Abstract
Background:
This study aimed to reveal changes in thyroid gland volume and functions in patients having undergone sleeve gastrectomy (SG).
Methods:
In this retrospective cohort study, the thyroid-stimulating hormone (TSH), free triiodothyronine (fT3), and free thyroxine (fT4) levels of 48 morbidly obese patients who underwent SG operations were examined within the last 1 year before the operation and in the sixth month after the operation and then compared. The volume of the thyroid gland and, if present, nodules of patients were also calculated through ultrasonographic measurements before and after SG.
Results:
It was determined that there was a decrease in fT3 and TSH levels, but an increase in fT4 levels, in the postoperative period. There was no significant difference in thyroid volume change. In addition, it was observed that the difference in weight before and after the operation was inversely proportional to the nodule width and average nodule size (p = 0.05 and p = 0.033).
Conclusions:
We have shown that there were some changes in the thyroid function tests carried out in the sixth month following SG compared with the preoperative period, but there was no statistically significant difference in the thyroid gland and nodule volumes.
Introduction
Obesity, which is considered to be a complex and multifactorial disease that negatively affects health, is now the second most important cause of preventable deaths after smoking. Together with changes in lifestyle and the advancement of the economy, obesity has become an increasingly important health problem. Recognized as a systemic disease, obesity can cause several health problems, including disorders of thyroid gland function, type 2 diabetes mellitus (DM), hyperlipidemia, hypertension, and cardiovascular diseases. 1
In obese patients, losing weight both helps in protection of health by reducing obesity-related complications and makes an economic contribution by diminishing health expenditures. Especially in treatment of severe obesity, the fact that the success rates of permanent treatment at desired levels are low with conventional approaches such as diet, exercise, and pharmacological treatment has greatly increased the interest in bariatric surgery methods. In recent years, bariatric surgery methods have come to the forefront as a viable alternative for patients considered suitable for surgery, for whom unsatisfactory results have been obtained through conventional treatment methods, and who are evaluated from a multidisciplinary perspective.2,3
Sleeve gastrectomy (SG), a volume-restricting method, is a bariatric surgery technique that is quite commonly applied. In addition to dramatic weight loss of up to 55–70%, SG can lead to a significant decrease in metabolic disorders accompanying obesity. With this method applied in the appropriate cases, recovery can be observed in cases of type 2 DM, by 66–81%; hypertension, by 61–78%; and dyslipidemia, by 42–67%.4–8
Bariatric surgery has metabolic effects and also effects such as decrease in joint pain, regression in sleep apnea, and improvement in psychosocial functions and depression.7–10 Another effect of bariatric surgery is seen on the thyroid gland. However, the effects of bariatric surgery on thyroid gland functions and especially on its structure are not clear, and limited studies in the literature have investigated this.11–14 In a meta-analysis, it was seen that thyroid-stimulating hormone (TSH), free triiodothyronine (fT3), and total T3 levels decreased and no significant change occurred in the levels of thyroxine (T4), free thyroxine (fT4), and reverse triiodothyronine (rT3) in general after the operation. 1 As a result of weight loss, a decrease in thyroid volume is known to occur, which is correlated with the amount of weight lost. 15 In a study published in 2016, it was shown that a significant reduction occurred in thyroid volumes of patients after loss of weight through metformin treatment and with diet and exercise. 12
In this study, we aimed to show postoperative changes in thyroid gland structure and functions in patients who underwent SG, for which the literature data were limited.
Materials and Methods
The research was conducted as a single-center (Health Sciences University Keçiören Training and Research Hospital), retrospective cohort study. Ninety-seven patients older than 18 years of age, having undergone an SG operation, and who applied to the Endocrinology and Metabolic Diseases Polyclinic for routine control in the postoperative sixth month between the dates of January 1, 2017, and December 31, 2019, which was performed under approval from the Institutional Review Board of our medical center, were evaluated. Of those 97 patients, 24 without detailed thyroid ultrasonography (USG) performed by the same endocrinologist with the same USG device in the Endocrinology and Metabolic Diseases Polyclinic within the year before SG were excluded from the study. Furthermore, those with any history of neck surgery, including thyroid surgery, with a history of longitudinal radiotherapy, with a history of thyroid radioactive iodine therapy, receiving antithyroid medication (propylthiouracil or methimazole), or receiving levothyroxine replacement therapy were excluded from the study (Fig. 1).

