2. Epidemiological Data

Following more active screening for CD in general populations and in at-risk groups, the frequency of diagnosis of PCD has increased significantly in recent years. It is often diagnosed in first-degree relatives of CD patients, and in people with autoimmune comorbidities, especially dermatitis herpetiformis [6]. PCD is believed to constitute as much as every fifth diagnosis of CD [6,7,8] with a greater frequency reported by some authors in recent years [8]. However, in view of the commonly-known limitations of CD tests, these findings may be significantly overestimated. Nevertheless, compared to overt CD, PCD tends to be diagnosed in slightly younger patients [8,9] and is observed more often in women [8]. This younger age of onset may support the hypothesis that it represents an earlier phase of overt CD.

3. Verification of a Diagnosis of PCD

In any case where PCD is suspected, the diagnosis must be verified. Most importantly, it is always necessary to determine whether sufficient amounts of gluten are present in the diet, and whether gluten is being limited or completely eliminated, either consciously or unconsciously. Consuming a low-gluten diet may result in the resolution of atrophic changes in the small intestine [10]. However, in such cases, it is proposed that the small intestine should be re-biopsied after gluten challenge. As gluten sensitivity varies so much between CD patients, it is difficult to set precise guidelines regarding the duration of the challenge, or the minimum amount of gluten necessary to elicit the development of advanced lesions; however, it seems that, for most adults, it should be sufficient to consume one to three slices of gluten-based bread over a period of two to six weeks [2].

In cases of suspected PCD, genetic tests should be performed on human leukocyte antigen (HLA) genes, which have been associated with the development of CD. Although nearly 30–40% of people in the general population have at least one of the HLA molecules (HLA-DQ2.5, DQ2.2, DQ8, DQ7.5), a negative genetic test result almost completely excludes the possibility of CD and PCD. However, patients with PCD may have a slightly different genetic profile to patients who demonstrate advanced atrophic changes in the mucosa of the small intestine. PCD patients are more likely to demonstrate the HLA-DQB1 * 0302 and HLA-DQB1 * 0603 alleles, and less likely to demonstrate DQB1 * 02 homozygosity [11]. In addition, they tend to demonstrate a different distribution of six gene polymorphisms (c-REL * G, one marker of KIAA1109/IL-2/IL-21, IL-21, IL-2, KIAA1109 and c-REL) [12].

Serological diagnostics of CD are usually performed with anti-tissue transglutaminase (anti-TG2), anti-endomysial (EmA) and anti-deamidated gliadin peptide antibodies (anti-DGP) [3]. Both positive serum anti-TG2 and EmA are needed for a diagnosis of PCD [1,2,13]. A positive anti-DGP test result would reinforce a diagnosis of PCD; however, due to the lower value of these antibodies in CD diagnostics, they are not included in the definition of PCD [3].

Currently, the most commonly used, most widely available and objective CD serological marker is anti-TG2. However, while it is characterized by the highest sensitivity of all currently available serological tests (about 95%), EmA is still considered the most specific in CD diagnostics (97–100%). The expertise of the laboratory and the selection of the test kit have a great effect on the accuracy of the CD antibody test results. The accuracy of the CD antibody test is strongly dependent on the expertise of the laboratory and the choice of test. Whenever the reliability of the test or circumstances of testing are questionable, for example, in cases where initial testing is performed with a rapid antibody-detection kit or by laypeople or untrained medical staff, any positive test result should be confirmed by a laboratory-based quantitative test. Scientific societies emphasize the need for constant quality control and systematic supervision of testing laboratories at the national and international level in order to increase the validity of serological tests [3].

