Prevalence and Clinical Picture of Celiac Disease in Italian
Down Syndrome Patients: A Multicenter Study
*Margherita Bonamico, *Paolo Mariani, *Helene Maria Danesi, *Massimo Crisogianni,
†Pinella Failla, ‡Gerolamo Gemme, §Alberto Rasore Quartino,?Aldo Giannotti,?Massimo Castro,
¶Fiorella Balli, #Margherita Lecora, #Generoso Andria, **Graziella Guariso, ††Orazio Gabrielli,
††Carlo Catassi, ‡‡Rosanna Lazzari, §§Nicoletta Ansaldi Balocco,??Stefano De Virgiliis,
¶¶Franco Culasso, †Corrado Romano, ##SIGEP, and ***Medical Genetic Group
*Department of Paediatrics, “La Sapienza” University, Rome; †I.R.C.C.S. Oasi Institute, Troina; ‡Department of Paediatrics,
University of Genoa, Genoa; §S. Martino Hospital, Genoa;?I.R.C.C.S., Bambino Gesù, Rome; ¶Department of Paediatrics,
University of Modena, Modena; #Department of Paediatrics, University Federico II, Naples; **Department of Paediatrics,
University of Padova, Padova; ††Department of Paediatrics, University of Ancona, Ancona; ‡‡Department of Paediatrics,
University of Bologna, Bologna; §§Department of Paediatrics, University of Turin, Turin;??Department of Paediatrics,
University of Cagliari, Cagliari; ¶¶Department of Experimental Medicine, Rome, Italy; ##Italian Society of Pediatric
Gastroenterology and Hepatology (SIGEP) participants: Francesca Ferretti (I.R.C.C.S. Bambino Gesù, Rome), Luigi Corvaglia
(Bologna), Maria Serenella Scotta (Varese), Massimo Spina and Novella Rotolo (Catania), Mariella Baldassarre (Bari),
Roberto Ferrari (Florence), and Franco Bascietto (Sulmona); ***Medical Genetic Group participants: Franco Colistro and
Maria Cristina Digilio (I.R.C.C.S. Bambino Gesù, Rome), Luigi Memo (Treviso), Livia Garavelli (Reggio Emilia), and
Sandra Brusa (Imola)
Background: A multicenter research study of Down syndrome
patients was carried out to estimate the prevalence of celiac
disease in patients with Down syndrome and to show clinical
characteristics and laboratory data of Down syndrome patients.
Methods: The authors studied 1,202 Down syndrome patients.
Fifty-five celiac disease patients (group 1) were compared with
55 immunoglobulin A antigliadin–positive antiendomysium
antibodies–negative patients (group 2) and with 57 immuno-
globulin A antigliadin–negative antiendomysium antibodies–
negative patients (group 3).
Results: Celiac disease was diagnosed in 55 of 1,202 Down
syndrome patients (4.6%). In group 1, weight and height per-
centiles were shifted to the left, whereas these parameters were
normally distributed in groups 2 and 3. In celiac patients, di-
arrhea, vomiting, failure to thrive, anorexia, constipation, and
abdominal distension were higher than in the other two groups.
Low levels of hemoglobinemia, serum iron, and calcium were
observed more frequently in group 1. The diagnosis of celiac
disease was made after a mean period of 3.8 years from the
initiation of symptoms. Sixty-nine percent of patients showed a
classic presentation, 11% had atypical symptoms, and 20% had
silent celiac disease. Autoimmune disorders were more fre-
quent (30.9%) in group 1 than in the other two groups exam-
ined (15%; P < 0.05).
Conclusions: This study reconfirms a high prevalence of celiac
disease in Down syndrome. However, the diagnostic delay, the
detection of atypical symptoms or silent form in one third of the
cases, and the increased incidence of autoimmune disorders
suggest the need for the screening of celiac disease in all Down
syndrome patients. JPGN 33:139–143, 2001. Key Words:
Celiac disease—Down syndrome—Multicenter study. © 2001
Lippincott Williams & Wilkins, Inc.
