| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Endocrinology-Diabetology, University of Antwerp, University Hospital Antwerp (C.E.M.D.B., I.H.D.L., L.F.V.G.), B-2650 Edegem; and the Belgian Diabetes Registry, B-1090 Brussels, Belgium
Address all correspondence and requests for reprints to: Christophe De Block, M.D., Department of Endocrinology-Diabetology, University of Antwerp, University Hospital Antwerp, Wilrijkstraat 10, B-2650 Edegem, Belgium. E-mail: christophe.deblock{at}uia.ua.ac.be
| Abstract |
|---|
|
|
|---|
The population studied consisted of 497 type 1 diabetics (men/women, 252/245; mean age, 40.8 ± 12.1 yr; mean duration of disease, 16.4 ± 10.4 yr; mean age at onset, 26.9 ± 13.5 yr; mean hemoglobin A1c, 8.1 ± 1.6%). Associated autoantibodies were present in 39% and PCA were present in 20.9% of the subjects, particularly in older patients. Gender, duration, and age at onset of diabetes did not influence the appearance of PCA. Antithyroid peroxidase antibodies (aTPO) were more frequent in PCA-positive patients than in those without PCA (33.6% vs. 22.4%; P = 0.025), suggesting an association between gastric and thyroid autoimmunity. We could demonstrate an association between PCA and the HLA DR5 haplotype (P = 0.001) as well, but not with HLA DR3 and/or DR4. In the PCA-positive group, iron deficiency anemia was detected in 15.4%, and pernicious anemia was found in 10.5% of subjects. These autoimmune gastric manifestations were significantly more prevalent in PCA-positive diabetics than in PCA-negative subjects, in whom the percentages were 6.9% and 0.5%, respectively (P = 0.01 and P < 0.0001). PCA were prevalent in 84.6% of patients with pernicious anemia. A gastroscopic and anatomopathoplogical examination performed in a subgroup of 30 patients with gastric symptoms revealed atrophic gastritis in 13 of 14 PCA-positive patients and in 9 of 16 PCA-negative subjects (P = 0.04). PCA were present in 59.1% of patients with atrophic gastritis.
In conclusion, a high prevalence of parietal cell antibodies and associated autoimmune gastric disease is present in PCA-positive type 1 diabetics, recommending its screening. Early detection of PCA and iron deficiency anemia, pernicious anemia, and atrophic gastritis and the subsequent care could reduce the morbidity of type 1 diabetes.
| Introduction |
|---|
|
|
|---|
In this article we report the results of a study on type 1 diabetes and associated gastric autoimmunity and compare them to literature data. This cross-sectional study aimed to determine the grade of autoimmunity and assess the pattern of prevalence of PCA by age, sex, duration, age at onset of diabetes, and human leukocyte antigen (HLA) DR haplotype. Furthermore, manifestations of gastric autoimmunity were studied to assess the clinical significance of being PCA positive.
| Subjects and Methods |
|---|
|
|
|---|
|
aTPO were assayed with a specific immunological test (Immutest, Henning, Berlin, Germany). Values greater than 100 U/mL were considered positive. Thyroid function was estimated by TSH (normal, 0.155.5 µU/mL) and freeT4 (free tetraiodothyronine; normal, 1124 pmol/L) levels. To detect PCA, indirect immunofluorescence (fluorescein isothiocyanate staining) was used with rat gastric mucosa as substrate (MeDiCa kit, Medical Diagnostics California, Carlsbad, CA). A titer higher than 1:20 was regarded as positive. Determining PCA by enzyme-linked immunosorbent assay is about 10-fold more sensitive and about equally specific compared to the indirect immunofluorescence method, as demonstrated by Karlsson et al. (24). However, data obtained with both techniques correlate well. In our study, AAA were also determined by indirect immunofluorescence, using kits from the same company (MeDiCa kit). Thus, for practical reasons the more routinely used immunofluorescence technique was performed in this study. AAA were determined by means of indirect immunofluorescence, using unfixed frozen sections of monkey adrenal tissue as substrate, with titers higher than 1:10 being scored as positive (MeDiCa kit, catalog no. 6001-AG). Antiintrinsic factor was assayed with the enzyme-linked immunosorbent assay technique, using purified intrinsic factor from pig gastric mucosa as substrate (Malakit); a titer higher than 1.1 was considered positive. Pernicious anemia was defined as a megaloblastic anemia with positive antiintrinsic factor and/or PCA.
