- Email: [email protected]
Thyroid autoantibodies in Thai type 1 diabetic patients: clinical significance and their relationship with glutamic acid decarboxylase antibodies
Diabetes Research and Clinical Practice 49 (2000) 107 – 111 www.elsevier.com/locate/diabres
Thyroid autoantibodies in Thai type 1 diabetic patients: clinical significance and their relationship with glutamic acid decarboxylase antibodies Chatchalit Rattarasarn a,*, Manuel Aguilar Diosdado b, Jose Ortego b, Rattana Leelawattana a, Supamai Soonthornpun a, Worawong Setasuban a, Somchit Jaruratanasirikul c, Narumon Patarakijvanich c a
Di6ision of Endocrinology and Metabolism, Department of Medicine, Prince of Songkla Uni6ersity, Songkhla 90110, Thailand b Seccion de Endocrinologia, Hospital Uni6ersitario Puerta del Mar, Cardiz, Spain c Department of Pediatrics, Prince of Songkla Uni6ersity, Songkhla, Thailand Received 6 September 1999; received in revised form 6 December 1999; accepted 14 March 2000
Abstract Objecti6e: To study the clinical significance of thyroid autoantibodies in Thai patients with type 1 diabetes and their relationship with glutamic acid decarboxylase antibodies (GAD65Ab). Methods: Thyroglobulin antibodies (TG-Ab) and thyroid peroxidase antibodies (TPO-Ab) were measured in 50 Thai type 1 diabetic patients. Forty-four patients also had GAD65Ab measured. Serum thyrotropin (TSH) was measured in all patients who had no history of thyroid disease regardless of thyroid antibody status. Clinical data including sex, age at onset and duration of diabetes, family history of diabetes, fasting c-peptide levels as well as frequencies of GAD65Ab were compared between patients with and without thyroid antibodies. GAD65Ab was also measured in 29 non-diabetic patients with hyperthyroid Graves’ disease or Hashimoto thyroiditis as a control group. Results: TG-Ab and TPO-Ab were positive in nine (18%) and 15 (30%) patients, respectively. Eight patients (16%) were positive for both antibodies. Two of 16 patients who were positive for TG-Ab or TPO-Ab had a previous history of hyperthyroidism prior to diabetes onset. Of the remainder, two were newly diagnosed with hyperthyroidism and one was found to have clinical hypothyroidism at the time of the study. None of 34 patients without thyroid antibodies had thyroid dysfunction. Eight patients with positive thyroid antibodies but without clinical thyroid dysfunction and 21 patients without thyroid antibodies were followed for up to 3 years, two patients of the first group developed hypothyroidism, whereas none of the latter developed thyroid dysfunction. The frequency of thyroid dysfunction at the time of initial study was significantly higher in patients with positive thyroid antibodies (3/14 vs. 0/34; P =0.021) and these patients who were initially euthyroid tended to have a higher risk of developing thyroid dysfunction (2/8 vs. 0/21; P= 0.069). The frequency of thyroid antibodies was significantly increased in females and in those who had positive GAD65Ab. GAD65Ab was negative in all of the non-diabetic patients with autoimmune thyroid disease. Conclusions: About one-fourth of Thai patients
* Corresponding author. Tel.: + 66-74-212070, ext. 1463; fax: + 66-74-429385. E-mail address: [email protected] (C. Rattarasarn). 0168-8227/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 0 0 ) 0 0 1 4 6 - 7
108
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
with type 1 diabetes without thyroid disease had thyroid antibodies. The frequency of thyroid antibodies was increased in female and in GAD65Ab positive patients. The presence of thyroid antibodies is associated with a higher frequency of and may predict a higher risk for thyroid dysfunction in Thai type 1 diabetic patients. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Thyroid antibody; Thyroid dysfunction; Glutamic acid decarboxylase antibody; Type 1 diabetes
1. Introduction Autoimmune thyroid disease is the most common autoimmune disorder associated with type 1 diabetes. The frequencies of thyroid autoantibodies are increased in type 1 diabetic patients with or without thyroid disorders and the presence of these autoantibodies in the latter group may predict the future development of thyroid dysfunction [1–4]. The close association between thyroid autoantibodies, particularly thyroid microsomal or thyroid peroxidase antibodies (TPO-Ab), and autoantibodies to islet antigens have been reported in type 1 diabetic patients [5 – 8]. Patients who were positive for glutamic acid decarboxylase antibodies (GAD65Ab) were more likely to be positive for TPO-Ab than those without. Although such positive association has been recently reported from several studies in Caucasian and Japanese populations, it could not be confirmed in some other studies [9 – 12]. For example, Chang et al. [12] found no association between GAD65Ab and TPO-Ab in Chinese type 1 diabetic children in Taiwan. This discrepancy is probably due to a lower prevalence of GAD65Ab in the Chinese population. Most of the studies of thyroid antibodies in type 1 diabetic patients are from Caucasian populations. Whether this can be extrapolated to Asian populations whose incidence of type 1 diabetes is much lower is uncertain. Although the frequencies of thyroid antibodies in Asian type 1 diabetes appear to be similar to those of Caucasians [11 – 13], their clinical significance has not been determined. In this study we examined the clinical significance of thyroid autoantibodies with regards to their role in prediction of thyroid dysfunction in Thai type 1 diabetic patients. We also examined the relationship between the presence of thyroid antibodies and clinical features of type 1
diabetes as well as their relationship with GAD65Ab.