In 1975, Bentley (1) described a boy with Down syn-
drome (DS) and celiac disease (CD); in the following
years, other reports of the association of these two con-
ditions appeared in the literature (2–4). More recently,
efforts have been made to evaluate the prevalence of DS
in CD patients (5,6) and the prevalence of CD in DS
patients (7–13). These studies have led to the conclusion
that CD is significantly more frequent in DS patients
than in the general population.
Although anemia, diarrhea, and, more rarely, growth
failure (1–8,10) have been described in DS patients af-
Received October 26, 2000; accepted April 19, 2001.
Address correspondence and reprint requests to Dr. Margherita Bo-
namico, Istituto di Clinica Pediatrica, Policlinico “Umberto I,” Viale
Regina Elena 324, 00161 Rome, Italy (e-mail: margherita.bonamico@
Journal of Pediatric Gastroenterology and Nutrition
33:139–143 © August 2001 Lippincott Williams & Wilkins, Inc., Philadelphia
fected by CD, the clinical meaning of this association is
still unclear because a comparison between DS patients
with or without CD has never been reported in the lit-
Celiac disease screening in DS patients is usually
based on the determination of CD-associated antibodies.
Although in some studies the positivity of antigliadin
antibodies (AGA) has been used to select DS patients for
intestinal biopsy, these markers show a high rate of false
positivity (3,11–13). Currently, antiendomysium anti-
bodies (EMA) represent the more reliable marker for this
We carried out a survey based on the largest series of
DS patients ever reported in the literature, with the fol-
lowing aims in mind: to evaluate the prevalence of CD in
DS patients and to define the clinical characteristics of
CD among DS patients by comparing DS patients with or
MATERIALS AND METHODS
We studied 1,202 DS patients (1,110 children and adoles-
cents, aged 15 months to 18 years, and 92 adults, aged 18 to 46
years; 609 males and 593 females), living in various Italian
regions. These patients were consecutively enrolled under the
auspices of the Italian Society of Pediatric Gastroenterology
and Hepatology and the Clinical Genetics Group of the Italian
Society of Pediatrics from various centers in northern, central,
southern, and insular Italian regions, thus making our sample
fairly representative of the entire population.
The diagnosis of DS was confirmed by karyotype in all
cases. All patients underwent a series of routine laboratory
tests: hemoglobin, serum iron, calcium, proteins, albumin, and
immunoglobulin A (IgA).
Celiac disease was diagnosed according to revised European
Society of Paediatric Gastroenterology, Hepatology and Nutri-
tion (ESPGHAN) criteria (18). Patients were selected for in-
testinal biopsy on the basis of EMA positivity, AGA IgA posi-
tivity, or both in children younger than 2 years of age, because
in this age group, EMA positivity may have a false-negative
result (19). Levels of IgA AGA were measured by enzyme-
linked immunosorbent assay by the Alfa-gliatest (Eurospital,
Trieste, Italy) (20). Levels of EMA IgA were evaluated by an
indirect immunofluorescence method (Eurospital, Trieste,
Italy). Sections from the distal portion of monkey esophagus
were used as a substrate, and fluorescein-labeled goat antihu-
man IgA antibody was used as the second antibody. The pa-
tients’ serum was diluted 1:5 in phosphate buffer at pH 7.2. The
presence of a brilliant green network pattern under a fluores-
cence microscope was taken as a positive result (20). Patients
with IgA deficiency were not included in the study. Intestinal
biopsies were performed by Watson capsule or by pediatric or
The clinical features of 55 CD patients diagnosed by ESP-
GHAN Criteria (36 males, aged 4 to 46 years; group 1) were
compared with those observed in 55 IgA AGA-positive EMA-
negative DS patients (33 males, aged 3 to 40 years; group 2)
and in 57 IgA AGA-negative EMA-negative DS patients (34
males, aged 4 to 38 years; group 3). Group 2 and group 3
patients were selected randomly from among the screened pa-
tients to be age and gender matched to group 1. The 18 symp-
tomatic patients belonging to group 2 underwent intestinal bi-
opsy and showed normal small bowel mucosa.
