Autoimmune thyroid disease in patients with anti-gad positive type 1 diabetes mellitus

Transcription

1 Cent. Eur. J. Med. DOI: /s z Central European Journal of Medicine Kamile Gul 1, Ihsan Ustun 2, Yusuf Aydin 3, Dilek Berker 3, Halil Kutlu Erol 4, Mustafa Unal 5, Ayse Ozden Barazi 6, Tuncay Delibasi 3, Serdar Guler 3 1 Ankara Ataturk Education and Research Hospital, Department of Endocrinology and Metabolism, 6533 Ankara, Turkey 2 Mersin State Hospital, Department of Endocrinology, Mersin Research Article 3 Ankara Numune Training and Research Hospital, Department of Endocrinology and Metabolism, 6100 Ankara, Turkey 4 University of Miami Miller School of Medicine, Department of Psychiatry and Behavioral Sciences, Miami, Florida, USA 5 Afyon State Hospital, Department of Endocrinology, Afyon, Turkey 6 Ankara Numune Training and Research Hospital, Department of Biochemistry, 6100 Ankara, Turkey Received 20 July 2008; Accepted 11 May 2009 Abstract: The aim of the study was to determine the frequency and titers of anti-thyroid peroxidase (Anti-TPO), anti-thyroglobulin (Anti-TG), and anti-glutamic acid decarboxylase (Anti-GAD) antibodies in Turkish patients with type 1 diabetes mellitus (DM), and to compare the frequency of anti-tpo and anti-tg titers in the presence or absence of anti-gad. A total of 104 patients including 56 males and 48 females with type 1 DM and their age-, gender-, and body mass index-matched control group, including 31 males and 27 females, 58 cases in total with an age range of years, were recruited into this study. In patients with type 1 DM, positive anti-gad was detected in 30.8% (n=32). In patients with positive anti-gad, rate of positive anti-tpo was 37.5%; however, in patients with negative anti-gad, the rate of positive anti-tpo was 9.7% and the difference was statistically significant (p=0.001). In patients with positive anti-gad, the rate of positive anti-tg was 18.8%. In patients with negative anti-gad, the rate of positive anti-tg was 2.8%, and the difference between them was statistically significant (p=0.005). In patients with positive and negative anti-gad, rates of both positive anti-tpo and anti-tg were 15.6% and 1.4%, respectively, with the difference showing statistical significance (p=0.004). Thyroid autoimmunity in type 1 DM patients with positive anti-gad was apparently higher; therefore, these patients should be followed more frequently and carefully. Keywords: Thyroid Autoantibodies Anti GAD Type 1 Diabetes Mellitus Versita Warsaw and Springer-Verlag Berlin Heidelberg. 1. Introduction Type 1 diabetes mellitus (DM) is frequently associated with other organ-specific autoimmune diseases, including autoimmune thyroid disease (ATD). Several studies in children and adults with type 1 DM have shown a high prevalence of thyroid autoantibodies; this has been found to be an indicator of ATD [1-4]. The similar pathogenesis of type 1 DM and ATD, and their tendency to occur together, suggest that their etiology may involve shared genetic factors. Both type 1 DM and ATD are organ-specific T-cell mediated diseases in which T-cell infiltration results in dysfunction of the target organ (the pancreatic islets in type 1 DM and the * iustun@yahoo.com

2 thyroid in ATD). Moreover, two immune regulatory genes, human leukocyte antigen and cytotoxic T lymphocyteassociated antigen 4, contribute to the susceptibility to both diseases [5]. Anti-thyroid peroxidase antibodies (Anti-TPO) and/or anti-thyroglobulin antibodies (Anti-TG) can be detected long before the changes in TSH and thyroid hormone levels occur. Thus, determination of these antibodies might be useful for early diagnosis of the disease before abnormal thyroid-stimulating hormone (TSH) values develop [6]. However, thyroid autoantibodies in the serum do not always appear in ATD, because differences in the sensitivity of autoantibody tests may arise from the design of the assay (e.g., competitive immunometric assay (RIA) versus two-site noncompetitive immunometric assay (IMA)), as well as from the physical method used for the signal (e.g., radioisotope versus chemiluminescence) [7,8]. Therefore in some cases there is a need for alternative approaches to confirm the diagnosis, such as ultrasound imaging and fine needle biopsy. Ultrasound studies of the thyroid gland have shown that typical patterns of parenchyma hypoechogenicity, as well as gland enlargement, may be present in patients with ATD [9]. In