2. Subjects and methods Fifty Thai patients with type 1 diabetes, 19 female and 31 male, who attended the diabetic clinic of Prince of Songkla University Hospital were studied. Their age at onset and duration of diabetes were 36.59 17.5 (S.D.) and 5.29 4.1 years, respectively. Seven patients were aged 5 15 years at onset of diabetes (childhood-onset), the remainder were aged \15 years at onset (adultonset). All had a history of ketonuria or ketoacidosis at onset or a history of primary or secondary failure to oral hypoglycemic agents within 3 years. Their fasting c-peptide levels were B 0.33 nmol/l and all were being treated with insulin at the time of study. Sera were obtained from all patients for the measurement of thyroglobulin antibodies (TG-Ab) and TPO-Ab. GAD65Ab was measured in 44 of 50 patients. Serum thyrotropin (TSH) was measured in all patients who had no previous history of thyroid disease regardless of thyroid antibody status. Clinical data of all patients which included sex, age at onset of diabetes, duration of diabetes, family history of diabetes and thyroid diseases, as well as a history of thyroid dysfunction, were obtained by reviewing the medical records and/or direct patient interview. GAD65Ab was also measured in 29 non-diabetic patients with a diagnosis of hyperthyroid Graves’ disease (defined as having diffuse goiter and exophthalmos with or without thyroid autoantibodies) or Hashimoto thyroiditis (defined as having goiter or a history of primary hypothyroidism and high titers of thyroid autoantibodies) as a control group.
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
GAD65Ab was measured by RIA (CIS, France) using recombinant human GAD65 as the antigen. For the assay, the test samples were incubated with 125I-labelled human recombinant GAD65, followed by addition of solid phase protein A to precipitate the labelled GAD65 – GAD65 antibody complexes. After centrifugation, the precipitates were counted for 125I and the amount of radioactivity in the precipitates was proportional to the concentration of GAD65Ab in the test sample. The intra-assay CV of the test was 2.9 – 3.1% and the inter-assay CV was 3.6 – 5.1%. This assay has been demonstrated to have 100% sensitivity and specificity by the 12th International Workshop. Serum was considered positive for GAD65Ab if it contained \1. 5 U/ml which was above the 99th percentile of 103 healthy, blood donor volunteers. TG-Ab and TPO-Ab were measured by the agglutination method (Serodia, Japan); the respective titers of ] 1:10 and ]1:100 were considered positive. TSH was measured by immunoradiometric assay (Biogenetech, France); the normal range was 0.25 – 4.00 mU/l. The intraassay CV of the test was 2.2 – 5.3% and the interassay CV was 6.8–8.7%. C-peptide was measured by double antibody RIA (Diagnostic Product, USA) with intra-assay CV of 2.5 – 5% and interassay CV of 3.6–8.8%.
3. Statistical analysis The x 2-test with continuity correction or Fisher’s exact probability test were used, where appropriate, to test the significance of frequency difference. Significance of difference between means was tested with the Mann – Whitney test. P B 0.05 denoted a statistically significant difference.
4. Results Of 50 type 1 diabetic patients, nine (18%) were positive for TG-Ab, 15 (30%) were positive for TPO-Ab and 8 (16%) were positive for both antibodies. All except one patient who were positive for TG-Ab were also positive for TPO-Ab. The
109
titers of TG-Ab and TPO-Ab were ranged from 1:20–1:5, 120 and 1:100–1:409, 600, respectively. Two of 16 patients who had TG-Ab or TPO-Ab had a previous history of hyperthyroidism prior to diabetes onset. Of the remainder, two were newly diagnosed with hyperthyroidism, one was found to have clinical hypothyroidism with a marked elevation of serum TSH and one had a mild elevation of serum TSH 10 mU/l without hypothyroid symptoms at the time of the study. Otherwise, serum TSH were within the normal range including 34 patients who were negative for thyroid antibodies. Eight patients who had TG-Ab or TPO-Ab without obvious thyroid dysfunction were followed for 199 8 (S.D.) months (range 8–35), two developed hypothyroidism during follow-up. Serum TSH was increased from 10.0 to 20.75 mU/l after 20 months in one patient and from 4.64 to 33.87 mU/l after 35 months in the other patient. None of 21 patients who were negative for thyroid antibodies developed thyroid dysfunction during 16.49 6.3 months (range 6–29) of follow-up. Patients with thyroid antibodies but without a previous history of thyroid diseases had a significantly higher frequency of thyroid dysfunction at the time of the initial study (3/14 vs. 0/34.1; P= 0.02 1) and tended to have a higher risk of developing thyroid dysfunction up to approximately 3 years of follow-up (2/8 vs. 0/21; P= 0.069) although the difference did not reach statistical significance. When patients were classified according to thyroid antibodies status as shown in Table 1, the antibodies were more frequently positive in female patients and in patients with positive GAD65Ab (P= 0.033 and 0.029, respectively). In non-diabetic patients with hyperthyroid Graves’ disease or Hashimoto thyroiditis, none were positive for GAD65Ab. Clinical characteristics which included age at onset of diabetes, duration of diabetes, family history of diabetes, as well as fasting c-peptide levels, were not different between patients with and without thyroid antibodies. The frequency of thyroid antibodies did not differ either between childhood (5 15 years) and adult-onset (\ 15 years) or between short (5 2 years) and long-duration (\ 2 years) of diabetes (data not shown).