A detailed questionnaire was completed to obtain informa-
tion (family or institution) about the patients, and about familial
gastroenterologic history. Special attention was given to feed-
ing habits (breast milk or formula, age of introduction of
gluten-containing foods). The gastrointestinal function was
also noted, particularly the features of CD, such as chronic
diarrhea, vomiting, failure to thrive, and anorexia. Information
on the presence of autoimmune or neoplastic conditions was
also obtained. All patients were receiving a gluten-containing
diet. Weight and height were evaluated using Down syndrome
percentile charts (DSPC)(21).
The Mann-Whitney U test and the Chi-square nonparametric
test were used to analyze the results.
Informed consent was obtained from the parents of all par-
ticipants, both for children and adults, before inclusion in the
study. The investigation was approved by the institutional ethic
As shown in Table 1, the screening of 1,202 DS pa-
tients with IgA AGA and EMA antibody determinations,
followed by intestinal biopsy in positive cases (48 chil-
dren and 7 adults), led to the diagnosis of CD in 55
patients (4.6%) by small intestinal biopsy (49 by endos-
copy and 6 by peroral capsule). Parents of eight EMA-
positive children and two EMA-positive adults did not
give permission for intestinal biopsy to be performed and
were not included among the 55 CD patients.
Almost all the 167 DS patients whose clinical charac-
teristics were compared lived with their families; only 2
of 55 patients affected by CD (group 1), one patient in
group 2 (IgA AGA positive, EMA negative), and one
patient in group 3 (IgA AGA negative, EMA negative)
were institutionalized. One hundred sixty-three of these
patients (97.6%) showed a complete trisomy 21; a mo-
saicism was observed in two DS patients with CD and in
two non-CD DS patients.
TABLE 1. Number and percentages of Down syndrome
patients (children and adults) with subtotal villous atrophy,
and number and percentages of Down syndrome patients
positive for AGA IgA and EMA
n ? 1,110
n ? 92
n ? 1,202
Subtotal villous atrophy
AGA IgA positive
AGA, antigliadin antibodies; EMA, antiendomysium antibodies;
IgA, immunoglobulin A.
M. BONAMICO ET AL.140
J Pediatr Gastroenterol Nutr, Vol. 33, No. 2, August 2001
No statistically significant differences were found
among the percentages of relatives with gastrointestinal
diseases in the three groups. Similarly, feeding habits
(breast milk duration and age of introduction of gluten-
containing foods) did not significantly differ among the
Weight and height percentile distributions of group 1
patients, referring to DS population percentiles, were
skewed because of the prevalence of patients with a rela-
tively lower weight and height, whereas in non-CD pa-
tients (groups 2 and 3), weight and height percentiles
showed a Gaussian distribution (Fig. 1 and 2).
As shown in Table 2, we observed in group 1 patients
higher percentages of cases with growth failure, diar-
rhea, and vomiting (P < 0.001), anorexia (P < 0.01),
and constipation (P < 0.05) than in groups 2 and 3.
In addition, we found in group 1 patients a higher fre-
quency of cases of low hemoglobinemia, low serum iron,
and low calcium (P < 0.01) than in the other two
Autoimmune disorders (more frequent hypothyroid-
ism, Hashimoto thyroiditis, type 1 diabetes, followed by
sporadic cases of autoimmune hepatitis and dermatitis
herpetiformis) were observed in 34 patients: 17 in group
1 (30.9%), four in group 2 (7%), and 13 in group 3
(22.4%; group 1 vs. group 2 plus group 3 P < 0.05). Two
patients from group 1 were affected by lymphoblastic
Among CD patients, 38 (69%) showed a classic form
of the disease, whereas 6 (11%) showed atypical symp-
toms (short stature in 5 patients and anemia in 2), and 11
(20%) showed a silent form. In 37 children and adoles-
cents, the median interval between the onset of symp-
toms and the diagnosis was 14 months (range, 2–127
months), and in 7 adults, the median interval was 36
months (range, 12–113 months).