Elevated, but not high, levels of anti-TG2 can occur in many conditions other than CD, such as autoimmune diseases, including especially inflammatory bowel diseases and primary biliary cirrhosis [14], as well as Goodpasture syndrome, granulomatosis with polyangiitis (formerly called Wegener’s granulomatosis), rheumatoid arthritis, systemic lupus erythematosus, systemic sclerosis, psoriasis [15] and type 1 diabetes mellitus [16]. Elevated levels of anti-TG2 are also found in non-autoimmune diseases such as connective tissue diseases [14], non-autoimmune cirrhosis [17] and linear IgA bullous dermatosis [15]. False positive results can be obtained for EmA in cases of Down’s syndrome [18], infantile cerebral palsy [19], infectious febrile illness [20] and end-stage heart failure [21], although such results are much less common than false-positive anti-TG2. Non-CD anti-TG2 and EmA seropositivity is often transient, and may occasionally be found in healthy subjects [22,23]. Spontaneous serological negativity is particularly common in patients with type 1 diabetes [24]. It is possible that patients with false-positive specific antibodies and a normal microscopic image of the mucosa of the small intestine may be misdiagnosed with PCD.

Patients with PCD typically demonstrate significantly lower levels of anti-TG2 than those with atrophic changes [7,25,26]. A small intestine biopsy is always required to establish the diagnosis of PCD. Pediatric patients with PCD do not meet the no-biopsy strategy criteria, as they rarely demonstrate high serum anti-TG2 concentration ≥ ten times the upper limit of the normal range [3].

Another key element in the correct diagnosis of PCD is reliable histopathological evaluation. As inflammatory changes are commonly found in foci in the small intestine (“patchiness”) and are restricted to the duodenal bulb (ultrashort coeliac disease), it has long been recommended to collect at least four [2] or five [3] biopsies of the small intestine mucosa, including at least one from the duodenal bulb. Taking fewer than four biopsies may result in a false negative result [27]. To avoid diagnostic errors, and to avoid overlooking atrophic changes in a patient with serological CD markers, it is important to follow the correct methodology when collecting small intestine biopsies: the tissue material should be spatially oriented on cellulose paper (orientation) and be fixed, stained and interpreted by an experienced histopathologist. It is recommended that immunohistochemical staining protocols intended for histopathological diagnosis should be based on anti-CD3 monoclonal antibodies [2,28].

In addition, taking into account the large high interobserver variability associated with such protocols [29], many authors suggest that in seropositive patients without typical atrophic changes, histopathological assessments should be revised by another expert, especially when the first assessment was performed in a non-specialist center [5,30]. Some rare variants of CD also exist, whose inflammatory lesions may be located beyond the reach of classic gastroduodenoscopy. Patients with such variants may benefit from a capsule endoscopy examination of the small intestine [31].

Diagnosis is also complicated by the low specificity of benign inflammatory lesions, i.e., lymphocytic duodenosis, which are classified as Marsh type 1. These lesions are also characterised by an increase in the number of intraepithelial lymphocytes to above 25 lymphocytes per 100 enterocytes in structurally-correct small intestine mucosa. In the general population, lymphocytic duodenitis may affect 5.4% of the population [32], with lymphocytic duodenosis being identified in a range of infectious diseases of the gastrointestinal tract, especially autoimmune atrophic gastritis [33], Helicobacter pylori-related gastritis, AIDS enteropathy, Whipple’s disease, small intestinal bacterial overgrowth and post-viral enteropathy [32], as well as hypersensitivity to milk, soy, fish, eggs or other nutrients [2].

The numbers of intraepithelial lymphocytes in the mucosa of the small intestine are also often elevated in intestinal disorders such as autoimmune enteropathy and Crohn’s disease, as well as in extraintestinal autoimmune disorders, such as autoimmune thyroiditis, type 1 diabetes, rheumatoid arthritis, multiple sclerosis, systemic lupus erythematosus and systemic sclerosis. Establishing a correct diagnosis is further complicated by the fact that patients with these diseases often have serological tests for CD, due to their common co-occurrence with CD, and these serological tests are more likely to return false positives [32]. The co-occurrence of mild enteropathy with elevated intraepithelial lymphocyte number may also be associated with pharmacotherapy with such agents as proton pump inhibitors, methotrexate, azathioprin, nonsteroidal anti-inflammatory drugs, olmesartan or ipilimumab [32].