A Neglected Point: Frailty in Older Adults with Differentiated Thyroid Cancer
Article information
Abstract
Background
This study investigated the risk of frailty in older adults with differentiated thyroid cancer (DTC) and the effect of thyroid-stimulating hormone (TSH) levels on frailty.
Methods
This single-center, cross-sectional study included 70 DTC patients aged ≥60 years with stable TSH levels during the previous year while receiving levothyroxine. Frailty was assessed using the fried frailty phenotype (FFP). Anterior thigh muscle thickness was measured by ultrasound, and the sonographic thigh adjustment ratio (STAR) index was calculated. Muscle strength was measured using a hand dynamometer. Physical activity was determined by the physical activity scale for the elderly (PASE).
Results
The median (interquartile range) age and follow-up time were 65 years (62 to 71) and 11 years (7.0 to 14.2), respectively. The median TSH level was 1.10 μIU/mL (0.49 to 1.62), and 58.6% of patients were prefrail/frail. Muscle mass and strength were reduced in 35.7% and 17.2% of patients, respectively. TSH levels were lower in those with prefrailty/frailty (P=0.002), low muscle mass (P=0.014), and low strength (P=0.037) than in their normal counterparts. TSH levels correlated negatively with FFP (P=0.001) and positively with the STAR index (P=0.034). TSH below 1.325 μIU/mL was associated with an increased frailty risk (area under the curve=0.719; P=0.001). Low TSH, female sex, low handgrip strength, and low PASE leisure time scores emerged as independent predictors of frailty (P<0.05).
Conclusion
Older adults with lower TSH levels due to DTC are at high frailty risk and have low muscle mass and strength. Therefore, TSH targets should be set based on a comprehensive evaluation with consideration of the risk-benefit ratio.
INTRODUCTION
Differentiated thyroid cancer (DTC) is the most prevalent malignancy of the endocrine system, and its incidence has increased recently due to advancements in diagnostic techniques. Patients with DTC are administered levothyroxine (LT4) following thyroidectomy to achieve euthyroidism and suppress thyroid-stimulating hormone (TSH) production [1]. The intensity and duration of TSH suppression are guided by postoperative risk assessments [2]. According to the American Thyroid Association (ATA) guidelines, setting target TSH levels based on postoperative risk classification and treatment response, and adjusting LT4 treatment accordingly, can reduce the risk of disease recurrence and metastatic spread [1]. As life expectancies increase, there is growing concern about the potential side effects of prolonged TSH-targeted LT4 substitution therapy, particularly in older adults monitored for DTC following ATA guidelines.
Frailty is a clinical condition characterized by an organism’s increased vulnerability to stressors, which leads to adverse health outcomes such as falls, delirium, and disability [3]. This condition often arises from a gradual decline in physiological reserves due to aging, influenced by genetic and environmental factors [4,5]. Clinically, frailty is assessed by indicators such as self-reported exhaustion, weight loss, reduced physical activity, slow walking speed, and low grip strength [6]. The relevance of frailty becomes more pronounced as life expectancy increases [3]. Additionally, older men with subclinical hyperthyroidism are at a heightened risk of frailty, likely due to diminished physical performance and muscular strength [7]. Other contributing factors to frailty include malnutrition, cognitive impairment, depression, loss of independence in daily activities, and low physical activity levels. It is crucial to identify and address the causes of frailty through comprehensive geriatric assessment [8]. Sarcopenia is regarded as one of the most significant biological risk factors for frailty [9]. The European Working Group on Sarcopenia in Older People 2 (EWGSOP2) defines sarcopenia as a progressive and widespread skeletal muscle disorder that increases the risk of adverse outcomes such as falls, fractures, physical disability, and death [10]. The diagnosis of sarcopenia is established by demonstrating low muscle mass along with low muscle strength [11].