Serologic HLA DR typing was performed in 280 patients using a two-color fluorescence technique as described by Van Rood et al. (25). The most recently published nomenclature of the HLA system was used (26).
Serum iron, total iron binding capacity, and a peripheral hematogram (hemoglobin, hematocrit, mean corpuscular volume, and mean corpuscular hemoglobin) were determined by routine laboratory tests to evaluate iron status. Serum gastrin (normal, <110 pg/mL) was measured using a RIA liquid technique. To collect some preliminary data giving an indication of whether atrophic gastritis is present in a high percentage of PCA-positive subjects, a gastroscopic and anatomopathological examination was performed in a subgroup of 30 patients (PCA+ vs PCA-, 14 vs. 16) with gastric symptoms.
The study was approved by the local ethical committee of the University Hospital of Antwerp. Informed consent was obtained from each patient in accordance with the Helsinki Declaration.
All data from this cross-sectional study were analyzed using the
statistical package SPSS version 8.0 (SPSS, Inc., Chicago,
IL). The unpaired t test was used for comparison of means
and the
2 test or Fishers exact test was
used to test frequency differences. Multiple logistic regression,
stepwise forward, was used to assess the strength and independency of
associations and to describe an adequate predictive model. Women were
scored 0, and men were scored 1. All tests were performed two-sided.
P < 0.05 or less was considered significant.
| Results |
|---|
|
|
|---|
|
|
Four PCA-positive diabetics had an overt pernicious anemia, and seven
PCA-positive and two PCA-negative patients showed a latent pernicious
anemia. In 84.6% of patients, characterized by this megaloblastic
anemia, PCA were present. The prevalence of this form of anemia was
significantly higher in PCA-positive (10.5%) than in PCA-negative
diabetics (0.5%; see Table 2
). No differences in gender distribution
were found.
Most patients showed the HLA DR3 and/or DR4 haplotype, as expected. No
difference in HLA DR type between men and women was detected. We did
observe that significantly more PCA-positive diabetics (17%) showed
the HLA DR5 haplotype compared to the PCA-negative group (3.5%;
P = 0.001). We could not find an association with HLA
DR3 and/or DR4 (see Table 2
).
Independent risk factors for having PCA were studied with stepwise forward logistic regression analysis. When the model tested age, gender, duration of disease, age at onset, HbA1c, insulin dose, aTPO, AAA, HLA DR17, gastrin, iron deficiency anemia, and pernicious anemia as covariates, only gastrin levels (ß = 1.74; P = 0.017) and the HLA DR5 haplotype (ß = 0.003; P = 0.049) were independently associated with PCA positivity. However, this model only included 158 patients. To detect possible associations among data from all 497 patients, insulin dose, AAA, gastrin, and HLA DR17 were removed from the previous model. Anti-TPO-positivity (ß = 0.86; P = 0.008), ferriprive anemia (ß = 0.92; P = 0.015), and pernicious anemia (ß = 1.61; P = 0.003) turned out to be associated with PCA positivity. Logistic regression analysis, stepwise forward, was also performed to study risk factors for microcytic hypochromic anemia. PCA (ß = 0.81; P = 0.034), gender (ß = -0.99; P = 0.007), and duration of diabetes (ß = 0.06; P = 0.0005) were associated with microcytic hypochromic anemia when the model tested age, gender, duration, age at onset, HbA1c, PCA, pernicious anemia, and HLA DR17 as independent risk factors.