110
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
5. Discussion The prevalence of thyroid antibodies in Asian type 1 diabetes has been previously reported [11 – 13]. This study differs from those studies in that most patients in this study had diabetes onset in adulthood, whereas patients in other studies had diabetes onset in childhood. Nevertheless, this study shows and confirms that the prevalence of thyroid antibodies in either childhood or adultonset type 1 diabetes Asian population is comparable to and not lower than those of Caucasians, as had been thought [2,14]. Similar to other studies [2,5,7,9,12], thyroid antibodies were more frequent in female than in male patients in this study. This could not be explained by a selection bias of the study given that the males outnumbered the females. The preponderance in females of thyroid antibodies is well recognized in both diabetics and non-diabetics. Why there is a sex discrepancy is unknown but may be related to the effects of sex steroids on the immunoregulatory process [15]. It is notable that all of our type 1 diabetic patients who had or developed thyroid dysfunction were female. In contrast to other studies from both Asian and Caucasian populations [2,11,16], the higher frequency of thyroid antibodies in patients with a longer duration of diabetes could not be demonstrated in this study. The association between thyroid autoantibodies and subsequent thyroid dysfunction in the general population was well established [16,17]. Several studies in Caucasian childhood type 1 diabetes
also demonstrated that patients who were positive for thyroid antibodies at diabetes onset more frequently had or developed thyroid dysfunction even after several years of diabetes. The frequencies of thyroid dysfunction in thyroid antibodies positive, newly diagnosed patients varied from 10 to 30%, the frequencies of which can increase up to 50% after several years of diabetes [2,4,6,18– 20]. Similar results in Caucasian adult-onset type 1 diabetes were recently reported by FernandezCastaner et al. [7]. However, the prevalence of thyroid dysfunction in Asian, particularly Chinese type 1 diabetes, appears to be lower than those of Caucasians. Tsai et al. reported that only two of 14 Chinese type 1 diabetic children with positive TPO-Ab developed thyroid dysfunction [13]. Our study, in contrast to the Tsai et al. study, shows that the prevalence of thyroid dysfunction in Thai type 1 diabetic patients corresponds to those of Caucasians. It is noticeable that all of the four patients with hyperthyroidism in this study developed the disease prior to or at the diagnosis of diabetes and three patients who developed hypothyroidism did so after 4, 8 and 9 years of diabetes. Our findings confirm the previous notion about thyroid dysfunction in childhood type 1 diabetes that hyperthyroidism commonly occurs at or prior to the diagnosis of diabetes, whereas hypothyroidism may occur at any time during the course of diabetes [21]. The positive association between GAD65Ab and thyroid antibodies in type 1 diabetes of this study is contrary to the study results of Chang et
Table 1 Clinical characteristics of diabetes and frequencies of GAD65Ab in type 1 diabetic patients with and without thyroid antibodiesa Characteristics
Thyroid Ab (+) (N =16)
Thyroid Ab (−) (N =34)
P
Female:male Age at onset of diabetes (years) Duration of diabetes (years) Family history of diabetes in first degree relative Fasting c-peptide (nmol/l)b
10:6 39.0 919.3 (range 10–73) 6.1 9 4.8 (range 0–20) 3/16 0.109 0.09 (range 0.00–0.32) 10/15
9:25 35.3 916.8 (range 9–76) 4.7 9 3.7 (range 0–14) 9/34 0.14 90.11 (range 0.00–0.33)
0.033 NS NS NS NS
GAD65Abc
8/29
Data are expressed as mean 9S.D. Thyroid Ab, thyroid antibodies; NS, not significant. N= 15 for thyroid Ab (+) and 32 for thyroid Ab (−). c N =15 for thyroid Ab (+) and 29 for thyroid Ab (−). a
b
0.029
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
al. [12] who reported no association between these antibodies in Chinese type 1 diabetic patients. Why the results of these two studies are at variance despite a similarity in frequencies of GAD65Ab and thyroid antibodies is unclear. The differences in age at onset of diabetes as well as the heterogeneity in genetic background may, in part, contribute to this discrepancy. In conclusion, a significant number of Thai patients with type 1 diabetes had thyroid antibodies particularly female patients and those who had positive GAD65Ab. The presence of thyroid antibodies is associated with a higher frequency of and may predict a higher risk for thyroid dysfunction. Therefore, thyroid antibodies should be screened in all type 1 diabetic patients particularly female and GAD65Ab positive patients and those with positive results should thyroid function tests be measured. Long-term follow-up is required in order to detect thyroid dysfunction in this group of patients. The prevalence of thyroid antibodies and thyroid dysfunction in Thai type 1 diabetic patients appear to be comparable to those of Caucasians. However, a population-based study with a larger number of patients should be performed to confirm our findings Acknowledgements We are grateful to Atchara Thamprasit for preparation of the manuscript and to Kalaya Leetanaporn for assisting in c-peptide analysis. References [1] C.L. Burek, N.R. Rose, K.E. Guire, et al., Thyroid autoantibodies in Black and in White children and adolescents with type 1 diabetes mellitus and their first degree relatives, Autoimmunity 7 (1990) 157–167. [2] R. Lorini, G. d’Annunzio, L. Vitah, S. Scaramuzza, IDDM and autoimmune thyroid disease in the pediatric age group, J. Pediatr. Endocrinol. Metab. 9 (1996) 89–94. [3] B. Lindberg, U.-B. Ericsson, R. Ljung, S.-A. Ivarsson, High prevalence of thyroid autoantibodies at diagnosis of insulin-dependent diabetes mellitus in Swedish children, J. Lab. Clin. Med. 130 (1997) 585–589. [4] M.J. McKenna, R. Herskowitz, J.J. Wolfisdorf, Screening for thyroid disease in children with IDDM, Diabetes Care 13 (1990) 801 – 803.
