International Journal of Clinical Medicine, 2011, 2, 201-205
doi:10.4236/ijcm.2011.23033 Published Online July 2011 (http://www.SciRP.org/journal/ijcm)
201
The Prevalence of Celiac Disease in Patients with
Diabetes Mellitus
Deniz Gökalp1, Senay Arıkan1, Mustafa Yakut2, Yekta Tüzün3, Alparslan Kemal Tuzcu1, Mithat
Bahçeci1, Senay Arikan1, İrfan Soykan2*
1Endocrinology, Dicle University Medical School, Diyarbakır, Turkey; 2Gastroenterology Ankara University Medical School, An-
kara, Turkey; 3Gastroenterology, Dicle University Medical School, Diya rbakır, Turkey.
Received December 19th, 2010; revised April 6th, 2011; accepted April 26th, 2011.
ABSTRACT
Aim: To determine the prevalence o f celiac disease related autoantibo dies in patients with type 1 and 2 diabetes melli-
tus (DM), and to compare these results with the general population. Methods: In total, 137 consecutive patients with
type 1 DM, 172 with type 2 DM and 113 age-sex matched control subjects were included into the study. Antigli-
adin-autoantibodies (AGA) IgG and IgA, and endomysial-antibodies (EMA) IgG and IgA antibodies were determined.
Patients who were positive for one or more were o ffered a gastroduodeno scopic examination. Results: AGA IgG posi-
tivity was detected in 38.7% (53/137) patients with type 1 DM in 26.2% (45/172) patients with type 2 DM and in 16.8 %
(19/113) control subjects (significant differences). AGA IgA positivity was detected in 24.8 % (34/137) patients with type
1 DM, in 9.3% (16/172) patients with type 2 DM and in 3.5% (4/113) control subjects (significant differences). EMA
IgG positivity was detected in 10.2% (14/137) patients with type 1 DM in 0.6% (1/176) patients with type 2 DM and
0.9% (1/113) control subjects (significant differences). EMA IgA positivity was detected in 11.7% (16/137) patients with
type 1 DM in 0.6% (1/172) patients with type 2 DM and in none of control subjects. EMA IgA positivity was sig- nifi-
cantly higher in patients with type 1 DM as compared with patients with type 2 DM and controls. Conclusions: High
prevalence of celiac disease at the diagnosis of type 1 DM is observed. Serological markers are useful for id enti- fying
celiac disease patien ts with type 1 DM.
Keywords: Celiac Disease, Diabetes Mellitus, AGA IgA, AGA IgG, EMA IgA, EMA IgG
1. Introduction
Celiac disease (CD) is proved to be a T-cell mediated
gluten intolerance in genetically predisposed individuals
and it is made distinct and recognizable by villus atroph y
and malabsorption of small intestine. In some people
who are genetically apt to CD, it becomes visible after
ingestion of foods containing gluten and related proteins
(Alaedini and Green, 2005). CD shows a noticeable
regional frequence. Western Europe is the region where
CD is known to be prevalent. That CD is widespread
with an incidence of nearly 1% [1], has been unveiled by
serological screening studies .
It has been confirmed that in patients with type 1 DM
has the high ratio of general existence of CD. This
existence of relation is declared as 2.3% to 7% [2] which
are higher than general population. Originally, both of
the diseases are known as autoimmune and as a result of
an interaction b etween genetic and environmental factors,
they could be obvious. However, in patients with type 2
DM, CD related autoantibodies are not so common [3,4].
Short stature, sideropenic anemia, or hypertransa-
minasemia are several symptoms which are non-classical
forms of the disease that most patients who have
diabetes and CD. In many cases, patients are completely
symptom-free [5]. The risk of ever-evolving autoimmune
diseases, malignancy, osteoporosis, infertility, and intes-
tinal lymphomas is proportional to the time of exposure
to gluten in CD patients [6,7]; thus, for the early diag-
nosis of silent celiac disease, using serologic screening
tests is vital. AGA has merely 80% sensitivity and
specificity. [8] EMA has been proved to be a more
trustworthy marker than AGA since it has 90% of
sensitivity and specificity [10]. In screen ing purposes for
CD in diabetic patients, the united detection of anti-
gliadin and antiendomysium antibodies is being used
because its a sensitivity and specificity is 100%. In order
to uncover silent celiac disease in diabetic patients, these

The Prevalence of Celiac Disease in Patients with Diabetes Mellitus
202
antibodies are a reliable method.