In the literature, the relationship between TSH suppression and both low muscle strength and sarcopenia in patients with DTC has been evaluated [12]. However, to the best of our knowledge, no study has investigated the relationship between TSH levels and the risk of frailty in older adults with DTC. Therefore, the aim of this study was to assess the risk of frailty in this population and to determine the effect of TSH levels, as categorized by risk status, on frailty.
METHODS
Identification of participants
This study was designed as a single-center, multidisciplinary, cross-sectional analysis. It included patients aged 60 years and older who visited an endocrinology outpatient clinic for routine DTC management, had undergone a total thyroidectomy, and had maintained stable TSH levels for at least the past year following a minimum of 2 years of follow-up. Stability in TSH levels was defined as having TSH values within the target range for the patient’s specific risk group, as determined by the ATA thyroid carcinoma risk classification, without any changes in their hormone replacement dose over the last year. Patients were categorized based on the ATA thyroid carcinoma risk classification, and their treatments were tailored accordingly. The TSH level, along with free triiodothyronine (fT3) and free thyroxine (fT4) levels, was calculated by averaging three measurements taken over the previous year. Exclusion criteria included patients with overt hyperthyroidism or hypothyroidism, hypocalcemia, neuromuscular diseases, a history of major orthopedic surgery, osteoporosis or osteoporotic fractures, psychiatric disorders including depression, any malignancy other than thyroid carcinoma, heart failure, chronic lung, liver, or kidney disease, hypogonadism, and those receiving glucocorticoid therapy, testosterone replacement therapy, or any other medications known to influence sarcopenia and frailty.
Frailty assessment
A comprehensive geriatric assessment, which included evaluations for frailty and sarcopenia, was conducted by three geriatricians. Frailty was assessed using the fried frailty phenotype (FFP), a method that is both valid and reliable [6,8]. The FFP consists of five criteria: self-reported exhaustion, weight loss, reduced physical activity, slow walking speed, and low grip strength. Individuals exhibiting one or two of these criteria were classified as prefrail, while those exhibiting three were classified as frail. Those who did not meet any of the criteria were defined as robust [6,8].
Anthropometric measurements and sarcopenia assessment
Height and weight were measured using a Tanita BC-418 MA Body Composition Analyzer (TANITA Corp., Tokyo, Japan). Body mass index (BMI) was calculated by dividing weight in kilograms by the square of height in meters. Measurements of muscle strength and muscle mass were conducted to assess sarcopenia, following the EWGSOP2 criteria. Handgrip strength (HGS) was measured using a Jamar hydraulic hand dynamometer. Three measurements were taken from the dominant hand, and the highest value was recorded. According to the EWGSOP2 criteria, HGS values below 27 kg for men and 16 kg for women indicated low muscle strength [13]. For the detection of low muscle mass, the thickness of the anterior thigh muscles (rectus femoris and vastus intermedius) was measured. This was done at the midpoint between the anterior superior iliac spine and the upper end of the patella using a GE Logiq P5 ultrasonography machine (GE HealthCare, Chicago, IL, USA). The measurements were performed by a geriatrician certified in musculoskeletal ultrasonography (Esra Cataltepe). To ensure accuracy, three ultrasound images were taken for each measurement, and the average value was recorded. The sonographic thigh adjustment ratio (STAR) index was calculated by dividing the anterior thigh thickness by the BMI. According to the STAR index, values below 1.4 for men and 1.0 for women were indicative of low muscle mass [14]. A diagnosis of sarcopenia was made when both low muscle strength and muscle mass were present [11].
Comprehensive geriatric assessment
All participants underwent a comprehensive geriatric assessment that included evaluations of function, cognition, mood, and nutrition. Cognitive function was assessed using the Mini-Mental State Examination (MMSE), where a score below 24 out of 30 indicated cognitive impairment [15]. Nutritional status was evaluated with the Mini Nutritional Assessment-Short Form (MNA-SF); a score below 12 signified nutritional risk [16]. Mood was assessed for depression using the Yesavage Geriatric Depression Scale (GDS), with a score of 7 or higher suggesting a risk of depression [17]. Functional capacity was measured using the Katz index of independence in Activities of Daily Living (ADL) and the Lawton-Brody Instrumental Activities of Daily Living (IADL) scales. The ADL was scored on a 6-point scale, and the IADL on an 8-point scale, with lower scores indicating greater dependency in daily activities [18,19]. Physical activity levels, including leisure time, household, and work activities, were assessed using the physical activity scale for the elderly (PASE) questionnaire. Both the PASE-leisure time score and the PASE-total score, which is the sum of the scores for leisure time, household, and work activities, were calculated and recorded [20].