111
[5] L. Groop, A. Miettinen, P.-H. Groop, S. Meri, S. Koskirnies, G.F. Bottazzo, Organ-specific autoimmunity and HLA-DR antigens as markers for b-cell destruction in patients with type II diabetes, Diabetes 37 (1988) 99 – 103. [6] C.F. Verge, N.J. Howard, M.J. Rowley, et al., Anti-glutamate decarboxylase and other antibodies at the onset of childhood IDDM: a population-based study, Diabetologia 37 (1994) 1113 – 1120. [7] M. Fernandez-Castaner, A. Molina, L. Lopez-Jimenez, J.M. Gomez, J. Soler, Clinical presentation and early course of type 1 diabetes in patients with and without thyroid autoimmunity, Diabetes Care 22 (1999) 377 – 381. [8] E. Kawasaki, H. Takino, M. Yano, et al., Autoantibodies to glutamic acid decarboxylase in patients with IDDM and autoimmune thyroid disease, Diabetes 43 (1994) 80 – 86. [9] M. Neufeld, N.K. Maclaren, W.J. Riley, et al., Islet cell and other organ-specific antibodies in U.S. Caucasians and Blacks with insulin-dependent diabetes mellitus, Diabetes 29 (1980) 589 – 592. [10] The Diabetes Incidence Study in Sweden Group, M. Landin-Olsson, F.A. Karlsson, A. Lernmark, G. Sundkvist, Islet cell and thyrogastric antibodies in 633 consecutive 15- to 34-yr-old patients in the Diabetes Incidence Study in Sweden, Diabetes 41 (1992) 1022 – 1027. [11] A-C. Thai, W.-Y. Ng, K.-E. Lui, J.-S. Cheah, Islet cell and thyroid autoimmunity in Chinese patients with IDDM, Diabetes Care 18 (1995) 586 – 587. [12] C.-C. Chang, C.-N. Huang, L.-M. Chuang, Autoantibodies to thyroid peroxidase in patients with type 1 diabetes in Taiwan, Eur. J. Endocrinol. 139 (1998) 44 – 48. [13] W.-Y. Tsai, J.-S. Lee, Thyroid disease in Chinese children with IDDM, Diabetes Care 16 (1993) 1314 – 1315. [14] G.W.K. Wong, Absence of thyroid disease in Chinese children with IDDM, Diabetes Care 16 (1993) 404 – 405. [15] N.R. Farid, Immunogenetics of autoimmune thyroid disease, Endocrinol. Metab. Clin. North Am. 16 (1987) 229 – 245. [16] B.R. Hawkins, P.S. Cheah, R.L. Dawkins, et al., Diagnostic significance of thyroid microsomal antibodies in randomly selected population, Lancet ii (1980) 1057 – 1059. [17] M.P.J. Vanderpump, W.M.G. Tunbridge, J.M. French, et al., The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham Survey, Clin. Endocrinol. 43 (1995) 55 – 68. [18] G.M. Bright, R.M. Blizzard, D.L. Kaiser, W.L. Clarke, Organ-specific autoantibodies in children with common endocrine diseases, J. Pediatr. 100 (1982) 8 – 14. [19] W.J. Riley, N.K. Maclaren, D.C. Lezotte, R.P. Spillar, A.L. Rosenbloom, Thyroid autoimmunity in insulin-dependent diabetes mellitus: the case for routine screening, J. Pediatr. 98 (1981) 350 – 354. [20] L.M. Prina Cerai, G. Weber, F. Meschi, et al., Prevalence of thyroid autoantibodies and thyroid autoimmune disease in diabetic children and adolescents, Diabetes Care 17 (1994) 782 – 783. [21] G. Jefferson, The clinical approach to thyroid disorders associated with childhood insulin-dependent diabetes mellitus, J. Pediatr. Endocrinol. Metab. 9 (1996) 95 – 100.
