Application of thermography in the diagnostic investigation of thyroid nodules

Abstract

Thyroid nodules (TN) are common in the general population, and the clinical importance of diagnosing thyroid nodules is based on excluding the possibility of thyroid cancer, which occurs in 7–15% of cases. The thyroid gland, owing to its superficial location, is easily accessible via thermography, a noninvasive method of recording body temperature that measures infrared radiation emitted by the body surface. Therefore, this study aimed to evaluate the temperature differences between benign and malignant TN by using thermography. We conducted a cross-sectional study where 147 TN were divided into two groups: the first group included 120 benign nodules and the other included 27 malignant nodules. All the nodules were subjected to ultrasound, fine needle aspiration biopsy, and thermography. On analyzing the thermography results, the benign nodules had a higher temperature at the beginning of the thermography evaluation, and the malignant nodules showed a higher temperature in the middle and at the end (Ft). Using the relationships, it was observed that the temperature delta (ΔT), ΔT nodule/ΔT healthy, ΔT nodule minus ΔT healthy, and nodule Ft minus Ft of the healthy region were higher in malignant nodules. The ROC curve analysis of ΔT demonstrated a cutoff point of 2.38°C, with a sensitivity of 0.963 and specificity of 0.992. Malignant nodules have higher temperatures than benign nodules on thermographic evaluation. This finding suggests that thermography can be a useful tool in the diagnosis of thyroid nodules.

THYROID NODULES (TNS) are common in the general population. According to population-based studies in adults living in iodine-sufficient areas, approximately 4–7% of women and 1% of men have palpable TNs [1]. However, the prevalence of nodules on ultrasound is substantially higher, occurring in ~68% of the population [2].

Current guidelines recommend that all patients with palpable TNs undergo ultrasound examination [3, 4]. Although most TNs are benign, malignancy must be excluded. Thyroid cancer accounts for 7–15% of all TNs [5]. The American Cancer Society’s most recent estimate for thyroid cancer in the United States in 2020 was ~52,890 new cases (12,720 in men and 40,170 in women) [6]. Factors that increase the risk of thyroid cancer include a history of childhood head and neck radiation therapy or ionizing radiation exposure; history of thyroid carcinoma in the family or a first-degree relative; and rapid nodule growth or hoarseness [4].

Fine needle aspiration biopsy (FNAB) is the gold standard diagnostic tool for TNs. Although the specificity of FNAB is >95%, indeterminate FNAB results were obtained in 15–30% of cases including follicular lesions of unknown significance and follicular lesions or neoplasms [7]. These indeterminate results put the patient and the surgeon in a treatment dilemma. Therefore, the development of other preoperative diagnostic techniques for such cases can prevent unnecessary diagnostic surgery.

The thyroid gland, owing to its superficial location, is easily accessible via several diagnostic methods such as scintigraphy, ultrasound, thermography, and FNAB [8]. Thermography is a noninvasive, non-contact system for recording body temperature by measuring the infrared radiation emitted by the body surface. Typically, the thyroid does not appear on the thermogram. However, the thyroid gland is a superficial gland that emits heat easily detectable through the skin when it presents nodular or diffuse hypermetabolism [9].

Thermography can detect changes in temperature generated by malignant tumors, since the malignant tumors require nutrients during growth, resulting in the development of new blood vessels. This process, known as angiogenesis, leads to an increase in blood supply [10, 11]. An increased metabolic rate and angiogenesis in tumors can cause an increase in local temperature. In addition, the increased levels of nitric oxide produced owing to the proliferation of malignant cells results in local vasodilation, subsequently leading to heat emission [12].

As the thyroid is a superficial gland [8] and malignant tumors have a higher temperature than the surrounding area [12], the use of thermography may help in the detection of these tumors.

The objective of the current study was to compare the temperatures of benign and malignant TNs by using thermography.

Materials and Methods

Study design

This was a cross-sectional study conducted at a public university hospital.

Study population

The patients were recruited from the Endocrinology and Surgery outpatient clinics of the Hospital Universitário Antônio Pedro, Universidade Federal Fluminense, between November 2016 and February 2020. A total of 128 patients were recruited and 113 were included in the study. Of the patients included, 34 contributed with more than one nodule, since they met the inclusion criteria of the study. Accordingly, sampling was performed, in which pairing was done according to the proportion of malignant and benign nodules in the population. The sample was divided into two groups: malignant nodules (n = 27) and benign nodules (n = 120). See Fig. 1.

