Diagnostic Dilemmas in Celiac Disease

Michael X Ma; Mina John; Geoffrey M Forbes


Expert Rev Gastroenterol Hepatol. 2013;7(7):643-655. 

In This Article

Five-year View

Our discussion focuses upon predictions and wishes regarding diagnostic aspects of CD, mindful that we also expect significant advances to occur with respect to treatment in the foreseeable future.

Novel Diagnostics

The applicability of current CD diagnostic algorithms involving serology and duodenal biopsy is limited by reliance on continued dietary intake of gluten. For patients who adopt a GFD prior to testing, there may be significant diagnostic uncertainty as discussed earlier in this review. For these patients, novel tests that rely on shorter duration gluten challenge may circumvent the problems currently faced.

Unlike serum antibodies, memory CD4 T cells can be stimulated to proliferate after low levels of antigen exposure in vivo. These cells are also the key players in downstream activation of CD8 intraepithelial cytotoxic T cells, which mediate enterocyte destruction and villous atrophy as well as providing T cell help to B cells, producing antibodies against tTG and DGP. The identification of mucosal gliadin-specific memory CD4 T cells (GSTCs)[85,86] might permit the development of cellular detection assays, particularly for use in patients who are already 'gluten free'. The fact that both the common immunogenic gliadin epitopes and the molecules (HLA-DQ2 and -DQ8) that present them to CD4 T cells are so well defined has enabled various methods to detect GSTCs to be investigated in the diagnostic context. Most have employed a short-term gluten challenge (3 g gluten per day for 3 days) in order to increase the number of antigen-specific T cells in vivo, while minimizing relapse of symptoms. For example, in a study of 13 CD patients on a GFD, HLA-DQ2-gliadin tetramers were able to detect GSTCs in the peripheral blood of 11 of these patients, after a 3-day oral gluten challenge.[87] Major histocompatibility complex tetramers refer to fluorescein-labeled tetrameric complexes of synthetic HLA-peptide units, which bind antigen-specific circulating peripheral blood memory CD4 T cells and are detectable by flow cytometry.[87,88] Alternative methods based on cytokine release assays or lymphoproliferative responses also show promise as potential diagnostic assays.[89,90] These techniques are yet to translate into routine clinical practice, and to do so, will require demonstrating a greater level of sensitivity and specificity for CD diagnosis compared with the current algorithm. The cost and scalability of these relatively labor intensive assays will also need to be favorable before adoption into routine diagnostics.

Detection of celiac antibodies in the supernatant of duodenal biopsy specimens incubated with gliadin peptides has also been suggested as a diagnostic test for patients with suspected seronegative CD.[91,92] The physiologic basis for this approach stems from evidence that celiac antibodies are produced in the small intestinal mucosa, and that secretion of these antibodies is dependent on small intestinal mucosal transglutaminase 2-specific IgA deposits.[93] In a study of 418 patients with CD and 705 non-CD controls, anti-EMA antibodies were present in the culture medium of 98% of CD patients and absent in 99% of non-CD controls.[92] In contrast, the sensitivity and specificity of celiac serology (anti-EMA and/or anti-tTG) in detecting CD in this study was 80% and 95%, respectively. A separate study measuring anti-tTG antibody in the culture medium of duodenal biopsy specimens from 273 patients with suspected CD, found a 98% sensitivity and 100% specificity for CD diagnosis.[91] Given the established utility of conventional celiac serology and duodenal biopsy, any future role of duodenal mucosal culture will likely be as an adjunct to histology, undertaken in specialized laboratories, when the diagnosis of CD remains unclear after serology, histology and HLA haplotype testing.

No Gastroscopy?

We previously discussed recent propositions to permit the diagnosis of CD to be made on serological testing alone. This will inevitably be the subject of further evaluations in the coming years, in attempts to enhance the ease of diagnosis without endoscopy, especially in children, and where difficulties in accessing endoscopic services exist. However, a GFD is a difficult undertaking; it has significant cost, psychosocial and quality-of-life implications, and our current view is that the highest level of diagnostic certainty is appropriate to aim for. If other studies replicate the highest level of positive and negative predictive values described with recent combination four-assay testing, then it is conceivable that duodenal histology may become unnecessary.

Diagnostic Evaluation: Less Inadvertent Gluten Ingestion

Moving from predictions to wishes, we anticipate ongoing improvements to the labeling and testing of gluten-free foods, to assist patients with CD maintain a GFD with less opportunity for inadvertent gluten ingestion. The introduction of a 20 ppm gluten threshold by the Codex Alimentarius Committee sets a reference standard to which food regulatory authorities and industry may base gluten-free products. A recent European multinational study investigating the concentration of gluten in gluten-free foods showed that 99.5% of the analyzed items had a gluten concentration of <20 mg/kg.[65] Similar standards should ideally be targeted in other countries. Additionally, unintentional gluten exposure from cross-contamination during food preparation remains a common cause for ongoing symptoms. Improved understanding of the GFD and practicalities on how to maintain it, by health practitioners, patients and their families may alleviate this problem. A collaborative multidisciplinary approach in GFD monitoring involving doctors, dieticians, local celiac societies, patients and their families should be the focus of future health initiatives. Through confidence that foods labeled as 'gluten free' do indeed meet such criteria, and that a GFD has no gluten contamination, the diagnostic work-up of nonresponsive CD patients will be much more efficient and rewarding.

Earlier Diagnosis

Finally, we particularly wish to see earlier diagnosis of unrecognized CD, especially in those with symptoms, with subsequent timely and appropriate institution of treatment. Given that CD is a relatively common condition with global prevalence and increasing incidence, the importance of this issue should not be underestimated. Greater detection of CD is likely achieved by further education of health practitioners, earlier use of serologic testing in symptomatic subjects and enhanced community awareness of CD by public health messages. The efficacy of improving detection of unrecognized CD by education campaigns was demonstrated in a prevalence study from Finland, where the estimated detection rate of CD was 70%.[10] A number of practical methods were utilized, including the distribution of nationwide CD diagnostic and treatment guidelines to all physicians and educational workshops at each central hospital organized by the national celiac society in conjunction with CD experts. The above-mentioned approaches encourage case finding of CD, rather than mass population screening, which has been shown to be an ineffective method of CD detection.[33] Case finding is particularly suited to CD diagnosis, given the condition's myriad of potential clinical manifestations, and when coupled with greater physician understanding of the disease and improved community awareness, has the potential to be a major factor in the paradigm of exposing more of the celiac iceberg in coming years.