Version May 2024
INTRODUCTION — Subclinical hypothyroidism is defined biochemically as a normal serum free thyroxine (T4) concentration in the presence of an elevated serum thyroid-stimulating hormone (TSH) concentration. Some patients with subclinical hypothyroidism may have vague, nonspecific symptoms suggestive of hypothyroidism, but attempts to identify patients clinically have not been successful [1,2]. Thus, this disorder can only be diagnosed on the basis of laboratory test results.
This topic will review the diagnosis and management of subclinical hypothyroidism. The clinical manifestations, diagnosis, and management of overt hypothyroidism, as well as subclinical and overt hypothyroidism during pregnancy, are reviewed separately.
●(See "Clinical manifestations of hypothyroidism".)
●(See "Diagnosis of and screening for hypothyroidism in nonpregnant adults".)
●(See "Laboratory assessment of thyroid function".)
●(See "Treatment of primary hypothyroidism in adults".)
EPIDEMIOLOGY — In population-based studies, the prevalence of subclinical hypothyroidism ranges from 4 to 15 percent [1,3-7]. In the United States Third National Health and Examination Survey (NHANES III), which excluded subjects with known thyroid disease, 4.3 percent of 16,533 people had subclinical hypothyroidism [8]. The prevalence rises with age, is higher in females than males, and is lower in Black persons than in White persons [5-8]. However, the prevalence is determined by the upper limit of normal for serum TSH. If the upper limit of normal rises with age, as appears to be the case, then the prevalence may not be as high as has been previously thought. In one study of individuals over age 70, the prevalence of a diagnosis of subclinical hypothyroidism fell from 29.6 to 3 percent if age-specific rather than uniform normal reference ranges were used [9]. (See 'Diagnosis' below.)
In Europe, where iodine intake is variable, subclinical hypothyroidism is more prevalent in areas of iodine sufficiency. In one study, the prevalence of subclinical hypothyroidism ranged from 4.2 percent in iodine-deficient areas to 23.9 percent in an area of abundant iodine intake, despite a similar prevalence of patients with high serum concentrations of antithyroid peroxidase (anti-TPO) antibodies [10].
ETIOLOGY — The causes of subclinical hypothyroidism are the same as those of overt hypothyroidism (table 1). (See "Disorders that cause hypothyroidism".)
Most patients have chronic autoimmune (Hashimoto's) thyroiditis with high serum concentrations of antithyroid peroxidase (anti-TPO, formerly called antithyroid microsomal) antibodies [11]. Other major causes include prior ablative or antithyroid drug therapy for hyperthyroidism caused by Graves' disease; prior partial thyroidectomy; external radiation therapy in patients with Hodgkin lymphoma, leukemia, or brain tumors; inadequate T4 (levothyroxine) replacement therapy for overt hypothyroidism; and drugs impairing thyroid function.
CLINICAL FINDINGS — Most patients with subclinical hypothyroidism have serum TSH levels <10 mU/L and are asymptomatic. In particular, older patients with subclinical hypothyroidism appear to be asymptomatic, although many euthyroid older individuals also have symptoms that might be construed as being related to hypothyroidism, including dry skin, constipation, and low energy [12-14]. In a population-based, prospective study of 558 individuals in the Netherlands who were screened for hypothyroidism during the month of their 85th birthday and again three years later, 5 percent had subclinical hypothyroidism [12]. At baseline, there was no association between baseline serum TSH concentration and cognitive function, depressive symptoms, or disability in activities of daily living. Although measures of performance declined over time, increased serum TSH at baseline was associated with a slower decline in ability to perform activities such as preparing one's own meals, shopping for groceries and personal items, managing one's money, using the telephone, and doing housework. In another community-based study of individuals ≥65 years, subclinical hypothyroidism was not associated with depression, anxiety, or cognition [13].
Some patients with subclinical hypothyroidism, however, may have vague, nonspecific symptoms suggestive of hypothyroidism, such as fatigue and constipation, but attempts to identify patients clinically have not been successful [2,11].
DIAGNOSIS — The diagnosis of subclinical hypothyroidism is based upon biochemical testing alone.
Subclinical hypothyroidism is defined as [11]:
●Normal serum free T4
●Elevated TSH
It may occur in the presence or absence of mild symptoms of hypothyroidism.
In most circumstances, the initial screening test for thyroid disease is the serum TSH. If the serum TSH concentration is elevated, the TSH measurement should be repeated along with a serum free T4 before making the diagnosis of subclinical hypothyroidism. Because the serum TSH concentration can be transiently elevated, a serum TSH measurement should be repeated after one to three months to confirm the diagnosis. An exception, however, is pregnant women or women undergoing infertility evaluation or treatment, in whom a TSH and free T4 should be repeated immediately and T4 (levothyroxine) treatment started if an elevated TSH is confirmed. (See "Hypothyroidism during pregnancy: Clinical manifestations, diagnosis, and treatment", section on 'Indications for treatment'.)
●Definition of elevated TSH
•For nonpregnant adults, an elevated serum TSH is defined as a TSH concentration above the upper limit of the normal TSH reference range, which is typically 4 to 5 mU/L in most laboratories. However, some experts suggest that the true upper limit is only 2.5 or 3 mU/L in healthy individuals without thyroid disease, while others argue that the serum TSH distribution shifts towards higher values with age, independent of the presence of antithyroid antibodies [15,16]. In this case, the upper limit of normal could be as high as 6 to 8 mU/L in healthy octogenarians. Considerable controversy remains over the appropriate upper limit of normal for serum TSH. This topic is reviewed in more detail separately. (See "Laboratory assessment of thyroid function", section on 'Serum TSH'.)
