Ultrasound assessment of thyroid gland volume in diabetic patients without overt thyroid disease.

BACKGROUND: Diabetes mellitus and thyroid disease are known to mutually influence each other. Thyroid disease can worsen glycaemic control in diabetes, and patients with diabetes mellitus have increased incidence of thyroid disorders such as increase in size, compared to the normal population. AIMS/
OBJECTIVES: The aim of the study was to sonographically assess thyroid gland volume in Nigerian adult diabetic patients and compare with apparently healthy adults (controls). SUBJECTS AND METHODS: The study setting was the Department of Radiology, University of Benin Teaching Hospital (UBTH) Nigeria. The thyroid gland in 120 diabetic subjects and equal number of apparently healthy controls was scanned with a 5-12 MHz linear transducer of a SONOACE X4 Machine. Thyroid gland volume was assessed. Statistical analysis was done with Statistical Package for the Social Sciences (SPSS) version 17 (SPSS Inc, Chicago, IL, USA).
RESULTS: Diabetics had significantly increased thyroid volume compared to age matched male and female control subjects (11.5 ± 5.2 cm3 vs 7.4 ± 1.9 cm3; P< 0.001 for males) and (9.9 ± 6.2 cm3 vs 7.1 ± 3.4 cm3; P< 0.001 for females) respectively. Gender did not significantly influence thyroid volume among diabetics.
CONCLUSION: Diabetics have higher thyroid gland dimensions, compared to apparently healthy subjects. Gland proliferation from circulating insulin may play a role. This is not influenced by gender.

Introduction

Thyroid disease is common in the general population, with a prevalence of 6.6%.[1] These disorders include goiter, thyroiditis, and neoplasia. Diabetes mellitus is a disease condition characterized by absolute or relative deficiency of insulin or its action,[2] resulting in hyperglycemia and disorders of carbohydrate, protein, and lipid metabolism.

Thyroid disease and diabetes mellitus are the two most common endocrine disorders encountered in clinical practice and have been known to mutually influence each other.[13] Increased circulating insulin in diabetes has a proliferative effect on thyroid tissue, while hyperthyroidism increases liver gluconeogenesis and gastrointestinal absorption of glucose, leading to worsening glycemic control. Hypothyroidism may increase susceptibility to hypoglycemia.[13]

Thyroid disorders are more common in type 1 diabetes,[3] and the prevalence is influenced by advancing age, and gender.[4] Some of the disorders include subclinical hypothyroidism, hyperthyroidism, and postpartum thyroiditis, and may manifest in the form of glandular enlargement. Therefore, regular screening of the thyroid gland in diabetic patients will allow early detection and treatment of these disorders.

Ultrasonography is a convenient affordable and safe method of imaging the thyroid gland.[5] It is superior to palpation, faster, cheaper, and more efficient for estimating thyroid volume and parenchymal abnormality than computerized tomography and magnetic resonance imaging.[56]

Although there are reports on ultrasound assessment of the thyroid gland in healthy Nigerians, such studies among Nigerian diabetics are scanty in the literature. The aim of this study was to contribute to data on thyroid gland dimensions among diabetic patients.

Subjects and Methods

This was a prospective ultrasound evaluation of the thyroid gland in 120 adult diabetic patients and equal number of age and sex-matched controls. The study period spanned 6 months, from June to December 2013. The diabetic patients were those attending the consultant outpatient clinic of the University of Benin Teaching Hospital (UBTH) Nigeria with no overt signs of thyroid disease while the control group consisted of apparently healthy volunteer adults already screened for thyroid disease and diabetes at the Centre for Disease Control of UBTH. Approval to conduct the study was obtained from Ethics and Research Committee of the hospital.

Study subjects aged <18 years, including those who were pregnant or with previous history of neck trauma, surgery, or irradiation were excluded. Informed consent was obtained from selected subjects, and bio-data, including age, recorded. Height and weight were measured, with meter height scale (Avery Co., Ltd., England) and calibrated weighing scale (Medfield Equipment and Scientific Ltd., England) respectively.

Scanning was performed with the 5–12 MHz linear transducer of an SONOACE X4 ultrasound machine (Medison Inc., Korea 2010). Each subject was assessed in the supine position with the head extended. Using the common carotid artery and internal jugular vein as landmarks, the thyroid gland was identified, and maximum anteroposterior (AB) and mediolateral (CD) dimension of each thyroid lobe was measured on the transverse scan [Figure 1]. The craniocaudal dimension was also measured on a longitudinal scan [Figure 2]. All measurements were taken thrice, and average values recorded to minimize intra-observer error. The volume of each lobe was calculated by the ellipsoid formula thus;[7]

Figure 1

Ultrasound image (and line diagram) of the right lobe of the thyroid gland on transverse scan showing the measurement planes

Figure 2

Ultrasound image (and line diagram) of the right lobe of the thyroid on longitudinal scan showing measurement planes

Volume = Anteroposterior (thickness) × Mediolateral (width) × Craniocaudal (length) × 0.49 (conversion factor)

The volume of the entire thyroid gland was calculated by addition of the volumes of each lobe. Body mass index (BMI) for each subject was calculated by dividing the weight in kilograms by the square of the height in square meters. Body surface area (BSA) was calculated using the Dubois formula.[8]

BSA = 0.20247 × Height (m)0.725 × Weight (kg)0.425

Data obtained were analyzed using Statistical Package for the Social Sciences, SPSS version 17 (SPSS Inc., Chicago, IL, USA).