Ethical considerations
This study was conducted in accordance with the principles outlined in the Declaration of Helsinki. The study protocol received approval from the Ethics Committee of Gazi University on August 8, 2022 (approval number 650; 08.08.2022-650). Written informed consent was obtained from all participants.
Statistical analysis
IBM SPSS Statistics 22.0 (IBM Co., Armonk, NY, USA) was utilized for statistical analyses, MedCalc statistical software for receiver operating characteristics (ROC) curve analysis, and R package program version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria) for graphical representation. The normality of continuous variables was assessed using the Shapiro-Wilk test, and the homogeneity of variance was evaluated with the Levene test.
Continuous variables with a normal distribution were presented as the mean±standard deviation. Those not normally distributed were presented as the median and interquartile range (IQR). The Pearson chi-square test was used for categorical data, the Student t test for normally distributed continuous data, and the Mann-Whitney U test for data that were not normally distributed. Spearman correlation analysis was conducted to explore potential correlations.
The risk of frailty was predicted by analyzing the TSH value using ROC curve analysis. The results, including the area under the curve (AUC), optimal cut-off values based on the Youden index, and measures of sensitivity and specificity, were presented.
Univariate binary logistic regression analysis was conducted to evaluate the effects of age, sex, disease duration, the presence of hypertension (HT) and diabetes mellitus (DM), BMI, HGS, TSH levels, mental status as measured by the MMSE, nutritional status assessed using the MNA-SF, depression risk score based on the GDS, leisure time physical activity measured by the PASE, and the STAR index for the risk of frailty. Variables that exhibited a significant association with frailty in the univariate analysis were subsequently included in the multivariate logistic regression analysis.
P values <0.05 were deemed statistically significant. A post hoc power analysis of the study was conducted using the G* Power 3.1 program (Heinrich Heine University Düsseldorf, Düsseldorf, Germany). Considering the TSH levels in the prefrail/frail and robust groups, with a margin of error set at 0.05, a sample size of n1=41 and n2=29, and an effect size of 0.71 according to Cohen’s d method, the power of the study was determined to be 0.80.
RESULTS
Seventy older adults with DTC were included in the study. The median age was 65 years (IQR, 62 to 71), 64.3% of the patients were female, 74.3% had HT, and 32.9% had type 2 DM. All of the patients had normal calcium (median, 9.50 mg/dL [IQR, 9.2 to 9.9]), albumin (median, 4.30 g/dL [IQR, 4.17 to 4.40]), phosphorus (median, 3.70 mg/dL [IQR, 3.30 to 4.10]), and creatinine (0.85±0.23 mg/dL) levels.