adults with longstanding type 1 DM, 27% to 44% of patients have thyroid autoantibodies (TA) [10,11], while 23% to 25% of these patients develop abnormal TSH values (sub- or clinical hypo- or hyperthyroidism) [6,7]. Type 1 DM is a predominantly immune-mediated type of diabetes that results from chronic progressive T-cell-mediated autoimmune destruction of the betacells of the pancreas. Markers of the process are various, including autoantibodies to glutamate decarboxylase, to islet cell and to insulin. At least one of these is present in 85% to 98% of newly diagnosed children. Patients with immune-mediated type 1 DM are also prone to other autoimmune disorders [12]. There have also been reports indicating that ATD patients have a high prevalence of autoantibodies (islet cell antibodies, anti- GAD) specific for type 1 DM [13-15]. The present study aimed to determine the frequency and titers of anti-tpo, anti-tg, and anti-gad in Turkish patients with type 1 DM, and to compare the frequency of anti-tpo, and anti-tg in the presence or absence of anti-gad. Additionally, abnormal TSH values, TA, and thyroid ultrasonography (US) abnormalities were evaluated. 2. Material and Methods The study was conducted between July 2003 and June 2006 at the inpatient and outpatient clinics of the Endocrinology and Metabolism Clinic of Ankara Numune Education and Research Hospital, Ankara, Turkey. It included 56 males and 48 females (104 patients) with type 1 DM; the age-, gender-, and body mass index (BMI)-matched control group included 31 males and 27 females (58 controls), with an age range of years. All patients and healthy controls were included in the study following the completion of voluntary informed consent forms; the study was conducted according to the stipulations of the Declaration of Helsinki. The mean ages of subjects with type 1 DM was 26.43±7.69 years; the median duration of diabetes was 6 years (range 0 33), and the diagnosis of type 1 DM was based on the published criteria [16], with a typical clinical history of diabetic ketoacidosis and requirement for insulin [17]. The recruited cases for the control subjects were selected from the patients who had visited the hospital for a general health examination. All subjects were clinically euthyroid and had no history of DM. Both the patient and the control groups had no known previous thyroid disease, and participants were not under any medication that could affect thyroid function tests. All patients with type 1 DM were under intensive therapy at the time that they were included in the study. The total insulin dose was calculated as regular insulin, and duration of diabetes was recorded as years starting from the date of diagnosis. Patients with a smoking history had been actively smoking for at least 3 years. Height and body weight were measured in all participants, and their body mass index [weight (kg)/ square of height (m 2 )] was calculated. Venous blood samples were taken to determine sensitive thyrotropin (TSH), thyroid hormones [free triiodothyronine (ft3), free thyroxine (ft4)], anti-tpo, anti-tg, anti-gad, and glycosylated hemoglobin (HbA1c) after a 12 hour fast. Anti-TPO, anti-tg and anti-gad levels above the normal ranges were defined as positive, whereas the levels at the normal range were defined as negative. TSH values below or above the normal range were defined as abnormal Biochemical measurements Thyroid specific antibodies (Anti-TG and Anti-TPO) were measured by, electrochemiluminescence immunoassay (ECLIA) (Elecsys 2010, Roche Diagnostics, USA). The normal ranges were IU/mL and 0 34 IU/mL for Anti-TG and Anti-TPO, respectively. Glutamic acid decarboxylase antibodies (normal range: 0 1 IU/mL) were measured by immunoradiometric assay (IRMA) (Cis Bio International). Serum ft4 (normal range: ng/dl), serum ft3 (normal range: pg/ml), and serum TSH (normal range: uiu/ ml) were measured by Chemiluminescent Microparticle