Data comparison/tests of significance were done with Student's t-test and Pearson correlation coefficient for continuous variables. At 95% of the confidence interval, two-tailed P ≤ 0.05 were considered significant.

Results

One hundred and twenty adult diabetic subjects and an equal number of controls were studied. The age range for the diabetic group was 20–76 years (mean: 54.5 ± 13.5 years), comprising 52 males (43%) and 68 females (57%). The bio-data of the study population is as shown in Table 1. Using unpaired t-test, the differences in age, height, weight, BMI and BSA between diabetic patients and controls was not statistically significant. Diabetic subjects had a statistically significant higher thyroid gland volume compared to controls which are shown in Table 1.

Table 1

General data of the studied population showing the mean age, weight, height, BMI, BSA, thyroid volume (including standard deviation) in diabetics and controls

Variables	Diabetics (n=120)	Control (n=120)	P value
Age (years) (mean±SD)	54.5±13.5	52.0±12.3	0.140
Weight (kg) (mean±SD)	74.8±12.0	73.5±10.5	0.386
Height (m) (mean±SD)	1.7±0.1	1.65±0.1	0.808
Body mass index (kg/g2) (mean±SD)	27.5±4.5	26.85±3.9	0.236
Body surface area (m2) (mean±SD)	1.82±0.2	1.81±0.39	0.654
Thyroid volume (cm3) (mean±SD)	10.6±5.9	7.18±2.9	0.0001*

SD=Standard deviation, n=Number, %=Percentage, *=Significant

Thyroid volume was significantly increased in male and female diabetic subjects compared to their sex-matched controls, as depicted in Table 2. However, the difference in thyroid volume between male and female diabetic patients was not statistically significant [Table 3].

Table 2

Comparison of mean thyroid volume, age, weight, height, BMI and BSA in male and female diabetics patients and controls

Subject gender	Age (years) (mean±SD)	Weight (kg) (mean±SD)	Height (m) (mean±SD)	Body mass index (kg/m2) (mean±SD)	Body surface area (m2) (mean±SD)	Thyroid volume (cm3) (mean±SD)
Males
Diabetics (n=52)	56.5±14.5	70.9±11.4	1.7±0.1	24.5±4.2	1.9±0.1	11.5±5.2
Control (n=44)	51.4±13.6	71.2±1.0	1.7±0.1	24.8±3.5	1.8±0.1	7.4±1.9
P value	0.081	0.907	0.491	0.723	0.199	0.0001*
Females
Diabetics (n=68)	52.9±12.5	75.7±13.9	1.6±0.1	28.8±4.7	1.8±0.2	9.9±6.2
Control (n=76)	52.4±11.5	74.9±10.7	1.6±0.1	28.2±3.6	1.8±0.1	7.1±3.4
P value	0.776	0.709	0.414	0.363	0.677	0.001*

SD=Standard deviation, n=Number, %=Percentage, *=Significant

Table 3

Data showing age, weight, height, BMI, BSA, thyroid volume and presence of nodules between male and female diabetics

Variables	Male (n=52)	Female (n=68)	P value
Age (years) (mean±SD)	56.5±14.5	52.9±12.5	0.157
Weight (kg) (mean±SD)	70.9±11.5	75.7±13.9	0.043*
Height (m) (mean±SD)	1.7±0.1	1.6±0.1	0.000*
Body mass index (kg/m2) (mean±SD)	24.5±4.2	29.1±4.7	0.000*
Body surface area (m2) (mean±SD)	1.9±0.1	1.79±0.17	0.030*
Thyroid volume (cm3) (mean±SD)	11.5±5.2	9.9±6.2	0.145

SD=Standard deviation, n=Number, %=Percentage, *=Significant

Correlation values of thyroid volume with age, weight and height, for male and female diabetic patients and their respective controls as well as duration of illness (for diabetic patients only) are shown in Table 4a and b. For male diabetic subjects, there was no significant correlation between thyroid volume and age (r = 0.174; P = 0.217), weight (r = 0.131; P = 0.355), height (r = 0.164; P = 0.245), or duration of illness (r = 0.148; P = 0.296). Among female diabetic subjects, there was no significant correlation between thyroid volume and age (r = 0.229; P = 0.060), weight (r = 0.135; P = 0.273), or duration of illness (r = 0.170; P = 0.165). For male controls, there was significant correlation between thyroid volume and weight (r = 0.342; P = 0.023). Female controls had significant correlation between thyroid volume and height (r = 0.398; P = 0.000). There was no significant correlation with age (r = 0.022; P = 0.847).

