Is subclinical hypothyroidism associated with cardiovascular disease in the elderly?

Introduction

Subclinical hypothyroidism (SCH) is biochemically defined as a state of an elevated serum thyrotropin (TSH) level and a normal free thyroxine (FT4) level (1). A cross-sectional study in the United States showed that the prevalences of overt hypothyroidism and SCH are 0.4 and 9%, respectively, and that the prevalence of SCH increases to more than 20% in women aged 75 years or older (2, 3). Patients with hypothyroidism, including SCH, may have risks for coronary artery disease events, heart failure and mortality (1, 4, 5). A study using data from the National Health and Nutrition Examination Survey in the United States revealed that SCH was associated with all-cause mortality mediated by cardiovascular disease (CVD) (6). In line with that study, we define CVDs as coronary artery diseases and nonfatal CVD events such as congestive heart failure, angina, myocardial infarction or stroke in this review. Reanalysis of the Whickham survey cohort, which is a historical cohort study in which the association between incident ischemic heart disease (IHD) events and IHD-related mortality was investigated with SCH followed up for a period of 20 years in Great Britain, revealed that SCH patients had higher systolic and diastolic blood pressures and hypercholesterolemia than those in euthyroid patients (7). An association between incident IHD events and IHD-related mortality in the SCH group was also shown in that study (7). Of note, that study was the first study showing that treatment with levothyroxine for SCH patients might reduce the rates of IHD-related morbidity and mortality (7). Recently, it has been revealed that SCH is associated with various risk factors for CVDs, including dyslipidemia, hypertension, atherosclerosis, obesity and metabolic disorders, which lead to cardiovascular mortality and morbidity (1, 5, 8, 9). Considering that SCH is associated with CVD/CVD risk factors and that the prevalence of SCH is higher in the elderly than in the younger population, it is crucial to clarify the relationship between SCH and CVD in the elderly. Of note, a subgroup analysis from a meta-analysis showed that while SCH was significantly associated with increased risk of CVD in individuals younger than 65 years of age, this association was not observed in individuals aged 65 years or older (10). This suggests that the clinical consequences of SCH may be age-dependent, which could impact treatment recommendations for elderly patients. We herein provide recent evidence regarding the association between SCH and CVD, focusing on treatment recommendations for SCH in elderly patients.

Clinical characteristics of SCH

SCH is mainly caused by Hashimoto’s thyroiditis, and positive thyroid peroxidase (TPO) antibodies are a risk for the progression of SCH to overt hypothyroidism (1, 4). Based on data from the Cardiovascular Health Study, the natural history of SCH in the elderly was examined over a period of 4 years (11). The serum TSH levels were normalized over a period of 2 years in 46% of the participants aged 65 years or older whose serum TSH levels were between 4.5 and 6.9 mIU/L, in 10% of the participants with serum TSH levels of 7–9.9 mIU/L and in 7% of the participants with TSH levels of 10 mIU/L or higher (11). The rate of normalization of serum TSH levels in SCH patients without TPO antibodies was 48%, whereas it was 15% in SCH patients with TPO antibodies (11). Another recent study including 2,335 participants showed that serum TSH levels were normalized without treatment in 60.8% of adults aged 65 years or older with SCH, defined as at least one elevated TSH measurement, after a median follow-up period of 1 year (12). Serum TSH levels were still normalized after one more year in 39.9% of older adults with persistent SCH, defined as at least two elevated TSH measurements taken more than 3 months apart (12). In that study, independent determinants associated with TSH normalization were younger age, female sex, lower initial TSH level, higher initial FT4 level, absence of TPO antibodies and a second measurement in summer (12). The representative natural history of SCH is summarized in Fig. 1.

View Full Size

Representative natural history of SCH. The main cause of SCH is Hashimoto’s thyroiditis. While SCH has a risk for progression to overt hypothyroidism, elevated serum TSH levels of SCH may be normalized without treatment.

Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0601

• Download Figure
• Download figure as PowerPoint slide

Our previous study, in which the relationships between chief complaints and thyroid function in general practice were examined, showed that patients with dizziness had a significantly lower serum FT4 level than that in asymptomatic patients (13). In addition, there was a negative correlation between serum FT4 level and age in patients with dizziness but not in patients without dizziness (13). Patients with SCH can be asymptomatic and some patients, especially elderly patients, have unspecific symptoms such as fatigue, shortness of breath and dizziness (4, 14). Since SCH may progress to overt hypothyroidism, clinicians should detect SCH as early as possible (4). To unveil clinical characteristics of latent SCH in general practice, our group investigated 958 outpatients who had been examined for thyroid functions (14). Our study revealed that 9.1% of the analyzed patients had SCH and that 53.8% of the SCH patients had anti-thyroid antibodies (14). The SCH patients were significantly older than the euthyroid patients and they showed higher serum aspartate aminotransferase and low-density lipoprotein cholesterol (LDL-C) levels and lower estimated glomerular filtration rate (eGFR) than those in the euthyroid patients (14). Of note, SCH patients less than 65 years of age had lower eGFRs and higher serum LDL-C levels than those in the euthyroid patients, whereas eGFRs and serum LDL-C levels were not different between the SCH patients and euthyroid patients aged 65 years or older (14). Age-related decreases in red blood cells and serum albumin were more pronounced in the SCH group than in the euthyroid group (14). That study indicated that age-related biochemical changes may be a clue for detecting SCH. In the analysis, clinical data for patients with SCH were used; however, the analysis included patients with heterogeneous backgrounds, reflecting real-world data in a general practice setting. Other thyroid conditions, including non-thyroidal illness, might have affected the results. A prospective multicenter study using age-specific cohorts will be needed in the future.

SCH and CVD risk factors

As shown in Fig. 2, SCH is considered to be associated with several CVD risk factors including dyslipidemia, hypertension and atherosclerosis. It was reported that SCH patients have cardiovascular abnormalities including impaired vascular relaxation, increased arterial stiffness, high blood pressure and endothelial dysfunction (15, 16). In patients with SCH, increased TSH levels may be associated with atherosclerosis through vascular inflammation, oxidative stress, autophagy, hyperhomocysteinemia and cardiometabolic dysfunctions including insulin resistance, dyslipidemia and hypertension (17). Normalization of TSH levels by levothyroxine treatment may prevent the development of atherosclerosis (17). Carotid intima-media thickness (CIMT) is a surrogate marker of subclinical atherosclerotic alterations and is associated with risks of myocardial infarction and stroke (18). SCH, especially SCH with TSH levels of 10 mIU/L or higher, was shown to be associated with an increased CIMT, possibly due to dyslipidemia, hypertension and elevated TSH, in a meta-analysis of results of eight observational studies with 3,602 patients (19). Levothyroxine treatment in SCH patients can reduce CIMT, with which improved lipid profiles may be associated, according to meta-analyses (20, 21). On the other hand, a randomized, double-blind, placebo-controlled trial nested within the TRUST (Thyroid Hormone Replacement for Untreated Older Adults With SCH) trial, which is the largest randomized-controlled trial (RCT) of SCH treatment that included 737 SCH patients aged 65 years or older, showed that levothyroxine treatment for participants aged 65 years or older with SCH was not associated with mean CIMT (22). According to the results of a meta-analysis of the association between SCH and lipid profiles, SCH may correlate with alterations in lipid pattern: increased concentrations of serum total cholesterol, LDL-C and triglyceride levels (23). Some studies have suggested that levothyroxine administration improved elevated levels of serum total cholesterol and LDL-C in SCH patients (24, 25). Hypothyroidism is vital as a cause of secondary dyslipidemia, and screening for hypothyroidism is recommended for patients with dyslipidemia.

View Full Size

Associations of SCH with CVD risk factors, CVD and mortality. SCH is associated with various risk factors of CVD, including dyslipidemia, hypertension and atherosclerosis. There are associations between SCH and risks of CVD and mortality in younger patients aged less than 65 years but not in older patients aged over 65 years.

Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0601

• Download Figure
• Download figure as PowerPoint slide

Although the association between SCH and hypertension is still unclear, in a recent meta-analysis study, pooled analysis of prospective cohort studies showed that levothyroxine therapy for SCH patients significantly lowered both systolic blood pressure and diastolic blood pressure, indicating a possible link between SCH and blood pressure, which might lead to hypertension (26). On the other hand, the TRUST trial showed that levothyroxine treatment did not reduce blood pressure (27). Doppler echocardiography in patients with SCH revealed left ventricular diastolic dysfunction, which might be improved by levothyroxine treatment (28). SCH is thought to increase blood pressure via reduced thyroid hormone-mediated endothelial-dependent vasodilation (29). SCH may also increase renin–angiotensin–aldosterone axis activation, vasoconstriction and sympathetic activity and reduce renal blood flow and GFR, which are risks for heart failure (29). A cross-sectional study focusing on older adults aged more than 65 years with CVD risk factors showed that SCH in older adults with CVD risk factors is associated with low GFR, which may lead to a risk of CVD (30). The results of studies on the relationship between SCH and metabolic syndrome have been conflicting (15). However, since both SCH and metabolic syndrome are associated with atherosclerosis and liver and kidney diseases, this association may enhance the negative effects, leading to worse cardiovascular outcomes (15).

Relationship between SCH and CVD in the elderly

A cross-sectional study using randomly selected Medicare patients aged 65 years or older showed that the prevalence of coronary heart disease was significantly higher in patients with serum TSH levels greater than 10 mIU/L than in patients whose serum TSH levels were less than or equal to 4.6 mIU/L (52.6 versus 25.0%; P = 0.007) (31). A database study using data from patients seen in the Cleveland Clinic Preventive Cardiology Clinic showed that the prevalence of coronary heart disease and all-cause mortality was higher in patients with serum TSH levels of 6.1–10 mIU/L or greater than 10 mIU/L (32). In those groups, the mortality rate was high in untreated patients and those aged less than 65 years but not in patients aged 65 years or over (32). Whether or not patients have TPO antibodies was not associated with risks of coronary heart disease mortality and events in SCH patients (33).

The association between SCH and CVD was investigated using the atherosclerotic CVD risk score in an observational study in a prospective Korean cohort that was followed up for 12 years (34). The SCH group with the highest quartile of serum TSH levels (>6.57 uIU/L) had significantly increased risks of all-cause mortality (hazard ratio (HR), 2.12; 95% confidence interval (CI), 1.27–3.56) and CVD events (HR, 1.92; 95% CI, 1.21–3.04) compared to those in the euthyroid group (34). Subgroup analysis by CVD risk showed that the high CVD risk group only had high risks of all-cause mortality (HR, 2.18; 95% CI, 1.22–3.87) and CVD events (HR, 2.42; 95% CI, 1.35–4.33) compared with those in the euthyroid group (34). A large, prospective cohort study, in which 3,233 community-dwelling U.S. men aged 65 years or older were enrolled, showed that SCH diagnosed by a single set of baseline thyroid function tests was not associated with an increased risk of all-cause or cardiovascular mortality (35). In addition, the same corresponding author conducted another large population-based study that included older patients aged 65 years or older with persistent SCH and found that there was no association between persistent SCH and CVD, heart failure or cardiovascular mortality (36). A recent meta-analysis conducted by Moon et al. revealed that SCH was associated with CVD (relative risk (RR), 1.33; 95% CI, 1.14–1.54) and with all-cause mortality (RR, 1.20; 95% CI, 1.07–1.34) (10). Subgroup analysis of studies with a mean participant age of <65 years showed significant associations between SCH and risk of CVD (RR, 1.54; 95% CI, 1.21–1.96) and between SCH and all-cause mortality (RR, 1.28; 95% CI, 1.10–1.48); on the other hand, a meta-analysis of studies with a mean participant age of ≧65 years showed no significant association between SCH and CVD (RR, 1.07; 95% CI, 0.97–1.18) or between SCH and all-cause mortality (RR, 1.13; 95% CI, 0.97–1.33) (10). In addition, a subgroup analysis showed that SCH patients with high CVD risk had a significantly higher risk of all-cause mortality (RR, 1.66; 95% CI, 1.41–1.94), whereas SCH patients with low CVD risk did not (10). Figure 2 summarizes the associations between SCH and CVD risk and mortality.