3 K. Gul et al. Table 1. Characteristic data of the patient and control groups. Results are presented as median (range). Diabetics (n=104) Controls (n=58) P Age (years) 26 (16-50) 27 (15-46) 0.08 BMI (kg/m 2 ) ( ) ( ) 0.23 Gender (F/M) 48/56 27/ Smoking (%) Thyroid volume (ml) (Right) 6.98 ( ) 6.08 ( ) Thyroid volume (ml) (Left) 7.01 ( ) 5.35 ( ) Total Thyroid volume (ml) ( ) ( ) TSH (µiu/ml) ( ) 1.51 ( ) FT3 (pg/ml) 2.42 ( ) 2.88 ( ) <0.001 FT4 (ng/dl) 1.05 ( ) 1.02 ( ) Anti-GAD (U/ml) 0.27 (0,01-258,81) ( ) <0.001 Anti-TPO (IU/ml) ( ) 5.80 ( ) Anti-TG (IU/ml) 16.55( ) ( ) Immunoassay (CMIA) (Architect i2000sr, Abbott Diagnostics). HbA1c (normal range 4% to 6%) was determined by immunoturbidimetric assay (C800, Abbott Diagnostics, USA) Thyroid US The thyroid volume was calculated on the basis of an ultrasound (US) scanning procedure, using a 12 MHz compound scanner (General Electric LOQÝC 400, USA). The investigator (T.D.) performed all the US examinations blinded to the knowledge of the group to which the subjects belonged (DM or control). Intraobserver variation was found to be 7.8%. The different echo patterns observed on US were divided into four groups: (0) normal thyroid gland; (I) mild diffuse hypoechogenicity; (II) moderate diffuse hypoechogenicity; and (III) non-homogeneous hypoechogenicity (heterogeneous). The nodular patterns found by US were classified as three groups: I) Normal II) Uninodular, and III) Multinodular Statistical Analysis Data analysis was performed by the SPSS for Windows Descriptive statistics were presented as median (min-max) for continuous variables. Categorical comparisons were made using the Chi-square or Fisher s exact test, where appropriate. Continuous variables were analyzed using the Mann-Whitney U test. Spearman rho statistics were used to determine the degree of associations between variables. A p value less than 0.05 was considered statistically significant. 3. Results There was no statistically significant difference between patient and control groups in terms of age, BMI, smoking, and gender (p>0.05). In the patient group, median duration of diabetes was 6 years (0 33), median value of the total insulin dose was 54 U/day (18 135), and median HbA1c levels were 10.4% ( ). While there was no difference between groups in terms of TSH and ft4 levels, ft3 levels were statistically significantly higher in the control group. There were statistically significant differences between groups in terms of anti-gad and anti-tpo, whereas anti-tg showed no statistically significant difference. The median value of total thyroid volumes were ml (6.5 32) and ml ( ) in the patient and control groups, respectively; the difference was statistically significant (p=0.002). Characteristics of the patient and control groups are shown in Table 1. In the patient group, the total thyroid volume was not correlated with anti-tpo, anti-tg, anti-gad, TSH, BMI, or gender. In the control group, total thyroid volume showed a correlation with TSH (p=0.001, r=0.41), whereas no correlation was established with anti-tpo and anti-tg. When the analysis was performed by gender in control group, total thyroid volume was higher in males as compared with females; the difference was statistically significant (p=0.04). Thyroid parenchyma echogenity was normal in 62.5% (n=65) of study patients and 91.4% (n=53) of the control group. There was statistically significant difference between groups in terms of parenchyma echogenicity (χ 2 =15.697, p<0.001). There was no significant difference between the two groups in terms of number of nodules (p=0.375).

4 Table 2. Ratios of positive TA in patient and control groups. Diabetics (n=104) Control (n=58) p Anti-TPO positive (%) 18.3 (n=19) 6.9 (n=4) Anti-TG positive (%) 7.7 (n=8) 10.3 (n=6) Anti-TPO and/or Anti-TG positive (%) 20.2 (n=21) 10.3 (n=6) Anti-TPO & Anti-TG positive (%) 5.8 (n=6) 6.9 (n=4) Table 3. Ratios of positive TA in patients with normal and abnormal TSH values. Anti-TPO Anti-TG Anti-TPO & Anti-TG Anti-TPO and/or Anti-TG Normal TSH (%) Abnormal TSH (%) Presence of nodules was not correlated with anti-tpo, anti-tg, or TSH in either group. In the patient group, 37.5% (n=39) had an abnormal morphology on US. The rate of abnormal findings were as follows; 15.4% (n=16) with mild and 17.3% (n=18) with moderately diffuse hypoechogenicity, while 4.8% (n=5) showed non-homogeneous hypoechogenicity. The most frequent abnormality was mild to moderate diffuse hypoechogenicity, found in 32.7% (n=34) of the patient group. Anti-TPO positivity in patients and control group was 18.3% and 6.9%, respectively (p=0.047). Anti-TPO and/or Anti-TG positivity in the patients and control group were 20.2% and 10.3%, respectively; this was statistically insignificant (p=0.107). The ratios of TA in patient and control groups are presented in Table 2. In the patient group, 8 (14.3%) of 56 males and 13 (27.1%) of 48 females had high anti-tpo and/or anti-tg. The