As we scanned the literature, we did not encounter any
data about the frequency of CD related autoantibodies in
patients with DM in the Southeastern population of
Turkey. For this reason, in this study, we evaluated the
prevalence of CD related autoantibodies in patients with
type 1 and 2 DM and found a higher frequency of AGA
IgA, IgG and EMA IgA, IgG positivity in patients with
DM compared with the general populatio n.
2. Materials and Methods
The study was performed in Dicle University, Depart-
ment of Endocrinology between January 2002 and De-
cember 2008. In total, consecu tive 137 patients with type
1 DM, 172 patients with type 2 DM and 113 age-sex
matched contro l subjects were included into the study.
The laboratory evaluations of all patients and control
subjects were performed according to standart methods at
Dicle University Medicine Faculty Laboratory services.
AGA IgG and IgA, and EMA IgG and IgA antibodies
were determined by immunofluorescence method with
Euroimmun immunofluorescence autoantibody determi-
nation kits. Total IgA level was detected by nefelometric
technique by Beckmancoulter image to be able to ex-
clude patients or control subjects with IgA deficiency.
Spesific questions were asked and laboratory examina-
tions were performed by same physician to define signs
or symptoms of CD; development of diarrhea, hypo-
gonadism, anemia, weight loss, osteoporosis, dermatitis
herpetiformis and elevation of trans aminase s .
Patients who were positive for (AGA Ig A, EMA IgG,
and EMA IgA) one or more were offered a gastroduo-
denoscopic examination. In the intestinal histopathology-
cal analysis, more than three biopsy specimens were
taken from the second part of duodenum during gastro-
duodenoscopy. Hematoxylineosin staining was used in
these specimens. Slides were graded by conventional
histology as normal, with partial villous atrophy, and
with subtotal villous atrophy.
3. Statistical Analysis
Mean value, standard deviation (SD), minimum and
maximum values were calculated. Independent t test was
used to determine the mean values of the variables. The
relation of categorical variables was analyzed with the
Chi-square test. A p value <0.05 was accepted as statis-
tically significant.
4. Results
In total 137 patients with type 1 DM, 172 patients with
type 2 DM and 113 healthy subjects were included into
the study. Out of 137 patients with type 1 DM, 75 were
female and 62 were male, and out of 172 patients with
type 2 DM, 88 were female and 84 were male. Out of
113 control subjects, 62 were female and 51 were male.
Median age of patients with of type 1 DM was 28
(ranged from 16 to 39), median age of patients with type
2 DM was 42 (ranged from 29 to 68) and the control
subjects was 39 (ranged from 29 to 48). There was no
statistically significant difference between patients with
type 1 DM, type 2 DM and control subjects. Median age
of patients with type 1 DM was lower than patients with
type 2 DM and control subjects (p < 0.0001). There is no
statistically significant difference between median age of
patients with type 2 DM and control subjects. None of
the patients and control subjects has IgA deficiency.
AGA IgG positivity was detected in 53 out of 137
(38.7%) patients with type 1 DM, in 45 out of 172
(26.2%) patients with type 2 DM and in 19 out of 113
(16.8%) control subjects. There was a statistically sig-
nificant difference in terms of AGA IgG positivity be-
tween patients with type 1 DM and control subjects, and
patients with type 1 DM and type 2 DM patients (p =
0.000 and p = 0.019, respectively). However, there was
no statistically significant difference in terms of AGA
IgG positivity between p atients with type 2 DM and con-
trol subjects.
AGA IgA positivity was detected in 34 out of 137
(24.8%) patients with type 1 DM, in 16 out of 172 (9.3%)
patients with type 2 DM, and in 4 o ut of 113 (3.5%) con-
trol subjects.
There was a statistically significant difference in terms
of AGA IgA positivity between patients with type 1 DM
and control subjects (p = 0.000), There was no statistic-
cally significant difference in p atients with type 2 DM as
compared with controls. There was a statistically signify-
cant difference in terms of AGA IgA positivity between
patients with t y pe 1 DM and t y pe 2 DM (p = 0.000),
EMA IgG positivity was detected in 14 out of 137
(10.2%) patients with type 1 DM, in 1 out of 176 (0.6%)
patients with type 2 DM and 1 out of 113 (0.9%) control
subjects. EMA IgG positivity was statistically signify-
cantly higher in patients with type 1 compared with con-
trols subjects and patients with types 2 DM (p = 0.002, p
= 0.000). There was no statistically significant difference
between types 2 DM and control subjects in terms of
EMA IgG positivity.
EMA IgA positivity was detected in 16 out of 137
(11.7%) patients with type 1 DM, in 1 out of 172 (0.6%)
patients with type 2 DM and in none of control subjects.
EMA IgA positivity was statistically significantly higher
in patients with type 1 DM as compared with patients

The Prevalence of Celiac Disease in Patients with Diabetes Mellitus203
with DM type 2 and controls (p = 0.000, p = 0.000).