•For women trying to conceive who have ovulatory dysfunction or infertility, elevations in TSH can be defined using first trimester-specific TSH reference ranges. (See "Overview of thyroid disease and pregnancy", section on 'Trimester-specific reference ranges'.)
•For pregnant women, elevations in TSH should be defined using population and trimester-specific TSH reference ranges. (See "Overview of thyroid disease and pregnancy", section on 'Trimester-specific reference ranges'.)
Differential diagnosis — There are several causes of a high serum TSH concentration that do not properly fit the definition of subclinical hypothyroidism. These include the following circumstances:
●During the period of recovery from nonthyroidal illness, where a transiently elevated serum TSH is seen after a period of TSH suppression. (See "Thyroid function in nonthyroidal illness".)
●Following the hyperthyroid phase of subacute, painless, or postpartum thyroiditis, where mild hypothyroidism is usually, but not always, transient. (See "Overview of thyroiditis".)
●Assay variability.
●The presence of heterophilic antibodies can interfere with TSH measurements in immunometric assays. These human anti-mouse gamma globulins can bridge the two mouse monoclonal antibodies (solid phase antibody and signal antibody) and cause spuriously elevated readings for TSH [17]. Rarely, heterophilic antibodies block binding of one of the mouse monoclonal antibodies to TSH and result in spuriously low readings for TSH. Nonlinearity of the serum TSH concentration with dilution of the patient's serum suggests interference. Addition of nonimmune homologous mouse immunoglobulins has reduced this type of assay interference. Commercial assays exist for detecting human anti-mouse antibodies (HAMA).
●Rheumatoid factors may cause similar interference in immunometric assays [17].
●Autoantibodies to TSH have also been described that create TSH-anti-TSH immunoglobulin G (IgG) complexes, also called "macro-TSH," which lack biologic activity but may be immunoreactive and cause spuriously high TSH values (often >100 mU/L) in euthyroid (normal free T4 and triiodothyronine [T3] levels) individuals [18-20]. In one study, 1.6 percent of patients diagnosed with subclinical hypothyroidism actually had macro-TSH [21]. Autoantibodies to TSH can be detected by removal of the IgG-TSH complexes with polyethylene glycol or protein A or G and then repeating the assay on the immunosubtracted sera [22].
●Untreated adrenal insufficiency.
●TSH-producing pituitary adenomas, where the elevated TSH is associated with elevated serum free T4 and/or T3 concentrations. Patients with subclinical hypothyroidism have normal free T4 levels. (See "TSH-secreting pituitary adenomas".)
●Resistance to thyroid hormone and rare mutations of the TSH receptor. In patients with resistance to thyroid hormone, the elevated TSH is associated with elevated serum free T4 and/or T3 concentrations. In contrast, patients with subclinical hypothyroidism have normal free T4 levels. (See "Resistance to thyroid hormone and other defects in thyroid hormone action".)
Patients with resistance to TSH secondary to alterations in the TSH receptor have high serum TSH concentrations and normal or low serum free T4 and T3 concentrations. (See "Resistance to thyrotropin and thyrotropin-releasing hormone".)
●Central hypothyroidism, where up to 25 percent of patients have a mildly elevated serum TSH (up to approximately 10 mU/L) and a low or low-normal free T4. (See "Central hypothyroidism".)
●Class II or III obesity, where high TSH is mediated centrally by leptin and TSH falls with weight loss (eg, after bariatric surgery) [23-25].
IDENTIFYING THE CAUSE
●History and physical examination – Although most patients with subclinical hypothyroidism are asymptomatic, patients should be questioned about symptoms of hypothyroidism in addition to past treatment for hyperthyroidism, a history of overt hypothyroidism, and use of medications that may impair thyroid hormone absorption or function (table 1 and table 2). In addition, they should be examined for the presence of thyroid gland enlargement [26]. (See "Clinical manifestations of hypothyroidism".)
●Laboratory testing – We do not routinely measure thyroid antibodies in patients with subclinical hypothyroidism. Most patients with subclinical hypothyroidism have chronic autoimmune (Hashimoto's) thyroiditis. We measure antithyroid peroxidase (anti-TPO) antibodies when the decision to treat or to monitor is not obvious. The presence of anti-TPO antibodies indicates Hashimoto's thyroiditis and predicts the likelihood of progression to overt hypothyroidism. Therefore, they may be useful in making management decisions. Antithyroglobulin antibodies could also be obtained, but they are less specific for Hashimoto's thyroiditis since they can be positive in patients with non-autoimmune nodular goiter, and they do not predict progression to overt hypothyroidism. (See 'Progression to overt hypothyroidism' below and 'Management' below.)
●Imaging – We do not routinely obtain thyroid ultrasound or other imaging in patients with subclinical hypothyroidism and normal thyroid examination. However, thyroid ultrasound is an important component of the evaluation of thyroid nodules and goiter. (See "Diagnostic approach to and treatment of thyroid nodules", section on 'Initial' and "Clinical presentation and evaluation of goiter in adults", section on 'Approach to evaluation'.)
CONSEQUENCES OF SUBCLINICAL HYPOTHYROIDISM
Progression to overt hypothyroidism — A substantial proportion of patients with subclinical hypothyroidism eventually develop overt hypothyroidism. In prospective studies with nearly 10 to 20 years of follow-up, the cumulative incidence of overt hypothyroidism ranges from 33 to 55 percent [27-29]. The annual rate of progression to overt hypothyroidism ranges from 2 to 4 percent.