difference was not statistical significant (p=0.08). The control group also showed no correlation between gender and TA. In both groups, TA showed no correlation with age and BMI (p>0.05). Further, TA was not correlated with duration of diabetes and total daily insulin dose (p>0.05). In the patient group, 14 (13.5%) of 104 cases had abnormal TSH values. Seven patients had subclinical hypothyroidism, 2 had Graves disease, and 5 had subclinical hyperthyroidism. In these 14 patients having abnormal TSH values, 12 had concomitant morphologic abnormalities on thyroid US (4 patients had nonhomogeneous hypoechogenicity, 7 had moderate diffuse hypoechogenicity, and 1 had mild diffuse hypoechogenicity), whereas the findings of 2 patients with subclinical hyperthyroidism were within the normal limits. In the control group, 2 of 58 patients had abnormal TSH values (3.4%). In terms of abnormal TSH values, there was a statistically significant difference between the patient and control groups (p=0.04). Among the 14 patients who had abnormal TSH values, 5 of them had anti-tpo, 5 had anti-tg, 7 had anti-tpo and/or anti-tg, and 3 had anti-tpo & anti- TG positivity. Among the 90 patients who had normal TSH, 14 had anti-tpo, 3 had anti-tg, 14 had anti-tpo and/or anti-tg, and 3 had anti-tpo & anti-tg positivity. Differences in ratios of positive anti-tg, anti-tpo & anti-tg, and anti-tpo and/or anti-tg were statistically significant between patients with abnormal TSH and normal TSH values (p<0.001; p=0.007; p=0.003, respectively). However, anti-tpo positivity was similar in patients with both abnormal TSH values and normal TSH values (p=0.07) (Table 3). In patients with Type 1 DM, anti-gad positivity was detected in 30.8% (n=32), but anti-gad was negative in 69.2% of the patients (n=72). Thirty-two of 104 patients, 16 (28.6%) male and 16 (33.3%) female, were anti- GAD positive (p=0.38). Diabetic patients were divided into two groups in terms of disease duration: one group having disease duration of less than one year (n=23) and the other group, more than one year (n=81). Anti- GAD positivity was present in 43.5% (10 of 23 patients) of patients whose duration of diabetes was less than one year, whereas it was positive in 27.2% (22 of 81 patients) of patients whose duration of diabetes was more than one year; the difference between these groups was not statistically significant (p=0.14). Among the 32 patients whose anti-gad antibodies were positive, 12 had anti-tpo (37.5%), 6 had anti-tg (18.8%), 13 had anti-tpo and/or anti-tg (40.6%), and 5 had anti-tpo & anti-tg (15.6%) positivity. Among the 72 patients whose anti-gad antibodies were negative, 7 had anti-tpo (9.7%), 2 had anti-tg (2.8%), 8 had anti- TPO and/or anti-tg (11.1%), and 1 had anti-tpo & anti- TG (1.4%) positivity. There was a statistically significant difference between the anti-gad positive and anti- GAD negative patients in terms of anti-tpo, anti-tg, anti-tpo & anti-tg, anti-tpo and/or anti-tg positivity (p=0.001, p=0.005, p=0.004, p=0.001, respectively (Figure 1). Thirty-nine patients had thyroid parenchyma abnormality detected by thyroid ultrasonography; in

5 K. Gul et al. Figure 1. Percentage of TA in anti-gad positive and negative patients with type 1 DM. Percentage (%) terms of thyroid autoimmunity, 9 patients had anti-tpo (23.07%), 5 patients had anti-tg (12.8%), 4 patients had anti-tpo & anti-tg (10.26%), 10 patients had anti-tpo and/or anti-tg (25.64%) positivity, and 12 patients had abnormal TSH values (30.8%). Thyroid parenchyma was normal in 65 patients; among these patients, 10 had anti- TPO (15.4%), 3 had anti-tg (4.6%), 2 had anti-tpo & anti-tg (3.1%), 11 had anti-tpo and/or anti-tg (16.9%) positivity and 2 of them had abnormal TSH values (3.1%). The association of morphological abnormalities on US and positive anti-tg, anti-tpo & anti-tg were statistically significant (p=0.01; p=0.01), while positive anti-tpo, anti-tpo and/or anti-tg showed no statistical significance (p=0.12; p=0.96). The association between morphological abnormalities on US and abnormal TSH values were statistically significant (p<0.001). 