What evidence should be enough for the diagnosis of celiac disease
Female patient, 7 years old, was referred to paediatric gastroenterologist due to positive point-of-care (POC) test for coeliac disease, which was taken when she enrolled in school, as a part of public health project in Croatia.
She is an only child in the family, with positive family history for irritable bowel syndrome (IBS), but negative for other gastrointestinal diseases, including coeliac disease. She was born from a normal pregnancy, breastfed up to 3 years and progressed well. Due to recurrent urinary tract infections, she is under treatment by paediatric nephrologist, and under treatment by an allergologist due to multiple inhalation and nutritional allergies.
After positive POC test for coeliac disease, her primary paediatrician ordered blood tests, including complete blood count, inflammation markers, liver enzymes, total immunoglobulins, antibodies to tissue transglutaminase (anti-tTG IgA) and deamidated gliadin peptide antibodies (DGP IgG). The results returned positive with anti-tTG levels of 40.8 (ULN is 7), with normal levels of immunoglobulins and normal levels of DGP. Other blood results were normal, and patient was referred to paediatric gastroenterologist.
At first visit, our patient was a well appearing child, with no complaints. Her appetite was normal, and she had regular, sometimes softer stools. Sometimes, after a larger meal, she complained of an abdominal pain. Due to nutritional allergies, she does not eat food with cow milk proteins (including dairy products), eggs, nuts, citrus fruits, seafood, peanuts, strawberries and honey.
Given the positive findings of tTG, we performed a duodenal biopsy. Until then, no restrictions regarding gluten were suggested. At histology, the sample was correct in orientation, with maintained architecture and ratio of villi and crypts. In lamina propria there were dense mononuclear cells, especially plasma cells and in some samples dense eosinophils (up to 50 eo/VVP), and in some samples increased number IEL (> 30 IEL / 100 enterocytes). Histologic finding suggested lymphocytic enteritis (Marsh stage I), but the pathologist concluded that dense accumulations of plasma cells and eosinophils may indicated an aetiology other than celiac disease.
The patient was advised to continue gluten containing diet and further workup was proposed. At her next check-up after 4 months, there were no new complaints. The patient was in a good general condition, gained on body weight, with normal stools. Parasitic infection was excluded. Genetic markers related to coeliac disease came back positive (HLA DQ2 positive in cis position), EMA was positive (2+) and tTG antibodies were 29 U/ml (ULN is 10).
The patient was advised to continue diet with gluten and to repeat the findings of complete blood count, anti-tTG and EMA serology in 6 months and report the findings.
Should we diagnose coeliac disease based on Marsh I histology and weakly positive serology?
Alternatively, should we leave our patient on a diet without restrictions concerning gluten and repeat serology and duodenal biopsy? If the latter, after what time?
This is an interesting case showing that diagnosing celiac disease in a child can be very challenging.
Diagnostic procedure is in complete concordance with the 2012/2020 ESPGHAN guidelines, which in case of a child with suspected CD propose initial serological testing with TGA (IgA). In this girl the results were abnormal, however the levels did not reach 10xULN (upper limit of normal), which would allow for a “no biopsy” approach. Therefore, a biopsy and histopathological examination of samples was performed. Mucosal changes as described by pathologist were Marsh 1 and increased eosinophils in mucosa suggested a possible eosinophilic pathology. Patient was left on a gluten containing diet and after 4 months she was symptomless. However, both TGA and EMA were elevated. A girl was also found to be HLA DQ2 positive which is consistent with possible CD. She is reappointed for another exam in 6 months while still on gluten containing diet.
This girl can be considered to have a potential CD: serology +, histology -, symptoms -, genetics +. Increased EMA are indicative of a true immune deterioration consistent with CD. Careful observation with repeated serology tests at least twice a year with earlier re‑appointment in case of clinical deterioration is advised. I would also suggest a repeated biopsy and storing of frozen samples for immunohistopathology (it can also be performed on paraffine sections in some centres) in order to test for TGA deposits in intestinal mucosa. In case of mucosal deterioration or in case of persistency of EMA (or high TGA) a trial of GFD could be discussed with parents in order to assess clinical impact of dietary intervention. This approach should also be considered with clinical deterioration.