Fig. 1

Flowchart showing the selection of study participants

ATA, American Thyroid Association; FNAB, Fine needle aspiration biopsy

Patients of either sex were included if they were aged ≥18 years and showed the presence of ≥1 TN with a size ≥10 mm and a TSH within the normal reference range. Patients with fever and active inflammatory/infectious processes in the cervical region were excluded. In addition, patients with nodules >4 cm or nodules with a high suspicion of malignancy according to the American Thyroid Association (ATA) criteria were included only if they presented two FNAB with Bethesda Category II result [7].

Patients were included after signing an informed consent form. The study was approved by the Research Ethics Committee of the Universidade Federal Fluminense (CAAE, registered at Brazilian Ministry of Health under project number 57078516.8.0000.5243).

Methods

After being included in the study, all the patients underwent clinical and laboratory evaluations and cervical ultrasound, FNAB, and thyroid thermography. When patients showed surgical indications, they were referred for surgery.

Thyroid nodules were classified as benign or malignant according to the cytopathological results obtained on FNAB (for benign nodules) and by the histopathological result (for malignant nodules). The nodules were classified according to the Bethesda System [7]. Those with Category II results were allocated to the benign nodule group. Nodules treated with surgery and with histopathological results of malignancy were allocated to the malignant nodule group.

Patient data selected for analysis were sex, age, thyroid stimulating hormone (TSH) values, history of head and neck irradiation, thyroid antibodies (anti-thyroperoxidase and anti-thyroglobulin), ultrasonographic characteristics of TNs such as nodule size (small, <2 cm, medium, 2–4 cm, and large, >4 cm), Chammas classification on Doppler ultrasonography [13], FNAB result of TNs by the Bethesda classification and histopathological results of TNs.

Thermographic imaging

The thermographic images were obtained using a FLIR Thermal Camera SC 620 [14]. The sensitivity of the camera is <0.04°C, and the images had dimensions of 640 × 480 pixels.

Images were acquired by the following protocol. Initially, the physical conditions of the environment were assessed; the temperature in the examination room should be maintained between 21°C and 25°C, and there should be no heat source close to the procedure area. Doors and windows should be closed during the examination to avoid uncontrolled airflow in the room. In addition, there should be no objects that generate any thermal interference. Then, the parameters of the thermographic camera were configured: the camera should be positioned on the tripod and then turned on, waiting 10 minutes for its calibration. The following parameters were set before the procedure: emissivity (fixed value equal to 0.98 due to the emissivity of human skin), object temperature (axillary temperature of the patient measured with a digital thermometer), distance from the patient (about 0.6 meters, depending on the width and length of the patient’s neck), atmospheric temperature, and relative humidity (measured with the thermo-hygrometer). Recommendations were provided to the patient: should not eat or drink in the last 30 minutes before the procedure, and should not have makeup, cream, perfumes, ointments, or any substance in the neck area. In addition, the patient was asked to remove earrings and any adornment on the neck before the examination. The neck area and upper chest should be exposed. Finally, the patient should remain seated in the examination room for 10 minutes before the procedure to achieve thermal balance.

For capturing thyroid images with thermography, the following steps were performed: after measuring the patient’s axillary temperature and checking the ambient temperature and humidity, these parameters were entered in the camera software. The patient was asked to sit in a chair in front of the camera, with the head tilted back and resting on the head support. The examiner combined the center of the Camera System Box tool with the prominence of the larynx and checked whether the area of interest was in the center of the screen and the camera was parallel to the patient. The patient’s neck area was cooled evenly with a flow of air from a ventilator until the temperature indicated by the toolbox reached 31.5°C. The camera was configured to capture an image every 15 seconds for 5 minutes, resulting in a total of 20 images. The patient was asked not to move during the procedure.

After obtaining the thermographic images, the location of the nodule in the images was performed with information from the thyroid ultrasound.

Subsequently, a temperature variation in 30 points was analyzed in the region of interest. A total of 15 of them were chosen randomly from healthy regions of the neck (anterior cervical) symmetrically opposite to the nodule and the other 15 were chosen randomly from the regions identified as the nodule (Fig. 2). The images were analyzed temporally considering their initial temperature (It) [nodule and healthy region] in image 1, the medium temperature (Mt) [nodule and healthy region] in image 10, and final temperature (Ft) [nodule and healthy region] in image 20.