Thyroid autoantibodies in Thai type 1 diabetic patients: clinical significance and their relationship with glutamic acid decarboxylase antibodies Chatchalit Rattarasarn a,*, Manuel Aguilar Diosdado b, Jose Ortego b, Rattana Leelawattana a, Supamai Soonthornpun a, Worawong Setasuban a, Somchit Jaruratanasirikul c, Narumon Patarakijvanich c a
Di6ision of Endocrinology and Metabolism, Department of Medicine, Prince of Songkla Uni6ersity, Songkhla 90110, Thailand b Seccion de Endocrinologia, Hospital Uni6ersitario Puerta del Mar, Cardiz, Spain c Department of Pediatrics, Prince of Songkla Uni6ersity, Songkhla, Thailand Received 6 September 1999; received in revised form 6 December 1999; accepted 14 March 2000
Abstract Objecti6e: To study the clinical significance of thyroid autoantibodies in Thai patients with type 1 diabetes and their relationship with glutamic acid decarboxylase antibodies (GAD65Ab). Methods: Thyroglobulin antibodies (TG-Ab) and thyroid peroxidase antibodies (TPO-Ab) were measured in 50 Thai type 1 diabetic patients. Forty-four patients also had GAD65Ab measured. Serum thyrotropin (TSH) was measured in all patients who had no history of thyroid disease regardless of thyroid antibody status. Clinical data including sex, age at onset and duration of diabetes, family history of diabetes, fasting c-peptide levels as well as frequencies of GAD65Ab were compared between patients with and without thyroid antibodies. GAD65Ab was also measured in 29 non-diabetic patients with hyperthyroid Graves’ disease or Hashimoto thyroiditis as a control group. Results: TG-Ab and TPO-Ab were positive in nine (18%) and 15 (30%) patients, respectively. Eight patients (16%) were positive for both antibodies. Two of 16 patients who were positive for TG-Ab or TPO-Ab had a previous history of hyperthyroidism prior to diabetes onset. Of the remainder, two were newly diagnosed with hyperthyroidism and one was found to have clinical hypothyroidism at the time of the study. None of 34 patients without thyroid antibodies had thyroid dysfunction. Eight patients with positive thyroid antibodies but without clinical thyroid dysfunction and 21 patients without thyroid antibodies were followed for up to 3 years, two patients of the first group developed hypothyroidism, whereas none of the latter developed thyroid dysfunction. The frequency of thyroid dysfunction at the time of initial study was significantly higher in patients with positive thyroid antibodies (3/14 vs. 0/34; P =0.021) and these patients who were initially euthyroid tended to have a higher risk of developing thyroid dysfunction (2/8 vs. 0/21; P= 0.069). The frequency of thyroid antibodies was significantly increased in females and in those who had positive GAD65Ab. GAD65Ab was negative in all of the non-diabetic patients with autoimmune thyroid disease. Conclusions: About one-fourth of Thai patients
* Corresponding author. Tel.: + 66-74-212070, ext. 1463; fax: + 66-74-429385. E-mail address: [email protected] (C. Rattarasarn). 0168-8227/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 0 0 ) 0 0 1 4 6 - 7
108
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
with type 1 diabetes without thyroid disease had thyroid antibodies. The frequency of thyroid antibodies was increased in female and in GAD65Ab positive patients. The presence of thyroid antibodies is associated with a higher frequency of and may predict a higher risk for thyroid dysfunction in Thai type 1 diabetic patients. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Thyroid antibody; Thyroid dysfunction; Glutamic acid decarboxylase antibody; Type 1 diabetes
1. Introduction Autoimmune thyroid disease is the most common autoimmune disorder associated with type 1 diabetes. The frequencies of thyroid autoantibodies are increased in type 1 diabetic patients with or without thyroid disorders and the presence of these autoantibodies in the latter group may predict the future development of thyroid dysfunction [1–4]. The close association between thyroid autoantibodies, particularly thyroid microsomal or thyroid peroxidase antibodies (TPO-Ab), and autoantibodies to islet antigens have been reported in type 1 diabetic patients [5 – 8]. Patients who were positive for glutamic acid decarboxylase antibodies (GAD65Ab) were more likely to be positive for TPO-Ab than those without. Although such positive association has been recently reported from several studies in Caucasian and Japanese populations, it could not be confirmed in some other studies [9 – 12]. For example, Chang et al. [12] found no association between GAD65Ab and TPO-Ab in Chinese type 1 diabetic children in Taiwan. This discrepancy is probably due to a lower prevalence of GAD65Ab in the Chinese population. Most of the studies of thyroid antibodies in type 1 diabetic patients are from Caucasian populations. Whether this can be extrapolated to Asian populations whose incidence of type 1 diabetes is much lower is uncertain. Although the frequencies of thyroid antibodies in Asian type 1 diabetes appear to be similar to those of Caucasians [11 – 13], their clinical significance has not been determined. In this study we examined the clinical significance of thyroid autoantibodies with regards to their role in prediction of thyroid dysfunction in Thai type 1 diabetic patients. We also examined the relationship between the presence of thyroid antibodies and clinical features of type 1
diabetes as well as their relationship with GAD65Ab.