Table 4a

Correlation of thyroid volume with age, weight, height and duration of illness of subjects in male diabetics and controls

	Thyroid volume		Age		Weight (kg)		Height (m)		Duration of Illness
	r	P	r	P	r	P	r	P	r	P
Diabetic diabetics
Thyroid volume (cm3)	1		0.174	0.217	0.131	0.355	0.164	0.245	0.148	0.296
Age (years)	0.174	0.217	1		0.003	0.985	−0.116	0.413	0.249	0.075
Weight (kg)	0.131	0.355	0.003	0.985	1		0.062	0.66	−0.102	0.47
Height (m)	0.164	0.245	−0.116	0.413	0.062	0.66	1		0.073	0.608
Duration of illness	0.148	0.296	0.249	0.075	−0.102	0.47	0.073	0.608	1
Control
Thyroid volume (cm3)	1		−0.332*	0.028	0.342*	0.023	0.477**	0.001
Age (years)	−0.332*	0.028	1		−0.01	0.947	−0.472**	0.001
Weight (kg)	0.342*	0.023	−0.01	0.947	1		0.263	0.084
Height (m)	0.477**	0.001	−0.472**	0.001	0.263	0.084	1

*Correlation is significant at the 0.05 level (2-tailed), **Correlation is significant at the 0.01 level (2-tailed), r=Pearson correlation, and P=Significant/P value

Table 4b

Correlation of thyroid volume with age, weight, height and duration of illness of subjects in both male and female diabetics and controls

	Thyroid volume		Age		Weight (kg)		Height (m)		Duration of illness
	r	P	r	P	r	P	r	P	r	P
Diabetics patients
Thyroid volume (cm3)	1		0.229	0.06	0.135	0.273	0.682**	0	0.17	0.165
Age (years)	0.229	0.06	1		0.092	0.458	0.176	0.151	0.145	0.238
Weight (kg)	0.135	0.273	0.157	0.2	1		0.535**	0	0.129	0.295
Height (m)	0.682**	0	0.176	0.151	0.357**	0.003	1		0.260*	0.032
Duration of illness	0.17	0.165	0.145	0.238	0.022	0.861	0.260*	0.032	1
Control
Thyroid volume (cm3)	1		−0.022	0.847	0.176	0.129	0.398**	0
Age (years)	−0.022	0.847	1		0.167	0.149	−0.352**	0.002
Weight (kg)	0.176	0.129	0.167	0.149	1		0.443**	0
Height (m)	0.398**	0	−0.352**	0.002	0.443**	0	1

*Correlation is significant at the 0.05 level (2-tailed), **Correlation is significant at the 0.01 level (2-tailed), r=Pearson correlation, and P=significant/P value

Scatter diagrams showing Pearson's correlation coefficients are illustrated in Figures 3–6.

Figure 3

Scatter plot for thyroid volume against age, height, weight, and duration of illness in male diabetics

Figure 6

Regression graph for thyroid volume against age, height and weight for female controls

Scatter plot of thyroid volume against age, height, weight, and duration of illness for female diabetics

Regression graph for thyroid volume against age, height and weight for male controls

Discussion

Thyroid disease and diabetes mellitus are the two most common endocrine diseases in clinical practice and have been shown to influence each other.[3] Thus, regular screening for thyroid disease in diabetics will allow early detection and possible correction of any such disease as this may improve general patient care.

Total thyroid volume in all diabetic subjects in this study was significantly higher than that of controls. Thyroid volume was also significantly increased in male and female diabetics compared to their age-matched controls. This is in agreement with a previous study by Junik et al.[9] who assessed thyroid volume by ultrasonography in Polish diabetic patients and found it to be significantly higher than that of their age-matched controls. The possible explanation for this is the higher prevalence of autoimmune thyroiditis in their study subjects and hyperinsulinism in types 1 and 2 diabetics respectively, and the proliferative effect of these on thyroid tissue. Similar observations were made by Gómez et al.[10] in Spain, Bianchi et al.[11] and Okten et al.[12] in Italy and Turkey respectively, in their studies on diabetic subjects, for reasons similar to those of the Polish study. However, Hansen et al.[13] and Darendeliler et al.[14] working in Finland and Turkey respectively did not find any significant difference in thyroid volume between diabetic patients and apparently healthy subjects. This is probably due to fewer incidence of autoimmune thyroiditis among diabetic patients studied.

In this study, thyroid volume was not statistically different between male and female diabetics. The probable implication of this finding is that gender does not influence thyroid size among diabetics. There is, however, paucity of similar studies in the literature for comparison.

No significant difference in thyroid volume was found between apparently healthy male and female controls in this study. This is contrary to the findings of Ahidjo et al.[15] in their study on thyroid volume in normal individuals in North-Eastern Nigeria, where males showed significant higher values than females. Possible explanations for this would include influence of ethnicity on thyroid size.

The biophysical characteristics age, weight, height, and duration of illness did not significantly influence thyroid volume in both male and female diabetics. This was also stated by Junik et al.[9] in their study. It can thus be inferred that age and anthropometry do not influence thyroid gland size in adult diabetics.

Conclusion

Thyroid volume is significantly higher in diabetic patient than age and sex-matched apparently healthy controls. This is not influenced by gender.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.