Fig. 2

Thermography of the anterior cervical region of a patient with a thyroid nodule. Marking of 15 points within the nodule (a) and 15 points in the healthy region of the neck (anterior cervical) symmetrically opposite the nodule (b)

Statistical analysis

For a test power of 80%, 147 TNs were selected, of which 120 were benign and 27 were malignant. Quantitative data were expressed as the median with interquartile range (IQR), and qualitative data were expressed as the percentage. The age, TSH level, temperature parameters, nodule size, and Chammas classification were compared between patients with malignant and benign nodules by using the Mann-Whitney test. The same test was used to compare the temperature parameters of the nodules according to sex. Categorical variables were compared using the chi-square or Fisher’s exact test. Using the Spearman correlation test, the association between the size of the nodule, age, TSH level and the temperature parameters were analyzed. The ROC curve was used to assess the sensitivity and specificity of the temperature delta. The level of significance was set at 0.05. All data were analyzed using IBM SPSS Statistics software version 20.0 (SPSS, Inc. Chicago, IL, USA).

Results

Participant characteristics

A total of 113 patients were included, and 147 nodules were analyzed (120 benign TNs and 27 malignant TNs); most of the participants in both the groups were female (95.8% vs. 70.4%), with no history of radiation (0% vs. 0%) and nodules <4 cm (100% vs. 85.1%). Only the group of malignant nodules had large nodules (>4 cm).

The patients were younger in the malignant nodule group (41 years vs. 58 years, p < 0.0001), had higher TSH values (2.376 mU/L vs. 1.488 mU/L, p < 0.0001), and had a higher Chammas classification (3–4 vs. 2–3, p < 0.0001). There was a trend towards a higher frequency of thyroid antibodies (anti-thyroperoxidase and anti-thyroglobulin) positivity in the malignant group of TN when compared with the benign group (22.2% vs. 9.16%, p = 0.0885). On the other hand, there was a trend towards a higher frequency of the cyst component in the group of benign nodules (18.3% vs. 3.7%, p = 0.059). The characteristics of the participants in the benign and malignant TN groups are shown in Table 1.

Table 1 Patients’ characteristics in the benign and malignant group of thyroid nodules

Characteristic	Benign nodule group (n = 120)	Malignant nodule group (n = 27)	p-value
Age (years)	58 (53–64)	41 (35–60)	0.0001**
Sex
Female, n (%)	115 (95.8%)	19 (70.4%)	0.0001*
Male, n (%)	5 (4.2%)	8 (26.6%)
TSH (mU/L)	1.488 (1.00–2.00)	2.376 (2.00–3.00)	0.0001**
Nodule size (cm)	2.00 (1.5–2.8)	2.00 (1.5–3.2)	0.242
Size classification
Small, n (%)	60 (50.0%)	11 (40.7%)
Medium, n (%)	60 (50.0%)	12 (44.4%)	0.0001*
Large, n (%)	0	4 (14.9%)
Chammas classification	2 (2–3)	3 (3–4)	0.0001**
Positive Antibodies (Anti Thyroperoxidase and Anti Thyroglobulin), n (%)	11 (9.16%)	6 (22.2%)	0.0885*

The data are expressed as the median (with interquartile range) for continuous variables and as a percentage for categorical variables (IQR: interquartile range; * p < 0.05 [chi-squared test); ** p < 0.05 [Mann-Whitney test]).

IQR, interquartile range; TSH, thyroid stimulating hormone

Among the malignant nodules, the histological types were papillary carcinoma (PTC) in 19 cases (70.4%), follicular carcinoma (FTC) in 5 cases (18.5%), and medullary carcinoma in 3 cases (11.1%).

Temperature analysis

The It was higher in the benign nodule group than in the malignant nodule group; however, the Mt, Ft, and temperature delta (ΔT) were higher in the malignant nodule group (Fig. 3). The It, Mt and Ft of the healthy region were higher in the benign group than in the malignant group.

Fig. 3

Initial temperature (It), medium temperature (Mt), and final temperature (Ft) of the nodules in the benign and malignant groups

The ΔT of the nodule divided by the ΔT of the healthy region as well as the ΔT of the nodule minus the ΔT of the healthy region and the nodule Ft minus the Ft of the healthy region were higher in the malignant nodule group. The comparison of the temperatures of the benign and malignant nodule groups is shown in Table 2.