The results of our multicenter study reconfirm the high
prevalence of CD in DS patients (4.6%). In fact, among
1,202 children and adults with DS, it was possible to
single out 55 biopsy-proven cases of CD. Nevertheless,
if we also take into account the 10 symptomatic IgA
AGA- and EMA-positive patients who did not undergo
intestinal biopsy, the prevalence of CD in our series
could increase further from 4.6% to 5.4%. Moreover,
these patients could be considered CD patients on the
basis of the excellent predictive value concerning the
morphologic status of the mucosa given by the combined
presence of these two antibodies (20,22). These preva-
lences are in agreement with those observed by Storm
(7), Castro et al. (8), Csizmadia et al. (9), and Zubillaga
et al. (10), who screened for CD in smaller series of DS
patients, from 70 (11) to 155 (8) patients. They found
prevalences of 2.6%, 4.5%, 7%, and 4.3%, respectively,
which far exceed the prevalence of CD in the general
population. Therefore, we can assume that this associa-
tion is not the result of chance.
The originality of our study lies in the search for the
characterization of CD in DS through the comparison of
group 1, made up of DS patients with known CD, with
two groups of DS patients without CD (group 2: IgA
AMA positive, EMA negative, with or without a nega-
tive biopsy result; group 3: IgA AMA negative, EMA
negative, with or without a negative biopsy result). Eigh-
teen group 2 patients with growth failure, iron defi-
ciency, or both were submitted to intestinal biopsy, and
the small bowel mucosa was normal. We did not perform
intestinal biopsy in all the remaining asymptomatic
group 2 patients for ethical reasons, as the recent litera-
ture on the association of CD and DS (10,11,13,14,15)
shows that most CD patients have been detected among
FIG. 1. Weight percentiles in Down syndrome patients of the three groups. Group 1: 55 patients affected by celiac disease; group 2: 55
immunoglobulin A (IgA) antigliadin antibody (AGA)–positive and antiendomysium antibody (EMA)–negative patients; group 3: 57 IgA
AGA–negative and EMA-negative patients.
CELIAC DISEASE IN DOWN SYNDROME PATIENTS141
J Pediatr Gastroenterol Nutr, Vol. 33, No. 2, August 2001
EMA-positive DS patients. On rare occasions, intestinal
villous atrophy was found in AGA IgA-positive and
EMA-negative patients. All such patients were always
younger than 2 years of age. The possibility of finding
IgA AGA positivity in patients with a normal mucosa
may be the result of the general elevated immunoglob-
ulin levels in patients with DS as compared with the total
population (23), particularly for food antigens (24) or to
a permeability defect (25).
Our most important result is the clinical presentation
of CD, which in the vast majority of cases is symptom-
atic. Indeed, growth failure, a common feature in CD,
was present in more than half of the DS patients with CD
and was much more frequent (P < 0.001) compared with
the other two groups. The characteristic gastrointestinal
symptoms of CD patients (diarrhea and vomiting) show
a higher prevalence in DS patients with CD (P < 0.001).
The presence of anorexia and constipation were also
more frequent in DS patients with CD than in the other
two groups (see Table 2).
Among laboratory data, the levels of hemoglobin, se-
rum iron, and, to a lesser degree, serum calcium were
lower in group 1. It is interesting to note that the in-
creased levels of aminotransferases in group 1 and group
2 patients. The possible transient increase of such en-
zymes has already been observed in CD (26–28). In
some group 2 patients with high levels of aminotransfer-
ases, an increased permeability or a lesion of the small
bowel may be present that permits gliadin peptide to
penetrate and induce AGA production, as reported in
food intolerance (25,26). In this condition, an increase of
aminotransferase levels has been observed also (27).
TABLE 2. Clinical characteristics and laboratory data of Down syndrome patients
(n ? 55) (%)
(n ? 55) (%)
(n ? 57) (%)
Low serum iron
Low serum calcium
Low serum albumin
Increased levels of AST
Increase levels of ALT
P < 0.001
P < 0.001
P < 0.001
P < 0.01
P < 0.05
P < 0.01
P < 0.01
P < 0.01
AST, aspartate aminotransferase; ALT, alanine aminotransferase.
Group 1: 55 Down syndrome celiac disease subjects. Group 2: 55 IgA antigliadin-positive, antiendo-
mysium antibodies–negative Down syndrome patients. Group 3: 57 IgA antigliadin-negative, antiendo-
mysium antibodies–negative Down syndrome subjects.