Regarding the association between SCH and heart failure, a pooled analysis was conducted using data from 25,390 participants with 216,248 person-years of follow-up supplied from six prospective cohorts in the United States and Europe (37). It was found that the risk of heart failure events was increased in patients with serum TSH levels of 10 mIU/L or higher (37). A prospective cohort study that included 5,316 patients aged 70–82 years old with known cardiovascular risk factors or previous CVD showed that SCH only with TSH levels of more than 10 mIU/L was associated with heart failure (38). Barasch et al. conducted a large cross-sectional analysis within the Cardiovascular Health Study, a prospective, community-based, epidemiologic observational study, to assess cardiovascular risk factors and outcomes in older adults (39). They investigated the impact of thyroid dysfunction on cardiac abnormalities by using blood biomarkers and echocardiography (39). They found that SCH was associated with higher levels of N-terminal pro-brain-type natriuretic peptide, NT-proBNP, consistent with diastolic overload, but not with echocardiographic measures (39). In this study, participants were recruited from four urban areas in the United States, and 5,201 participants were enrolled, of whom 3,678 had measured TSH (39). Participants taking medications that interfere with thyroid function (except for those with exogenous hyperthyroidism) (n = 311), those with prevalent heart failure (n = 143) and those with atrial fibrillation (n = 61) were excluded (39). The final sample of 3,163 participants was analyzed (39). Among the participants, 2,477 were euthyroid, 465 had SCH, 47 had overt hypothyroidism, 129 had exogenous subclinical hyperthyroidism due to thyroid hormone supplementation and 45 had endogenous subclinical hyperthyroidism (39). This process ensured that the association between thyroid dysfunction and cardiac abnormalities was directly analyzed; however, other thyroid-related factors, including non-thyroidal illnesses, may also have influenced the results. A meta-analysis, including 13 studies, on the association between hypothyroidism and clinical outcomes in patients with heart failure showed that SCH and hypothyroidism was associated with all-cause mortality for patients with heart failure (pooled RR, 1.41; 95% CI, 1.15–1.74) (40).

Effects of levothyroxine treatment on CVD in elderly SCH patients

A retrospective study using the United Kingdom General Practitioner Research Database showed that levothyroxine therapy for SCH was not associated with fewer IHD events in patients aged 70 years or older with serum TSH levels of 5.01–10.00 mIU/L during a median follow-up period of 7.6 years; on the other hand, there were fewer IHD events and reduced all-cause mortality in younger SCH patients aged 40–70 years (41). A more extensive retrospective cohort study was conducted using Danish primary care patients aged 18 years or older (n = 628,953), 1.9% of whom had SCH (n = 12,212) (42). During a median follow-up period of 5.0 years, levothyroxine treatment was not associated with the incidence of myocardial infarction (incidence rate ratio (IRR), 1.08; 95% CI, 0.81–1.44), cardiovascular death (IRR, 1.02; 95% CI, 0.83–1.25) or all-cause mortality (IRR, 1.03; 95% CI, 0.90–1.19); however, subgroup analysis revealed that levothyroxine treatment was associated with all-cause mortality in patients aged less than 65 years (IRR, 0.63; 95% CI, 0.40–0.99) (42). Furthermore, a registry-based Danish cohort study, in which 1,192 older patients (mean age, 74 years) with SCH and heart disease were included, showed that levothyroxine treatment was not associated with a significant benefit or risk of all-cause mortality, major adverse cardiovascular events or hospital admission (43). A pooled analysis of two large RCTs, TRUST and IEMO80+ (the Institute for Evidence-Based Medicine in Old Age 80-plus thyroid trial), in which the effects of levothyroxine treatment on cardiovascular outcomes in older SCH patients were investigated, showed that levothyroxine treatment did not significantly affect the risks of cardiovascular outcomes, including fatal and non-fatal cardiovascular events (HR, 0.74; 95% CI, 0.41–1.35), atrial fibrillation (HR, 0.69; 95% CI, 0.32–1.52), heart failure (HR, 0.41; 95% CI, 0.13–1.35) and all-cause mortality (HR, 1.28; 95% CI, 0.54–3.03), in older SCH patients irrespective of a history of CVD and age (44). In a meta-analysis that included five observational studies and two RCTs, the effect of thyroid hormone treatment on mortality in adult SCH patients was investigated (45). Levothyroxine treatment was not significantly associated with all-cause or cardiovascular mortality, whereas subgroup analyses revealed that levothyroxine treatment in younger SCH patients aged less than 65–70 years was associated with lower all-cause mortality (pooled RR, 0.50; 95% CI, 0.29–0.85) and lower cardiovascular mortality (pooled RR, 0.54; 95% CI, 0.37–0.80) (45). No significant association between treatment and mortality was found in older SCH patients aged 65–70 years or older (45). In addition, a systematic review and meta-analysis performed to evaluate cardiovascular and bone health outcomes in elderly SCH patients showed that levothyroxine treatment was not significantly associated with cardiovascular outcomes (HR, 0.89; 95% CI, 0.71–1.12) and bone health outcomes in SCH patients over 65 years of age (46). The effects of levothyroxine therapy for SCH on CVD, CVD risk factors and mortality are summarized in Fig. 3.

