Effects of Exogenous Insulin Therapy on Thyroid Nodule Size in Patients with Type 2 Diabetes Mellitus

Objectives: Recent evidence suggests that insulin resistance and endogenous hyperinsulinaemia are related to thyroid nodule growth and development, but the effect of exogenous insulin therapy on thyroid nodules is unknown. The objectives of this study are to investigate the effects of exogenous insulin on thyroid nodule size in diabetic patients. Methods: retrospective analysis of patients with nodular thyroid disease and type 2 diabetes mellitus receiving metformin alone or insulin-and-metformin. Thyroid ultrasonography was used to monitor changes in nodule size: ≥2 mm and ≥20% being statistically signifi cant. Results: 34 insulin-treated (73 nodules) and 53 metformin-only (135 nodules) patients were followed for 5.9-67.9 months (mean 24.5±1.8). Nodule size ranged from 0.17-5.25 cm (mean 1.29±0.08). There were no signifi cant differences in serum TSH, free-T4 or Vitamin D between the two groups, but haemoglobin A1C was higher in insulin-treated patients at baseline and during follow-up (8.75±0.12% vs. 6.48±0.08%; P-value <0.001). In insulin-treated patients, the number of nodules with no change, increase or decrease in size were 49/73 (67.12%), 13/73 (17.81%) and 11/73 (15.07%) at the end of follow-up. The respective number of nodules in the metformin-only group were 105/135 (77.78%), 20/135 (14.81%) and 10/135 (7.41%). Chi-square test for type of treatment and changes in nodule size was not signifi cant (P-value 0.15). Conclusions: Compared to metformin alone, treatment with a combination of exogenous insulin and metformin was not associated with thyroid nodule size or thyroid volumes. The effects of exogenous insulin alone, without metformin, on nodular thyroid disease warrant further evaluation. Research Article Effects of Exogenous Insulin Therapy on Thyroid Nodule Size in Patients with Type 2 Diabetes Mellitus Samer El-Kaissi1*, Badurudeen Mahmood Buhary2 and Suphia Sherbeeni2 1Cleveland Clinic Abu Dhabi, Al Maryah Island, Abu Dhabi, United Arab Emirates 2King Fahad Medical City, Riyadh, Kingdom of Saudi Arabia Dates: Received: 09 December, 2016; Accepted: 29 December, 2016; Published: 02 January, 2017 *Corresponding author: Samer El-Kaissi, Cleveland Clinic Abu Dhabi, Al Maryah Island, Abu Dhabi, United Arab Emirates, E-mail: https://www.peertechz.com


Introduction
Nodular thyroid disease (NTD) and type 2 diabetes mellitus (T2DM) are common disorders with increasing prevalence throughout the world. While not directly linked to each other, T2DM has been associated with a greater serum TSH, increased thyroid volume and a higher prevalence of NTD [1].
While there are no studies linking exogenous insulin treatment in T2DM and thyroid nodule growth, there is mounting evidence that insulin resistance and endogenous hyperinsulinaemia are associated with nodule development [2] and that reduction of insulin resistance has favourable effects on nodule size [3,4]. Hyperinsulinaemia may also be associated with an increased risk of differentiated thyroid carcinoma [5] and persistence of disease post-operatively [6].
Similarly, acromegalic patients with elevated growth hormone and insulin-like growth factor 1 (IGF-1) are at an increased risk for the development of goitre, thyroid nodules [7], and thyroid malignancy [8]. Growth hormone replacement therapy in growth hormone defi cient patients has been associated with increased total thyroid volume and thyroid nodule development [9].
Human insulin and its analogues stimulate IGF-1 receptors potentially leading to cellular proliferation, but the effects of exogenous insulin on thyroid nodule growth are currently unknown. The objective of this study is to compare the effects of metformin alone vs. metformin and exogenous insulin therapy on thyroid nodule size in T2DM patients with nontoxic multinodular goitres.