A subgroup of 30 patients (PCA+ vs. PCA-, 14 vs. 16) demonstrating gastric symptoms underwent gastroscopy. Atrophic gastritis was documented by gastroscopic and anatomopathological examination in 73.3% of this subgroup with gastric symptoms. In the PCA-positive group, 13 of 14 patients (92.8%) were detected with atrophic gastritis compared to a significantly lesser number of 9 of 16 diabetics in the PCA-negative group (56.3%; P = 0.0395; OR, 10.11). PCA were prevalent in 59.1% of patients with atrophic gastritis. Gender did not influence the prevalence of atrophic gastritis. In diabetics with atrophic gastritis more patients with iron deficiency anemia were found (36.4%) compared to those without atrophic gastritis (12.5%), although this difference was not statistically significant. Hypergastrinemia was significantly more prevalent in patients with atrophic gastritis (72.7%) than in those who did not have atrophic gastritis (25%; P = 0.034; OR, 8.00).
| Discussion |
|---|
|
|
|---|
Type 1 diabetic adults have a high grade of associated autoimmunity. In the present study 39% of the patients showed one or more autoantibodies. The prevalence of PCA is high in our study, namely 20.9%, compared to previously reported prevalences ranging between 528% in type 1 diabetics and 1.412% in nondiabetic controls (11, 14, 17, 19, 28).
PCA positivity was found significantly more at an older age, although multiple logistic regression did not withhold age as an independent risk factor. Neither duration of disease, age at onset of diabetes, nor gender triggered the appearance of these autoantibodies. The association we found between gastric and thyroid autoimmunity has been described previously (16). Regarding the HLA system, Gorsuch et al. reported that thyrogastric autoimmunity did not segregate with the HLA-bound high risk haplotypes of type 1 diabetes mellitus (29). In the present study a positive association was detected between PCA positivity and the HLA DR5, but not DR3 and/or DR4 haplotype. To our knowledge this has not been previously described. We must indicate however that although this is a statistically significant association, the attributable fraction for the allele is low.
The clinical importance of gastric parietal cell autoantibodies is
demonstrated by the association with manifestations of autoimmune
gastropathy (see Fig. 2
). PCA are targeted against the gastric proton
pump, the H+/K+-adenosine
triphosphatase (30, 31, 32), resulting in decreased gastric acid secretion
and decreased iron absorption. This can lead to iron deficiency anemia,
which is a common and first, although frequently overlooked,
manifestation of autoimmune atrophic gastritis (20, 21, 22). Patients with
iron deficiency, mainly PCA-positive subjects detected during an
initial screening study (22), were supplemented with iron, resulting in
normalization of the serum iron levels, as shown in Table 2
. Elevated
levels of gastrin are also associated with autoimmune gastropathy.
However, besides arising from the hypo- or achlorhydric situation in
the case of atrophic gastritis, elevated gastrin levels in type 1
diabetics can be caused by hypo- or hyperglycemia, both of which can
lead to reduced gastric acid secretion (33, 34). In the present study
no correlation was found between levels of gastrin and
HbA1c. However, hypergastrinemia was
significantly associated with being PCA positive. On the other hand,
many type 1 diabetic patients, and not only those with gastroparesis,
show a triple increase in fasting and postprandial gastrinemia.