4. Discussion Anti-TPO Anti-TG Anti-TPO& TG Anti-TPO and/or Anti- TG Anti-GAD negative Anti-GAD positive Statistically significant differences are pointed out respectively; Anti TPO (χ 2 =11.448, p<0.001), Anti TG (p=0.010) (Fisher s Exact test), Anti TPO & TG (p=0.010) (Fisher s Exact test), Anti TPO and/or Anti TG (χ 2 =11.975, p<0.001). Type 1 DM is often related to other autoimmune diseases, but the most frequent concomitant disease is autoimmune thyroiditis. Several studies have confirmed the higher prevalence of autoimmune thyroiditis in patients with type 1 diabetes, when compared with the non-diabetic population [18,19]. In the study of Völzke et al., diabetic patients had higher levels of anti- TPO and higher proportions of positive anti-tpo than non-diabetic individuals. In that study, ft3 levels were lower in diabetic patients [20]. In the present study, there was a statistically significant difference between patient and control groups with regard to anti-tpo and ft3 levels. Low ft3 levels, which were determined in particular in poorly controlled diabetic patients, could result from euthyroid sick syndrome. The frequency of thyroid-related autoantibodies (anti-tpo and/or anti- TG) was reported as 21.6% in a multicenter study from Germany [21]. In the present study, the rate of positive anti-tpo and/or anti-tg was 20.2% and 10.3% in the patient and control groups, respectively. In the study by Premawardhana et al., patient and control groups had an anti-tg positivity of 25% and 33%, respectively, and for anti-tpo, these rates were as 22% and 18%, respectively; there was no statistical difference between these groups [22]. Previous studies have reported that TA is more frequent among girls than boys [21,23], while the present authors and others [19,24] have not found such a difference. In type 1 diabetic patients, prevalence of abnormal TSH was reported in various proportions. In a study of Karaguzel et al., subclinical hyper- and/or hypothyroidism proportion was 5.3% in type 1 diabetic patients [2]. In another study including 58 patients with type 1 diabetes, this rate was 33% [25]. In the present study, the rate of abnormal TSH was 13.5%, and TA showed no difference in terms of gender distribution. However, 71.4% of the patients with abnormal TSH values were female. Possible explanations for this difference could be ethnic variations, age, difference in iodine intake, and a more sensitive assay in the latter study, or alternatively, immunological abnormalities might be presented at the time of diagnosis of type 1 DM before the initiation of insulin treatment. Some of the studies showed that anti-tpo positivity was more sensitive than anti-tg in terms of indicating abnormal TSH values [25,26]. On the other hand, Hansen et al. showed that both anti-tg and anti-tpo indicate abnormal TSH values in similar proportions, but if anti-tpo had been used alone, 24% of patients might have been missed [19]. In the present study, when patients with abnormal TSH and normal TSH values were compared according to their TA positivity rates, there was no significant difference in anti-tpo positivity, whereas anti-tg; anti-tpo and/or anti-tg; and anti- TPO & anti-tg positivity were statistically significantly higher in the abnormal TSH group. Thus, measuring anti- TPO only would give incorrect information. According to this result, anti-tpo and anti-tg should be evaluated collectively. Anti-GAD frequency in type 1 DM has varied from 25% to 70%, and it has been reported more frequently in females than males [27-29]. In our study, anti-gad frequency was 30.8%, and there was no statistical difference between females and males. It is known that anti-gad is positive in more than 70% of children with recent-onset type 1 diabetes, and that its level seems to decrease with the duration of the disease and decreasing number of residual beta cells [30,31]. In our study, an increased anti-gad positivity was observed in patients

6 whose duration of diabetes was less than one year compared with that of those whose duration of diabetes was longer than one year; but the difference between two groups was statistically insignificant. Consistent with our results, Chang and his colleagues determined no relations between anti-gad positivity and duration of diabetes and gender [10]. In view of the association between anti-gad and type 1 DM with thyroid autoimmunity, when anti-gad positive type 1 diabetic patients of the present study were compared with anti-gad negative diabetic patients, the prevalence of positive TA (anti-tpo; anti- TG; anti-tpo & anti-tg; anti-tpo and/or anti-tg) was statistically significant. Shiau et al. showed a significantly higher frequency of anti-tpo/anti-tg among anti-gad positive than