The risk of progression is related to the initial serum TSH concentration (increased with TSH values higher than 12 to 15 mU/L) and the presence of antithyroid peroxidase (anti-TPO) antibodies [27,30,31]. In a study in which more than 1700 subjects were followed for 20 years, for example, women with both high serum TSH and high thyroid antibody concentrations developed hypothyroidism at a rate of 4.3 percent per year (cumulative incidence 55 percent) [28]. In another study in which 82 women were observed for 9.2 years, the cumulative incidence of overt hypothyroidism was 0 percent for subjects with initial TSH concentrations of 4 to 6 mU/L [27].
The underlying disease also may be a determinant of the risk of overt hypothyroidism [29]. Patients who have autoimmune thyroid disease or received radioiodine therapy or high-dose therapeutic external radiotherapy are likely to progress to overt hypothyroidism. In contrast, subclinical hypothyroidism is likely to persist in those who have had thyroid surgery for indications other than hyperthyroidism or in those who received low-dose external radiotherapy for benign conditions during childhood.
Spontaneous recovery has also been described in patients with subclinical hypothyroidism, although the frequency of this phenomenon is unclear [27,31-33]. In a study of 422,242 persons without known thyroid disease, serum TSH was elevated (5.5 to ≤10 mU/L) in 3 percent [33]. During the five-year follow-up period, TSH levels became normal without treatment in 62 percent of patients. Normalization of serum TSH concentrations is more likely to occur in patients with negative antithyroid antibodies, serum TSH levels <10 mU/L, and within the first two years after diagnosis [32].
Cardiovascular disease — Subclinical hypothyroidism may be associated with an increased risk of cardiovascular disease (CVD; eg, coronary heart disease [CHD], heart failure), particularly when the serum TSH concentration is above 10 mU/L. However, the data are conflicting due to differences in the patient populations studied and study designs.
●CHD – Some [34-39], but not all [40-42], observational studies report an increased risk of CHD in patients with subclinical hypothyroidism. A meta-analysis of patient-level data from seven prospective cohort studies (25,977 participants, 2020 with subclinical hypothyroidism) showed a significant trend of increased risk of CHD events (nonfatal myocardial infarction, CHD death, hospitalization for angina, or coronary revascularization) at higher serum TSH concentrations [43]. Compared with euthyroid individuals, patients with TSH ≥10 mU/L had a significant increase in CHD events (38.4 versus 20.3 events/1000 person-years; hazard ratio [HR] 1.89, 95% CI 1.28-2.80). In contrast, minimal TSH elevations (4.5 to 6.9 mU/L) were not associated with an increased risk (HR 1.00, 95% CI 0.96-1.43). The risk estimates did not differ according to age, sex, or presence of preexisting CVD. In a separate meta-analysis, risk estimates did not differ by TPO antibody status [44].
●Heart failure – In a pooled analysis of patient-level data from six prospective cohort studies (25,390 participants, 2068 with subclinical hypothyroidism), there was a significant trend for increased risk of heart failure events at higher TSH concentrations [45]. Compared with euthyroid controls, patients with a TSH between 10 and 19.9 mU/L had a significant increase in heart failure (40 events in 224 participants [17.9 percent] versus 1762 events in 22,674 controls [7.8 percent]; HR 1.86, 95% CI 1.27-2.72) [45].
The increased risk of heart failure in those with a serum TSH between 7.0 and 9.9 mU/L was not statistically significant (54 events in 422 participants [12.8 percent]; HR 1.65, 95% CI 0.84-3.23), and minimal TSH elevations (4.5 to 6.9 mU/L) were not associated with an increased risk (HR 1.01, 95% CI 0.81-1.26).
In a subsequent study of 1100 consecutive patients with heart failure, those patients with subclinical hypothyroidism, compared with those who were euthyroid, had impaired exercise capacity, higher pulmonary artery pressures, and increased cardiovascular events [46].
●Stroke – Overall, there does not appear to be an association between subclinical hypothyroidism and risk of stroke [47]. However, in younger patients and in those with higher TSH concentrations, there may be an increased risk. This was illustrated by an analysis of patient-level data from 17 prospective cohort studies (47,573 adults, 3451 with subclinical hypothyroidism) [48]. Overall, when compared with euthyroidism, subclinical hypothyroidism was not associated with risk for stroke events, which occurred in 9.3 versus 6.6 percent of euthyroid participants (HR 1.05, 95% CI 0.91-1.21) or for fatal stroke (2.8 versus 2.1 percent, HR 1.07, 95% CI 0.80-1.42). However, in predefined subgroup analyses, there was an increased risk of stroke events (64 total events in 8555 total participants, HR 3.32, 95% CI 1.25-8.80) in patients 18 to 49 years of age and an increased risk of fatal stoke in patients 18 to 49 years (HR 4.22, 95% CI 1.08-16.55) and 50 to 64 years (HR 2.86, 95% CI 1.31-6.26). There was no increased risk for those 65 to 79 years or ≥80 years. The subgroup analysis was limited by the small number of events within each subgroup. There was a nonsignificant pattern of increased risk of fatal stroke with higher TSH concentrations.
●Lipids – Some, but not all, studies show an association between elevated TSH and total and low-density lipoprotein (LDL) cholesterol concentrations [11]; however, cholesterol levels were not always adjusted for age, and it is not known whether this difference is clinically important with regard to CHD risk. In one of the largest cross-sectional studies to date (25,862 participants, median age 56 years), patients with modest elevations of serum TSH (between 5.1 and 10 mU/L) had significantly higher mean total cholesterol concentrations than those who were euthyroid (223 versus 216 mg/dL [5.6 versus 5.8 mmol/L]) [6]. (See "Lipid abnormalities in thyroid disease", section on 'Subclinical hypothyroidism'.)