Helicobacter pylori infection is another possible reason for
hypergastrinemia (35). Although there are indications that H.
pylori can trigger autoimmune inflammation, its role in autoimmune
gastritis is unknown (36, 37, 38).
|
Parietal cell antibodies are recognized as a marker of an autoimmune type A gastritis, because they are cytotoxic against the gastric mucosa (17). The autoimmune process results in mucosal atrophy and decreased gastric acid and intrinsic factor secretion, subsequently causing hypergastrinemia and vitamin B12 deficiency (42, 43), as discussed above. In this study the presence of atrophic gastritis was investigated in a subgroup of 30 patients. For ethical reasons, at this point only those patients demonstrating gastric symptoms were studied. Atrophic gastritis was present in a very high percentage (73.3%) of this subgroup, although this may be overestimated because only those with gastric symptoms were studied. Supposing patients without symptoms would have much less incidence of atrophic gastritis, the prevalence of this type of gastritis in the whole type 1 diabetic population would be about 4.58%. In the PCA-positive group significantly more were detected with atrophic gastritis than in the group without PCA. The prevalence of PCA in patients with atrophic gastritis was 59.1%. This is in line with data from the literature, where PCA could be found in 60% of patients with atrophic gastritis. The observed associations between PCA and hypergastrinemia, on the one hand, and atrophic gastritis, on the other hand, could be confirmed by our study results. The fact that iron deficiency anemia was not associated with atrophic gastritis is because of the small number of patients in this substudy.
Data from this study and the literature indicate that PCA are good markers of autoimmune gastric manifestations, such as iron deficiency anemia, pernicious anemia, and atrophic gastritis. The well known complications of these three disorders can influence the prognosis of the diabetic patient. Iron deficiency anemia can severely influence cellular immunity, intestinal function, growth, work capacity, cardiopulmonary status, and intellectual performance (44, 45). Pernicious anemia can cause neurologic complications, such as paraesthesia, loss of perception of vibration, and proprioception, which are amenable to vitamin B12 therapy, and is recognized as an early stage of gastric adenocarcinoma (46, 47). Atrophic gastritis itself is also regarded as a possible precursor of gastric adenocarcinoma, which manifests itself only very late in the disease process. Therefore, particularly in anemic patients, screening for PCA and, in case of positivity, additional evaluation, including gastroscopy with biopsy, are needed to take necessary precautions to protect against a bad outcome of these disease processes.
Conclusions
We can conclude that type 1 diabetic patients, particularly at an older age, have a high prevalence of PCA. These PCA can be used as a marker for iron deficiency anemia and hypergastrinemia associated with autoimmune atrophic gastritis and for pernicious anemia. We strongly suggest that physicians faced with a diabetic patient who has anemia and/or gastrointestinal symptoms should measure PCA to help define the pathogenesis of the problem. Based on the high grade of PCA positivity and looking at the adverse impact of these diseases on the health of the diabetic patient in general, screening for PCA and, in case of positivity, an adequate work-up are warranted.
| Acknowledgments |
|---|
We thank the members of the Belgian Diabetes Registry: E. Balasse (Brussels), H. Becq (Wilrijk), A. Bodson (Jumet), L. Claeys (Zoersel), M. Coeckelberghs (Antwerp), J.-L. Coolens (Hasselt), W Coucke (Roeselare), E. Couturier (Brussels), R. Craen (Ghent), J.-C. Daubresse (Charleroi), P. Decraene (Bonheiden), I. De Leeuw (Antwerp), C. Delvigne (Antwerp), J. De Schepper (Brussels), M. Du Caju (Antwerp), F. Féry (Brussels), N. Gaham (Braine-lAlleud), J. Gérard (Liege), C. Gillet (Brussels), J. Guillot (Seraing), C. Herbaut (Mons), B. Keymeulen (Brussels), G. Krzentowski (Jumet), G. Lamberigts (Brugge), M. Letiexhe (Liege), K. Logghe (Roeselare), J. Monballyu (Ekeren), D. Nicolaij (Kortrijk), F. Nobels (Aalst), M.-C. Pelckmans (Lier), A. Purnode (Brussels), M.-P. Roggemans (Brussels), R. Rooman (Antwerp), R. Rottiers (Ghent), A. Scheen (Liege), J. Schutyser (Kortrijk), L. Terriere (Hoboken), J. Teuwen (Kapellen), G. Thenaers (Hasselt), K. Van Acker (Antwerp), P. Van Crombrugge (Aalst), E. Vandenbussche (Herentals), L. Van Gaal (Antwerp), S. Van Imschoot (Brugge), S. Vanneste (Zoersel), P. Van Rooy (Antwerp), and J. Vertommen (Antwerp).
| Footnotes |
|---|
Received September 23, 1998.
