Rationale and impact
- Information for people with thyroid disease, their families and carers
- Indications for tests for thyroid dysfunction
- Tests when thyroid dysfunction is suspected
- Tests for people with confirmed primary hypothyroidism
- Managing primary hypothyroidism
- Follow-up and monitoring of primary hypothyroidism
- Managing and monitoring subclinical hypothyroidism
- Tests for people with confirmed thyrotoxicosis
- Initial management in primary/non-specialist care for people with thyrotoxicosis
- Treatment for adults with Graves' disease
- Treatment for adults with toxic nodular goitre
- Treatment for children and young people with Graves' disease or toxic nodular goitre
- Antithyroid drugs for people with hyperthyroidism
- Follow-up and monitoring of hyperthyroidism
- Managing and monitoring subclinical hyperthyroidism
- Investigating non-malignant thyroid enlargement with normal thyroid function
- Managing non-malignant thyroid enlargement
These sections briefly explain why the committee made the recommendations and how they might affect practice. They link to details of the evidence and a full description of the committee's discussion.
The committee agreed it was important for all people with thyroid disease to understand the disease, the goals of treatment and the complex interaction between thyroid function tests and symptoms. The specific recommendations for people with hypothyroidism centred around the use of thyroid hormone replacement. Levothyroxine is frequently taken incorrectly, which can lead to suboptimal treatment. The specific recommendations for thyrotoxicosis centred around what people should know about their treatment options and the consequences of untreated thyrotoxicosis. The specific recommendations on thyroid enlargement focused on reassuring people that enlargement is common and generally non-cancerous, while also alerting them to red flag symptoms that should prompt further action.
The recommendations provide guidance on the type of information and support that should be provided to people with thyroid disease so that they can make informed decisions about management. The committee noted that currently there is variation in the information provided. In some areas there may be additional resource use, for example, if longer or more consultations are needed. But this may lead to later benefits and reductions in resource use, if, for example, it leads to better understanding and use of medication.
Full details of the evidence and the committee's discussion are in evidence review A: information for people with thyroid disease.
The committee noted that thyroid dysfunction affects many systems in the body, and the symptoms are often non-specific. They agreed, based on evidence and their experience, that most single common symptoms alone are not predictive of thyroid dysfunction. The decision to test should be based on an overall clinical suspicion, taking into account the nature and severity of symptoms, clinical signs and coexisting conditions.
The evidence showed that type 1 diabetes, an autoimmune disease, is associated with thyroid dysfunction. In the committee's experience, this is also likely to apply to other autoimmune diseases and justifies testing for thyroid dysfunction in these conditions.
There was little evidence on thyroid disease in people with atrial fibrillation. However, the committee agreed that the potential importance of thyroid disease and its impact on the treatment of atrial fibrillation is sufficient to justify testing.
Limited evidence showed that depression can be associated with thyroid dysfunction. The committee agreed that, in their experience, this can also apply to anxiety.
The committee noted that in children and young people, thyroid dysfunction may be accompanied by abnormal growth or a change in behaviour or school performance that is unexplained by other factors. They agreed, based on their experience, that testing for thyroid dysfunction should be considered for children and young people presenting with those features. Thyroid function tests may be abnormal in people who are acutely unwell with non-thyroid disease and the committee agreed that decisions on treatment of thyroid dysfunction should not be based solely on these results. Tests for thyroid dysfunction should be performed when the acute illness has resolved, unless the acute illness may be due to thyroid dysfunction.
Evidence showed that type 2 diabetes is not associated with thyroid dysfunction, so the committee concluded that thyroid function tests should not be performed solely because a person has this condition.
The recommendations to test for thyroid dysfunction in people with autoimmune disease, atrial fibrillation, anxiety or depression are broadly in line with current practice. Checking thyroid function in people with type 2 diabetes and during acute illness is also current practice in some areas. The recommendations to avoid this should be cost saving for the NHS.
Full details of the evidence and the committee's discussion are in evidence review B: indications for testing.