* P is relative to the comparison between group 1 versus groups 2 and 3 patients.
FIG. 2. Height percentiles in Down syndrome patients of the three groups. Group 1: 55 patients affected by celiac disease; group 2: 55
immunoglobulin A (IgA) antigliadin antibody (AGA)–positive, antiendomysium antibody (EMA)–negative patients; group 3: 57 IgA AGA–
negative and EMA-negative patients.
M. BONAMICO ET AL.142
J Pediatr Gastroenterol Nutr, Vol. 33, No. 2, August 2001
In the general population, the ratio between symptom-
atic and silent forms of CD is 1:8 (29), but this ratio is
reversed to 4:1 in our DS patients affected by CD. In fact,
80% of DS patients with CD showed a symptomatic
form of CD. One could hypothesize that, whereas in
CD patients without DS, mechanisms of compensation
may be present so that for a long time enteropathy may
exist without symptoms, in DS patients these mecha-
nisms are less able to overcome the overt clinical mani-
festation of CD.
Paradoxically, the very high prevalence of typical and
atypical forms of CD in DS patients could lead us to
consider that a DS population screening is not necessary.
However, we discovered that a long period of time
passes from onset of symptoms to diagnosis. This would
seem to be caused by a misinterpretation of symptoms
that could be present in persons with DS. The hypothesis
of malabsorption frequently is not considered in DS pa-
tients who live with families and not in institutions, even
if they undergo regular health check-ups. The screening
of CD in DS is also necessary because of the risk of these
patients being affected by autoimmune disorders and
cancer, common to both CD and DS. Indeed, group 1
showed a higher frequency of autoimmune disorders
compared with the other two groups (P < 0.01). The
young age of our patients (mean age, 6.5 years) could
lower the prevalence of these disorders, whose onset usu-
ally occurs during adolescence (30). It is difficult to com-
ment regarding malignancy because of the small number
of patients and their young ages.
In conclusion, the following findings of CD in this
population are as follows: the high frequency of symp-
tomatic forms suggests an increased expression of CD in
these impaired patients and the presence of failure to
thrive, anemia, and increased autoimmune disorders
should prompt an early diagnosis. Therefore, the results
of our multicenter study show that a systematic screening
for CD in DS patients is necessary.
1. Bentley D. A case of Down’s syndrome complicated by retino-
blastoma and celiac disease. Pediatrics. 1975;56:131–3.
2. Nowak TV, Ghisham FK, Schulze-Delrieu K. Celiac sprue in
Down’s syndrome: considerations on a pathogenic link. Am J Gas-
3. Ruch W, Schurmann K, Gordon P, et al. Coexistent coeliac dis-
ease, Graves’ disease and diabetes mellitus type I in a patient with
Down syndrome. Eur J Pediatr. 1985;144:89–90.
4. Santer R, Sievers E, Oldigs HD. Celiac disease in Down’s syn-
drome [letter]. J Pediatr Gastroenterol Nutr. 1991;13:121.
5. Amil Dias J, Walker-Smith J. Down’s syndrome and celiac dis-
ease. J Pediatr Gastroenterol Nutr. 1990;10:41–3.
6. Granditsch G, Rossipal E. Down’s syndrome and celiac disease
[letter]. J Pediatr Gastroenterol Nutr. 1990;11:279.
7. Storm W. Prevalence and diagnostic significance of gliadin anti-
bodies in children with Down syndrome. Eur J Pediatr. 1990;149:
8. Castro M, Crinò A, Papadatou B, et al. Down’s syndrome and
celiac disease: the prevalence of high IgA-antigliadin antibodies
and HLA-DR and DQ antigens in trisomy 21. J Pediatr Gastro-
enterol Nutr. 1993;16:265–8.
9. Csizmadia CG, Mearin ML, Oren A, et al. Accuracy and cost-
effectiveness of a new strategy to screen for celiac disease in
children with Down syndrome. J Pediatr. 2000;137:756–61.
10. Zubillaga P, Vitoria JC, Arrieta A, et al. Down’s syndrome and
coeliac disease. J Pediatr Gastroenterol Nutr. 1993;16:168–70.