In all patients, fT3 (3.09±0.32 pg/mL) and fT4 (1.11±0.13 ng/dL) values were within the normal range. The median weight-based LT4 dose was 1.37 µg/kg/day (IQR, 1.17 to 1.58) and the median TSH was 1.10 µIU/mL (IQR, 0.49 to 1.62). The median follow-up was 11 years (IQR, 7.0 to 14.2). The percentage of patients with low TSH (<0.5 µIU/mL) was 27.1%. The TSH level ranged between 2.0 and 4.0 µIU/mL in 12.9% of patients, who were considered to be at low risk in the initial evaluation, exhibited an excellent response to treatment, showed no risk of biochemical and structural disease, and had a follow-up period of more than 10 years. In the first risk assessment following surgery, 65.7% of the cases received radioactive iodine treatment as they were categorized in the intermediate or high-risk groups. The most recent patient received treatment 2 years ago. TSH levels were between 0.5 and 2.0 µIU/mL in 60.0% of patients initially evaluated as intermediate or high risk who had an excellent response to treatment, were clinically and biochemically free of disease, and had a low recurrence risk. TSH levels were between 0.1 and 0.5 µIU/mL in 22.9% of the patients initially evaluated as high risk with an incomplete biochemical response to therapy, indicated by high thyroglobulin levels or a tendency to increase despite no proven structural disease. TSH was suppressed to <0.1 µIU/mL in 4.3% of the patients initially evaluated as high risk with an incomplete structural response to therapy. Among high-risk patients with an incomplete structural response to therapy, there was no evidence of distant organ metastasis, and the disease was confined to the local region. Two patients with locoregional disease underwent a second operation after their initial diagnosis. One patient declined the second operation due to the potential risk of complications related to the proximity of the residual tissue to the recurrent laryngeal nerve. These three cases were considered high-risk, and the goal of TSH suppression was maintained. The results of muscle mass and muscle strength, along with a comprehensive geriatric assessment, are presented in Table 1.
Factors Affecting Frailty in Older Adults with Differentiated Thyroid Carcinoma
According to FFP, 1.4% of the patients were frail, 57.2% were prefrail, and 41.4% were robust. Given that only one patient was frail, the analysis compared the combined prefrail/frail group (58.6%) with the robust group. As shown in Table 1, there were no significant differences between the groups in terms of age (P=0.443), disease duration (P=0.155), the prevalence of DM (P=0.580) and HT (P=0.078), vitamin D levels (P=0.208), fT4 levels (P=0.570), fT3 levels (P=0.490), nutritional risk (P=0.937), cognitive functions (P=0.735), depression risk (P=0.883), and independence in daily living activities (P=0.563). No differences in frailty were observed between the groups when patients were categorized by age: those under 65 years and those 65 years and older, as well as those under 70 years and those 70 years and older, with P values of 0.518 and 0.785, respectively. However, the prefrail/frail group exhibited a lower TSH value compared to the robust group (P=0.002). Additionally, this group had higher percentages of women (P=0.019), higher BMI (P=0.036), and a higher prevalence of patients with low TSH levels (<0.5 µIU/mL) (P=0.035), low muscle mass (P=0.003), low muscle strength (P=0.001), and sarcopenia (P=0.007). The classification of DTC based on initial risk status and treatment response also differed between the prefrail/frail group and the robust group (P=0.010). Furthermore, the PASE-leisure time score (P=0.006), STAR index (P=0.003), and HGS (P=0.019) were lower in the prefrail/frail group than in the robust group.
The association between muscle mass, muscle strength, sarcopenia, and TSH levels was evaluated across all groups. The TSH levels in the group with low muscle mass, measured at 0.71 µIU/mL (IQR, 0.24 to 1.31), were significantly lower than those in the group with normal muscle mass, which recorded levels of 1.20 µIU/mL (IQR, 0.68 to 1.76) (P=0.014). Similarly, the TSH levels in the low muscle strength group, at 0.66 µIU/mL (IQR, 0.24 to 1.21), were significantly lower than those in the normal muscle strength group, which were 1.18 µIU/mL (IQR, 0.51 to 1.68) (P=0.037). Although TSH levels were lower in patients with sarcopenia, at 0.61 µIU/mL (IQR, 0.19 to 1.26), than in those without sarcopenia, at 1.12 µIU/mL (IQR, 0.50 to 1.66), this difference was not statistically significant in older adult patients with DTC (P=0.075).
As shown in Table 2, the risk of frailty was positively correlated with BMI (P=0.046) and negatively correlated with TSH (P=0.001), HGS (P<0.001), and muscle mass (P=0.007). The STAR index showed a positive correlation with TSH (P=0.034), PASE total score (P=0.025), and HGS (P<0.001), and a negative correlation with BMI (P<0.001). Additionally, muscle strength was negatively correlated with BMI (P=0.002); however, there was no significant correlation between muscle strength and TSH (P=0.127).