There was no statistically significant difference in terms
of EMA IgA positivity of patients between with type 2
DM and controls subjects. AGA-IgA and EMA-IgA
positivity was found in 10 out 137 patients with typa 1
DM, in out of 172 patients with type 2 DM and in none
of control subjects.Charasteristics of patients with type 1
DM, type 2 DM and control subjects were shown in Ta-
ble 1.
Of the 309 patients and 113 controls, 55 (13.4%) were
positive for at least one of the three antibodies (AGA IgA,
EMA IgG, and EMA IgA), and 41 accepted to undergo
an endoscopy. Sixteen patients 10.9% of the DM1 pa-
tients, 0.6% (n = 1of DM2 patients were histologically
diagnosed with CD. Most of these patients had no symp-
toms of CD.
5. Discussion
CD accounts for one of the best-understood examples of
an environmentally activated autoimmune disease, with
genetic risk factors (HLA-DQ2/DQ8), environmental
triggers gliadin, and autoantigenic targets tissue trans-
glutaminase being well defined [11]. Villous atrophy and
crypt hyperplasia, which ranges from silent asympto-
matic forms to active malabsorption syndromes are the
abnormalties in the small intestinal mucosa that show us
CD is a heterogenous disorder . Among patients with
celiac sprue, a large number of diseases such as Derma-
titis herpetiformis, type 1 DM, autoimmune thyroid di-
sease happen to be more frequent.
In patients with typ e 1 DM, its clinical and pathologic
manifestations, and typ ical gastrointestinal symptoms are
rare [12]. That screening is necessary to detect CD in
patients with type 1 DM is suggested by the increasing
numbers of reports of milder, less symptomatic forms of
CD without overt signs of malabsorption.
Undoubtedly, modern serology has become a reliable
method for better targeting patients to biopsy although
small intestinal biopsy has been the value test for
diagnosis [13] . In order to screen high risk groups and
the general population, both IgG and IgA AGA
anti-bodies have been extensively used. They are cheap,
easy to measure, and have reasonably high sensitivity
Table 1. Characteristics of patients and control subjects.
DM type 1
(n = 137) DM type 2 (n = 172) Controls (n = 113)
Female/male 66/62 88/84 62/51
Median age (year) 28(16 - 39) 42(29 - 68) 39(29 - 48)
AGA IgG positive 53(38.7%) 45(26.2%) 19(16.8%)
IgA positive 34(24.8%) 16(9.3%) 4(3.5%)
EMA IgG positive 14(10.2%) 1(0.6%) 1(0.9%)
IgA positive 16(11.7%) 1(0.6%) 0
and specificity [14]. On the other hand, the positive
foretelling value of this test is low for population
screening [15]. In contrast, with high predictive value
[15,16], EMA IgA is a highly sensitive (93% - 98%) and
specific (99% - 100%) test. Some investigators suggested
that in the presence of a positive EMA, a duodenal
biopsy is not necessary for diagnosis by the use of this
high specificity [17]. Since the majority of reports
indicating approximately 90% sensitivity, we understand
that generally the sensitiv ity o f th e EMA is excellen t [1 8 ].
The degree of mucosal damage is correlated with the titer
of EMA [19].
Two recent screening programs for CD in a cohort of
American type 1 DM patients, found a prevalence of
6.4% [20,21], just as the same with European patients
[22]. In an adult Tu rkish population with type 1 DM, the
prevalence of CD is reported as 2.45% [23]. When we
compare Turkish type 1 diabetes mellitus patients with
American and European patients, the high prevalence of
EMA antibodies (EMA IgG positivity 10.2%, EMA IgA
positivity 11.7%) may be caused by excessive bread
consumption in our region and the prevalence of gastro-
intestinal infection in children.
It is suggested by the previous studies that an inde-
pendent trigger effect on the development of CD might
be implemented by type 1 diabetes [24] The reaction to
gliadin in susceptible individuals could be triggered by
immunoregulatory disturbances in type 1 DM . Both type
1 DM and CD are associated with certain HLA alleles of
the DQB1 an d DQA1 locus. Some of the co-existance of
the two diseases may be explained by the HLA alleles;
however, the reason why type 1 DM diagnosis precedes
CD diagnosis is still unknown. It has been assumed that
there may be a nonspecific activation of the immune
system directed toward several dietary proteins attri-
butable to defective immunoregulation or loss of immu-
nologic tolerance of a variety of ingested antigens in
patients newly diagnosed with type 1 DM [25].