2. Subjects and methods Fifty Thai patients with type 1 diabetes, 19 female and 31 male, who attended the diabetic clinic of Prince of Songkla University Hospital were studied. Their age at onset and duration of diabetes were 36.59 17.5 (S.D.) and 5.29 4.1 years, respectively. Seven patients were aged 5 15 years at onset of diabetes (childhood-onset), the remainder were aged \15 years at onset (adultonset). All had a history of ketonuria or ketoacidosis at onset or a history of primary or secondary failure to oral hypoglycemic agents within 3 years. Their fasting c-peptide levels were B 0.33 nmol/l and all were being treated with insulin at the time of study. Sera were obtained from all patients for the measurement of thyroglobulin antibodies (TG-Ab) and TPO-Ab. GAD65Ab was measured in 44 of 50 patients. Serum thyrotropin (TSH) was measured in all patients who had no previous history of thyroid disease regardless of thyroid antibody status. Clinical data of all patients which included sex, age at onset of diabetes, duration of diabetes, family history of diabetes and thyroid diseases, as well as a history of thyroid dysfunction, were obtained by reviewing the medical records and/or direct patient interview. GAD65Ab was also measured in 29 non-diabetic patients with a diagnosis of hyperthyroid Graves’ disease (defined as having diffuse goiter and exophthalmos with or without thyroid autoantibodies) or Hashimoto thyroiditis (defined as having goiter or a history of primary hypothyroidism and high titers of thyroid autoantibodies) as a control group.
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
GAD65Ab was measured by RIA (CIS, France) using recombinant human GAD65 as the antigen. For the assay, the test samples were incubated with 125I-labelled human recombinant GAD65, followed by addition of solid phase protein A to precipitate the labelled GAD65 – GAD65 antibody complexes. After centrifugation, the precipitates were counted for 125I and the amount of radioactivity in the precipitates was proportional to the concentration of GAD65Ab in the test sample. The intra-assay CV of the test was 2.9 – 3.1% and the inter-assay CV was 3.6 – 5.1%. This assay has been demonstrated to have 100% sensitivity and specificity by the 12th International Workshop. Serum was considered positive for GAD65Ab if it contained \1. 5 U/ml which was above the 99th percentile of 103 healthy, blood donor volunteers. TG-Ab and TPO-Ab were measured by the agglutination method (Serodia, Japan); the respective titers of ] 1:10 and ]1:100 were considered positive. TSH was measured by immunoradiometric assay (Biogenetech, France); the normal range was 0.25 – 4.00 mU/l. The intraassay CV of the test was 2.2 – 5.3% and the interassay CV was 6.8–8.7%. C-peptide was measured by double antibody RIA (Diagnostic Product, USA) with intra-assay CV of 2.5 – 5% and interassay CV of 3.6–8.8%.
3. Statistical analysis The x 2-test with continuity correction or Fisher’s exact probability test were used, where appropriate, to test the significance of frequency difference. Significance of difference between means was tested with the Mann – Whitney test. P B 0.05 denoted a statistically significant difference.
4. Results Of 50 type 1 diabetic patients, nine (18%) were positive for TG-Ab, 15 (30%) were positive for TPO-Ab and 8 (16%) were positive for both antibodies. All except one patient who were positive for TG-Ab were also positive for TPO-Ab. The
109
titers of TG-Ab and TPO-Ab were ranged from 1:20–1:5, 120 and 1:100–1:409, 600, respectively. Two of 16 patients who had TG-Ab or TPO-Ab had a previous history of hyperthyroidism prior to diabetes onset. Of the remainder, two were newly diagnosed with hyperthyroidism, one was found to have clinical hypothyroidism with a marked elevation of serum TSH and one had a mild elevation of serum TSH 10 mU/l without hypothyroid symptoms at the time of the study. Otherwise, serum TSH were within the normal range including 34 patients who were negative for thyroid antibodies. Eight patients who had TG-Ab or TPO-Ab without obvious thyroid dysfunction were followed for 199 8 (S.D.) months (range 8–35), two developed hypothyroidism during follow-up. Serum TSH was increased from 10.0 to 20.75 mU/l after 20 months in one patient and from 4.64 to 33.87 mU/l after 35 months in the other patient. None of 21 patients who were negative for thyroid antibodies developed thyroid dysfunction during 16.49 6.3 months (range 6–29) of follow-up. Patients with thyroid antibodies but without a previous history of thyroid diseases had a significantly higher frequency of thyroid dysfunction at the time of the initial study (3/14 vs. 0/34.1; P= 0.02 1) and tended to have a higher risk of developing thyroid dysfunction up to approximately 3 years of follow-up (2/8 vs. 0/21; P= 0.069) although the difference did not reach statistical significance. When patients were classified according to thyroid antibodies status as shown in Table 1, the antibodies were more frequently positive in female patients and in patients with positive GAD65Ab (P= 0.033 and 0.029, respectively). In non-diabetic patients with hyperthyroid Graves’ disease or Hashimoto thyroiditis, none were positive for GAD65Ab. Clinical characteristics which included age at onset of diabetes, duration of diabetes, family history of diabetes, as well as fasting c-peptide levels, were not different between patients with and without thyroid antibodies. The frequency of thyroid antibodies did not differ either between childhood (5 15 years) and adult-onset (\ 15 years) or between short (5 2 years) and long-duration (\ 2 years) of diabetes (data not shown).