In addition, subclinical hypothyroidism has been associated with an increase in a number of other cardiovascular risk factors and surrogate cardiovascular endpoints, including markers of inflammation [49-52] and endothelial function [53-55]. Some patients with subclinical hypothyroidism also have diastolic dysfunction and increased peripheral vascular resistance, as noted in patients with overt hypothyroidism [56]. In contrast, another study found no abnormalities in left ventricular mass or function in patients with serum TSH concentrations between 3.5 and 10 mU/L compared with those with normal TSH [57]. (See "Cardiovascular effects of hypothyroidism".)
Mortality — Studies of the effect of subclinical hypothyroidism on mortality show conflicting results. In some [36,37,39,58-61], but not all [12,41,62,63], studies, patients with subclinical hypothyroidism have an increased risk of cardiovascular and/or all-cause mortality.
In a meta-analysis of patient-level data from 11 prospective cohort studies, the risk of cardiovascular mortality, but not all-cause mortality, increased with higher concentrations of TSH and was significantly increased in patients with TSH concentrations ≥10 mU/L (HR 1.58, 95% CI 1.10-2.27) or between 7.0 and 9.9 mU/L (HR 1.42, 95% CI 1.03-1.95) [43]. In contrast, minimal elevations of TSH (4.5 to 6.9 mU/L) were not associated with cardiovascular or all-cause mortality.
In another meta-analysis of patient-level data from four prospective cohort studies that evaluated community-dwelling older individuals (age 80 to 109 years), subclinical hypothyroidism did not reduce or increase mortality (HR 0.9, 95% CI 0.7-1.2) [64].
In one prospective study included in the meta-analyses, older individuals (>85 years) in the Netherlands with untreated subclinical hypothyroidism (the majority with TSH between 4.8 and 10 mU/L) actually had a lower rate of cardiovascular and all-cause mortality [12].
Other studies report a range of results, as illustrated by the following:
●In a retrospective cohort study from Denmark, subclinical hypothyroidism with a TSH of 5 to 10 mU/L was associated with a reduction in all-cause mortality [65].
●In a prospective cohort study from the United States, older individuals with untreated subclinical hypothyroidism had neither increased nor decreased mortality over a median follow-up period of five years [66].
●In a study using the Third National Health and Nutrition Examination Survey (NHANES III), subclinical hypothyroidism (median TSH 6.3 mU/L) compared with euthyroidism was associated with greater mortality in those with heart failure but not in those without heart failure [67].
●In a retrospective cohort study of community-dwelling persons (mean age 83 years) from Israel, subclinical hypothyroidism (serum TSH >5 mU/L and above) was associated with an increase in all-cause mortality but only in the first year of follow-up [68].
Reproductive abnormalities — Although women with overt hypothyroidism may have either oligo- or amenorrhea, which can result in decreased fertility, the effects of subclinical hypothyroidism on reproduction are not well defined.
In one study that evaluated the prevalence of subclinical hypothyroidism among 244 women with infertility (approximately half of whom had ovulatory dysfunction), subclinical hypothyroidism was diagnosed in 13.9 percent compared with 3.9 percent of control women (healthy women with confirmed fertility) [69]. The difference in TPO-antibody positivity between infertile and fertile women was not statistically significant. However, in a meta-analysis of four studies, the presence of thyroid autoimmunity prior to conception increased the risk of unexplained subfertility (odds ratio [OR] 1.5, 95% CI 1.1-2.0) [70].
In a subsequent prospective cohort study, healthy women ages 18 to 40 years with regular menstrual cycles who were attempting to conceive and had one or two prior pregnancy losses but no history of infertility, were evaluated preconception and followed during pregnancy attempts (and throughout pregnancy if they conceived) [71]. The mean preconception TSH was 2.1 mU/L (2.5 to 97.5 percentile 0.56-5.1). There was no difference in time to pregnancy in women with preconception TSH levels <2.5 versus ≥2.5 mU/L or in women with or without TPO antibodies. Whether women with preconception subclinical hypothyroidism (TSH levels above the usual normal range [eg, >5 mU/L]) have decreased fertility remains uncertain.
In men, subclinical hypothyroidism is associated with increased sperm DNA fragmentation [72].
There is evidence of an increased risk of miscarriage in women with subclinical hypothyroidism and in euthyroid women with TPO antibodies. The effects of subclinical hypothyroidism during pregnancy are reviewed separately. (See "Recurrent pregnancy loss: Evaluation", section on 'Thyroid function' and "Hypothyroidism during pregnancy: Clinical manifestations, diagnosis, and treatment", section on 'Subclinical hypothyroidism' and "Overview of thyroid disease and pregnancy", section on 'Thyroid peroxidase antibodies in euthyroid women'.)
Nonalcoholic fatty liver disease — In a cross-sectional study, nonalcoholic fatty liver disease (NAFLD) was correlated with serum TSH levels. Thirty and 36 percent of individuals with subclinical or overt hypothyroidism, respectively, had typical ultrasonographic findings of NAFLD (versus 20 percent of controls), while 20 and 26 percent of individuals with subclinical or overt hypothyroidism had abnormal liver enzymes [73]. Among patients with NAFLD, the prevalence of nonalcoholic steatohepatitis (NASH) and fibrosis was 58 and 25 percent, respectively, in patients with subclinical hypothyroidism versus 37 and 11 percent among those with normal thyroid function [74].