anti-gad negative type 1 DM patients [32]. Kawasaki et al. demonstrated that high levels of anti-gad were present in type 1 diabetic patients with ATD [14]. These observations are in agreement with the studies of Marino et al [27] and Rattarasarn et al [33]; there is a suggestion that this may represent a subgroup associated with polyendocrine autoimmunity. We have concluded that the detection of positive anti-gad in diabetic patients could be considered a predictive factor for progression of thyroid autoimmunity. In the study by Chang et al. on type 1 diabetics in Taiwan, it was demonstrated that the presence of positive anti-tpo in the presence of anti-gad was 23.2%. This difference was not statistically significant when compared with the 22% positivity among cases negative for anti-gad [10]. Premawardhana et al. did not find any statistical difference between anti-gad65/anti-ia-2 positive and negative type 1 DM patients in terms of TA [22]. This may reflect ethnic differences, difference in iodine intake, or genetic heterogeneity in the role of TA and anti-gad autoimmunity in type 1 DM with thyroid autoimmunity. Thyroid US is often used in the diagnosis of ATD. In diabetics, a high prevalence of morphological abnormalities by thyroid US has been demonstrated. In their study, Hansen et al. showed that in patients with type 1 diabetes, prevalence of abnormal morphology on US was 42% [19]. These findings illustrated that most prominent US pathology in type 1 DM were thyroid parenchyma hypoechogenicity compatible with ATD [9]. In the present study, morphological abnormalities were detected by US in 37.5% of patients with type 1 DM patients, with the most common finding being mild and moderate diffuse hypoechogenicity, in accordance with the literature (33.7%). This finding is not supported by Darendeliler et al., whose study showed US abnormalities in only 2 of 83 young diabetics [34]. However, in some studies, US hypoechogenicity is a valuable prognostic marker in ATD, predicting the development of hypothyroidism [7]. In our study, all patients with hypothyroidism had morphological abnormalities; however, the groups showed no statistically significant difference in terms of thyroid nodules. Hansen et al. could not demonstrate a relation between US hypoechogenicity and TA [19]. On the other hand, when the associations of morphological abnormalities on US and TA in the present study were analyzed, it was revealed that, although positive anti- TG and positive anti-tg & anti-tpo were statistically significant, positive anti-tpo alone carried no statistical significant association with US abnormality. This finding could be another indicator of positive anti-tpo that suggests the presence of ATD together with anti-tg. Furthermore, when the proportions of abnormal TSH in patients with morphological abnormality in US (33.8%) and normal morphology (3.1%) were compared, the difference was statistically significant p<0.001). Thus, morphological abnormality on US might predict abnormal TSH values. Some studies have demonstrated that, in diabetics, thyroid volume is larger than control cases [35,36]. On the other hand, other studies haven t detected any statistically significant difference between type 1 diabetic patients and controls in terms of thyroid volume [19,20]. In a study including 45 patients with type 1 DM, although thyroid volume was larger in diabetics than control cases, thyroid volume was not related with TSH levels and antithyroid antibodies [37]. In the present study, total thyroid volume in type 1 DM patients was also larger than control cases, in agreement with the literature (p=0.002), and total thyroid volume was not correlated with anti-tpo, anti-tg, anti-gad, or TSH. This increase in thyroid volume could be the consequence of diabetes-related autoimmunity. 5. Conclusion The present study showed that the rates of TA and abnormal TSH values and morphological abnormalities found by US are increased in patients with type 1 diabetes. These data support the recommendation for regular examinations of thyroid antibodies and thyroid function assessments in these patients. The TA of patients with positive anti-gad was apparently higher when compared with the patients with negative anti- GAD. According to this result, positive anti-gad could predict a higher risk for thyroid dysfunction. Thus, patients with positive anti-gad should be followed more frequently and more carefully.

