DIABETICMedicine DOI: 10.1111/dme.12342
Short Report: Treatment Type 2 diabetes mellitus and hypothyroidism: the possible influence of metformin therapy L. A. Distiller, E. S. Polakow and B. I. Joffe Centre for Diabetes and Endocrinology, Johannesburg, South Africa Accepted 18 October 2013
Abstract Aims To evaluate the frequency with which hypothyroidism is associated with Type 2 diabetes, to examine gender and ethnic group differences, and to assess the possible impact of metformin therapy. To compare the prevalence of hypothyroidism in a cohort of people with Type 2 diabetes with a previously published cohort of people with Type 1 diabetes from the same centre.
We randomly surveyed the records of 922 people with Type 2 diabetes (576 men and 342 women) to identify diagnoses of hypothyroidism, based on current thyroxin replacement therapy (with previous biochemical confirmation). Four subjects had secondary hypothyroidism after radio-iodine therapy for primary hyperthyroidism and were excluded from the analysis. The prevalence of primary hypothyroidism was documented in the remaining 918 subjects. We assessed the association of metformin therapy with hypothyroidism.
Method
Results The overall prevalence of primary hypothyroidism was 11.8% (women: 22.5%; men: 5.4%, P < 0.001) in subjects with Type 2 diabetes. Inter-ethnic differences were noted, with the highest prevalence among white subjects. The prevalence of hypothyroidism was lower in subjects with Type 2 diabetes who were receiving metformin therapy (P < 0.01), and this difference was greater when assessing those who developed primary hypothyroidism after starting metformin therapy (P < 0.001) Conclusions Our results support a relatively close association between diabetes and hypothyroidism. Hypothyroidism was more common in the cohort of white subjects than in other ethnic groups. The use of metformin therapy in people with Type 2 diabetes was associated with a significantly lower prevalence of diagnosed hypothyroidism.
Diabet. Med. 31, 172–175 (2014)
Introduction Diabetes mellitus and primary hypothyroidism are two of the most common endocrine disorders seen in clinical practice. The former consists of two main subtypes: Type 1, which is usually of autoimmune aetiology, and Type 2, which has a multifactorial genetic and environmental causation [1,2]. Primary hypothyroidism, by contrast, frequently results from autoimmune destruction of thyroid follicles by circulating auto-antibodies (autoimmune thyroiditis) [3]. The potential co-existence of these two common endocrine disorders in the same patient is of considerable interest [4]. Type 1 diabetes and primary hypothyroidism share an autoimmune predisposition, while Type 2 diabetes might be linked to hypothyroidism by virtue of the concurrence of two common endocrine conditions. In addition, recent
Correspondence to: Larry A. Distiller. E-mail: [email protected]
172
literature suggests that metformin therapy may exert an independent effect on thyroid function by suppressing endogenous thyroid-stimulating hormone [5]. The purpose of the present investigation, therefore, was to evaluate formally the frequency of this association with Type 2 diabetes in a large cohort of people attending a diabetes clinic in Johannesburg, South Africa, and to assess whether metformin therapy may affect the prevalence of this disorder.
Methods We randomly surveyed the records of 922 people with previously diagnosed Type 2 diabetes at the Centre for Diabetes and Endocrinology in Johannesburg. The sample contained 577 men and 345 women, of whom the majority were white, 86 were of Asian ethnicity, 119 were black and 32 were mixed race. The diagnosis of Type 2 diabetes was based on the American Diabetes Association criteria [6].
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
Research article
DIABETICMedicine
What’s new? • An association of metformin therapy with the prevalence of hypothyroidism in people with Type 2 diabetes has not previously been described. The diagnosis of hypothyroidism in a particular subject was based on current thyroxine sodium replacement therapy with biochemical confirmation of an elevated serum thyroid-stimulating hormone level before therapy had been instituted. In four subjects, the hypothyroidism was attributable to radioactive iodine therapy for primary hyperthyroidism and these subjects were excluded from the analysis. The remainder were considered to have primary hypothyroidism. Thyroid antibody titres were not available in all cases, and the underlying pathology of the primary hypothyroidism could not be confirmed. We tabulated the number and percentage of subjects with hypothyroidism in each diabetes subgroup and calculated the male: female distribution. We used the Yates corrected chi-squared test to determine significant differences between the diabetes subgroups and sexes when indicated.
Results Table 1 summarizes the percentage of subjects with hypothyroidism in the 918 subjects with Type 2 diabetes, together with a comparison of the different ethnic groups. The overall prevalence was 11.8%, with women having a significantly higher prevalence than men (22.5 vs 5.4%; P < 0.001). Inter-ethnic comparisons showed the highest prevalence of hypothyroidism in white subjects, followed by those of Asian ethnicity, then mixed race and black subjects. There was no significant difference between the age of the subjects and the
duration of diabetes across the different ethnic groups. There was a large range in BMI in this population, varying from 26 to 48 kg/m2. The majority of subjects with Type 2 diabetes were receiving oral metformin therapy. In all cases, metformin therapy had been started at the time of diagnosis of Type 2 diabetes. A total of 43% of subjects were on insulin therapy, with or without metformin, and 61% were in addition receiving either sulphonylurea treatment or a dipeptidyl peptidase-4 inhibitor. The remaining 58% were receiving either metformin, as monotherapy or with the addition of either sulphonylurea, or a dipeptidyl peptidase-4 inhibitor or both. Table 2 shows the association between metformin use and the prevalence of hypothyroidism. In all groups, with the exception of black women, the percentage of subjects on metformin with hypothyroidism was lower than those not receiving metformin (P < 0.01 for all subjects). This finding was enhanced when those who developed hypothyroidism before commencing metformin, and therefore before the diagnosis of diabetes, were excluded.
Discussion Diabetes and primary hypothyroidism are common companions, as suggested by the 11–20 % concordance of both endocrine disorders in the present large cohort study. This is substantially higher than the reported 4% prevalence of hypothyroidism alone in the general Western population [3,7]. The prevalence of hypothyroidism in the general population of South Africa is not known; however, the prevalence of hypothyroidism in those with Type 2 diabetes across several ethnic groups has been reported by a number of studies to be significantly higher than that in the general population [8–10]. The present study supports those findings.
Table 1 Prevalence of hypothyroidism in subjects with Type 2 diabetes Ethnic group and gender White Women Men Asian Women Men Black Women Men Mixed race Women Men All ethnic groups Women Men Total
Number of subjects
Mean (range) age, years
Mean (range) duration of diabetes, years
Prevalence of hypothyroidism, n (%)
245 439
65.7 (36–85) 65.2 (48–82)
10.3 (1–24) 11.3 (2–17)
68 (27.3) 26 (5.9)
30 56
58.1 (37–78) 66 (63–79)
10.4 (1–22) 9.8 (3–23)
7 (23.3) 3 (5.4)
51 68
67.4 (34–79) 66.9 (41–86)
12.8 (1–27) 10.7 (2–19)
1 (2.0) 2 (3.1)
16 16
58.3 (32–77) 60.6 (39–83)
9.4 (2–14) 11.2 (1–17)
1 (6.3) 0 (0)
342 576 918
65.1 (34–86)
10.2 (1–27) 10.9 (1–26)
77 (22.5)* 31 (5.4) 77 (342)
*Prevalence of hypothyroidism significantly higher in women than men (P < 0.001).
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
173
Diabetes and hypothyroidism L. A. Distiller et al.
DIABETICMedicine Table 2 Prevalence of hypothyroidism related to metformin use in subjects with Type 2 diabetes
Ethnic group and gender White Women Men Asian Women Men Black Women Men *All ethnic groups Women Men
Subjects not on concurrent metformin, n/N (%)
Subjects on concurrent metformin, n/N (%)
Subjects who developed hypothyroidism after starting metformin, n/N (%)
14/39 (35.9) 4/47 (8.5)
53/206 (25.7) 22/392 (5.6)
9/206 (4.4) 6/392 (1.5)
4/7 (57.1) 1/9 (11.1)
3/23 (13.0) 2/47 (4.1)
0/23 (0) 0/47 (0)
0/5 (0) 2/9 (22.2)
1/46 (2.2) 0/59 (0)
0/46 (0) 0/59 (0)
18/54 (33.3) 7/66 (10.6)
58/288 (20.2) 24/510 (4.7)
9/288 (3.1) 6/510 (1.2)
*Significantly lower prevalence of hypothyroidism in subjects receiving metformin (P < 0.01) and in those in whom hypothyroidism was diagnosed after starting metformin treatment (P < 0.001).
We have previously reported a prevalence of hypothyroidism in people with Type 1 diabetes of 20.2 % (women: 30.9 %; men: 10.1 %), which is significantly higher than in the present cohort of people with Type 2 diabetes (P < 0.01), both for women and men [11]. Reasons for this close association appear to differ in the two diabetes subgroups. In Type 1 diabetes, a common autoimmune predisposition seems likely to be the cause, with recent studies showing a shared genetic susceptibility to both conditions [12]. In Type 2 diabetes, reasons for the close association are more complex. It is unlikely to be just a coincidence of two common endocrine conditions, since among people in this group there is a higher frequency of hypothyroidism than in the general population. Dietary deficiency of iodine is an intriguing possibility, in view of a recent report highlighting a deficiency of dietary iodine intake in three major US weight loss programmes [13]. There is also the possibility that hypothyroidism and hyperglycaemia may each adversely influence the other disorder of metabolism [14]. A possible effect of metformin therapy in lowering serum thyroid-stimulating hormone concentrations in people with diabetes and primary hypothyroidism has been suggested [5] and this may be supported by recent evidence that metformin inhibits goitre formation in people with Type 2 diabetes [15]. While the association between metformin usage and a lower prevalence of hypothyroidism does not necessarily imply a ‘protective’ effect of metformin therapy, the results of the present simple survey raise this possibility. Further studies are clearly required to clarify this finding, including the likelihood that metformin-induced suppression of thyroid-stimulating hormone may result in the diagnosis of
174
hypothyroidism being overlooked when thyroid-stimulating hormone screening is the only method used. The present study has some limitations. Detailed dietary histories, particularly concerning iodine intake in the different ethnic groups were generally unavailable. Because of logistic and economic constraints, we were unable to obtain a suitably matched control group of subjects without diabetes to assess the prevalence of primary hypothyroidism in our own local community, and relied instead on international databases to obtain this information. We also made no attempt to relate the subjects’ BMI or the presence of macroangiopathy to the finding of hypothyroidism, factors which have previously been reported to be factors possibly associated with the occurrence of hypothyroidism [4]. The present findings raise the interesting possibility of an association between metformin therapy and hypothyroidism in the population with Type 2 diabetes, although we have not established whether this may be attributable to a specific protective effect or a masking of underlying hypothyroidism by metformin. This finding is contrary to the report by Diez and Iglesias [4], who noted that metformin therapy was a significant risk factor for hypothyroidism. Their results were obtained by a screening programme and ours was a retrospective analysis, and this different study design may partially explain the different findings. Nevertheless, the present short report acts as an alert to a possible association that needs further exploration in larger studies. Finally, our findings raise the question as to whether screening for thyroid dysfunction should form part of the diagnostic evaluation for people with Type 2 diabetes, and whether in these people, a serum thyroid-stimulating hormone level test alone may be sufficient for those on metformin therapy. Clearly, screening for hypothyroidism is indicated for subjects with Type 1 diabetes; for people with Type 2 diabetes, this recommendation is less clear-cut and would appear to depend on factors such as sex, ethnic origin and probably age. Similar guidelines have been presented by other authorities [16].
Funding sources
None.
Competing interests
None declared.
References 1 Groop L. Genetic and metabolic heterogeneity of diabetes. Int Diabetes Monitor 2000; 12: 1–6. 2 De Fronzo RA. Pathogenesis of type 2 diabetes mellitus. Med Clin N Am 2004; 88: 787–835. 3 Hollowell JH, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA et al. Serum TSH, T4, and thyroid antibodies in the United States population (1988 to 1994): National Health and
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
Research article
4
5
6 7 8
9
10
Nutrition Examination Survey (NHANES III). J. Clin Endocrinol Metab 2002; 87: 489–499. Diez JJ, Iglesias P. An analysis of the relative risk for hypothyroidism in people with Type 2 diabetes. Diabet Med 2012; 29: 1510– 1514. Capelli C, Rotondi M, Pinola I, Agosti B, Gandossi F, Valentini U et al. TSH lowering effect of metformin in Type 2 diabetic people. Diabetes Care 2009; 32: 1589–1590. American Diabetes Association. Standards of medical care in diabetes - 2013. Diabetes Care 2013; 36(Suppl. 1): S11–S13. Olsen AH, Kolsberg G, Coffey J. Should we screen women for hypothyroidism? Journal of Family Practise 2004; 53: 1–4. Demitrost L, Ranabir .Thyroid Dysfunction in type 2 diabetes mellitus: A retrospective study. Indian J Endocrinol Metab 2012: 16 (Supp 2): S334–S335 Ghazali SM, Abbiyesuku FM. Thyroid dysfunction in type diabetics seen at the University College Hospital, Ibadan. Nigeria. Niger J Physiol Sci 2010; 28: 173–179. Kadiyala R, Peter R, Okosieme OE. THroid dysfunction in patients with diabetes: Clinical implications and screening strategies. Int J Clin Pract 2012; 64: 1130–1139.
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
DIABETICMedicine 11 Joffe B, Distiller L, Landau S, Blacking L, Klisiewicz A. Spectrum of autoimmune disorders in Type 1 diabetes – a cross sectional clinical audit. J Diabetes Metab 2010; 1: 112. doi:10. 4172/2 155-6156. 1000112. 12 Huber A, Menconi F, Corathers S., Jacobsen EM, Tamer Y. Joint genetic susceptibility to Type 1 diabetes and autoimmune thyroiditis: from epidemiology to mechanisms. Endoc Rev. 2008; 29: 697–725. 13 Kuriti M, Pearce E N, Braverman L E, He X, Leung A M. Iodine content of U.S. Weight Loss food. The Endocrine Society’s 95th Annual Meeting San Francisco 2013: Poster Board MON-457 (Abstract) 14 Brenta G. Diabetes and thyroid disorders. Br J Diabetes Vasc Dis 2010; 10: 172–177. 15 Itterman T, Markus MR, Schipf S, Derwahl M, Meisinger C, Volzke H. Metformin inhibits goitrogenous effects of type 2 diabetes. Eur J Endocrinol 2013; 169: 9–15. 16 American Association of Clinical Endocrinologists. The American Association of Clinical Endocrinologists medical guidelines for the management of diabetes mellitus: the ACCE system of intensive diabetes self-management, 2000 update. Endoc Pract 2000; 6: 43084.
175
Short Report: Treatment Type 2 diabetes mellitus and hypothyroidism: the possible influence of metformin therapy L. A. Distiller, E. S. Polakow and B. I. Joffe Centre for Diabetes and Endocrinology, Johannesburg, South Africa Accepted 18 October 2013
Abstract Aims To evaluate the frequency with which hypothyroidism is associated with Type 2 diabetes, to examine gender and ethnic group differences, and to assess the possible impact of metformin therapy. To compare the prevalence of hypothyroidism in a cohort of people with Type 2 diabetes with a previously published cohort of people with Type 1 diabetes from the same centre.
We randomly surveyed the records of 922 people with Type 2 diabetes (576 men and 342 women) to identify diagnoses of hypothyroidism, based on current thyroxin replacement therapy (with previous biochemical confirmation). Four subjects had secondary hypothyroidism after radio-iodine therapy for primary hyperthyroidism and were excluded from the analysis. The prevalence of primary hypothyroidism was documented in the remaining 918 subjects. We assessed the association of metformin therapy with hypothyroidism.
Method
Results The overall prevalence of primary hypothyroidism was 11.8% (women: 22.5%; men: 5.4%, P < 0.001) in subjects with Type 2 diabetes. Inter-ethnic differences were noted, with the highest prevalence among white subjects. The prevalence of hypothyroidism was lower in subjects with Type 2 diabetes who were receiving metformin therapy (P < 0.01), and this difference was greater when assessing those who developed primary hypothyroidism after starting metformin therapy (P < 0.001) Conclusions Our results support a relatively close association between diabetes and hypothyroidism. Hypothyroidism was more common in the cohort of white subjects than in other ethnic groups. The use of metformin therapy in people with Type 2 diabetes was associated with a significantly lower prevalence of diagnosed hypothyroidism.
Diabet. Med. 31, 172–175 (2014)
Introduction Diabetes mellitus and primary hypothyroidism are two of the most common endocrine disorders seen in clinical practice. The former consists of two main subtypes: Type 1, which is usually of autoimmune aetiology, and Type 2, which has a multifactorial genetic and environmental causation [1,2]. Primary hypothyroidism, by contrast, frequently results from autoimmune destruction of thyroid follicles by circulating auto-antibodies (autoimmune thyroiditis) [3]. The potential co-existence of these two common endocrine disorders in the same patient is of considerable interest [4]. Type 1 diabetes and primary hypothyroidism share an autoimmune predisposition, while Type 2 diabetes might be linked to hypothyroidism by virtue of the concurrence of two common endocrine conditions. In addition, recent
Correspondence to: Larry A. Distiller. E-mail: [email protected]
172
literature suggests that metformin therapy may exert an independent effect on thyroid function by suppressing endogenous thyroid-stimulating hormone [5]. The purpose of the present investigation, therefore, was to evaluate formally the frequency of this association with Type 2 diabetes in a large cohort of people attending a diabetes clinic in Johannesburg, South Africa, and to assess whether metformin therapy may affect the prevalence of this disorder.
Methods We randomly surveyed the records of 922 people with previously diagnosed Type 2 diabetes at the Centre for Diabetes and Endocrinology in Johannesburg. The sample contained 577 men and 345 women, of whom the majority were white, 86 were of Asian ethnicity, 119 were black and 32 were mixed race. The diagnosis of Type 2 diabetes was based on the American Diabetes Association criteria [6].
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
Research article
DIABETICMedicine
What’s new? • An association of metformin therapy with the prevalence of hypothyroidism in people with Type 2 diabetes has not previously been described. The diagnosis of hypothyroidism in a particular subject was based on current thyroxine sodium replacement therapy with biochemical confirmation of an elevated serum thyroid-stimulating hormone level before therapy had been instituted. In four subjects, the hypothyroidism was attributable to radioactive iodine therapy for primary hyperthyroidism and these subjects were excluded from the analysis. The remainder were considered to have primary hypothyroidism. Thyroid antibody titres were not available in all cases, and the underlying pathology of the primary hypothyroidism could not be confirmed. We tabulated the number and percentage of subjects with hypothyroidism in each diabetes subgroup and calculated the male: female distribution. We used the Yates corrected chi-squared test to determine significant differences between the diabetes subgroups and sexes when indicated.
Results Table 1 summarizes the percentage of subjects with hypothyroidism in the 918 subjects with Type 2 diabetes, together with a comparison of the different ethnic groups. The overall prevalence was 11.8%, with women having a significantly higher prevalence than men (22.5 vs 5.4%; P < 0.001). Inter-ethnic comparisons showed the highest prevalence of hypothyroidism in white subjects, followed by those of Asian ethnicity, then mixed race and black subjects. There was no significant difference between the age of the subjects and the
duration of diabetes across the different ethnic groups. There was a large range in BMI in this population, varying from 26 to 48 kg/m2. The majority of subjects with Type 2 diabetes were receiving oral metformin therapy. In all cases, metformin therapy had been started at the time of diagnosis of Type 2 diabetes. A total of 43% of subjects were on insulin therapy, with or without metformin, and 61% were in addition receiving either sulphonylurea treatment or a dipeptidyl peptidase-4 inhibitor. The remaining 58% were receiving either metformin, as monotherapy or with the addition of either sulphonylurea, or a dipeptidyl peptidase-4 inhibitor or both. Table 2 shows the association between metformin use and the prevalence of hypothyroidism. In all groups, with the exception of black women, the percentage of subjects on metformin with hypothyroidism was lower than those not receiving metformin (P < 0.01 for all subjects). This finding was enhanced when those who developed hypothyroidism before commencing metformin, and therefore before the diagnosis of diabetes, were excluded.
Discussion Diabetes and primary hypothyroidism are common companions, as suggested by the 11–20 % concordance of both endocrine disorders in the present large cohort study. This is substantially higher than the reported 4% prevalence of hypothyroidism alone in the general Western population [3,7]. The prevalence of hypothyroidism in the general population of South Africa is not known; however, the prevalence of hypothyroidism in those with Type 2 diabetes across several ethnic groups has been reported by a number of studies to be significantly higher than that in the general population [8–10]. The present study supports those findings.
Table 1 Prevalence of hypothyroidism in subjects with Type 2 diabetes Ethnic group and gender White Women Men Asian Women Men Black Women Men Mixed race Women Men All ethnic groups Women Men Total
Number of subjects
Mean (range) age, years
Mean (range) duration of diabetes, years
Prevalence of hypothyroidism, n (%)
245 439
65.7 (36–85) 65.2 (48–82)
10.3 (1–24) 11.3 (2–17)
68 (27.3) 26 (5.9)
30 56
58.1 (37–78) 66 (63–79)
10.4 (1–22) 9.8 (3–23)
7 (23.3) 3 (5.4)
51 68
67.4 (34–79) 66.9 (41–86)
12.8 (1–27) 10.7 (2–19)
1 (2.0) 2 (3.1)
16 16
58.3 (32–77) 60.6 (39–83)
9.4 (2–14) 11.2 (1–17)
1 (6.3) 0 (0)
342 576 918
65.1 (34–86)
10.2 (1–27) 10.9 (1–26)
77 (22.5)* 31 (5.4) 77 (342)
*Prevalence of hypothyroidism significantly higher in women than men (P < 0.001).
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
173
Diabetes and hypothyroidism L. A. Distiller et al.
DIABETICMedicine Table 2 Prevalence of hypothyroidism related to metformin use in subjects with Type 2 diabetes
Ethnic group and gender White Women Men Asian Women Men Black Women Men *All ethnic groups Women Men
Subjects not on concurrent metformin, n/N (%)
Subjects on concurrent metformin, n/N (%)
Subjects who developed hypothyroidism after starting metformin, n/N (%)
14/39 (35.9) 4/47 (8.5)
53/206 (25.7) 22/392 (5.6)
9/206 (4.4) 6/392 (1.5)
4/7 (57.1) 1/9 (11.1)
3/23 (13.0) 2/47 (4.1)
0/23 (0) 0/47 (0)
0/5 (0) 2/9 (22.2)
1/46 (2.2) 0/59 (0)
0/46 (0) 0/59 (0)
18/54 (33.3) 7/66 (10.6)
58/288 (20.2) 24/510 (4.7)
9/288 (3.1) 6/510 (1.2)
*Significantly lower prevalence of hypothyroidism in subjects receiving metformin (P < 0.01) and in those in whom hypothyroidism was diagnosed after starting metformin treatment (P < 0.001).
We have previously reported a prevalence of hypothyroidism in people with Type 1 diabetes of 20.2 % (women: 30.9 %; men: 10.1 %), which is significantly higher than in the present cohort of people with Type 2 diabetes (P < 0.01), both for women and men [11]. Reasons for this close association appear to differ in the two diabetes subgroups. In Type 1 diabetes, a common autoimmune predisposition seems likely to be the cause, with recent studies showing a shared genetic susceptibility to both conditions [12]. In Type 2 diabetes, reasons for the close association are more complex. It is unlikely to be just a coincidence of two common endocrine conditions, since among people in this group there is a higher frequency of hypothyroidism than in the general population. Dietary deficiency of iodine is an intriguing possibility, in view of a recent report highlighting a deficiency of dietary iodine intake in three major US weight loss programmes [13]. There is also the possibility that hypothyroidism and hyperglycaemia may each adversely influence the other disorder of metabolism [14]. A possible effect of metformin therapy in lowering serum thyroid-stimulating hormone concentrations in people with diabetes and primary hypothyroidism has been suggested [5] and this may be supported by recent evidence that metformin inhibits goitre formation in people with Type 2 diabetes [15]. While the association between metformin usage and a lower prevalence of hypothyroidism does not necessarily imply a ‘protective’ effect of metformin therapy, the results of the present simple survey raise this possibility. Further studies are clearly required to clarify this finding, including the likelihood that metformin-induced suppression of thyroid-stimulating hormone may result in the diagnosis of
174
hypothyroidism being overlooked when thyroid-stimulating hormone screening is the only method used. The present study has some limitations. Detailed dietary histories, particularly concerning iodine intake in the different ethnic groups were generally unavailable. Because of logistic and economic constraints, we were unable to obtain a suitably matched control group of subjects without diabetes to assess the prevalence of primary hypothyroidism in our own local community, and relied instead on international databases to obtain this information. We also made no attempt to relate the subjects’ BMI or the presence of macroangiopathy to the finding of hypothyroidism, factors which have previously been reported to be factors possibly associated with the occurrence of hypothyroidism [4]. The present findings raise the interesting possibility of an association between metformin therapy and hypothyroidism in the population with Type 2 diabetes, although we have not established whether this may be attributable to a specific protective effect or a masking of underlying hypothyroidism by metformin. This finding is contrary to the report by Diez and Iglesias [4], who noted that metformin therapy was a significant risk factor for hypothyroidism. Their results were obtained by a screening programme and ours was a retrospective analysis, and this different study design may partially explain the different findings. Nevertheless, the present short report acts as an alert to a possible association that needs further exploration in larger studies. Finally, our findings raise the question as to whether screening for thyroid dysfunction should form part of the diagnostic evaluation for people with Type 2 diabetes, and whether in these people, a serum thyroid-stimulating hormone level test alone may be sufficient for those on metformin therapy. Clearly, screening for hypothyroidism is indicated for subjects with Type 1 diabetes; for people with Type 2 diabetes, this recommendation is less clear-cut and would appear to depend on factors such as sex, ethnic origin and probably age. Similar guidelines have been presented by other authorities [16].
Funding sources
None.
Competing interests
None declared.
References 1 Groop L. Genetic and metabolic heterogeneity of diabetes. Int Diabetes Monitor 2000; 12: 1–6. 2 De Fronzo RA. Pathogenesis of type 2 diabetes mellitus. Med Clin N Am 2004; 88: 787–835. 3 Hollowell JH, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA et al. Serum TSH, T4, and thyroid antibodies in the United States population (1988 to 1994): National Health and
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
Research article
4
5
6 7 8
9
10
Nutrition Examination Survey (NHANES III). J. Clin Endocrinol Metab 2002; 87: 489–499. Diez JJ, Iglesias P. An analysis of the relative risk for hypothyroidism in people with Type 2 diabetes. Diabet Med 2012; 29: 1510– 1514. Capelli C, Rotondi M, Pinola I, Agosti B, Gandossi F, Valentini U et al. TSH lowering effect of metformin in Type 2 diabetic people. Diabetes Care 2009; 32: 1589–1590. American Diabetes Association. Standards of medical care in diabetes - 2013. Diabetes Care 2013; 36(Suppl. 1): S11–S13. Olsen AH, Kolsberg G, Coffey J. Should we screen women for hypothyroidism? Journal of Family Practise 2004; 53: 1–4. Demitrost L, Ranabir .Thyroid Dysfunction in type 2 diabetes mellitus: A retrospective study. Indian J Endocrinol Metab 2012: 16 (Supp 2): S334–S335 Ghazali SM, Abbiyesuku FM. Thyroid dysfunction in type diabetics seen at the University College Hospital, Ibadan. Nigeria. Niger J Physiol Sci 2010; 28: 173–179. Kadiyala R, Peter R, Okosieme OE. THroid dysfunction in patients with diabetes: Clinical implications and screening strategies. Int J Clin Pract 2012; 64: 1130–1139.
ª 2013 The Authors. Diabetic Medicine ª 2013 Diabetes UK
DIABETICMedicine 11 Joffe B, Distiller L, Landau S, Blacking L, Klisiewicz A. Spectrum of autoimmune disorders in Type 1 diabetes – a cross sectional clinical audit. J Diabetes Metab 2010; 1: 112. doi:10. 4172/2 155-6156. 1000112. 12 Huber A, Menconi F, Corathers S., Jacobsen EM, Tamer Y. Joint genetic susceptibility to Type 1 diabetes and autoimmune thyroiditis: from epidemiology to mechanisms. Endoc Rev. 2008; 29: 697–725. 13 Kuriti M, Pearce E N, Braverman L E, He X, Leung A M. Iodine content of U.S. Weight Loss food. The Endocrine Society’s 95th Annual Meeting San Francisco 2013: Poster Board MON-457 (Abstract) 14 Brenta G. Diabetes and thyroid disorders. Br J Diabetes Vasc Dis 2010; 10: 172–177. 15 Itterman T, Markus MR, Schipf S, Derwahl M, Meisinger C, Volzke H. Metformin inhibits goitrogenous effects of type 2 diabetes. Eur J Endocrinol 2013; 169: 9–15. 16 American Association of Clinical Endocrinologists. The American Association of Clinical Endocrinologists medical guidelines for the management of diabetes mellitus: the ACCE system of intensive diabetes self-management, 2000 update. Endoc Pract 2000; 6: 43084.
175