Determination of patient groups. LT4, levotyroxin; RAI-U, radioactive iodine uptake; USG, ultrasonography.
Height; weight; body–mass index (BMI); fT3, fT4, and TSH values; preoperative thyroid USG findings; age; and gender data of the 48 volunteers included in the study (which were recorded within 1 year before SG and in the sixth month after the operation) were gathered from the hospital information system and patient file records. The same endocrinologist who had performed the first USG measurements for the volunteers in the sixth postoperative month performed a thyroid USG again using the same USG device and those findings were also recorded.
Ultrasonographic examination was performed with a Hitachi EUB-7000. In thyroid USG, gland sizes and nodule sizes, if any, were recorded. The average nodule size was calculated by the following formula: [nodule width (mediolateral length in transverse section)+nodule length (anteroposterior length in transverse section)+nodule height (mediolateral length in longitudinal section)]/3, and results are given in millimeters. Thyroid gland and nodule volumes were calculated by the following ellipsoid formula: [width (mediolateral length in transverse section) × length (anteroposterior length in transverse section) × height (longitudinal section height) × π/6], and results are given in cubic millimeters. 16 The dominant nodule dimensions of the participants with multinodular goiter were recorded.
BMI was calculated by the following formula: body weight (kilograms)/square of the height (square meters).
Thyroid function tests were measured from peripheral venous blood taken between 08:00 and 09:00 am after at least 10 h of fasting. Serum TSH, fT3, and fT4 measurements were performed using chemiluminescence on an automatic Abbott i2000 immunoassay device. The TSH reference interval is 0.27–4.3 mIU/L, the fT3 reference interval is 2.3–4.2 pg/mL, and the fT4 reference interval is 0.93–1.7 ng/dL.
The study was conducted in accordance with the Declaration of Helsinki, and written and verbal informed consent was obtained from all patients (2012-KAEK-15/1974).
Statistical analyses
Data obtained in this research were analyzed using SPSS 21 (IBM, Armonk, NY). For descriptive statistics, categorical variables are shown as numbers and percentages and numerical variables are shown as means, standard deviations (SDs), medians, minimum and maximum values, and interquartile ranges (IQRs). The compliance of numerical variables with normal distribution was assessed using the Shapiro–Wilk test. The comparison of the pre- and post-treatment data was performed with the dependent groups t-test and values summarized as median ± SD based on fulfillment of parametric conditions. The Wilcoxon test was performed and values summarized as medians and IQRs in nonparametric conditions. In correlation analyses, Pearson and Spearman correlation tests were used depending on the fulfillment of parametric conditions. Cases in which the type 1 error level was below 5% (p < 0.05) were considered statistically significant.
Results
Sample characteristics
Of the 48 patients included in the study group, 42 (87.5%) were female. The mean age of patients was 41.4 ± 11.5 years. While 20 (41.7%) of the patients had thyroid nodules, 8 had a multinodular goiter.
Weight and BMI changes
The mean preoperative weight of the study group was 121.04 ± 14.34 kg and the mean postoperative weight was 89.75 ± 13.56 kg, which indicates a significant decrease (p < 0.001). Similarly, the BMI, which was 45.37 kg/m2 preoperatively, had decreased to 33.42 kg/m2 in the sixth postoperative month (p < 0.001). When compared with their preoperative values, the postoperative weight loss of patients was at least 8 kg and at most 53 kg; the mean weight loss was 31.3 ± 9.3 kg. The mean decrease in BMI was 12.11 ± 3.62 kg/m2.
