Risk Factors for Predicting Osteoporosis in Patients Who Receive Thyrotropin Suppressive Levothyroxine Treatment for Differentiated Thyroid Carcinoma

Objectives: Endogenous hyperthyroidism accelerates bone turnover and shortens the normal bone remodeling cycle, which results in reduced bone density. It is estimated that suppressive levothyroxine (LT4) therapy also decreases bone density. The aim of this study was to define risk factors for osteoporosis development in patients under thyrotropin-stimulating hormone (TSH) suppressive treatment for differentiated thyroid cancer (DTC). Methods: Patients with a diagnosis of low or intermediate risk group DTC according to the American Thyroid Association 2015 guidelines and who have been receiving LT4 suppression therapy and were physically fit to undergo femur and lumbar vertebra bone density study were included in the study. Patients lacking information on demographic data, medical history, preoperative thyroid hormone status, or routine follow-up data were excluded from the study. A study form consisting of patient information on possible risk factors for osteoporosis such as gender, age, menopausal status, smoking, family history of osteoporosis, preoperative thyroid hormone status, postoperative hypoparathyroidism history, mean serum TSH levels, and duration of TSH suppression was created and filled out for each participant. Bone mineral densitometries of the femur and lumbar vertebrae were measured along with serum vitamin D and parathyroid hormone levels. Results: During TSH suppression (mean 7.2±4.5 years, range: 1-26), osteoporosis was detected in 89 (9.6%) patients. The mean time to develop osteoporosis was significantly different in patients with or without a family history of osteoporosis (15.3±0.4 versus 20.3±0.6 years; p=0.002). Similarly, the mean time to develop osteoporosis for was found to be significantly shorter in postmenopausal patients than that for premenopausal women (18.6±0.7 versus 20.4±0.4 years; p<0.001). Male gender (p<0.001), a family history of osteoporosis (p=0.001) and menopausal state (p<0.001) were identified as independent predictive factors for developing osteoporosis. Conclusion: Postmenopausal women, men, and patients with a family history who receive TSH-suppression treatment have a tendency to develop osteoporosis.


Introduction
Differentiated thyroid carcinoma (DTC) is the most common endocrine neoplasia. Although the incidence of DTC is increasing its mortality rate remains stable (1,2). After initial treatment with a thyroidectomy with/without radioiodine treatment, patients are treated with levothyroxine (LT4) therapy to suppress thyrotropin-stimulating hormone (TSH) since suppression of serum TSH levels reduces tumor recurrence rates (3). Endogenous hyperthyroidism has been shown to reduce bone density because hyperthyroidism accelerates bone turnover and shortens the normal bone remodeling cycle (4). For this reason, suppression LT4 therapy might cause a decrease in bone density. Considering the long life expectancy for DTC patients, treatment related comorbidities could affect their quality of life.
Although several studies have been designed to explore the correlation between bone density changes and LT4 treatment, conflicting findings have been reported (4,5). Most of the reported studies have included small number of patients, and despite this limitation bone density seems to decrease in at least some DTC patients. In this large series, we analyzed DTC patients' bone density after considering several demographic features, comorbidities, and treatment-related risk factors. In our analysis, we aimed to define additional risk factors for developing osteoporosis in patients who received TSH-suppressive LT4 treatment for DTC.

Patients
This study included prospective and retrospective components and was approved by the Institutional Ethical Committee of Ankara University Medical Faculty (approval number: 11-489-16). After receiving informed consent for the prospective component, patient inclusion was continued for the period between June 2016 and Jan 2018. All patients had received radioiodine treatment for DTC in Ankara University Medical Faculty Department of Nuclear Medicine. Patient inclusion criteria were based on several parameters: (1) low or intermediate risk group DTC diagnosis according to the American Thyroid Association (ATA) 2015 guidelines; (2) receiving LT4 suppression therapy after initial treatment; (3) available routine followup data after initial therapy; (4) known preoperative thyroid hormone status; (5) sufficient and available demographic and medical history data to fill-in the study form; and (6) be physically fit enough to undergo a femur and lumbar vertebra bone density study (6).

Statistical Analysis
The differences in proportions between groups were compared by using chi-square test. The survival estimations were performed using the method of Kaplan-Meier algorithm, and the comparison between groups was evaluated with the log-rank test. Multiple Cox regression proportional hazard model was used to determine independent predictors of osteoporosis development (7). p value less than 0.05 was considered as significant. SPSS version 20.0 (IBM, Chicago, Illinois, USA) was used for statistical analyses.

Patients
A total of 929 patients (813 female, 116 male, mean age: 52.33±7.2) who received TSH suppression therapy for DTC were included. Patient descriptive data are presented in Table 1.