View Full Size

Impact of levothyroxine (LT4) treatment on SCH. SCH is associated with mortality through CVD. LT4 treatment for SCH may improve risk factors for CVD, including atherosclerosis, dyslipidemia and hypertension. LT4 administration is associated with lower mortality in young SCH patients (<65–70 years old) but not in patients aged 65–70 years or older.

Citation: Endocrine Connections 14, 5; 10.1530/EC-24-0601

• Download Figure
• Download figure as PowerPoint slide

To investigate the impact of levothyroxine treatment on cardiac function in SCH patients, an RCT nested within the TRUST trial was conducted (47). Swiss participants aged 65 years or older with SCH were randomized to receive levothyroxine replacement therapy or a placebo (47). A total of 185 participants were enrolled and treated for a median duration of 18.4 months (47). Levothyroxine therapy did not have any effect on heart function including mean left ventricular ejection fraction and E/e’ ratio (47). Recently, a randomized, multicenter, open-label trial, in which the effects of levothyroxine treatment for SCH patients with heart failure with reduced ejection fraction (HFrEF) were investigated, showed that levothyroxine treatment addition to standard HFrEF treatment improved the results of a 6-minute walk test, New York Heart Association classification and thyroid function within 6 months (48). However, specific trials on levothyroxine therapy for SCH patients with heart failure, both HFrEF and heart failure with preserved ejection fraction, are scarce and should be performed in the future (48).

How to treat SCH in elderly patients

It is still unclear whether levothyroxine therapy for SCH reduces CVD risk factors, CVD and CVD mortality, especially in elderly patients (9, 29). Since older patients have a relatively high risk of adverse effects from levothyroxine treatment, including atrial fibrillation and cardiovascular mortality due to iatrogenic thyrotoxicosis, overdiagnosis and overtreatment should be avoided in older patients (4). Most of the international guidelines recommend that levothyroxine replacement therapy should be individualized based on age, serum TSH level, symptoms, CVD risk and other comorbidities (29). Recommendations for SCH treatment in elderly patients are different in international guidelines (49, 50, 51). The European Thyroid Association recommends levothyroxine replacement therapy for patients younger than 65–70 years of age with serum TSH levels over 10 mIU/L, even if they do not have symptoms (49). Younger SCH patients (<65–70 years) with serum TSH levels less than 10 mIU/L, who have symptoms suggestive of hypothyroidism, should be considered as candidates for replacement therapy (49). Elderly SCH patients older than 80–85 years of age with elevated serum TSH levels of 10 mU/L or less should be carefully followed without levothyroxine treatment (49). The Latin American Thyroid Society recommends that all patients with persistently elevated serum TSH levels of 10 mU/L or more should be treated since they may progress to overt hypothyroidism and have higher risks of congestive heart failure, CVD and mortality (50). In SCH patients younger than 65 years of age, with a mild increase in serum TSH levels (4.5–10 mU/L), particularly with persistent elevation of more than 7 mU/L, those with increased cardiovascular risk should be considered for treatment (50). The guidelines recommend against routine treatment for SCH patients aged over 65 years and those aged over 80 years with serum TSH levels less than 10 mU/L (50). On the other hand, a clinical practice guideline published by an international expert panel strongly recommends against levothyroxine treatment in adults with SCH, except for women trying to become pregnant or patients with serum TSH levels more than 20 mIU/L (51). The panel presents a concern about the lifelong treatment burden and uncertain potential harms (51). The recommendation may not apply to those with severe symptoms or young adults aged 30 years or younger (51).