7 K. Gul et al. References [1] González GC., Capel I., Rodríguez-Espinosa J., Mauricio D., de Leiva A., Pérez A., Thyroid autoimmunity at onset of type 1 diabetes as a predictor of thyroid dysfunction, Diabetes Care, 2007, 30, [2] Karagüzel G., Simşek S., Değer O., Okten A., Screening of diabetes, thyroid, and celiac diseasesrelated autoantibodies in a sample of Turkish children with type 1 diabetes and their siblings, Diabetes Res Clin Pract., 2008, 80, [3] Mantovani RM., Mantovani LM., Dias VM., Thyroid autoimmunity in children and adolescents with type 1 diabetes mellitus: prevalence and risk factors, J Pediatr Endocrinol Metab., 2007, 20, [4] Libman IM., Sun K., Foley TP., Becker DJ., Thyroid autoimmunity in children with features of both type 1 and type 2 diabetes, Pediatr Diabetes, 2008, 9, [5] Levin L., Tomer Y. The etiology of autoimmune diabetes and thyroiditis: evidence for common genetic susceptibility, Autoimmun Rev., 2003, 2, [6] Bárová H., Perusicová J., Hill M., Sterzl I., Vondra K., Masek Z., Anti-GAD-positive patients with type 1 diabetes mellitus have higher prevalence of autoimmune thyroiditis than anti-gad-negative patients with type 1 and type 2 diabetes mellitus, Physiol Res., 2004, 53, [7] Sostre S., Reyes MM., Sonographic diagnosis and grading of Hashimoto's thyroiditis, J Endocrinol Invest., 1991, 14, [8] D. Thyroid Autoantibodies (TPOAb, TgAb and TRAb), Thyroid, January 2003,13, [9] Marcocci C., Vitti P., Cetani F., Catalano F., Concetti R., Pinchera A., Thyroid ultrasonography helps to identify patients with diffuse lymphocytic thyroiditis who are prone to develop hypothyroidism, J Clin Endocrinol Metab., 1991, 72, [10] Chang CC., Huang CN., Chuang LM., Autoantibodies to thyroid peroxidase in patients with type 1 diabetes in Taiwan, Eur J Endocrinol., 1998, 139, [11] McCanlies E., O'Leary LA., Foley TP., Kramer MK., Burke JP., Libman A., et al., Hashimoto's thyroiditis and insulin-dependent diabetes mellitus: differences among individuals with and without abnormal thyroid function, J Clin Endocrinol Metab., 1998, 83, [12] A.W. Norris., J.I. Wolfsdorf., Diabetes mellitus, in: C. Brook, P. Clayton, R. Brown (eds): Brook s Clinical Pediatric Endocrinology, 5th ed., Blackwell Publishing, Oxford, 2005 [13] Yamaguchi Y., Chikuba N., Ueda Y., Yamamoto H., Yamasaki H,. Nakanishi T., et al., Islet cell antibodies in patients with autoimmune thyroid disease, Diabetes, 1991, 40, [14] Kawasaki E., Takino H., Yano M., Uotani S., Matsumoto K., Takao Y., et al., Autoantibodies to glutamic acid decarboxylase in patients with IDDM and autoimmune thyroid disease, Diabetes, 1994, 43, [15] Verge CF., Howard NJ., Rowley MJ., Mackay IR., Zimmet PZ., Egan M., et al., Anti-glutamate decarboxylase and other antibodies at the onset of childhood IDDM: a population-based study, Diabetologia, 1994, 37, [16] Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. National Diabetes Data Group, Diabetes, 1979, 28, [17] Chuang LM., Jou TS., Hu CY., Wu HP., Tsai WY., Lee JS., et al., HLA-DQB1 codon 57 and IDDM in Chinese living in Taiwan, Diabetes Care, 1994, 17, [18] Jaeger C., Hatziagelaki E., Petzoldt R., Bretzel RG., Comparative analysis of organ-specific autoantibodies and celiac disease-associated antibodies in type 1 diabetic patients, their firstdegree relatives, and healthy control subjects, Diabetes Care, 2001, 24, [19] Hansen D., Bennedbaek FN., Hansen LK., Hoier- Madsen M., Jacobsen BB., Hegedüs L., Thyroid function, morphology and autoimmunity in young patients with insulin-dependent diabetes mellitus, Eur J Endocrinol., 1999, 140, [20] Völzke H., Krohn U., Wallaschofski H., Lüdemann J., John U., Kerner W., The spectrum of thyroid disorders in adult type 1 diabetes mellitus, Diabetes Metab Res Rev., 2007, 23, [21] Kordonouri