11. Failla P, Ruberto C, Pagano MC, et al. Coeliac disease in Down’s
syndrome with HLA serological and molecular studies. J Pediatr
Gastroenterol Nutr. 1996;23:303–6.
12. Bonamico M, Rasore-Quartino A, Mariani P, et al. Down’s syn-
drome and coeliac disease: usefulness of antigliadin and antiendo-
mysium antibodies. Acta Paediatr. 1996;85:1503–5.
13. Gale L, Wimalaratna H, Brotodiharjo A, et al. Down’s syndrome
is strongly associated with coeliac disease. Gut. 1997;40:492–6.
14. Carlsson A, Axelsson I, Borulf S, et al. Prevalence of IgA-
antigliadin antibodies and IgA-antiendomysium antibodies related
to celiac disease in children with Down’s syndrome. Pediatrics.
15. Morris MA, Biagi F, Ellis HJ, et al. Are Down syndrome and
coeliac disease associated? Gut. 1997;41:724–5.
16. Hansson T, Annerén G, Sjoberg O, et al. Celiac disease in relation
to immunologic serum markers, trace elements and HLA-DR and
DQ antigens in Swedish children with Down syndrome. J Pediatr
Gastroenterol Nutr. 1999;29:286–92.
17. Zachor DA, Mroczek-Musulman E, Brown P. Prevalence of celiac
disease in Down syndrome in the United States. J Pediatr Gastro-
enterol Nutr. 2000;31:275–9.
18. Walker-Smith JA, Guandalini S, Schmitz J, et al. Revised criteria
for diagnosis of coeliac disease. Arch Dis Child. 1990;65:909–11.
19. Burgin-Wolff A, Gaze H, Hadziselimovic F, et al. Antigliadin and
antiendomysium antibody determination for coelic disease. Arch
Dis Child. 1991;66:941–7.
20. Savilhati E, Perkkio M, Kalimo K. IgA antigliadin antibodies: a
marker of mucosal damage in childhood coeliac disease. Lancet.
21. Cronk C, Crocker AC, Pueschel SM, et al. Growth charts for
children with Down syndrome: 1 month to 18 years of age. Pedi-
22. Cataldo F, Ventura A, Lazzari R, et al. Antiendomysium antibodies
and coeliac disease: solved and unsolved questions. An Italian
multicenter study. Acta Paediatr. 1995;84:1125–31.
23. Burgio GR, Ugazio AG, Nespoli L, et al. Derangements of immu-
noglobulin levels, phytohemagglutinin responsiveness and T and B
cell markers in Down’s syndrome at different ages. Eur J Immunol.
24. Reichelt KL, Lindback T, Scott H. Increased levels of antibodies to
food proteins in Down syndrome. Acta Paediatr Jpn. 1994;36:
25. Ferrante E, Bonamico M, Mariani P, et al. High serum antigliadin
antibodies in children with Down syndrome and in a subject with
18q-syndrome [abstract]. Eur J Epidemiol. 1991;7:732–5.
26. Lindberg T, Berg NO, Borulf S, et al. Liver damage in coeliac
disease or other food intolerances in childhood. Lancet. 1978;18:
27. Bonamico M, Pitzalis G, Culasso F, et al. Il danno epatico nella
malattia celiaca del bambino. Minerva Pediatr. 1986;38:959–62.
28. Volta U, De Franceschi L, Lari F, et al. Coeliac disease hidden by
cryptogenic hypertransaminasaemia. Lancet. 1998;352:26–9.
29. Catassi C, Fabiani E, Ratsch IM, et al. The coeliac iceberg in Italy.
A multicentre antigliadin antibodies screening for coeliac disease
in school-age subjects. Acta Paediatr Suppl. 1996;412:29–35.
30. Toscano V, Conti FG, Anastasi E, et al. Importance of the diet in
the induction of endocrine autoimmunity and endocrine dysfunc-
tion in adolescent celiac patients. Am J Gastroenterol. 2000;35:
CELIAC DISEASE IN DOWN SYNDROME PATIENTS143
J Pediatr Gastroenterol Nutr, Vol. 33, No. 2, August 2001