No evidence was identified on which tests should be used when thyroid dysfunction is suspected so the committee used their experience to develop the recommendations. The committee agreed that in general TSH alone is an appropriate first test for people in whom thyroid dysfunction is suspected. Subsequent tests (cascading) are only needed if TSH is abnormal (with FT4 if the TSH suggests hypothyroidism and both FT4 and FT3 if the TSH suggests hyperthyroidism). This approach reduces unnecessary testing compared with simultaneous TSH, FT4 and FT3 testing for all people. However, tests should be done in a way to minimise potential delays and the need for additional appointments, for example, by laboratories keeping original samples and performing subsequent tests on the same samples. The committee agreed based on their experience that this approach did not apply to adults in whom secondary thyroid dysfunction is suspected or in children and young people, where both TSH and FT4 are needed by default because of the differing likely causes of dysfunction. The committee further agreed that tests may need repeating when new symptoms develop or worsen, but that this should not be within 6 weeks of the last test because this is unlikely to provide new information.
Where FT4 is currently a routine test for thyroid dysfunction, cascading will reduce NHS costs by avoiding extra tests for people with a TSH within the reference range. In areas where FT3 is not currently being measured, cascading will mean a cost increase. But this will be offset by the benefits of correctly diagnosing and managing thyrotoxicosis.
Full details of the evidence and the committee's discussion are in evidence review C: thyroid function tests.
No evidence was identified on the use of antibodies to investigate hypothyroidism so the committee used their experience to develop the recommendations. They agreed that testing for thyroid peroxidase antibodies (TPOAbs) may be useful in the early investigation of the underlying cause of hypothyroidism. However, for adults there was no role for remeasuring TPOAbs because changes in levels are unlikely to guide treatment decisions. The committee agreed that for young people it may be useful to repeat TPOAbs at the point of transition to adult services.
The recommendations broadly reflect current practice so the committee agreed there should be no substantial change. By avoiding re-testing in adults, there could be some cost savings.
Full details of the evidence and the committee's discussion are in evidence review D: tests for confirmed primary hypothyroidism.
The committee agreed that hypothyroidism needs thyroid hormone replacement. Potential treatments are levothyroxine, usually prescribed to everyone, liothyronine, which is sometimes prescribed when levothyroxine fails, and natural thyroid extracts (which is currently unlicensed for use in the UK). Overall the evidence from 7 randomised controlled trials suggested that combination treatment with levothyroxine and liothyronine did not offer any important health benefits compared with levothyroxine monotherapy and was significantly more expensive. However, the committee noted that some of the trials did show some small benefits in specific quality of life domains and anecdotal evidence from some committee members suggested beneficial effects of combination treatment with levothyroxine and liothyronine in small subgroups of patients. The committee were aware that some people reported still feeling unwell with levothyroxine monotherapy and agreed that in this group adding liothyronine could potentially have greater benefit than in the general population with hypothyroidism, although there are no trials in this population. Some evidence suggested that combination therapy with levothyroxine and liothyronine could be harmful because it may suppress the production of TSH and its long-term adverse effects are uncertain. The committee was aware that the use of combination therapy is a critical issue in hypothyroidism. They could not recommend liothyronine either alone or in combination treatment based on the evidence available and its current list price but agreed a research recommendation to help inform future guidance in this important area. See also NHS England liothyronine advice for prescribers.
The committee agreed that the evidence for natural thyroid extracts showed no benefit over levothyroxine. The committee also noted that the proportion of T3 to T4 is higher in natural thyroid extracts than produced in the human body and the adverse effects are uncertain. Natural thyroid extracts are an unlicensed medication in the UK and overall the committee agreed they should not be offered.
Some evidence showed that a high starting dose of levothyroxine produced more rapid improvements in quality of life than a lower starting dose followed by titration. The committee agreed that this was also their experience and therefore recommended a high starting dose (1.6 micrograms per kilogram body weight per day) in adults unless contraindicated (adults over 65 or with a history of cardiovascular disease). Although evidence about dosing was very limited, the committee agreed that adults over 65 years are more likely to have cardiovascular comorbidities. Most studies of hypothyroidism and subclinical hypothyroidism use 65 as a cut-off when defining older adults. The committee agreed to recommend a lower starting dose with titration for people over 65.