Risk Factors to Development of Osteoporosis
During TSH suppression (mean 7.2±4.5 years, range: 1-26), osteoporosis was detected in 89 (9.6%) patients. The rate of osteoporosis in patients with and without a family history of osteoporosis was 13% and 8%, respectively. Osteoporosis detection rates were calculated as 0.6%, 15%, and 12% in premenopausal and postmenopausal women and men, respectively. Preoperative hyperthyroidism was found to be significantly correlated with the presence of osteoporosis based on chi-squared analysis with 15% versus 8% (p=0.003); however, this significance was lost in multivariate Cox regression analysis. Although mean serum TSH levels were not significant factors for the presence of osteoporosis, osteoporosis detection rates seem to decrease in patients with TSH levels >0.4 mIU/L. Osteoporosis rates for different risk groups are summarized in Table 2.   Table 3. The mean time to develop osteoporosis among patient groups according to risk factors was not significant in Kaplan-Meier analysis (p>0.05). Kaplan-Meier curves for different groups are presented in Figures 1A, 1B, 1C (Table 4). According to multiple Cox regression proportional hazard analysis, other risk factors were not found to be significant (p>0.05).

Discussion
It is considered that remnant DTC cells behave in a manner similar to benign thyrocytes from which they originated. TSH stimulates the number, size and activity of thyrocytes (8). The rationale for this approach was based on observations that the incidence of thyroid cancer is correlated with serum TSH levels in the normal population (9). Hence, the primary aim of TSH suppression therapy is to lower endogenous TSH levels to reduce the risk of disease recurrence.
In the literature, different outcomes have been reported concerning the benefits of long-term TSH suppression in DTC patients. Lower serum TSH levels have been shown to be an independent predictor for disease progression in patients with a high risk of tumor recurrence. Interestingly, a similar effect has not been demonstrated on patients with stage 1 or 2 disease (10). Moreover, a meta-analysis including 10 studies did not demonstrate any benefits  (12). Similarly, Hovens et al. (13) have reported results of 366 patients treated with total thyroidectomy followed by radioiodine treatment. They found that serum TSH levels >4.5 mU/L was an independent predictor for death, and TSH levels >2 mU/L were also associated with DTC-related deaths and recurrence in patients with T1-3, M0 tumors. Also, Pujol et al. (14) reported that suppressed serum TSH levels were associated with an increase in relapse-free survival in patients with DTC.
Controversial results based on these analyses have led to discussions about optimal TSH level and duration of suppression for low-intermediate risk group patients with respect to therapy-related side effects. Thyroid hormones act directly on the skeleton, and endogenous hyperthyroidism is known to be related with a high risk of osteoporosis (15). Known risks of iatrogenic overt or subclinical hyperthyroidism are osteoporosis, osteopenia, and/or atrial fibrillation. For this reason, slightly subnormal or normal TSH levels are recommended for long term periods (16). Two cohort studies have demonstrated that postmenopausal DTC patients with fully suppressed TSH levels have a high risk of osteoporosis (17,18). In our study, we aimed to analyze additional risk factors for developing osteoporosis in a large cohort. A family history of osteoporosis and menopausal status were found to be significant factors favoring osteoporosis development. The presence of preoperative hyperthyroidism could also be another risk factor. Interestingly, we could not find any significant correlation between mean serum TSH levels and presence of osteoporosis. However, osteoporosis detection rates tend to decrease in patients with mean TSH levels >0.4 mIU/L.  Another interesting finding of our analysis was that male patients who received TSH suppression therapy were found to have a 20-fold increase in developing osteoporosis as compared to premenopausal women. Most studies have included only female patients for osteoporosis analysis.
Reverter et al. (19) analyzed bone mineral densities and bone fractures in male patients receiving long-term TSH suppressive therapy. They compared bone mineral density and bone turnover parameters from 33 DTC patients with age-and body mass index-matched control groups. They did not find any significant differences between bone turnover parameters, including the T and Z scores, between groups.
We could not compare our study parameters with an agematched group. However, the number of included male patients in this study was higher than that in the previous study. A total of 66 patients and 67 controls were included in a recent meta-analysis on the effects of TSH suppression in men. The authors did not find any significant correlation between TSH suppression and lower BMD values in men (20). For this reason, the osteoporosis rate in male patients receiving TSH suppression therapy needs further clarification with prospective randomized control studies.
The ATA 2015 guidelines recommend 0.1-0.5 mU/L levels as an initial TSH goal for low-risk group patients with indeterminate or incomplete response as well as for intermediate risk group patients based on data on this subject. The guide recommends continuation therapy with 0.5-2.0 mU/L levels for low and intermediate risk group patients with excellent response. It is reasonable for clinicians to consider disease stage, response to initial treatment, and personal risk factors to develop osteoporosis in order to personalize a patient's TSH suppression therapy. The risk of disease recurrence and TSH suppression-related risks should be balanced. Postmenopausal women, men, and patients with a family history of osteoporosis have a high rate of osteoporosis under TSH-suppression LT4 treatment. Preoperative hyperthyroidism and mean serum TSH levels seem to be possible predictors of developing osteoporosis, although not statistically significant.

Conclusion
The current data suggest that personalized TSH suppression treatment, based on DTC risk group and patient-related risk factors to develop osteoporosis, might be beneficial. The authors declared that this study received no financial support.