Correlation Analysis between Frailty, Clinical Markers of Sarcopenia, and Risk Factors in Older Adults with Differentiated Thyroid Carcinoma
Upon analyzing the TSH value to predict the risk of frailty using ROC curve analysis, it was found that patients with a TSH below 1.325 µIU/mL exhibited an increased risk of frailty. The sensitivity and specificity were 80.5% and 58.6%, respectively (AUC=0.719; P=0.001) (Fig. 1).
Receiver operating characteristic (ROC) analysis of thyroid-stimulating hormone (TSH) levels for the risk of frailty. AUC, area under the curve.
As shown in Table 3, variables that demonstrated a significant association with frailty risk in the univariate analysis were subsequently included in the multivariate logistic regression analysis. The findings from the multivariate logistic regression analysis revealed that low TSH levels, female sex, reduced HGS, and a low PASE leisure time score were independently linked to an increased risk of frailty (P<0.05).
Logistic Regression Analysis of Factors Related to Frailty Risk in Older Adults with Differentiated Thyroid Carcinoma
DISCUSSION
To the best of our knowledge, this is the first study to explore the relationship between TSH levels and frailty in older adults with DTC who are on LT4 substitution therapy. In the group identified as prefrail or frail, which also exhibited low muscle mass and strength, TSH levels were significantly lower than those in the non-frail group. Furthermore, a significant negative correlation was observed between TSH levels and the risk of frailty. More importantly, relatively low TSH levels were identified as an independent risk factor for frailty, alongside being female, having low HGS, and engaging in low levels of leisure-time physical activity.
In the present study, muscle mass was assessed by measuring the anterior thigh muscle thickness with ultrasound and calculated using the STAR index. Patients with low muscle mass exhibited lower TSH levels compared to those with normal muscle mass. Similarly, median TSH levels were lower in patients with diminished muscle strength than in those with normal muscle strength. These findings align with existing literature that has demonstrated a link between low TSH levels and reduced muscle strength in both middle-aged and older adult patients diagnosed with DTC [12,21]. It is noteworthy that a prior study assessing appendicular muscle mass did not find a correlation between low TSH levels and reduced muscle mass in older adult patients with DTC [12]. However, in our study, the use of ultrasound to evaluate muscle mass, a relatively novel approach, may have enhanced our ability to detect low muscle mass. This method offers benefits in terms of accuracy, accessibility, and minimal patient burden.
In the present study, muscle mass was positively correlated with TSH and PASE scores. Our study also demonstrated a positive impact of increased physical activity on mid-thigh muscle mass despite the negative effect of low TSH levels. Conversely, we observed a negative correlation between BMI and both HGS and the STAR index. This relationship between higher BMI and decreased muscle mass and strength indicates the presence of sarcopenic obesity [22]. Sarcopenic obesity is a recently identified form of obesity marked by low muscle mass, diminished muscle strength or quality, and increased adiposity [23]. The coexistence of sarcopenia and obesity is associated with heightened risks of metabolic dysfunction, frailty, and mortality [24,25]. Furthermore, our study demonstrated a positive association between BMI and frailty, aligning with findings reported in the existing literature [26].
While there are limited studies evaluating the effects of stable TSH levels on muscles in DTC, to our knowledge, no studies have assessed the risk of frailty [27]. In our older adult patients with DTC, those in the prefrail/frail group had lower TSH levels compared to the robust group. Additionally, there was a negative correlation between TSH and FFP. A previous study indicated that men with low TSH values exhibited higher levels of muscle weakness and exhaustion, whereas women with low TSH values were more likely to meet all Fried criteria except for exhaustion [28]. In our study, female sex was identified as a risk factor for frailty in regression analyses, aligning with existing literature [9,29]. Although DM and HT have been reported as predictive factors for frailty in other studies, our study did not find an independent association in univariate analyses [30]. This lack of association might be attributed to our sample size. Additionally, low muscle mass was observed in the prefrail/frail group. However, the effect of low muscle mass was not significant in multivariate regression analysis. This could be due to decreased anterior thigh muscle thickness being one of the early signs of sarcopenia [14]. Instead, as expected, low muscle strength and low physical activity, which are also indicators of sarcopenia severity [13], were identified as strong risk factors for frailty. Exercise and a high level of daily activity are protective factors against both sarcopenia and frailty [31,32].