The committee were unable to make recommendations on iodine or selenium supplements because of a lack of evidence.
The recommendations on thyroid hormone replacement and natural thyroid extract reinforce current practice and are not expected to have a significant cost impact. Currently everyone is offered levothyroxine as thyroid hormone replacement and small subgroups of people who do not feel well on levothyroxine are sometimes offered liothyronine.
Full details of the evidence and the committee's discussion are in evidence review E: managing hypothyroidism.
Evidence showed no clinically important benefits of maintaining TSH levels in the lower rather than higher end of the TSH reference range. Given the need for additional medication to achieve a TSH level in the lower end of the reference range, with the potential for adverse effects and increased cost, the committee concluded that as a starting point TSH levels could be maintained at any point within the reference range. Nevertheless, the committee acknowledged that some people may still have troublesome symptoms even with TSH levels in the reference range. Therefore, they recommended adjusting the dose of levothyroxine if symptoms persist to achieve optimal wellbeing for individual patients. The committee also agreed that it was important not to use doses high enough to cause TSH suppression or thyrotoxicosis.
The committee agreed that TSH levels can take up to 6 months to return to the reference range if they have previously been very high or have been high for a long time. They agreed that healthcare professionals should take this into account when adjusting doses, to avoid large dose increases that could cause thyrotoxicosis.
The committee based recommendations about the timing of testing on their experience. They made separate recommendations for children under 2 years because in this age group the impact of poorly treated hypothyroidism can be most severe, and the child is developing at such a rate that frequent dose changes may be needed.
The committee used their experience to agree which thyroid function tests are needed for monitoring. They followed the general principle that once TSH has stabilised in the reference range, TSH testing alone is enough if the person has no symptoms suggesting thyroid dysfunction.
They agreed that children should have more frequent monitoring to ensure that dose adjustments are made promptly and because in the very young, under-treatment can lead to serious neurodevelopmental consequences.
The committee agreed that generally the testing strategies are in line with current practice, although there may be some variation across the country. The recommendations should help patients (and their support groups) to advocate for their own monitoring and treatment. Doctors will have clear guidance on monitoring to assist with consultations.
Full details of the evidence and the committee's discussion are in evidence review F: monitoring thyroid disease.
There was little evidence on treatment for people with subclinical hypothyroidism, with most of the evidence relating to older adults. The committee agreed that as most studies used 65 years as a cut-off it was appropriate to define older adults as over 65 and make separate recommendations for this group.
The committee discussed the tendency to over-rely on TSH levels when making decisions about treatment. They agreed that factors suggesting underlying thyroid disease should also be taken into account when deciding whether or not to treat subclinical hypothyroidism.
The committee noted that a TSH level of 5 to 10 mlU/litre might return to the reference range without treatment in around half of people, whereas a TSH level above 10 mlU/litre is less likely to do so and is more often associated with symptoms. They therefore agreed that levothyroxine should be considered for all adults with a TSH level of 10 mlU/litre or more because this may improve symptoms and may have long-term benefits including on cardiovascular outcomes. For people with a TSH level lower than 10 mIU/litre, the committee agreed based on their experience that treatment was less likely to have a benefit but that the balance of risks to benefits was most favourable for adults under the age of 65. The committee noted that for people over 65 there was less likely to be an improvement in symptoms and the potential for harms from suppressing TSH (such as atrial fibrillation) is greater. The committee agreed that the trial of levothyroxine treatment should be stopped if symptoms persist with TSH levels within the reference range, as they are likely to be due to causes other than hypothyroidism.
The committee also agreed that the recommendations on testing TPOAbs in overt hypothyroidism applied to subclinical hypothyroidism because testing would help to inform the decision on whether or not to treat.
Because of the small amount of evidence available on the treatment of subclinical hypothyroidism, the committee made recommendations for research on selenium and iodine.