Neuropsychiatric symptoms — Several reports suggest that subclinical hypothyroidism is associated with neuropsychiatric diseases [75-78]. However, other studies do not demonstrate an association of subclinical hypothyroidism with depression, anxiety, or cognitive dysfunction [12,13,79-82]. As an example, in a prospective study of 220,545 middle-aged individuals followed for two years, of whom 4384 had subclinical hypothyroidism and 7323 developed incident depression, there was no correlation between subclinical hypothyroidism and depression [82].
Potential consequences
●In some [83-85], but not all [86,87], studies in middle-aged adults, increasing serum TSH concentrations within the normal range or slightly above normal were associated with a modest increase in body weight. In older women (>65 years), subclinical hypothyroidism (mean TSH 6.7 mU/L) compared with euthyroidism (TSH 2.2 mU/L) was associated with a slightly higher baseline weight (0.51 kg higher baseline weight per 1 mU/L higher TSH level) but not with weight change over time [88], and T4 (levothyroxine) treatment of patients with subclinical hypothyroidism was not associated with weight loss [89]. There was no relationship between TSH and weight in older men.
●In one study, 21 of 33 patients (64 percent) with subclinical hypothyroidism had a higher frequency of neuromuscular symptoms (weakness, fatigue, paresthesias, cramps), as compared with 6 of 44 normal subjects (14 percent) [90]. However, in another study of more than 2000 older adult individuals, functional mobility (gait speed and walking endurance) was better in those with mild elevations in TSH (4.5 to <7.0 mU/L) than in those with TSH values within the normal range (0.4 to <4.5 mU/L) [14].
●Subclinical hypothyroidism may also be associated with defects in verbal memory and executive functioning [91,92]. These defects correct with T4 therapy and are thought to reflect abnormal hippocampal function rather than general cognitive slowing. (See "Neurologic manifestations of hypothyroidism".)
●In a population-based study, subclinical hypothyroidism was associated with an increased risk of Alzheimer disease in women but not in men [93]. (See "Neurologic manifestations of hypothyroidism".)
●In one study, patients with subclinical hypothyroidism were more likely to have common bile duct stones, thought to be secondary to sphincter of Oddi dysfunction [94,95].
●Using Mendelian randomization analysis, increasing TSH (even within the normal reference range) was associated with reduced eGFR [96].
EFFECTS OF THYROID HORMONE REPLACEMENT — The fundamental clinical question regarding patients with subclinical hypothyroidism is whether they should be treated with thyroid hormone (T4 [levothyroxine]). Based upon the natural history alone, most experts and all professional groups recommend that treatment should be started to prevent progression to overt hypothyroidism in patients with serum TSH values ≥10 mU/L. The treatment of patients with TSH values between 4.5 and 10 mU/L remains controversial as relatively short-term randomized trials have not shown a benefit with treatment. In general, the data are limited by differences among the patient populations studied such as age, TSH concentrations, presence of antithyroid antibodies, and history of prior thyroid disease. (See 'Candidates for T4 replacement' below.)
Hypothyroid signs and symptoms — The results of systematic reviews and meta-analyses of trials in which most patients had TSH <10 mU/L suggest that treatment of subclinical hypothyroidism with T4 does not result in significant improvement in hypothyroid symptoms, general quality of life, depression, and/or cognitive function [97,98].
As an example, in a meta-analysis of 21 clinical trials (2192 nonpregnant adults with baseline TSH values ranging from 4.4 to 12.8 mU/L), there were no differences in hypothyroid signs or symptoms (standardized mean difference [SMD] 0.01, 95% CI -0.12 to 0.14) or general quality of life (SMD -0.11, 95% CI -0.25 to 0.03) between the T4 and placebo groups [98].
In the largest trial included in the meta-analysis, 737 adults (mean age 74.4 years) with subclinical hypothyroidism (mean TSH 6.4 mU/L) were randomly assigned to receive T4 or placebo [99]. After one year, there was no difference in hypothyroid symptoms or tiredness scores. Prespecified subgroup analysis according to baseline TSH level did not reveal a subgroup of patients who benefitted from T4. In subsequent analyses of this trial, the following findings were reported:
●There was no difference between the two groups in hypothyroid symptoms or tiredness scores in the subgroup of participants ≥80 years (data was combined with data from another trial of levothyroxine in 105 individuals ≥80 years with subclinical hypothyroidism) [100].
●There was no difference in depressive symptoms between the two groups [101].
●There was no improvement in gait speed, handgrip strength, or muscle mass between the two groups (data was combined with data from another trial) [102].
There are few trials specifically evaluating T4 replacement in patients with baseline serum TSH concentration ≥10 mU/L [103,104]. In these trials, T4 replacement resulted in improved hypothyroid symptoms.
In patients with subclinical hypothyroidism and goiter secondary to autoimmune thyroiditis, T4 treatment may also decrease the size of the goiter. In a study of 13 patients with subclinical hypothyroidism, treatment with T4 was associated with a significant decrease (median 80 percent) in thyroid volume determined by ultrasound [105]. (See "Thyroid hormone suppressive therapy for thyroid nodules and benign goiter", section on 'Nontoxic goiter'.)
Cardiovascular disease
Major adverse cardiovascular events — Although T4 replacement has been shown to improve a number of cardiovascular risk factors and surrogate cardiovascular endpoints in patients with subclinical hypothyroidism, including cardiac output, left ventricular ejection fraction, dyslipidemia, markers of inflammation, vascular smooth muscle proliferation, vascular reactivity, endothelial function, and carotid intima media thickness, data demonstrating its ability to decrease cardiovascular events are limited [51,53,56,106-114].