Changes in thyroid gland function
When the preoperative and postoperative thyroid function test results were compared, a mean decrease of 0.47 ± 0.47 mg/dL was observed in fT3 levels (p < 0.001), a mean decrease of 0.38 ± 0.71 mg/dL was observed in TSH levels (p = 0.001), and a mean increase of 0.99 ± 0.18 mg/dL was observed in fT4 levels (p = 0.001).
Changes in thyroid gland volume
The median thyroid volume levels were 3671.4 mm3 preoperatively and 3192.3 mm3 postoperatively. There was no significant difference in median thyroid volume change (p = 0.218) (Table 1).
Thyroid Function Tests and Thyroid Volumes Before and After Bariatric Surgery
IQR, interquartile range; fT3, free triiodothyronine; fT4, free thyroxine; SG, sleeve gastrectomy; TSH, thyroid-stimulating hormone.
Correlations
No significant correlation was observed between the patients' pre- and postoperative weights, BMI differences, thyroid function tests (fT3, fT4, and TSH), thyroid gland sizes, or thyroid gland volume differences (p > 0.05). As the pre- and postoperative weight difference increased, a decrease was observed in the nodule width difference (p = 0.05; r = −0.445) and mean nodule size difference (p = 0.033; r = −0.479) in patients with thyroid nodules. A significant correlation was particularly observed between the BMI difference and nodule width difference; this was a moderate correlation in the opposite direction of the others (p = 0.05; r = −0.444).
A positive moderate correlation (r = 0.519; p = 0.016) was found between postoperative nodule volume and preoperative fT4 values, while a negative moderate correlation (r = −0.537; p = 0.012) was found between postoperative nodule volume and postoperative TSH values (Table 2).
Correlation Analysis Between the Measurements Before and After Sleeve Gastrectomy
Moderately significant positive correlation between post-treatment nodule volume and pre-treatment fT4 values and a moderately significant negative correlation between post-treatment nodule volume and post-treatment TSH values.
Discussion
In our study, it was shown once more that SG is an effective weight loss method in the treatment of obesity and it was also revealed that it could lead to some changes in the functions of the thyroid gland and decreases in thyroid volume. A decrease in fT3 and TSH and an increase in fT4 were determined in the postoperative period. Interestingly, it was observed that the pre- and postoperative weight difference was inversely proportional to the nodule size and volume.
In many studies conducted to date, a relationship between bariatric surgery and thyroid function tests has been indicated.1,17 However, our study is the first research investigating the direct relationship of SG (which is an obesity treatment method that has been frequently practiced in recent years) with the thyroid gland volume. In our study, a decrease in thyroid volume was observed when changes from the preoperative to the postoperative period were compared. However, this decrease was not found to be statistically significant. The reason for this may be the fact that our study population consisted of a relatively small number of people. We think that this difference will be statistically significant in measurements made in a larger series. In addition, the fact that the postoperative evaluation was performed after a relatively short time (6 months) may have caused this result. In these patients, a more significant change may possibly occur in later time periods. In some studies, it was indicated that insulin resistance caused an increase in thyroid gland volume and the number and volume of nodules.18,19 This increase in volume is thought to be a result of the effect of increased TSH level mostly based on obesity as well as iodine uptake disorder due to the diet followed and hormonal changes. 18 As the reason for this, it was suggested that insulin-like growth factor 1 (IGF-1) caused the proliferation of thyrocytes with a TSH-mediated mechanism. 20 Previously, a decrease was shown in the thyroid gland and nodule volume with the reduction of insulin resistance with metformin therapy.12,21 It is known that SG is a potent resource in the treatment of insulin resistance.22,23 Therefore, we think that there is a decrease in the TSH level and consequently in the thyroid volume in the postoperative period.
