The concept of a causal relationship between the ingestion of gluten and the occurrence of symptoms in the absence of celiac disease (CD) and wheat allergy was first described in the late 1970s by Cooper and Ellis [1
]. This clinical entity has been termed non-celiac gluten or wheat sensitivity (NCGS or NCWS) [3
]. Over the past several years, NCWS has gained significant interest and the number of individuals embracing a gluten-free diet is rapidly growing [5
]. The discussion of whether or not gluten can cause symptoms in the absence of CD is confused by a popular phenomenon of people who avoid gluten-containing food in the light of a healthier lifestyle which is related to the fast growth of the gluten-free market [6
]. This theory that grains by means of their composition are unhealthy, should be distinguished from the question as to whether gluten can cause symptoms in the absence of CD and wheat allergy.
As defined by the 2015 Salerno Expert’s Criteria [4
], NCWS includes both intestinal and extra-intestinal symptoms which are related to the ingestion of gluten-containing food after exclusion of CD and wheat allergy. Most common symptoms include bloating, abdominal pain, diarrhea, tiredness and headache [4
]. These symptoms display a significant overlap with the irritable bowel syndrome (IBS), which is one of the most common disorders in daily practice [7
]. Although NCWS patients report that their symptoms are induced by gluten, it has been hypothesized that their symptoms may in fact be induced by other compounds in grains, among which a group of carbohydrates, referred to as fermentable oligo-, di-, monosaccharides and polyols (FODMAPs) have gained substantial interest [9
Despite the overwhelming current interest in NCWS, the actual prevalence is difficult to establish in the absence of a gold standard. The number of studies that have addressed this question are sparse and the outcome varied widely between 0.6% and 13% [3
Here, we studied the population prevalence of self-reported gluten sensitivity (srGS) in a large cohort of adults in the Dutch population.
A total of 785 adults completed the questionnaire, of whom 66% were recruited in dental practices, 28% at markets and 6% at a university. Two individuals had an established diagnosis of CD and were excluded from further analysis. Mean age at the time of survey was 47 years old (sd: 18 years, range 18–90) with a slight predominance of women (60%). The majority of questionnaires (57%) was completed in the urban region.
A total of 49 individuals (6.2%) indicated symptoms after the ingestion of gluten-containing foods. Such srGS individuals were younger (39 vs. 47 years old, p = 0.001), predominantly female (80% vs. 58%, p < 0.01) and mostly lived in the urban region (76% vs. 56%, p < 0.01) compared with the controls. Although not statistically significant, there was a trend for a higher education level in srGS individuals (49% with a high education level vs. 39%, not statistically significant).
3.1. Characteristics of Self-Reported Gluten Sensitivity
The most frequently reported intestinal symptoms in srGS were bloating, abdominal discomfort and flatulence. Tiredness and headache were the most frequent extra-intestinal symptoms reported, as shown in Figure 2
. Especially bread (n =
32, 65%), pizza (n =
15, 31%) and pasta (n =
18, 37%) induced these symptoms. Interestingly, 35% (n =
17) of the srGS respondents reported a reduction of clinical signs when consuming spelt bread.
The median duration of symptoms was four years (range 0 months–40 years) at the time of the survey. Most srGS individuals (n =
16, 33%) reported symptoms nearly every day after eating gluten, with the onset of symptoms 1 to 6 h after ingestion of a gluten-containing food (n =
23, 47%). In most srGS individuals, these symptoms resolved within hours (n =
29, 59%). More details are shown in Figure 3
The prevalence of individuals fulfilling the Rome III criteria for IBS in srGS was 37% vs. 9% in controls (p <
0.001). Medical history showed more often anxiety, anemia, chronic headache, IBS and gastro-intestinal reflux disease in srGS individuals. Table 1
shows basic demographic information and medical history. Family medical history was more often positive for CD, thyroid disease and IBS in srGS (Table 2
Some srGS individuals reported self-initiated dietary changes, of whom two (4%) followed a strict gluten-free diet and 21 (43%) a gluten-restricted diet.
