More and more often, the clinician dealing with chil-dren or adults suspected of celiac disease is copingwith the diagnostic and therapeutic dilemma of the“potentialceliac patient,”that is, the patient who presents with elevatedspecific celiac serum antibodies but has a normal (or almostnormal) duodenal biopsy. How to handle these people? Can wepredict those who will evolve into full-blown celiac disease ifleft on a gluten-containing diet? Are we risking under-diagnosing or over-diagnosing celiac disease in these in-dividuals? In their article,“Progression of Celiac Disease inChildren with Antibodies Against Tissue Transglutaminaseand Normal Duodenal Architecture”published in this journal,Auricchio et al 1 follow a large cohort of children with potentialceliac disease to try to answer these questions.
There is a problem with both the under-diagnosis as wellas the over-diagnosis of celiac disease. Under-diagnosis is awell-known, worldwide problem, which is particularly rele-vant in the United States where the rate of diagnosis has laggedbehind other countries. 2,3
A considerable body of research hasaddressed why a substantial portion of patients with celiacdisease are not diagnosed. Factors identified include minimalor absent symptoms, failure to test high-risk groups,4 inappropriate interpretation of serologic testing, aninadequate number of duodenal biopsies,5 and theinadequate pathological interpretation of duodenal biopsies.6
Over-diagnosis is when people are inappropriately toldthey have celiac disease when they do not. This latter groupwouldbecommittedtoalife-long,unnecessarygluten-freediet.This area, however, has not been investigated adequately.7,8
After the diagnosis of celiac disease, a gluten-free diet isadvised. Because celiac disease is currently incurable, thiscommitment is life-long. Recent research has demonstratedthat adherence to a gluten-free diet can become the majorfactor affecting quality of life.9 Thisfinding is especiallyevident considering social, functions such as eating out ofthe home and travel. In fact, the most vigilant patients, thosewith the greatest knowledge of the diet, have a lower qualityof life 10 and may exhibit maladaptive eating behaviors thatare forbearers of eating disorders. 11
Therefore, the diagnosisof celiac disease clearly requires great precision.The diagnosis of celiac disease requires several steps: thepatient, while eating gluten, contacts their health care provider;the clinician has the diagnostic consideration of celiac disease;serologic testing is ordered, most appropriately a tissue trans-glutaminase IgA antibody (tTG IgA), often combined with orfollowed by an anti-endomysium IgA antibody, particularly inchildren; and, if serologic testing is positive, an endoscopicduodenal biopsy is performed. In this setting, thefindings ofvillous atrophy together with intraepithelial lymphocytosis in-dicates active celiac disease and dictates the prescription of agluten-free diet.
The circumstances may arise when, often sur-prisingly, a positive serology is not associated with the presenceof villous atrophy. Instead, the duodenal mucosa may be normalor show only an intraepithelial lymphocytosis and a diagnosis ofactive celiac disease cannot be made.
This intriguing subset ofpatients has been labelled as having“potential celiac disease.” 12
The cohort described by Auricchio et al 1 in this issue of Gastroenterologyincluded 280 children. They had bothpositive tTG IgA and endomysial antibodies on >2 occasionsand normal duodenal biopsies. Over a median follow-uptime of 60 months (from a minimum of 18 months to amaximum of 12 years), 42 children (15%) developed villousatrophy, 89 (32%) lost their positive antibodies, and theremainder of these children had persistent positive serologictests.
The cumulative incidence of progression to villous at-rophy was 43% at 12 years. In multivariate analysis, thebaseline factors most strongly associated with developmentof villous atrophy were the number ofgdintraepitheliallymphocytes, age (older age), and homozygosity for HLA DQ2.Adding to the analysis the persistence of positive tTG IgA at 2 years after diagnosis allowed the correct identification ofpatients who will develop villous atrophy 85% of the time.
As mentioned, this phenomenon of potential celiac dis-ease can also be seen in adults. 13,14 Adults, too, if left on agluten-containing diet, can progress to villous atrophy, losetheir positive celiac antibodies, or persist with positiveserologic tests without progression to villous atrophy (thelatter perhaps meeting the definition of false-positive tTG?). What should be the management of patients with posi-tive serologic tests for celiac disease and normal biopsies?Ultimately, the management must be the result of a thor-ough, well-informed agreement reached between the expert doctor and the patient (or their caregivers), but general advice can be given as follows.
For adults, a biopsy is usually recommended for thediagnosis of celiac disease.18 However, for children the European Society for Paediatric GastroenterologyHepatology and Nutrition has published evidence-basedguidelines allowing diagnosis without a biopsy 19 in asymptomatic child with tTG IgA antibodies>10 times theupper level of normal, a positive anti-endomysium IgAantibody in a sample drawn on a different occasion, andconsistent celiac HLA, because this approach has a positivepredictive value nearing 100%.20 It must be noted that noneof the children in the study by Auricchio et al1who werelabelled as having potential celiac disease were initiallysymptomatic or had tTG IgA levels that were>10 timesthe upper level of normal, thus reinforcing the concept thatnone of them would have been improperly labelled ashaving celiac disease even with these less restrictiveguidelines. This biopsy avoiding policy may well be appliedto adults in the near future.21
It is essential that diagnostic guidelines be followed toensure that the diagnosis of celiac disease is well-founded.The disease, and its treatment, are considered life-long:we would not want a child or an adult to be committed toa socially restrictive diet unnecessarily; neither, of course,would we want to leave a patient with celiac disease on aharmful gluten-containing diet.
PETER H. R. GREEN
Celiac Disease Center
Columbia University Medical Center
New York, New York
Celiac Disease Centerand
Section of Pediatric Gastroenterology, Hepatology and NutritionUniversity of ChicagoChicago, Illinois