Table 2 Comparison of the temperature parameters between benign and malignant nodules

	Benign nodule group (n = 120)	Malignant nodule group (n = 27)	p-value
It nodule (°C)	31.68 (31.36–31.95)	31.22 (30.96–31.53)	0.0001*
Mt nodule (°C)	33.01 (32.56–33.23)	33.46 (33.09–33.93)	0.0001*
Ft nodule (°C)	33.38 (33.11–33.64)	33.99 (33.90–34.48)	0.0001*
ΔT nodule (°C)	1.73 (1.50–1.91)	2.95 (2.74–3.24)	0.0001*
It healthy (°C)	30.79 (30.51–31.22)	30.52 (29.95–30.88)	0.008*
Mt healthy (°C)	32.38 (31.98–32.75)	32.06 (31.74–32.15)	0.007*
Ft healthy (°)	32.90 (32.45–31.14)	32.38 (32.16–32.60)	0.0001*
ΔT healthy (°C)	2.00 (1.66–2.14)	1.94 (1.59–2.22)	0.542
ΔT nodule/ΔT healthy (°C)	0.90 (0.79–1.01)	1.57 (1.46–1.77)	0.0001*
ΔT nodule–ΔT healthy (°C)	–0.195 (–0.41–0.02)	1.11 (0.97–1.28)	0.0001*
It nodule–It healthy (°C)	0.78 (0.55–1.04)	0.63 (0.51–0.95)	0.305
Ft nodule–Ft healthy (°C)	0.59 (0.33–0.84)	1.83 (1.59–0.47)	0.0001*

The data are expressed as the median (with interquartile range) for continuous variables. Initial temperature; Ft, final temperature; Mt, medium temperature; ΔT, temperature delta

* p < 0.05 (Mann-Whitney test).

When separating the benign and malignant nodules according to size categories (<2 cm, 2–4 cm and >4 cm), the differences in the temperature parameters found in the comparison of the whole group of benign and malignant TN persisted (Table 3). However, analyzing the benign and the malignant group of TN, no significant correlation was found between the size of the nodules and the temperature parameters. Specifically, no correlation was found between the size of the nodules and It (r_s = 0.00008, p = 0.924), Mt (r_s = 0.050, p = 0.289), Ft (r_s = 0.053, p = 0.211), ΔT nodule (r_s = 0.054, p = 0.236), ΔT nodule/ΔT healthy (r_s = 0.040, p = 0.282), ΔT nodule - ΔT healthy (r_s = 0.048, p = 0.241), It nodule - It healthy (r_s = 0.001, p = 0.982) and Ft nodule - Ft healthy (r_s = 0.062, p = 0.143).

Table 3 Comparison of the temperature parameters between benign and malignant nodules, according to the size categories of the thyroid nodules

<2 cm	Benign (n = 60)	Malignant (n = 11)	p-value
It nodule (°C)	31.66 (31.44–31.93)	31.20 (30.78–31.27)	0.0001*
Mt nodule (°C)	33.00 (32.59–33.26)	33.23 (32.87–33.45)	0.062
Ft nodule (°C)	33.38 (33.14–33.58)	33.95 (33.88–33.99)	0.0001*
ΔT nodule (°C)	1.68 (1.41–1.90)	2.77 (2.70–3.33)	0.0001*
It healthy (°C)	30.77 (30.51–31.17)	30.52 (29.90–30.71)	0.032*
Mt healthy (°C)	32.33 (32.03–32.70)	32.06 (31.73–32.15)	0.075
Ft healthy (°C)	32.85 (32.50–33.08)	32.30 (32.01–32.46)	0.018*
ΔT healthy (°C)	2.01 (1.55–2.14)	1.86 (1.80–2.21)	0.539
ΔT nodule/ΔT healthy (°C)	0.90 (0.78–1.00)	1.54 (1.46–1.64)	0.0001*
ΔT nodule–ΔT healthy (°C)	–0.20 (–0.42–0.01)	1.09 (0.86–1.14)	0.0001*
It nodule–It healthy (°C)	0.84 (0.56–1.10)	0.61 (0.52–0.69)	0.139
Ft nodule–Ft healthy (°C)	0.56 (0.35–0.86)	1.70 (1.56–1.93)	0.0001*
2–4 cm	Benign (n = 60)	Malignant (n = 12)	p-value
It nodule (°C)	31.71 (31.26–31.96)	31.20 (30.81–31.51)	0.009*
Mt nodule (°C)	33.02 (32.54–33.32)	33.59 (33.15–34.01)	0.001*
Ft nodule (°C)	33.38 (33.08–33.72)	34.17 (33.89–34.49)	0.0001*
ΔT nodule (°C)	1.78 (1.53–1.93)	3.01 (2.81–3.22)	0.0001*
It healthy (°C)	30.80 (30.55–31.35)	30.61 (29.91–30.88)	0.047*
Mt healthy (°C)	32.46 (31.92–32.81)	32.03 (31.71–32.11)	0.013*
Ft healthy (°C)	32.92 (32.35–33.17)	32.32 (32.16–32.54)	0.004*
ΔT healthy (°C)	1.97 (1.71–2.13)	1.77 (1.56–2.23)	0.934
ΔT nodule/ΔT healthy (°C)	0.90 (0.81–1.02)	1.69 (1.48–1.82)	0.0001*
ΔT nodule–ΔT healthy (°C)	–0.19 (–0.39–0.05)	1.18 (1.05–1.33)	0.0001*
It nodule–It healthy (°C)	0.72 (0.53–1.03)	0.84 (0.31–1.04)	0.745
Ft nodule–Ft healthy (°C)	0.59 (0.32–0.81)	1.89 (1.66–2.09)	0.0001*
>4 cm	Benign (n = 0)	Malignant (n = 4)	p-value
It nodule (°C)	—	31.55 (34.45–32.06)	—
Mt nodule (°C)	—	34.05 (33.85–34.65)	—
Ft nodule (°C)	—	34.69 (34.44–35.06)	—
ΔT nodule (°C)	—	2.90 (2.72–3.51)	—
It healthy (°C)	—	30.83 (30.41–31.32)	—
Mt healthy (°C)	—	32.51 (32.16–32.77)	—
Ft healthy (°C)	—	32.75 (32.60–32.98)	—
ΔT healthy (°C)	—	1.97 (1.61–2.20)	—
ΔT nodule/ΔT healthy (°C)	—	1.57 (1.46–1.75)	—
ΔT nodule–ΔT healthy (°C)	—	1.09 (0.90–1.36)	—
It nodule–It healthy (°C)	—	0.85 (0.61–1.05)	—
Ft nodule–Ft healthy (°C)	—	1.94 (1.53–2.39)	—