In children and young people there are a number of different causes of a mild increase in TSH besides autoimmune thyroid disease. These include mild congenital hypothyroidism, a low iodine intake (for example, because of a special diet), intercurrent illness, adrenal insufficiency and the paradoxical 'increase' in TSH observed in children with secondary hypothyroidism. The committee agreed that there is no urgency to treat with levothyroxine if thyroid hormone levels are appropriate for age. It is important to make a diagnosis before offering any treatment. The committee recommended caution when considering levothyroxine for children and young people whose thyroid dysfunction was unexplained. However, they also noted that in the very young it would not be appropriate to wait as long as for adults (3 months) to confirm a raised TSH with a second test.
The committee highlighted that current practice for monitoring of subclinical hypothyroidism varies considerably, with unnecessary testing often being undertaken. They agreed that the recommendations are likely to reduce the number of tests overall.
Full details of the evidence and the committee's discussion are in evidence review G: managing subclinical hypothyroidism.
The committee agreed that it is crucial to determine the cause of thyrotoxicosis because this affects management decisions. Antithyroid drugs are unlikely to cure toxic nodular disease but may induce remission in Graves' disease. TRAbs testing provides confirmation of clinical features that suggest Graves' disease. Getting an accurate diagnosis sooner benefits the patient because the appropriate treatment options would be offered first. If TRAbs levels are high it is unlikely that antithyroid drugs will induce remission of Graves' disease.
Evidence suggested that in adults, the diagnostic accuracy of TRAbs testing for Graves' disease was high across different cut-off values. Evidence also showed the accuracy to be high in children. However, the evidence was limited in terms of the number of studies and the study sizes. But it was in line with the committee's experience and current practice, so they agreed to recommend TRAbs testing for Graves' disease in both adults and children and young people.
Evidence on the diagnostic accuracy of TPOAbs testing was not available in either adults or children. The committee agreed that based on their experience, TPOAbs testing alone is not likely to be as useful as TRAbs testing for the diagnosis of Graves' disease, but it could be used in children and young people where the absence of TRAbs but presence of TPOAbs indicates that thyrotoxicosis is more likely to resolve spontaneously.
Evidence for the diagnostic accuracy of ultrasound was limited in both adults and children. Based on clinical experience, the committee agreed that ultrasound was useful for the diagnosis of Graves' disease but only when there were palpable thyroid nodules.
The committee agreed that technetium scanning may be useful when TRAbs are negative and Graves' disease is suspected because generalised uptake on the scan suggests Graves' disease.
Over recent years, TRAbs testing has become more widely available and more centres in the UK are using it to confirm the diagnosis of Graves' disease. However, some centres continue to use TPOAbs testing. If TRAbs testing allows more accurate differentiation between the different causes of thyrotoxicosis, there are likely to be reductions in unnecessary antithyroid treatment (including surgery) for people with transient thyroiditis and more timely and appropriate treatment choices for people with toxic nodular hyperthyroidism. The committee anticipates that TRAbs testing will become standard best practice for all UK centres leading to a correct diagnosis of Graves' disease for more people.
The use of technetium scanning for adults who are TRAbs negative reflects current practice in most centres.
Although thyroid ultrasound has only a limited role in the investigation of suspected Graves' disease, many healthcare professionals offer this investigation. This often results in incidental findings of doubtful clinical significance leading to further unnecessary investigations and interventions. The recommendation will discourage healthcare professionals from using thyroid ultrasound routinely in the investigation of suspected Graves' disease, which is likely to be cost saving.
Full details of the evidence and the committee's discussion are in evidence review H: tests for people with confirmed thyrotoxicosis.
Based on their experience, the committee reminded healthcare professionals that transient thyrotoxicosis does not need definitive treatment. They recommended that short-term treatment with antithyroid drugs may be needed until decisions about the most appropriate treatment can be made in specialist care. The committee also agreed that treatment with antithyroid drugs before radioactive iodine would minimise the rise in circulating thyroid hormone levels following this treatment and so reduce the symptoms of thyrotoxicosis. They acknowledged that it is important to render the patient euthyroid with antithyroid drugs before surgery to ensure patient safety.
Full details of the evidence and the committee's discussion are in evidence reviews I, J, K, L: managing thyrotoxicosis.