110
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
5. Discussion The prevalence of thyroid antibodies in Asian type 1 diabetes has been previously reported [11 – 13]. This study differs from those studies in that most patients in this study had diabetes onset in adulthood, whereas patients in other studies had diabetes onset in childhood. Nevertheless, this study shows and confirms that the prevalence of thyroid antibodies in either childhood or adultonset type 1 diabetes Asian population is comparable to and not lower than those of Caucasians, as had been thought [2,14]. Similar to other studies [2,5,7,9,12], thyroid antibodies were more frequent in female than in male patients in this study. This could not be explained by a selection bias of the study given that the males outnumbered the females. The preponderance in females of thyroid antibodies is well recognized in both diabetics and non-diabetics. Why there is a sex discrepancy is unknown but may be related to the effects of sex steroids on the immunoregulatory process [15]. It is notable that all of our type 1 diabetic patients who had or developed thyroid dysfunction were female. In contrast to other studies from both Asian and Caucasian populations [2,11,16], the higher frequency of thyroid antibodies in patients with a longer duration of diabetes could not be demonstrated in this study. The association between thyroid autoantibodies and subsequent thyroid dysfunction in the general population was well established [16,17]. Several studies in Caucasian childhood type 1 diabetes
also demonstrated that patients who were positive for thyroid antibodies at diabetes onset more frequently had or developed thyroid dysfunction even after several years of diabetes. The frequencies of thyroid dysfunction in thyroid antibodies positive, newly diagnosed patients varied from 10 to 30%, the frequencies of which can increase up to 50% after several years of diabetes [2,4,6,18– 20]. Similar results in Caucasian adult-onset type 1 diabetes were recently reported by FernandezCastaner et al. [7]. However, the prevalence of thyroid dysfunction in Asian, particularly Chinese type 1 diabetes, appears to be lower than those of Caucasians. Tsai et al. reported that only two of 14 Chinese type 1 diabetic children with positive TPO-Ab developed thyroid dysfunction [13]. Our study, in contrast to the Tsai et al. study, shows that the prevalence of thyroid dysfunction in Thai type 1 diabetic patients corresponds to those of Caucasians. It is noticeable that all of the four patients with hyperthyroidism in this study developed the disease prior to or at the diagnosis of diabetes and three patients who developed hypothyroidism did so after 4, 8 and 9 years of diabetes. Our findings confirm the previous notion about thyroid dysfunction in childhood type 1 diabetes that hyperthyroidism commonly occurs at or prior to the diagnosis of diabetes, whereas hypothyroidism may occur at any time during the course of diabetes [21]. The positive association between GAD65Ab and thyroid antibodies in type 1 diabetes of this study is contrary to the study results of Chang et
Table 1 Clinical characteristics of diabetes and frequencies of GAD65Ab in type 1 diabetic patients with and without thyroid antibodiesa Characteristics
Thyroid Ab (+) (N =16)
Thyroid Ab (−) (N =34)
P
Female:male Age at onset of diabetes (years) Duration of diabetes (years) Family history of diabetes in first degree relative Fasting c-peptide (nmol/l)b
10:6 39.0 919.3 (range 10–73) 6.1 9 4.8 (range 0–20) 3/16 0.109 0.09 (range 0.00–0.32) 10/15
9:25 35.3 916.8 (range 9–76) 4.7 9 3.7 (range 0–14) 9/34 0.14 90.11 (range 0.00–0.33)
0.033 NS NS NS NS
GAD65Abc
8/29
Data are expressed as mean 9S.D. Thyroid Ab, thyroid antibodies; NS, not significant. N= 15 for thyroid Ab (+) and 32 for thyroid Ab (−). c N =15 for thyroid Ab (+) and 29 for thyroid Ab (−). a
b
0.029
C. Rattarasarn et al. / Diabetes Research and Clinical Practice 49 (2000) 107–111
al. [12] who reported no association between these antibodies in Chinese type 1 diabetic patients. Why the results of these two studies are at variance despite a similarity in frequencies of GAD65Ab and thyroid antibodies is unclear. The differences in age at onset of diabetes as well as the heterogeneity in genetic background may, in part, contribute to this discrepancy. In conclusion, a significant number of Thai patients with type 1 diabetes had thyroid antibodies particularly female patients and those who had positive GAD65Ab. The presence of thyroid antibodies is associated with a higher frequency of and may predict a higher risk for thyroid dysfunction. Therefore, thyroid antibodies should be screened in all type 1 diabetic patients particularly female and GAD65Ab positive patients and those with positive results should thyroid function tests be measured. Long-term follow-up is required in order to detect thyroid dysfunction in this group of patients. The prevalence of thyroid antibodies and thyroid dysfunction in Thai type 1 diabetic patients appear to be comparable to those of Caucasians. However, a population-based study with a larger number of patients should be performed to confirm our findings Acknowledgements We are grateful to Atchara Thamprasit for preparation of the manuscript and to Kalaya Leetanaporn for assisting in c-peptide analysis. References [1] C.L. Burek, N.R. Rose, K.E. Guire, et al., Thyroid autoantibodies in Black and in White children and adolescents with type 1 diabetes mellitus and their first degree relatives, Autoimmunity 7 (1990) 157–167. [2] R. Lorini, G. d’Annunzio, L. Vitah, S. Scaramuzza, IDDM and autoimmune thyroid disease in the pediatric age group, J. Pediatr. Endocrinol. Metab. 9 (1996) 89–94. [3] B. Lindberg, U.-B. Ericsson, R. Ljung, S.-A. Ivarsson, High prevalence of thyroid autoantibodies at diagnosis of insulin-dependent diabetes mellitus in Swedish children, J. Lab. Clin. Med. 130 (1997) 585–589. [4] M.J. McKenna, R. Herskowitz, J.J. Wolfisdorf, Screening for thyroid disease in children with IDDM, Diabetes Care 13 (1990) 801 – 803.