There appears to be an age-related difference in cardiovascular outcomes with T4 replacement. As an example, in a meta-analysis of seven studies (two randomized trials and five observational studies) including 21,055 participants, treatment with T4 was not associated with cardiovascular mortality [115]. In a subgroup analysis, T4 replacement was associated with reduced cardiovascular morality in younger patients (age less than 65 to 70 years [RR 0.54, 95% CI 0.37-0.80]), but had no effect in older patients. This discrepancy in younger and older individuals may be related to unintended overtreatment with T4, which is more likely to have harmful cardiovascular effects in older than in younger individuals [116].
Observational studies that suggest a cardiovascular benefit of T4 replacement in younger but not in older individuals include the following:
●In an analysis of the United Kingdom General Practitioner Research Database, ischemic heart disease events were less common among patients aged 40 to 70 years with subclinical hypothyroidism treated with T4 (68 events in 1634 treated patients [4.2 percent] versus 97 events in 1459 untreated patients [6.6 percent]; hazard ratio [HR] 0.61, 95% CI 0.39-0.95) [117]. However, in individuals over age 70 years, there was no benefit of treatment (104 events in 819 treated patients [12.7 percent] versus 88 events in 823 untreated patients [10.7 percent], HR 0.99, 95% CI 0.59-1.33) [117].
●In a retrospective study of 12,212 patients from Denmark with subclinical hypothyroidism, there was no difference in the risk of myocardial infarction or cardiovascular mortality between treated and untreated patients [118]. However, those under age 65 years who were treated with T4 had reduced overall mortality compared with those who were not treated (incidence rate 4.2 versus 6.5 per 1000 person-years, incidence rate ratio 0.63, 95% CI 0.40-0.99). For patients over age 65 years, there was no difference in overall mortality between treated and untreated patients.
Additional studies in older adults that suggest no cardiovascular benefit of T4 replacement include the following [119-121]:
●In a retrospective Danish study of 61,611 patients with cardiovascular disease (CVD; 1192 patients with subclinical hypothyroidism, mean age 73.6 years), there was no difference in the risk of all-cause mortality or major adverse cardiac events (myocardial infarction, cardiovascular mortality, stroke, or all-cause hospital admissions) between patients treated or not treated with levothyroxine [119].
●In a randomized trial of levothyroxine or placebo for an average of 18 months in 185 individuals ≥65 years with TSH 4.6 to 19.99 mU/L, there was no difference in systolic function measured as left ventricular ejection fraction, or diastolic function measured as the ratio (E/e') of early mitral inflow velocity (E) to mitral annular early diastolic velocity (e') [120]. There were few individuals with TSH ≥10 mU/L, and mean achieved TSH values were 3.55 and 5.29 mU/L in the treatment and placebo groups, respectively. Therefore, the findings are mainly applicable to older adults with mild subclinical hypothyroidism, in whom treatment is generally not recommended. (See 'Candidates for T4 replacement' below.)
One observational study in older adults suggests potential harm with T4 replacement:
●In a small case-control study of patients aged >65 years with TSH 4.2 to 10 mU/L who did or did not die between 2012 and 2016, levothyroxine treatment was associated with increased mortality (HR 1.19, 95% CI 1.03-1.38), but not atrial fibrillation (or fracture) [122].
Long-term clinical trials in younger and older adults are needed to assess whether thyroid hormone replacement reduces CVD outcomes in subclinical hypothyroidism. (See "Cardiovascular effects of hypothyroidism".)
Serum lipid and apoprotein concentrations — The effect of T4 replacement on lipids is uncertain. Some trials suggest benefit, but the results are inconsistent and the magnitude of the effect is of uncertain clinical significance [97]. In the meta-analysis of 12 clinical trials (nine trials with TSH concentrations <10, <12, or <15 mU/L) described above, seven of the trials analyzed lipid levels after treatment of subclinical hypothyroidism [89]. There were no significant effects of T4 replacement on total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, apolipoprotein A and B, or lipoprotein(a).
However, in several randomized trials of patients with subclinical hypothyroidism treated with T4 versus placebo, serum total and LDL cholesterol [53,104,106,123-125] and apoprotein B-100 concentrations [104,106,124] decreased significantly, whereas serum HDL cholesterol, triglyceride, and lipoprotein(a) concentrations did not change [104,125]. All but one of these trials included patients with serum TSH concentrations >10 mU/L.
Fertility — There are few data showing improved fertility outcomes in women with subclinical hypothyroidism treated with thyroid hormone. In observational studies of women with infertility and subclinical hypothyroidism, 44 to 84 percent of women treated with T4 successfully conceived during treatment [69,126,127].
In clinical trials of thyroid hormone replacement in patients with subclinical hypothyroidism undergoing assisted reproductive technologies (ART), treatment with T4 compared with placebo improved some pregnancy outcomes (eg, live birth rate, miscarriage rate) [128-130]. (See "Recurrent pregnancy loss: Management", section on 'Thyroid dysfunction and diabetes mellitus'.)
Other — There may be other benefits of thyroid hormone replacement, as illustrated by the following studies:
●In a study of 113 patients with chronic kidney disease and subclinical hypothyroidism, progression of renal failure was attenuated by treatment with T4 [131].
●Patients with coexisting iron deficiency anemia and subclinical hypothyroidism had a greater increase in hemoglobin when given both iron and thyroid hormone replacement when compared with those given iron alone [132].
MANAGEMENT — Thyroid hormone replacement should be administered only to patients with abnormal thyroid function tests. The degree of abnormality that warrants therapy is controversial. Thyroid hormone should not be prescribed for patients with hypothyroid symptoms but normal thyroid function. Although anecdotal reports suggest that T4 (levothyroxine) therapy may be beneficial in patients with symptoms of hypothyroidism but normal thyroid function tests, T4 was no more effective than placebo in increasing cognitive function and psychological well-being in a randomized crossover trial of 22 such patients [133].