Methods
This is a retrospective analysis of 87 patients with nontoxic NTD and long-standing T2DM treated with metformin alone (n = 53 patients; total of 135 nodules) or insulin-and-J Thyroid Res 1(1): 004-006. metformin (n = 34 patients; total of 73 nodules) for at least 6-months. Patients were followed for a mean 24.5 ± 1.8 months (range 5.9-67.9 months). A total of 111 nodules underwent fi ne needle aspiration based on clinical suspicion, and 4/77 (5.2%) insulin-treated and 3/138 (2.2%) metformintreated nodules were found to be malignant and excluded from the analysis. Nodule size ranged from 0.17-5.25 cm (mean 1.29±0.08). Thyroid ultrasonography was used to monitor changes in thyroid nodule size, with changes ≥2 mm and ≥20% in 2-dimensions deemed statistically signifi cant. Total thyroid volume (TTV) was estimated from the sum of the right and left thyroid lobes. The volume of each lobe was determined using the 3-dimensional elliptical shape volume formula.
All measurements on thyroid ultrasonography throughout the study were performed by one investigator (SEK). Statistical analysis was performed with the software program Minitab 14.12. Proportions were compared with Fisher's exact test while the sample means were evaluated with the 2-sample t-test.
Chi-square test was used to compare the treatment for diabetes and changes in nodule size in the two groups. Signifi cance was set at P value less than 0.05. The study was approved by our institution's research and ethics committee.

Results
A total of 34 insulin-and-metformin treated patients (73 nodules) and 53 metformin-only patients (135 nodules) were included in the analysis. Apart from a higher haemoglobin A1C in the insulin-treated patients, there were no signifi cant differences at baseline in the age, serum TSH, free-T4, vitamin D levels, mean nodule size or TTV of the two groups (Table   1). During follow-up, there were no signifi cant differences in the serum TSH, free-T4, or vitamin D levels between the two groups, although the haemoglobin A1C remained higher in the insulin-treated patients suggesting less stringent glycaemic control compared to metformin-only patients ( Table 2).

Discussion
In this study, type 2 diabetic patients treated with an insulin-metformin combination regimen did not have signifi cant changes in thyroid nodule size or thyroid volume over a mean follow-up duration of 24-months. These fi ndings are reassuring as many insulin-treated patients with T2DM often have co-existing NTD. However, it remains to be determined if exogenous insulin therapy, without metformin, does not impact the development and progression of NTD. Treatment with metformin has been shown to reduce insulin resistance and has been associated with a decrease in thyroid nodule size [3,4]. In a 6-months study of 66 women with insulin resistance and NTD, metformin therapy was associated with a 30% median reduction in nodule size while the combination of levothyroxine and metformin was associated with 55% reduction [10]. In our study, we could not evaluate the effects of metformin on nodule size as both groups were treated with metformin.
The relationship between exogenous insulin therapy and thyroid malignancy was not examined in this study. A total of seven patients with confi rmed thyroid malignancy (n=7) were excluded from the analysis. However, in a large database   study, Tseng et al., did not fi nd an increased risk of thyroid malignancy in insulin-treated diabetic patients [11].
Limitations of this study include its retrospective design, small sample size and relatively short follow-up duration as changes in thyroid nodule size may take years to occur. The unavailability of baseline thyroid nodule measurements before starting insulin therapy is another limitation, as changes in thyroid nodule size may occur within the fi rst few weeks of insulin therapy. Moreover, the insulin group were also treated with metformin which may have masked the effects of insulin on nodule growth. Lastly, the total number of insulin units per kilogram of body weight was not available. This may be important to determine if patients on a high dose of insulin, e.g. > 2 units/kg, are at a greater risk for thyroid nodule growth.

Conclusions
Compared to metformin alone, there was no association between insulin-metformin combination therapy and changes in nodule size of diabetic patients. Prospective studies are needed to examine this relationship further, especially in patients treated only with insulin rather than insulin and metformin.