We found a high prevalence of CD at the diagnosis of
type 1 DM in the region. It is important to treat sub-
clinical CD, because patients with undiagnosed CD are at
risk of long-term CD complications, such as infertility,
anemia, osteoporosis, and malignancy. It has been re-
vealed that untreated celiac disease is associated with
high morbidity an d incr eased mortality [ 25 ,26 ]. Althou gh
the presentation of patients with celiac disease may be
protean, serological markers are a cheap and non-inva-
sive method for clinicians in primary care and secondary
care to identify patients with this disease.
REFERENCES
[1] R. J. Farrell and C. P. Kelly, “Celiac Sprue and Refrac-

The Prevalence of Celiac Disease in Patients with Diabetes Mellitus
204
tory,” In: M. Feldman, L. Lawrence and S. Friedman, Eds.,
Sleisenger and Fordtran’s Gastrointestinal and Liver Dis-
ease: Pathophysiology, Diagnosis, Management, 2-Volu-
mes Set, Saunders, Philadelphia, 2006, pp. 2277-2306
[2] M. Dirks, “Preventive Health Care, 2004 Update: Screen-
ing Persons with Insulin-Dependent Diabetes Mellituswith
Serology for Celiac Disease: Systematic Review & Rec-
ommendations,” Canadian Medical Association Journal, In
press.
[3] J. C. Cruz, E. C. Rode, L. S. Gomez, et al., “Celiac Dis-
ease Associated Antibodies in Persons with Latent
Autoimmune Diabetes of Adult and Type 2 Diabetes,”
Autoimmunity, Vol. 40, No. 2, 2007, pp. 103-107.
doi:10.1080/08916930601118825
[4] C. Catassi, E. Fabiani, I. M. Rätsch, G. V. Coppa, P. L.
Giorgi, et al., “The Coeliac Iceberg in Italy. A Multicen-
tre Antigliadin Antibodies Screening for Coeliac Disease
in School-Age Subjects,” Acta Paediatrica, Vol. 85, No.
S412, 1996, pp. 29-35.
doi:10.1111/j.1651-2227.1996.tb14244.x
[5] I. De Vitis, G. Ghirlanda and G. Gasbarrini, “Prevalence
of Coeliac Disease in Type I Diabetes: A Multicentre
Study,” Acta Paediatrica, Vol. 85, No. S412, 1996, pp.
56-57. doi:10.1111/j.1651-2227.1996.tb14253.x
[6] W. Dieterich, T. Ehnis, M. Bauer, et al., “Identification of
Tissue Transglutaminase as the Autoantigen of Celiac
Disease,” Nature Me dicine, Vol. 3, 1997, pp. 797-801.
doi:10.1038/nm0797-797
[7] R. F. A. Logan, E. A. Rifkind, I. D. Turner and A. Fer-
guson, “Mortality in Celiac Disease,” Gastroenterology,
Vol. 97, No. 2, 1989, pp. 265-271.
[8] A. Lerner and E. Lebe nthal, “The Controversy of the Use
of Anti-Gluten Antibody (AGA) as a Diagnostic Tool in
Celiac Disease,” Journal of Pediatrc Gastroenterology
and Nutrition, Vol. 12, No. 4, 1991, pp. 407-409.
doi:10.1097/00005176-199105000-00001
[9] H. Ascher, M. Hahn-Zoric, L. A. Hansson, A. F. Kilander,
L. A. Nilsson and H. T. laskalova, “Value of Serologic
Markers for Clinical Diagnosis and Population Studies of
Coeliac Disease,” Scandinavian Journal of Gastroen-
terology, Vol. 31, No. 1, 1996, pp. 61-67.
doi:10.3109/00365529609031628
[10] A. Bürgin-Wolff, H. Gaze, F. Hadziselimovic, et al., “An-
tigliadin and Antiendomysium Antibody Determination
for Coeliac Disease,” Archives of Disease in Childhood,
Vol. 66, No. 8, 1991, pp. 941-947.
doi:10.1136/adc.66.8.941
[11] R. J. Farrell and C. P. Kelly, “Current Concepts: Celiac
Sprue,” The New England Journal of Medicine, Vol. 346,
2002, pp. 180-188. doi:10.1056/NEJMra010852
[12] R. Lorini, A. Scaramuzza, L. Vitali, et al., “Clinical As-
pects of Coeliac Disease in Children with Insulin-De-
pendent Diabetes Mellitus,” Journal of Pediatric Endo-
crinology Metabolism, Vol. 9, No. S, 1996, pp. 101-111.
doi:10.1515/JPEM.1996.9.S1.101
[13] M. Markku and C. Pekka, “Celiac Disease,” The Lancet,
Vol. 349, No. 9067, 1997, pp. 1755-1759.
doi:10.1016/S0140-6736(96)70237-4
[14] G. R. Corazza, F. Biagi, M. L. Andreani, et al., “Screen-
ing Test for Celiac Disease,” The Lancet, Vol. 349, No.