The treatment of pregnant women with normal serum TSH levels and positive antithyroid peroxidase (anti-TPO) antibodies is reviewed separately. (See "Overview of thyroid disease and pregnancy", section on 'Thyroid peroxidase antibodies in euthyroid women'.)
Candidates for T4 replacement — Although virtually all experts recommend treatment of patients with serum TSH concentrations >10 mU/L, the routine treatment of asymptomatic patients with TSH values between 4.5 and 10 mU/L remains controversial (algorithm 1) [11,26,134-136].
Our approach is as follows:
●TSH ≥10 mU/L – In view of data linking subclinical hypothyroidism with atherosclerosis and myocardial infarction and the increased risk of progression to overt hypothyroidism, we suggest treatment of patients with subclinical hypothyroidism and TSH levels ≥10 mU/L. (See 'Cardiovascular disease' above and 'Progression to overt hypothyroidism' above.)
This recommendation is consistent with that of a clinical consensus group comprised of representatives from the American Thyroid Association (ATA) and the American Association of Clinical Endocrinologists (AACE) and with the European Thyroid Association guidelines [26,137].
●TSH 7.0 to 9.9 mU/L – In view of the report of increased cardiovascular mortality in younger individuals with TSH in this range [43], we treat most patients under age 65 to 70 years with TSH 7.0 or higher. However, due to the absence of benefit of treating older patients [100,117,118] and the concern that older patients may have complications from unintended overtreatment, we consider treatment only in older patients (ie, those age >65 to 70 years) who have convincing symptoms suggestive of hypothyroidism. (See 'Hypothyroid signs and symptoms' above and 'Cardiovascular disease' above.)
●TSH ≤6.9 mU/L, and above the upper limit of the reference range – We treat patients <65 to 70 years who have convincing symptoms suggestive of hypothyroidism and who have serum TSH values above the upper limit of the reference range to 6.9 mU/L. Patients with high titers of anti-TPO antibodies, who may rapidly progress to overt hypothyroidism, and patients with goiter may also benefit from early treatment.
We suggest not treating older patients (>65 to 70 years) with subclinical hypothyroidism and TSH above the upper limit of the reference range to 6.9 mU/L, since TSH values in this range are age appropriate, and we prefer to avoid treatment of patients over age 80 years. The upper limit of the reference range could be as high as 6 to 8 mU/L in healthy octogenarians [16]. (See 'Diagnosis' above and 'Mortality' above and 'Major adverse cardiovascular events' above.)
●Infertility or attempting pregnancy – We suggest initiating T4 replacement in women with subclinical hypothyroidism (TSH values above first trimester-specific normal reference range with normal free T4) who are trying to conceive and who have ovulatory dysfunction or infertility. (See 'Fertility' above.)
This is consistent with Endocrine Society and ATA guidelines [138,139].
Arguments for treatment — Treatment will prevent progression to overt hypothyroidism, especially in those with serum TSH concentrations greater than 10 to 15 mU/L and high serum anti-TPO antibody concentrations. (See 'Effects of thyroid hormone replacement' above.)
There are few data to show benefit or harm of T4 treatment in patients with TSH values between 4.5 and 10 mU/L. Treatment in patients with lesser elevations in serum TSH concentrations may possibly ameliorate nonspecific symptoms of hypothyroidism, such as fatigue, constipation, or depression, and may decrease the size of goiter, if present.
Some experts suggest that the presence of risk factors for cardiovascular disease (CVD) is a reason for initiating treatment [11]. Treatment may improve cardiac contractility and serum lipid concentrations in some patients, correction of abnormal serum lipid concentrations may be cardioprotective, and there is little risk associated with monitored T4 replacement [134,135,140]. (See 'Cardiovascular disease' above.)
Arguments against treatment — Arguments against T4 treatment include its cost (for both the hormone and for monitoring its efficacy), the lifelong commitment to daily medication in asymptomatic patients, the potential risk of overtreatment and inducing symptoms from excess thyroid hormone, and the possible induction or exacerbation of angina pectoris or cardiac arrhythmia in susceptible patients [141], especially in view of data from a community survey showing that 41 percent of patients over age 65 years taking thyroid hormone replacement had a subnormal serum TSH [142].
Although these concerns are not usually sufficient to counterbalance the potential benefits of therapy in younger patients, we do recommend a higher TSH threshold for treating older patients, especially since the upper limit of normal for serum TSH may be higher in this age group. (See "Diagnosis of and screening for hypothyroidism in nonpregnant adults".)
If the patient is not treated, regular follow-up is indicated. (See 'Monitoring untreated patients' below.)
Dosing and monitoring — There are two approaches to initiating T4 therapy:
●One option is to start with the lowest dose necessary to normalize the serum TSH concentration, typically 25 to 50 mcg daily. This approach will avoid overtreatment and is most appropriate in older adults or in patients with underlying CVD.
●An alternative approach for younger patients with Hashimoto's thyroiditis who do not have autonomy (eg, a prior history of toxic adenoma or Graves' disease treated with radioiodine) and who have normal negative feedback is to initiate treatment at slightly below full replacement doses (1.6 mcg/kg/day). This approach obviates the need for periodic dose increases, should there be progressive autoimmune destruction of their gland. Alternatively, younger patients could be started on low-dose therapy, as is recommended for older patients.