O., Klinghammer A., Lang EB., Grüters-Kieslich A., Grabert M., Holl RW., Thyroid autoimmunity in children and adolescents with type 1 diabetes: a multicenter survey, Diabetes Care, 2002, 25, [22] Premawardhana LD., Wijeyaratne CN., Chen S., Wijesuriya M., Illangasekera U., Brooking H., et al., Islet cell, thyroid, adrenal and celiac disease related autoantibodies in patients with Type 1 diabetes from Sri Lanka, J Endocrinol Invest., 2006, 29, [23] Fröhlich-Reiterer EE., Hofer S., Kaspers S., Herbst A., Kordonouri O., Schwarz HP., et al., on behalf of the DPV-Wiss Study Group. Screening frequency

8 for celiac disease and autoimmune thyroiditis in children and adolescents with type 1 diabetes mellitus - data from a German/Austrian multicentre survey, Pediatr Diabetes, 2008, 9, [24] Lindberg B., Ericsson UB., Ljung R., Ivarsson SA., High prevalence of thyroid autoantibodies at diagnosis of insulin-dependent diabetes mellitus in Swedish children, J Lab Clin Med., 1997, 130, [25] Umpierrez GE., Latif KA., Murphy MB., Lambeth HC., Stentz F., Bush., A et al., Thyroid dysfunction in patients with type 1 diabetes: a longitudinal study, Diabetes Care, 2003, 26, [26] Kordonouri O., Hartmann R., Deiss D., Wilms M., Grüters-Kieslich A., Natural course of autoimmune thyroiditis in type 1 diabetes: association with gender, age, diabetes duration, and puberty, Arch Dis Child., 2005, 90, [27] Martino GV., Tappaz ML., Braghi S., Dozio N., Canal N., Pozza., G et al., Autoantibodies to glutamic acid decarboxylase (GAD) detected by an immuno-trapping enzyme activity assay: relation to insulin-dependent diabetes mellitus and islet cell antibodies, J Autoimmun., 1991, 4, [28] DeAizpurua HJ., Harrison LC., Cram DS., An ELISA for antibodies to recombinant glutamic acid decarboxylase in IDDM, Diabetes, 1992, 41, [29] Thivolet CH., Tappaz M., Durand A., Petersen J., Stefanutti A., Chatelain P., et al., Glutamic acid decarboxylase (GAD) autoantibodies are additional predictive markers of type 1 (insulin-dependent) diabetes mellitus in high risk individuals, Diabetologia, 1992, 35, [30] Bingley PJ., Bonifacio E., Williams AJ., Genovese S., Bottazzo GF., Gale EA., Prediction of IDDM in the general population: strategies based on combinations of autoantibody markers, Diabetes, 1997, 46, [31] Fajardo C., Piñón F., Carmona E., Sánchez-Cuenca JM., Merino JF., Carlés C., Influence of age on clinical and immunological characteristics of newly diagnosed type 1 diabetic patients, Acta Diabetol., 2001, 38, [32] Shiau MY., Tsai ST., Hwang J., Wu CY., Chang YH., Relationship between autoantibodies against glutamic acid decarboxylase, thyroglobulin/thyroid microsome and DNA topoisomerase II in the clinical manifestation of patients with type 1 diabetes mellitus in Taiwan, Eur J Endocrinol., 2000, 142, [33] Rattarasarn C., Diosdado MA., Ortego J., Leelawattana R., Soonthornpun S., Setasuban W., et al., Thyroid autoantibodies in Thai type 1 diabetic patients: clinical significance and their relationship with glutamic acid decarboxylase antibodies, Diabetes Res Clin Pract., 2000, 49, [34] Darendeliler FF., Kadioğlu A., Bas F., Bundak R., Günöz H., Saka N., et al., Thyroid ultrasound in IDDM, J Pediatr Endocrinol., 1994, 7, [35] Okten A., Akcay S., Cakir M., Girisken I., Kosucu P., Deger O., Iodine status, thyroid function, thyroid volume and thyroid autoimmunity in patients with type 1 diabetes mellitus in an iodine-replete area, Diabetes Metab., 2006, 32, [36] Junik R., Kozinski M., Debska-Kozinska K., Thyroid ultrasound in diabetic patients without overt thyroid disease, Acta Radiol., 2006, 47, [37] Bianchi G., Montanari P., Fabbri A., Gamberini A., Zoli M., Marchesini G., Thyroid volume in type 1 diabetes patients without overt thyroid disease, Acta Diabetol., 1995, 32, 49-52