The data are expressed as the median (with interquartile range) for continuous variables. Initial temperature; Ft, final temperature; Mt, medium temperature; ΔT, temperature delta

* p < 0.05 (Mann-Whitney test).

Although there were significant differences in age, sex, and TSH values between the benign and malignant groups, no correlation was found between age, TSH values and temperature parameters. Specifically, no correlation was found between age and It (r_s = 0.009, p = 0.051), Mt (r_s = 0.060, p = 0.053), Ft (r_s = 0.077, p = 0.062), ΔT nodule (r_s = 0.127, p = 0.068), ΔT nodule/ΔT healthy (r_s = 0.032, p = 0.151), ΔT nodule - ΔT healthy Δ (r_s = 0.067, p = 0.063), It nodule - It healthy (r_s = 0.003, p = 0.311), Ft nodule - Ft healthy (r_s = 0.093, p = 0.077). Also, no correlation was found between TSH and It (r_s = 0.002, p = 0.637), Mt (r_s = 0.045, p = 0.064), Ft (r_s = 0.080, p = 0.081), ΔT nodule (r_s = 0.069, p = 0.071), ΔT nodule/ΔT healthy (r_s = 0.060, p = 0.078), ΔT nodule - ΔT healthy (r_s = 0.076, p = 0.091), It nodule - It healthy (r_s = 0.025, p = 0.147) and Ft nodule - Ft healthy (r_S = 0.031, p = 0.063). Finally, no difference in the temperature parameters was found when comparing females and males, respectively: It (31.21 vs. 31.27, p = 0.621), Mt (33.43 vs. 33.98, p = 0.106), Ft (33.97 vs. 34.37, p = 0.066), ΔT nodule (2.86 vs. 3.22, p = 0.163); ΔT nodule/ΔT healthy (1.57 vs. 1.57 p = 0.856); ΔT nodule - ΔT healthy (1.09 vs. 1.14, p = 0.621); It nodule - It healthy (0.61 vs. 0.88, p = 0.075) and Ft nodule - Ft healthy (1.70 vs. 2.02, p = 0.067).

Since it is known that FTC has more blood flow than PTC [15], we compared the temperature parameters of the malignant nodules according to the histological type PTC and FTC, and no significant difference was observed.

A ROC curve analysis of ΔT was performed. The results demonstrate a statistically significant curve (AUC = 0.967, EP = 0.026; p < 0.0001; 95% CI = 0.916–1,000), and evaluating the coordinates of the ROC Curve, it was noted that the cutoff point that maximized specificity and sensitivity was the ΔT of 2.38°C, with sensitivity of 0.963 and specificity of 0.992. Thus, the ΔT had a greater capacity to correctly classify those who had a benign nodule (99.2% of cases) when compared to those who had a malignant nodule (96.3% of cases). In addition, it was noted that the positive predictive value was 0.963 and the negative predictive value was 0.992 (Fig. 4).