A few mechanisms were suggested to explain changes in thyroid hormones following bariatric surgery. First of all, increased adipose tissue is thought to be responsible for TSH and T3, the levels of which are increased in obese people; therefore, cytokines such as increased leptin and others were found to be reduced after bariatric surgery. 24 Second, bariatric surgery is thought to be responsible for these changes by an effect other than weight loss.11,25 It was argued that this phenomenon showed similarity to the fact that remission of type 2 diabetes after bariatric surgery occurred not only with loss of weight but also with the effect of the release of some gastrointestinal hormones. 26 Third, it was specified that changes in the deiodinase level caused by energy imbalance and nutritional change after weight loss might also play a role.27,28
In obesity, leptin secretion increases due to the increased fat mass and insulin resistance. In obese people, TSH and T3 levels are found to be higher when compared with individuals with normal weight. Although not completely confirmed, it is asserted that this difference is associated with the leptin level.24,29,30 Reviewing the literature, there is some evidence indicating a correlation among leptin, negative feedback of the thyroid hormones, and thyroid-related hormone expression.31,32 In obesity, the decrease in TSH and fT3 levels due to weight loss was explained by the decrease in leptin level.33,34 The fT3 level was found to be lower in individuals with leptin deficiency. 35 In a study conducted in rats, it was shown that SG led to a decrease in leptin level. 36 In our study, we consider the leptin decrease due to the loss of fat mass as one of the mechanisms causing a low level of fT3. However, again in a study conducted with rats, no correlation could be shown between SG and the leptin level. 37 Since these studies were conducted with a limited number of animal models and they were not prospective, randomized human studies, the relationship between SG and leptin level has not yet been clarified in the literature.
In our study, it was observed that there was a decrease in TSH and fT3, but an increase in fT4, after SG. In another study carried out with 73 patients, a decrease in TSH, fT3, and fT4 levels after bariatric surgery was reported. 38 Lips et al. observed an increase in fT4 levels 3 weeks after Roux-en-Y gastric bypass surgery and a regression to normal levels 3 months after the surgery. 39 The results for TSH and fT3 in our study were similar to the findings of those studies. However, the fT4 levels in these studies were measured at different times at different levels.1,14,38,39 If we count on the level of fT4 being low in cases of obesity, it seems normal that we would detect an increase in the level of fT4 after SG. 40
Determination of contradictory fT4 levels after bariatric surgery in the literature may be due to the differences in study designs. We think that the increase in fT4 in our study may have resulted from the facts that the numbers of participants in the different studies were different, our study population included both genders, laboratory results in the sixth postoperative month were taken as the basis, and other studies included bariatric surgery methods other than SG.
Duran et al. indicated a decrease in fT3 and TSH after SG, which supports our study. 14 Neves et al. also obtained findings that were similar to ours. 17 Consequently, it may be necessary to decrease the replacement dosage in patients who receive thyroid hormone replacement therapy after bariatric surgery. We could not establish this in our study since none of the enrolled patients were receiving hormone replacement therapy. For proof of this hypothesis, further randomized controlled studies are required.
The main limitation of our study was the absence of a control group. Moreover, the number of our participants was also a limitation. Furthermore, this research was designed as a preliminary study because of its unique pattern. Another limitation is that we were not able to specify the thyroid gland heterogeneity of patients or whether they had autoimmune thyroid diseases because the relationship between autoimmune thyroid diseases and bariatric surgery has still not been revealed in the literature.
Similarly to the literature, our study population consisted of mostly female patients. 41 For this reason, our results are less representative of the male gender. We believe that the relatively smaller number of our patients with nodular goiter, our inability to state the number of nodules in individuals, and the comparison that we could make only regarding the dominant nodules have affected the results.
It has been kept in mind that some absorption problems might occur after SG, and the lack of patients receiving antithyroid therapy or thyroid hormone replacement therapy in our study increases the reliability of our findings.13,42
Conclusions
While changes may occur in the results of thyroid function tests following SG (about which we are gaining more knowledge and experience with every passing day), thyroid/nodule volume interaction is seen to be possible. It is necessary to conduct randomized controlled studies with a larger number of participants to investigate this relationship.
Footnotes
Author Disclosure Statement
No competing financial interests exist.
Funding Information
No funding was received.
