What are the limits of serological tests and biopsy?

Diagnostic Tests Aren’t Perfect!

Serological tests

Anti-gliadin antibody (AGA): The search for serum (blood) antibodies produced and directed against gliadin is a common way to determine if gluten is present in our bloodstream and therefore if our health is threatened .

Our immune system can tell the difference between a gluten protein and another invading protein that is part of a bacterium or microbe, by developing specific antibodies against these proteins. Two classes of gluten-sensitive antibodies are found in blood tests: type A immunoglobulins (IgA) and type G immunoglobulins (IgG). We therefore distinguish between two anti-gliadin antibody tests: the anti-gliadin IgA antibody test and the anti-gliadin IgG antibody test. IgAs are more specific but less sensitive than IgGs.

The AgA IgG test identifies patients with potential celiac disease and IgA deficiency. Indeed, two to three percent of celiacs have an IgA deficiency and will therefore have negative results for the detection of anti-gliadin antibodies, anti-endomysium antibodies and anti-transglutaminase antibodies based on type A immunoglobulin. It is therefore important to ask the doctor to check if you have an IgA deficiency by carrying out a total IgA test so that possible negative results, based only on this type of immunoglobulin, are not interpreted as false negatives .

Limitations of the test:The positivity of anti-gliadin IgA and IgG antibodies correlates with the severity of atrophy and these prove less effective in screening gluten-intolerant patients with little or no atrophy. ( Antonio Tursi et al,. 2001 ) (Rostami K et al,.1999) (JA Abrams et al,.2004)

Anti-endomysial antibody (EMA)

The endomysium is a tissue that envelops the fiber of certain muscles. The development of the EMA test is based on the recognition, at the beginning of the 1980s, of the fact that antibodies attacking this tissue are present in the blood of 90% of celiacs consuming gluten whereas they disappear rapidly after elimination. gluten. This test was until the early 2000s the most widely used test for detecting celiac disease and is tending to be replaced by the detection of the anti-transglutaminase antibody.

Limitations of the test: The positivity of anti-endomysial antibodies also correlates with the severity of atrophy and these appear to be less effective in detecting gluten-intolerant patients with little or no atrophy. This test is also subject to interpretation by the examiner.
(Rostami K et al., 1999) ( Antonio Tursi et al., 2001 ) ( Abrams JA et al., 2004 )

Anti-transglutaminase (tTG) antibody

In the field of blood screening for celiac disease, the tTG test is the new entity. This is another step forward in solving the celiac disease puzzle. Transglutaminase is an enzyme that is an integral part of the endomysium and is involved in tissue repair. It is precisely the part of the endomysium that is attacked by anti-endomysium antibodies. This test is considered superior to the EMA and has replaced it because its cost is lower and also because its interpretation does not depend on the examiner as is the case for anti-endomysial antibodies.

Limitations of the test: Anti-transglutaminase antibody positivity also correlates with the severity of atrophy and the results may underestimate the true prevalence of celiac disease. ( Antonio Tursi et al,. 2003 ) ( Esteve et al,. 2006 ) ( Abrams JA et al., 2004 )

In summary: For screening for celiac disease, it is recommended to test for anti-gliadin IgG and IgA antibodies as well as anti-tTG IgA antibodies. The total IgA level should always be determined, because IgA deficiency is common in celiacs (3%). However, it should be kept in mind that serological tests as a whole are effective when the patient is in the final stage of celiac disease (Marsh IIIb-c), with subtotal or total villous atrophy, and prove to be less effective. in patients with mild enteropathy (Marsh I-IIIa). In short, all these tests can tell you is that you probably have gluten intolerance, but they can’t tell you that you don’t.

Table 1 (from Pietzak et al, 2001, data compiled from multiple studies)

AGA IgG57 – 10042 – 9820 – 9541 – 88
AGA IgA53 – 10065 – 10028 – 10065 – 100
EMA IgA *75 – 9896 – 10098 – 10080 – 95
Human tTg +98.598  

* Patients over 2 years old
+ IgG or IgA antibodies
PPV = positive predictive value
VPN= negative predictive value

The sensitivity of a test is its ability to give a positive result when disease is present. It contrasts with specificity , which is the ability of a test to give a negative result when disease is not present. Thus, the higher the specificity when the test is positive, the less chance there is of a false positive and the higher the sensitivity when the test is negative, the less chance there is that this is a false negative.

Intestinal biopsy

Everyone with a healthy gut has millions of small, finger-like projections called “villi” on the surface of the gut. They increase the absorption surface of the intestine and therefore its efficiency in absorbing nutrients. The atrophy of these villi in a diseased intestine will therefore reduce this absorption surface, resulting in a reduction in the absorption of nutrients present in food.

The common test used and considered in medicine to be the gold standard for diagnosing celiac disease is the intestinal biopsy performed by endoscopy. This is a procedure developed in the 1950s where the gastroenterologist inserts a tube through the patient’s mouth, this tube passes through the throat and stomach and ends in the small intestine. The gastroenterologist then takes small samples of skin from the surface of the intestine to then have them examined under a microscope by the pathologist. The biopsies must be multiple (six on average in the second and third duodenum) because the severity of the lesions can vary from one site to another, and well perpendicular to the mucosa. In common practice, if the biopsy reveals atrophy of the villi, a diagnosis of celiac disease is pronounced.

