Cardiac benefit from treating subclinical hypothyroidism
Overt hypothyroidism produces reversible changes in cardiovascular hemodynamics and in many of the modifiable cardiovascular risk factors for ASCVD and heart failure. Some prospective studies also indicate that treatment of subclinical hypothyroidism, including groups with minimally elevated TSH levels, results in improvement in surrogate markers for ASCVD such as atherogenic lipids (120–123) and carotid intima media thickness (126).
A meta-analysis of 10 longitudinal studies of subclinical hypothyroidism (119), which excluded patients with ASCVD at baseline, showed a relative risk of CHD of 1.2 when all studies were combined. When only higher quality studies were analyzed, the risk dropped to 1.02–1.08 depending on whether the study design allowed for adjudicated outcomes with or without knowledge of thyroid status. However, in studies with mean age younger than 65 years, the risk was 1.51 compared with 1.05 in studies with a mean age of 65 and over. Another meta-analysis, also done in 2008, of 15 studies with over 2500 participants with subclinical hypothyroidism, eight of which were also used in the aforementioned meta-analysis, showed elevated odds ratios for the incidence of ASCVD and cardiovascular all-cause mortality of 1.57 and 1.37 for those under 65 years, but not for those over 65 years (292).
A study from the Cleveland Clinic Preventive Cardiology Clinic of patients at high risk for ASCVD showed that those with TSH levels of 6.1–10 mIU/L as well as greater than 10 mIU/L who were under 65 years and not treated with thyroid hormone had higher all-cause mortality (118). Most recently a U.K. general practitioner database was analyzed to assess the impact of L-thyroxine treatment on fatal and nonfatal cardiac events in over 3000 individuals with subclinical hypothyroidism (TSH between 5.01 and 10 mIU/L) aged between 40 and 70 years and over 1500 individuals older than 70 years who were followed up for a median of 8 years. In the 50% of individuals between 40 and 70 years of age who were treated with L-thyroxine (87.4% women) the hazard ratio for ischemic heart disease events was reduced compared to the 50% of untreated individuals (82.5% women) (0.61, CI 0.49–0.92). This reduction was not evident in those older than 70 years, of whom 84.6% in the treatment group and 75.6% in the untreated group were women (293).
Yet other studies fail to show that an increased risk of cardiac disease in those with subclinical hypothyroidism is age dependent. The Cardiovascular Health Study followed 3000 patients 65 years or older with subclinical hypothyroidism who were initially free of heart failure. Those with TSH levels of 10 mIU/L or greater had an increased risk of heart failure (294). During the 20 years of follow-up in the Whickham Survey, an association was found between ASCVD and ASCVD-related mortality in those with subclinical hypothyroidism whose TSH values were between 6 and 15 mIU/L independent of age. When those treated with L-thyroxine were excluded, ASCVD-related morbidity and mortality were no longer evident (116). Additional large-scale studies in those with serum TSH values of 10 mIU/L or greater including a study of 11 prospective cohorts in the United States, Europe, Australia, Brazil, and Japan demonstrated an increase in ASCVD that was independent of age (115) while a study of six prospective cohorts with over 2000 patients had an increased incidence of heart failure in those up to 80 years of age (117).
The absence of randomized prospective controlled trials leaves us with several unresolved key issues pertaining to subclinical hypothyroidism, including whether or not L-thyroxine treatment will prevent the development of ASCVD or decrease the frequency of hospital admissions for heart failure and whether age is a critical determinant of risk for cardiac morbidity. A prospective study to assess both of these parameters is currently being planned.
Cognitive benefit from treating subclinical hypothyroidism
Some reports on mood, cognitive, and other objective brain function studies in subclinical hypothyroidism demonstrate the presence and reversal of deficits after treatment with L-thyroxine (295). However, other studies have not (296,297).
L-thyroxine and L-triiodothyronine combination therapy
An important question is whether a recent study had sufficient data to warrant revisiting why some patients seem to feel better on L-thyroxine/L-triiodothyronine combinations and whether we can identify them and safely treat them (136) with this combination.
A potential role for L-triiodothyronine monotherapy in lieu of L-thyroxine monotherapy was recently raised by a small randomized, double-blind crossover intervention study done comparing L-triiodothyronine monotherapy with L-thyroxine monotherapy in patients with hypothyroidism (298). Thrice daily dosing was employed for each. Comparable TSH levels were achieved. Mild weight loss and decreases in total cholesterol, LDL cholesterol, and apolipoprotein levels were seen without differences in cardiovascular function, insulin sensitivity, or quality of life with L-triiodothyronine monotherapy compared with L-thyroxine monotherapy. The small size and short duration of the study as well as thrice daily dosing presently precludes considering L-triiodothyronine monotherapy as an alternative to L-thyroxine monotherapy (298).