Fig. 4

The ROC curve of the nodule ΔT

Discussion

In the current study, the malignant nodules had a higher temperature during the thermografic evaluation. In addition, the ΔT, the ΔT of the nodule divided by the ΔT of the healthy region, as well as the ΔT of the nodule minus the ΔT of the healthy region and the Ft nodule minus the Ft of the healthy region, were higher in the malignant nodule group. The ROC curve analysis of ΔT demonstrated a cutoff point of 2.38°C, with a sensitivity of 0.963 and specificity of 0.992.

The patients were younger in the malignant nodule group than in the benign nodule group (41 years vs. 58 years, p < 0.0001). This characteristic is in line with data from the literature, such as the study by Lim H et al. (2017) that demonstrated that the median age at diagnosis of thyroid cancer was 47 years [16]. In addition, the TSH values were higher in the malignant nodule group than in the benign nodule group, although the values were within the normal range (2.376 mU/L vs. 1.488 mU/L, p < 0.0001). Several studies have shown a direct correlation between serum TSH levels and the risk of malignancy of TNs, possibly because TSH stimulates cell proliferation [17-19].

The assessment of intranodular vascularization using Doppler ultrasonography, by the Chammas classification, demonstrated greater central vascularization in malignant nodules (3–4 vs. 2–3, p < 0.0001), which is in agreement with data that intranodular vascularization was correlated with thyroid malignancy [13, 20].

Regarding the size of the nodules, only the malignant nodule group presented with nodules >4 cm, possibly indicating the relationship between the nodule size and the malignancy risk [21]. The histological types observed in our study were papillary (70.4%), follicular (18.5%), and medullary tumors (11.1%), which is in agreement with the literature [22, 23].

The current study analyzed the use of thermography for 27 malignant nodules and 120 benign nodules to determine whether thermography can be used as an additional tool for the investigation of TNs, especially the detection of malignant nodules.

Thermography is a technique in which the temperature of distant objects without physical contact, by capturing infrared radiation in the electromagnetic spectrum via thermographic cameras, wherein the radiated energy of the object is captured by the camera and later converted into temperature values [24].

The analysis of the thermograms showed that the Mt (33.46°C vs. 33.01°C, p < 0.0001), Ft (33.99°C vs. 33.38°C, p < 0.0001), and ΔT (2.95°C vs. 1.73°C, p < 0.0001) were higher in the malignant nodule group than in the benign nodule group. Only the It was lower in the malignant nodule group than in the benign nodule group (31.22°C vs. 31.68°C, p < 0.0001). These differences between the Mt, Ft, and ΔT, with higher values among malignant nodules, suggest that malignant nodules emit more heat than benign nodules do.

Angiogenic conversion is a fundamental component of primary tumor growth, i.e., the tumor’s ability to promote the formation of new capillaries from preexisting blood vessels. The pattern of vascular formation in malignant tumors is anomalous: the vessels are tortuous, dilated, permeable, and branched in various configurations [10, 11]. Owing to increased blood flow, malignant tumors may have a higher temperature than the surrounding region [12, 25]; this temperature difference could be captured via thermography and used to differentiate between benign and malignant TNs.

The tendency of a higher frequency of thyroid antibodies (anti-thyroperoxidase and anti-thyroglobulin) positivity in the malignant group of TN suggests that there is a greater number of patients with autoimmune thyroiditis (AT) in this group. Since vascularization of the thyroid gland may decrease in the later stages of AT [26], this may explain the lower temperature in the healthy region (It, Mt and Ft) found in this group compared to the group of benign TN.

On comparing the size of the nodules and the temperature parameters, both in the benign and malignant TNs, there was no significant correlation between the size of the nodules and the temperature parameters. A probable explanation that may justify this finding is the presence of cystic areas, that have no vascularization. This may have contributed to the absence of temperature difference considering the size of the TNs [27]. Nonetheless, there was no significant difference in the frequency of cystic component when comparing benign and malignant TN groups.

Among all the thermographic parameters analyzed, the ΔT stood out and the ROC curve analysis demonstrated a cutoff point of 2.38°C, with a sensitivity of 0.963 and specificity of 0.992. Thus, the temperature delta could be a parameter with clinical applicability in the differentiation between benign and malignant TN.