111
[5] L. Groop, A. Miettinen, P.-H. Groop, S. Meri, S. Koskirnies, G.F. Bottazzo, Organ-specific autoimmunity and HLA-DR antigens as markers for b-cell destruction in patients with type II diabetes, Diabetes 37 (1988) 99 – 103. [6] C.F. Verge, N.J. Howard, M.J. Rowley, et al., Anti-glutamate decarboxylase and other antibodies at the onset of childhood IDDM: a population-based study, Diabetologia 37 (1994) 1113 – 1120. [7] M. Fernandez-Castaner, A. Molina, L. Lopez-Jimenez, J.M. Gomez, J. Soler, Clinical presentation and early course of type 1 diabetes in patients with and without thyroid autoimmunity, Diabetes Care 22 (1999) 377 – 381. [8] E. Kawasaki, H. Takino, M. Yano, et al., Autoantibodies to glutamic acid decarboxylase in patients with IDDM and autoimmune thyroid disease, Diabetes 43 (1994) 80 – 86. [9] M. Neufeld, N.K. Maclaren, W.J. Riley, et al., Islet cell and other organ-specific antibodies in U.S. Caucasians and Blacks with insulin-dependent diabetes mellitus, Diabetes 29 (1980) 589 – 592. [10] The Diabetes Incidence Study in Sweden Group, M. Landin-Olsson, F.A. Karlsson, A. Lernmark, G. Sundkvist, Islet cell and thyrogastric antibodies in 633 consecutive 15- to 34-yr-old patients in the Diabetes Incidence Study in Sweden, Diabetes 41 (1992) 1022 – 1027. [11] A-C. Thai, W.-Y. Ng, K.-E. Lui, J.-S. Cheah, Islet cell and thyroid autoimmunity in Chinese patients with IDDM, Diabetes Care 18 (1995) 586 – 587. [12] C.-C. Chang, C.-N. Huang, L.-M. Chuang, Autoantibodies to thyroid peroxidase in patients with type 1 diabetes in Taiwan, Eur. J. Endocrinol. 139 (1998) 44 – 48. [13] W.-Y. Tsai, J.-S. Lee, Thyroid disease in Chinese children with IDDM, Diabetes Care 16 (1993) 1314 – 1315. [14] G.W.K. Wong, Absence of thyroid disease in Chinese children with IDDM, Diabetes Care 16 (1993) 404 – 405. [15] N.R. Farid, Immunogenetics of autoimmune thyroid disease, Endocrinol. Metab. Clin. North Am. 16 (1987) 229 – 245. [16] B.R. Hawkins, P.S. Cheah, R.L. Dawkins, et al., Diagnostic significance of thyroid microsomal antibodies in randomly selected population, Lancet ii (1980) 1057 – 1059. [17] M.P.J. Vanderpump, W.M.G. Tunbridge, J.M. French, et al., The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham Survey, Clin. Endocrinol. 43 (1995) 55 – 68. [18] G.M. Bright, R.M. Blizzard, D.L. Kaiser, W.L. Clarke, Organ-specific autoantibodies in children with common endocrine diseases, J. Pediatr. 100 (1982) 8 – 14. [19] W.J. Riley, N.K. Maclaren, D.C. Lezotte, R.P. Spillar, A.L. Rosenbloom, Thyroid autoimmunity in insulin-dependent diabetes mellitus: the case for routine screening, J. Pediatr. 98 (1981) 350 – 354. [20] L.M. Prina Cerai, G. Weber, F. Meschi, et al., Prevalence of thyroid autoantibodies and thyroid autoimmune disease in diabetic children and adolescents, Diabetes Care 17 (1994) 782 – 783. [21] G. Jefferson, The clinical approach to thyroid disorders associated with childhood insulin-dependent diabetes mellitus, J. Pediatr. Endocrinol. Metab. 9 (1996) 95 – 100.