Thyroid hormone analogues
Thyroid hormone’s effects are protean, affecting virtually every organ system. Efforts are underway to develop and study analogues that have selective beneficial effects on weight control, lipoproteins, and TSH suppression without inducing hypothyroidism or the most important negative consequences of hyperthyroidism on the heart and skeleton. Compounds studied to date include D-thyroxine (299), tiratricol (275), eprotiromone (KB 2115) (300,301), and diodothyropropionic acid (302). A recent prospective Phase II clinical trial of the thyroid hormone analogue eprotirome, designed to be a selective beta II receptor agonist, has been shown to lower both total cholesterol and Lp(a) without any change in thyroid hormone levels or untoward cardiovascular or bone effects (300). However, the development program for eprotirome has been discontinued due to adverse findings in preclinical studies. Further studies will be needed to confirm the benefit and lack of side effects of these agents.
Screening for hypothyroidism in pregnancy
It remains unclear if screening for hypothyroidism in pregnancy is beneficial. A consensus statement in 2004 (106) and clinical practice guidelines in 2007 (303) and 2011 (10) found insufficient data to support a 1999 (304) and restated 2005 recommendation (305) for universal screening for thyroid dysfunction during pregnancy, but rather recommended aggressive case finding.
Arguments for screening include the following:
- Limiting evaluation to women in high-risk groups misses 30% of pregnant women with overt or subclinical hypothyroidism (306).
- A study comparing universal screening to case finding found that there was a statistically significant difference in a composite endpoint of adverse obstetric and neonatal outcomes associated with treatment of thyroid dysfunction in low-risk women who were screened compared to those who were not (307).
- A cost-effectiveness model to evaluate universal screening, which was predicated on the effectiveness of thyroid hormone treatment in lowering the incidence of offspring with intelligence quotient (IQ)<85, concluded that a random TSH done during the first trimester of pregnancy would ultimately save $84 per pregnancy (308). However, this has not been confirmed by a recent randomized controlled trial (219).
However, questions remain about the utility of screening those at low risk for developing hypothyroidism (307) and whether screening and intervention earlier on in the first trimester (219) may be cost effective.
The Controlled Antenatal Thyroid Study in the United Kingdom and Italy examined the impact at 3 years of age of L-thyroxine treatment if free T4 is below the 2.5th percentile or if TSH is above the 97.5th percentile (219). Analyses failed to demonstrate a benefit when screening was performed around the end of the first trimester. Whether earlier intervention, different cognitive testing, or the same testing performed at age greater than 3 years would yield different results is uncertain. “A Randomized Trial of Thyroxine Therapy for Subclinical Hypothyroidism or Hypothyroxinemia Diagnosed During Pregnancy”, done under the auspices of the National Institute of Child Health and Human Development, is presently studying the IQ at 5 years of age following a universal screening versus case finding program.
Agents and conditions having an impact on L-thyroxine therapy and interpretation of thyroid tests
Conditions such as pregnancy and malabsorption, drugs, diagnostic agents, dietary substances, and supplements can have an impact on thyroid hormone economy, which may or may not result in a change in thyroid status. For example, orally administered estrogens increase TBG levels. While this does not alter thyroid status in euthyroid individuals with normal thyroid reserve, it may do so when there is either marginal thyroid reserve or established hypothyroidism. Drugs may have multiple effects on thyroid hormone metabolism. Notable examples include glucocorticoids and amiodarone. In a number of cases, the mechanisms by which agents alter thyroid status are not known. The impact that an agent or condition has on thyroid status may require clinicians to increase monitoring, adjust dosages, or instruct patients to change how and when they take L-thyroxine.
Major determinants of whether or not drugs and other substances will have an impact on thyroid status include the following:
- Duration of action
- Proximity to when thyroid hormone is taken
- Duration of treatment
- Iodine content
- Size of iodine pool
- Autoimmune thyroid disease
- Nodular thyroid disease
- Thyroid hormone status
- Genetic factors
The principal mechanisms and reasons that conditions, drugs, and other substances have an impact on thyroid status are the following:
- Effects on thyroid hormone metabolism:
- Peripheral metabolism
- Direct and indirect effects on the hypothalamic–pituitary–thyroid axis
- TSH secretion
- Direct and indirect effects on the thyroid gland
- Iodine uptake
- Hormone production
- Hormone secretion
- Thyroiditis (amelioration or development)
- Amelioration or development of Graves’ disease
Table 10 lists agents and some conditions that affect thyroid status—particularly if they are commonly used—and are likely to do so or to have a profound impact on it. However, some very commonly used drugs such as sulfonylureas or sulfonamides or foodstuffs such as grapefruit juice that may only have a minor impact have been included. Because of their potential importance, some drugs, such as perchlorate, iopanoic acid, and ipodate, are also listed even though they are not generally available. On the other hand, some drugs that are rarely used have been omitted. Agents may appear more than once if there is more than one known mechanism of action. A comprehensive review of this subject and references for each drug or condition is beyond the scope of these guidelines. The interested reader is encouraged to consult other sources for more information (309–311).