The evidence suggested that radioactive iodine produced better long-term outcomes than antithyroid drugs in terms of thyroid status, but with a greater risk of thyroid eye disease. There was no convincing evidence of a difference between radioactive iodine and surgery. The economic evidence showed that radioactive iodine offered a better balance of benefits and costs than surgery (total thyroidectomy) and was more cost effective than antithyroid drugs. Although exposure to radiation will always lead to some small increase in relative risk of cancer, the evidence showed that this did not translate into an absolute effect that was clinically important. The committee agreed nonetheless that continued follow-up of people who have undergone radioactive iodine treatment was important and the 'as low as reasonably practicable' (ALARP) principle applied. They also agreed to make a research recommendation on the long-term effectiveness and safety of exposure to radioactive iodine.
The committee agreed, based on the clinical and economic evidence, that radioactive iodine should be offered as first-line definitive treatment for most people with hyperthyroidism secondary to Graves' disease. However they noted a number of important exceptions and specified these in the recommendations. The committee acknowledged that circulating levels of thyroid hormones are likely to rise following radioiodine administration and pre-treatment with anti-thyroid drugs would make radioactive iodine safer and lead to reduced symptoms of thyrotoxicosis. The committee also agreed that the response to antithyroid drugs is better in some people than in others. For adults who are likely to have a particularly good response to antithyroid drugs (mild uncomplicated Graves' disease), radioactive iodine and a course of antithyroid drugs could be equally appropriate options.
Some studies have suggested that some people are more likely to relapse after antithyroid drugs. These include males, younger people, people who smoke, people with a large goitre, people with high levels of thyroid hormones at the time of diagnosis and people with high levels of TRAbs. However, most of the studies were small and retrospective. The committee agreed that it would be very helpful to confirm these findings in large prospective multi-centre studies. They made a research recommendation to inform future guidance.
The evidence did not identify a clinically important difference between a calculated or fixed strategy in terms of radioactive iodine dosing. A calculated strategy has an increased cost because of the need for imaging (usually ultrasound) and uptake measurements. There are theoretical benefits from a calculated strategy to administer a more precise dose that could reduce potentially unnecessary additional radiation exposure, but the evidence did not indicate that this precision translated to clinically important benefits. The committee's experience is that, in the UK, radioactive iodine is usually given without calculating the absorbed dose. The committee agreed that there was too much uncertainty around the impact of the differing strategies to make a recommendation and chose to make a research recommendation.
The evidence suggested no clinically important difference between surgical options (total and hemithyroidectomy) for Graves' disease, but tended towards a benefit of total thyroidectomy in terms of relapse rates and harm in terms of increased risk of hypoparathyroidism. These findings were consistent with the committee's own experience. The committee agreed to recommend total thyroidectomy for adults with Graves' disease having surgery. This was based on their experience that people opting for surgery are generally seeking a definitive treatment.
The committee was aware that the recommendations would result in radioactive iodine being offered as first-line definitive treatment to more people than currently. This is likely to be cost effective as shown by the economic evidence.
Full details of the evidence and the committee's discussion are in evidence reviews: I, J, K, L: managing thyrotoxicosis.
The committee used their experience and an extrapolation of the evidence from Graves' disease to recommend radioactive iodine as the first-line definitive treatment for hyperthyroidism secondary to multiple nodules. Surgery or life-long antithyroid drugs could be offered in some circumstances when radioactive iodine is inappropriate (for example, young mothers, people in nursing homes or people unable to adhere to radiation protection guidance) or surgery is likely to have additional benefits (for example, if malignancy is suspected). The committee used their experience to recommend that when surgery is chosen hemithyroidectomy should be considered for people with hyperthyroidism due to a single toxic nodule, and total thyroidectomy for people with hyperthyroidism and multiple toxic nodules. A hemithyroidectomy is a shorter procedure that removes less of the thyroid gland; this requires less time in hospital and leads to a lower risk of adverse effects like hypoparathyroidism and long-term hypothyroidism but has a greater risk of relapse of hyperthyroidism. The risk of relapse is greater for multiple toxic nodules, hence the different recommendations for different populations.
