal symptoms such as fatigue, headache, numbness, mental confusion (‘brain fog’), anxiety and sleep abnormalities are consistently reported, and may be the criteria for a differential diagnoses from irritable bowel syndrome.
Professor Gibson says that at the conclusion of his 2013 study he and his colleagues were perplexed that some subjects chose to stay on the gluten-free diet because they ‘felt better’ in a general way even if there was there was no measureable change in bowel symptoms.
So the team performed another pilot study in 22 NCGS patients to determine whether this change in mood was causal and quantifiable5. In this crossover study subjects were challenged with the high gluten and placebo diets for only three days and surveyed for expression of depressive traits (the Spielberger State Trait Personality Inventory). The high gluten diet was associated with higher depression scores as compared to the zero gluten diet.
While intrigued, Professor Gibson hopes that this tentative finding will not lead to more runaway hype: “We used one index only. [Jessica Biesiekierski] has just finished recruiting for another study in much more detail. If it does confirm there is an effect of gluten then we need to look at healthy people to prove that this is a specific effect,” he explained.
Speculatively he suggests that peptides could affect neurotransmitter production or that the effect might be an indirect response to by-products of floral metabolism of gluten. But there is much to be done to achieve a more definitive understanding of non-coeliac gluten sensitivity. More accurate diagnosis of this condition on the basis of informed used of biomarkers will contribute to this understanding in future.
Coeliac disease in-brief
Coeliac disease is the most common autoimmune disorder but four in five cases remain undiagnosed. Coeliac disease pathology results from an over-reactive immunogenic response against the gut endomysium. Ingested gluten is broken down by tissue transglutaminase (tTG) into gliadin and smaller protein fragments. Gliadin is taken up by helper T-cells in the gut vasculature and presented via the HLA complex to other immune cells.
But if patients have the DQ2/DQ8 variants of HLA there is fierce over-activation of the immune response against gliadin and tTG. The molecules released in this response damage the villi of the small intestine which has dire consequences for nutrient absorption. Iron-deficiency anaemia, fatigue and joint pain are common symptoms in people with untreated coeliac disease.
Diagnosis for coeliac disease usually begins with serological tests for antibodies produced in the reaction against gliadin. The most specific assays looks for endomysial antibodies although the more sensitive are against tissue transglutaminase, of which several variants have now been identified. In infants and others with low IgA levels the old anti-gliadin test might be used although this is the least reliable.
Dr Jason Tye-Din warns that all assays have their weaknesses and should not be relied upon as the sole diagnostic marker; “We’ve looked at 73 different labs and there’s about 11 different kits doing the transglutaminase antibody test and you get wildly different results.” He says the 10 minute point-of-care tests now available in clinics are pharmacies are even more prone to spurious results.
Diagnosis of coeliac disease is confirmed by classifying duodenal biopsies on the Marsh pathology scale. In the milder stages of various enteropathies there will be infiltration of the epithelium by lymphocytes and enlargement of the crypts surrounding the villi. Most chronic coeliac disease patients will show more severe atrophy of the villi that when complete results in shrunken crypts, a defined as Marsh stage III.
Both these serological and histopathological tests are only of use in people who are still consuming gluten products. If patients have already excluded gluten from their diet then it is necessary to challenge them for six weeks with at least four slices of bread per day to evoke a diagnostic response.
Coeliac disease has strong familial inheritance and diagnosis can be strengthened by HLA typing as 90 percent of coeliac disease patients are positive for the DQ2/DQ8 alleles. A negative HLA-DQ2/DQ8 haplotype can exclude coeliac disease diagnosis with great confidence but not all positives will have advanced enteric pathology.
There is also growing evidence for immune-mediated attack of gliadin-like epitopes in the nervous system (gluten ataxia), the skin (dermatitis herpatiformis) and the bone (osteoporosis)4. Some of these symptoms can present independently of the classical enteric pathology. Phase II trials are currently underway for a vaccine that would desensitise this adaptive immune reaction and permit coeliacs to consume gluten once more.
The FODMAP effect
20 percent of wheat protein is not gluten at all and some of these others peptides are potential allergens in their own right. Then there are small chain carbohydrates such as fructose, lactose and sugar alcohols known collectively as FODMAPs.
“There’s no doubt if you have a lot of bread you do feel bloated. We believe that is via the FODMAP effect,” says Professor Gibson. “If you go gluten-free you might go down 50 percent in FODMAP intake which might be enough to improve symptoms.”
FODMAPs are fermentable, poorly absorbed short-chain carbohydrates or FODMAPs not only contained in grains but also other fruits, honey, dairy and the well-known gas giants, onions. FODMAPs are not complicated to digest but they are osmotic and so draw water into the intestinal tract leading to stool motility.
The activation of mechanoreceptors in the gut would explain sensations of bloating and pain and if FODMAPs pass into the colon undigested, bacteria-driven fermentation will produce further odorous gases.
About 75 percent of people with irritable bowel syndrome respond well to a diet low in FODMAPs and Professor Gibson suspects that many people self-diagnosing with NCGS may instead fall into this group. Indeed in the 2013 gluten study all patients were put on a lead-in diet that was low in FODMAPs and this led to an elimination in bowel symptoms before the gluten re-challenge.
Irritable bowel syndrome affects 14 percent of Australians so the FODMAP effect might explain a large proportion of the people reporting benefits from giving up gluten-containing grains. But avoidance of wheat products altogether can be frustrating for patients and their dining companions as gluten-free products often come with a hefty price tag. By contrast, reducing FODMAP intake is more easy to manage in a balanced diet as described on the public information site of Monash University Department of Gastroenterology.
Professor Gibson explains that if you’re truly gluten intolerant, “you’ve got to look at every food label to see if there are hidden sources. Whereas in a low FODMAP diet, little bits don’t matter….We’re trying to put together something that’s possible to maintain long term.”
People who do abandon gluten entirely should make sure to supplement their nutrient intake from other sources. Coeliac patients who have excluded wheat products long-term are at risk of deficient intake of fibre and folate only partly due to malabsorption. Thiamin and vitamin A may also need to be supplemented by non-grain sources such as legumes, fruits, vegetables, meat, fish, chicken and eggs.
- Biesiekierski JR, et al. No effects of gluten in patients with self-reported non-celiac gluten sensitivity after dietary reduction of fermentable, poorly absorbed, short-chain carbohydrates. Gastroenterology. 2013 Aug;145(2):320-8 e1-3.
- Di Sabatino A, et al. Small Amounts of Gluten in Subjects With Suspected Nonceliac Gluten Sensitivity: A Randomized, Double-Blind, Placebo-Controlled, Cross-Over Trial. Clin Gastroenterol Hepatol. 2015 Sep;13(9):1604-12 e3.
- Tovoli F, et al. Clinical and diagnostic aspects of gluten related disorders. World J Clin Cases. 2015 Mar 16;3(3):275-84.
- Aziz I, et al. The spectrum of noncoeliac gluten sensitivity. Nat Rev Gastroenterol Hepatol. 2015 Sep;12(9):516-26.
- Peters SL, et al. Randomised clinical trial: gluten may cause depression in subjects with non-coeliac gluten sensitivity – an exploratory clinical study. Aliment Pharmacol Ther. 2014 May;39(10):1104-12.