A broader histological view of gluten-sensitive enteropathy has been proposed by Dr. Marsh.
Depending on the histological lesions, the progression of gluten-sensitive enteropathy can therefore be subdivided into three stages according to the modified Marsh classification:

Marsh I
Marsh II
Marsh IIIA
Marsh IIIB
Marsh IIIC
enteritis Lymphocytic enteritis with hyperplasia of crypts
Atrophy of intestinal villi: partial
Atrophy of intestinal villi:
subtotal Atrophy of intestinal villi: total

The first stage, Marsh I, is often referred to as the latent stage of celiac disease.
The second stage, Marsh II, is often referred to as the stage of silent celiac disease.
The third stage, Marsh III, is considered to be the typical stage of celiac disease.

Limitations of the test:Although the Marsh system provides pathologists with measurement templates for evaluating biopsies and provides them with clear criteria and guidelines for considering gluten-induced enteropathy, the majority of specialists rely on the criteria of ESPGAN (European Society for Pediatric Gastroenterology and Nutrition), which defines Marsh type I and II lesions in gluten-sensitive enteropathy as non-celiac lesions, thus not requiring a gluten-free diet despite the fact that many patients experience symptoms of gluten intolerance and show a marked improvement in these symptoms by removing gluten from their diet. Gluten sensitivity is indeed associated with minimal changes in the mucosa, not detectable by conventional light microscopy. Such lesions which mainly involve the microscopic villous structure, may involve a reduction in the absorptive surface of the intestine from the latent stage of the disease. (Sbarbati A et al,. 2003) (GR Corazza et al,. 2005) (Kaukinen et al, .2001 ) ( Esteve et al, .2006 ) (Tursi A, Brandimarte G ,. 2003)

Here is what world-renowned neurologist and researcher Dr. Hadjivassiliou says about gluten-associated neurological diseases in a study published in 2002:

“Only one-third of patients with neurological disorders associated with gluten sensitivity have villous atrophy on duodenal biopsy. Even some patients with biochemical markers of malabsorption such as low serological levels of vitamin B12, folate or vitamin D had normal duodenal histology.
These cases may illustrate the patchy nature of bowel involvement in celiac disease and the imprecise interpretation of duodenal biopsies by inexperienced histo-pathologists. Preliminary data based on counting the T cell subpopulation in the small intestinal epithelium suggest that these patients have potential celiac disease. There are, however, patients where the immunologic disorder is primarily directed to the nervous system with little or no gut damage.”


  • Tursi A, Brandimarte G, Giorgetti G, Gigliobianco A, Lombardi D, Gasbarrini G.
     AM J Gastroenterol 2001 May;96(5):1507-10
     Low prevalence of anti-gliadin and anti-endomysium antibodies
     to detect celiac disease sub- clinical or silent.
     Low prevalence of antigliadin and anti-endomysium antibodies in
     subclinical/silent celiac disease.
  • Abrams JA, Diamond B, Rotterdam H, Green PH.
     Department of Medicine, Columbia University College of Physicians and
     Surgeons, New York, New York, USA.
     Seronegative celiac disease: increased prevalence with lesser degrees
     of villous atrophy.
  • Antonio Tursi et al
     Journal of clinical Gastroenterology 2003; 36(3):219-221
     Serological level of anti-tTG correlated with the degree of
     intestinal damage.
     Anti-tTG Serum Level Correlates to Degree of Intestinal Damage.
  • Rostami K, Kerckhaert J, Tiemessen R, von Blomberg BM, Meijer JW, Mulder CJ.
     Am J Gastroenterol, 1999 Apr;94(4):888-94.
     Sensitivity of antiendomysium and antigliadin antibodies in untreated
     celiac disease: disappointing in clinical practice.
  • Esteve M, Rosinach M, Fernandez-Banares F, Farre C, Salas A, Alsina M, Vilar P, Abad-Lacruz A, Forne M, Marine M, Santaolalla R, Espinos JC, Viver JM.
    Gut. 2006 May 18;
    Spectrum of gluten sensitive enteropathy in first degree relatives of
    coeliac patients: clinical relevance of lymphocytic enteritis.
  • Kaukinen K, Maki M, Partanen J, Sievanen H, Collin P.
    Dig Dis Sci. 2001 Apr;46(4):879-87.
    Celiac disease without villous atrophy: call for revision
    of criteria.
     Celiac disease without villous atrophy: revision of criteria called for.
  • G R Corazza and V Villanacci
    Journal of Clinical Pathology 2005;58:573-574
    Coeliac disease : Some considerations on the histological diagnosis
  • Tursi A, Brandimarte G.
    J Clin Gastroenterol, Jan;36(1):13-7.
    The symptomatic and histologic response to a gluten-free diet in
    patients with borderline enteropathy.
  • Sbarbati A, Valleta E, Bertini M, Cipolli M, Morroni M, Pinelli L, Tato L.
    Dig Liver Dis. 2003 Nov;35(11):768-73.
    Gluten sensitivity and “normal” histology: is the
    intestinal mucosa really normal?
    Gluten sensitivity and ‘normal’ histology: is the intestinal mucosa really
  • M Hadjivassiliou, R A Grünewald and G A B Davies-Jones
    Journal of Neurology Neurosurgery and Psychiatry 2002;72:560-563
    Gluten sensitivity as a neurological illness.
  • JA Abrams, B. Diamond, H. Rotterdam, and PH Green
    Seronegative Celiac Disease: Increased Prevalence with
    Lesser Degrees of Villus Atrophy.
    Seronegative Celiac Disease: Increased Prevalence with Lesser
    Degrees of Villous Atrophy.