Full details of the evidence and the committee's discussion are in evidence reviews I, J, K, L: managing thyrotoxicosis.
No evidence was identified for treatments in children and young people with hyperthyroidism. Based on their experience, the committee recommended that antithyroid drugs should be first-line definitive treatment for children and young people with Graves' disease or hyperthyroidism secondary to a single or multiple toxic nodules. The committee agreed that the risks of radioactive iodine and surgery may be greater than for adults so it is important to get input from the multidisciplinary team and discuss these options with the child and their family.
When surgery is chosen, the committee recommended that this should be total thyroidectomy for Graves' disease or bilateral thyroid nodules. They based this on their experience and evidence in adults.
The recommendations to offer antithyroid drugs as first-line treatment are broadly in line with current practice and are unlikely to have a substantial cost impact. The potential risks and benefits (and therefore cost effectiveness) of radioactive iodine treatment in children are uncertain.
Full details of the evidence and the committee's discussion are in evidence reviews I, J, K, L: managing thyrotoxicosis.
Evidence showed a clinically important benefit in terms of normalising thyroid hormone levels and minor drug-related adverse events for methimazole or carbimazole compared with propylthiouracil. However, hypothyroidism was more frequent with methimazole or carbimazole. The committee questioned whether this was a result of over-treatment and thought hypothyroidism unlikely to be permanent. Because carbimazole, and not methimazole, is currently licensed in the UK, the committee recommended carbimazole as the drug of choice when offering an antithyroid drug for treating hyperthyroidism. The committee noted that carbimazole is not recommended in children under 2 years. In view of the potential risk of liver failure the committee agreed that propylthiouracil should not be the first choice of antithyroid drug. However, the committee noted the MHRA drug safety advice for carbimazole on contraception and the risk of acute pancreatitis and agreed that propylthiouracil is appropriate as an alternative for adults.
Evidence showed a clinically important benefit in terms of lack of relapse to hyperthyroidism and maintaining normal thyroid hormone levels with 12 to 18 months' treatment compared with 6 to 12 months' treatment. There was no clinically important difference in relapse following a longer treatment of more than 18 months.
The committee recommended that carbimazole should be offered for at least 12 to 18 months. They noted that this differed from the summary of product characteristics which advises 6 to 18 months, but agreed that the deviation was justified by the evidence. The committee agreed based on their experience and extrapolation from evidence in adults that the treatment duration for children should be reviewed every 2 years.
Evidence showed that block and replace (fixed high dose combined with levothyroxine) and titration (dose based on thyroid function tests) regimens of antithyroid drugs were similar in terms of minor drug-related adverse events (skin reactions). There were fewer relapses to hyperthyroidism with block and replace treatment compared with titration. But there was limited evidence suggesting more chance of agranulocytosis with block and replace regimens. The committee noted that block and replace treatment could theoretically provide greater stability and require fewer medical appointments than titration regimens. Therefore they recommended a choice of either regimen for adults with Graves' disease. Hyperthyroidism in children and young people should usually be managed with a dose titration regimen because of the increased risk of adverse events in this age group. The committee also made a recommendation for further research in this area.
The committee noted that people with hyperthyroidism secondary to a single or multiple toxic nodules will not go into remission and therefore discontinuing antithyroid drugs is not relevant. They agreed that titration regimens are generally more appropriate for this group.
Current practice in the UK, in adults and children, is a mix of block and replace (approximately 40%) and titration regimens (approximately 60%). The recommendations are broadly in line with current practice and unlikely to have significant resource impact.
Full details of the evidence and the committee's discussion are in evidence review J: management of thyrotoxicosis – antithyroid drugs.