The goal of therapy is to reduce the patient's serum TSH concentration into the normal reference range. Since the mean serum TSH for the general population is around 1.4 mU/L, with 90 percent having serum TSH levels <3.0 mU/L, many experts recommend a therapeutic TSH target of 0.5 to 2.5 mU/L in young and middle-aged patients.
In patients aged ≥65 to 70 years who are treated (whether with initial serum TSH levels >10 mU/L or between 7 and 9.9 mU/L), a target serum TSH of 3 to 6 mU/L, consistent with the normal age-related increase in serum TSH, is appropriate. (See "Treatment of primary hypothyroidism in adults", section on 'Goals of therapy'.)
Serum TSH should be evaluated six weeks after initiation of therapy, and if the target TSH has not been reached, the dose of T4 should be increased by 12.5 to 25 mcg a day. If the serum TSH is too low, the T4 dose should be decreased by the same amount. After any adjustment in T4 dose, the serum TSH should be rechecked in six weeks. Once the correct dose has been achieved, serum TSH levels can be assessed annually. (See "Treatment of primary hypothyroidism in adults", section on 'Dose and monitoring'.)
Monitoring untreated patients — For patients with subclinical hypothyroidism who do not receive thyroid hormone replacement, we repeat thyroid function tests (TSH, free T4) initially at six months and, if stable, yearly thereafter.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Hypothyroidism".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Hypothyroidism (underactive thyroid) (The Basics)")
●Beyond the Basics topics (see "Patient education: Hypothyroidism (underactive thyroid) (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Diagnosis – The diagnosis of subclinical hypothyroidism is based upon biochemical testing alone. It is defined biochemically as a normal serum free thyroxine (T4) concentration in the presence of an elevated serum thyroid-stimulating hormone (TSH) concentration. Subclinical hypothyroidism may occur in the presence or absence of mild symptoms of hypothyroidism. (See 'Diagnosis' above.)
For nonpregnant adults, an elevated serum TSH is defined as a TSH concentration above the upper limit of the normal TSH reference range, which is typically 4 to 5 mU/L in most laboratories. (See 'Diagnosis' above.)
For pregnant women, elevations in TSH should be defined using population and trimester-specific TSH reference ranges. For women trying to conceive who have ovulatory dysfunction or infertility, elevations in TSH can be defined using first trimester-specific TSH reference ranges. (See "Overview of thyroid disease and pregnancy", section on 'Trimester-specific reference ranges'.)
●Consequences of subclinical hypothyroidism – A substantial proportion of patients with subclinical hypothyroidism eventually develop overt hypothyroidism. Subclinical hypothyroidism may also be associated with an increased risk of cardiovascular disease (CVD; eg, coronary heart disease [CHD], heart failure), particularly when the serum TSH concentration is above 10 mU/L. (See 'Progression to overt hypothyroidism' above and 'Cardiovascular disease' above.)
●Management – The degree of TSH abnormality that warrants therapy is controversial.
•Selection of patients for therapy
-TSH ≥10 mU/L – For patients with subclinical hypothyroidism and TSH concentrations ≥10 mU/L, we suggest treatment with thyroid hormone (T4 [levothyroxine]) (Grade 2C) (algorithm 1). (See 'Candidates for T4 replacement' above.)
-TSH between 4.5 and 10 mU/L – The routine treatment of asymptomatic patients with TSH values between 4.5 and 10 mU/L is controversial. Our approach to treatment is based on patient age, degree of TSH elevation, and the presence or absence of symptoms (algorithm 1). For older patients (>65 to 70 years) with subclinical hypothyroidism and TSH ≤6.9 mU/L, and above the upper limit of the reference range, we suggest not treating (Grade 2C), since TSH values in this range are age appropriate and in view of the potential for both cardiovascular and skeletal morbidity associated with inadvertent overtreatment. (See 'Candidates for T4 replacement' above.)
-Infertility – For women with subclinical hypothyroidism (TSH values above first trimester-specific normal reference range with normal free T4) who are trying to conceive and who have ovulatory dysfunction or infertility, we suggest initiating T4 replacement (Grade 2B). (See 'Candidates for T4 replacement' above and 'Fertility' above.)
•Dosing – Synthetic T4 is the treatment of choice for correction of hypothyroidism.
-Older patients and/or underlying CVD – For older patients and those with underlying CVD, the initial dose of T4 is typically 25 to 50 mcg/day. This approach will avoid overtreatment.
For patients aged >65 to 70 years who are treated, whether with initial serum TSH levels >10 mU/L or 7 to 9.9 mU/L (algorithm 1), the goal of therapy is a serum TSH of 3 to 6 mU/L, consistent with the normal age-related increase in serum TSH. (See 'Dosing and monitoring' above and "Treatment of primary hypothyroidism in adults", section on 'Standard replacement therapy'.)
-Younger patients – For younger patients without a history of thyroid autonomy (eg, without a prior history of toxic adenoma or Graves' disease treated with radioiodine) and who have normal negative feedback, an alternative approach is to initiate treatment at slightly below full replacement doses (1.6 mcg/kg/day). The goal of therapy is to reduce the patient's serum TSH concentration into the lower half of the age-adjusted normal reference range. This approach obviates the need for periodic dose increases, should there be progressive autoimmune destruction of their gland. Alternatively, younger patients could be started on low-dose therapy, as is recommended for older patients. (See 'Dosing and monitoring' above.)
•Monitoring untreated patients – For patients with subclinical hypothyroidism who do not receive thyroid hormone replacement, we repeat thyroid function tests (TSH, free T4) initially at six months and, if stable, yearly thereafter. (See 'Monitoring untreated patients' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