The Rotterdam study identified a U-shaped relationship between the baseline frailty index and TSH [33], a finding that our study also supports. Our research indicates that TSH is an independent factor that increases the risk of frailty in older adults with DTC. Specifically, the risk of frailty is higher in patients with a TSH value below 1.325 µIU/mL. These results underscore the impact of even mildly low TSH levels on frailty in older patients with DTC. The long-term TSH monitoring targets for patients are determined based on their postoperative risk classification and response to treatment. During long-term follow-up, it is advisable to consider the patient’s risk status, as well as the presence of osteoporosis, arrhythmia, and increasing age (over 60), when setting TSH targets [1]. Do Cao and Wemeau [34] conducted a thorough evaluation of the risk-benefit ratio of long-term TSH suppression in DTC. They recommended adopting more flexible TSH cut-off values, especially for patients at high risk of cardiac side effects, osteopenia, and fractures [34]. Our findings elucidate the relationship between TSH and the risk of frailty in adults with DTC. Consequently, we recommend that older adult patients with DTC undergo both a frailty risk assessment and a comprehensive geriatric assessment as part of their long-term evaluation. If these patients are identified as being at high risk of frailty, caution should be exercised in administering TSH substitution therapy. It is advisable to maintain TSH levels near the upper limit of the specified values, taking into account the patient’s risk group.
Our study had some limitations. We focused on older adult patients with DTC who had maintained stable TSH levels over the past year and had no additional risk factors for frailty and sarcopenia. The stringent inclusion criteria resulted in a reduced number of participants, which may have decreased the statistical power of our findings. Additionally, the cross-sectional design of our study precluded the collection of prospective data. Another limitation is the variability in the duration for which patients had been taking LT4 before their evaluation for frailty. Disease duration was defined as the time elapsed since the pathological diagnosis following thyroidectomy, which was longer than that in previous studies examining the link between thyroid dysfunction and frailty [7,35]. Although the patients in our study had not experienced any changes in their LT4 dosage for at least 1 year and their TSH levels had remained within the same reference range based on risk classification, this could still represent a residual confounding factor in assessing frailty risk. Furthermore, the patients underwent a lengthy follow-up period, which resulted in a limited number of patients undergoing active TSH suppression. Nevertheless, our assessment of frailty risk based on TSH levels indicated a high risk even within the lower half of the normal TSH range.
In conclusion, to the best of our knowledge, our study is the first to explore the relationship between TSH levels and frailty in patients with DTC. Our results indicate that patients with DTC who maintain long-term, relatively low TSH levels may have reduced muscle mass and strength, and an elevated risk of frailty. Consequently, a comprehensive assessment is essential during the long-term follow-up of these patients. Furthermore, when establishing TSH targets for older adults with DTC at high risk of frailty, it is crucial to carefully weigh the risks and benefits.
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: M.C., E.C., H.D.V., E.C., Y.B., Y.K., M.M.Y., M.A., F.B.T., M.A.K., A.E.A. Acquisition, analysis, or interpretation of data: M.C., E.C., E.C., Y.B., Y.K. Drafting the work or revising: M.C., E.C., H.D.V., E.C., Y.B., Y.K., M.M.Y., M.A., F.B.T., M.A.K., A.E.A. Final approval of the manuscript: M.C., E.C., H.D.V., E.C., Y.B., Y.K., M.M.Y., M.A., F.B.T., M.A.K., A.E.A.
Acknowledgements
The authors thank all the study participants.