No evidence was identified on the most appropriate ways to monitor hyperthyroidism, so the committee made recommendations based on their experience. The precise timing of monitoring depends on the treatment chosen, but in general the committee aimed to minimise unnecessary testing while ensuring early treatment failures or adverse effects (for example, hypothyroidism) were identified. Regardless of treatment option chosen, the committee agreed that in the long term TSH alone is sufficient for monitoring, although TSH alongside FT4 and FT3 will be needed in the short term after treatment. Should TSH become abnormal, having been stable within the reference range for a prolonged period, further tests will be necessary. Short-term combined testing with TSH, FT4 and FT3 is needed to inform decisions about the need for additional courses of treatment or dose changes with antithyroid drugs. As no evidence was found to support a strategy of routinely monitoring full blood count and liver function tests, and these tests have a treatment burden for people with hyperthyroidism and a resource impact, the committee recommended that healthcare professionals do not test unless there is a clinical suspicion of specific adverse effects of treatment.
More people with Graves' disease are expected to have monitoring because radioactive iodine is more likely to be offered as first-line treatment. This extra cost is likely to be justified, because radioactive iodine is more cost effective than other treatments for hyperthyroidism. In general, these recommendations reinforce best practice, with earlier detection leading to earlier treatment, reduced costs of treating complications and improved quality of life. The recommendation that there is no need to monitor full blood count and liver function after antithyroid drug treatment unless liver dysfunction or agranulocytosis are suspected is likely to be cost saving.
Full details of the evidence and the committee's discussion are in evidence review F: monitoring thyroid disease.
There was no evidence available on treating subclinical hyperthyroidism so the committee used their experience to develop the recommendations. They agreed that treatment might be suitable if subclinical hyperthyroidism is persistent and appears to be caused by intrinsic thyroid disease. However, the committee noted that there is currently no evidence that treatment offers benefits and it can have adverse effects. Overall the committee agreed that treatment decisions should be made with specialist advice. Treatment would be appropriate in those most likely to benefit, in other words those with very suppressed TSH and other features suggesting thyroid disease. The committee also agreed that they could not make specific recommendations about when to treat subclinical hyperthyroidism in children as specialist input would also be needed for this.
Several large population-based observational studies have shown that subclinical hyperthyroidism is associated with an increased risk of atrial fibrillation, osteoporosis, dementia, and death, including death from cardiovascular disease. Although most people with subclinical hyperthyroidism have no symptoms, an important question is whether treatment could improve long-term outcomes (for example, atrial fibrillation and dementia). The committee agreed to make a research recommendation to inform future practice.
There was no evidence available on monitoring subclinical thyroid dysfunction so the committee based the recommendations on their experience.
The overall aim of these recommendations is to ensure that if the subclinical thyroid dysfunction needs treatment, this will be identified in a timely manner but without subjecting a person to a lot of unnecessary tests.
The committee agreed that for children and young people monitoring may need to be more frequent.
Current practice in managing subclinical hyperthyroidism is variable. Some people are offered antithyroid drugs or radioactive iodine; surgery is very rare. Many people are offered no treatment. The recommendations are likely to reduce inappropriate treatment and to be cost saving.
The recommendations for monitoring subclinical thyroid dysfunction reflect good current practice.
Full details of the evidence and the committee's discussion are in evidence review M: management of subclinical thyrotoxicosis.
Evidence showed that ultrasound using established criteria is accurate for determining whether thyroid nodules need fine needle aspiration to investigate potential malignancy. The committee noted that many referrals for thyroid ultrasound are based on incidental findings of other types of imaging (for example, CT scans performed for other indications). They agreed that thyroid ultrasound should only be performed when a full assessment indicates a likelihood of malignancy. Thyroid ultrasound of incidental findings should not be the default option because most incidental findings are not malignant and further investigation may cause harms in terms of the adverse effects of testing and patient anxiety.
The evidence showed that different ultrasound criteria generally assessed the same features and had similar accuracy for detecting malignancy. Rather than recommending a specific set of criteria, the committee chose to list the essential features of any grading system (the lesion's echogenicity, border, shape, vascularity, presence of microcalcifications and cervical lymphadenopathy). Based on their knowledge that nodule size does not determine the likelihood of malignancy, and the observation that the criterion (SRU) that includes nodule size results in significantly lower sensitivity and specificity, they agreed not to include nodule size in the list.