9048, 1997, pp. 325-326.
doi:10.1016/S0140-6736(05)62823-1
[15] G. Corrao, G. R. Corazza, M. L. Andreani, et al., “Sero-
logical Screening of Celiac Disease: Choosing the Opti-
mal Procedure according to Various Prevalence Values,”
Gut, Vol. 35, No. 6, 1994, pp. 771-775.
doi:10.1136/gut.35.6.771
[16] S. Sulkanen, T. Halttunen, K. Laurila, et al., “Tissue
Trans-Glutaminase Autoantibody Enzyme-Linked Imu-
nosorbent Assay in Detecting Celiac Disease,” Gastroen-
terology, Vol. 115, No. 6, 1998, pp. 1322-1328.
doi:10.1016/S0016-5085(98)70008-3
[17] T. Valdimarsson, L. Franzen, E. Grodzinsky, et al., “Is
Small Bowel Biopsy Necessary in Adults with Suspected
Celiac Disease and Iga Anti-Endomysium Antibodies?
100% Positive Predictive Value for Celiac Disease in
Adults,” Digestive Diseases and Sciences, Vol. 41, No. 1,
1996, pp. 83-87. doi:10.1007/BF02208588
[18] A. Rostom, C. Dube, A. Cranney, et al., “Celiac Disease,”
Evidence Report/Technology Assessment, No. 104, AHRQ
Publication No. 04-E029-2, Rockville, MD: Agency for
Healthcare Research and Quality, 2004.
[19] C. Sategna-Guidetti, R. Pulitano, S. Grosso and G. Fer-
foglia, “Serum IgA-Antiendomysium Antibody Titers as
a Marker of Intestinal Involvement and Diet Compliance
in Adult Celiac Sprue,” Journal of Clinical of Gastroen-
terology, Vol. 17, No. 2, 1993, pp. 123-127.
doi:10.1097/00004836-199309000-00007
[20] A. H. Talal, J. A. Murray, J. A. Goeken, et al., “Celiac
Disease in Adult Population with Insulin-Dependent
Diabetes Mellitus: Use of Endomysial Antibody Testing,”
The American Journal of Gastroenterology, Vol. 92, No.
8, 1997, pp. 1280-1284.
[21] M. J. Rensch, J. A. Merenich, M. Lieberman, et al., “Glu-
ten-Sensitive Enteropathy in Patients with Insulin-De-
pendent Diabetes Mellitus,” Annals of Internal Medicine,
Vol. 124, No. 6, 1996, pp. 564-567.
[22] A. K. Carlsson, I. E. Axelsson, S. K. Borulf, et al.,
“Prevalence of IgA-Antiendomysium and IgA-Antigli-
adin Autoantibodies at Diagnosis of Insulin-Dependent
Diabetes Mellitus in Swedish Children and Adolescents,”
Pediatrics, Vol. 103, 1999, pp. 1248-1252.
doi:10.1542/peds.103.6.1248
[23] C. Aygun, S. Uraz, T. Damci, Z. Osar, V. Yumuk, E.
Akdenizli and H. Ilkova, “Celiac Disease in an Adult
Turkish Population with Type I Diabetes Mellitus,” Di-
gestive Diseases and Sciences, Vol. 50, No. 8, 2005, pp.
1462-1466. doi:10.1007/s10620-005-2862-8
[24] M. Hummel, E. Bonifacio, M. Stern, J. Dittler, A.
Schimmel and A. G. Ziegler, “Development of Celiac
Disease-Associated Antibodies in Offspring of Parents
with Type I Diabetes,” Diabetologia, Vol. 43, No. 8,
2000, pp. 1005-1011. doi:10.1007/s001250051483
[25] G. K. T. Holmes, “Long-Term Health Risks for Unrecog-

The Prevalence of Celiac Disease in Patients with Diabetes Mellitus
205
nized Coeliac Patients,” In: S. Auricchio and J. K. Visa-
korpi, Eds., Common Food Intolerance, I, Epidemiology
of Coeliac Disease, Karger, Basel, 1992, pp. 105-118.
[26] J. West, R. F. Logan, C. J. Smith, R. B. Hubbard and T. R.
Card, “Malignancy and Mortality in People with Coeliac
Disease: Population Based Cohort Study,” British Medi-
cal Journal, Vol. 329, No. 7468, 2004, pp. 716-719.