Our results, which demonstrated a temperature difference between malignant and benign nodules, are in line with the findings of previous studies. In 1972, Samuels reported that, of the 3 nodules with a histological diagnosis of malignancy, 2 were “hot” on thermography; however, the temperature of these tumors was not described [28]. In 1974, Galli et al. reported 6 cases of histologically confirmed malignant tumors that demonstrated an evidently increased temperature on the thermogram, with values close to those recorded for hyperthyroidism; however, the temperature was not reported [29]. In a study by Kamardin et al., 201 patients with thyroid pathologies and surgical indications underwent thermography, and the correct diagnosis was made in 59 of the 66 patients with thyroid carcinoma; however, the parameter used for thermographic differentiation of the nodules was not clear [30]. Another study by D’Arbo et al. in 1988 evaluated the usefulness of thermography for the selection of TNs for surgery. A difference of 0.9°C was observed on the thermograms between benign and malignant lesions of the thyroid [31]. More recently, Aweda et al. evaluated 37 patients with thyroid disease and 16 healthy volunteers. A significant difference was observed in the temperatures of the malignant nodules (37.63 ± 0.29°C) compared to healthy patients (36.21 ± 0.73°C), although the thermographic parameters were not described again [32].

A limitation of this study was that the thickening of the neck fat layer was not evaluated. The surface temperature of the cervical region may be associated with the thickness of the fat layer owing to the thermal insulation effect of the adipose tissue [33].

Conclusions

The findings of the current study showed that malignant nodules have higher temperature than benign nodules on thermographic evaluation. These findings demonstrate that thermography could be a useful tool for the investigation of malignant TNs. However, further studies are needed to prove the effectiveness of thermography for the investigation of these nodules.

Declarations

Author disclosure statement

None of the authors have any potential conflicts of interest associated with this research.

Financial support

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES). This was not an industry-supported study.

Authors’ contributions

CAPF, MANS, AC, RACF and GABL: Conception or design of the work. CPD, JRGM, MBHM, MEOMSC, CGF, IBB, CF, MFOR and AP: Data collection. CPD, JRGM, AC, RACF and GABL: Data analysis and interpretation. CPD: Drafting the article. RACF and GABL: critical revision of the article. All the authors provided Final approval of the version to be published.