Eight srGS individuals (16%) visited their general practitioner, five (10%) visited a medical specialist, three (6%) visited an alternative healthcare professional, and two (4%) a dietician. The median time before consulting a healthcare professional after onset of the symptoms was two years (range 0 months–32 years). Six individuals (12%) underwent upper endoscopy examination. None of them reported a diagnosis of CD or other diagnosis after upper endoscopy examination.
3.2. Fermentable Oligosaccharides, Disaccharides, Monosaccharides and Polyols (FODMAPs)
Of all srGS individuals, 74% reported abdominal discomfort related to at least one high FODMAP-containing product compared to 22% of the controls (OR 10.0 (95% confidence interval 5.2–19.3), p <
0.001) with a predominance of legume, cabbage, onions and leek (Table 3
This study confirms that a significant part of the general adult population reports sensitivity to gluten-containing foods. The percentage is substantially less (6.2%) than in a recent comparable UK study (13%) [12
]. Second, we showed that self-reported gluten sensitivity (srGS) individuals more frequently reported symptoms upon consumption of products high in FODMAPs. Third, quite surprisingly, we found that only a small number of the srGS individuals visited a doctor or ever consumed a self-initiated strict gluten-free diet (GFD).
Why srGS was found to be less prevalent in The Netherlands compared to the UK is unknown. It may be related to differences of media attention, but data to support this are lacking.
Despite the large growth of the gluten-free market [6
] and the popularity of gluten-free products, knowledge about gluten is still low in the general population as described in Australian and UK literature [17
]. Indeed, in our survey, a high percentage of srGS individuals (35%) reported no symptoms when consuming spelt bread.
In this study, type and onset of symptoms after consumption of gluten were comparable with other non-celiac wheat sensitivity (NCWS) studies [12
]. As shown in Table 1
, anxiety, chronic headache and gastro-intestinal reflux disease were more common in the srGS group compared with controls. These subjective health complaints are also common symptoms in patients with irritable bowel syndrome (IBS) and self-reported food intolerance in general [21
A significant number of NCWS patients fulfills the Rome III IBS criteria with a strong overlap between NCWS and IBS symptoms [12
]. Foods which are reported to be associated with IBS symptoms are commonly rich in gluten, wheat and carbohydrates [23
]. Therefore NCWS could be seen as part of IBS with a gluten-free diet as treatment strategy. Although the pathophysiology of IBS is still not well understood, food could affect a variety of physiologic parameters important in IBS such as visceral perception, motility, permeability, microbiota composition, brain-gut interactions, neuro-endocrine function and immune activation [24
It is well known that stress and anxiety may exacerbate or contribute to gastrointestinal symptoms [25
]. Indeed, individuals with srGS often reported an association between increase of abdominal symptoms and stress (84% vs. 48% in the control group, OR 5.7, p <
The mechanisms by which gluten causes symptoms in individuals without celiac disease (CD) is unknown. There are some indications that NCWS belongs to the group of the gluten-related disorders. The relatively large number of relatives with CD in srGS individuals could indicate a shared (genetic) predisposition, although current literature is not consistent about the relationship between HLA-DQ2 and NCWS [26
]. Another indication for (mild) immune activation in NCWS comes from a study which showed that serum levels of soluble CD14 and lipopolysaccharide-binding protein as well as antibody reactivity to microbial antigens are elevated in NCWS patients with resolution after a GFD [29
Whether or not such immune activation is triggered by gluten is not yet established. At this point, it cannot be excluded that other ingredients in grains, including wheat germ agglutinin [30
] and amylase-trypsin inhibitors [31
], are in fact responsible for these signs of immune activation.
An alternative explanation is that symptoms are not immune mediated, but caused by the result of other mechanisms. One such mechanism is luminal distention of the intestine via a combination of osmotic effects and gas production caused by bacterial fermentation of poorly absorbed short-chain carbohydrates referred to as FODMAPs [32
]. FODMAPs can be found in a variety of products, including grains. In support of the FODMAP theory, we found that individuals reporting gluten sensitivity more often reported symptoms after consuming products high in FODMAPs and less symptoms after eating spelt bread.
It is remarkable that only a fraction of srGS individuals started a GFD and that only 16% of them had visited their general practitioner due to srGS. This indicates that apparently burden of symptoms was not severe enough to change the diet or might be related to costs and availability of gluten-free products [34