Future perspectives and conclusions

Several points should be addressed to improve the diagnosis and management of SCH and to understand the relationship between SCH and CVD in elderly patients. Standardization of FT4 values and harmonization of TSH measurement would be needed, especially for elderly patients, since calibration differences between manufacturers can be reduced by standardization/harmonization (9, 52). The upper bound of the TSH reference interval increases with age, which may represent a normal compensatory phenomenon in older adults aged 65 years or older (52). Some studies have suggested that many older individuals may be misdiagnosed with SCH if age-specific serum TSH reference ranges are not used (4, 53, 54). In the future, the thresholds of serum TSH levels for age-specific reference ranges of elderly patients should be investigated for appropriate diagnosis and treatment of SCH (4, 9). Larger RCTs using age-specific diagnostic criteria of SCH should be conducted to clarify the relationship between SCH and CVD in elderly patients and to examine the effects of levothyroxine treatment, with focus on specific SCH populations including elderly patients, specific serum TSH levels, TPO antibody status and various cardiovascular risks (4). More country-specific guidelines for SCH should be made. For example, the prevalence of anti-thyroid autoantibodies varies in different countries. A study using data from general health checkups in Japan, an iodine-sufficient country, showed that the prevalence of anti-thyroglobulin antibodies is higher than that of anti-TPO antibodies (9, 55). Japanese SCH patients had a higher prevalence of anti-thyroglobulin antibodies than that of anti-TPO antibodies (9, 55). The significance of each anti-thyroid autoantibody in SCH should be investigated.

Recent studies have suggested that SCH is associated with CVD risk factors, CVD and cardiovascular mortality and that levothyroxine treatment is associated with lower mortality in younger patients but not in elderly patients. Larger RCTs using age-specific reference ranges of thyroid function tests are needed to make a conclusion about the effectiveness of levothyroxine treatment for reducing CVD risk factors, CVD and cardiovascular mortality in elderly SCH patients.

Supporting information

We identified articles cited in this review through a PubMed search with the following terms: ‘SCH’ and ‘CVDs’ [Mesh] or a combination with (‘aged’ [Mesh] OR ‘elderly’ OR ‘older’). The article types were ‘clinical trial’, ‘guideline’, ‘meta-analysis’, ‘randomized clinical trial’, ‘review’, ‘systematic review’ and ‘observational study’. The search was performed from database inception until September 30, 2024. The reference lists of the cited papers were also reviewed. In this review, we referenced various types of studies, including observational studies, RCTs, reviews and meta-analyses. To ensure the quality of the present review, we evaluated the quality of the key systematic reviews (10, 45) that influenced the conclusion on the association between SCH and CVD/CVD risk factors in the elderly. This evaluation was conducted using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2020 checklists (56). These articles declared that they followed the PRISMA guidelines, and we checked that (see Supplementary Materials).

Supplementary materials

This is linked to the online version of the paper at https://doi.org/10.1530/EC-24-0601.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.

Funding

This work did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.

Author contribution statement

KY wrote the first draft and managed the submission process. YN, YS, AS, KO and KH checked the manuscript. FO organized the writing of the manuscript.