References

• 1   Vanderpump  MP,  Tunbridge  WM,  French  JM,  Appleton  D,  Bates  D, et al. (1995) The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham survey. Clin Endocrinol (Oxf) 43: 55–68.
• 2   Guth  S,  Theune  U,  Aberle  J,  Galach  A,  Bamberger  CM (2009) Very high prevalence of thyroid nodules detected by high frequency (13 MHz) ultrasound examination. Eur J Clin Invest 39: 699–706.
• 3   Rosário  PW,  Ward  LS,  Carvalho  GA,  Graf  H,  Maciel  RM, et al. (2013) Thyroid nodules and differentiated thyroid cancer: update on the Brazilian consensus. Arq Bras Endocrinol Metabol 57: 240–264.
• 4   Haugen  BR,  Alexander  EK,  Bible  KC,  Doherty  GM,  Mandel  SJ, et al. (2016) 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid 26: 1–133.
• 5   Davies  L,  Welch  HG (2014) Current thyroid cancer trends in the United States. JAMA Otolaryngol Head Neck Surg 140: 317–322.
• 6  American Cancer Society (2020). www.cancer.org. Accessed on June 25, 2020.
• 7   Cibas  ES,  Ali  SZ (2017) The 2017 Bethesda System for Reporting Thyroid Cytopathology. Thyroid 27: 1341–1346.
• 8   Burman  KD,  Wartofsky  L (2016) Thyroid nodules. N Engl J Med 374: 1294–1295.
• 9   Weber  J (1978) Thermography of thyroid body. Technique, semiology and classification of thermographic images. Acta Thermographica 3: 18–22.
• 10   Sprindzuk  MV (2010) Angiogenesis in malignant thyroid tumors. World J Oncol 1: 221–231.
• 11   Rajabi  S,  Dehghan  MH,  Dastmalchi  R,  Jalali Mashayekhi  F,  Salami  S, et al. (2019) The roles and role-players in thyroid cancer angiogenesis. Endocr J 66: 277–293.
• 12   Lee  MY,  Yang  CS (2010) Entropy-based feature extraction and decision tree induction for breast cancer diagnosis with standardized thermograph images. Comput Methods Programs Biomed 100: 269–282.
• 13   Chammas  MC,  Gerhard  R,  de Oliveira  IR,  Widman  A,  de Barros  N, et al. (2005) Thyroid nodules: evaluation with power Doppler and duplex Doppler ultrasound. Otolaryngol Head Neck Surg 132: 874–882.
• 14  Flir Systems, ThermaCamTM S45, Operator’s Manual, April 2005, plub. n° 1 55580007.
• 15   Sobrinho-Simões  M,  Eloy  C,  Magalhães  J,  Lobo  C,  Amaro  T (2011) Follicular thyroid carcinoma. Mod Pathol 24 Suppl 2: S10–S18.
• 16   Lim  H,  Devesa  SS,  Sosa  JA,  Check  D,  Kitahara  CM (2017) Trends in thyroid cancer incidence and mortality in the United States, 1974–2013. JAMA 317: 1338–1348.
• 17   Boelaert  K,  Horacek  J,  Holder  RL,  Watkinson  JC,  Sheppard  MC, et al. (2006) Serum thyrotropin concentration as a novel predictor of malignancy in thyroid nodules investigated by fine-needle aspiration. J Clin Endocrinol Metab 91: 4295–4301.
• 18   Haymart  MR,  Repplinger  DJ,  Leverson  GE,  Elson  DF,  Sippel  RS, et al. (2008) Higher serum thyroid stimulating hormone level in thyroid nodule patients is associated with greater risks of differentiated thyroid cancer and advanced tumor stage. J Clin Endocrinol Metab 93: 809–814.
• 19   Golbert  L,  de Cristo  AP,  Faccin  CS,  Farenzena  M,  Folgierini  H, et al. (2017) Serum TSH levels as a predictor of malignancy in thyroid nodules: a prospective study. PLoS One 12: e0188123.
• 20   Papini  E,  Guglielmi  R,  Bianchini  A,  Crescenzi  A,  Taccogna  S, et al. (2002) Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 87: 1941–1946.
• 21   Mehta  RS,  Carty  SE,  Ohori  NP,  Hodak  SP,  Coyne  C, et al. (2013) Nodule size is an independent predictor of malignancy in mutation-negative nodules with follicular lesion of undetermined significance cytology. Surgery 154: 730–738.
• 22   Aschebrook-Kilfoy  B,  Ward  MH,  Sabra  MM,  Devesa  SS (2011) Thyroid cancer incidence patterns in the United States by histologic type, 1992–2006. Thyroid 21: 125–134.
• 23   Wiltshire  JJ,  Drake  TM,  Uttley  L,  Balasubramanian  SP (2016) Systematic review of trends in the incidence rates of thyroid cancer. Thyroid 26: 1541–1552.
• 24   Jones  BF (1998) A reappraisal of the use of infrared thermal image analysis in medicine. IEEE Trans Med Imaging 17: 1019–1027.
• 25   Hu  L,  Gupta  A,  Gore  JP,  Xu  LX (2004) Effect of forced convection on the skin thermal expression of breast cancer. J Biomech Eng 126: 204–211.
• 26   Schulz  SL,  Seeberger  U,  Hengstmann  JH (2003) Color Doppler sonography in hypothyroidism. Eur J Ultrasound 16: 183–189.
• 27   Salabè  GB (2001) Pathogenesis of thyroid nodules: histological classification? Biomed Pharmacother 55: 39–53.
• 28   Samuels  BI (1972) Thermography: a valuable tool in the detection of thyroid disease. Radiology 102: 59–62.
• 29   Galli  G,  Salvo  D,  Troncone  L,  De Rossi  G (1974) Combined thermography and isotope scanning in thyroid pathology. Acta Radiol Diagn (Stockh) 15: 656–661.
• 30   Kamardin  LN,  Kuz’michev  AS (1983) Thermography in the differential diagnosis of nodular goiter and thyroid cancer. Vestn Khir Im II Grek 130: 70–74 (In Russian).
• 31   D’Arbo  ML,  Andrade  J,  Cherri  J,  Moriya  T,  Piccinato  CE, et al. (1988) “Role of thermography in the selection of thyroid nodules for surgical indication”. Arq Bras Endocrinol Metabol 32: 97–99 (In Portuguese).
• 32   Aweda  MA,  Adeyomoye  AO,  Abe  GA (2012) Thermographic analysis of thyroid diseases at the Lagos university teaching hospital, Nigeria. Advances in Applied Science Research 3: 2027–2032.
• 33   Speakman  JR (2018) Obesity and thermoregulation. Handb Clin Neurol 156: 431–443.