Revised January 28, 1999.
Revised April 28, 1999.
Accepted July 13, 1999.
| References |
|---|
|
|
|---|
-subunit of the
human H,K-adenosine triphosphate. Scand J Gastroenterol. 29:122127.[Medline]
This article has been cited by other articles:
![]() |
N. Alonso, M. L. Granada, I. Salinas, A. M. Lucas, J. L. Reverter, J. Junca, A. Oriol, and A. Sanmarti Serum Pepsinogen I: An Early Marker of Pernicious Anemia in Patients with Type 1 Diabetes J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5254 - 5258. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. W. S. Rosario, J. S. Reis, M. A. Borges, R. Amim, M. R. Calsolari, S. C. Silva, and S. Purisch Extrapancreatic Autoimmunity in Patients With Latent Autoimmune Diabetes of Adults Diabetes Care, February 1, 2005; 28(2): 496 - 497. [Full Text] [PDF] |
||||
![]() |
M. Segni, O. Borrelli, I. Pucarelli, G. Delle Fave, A. M. Pasquino, and B. Annibale Early Manifestations of Gastric Autoimmunity in Patients with Juvenile Autoimmune Thyroid Diseases J. Clin. Endocrinol. Metab., October 1, 2004; 89(10): 4944 - 4948. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. E.M. De Block, G. Colpin, K. Thielemans, W. Coopmans, J. J.P.M. Bogers, P. A. Pelckmans, E. A.E. Van Marck, V. Van Hoof, M. Martin, I. H. De Leeuw, et al. Neuroendocrine Tumor Markers and Enterochromaffin-Like Cell Hyper/Dysplasia in Type 1 Diabetes Diabetes Care, June 1, 2004; 27(6): 1387 - 1393. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Zendehdel, O. Nyren, C.-G. Ostenson, H.-O. Adami, A. Ekbom, and W. Ye Cancer Incidence in Patients With Type 1 Diabetes Mellitus: A Population-Based Cohort Study in Sweden J Natl Cancer Inst, December 3, 2003; 95(23): 1797 - 1800. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. E.M. De Block, I. H. De Leeuw, J. J.P.M. Bogers, P. A. Pelckmans, M. M. Ieven, E. A.E. Van Marck, K. L. Van Acker, and L. F. Van Gaal Autoimmune Gastropathy in Type 1 Diabetic Patients With Parietal Cell Antibodies: Histological and clinical findings Diabetes Care, January 1, 2003; 26(1): 82 - 88. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. E.M. De Block, I. H. De Leeuw, P. A. Pelckmans, D. Callens, E. Maday, and L. F. Van Gaal Delayed Gastric Emptying and Gastric Autoimmunity in Type 1 Diabetes Diabetes Care, May 1, 2002; 25(5): 912 - 917. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. L. Lethagen, U.-B. Ericsson, B. Hallengren, L. Groop, and T. Tuomi Glutamic Acid Decarboxylase Antibody Positivity Is Associated with an Impaired Insulin Response to Glucose and Arginine in Nondiabetic Patients with Autoimmune Thyroiditis J. Clin. Endocrinol. Metab., March 1, 2002; 87(3): 1177 - 1183. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. E. M. De Block, I. H. De Leeuw, K. Decochez, F. Winnock, J. Van Autreve, C. M. Van Campenhout, M. Martin, and F. K. Gorus The Presence of Thyrogastric Antibodies in First Degree Relatives of Type 1 Diabetic Patients Is Associated with Age and Proband Antibody Status J. Clin. Endocrinol. Metab., September 1, 2001; 86(9): 4358 - 4363. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |