Introduction
It is well known that glucose abnormalities are frequent in hyperthyroidism, but little is known about the contributing factors. The aim of our study is to analyze the predictive factors for glucose abnormalities in hyperthyroid subjects.

Material and methods
376 hyperthyroid subjects (278 females, 98 males) were studied to look for glucose abnormalities (diabetes or prediabetes) by performing fasting glycaemia and/or an oral glucose tolerance test. Their mean age was 44.83 years with a range of age between 18 and 86 years. Patients who were known as diabetics were not included. Then, we studied correlations with age, gender, BMI, personal and familial background, aetiology of hyperthyroidism, and thyroid hormones levels.

Results
Forty four percent had glucose abnormalities: 12% had diabetes whereas 32% had prediabetes. Mean age was 41.30 ± 16.56 years for patients with normoglycaemia status, 47.91 ± 15.84 years for patients with prediabetes (p<0.001 versus normal) and 53.15 ± 15.82 years for patients with diabetes (p<0.001 versus normoglycemic and <0.025 versus prediabetes). Hyperglycemia was reported in 44.58% of women and 40.38% of men (p>0.10).

According to hyperthyroidism etiologies, hyperglycemia was observed in 54.87% among patients having nodular toxic goitre and 35.33% in patients having Graves’ disease (p<0.001). Mean body mass index was 23.36 ± 4.50 kg/m² in normoglycemic patients, 23.65 ± 4.30 kg/m² in patients with prediabetes and 24.26 ± 4.50 kg/m² in patients with diabetes (p>0.10). Free tri-iodothyronine (FT3) levels were 19.84 ± 16.14, 19.31 ± 15.77 and 16.60 ± 14.56 pmol/L in normoglycemic, prediabetic and diabetic patients re-spectively (p>0.10). Free thyroxine (FT4) levels were 41.96 ± 29.21, 49.16 ± 22.79 and 41.39 ± 22.21 pmol/L in normoglycemic, prediabetic and diabetic patients respectively (p>0.10). Familial history of diabetes was reported in 23.33% of normoglycemic, 21.87% of prediabetic and 40.42% of diabetic patients (p<0.05 for diabetic versus normoglycemic and prediabetic). Personal background of high blood pressure was reported in 6.69 % of normoglycemic, 10.76 % of prediabetic and 16 % of diabetic patients (p<0.05 for diabetic versus normoglycemic and prediabetic patients).

Discussion
Glucose abnormalities are frequent in our study, 12% for diabetes mellitus and 32% of prediabetes. Our diabetes mellitus frequency is higher than the literature one: 2-8% (Dimitriadis and Raptis, 2001), maybe because some of our patients had an unknown diabetes before their hyperthyroidism. For predictive factors, as expected age seemed to play a great role. This was confirmed by Paul who found 85% of glucose intolerance in old subjects with hyperthyroidism (Paul et al., 2004). On the etiological side, glucose intolerance states were more frequent in toxic nodular goitres than in Graves’ diseases, probably because patients with toxic nodular goitres were older. Backgrounds of personal hypertension or family diabetes mellitus are important too. For thyroid hormones concentrations curiously we did not find any correlation, although thyroid hormones may have numerous effects on glucose homeostasis. Thyroid hormone excess may precipitate glucose intolerance in patients at risk. During hyperthyroidism, there is an increased rate of degradation and an increased secretion of inactive precursors of insulin (Dimitriadis et al., 1985). Thyroid hormones excess increases gut absorption of glucose. Endogenous production of glucose is enhanced in hyperthyroidism through several mechanisms (Wennlund et al., 1986). It is known that hyperthyroidism is associated with insulin resistance as HOMA-IR values and insulin levels are positively correlated with thyroid hormones levels in hyperthyroid subjects. The mechanism of insulin resistance is not completely elucidated. It may be induced by adi-pocytokines. Adiponectin is decreased and resistin is increased in hyperthyroidism. Leptin and ghrelin may also play a role in these anomalies (Hage et al., 2001). Thyroid hormones may also affect energy expenditure by acting on hypothalamic pathways. Thyroid hormones have a repressive effect on MC4R which helps to conserve energy in hyperthyroid states (Decherf et al., 2010).

Conclusion
Glucose abnormalities are frequent in hyperthyroidism. That one may induce hyperglycaemia through different mechanisms. Age, nodular thyroid disease, hypertension and family diabetes are risk factors for glucose intolerance. Thyroid hormones excess seems to precipitate glucose abnormalities in patients who are prone to type 2 diabetes. Seeking for glucose ab-normalities in hyperthyroidism is more than necessary.

References
Decherf S., Seugnet I., Kouidhi S., Lopez-Juarez A., Clerget-Froidevaux MS., and Demeneix B.A., 2010, Thyroid hormone exerts negative feedback on hypothalamic type 4 Melanocortin receptor expression, Proceedings of the National Academy of Sciences of the United States of America, 107(9): 4471–4476
http://dx.doi.org/10.1073/pnas.0905190107

Dimitriadis G., Baker B., Marsh H., et al., 1985, Effect of thyroid hormone excess on action, secretion, and metabolism of insulin in humans, The American journal of physiology, 248(5): E593–E601

Dimitriadis G.D., and Raptis S.A., 2001, Thyroid hormone excess and glucose intolerance, Exp. Clin. Endocrinol. Diabetes, 109: S225-S239
http://dx.doi.org/10.1055/s-2001-18584

Hage M., Zantout MS., and Azar ST., 2011, Thyroid disorders and diabetes mellitus, Journal of Thyroid Research, 1-7
http://dx.doi.org/10.4061/2011/439463

Paul D.T., Mollah F.H., Alam M.K., Fariduddin M., Azad K., and Arslan M.I., 2004, Glycemic status in hyperthyroid subjects, Mymensingh Med. J., 13(1): 71-75

Wennlund A., Felig P., Hagenfeldt L., and Wahren J., 1986, Hepatic glucose production and splanchnic glucose exchange in hyperthyroidism, J. Clinical Endocrinol. Metab., 62: 174–180
http://dx.doi.org/10.1210/jcem-62-1-174