Because healthcare professionals need to be able to re-visit ultrasound findings, the committee agreed that the grading system used for clinical decision making, including the specific nodule features examined to assess malignancy, should be specified in ultrasound reports. For the same reason, they agreed that ultrasound reports should also confirm that both lobes have been assessed and document the assessment of cervical lymph nodes.
The committee agreed that the recommendations should apply to children and young people as well as adults.
Evidence indicated that performing fine needle aspiration under ultrasound guidance provides greater sensitivity and specificity for determining the malignancy of thyroid nodules compared with palpation guidance.
The evidence also showed that with ultrasound guidance an inadequate sample is less likely than with palpation. Inadequate sampling is likely to add to the overall cost of palpation-guided fine needle aspiration because samples may need to be repeated or require more extensive investigation. The committee also noted that there were additional benefits of ultrasound guidance because ultrasonographic characteristics identified by simultaneous imaging provide more information about the risk of malignancy before cytology results are obtained.
Ultrasound is currently already used to assess the likelihood of thyroid malignancy, so the recommendations are not likely to make a significant impact. In the UK, the most commonly used ultrasound criteria are those of the British Thyroid Association, which are in line with the recommendations in this guideline.
The recommendation reflects current practice so the committee agreed that it should not have a significant impact.
Full details of the evidence and the committee's discussion are in evidence review N: imaging for fine needle aspiration and evidence review O: ultrasound guidance for fine needle aspiration.
The committee noted that there was little evidence on the efficacy of surgery for nodules; this was related to the challenges of comparing surgical and non-surgical interventions. In general, the committee agreed that surgery would be appropriate for nodules or enlargement causing symptoms, if there has been no response with other options or if there is true compression of nearby organs (for example, tracheal narrowing).
Aspiration is routinely done before more intensive intervention for cystic nodules because it is simple and can be done in the same appointment as preliminary investigation by a radiologist (or endocrinologist). The evidence generally showed a benefit of ethanol ablation over levothyroxine and equivalence with radiofrequency ablation.
The evidence showed no clinically important effect of levothyroxine on non-cystic nodules and a benefit of radiofrequency ablation and laser ablation. There was no evidence identified on radioactive iodine ablation although the committee noted that it is very commonly used in the UK for diffuse goitres that are causing symptoms, particularly if there is demonstrable radionuclide uptake. The committee also noted that the more recently developed techniques for percutaneous thermal ablation (for example, high-intensity focused ultrasound and microwave ablation) may be appropriate for some people but are not widely available. They made a research recommendation on percutaneous thermal ablation to inform future practice. The committee agreed not to recommend the use of levothyroxine due to the evidence suggesting no clinically important benefit for most outcomes and their awareness of adverse effects (for example, TSH suppression and increasing cardiovascular risk).
The committee noted that there was no evidence in children. They agreed that a healthcare professional should refer a child with thyroid enlargement to an appropriate multidisciplinary team, to ensure appropriate management as early as possible.
There was no evidence on monitoring thyroid enlargement but the committee agreed that, given the accuracy of ultrasound imaging, the risk of missing a malignancy or malignant transformation in an enlarged thyroid gland is low. However, they agreed that worsening symptoms or development of new symptoms may warrant repeating the ultrasound and TSH measurement. Suspicion of malignancy or compression would warrant repeating these tests.
The recommendations on surgical referral for goitre and non-malignant thyroid nodules are broadly in line with current clinical practice and therefore not expected to have significant impact.
The recommendations on managing non-malignant thyroid enlargement are also unlikely to have substantial resource impact. Radiofrequency ablation and laser ablation are not currently widely available. However it is current practice to only provide interventions to adults with nodules causing symptoms, as outlined in the recommendations. This limits the number of adults needing these interventions. Referral of children to a multidisciplinary team is not likely to occur often because thyroid enlargement is rare in this population.
The recommendation broadly reflects current practice. The committee recognised that there may be some centres in which routine monitoring is done, and the recommendation is likely to reduce this.
Full details of the evidence and the committee's discussion are in evidence review P: management of non-malignant thyroid enlargement, and evidence review F: monitoring thyroid disease.