Thyroid testing: When to worry (not often) and when to reassure


Thyroid problems can be worrisome, but relatively few children referred for workup for an abnormal thyroid test have clinically significant disease. Choosing the right screening tests can prevent these costly encounters without compromising children whose test results warrant close follow-up and possible intervention.

DR. KAPLOWITZ is chief of endocrinology, Children's National Medical Center, and professor of pediatrics, George Washington University School of Medicine, Washington, D.C.

Referrals for evaluation of "abnormal" thyroid tests account for at least 25% of appointment requests in our department of pediatric endocrinology. The great majority of these patients do not actually have thyroid disease, need no therapy, and in many cases, no follow-up testing. They nevertheless arrive at the endocrinologist's office for three basic reasons:

  • Because thyroid tests were ordered in the absence of any signs or symptoms suggestive of thyroid disease, or
  • Because so many tests were ordered that there is a high probability of at least one "abnormal" result-i.e., a result that falls outside what the lab defines as the normal range, or
  • Because the child's primary care practitioner does not differentiate between abnormalities that require referral and possible treatment and those that have no clinical significance.

This article will review some basic thyroid pathophysiology, pinpoint the indications for thyroid testing in children, and offer examples of test results that (most of the time) do not require referral to an endocrinologist.

How thyroid hormones regulate thyroid function

Thyroid hormones l-thyroxine (known as T4) and triiodothyronine (known as T3) have many important functions, but those that are most critical in children are promotion of normal myelination during brain development in the first two to three years of life and normal skeletal growth. In addition, thyroid hormones regulate the rate of metabolism. Thus a severe deficiency may result in fatigue, feeling cold, low heart rate, constipation, and slow growth if the problem has been present for long enough. It is important to note that mild but biochemically detectable deficiency is often asymptomatic, and that the actual symptoms of thyroid hormone deficiency (or excess) vary greatly from child to child.

T3 is more potent as a thyroid hormone than T4, but its blood levels are lower. About 20% of T3 comes directly from thyroid synthesis and 80% from peripheral metabolism, since many tissues can convert T4 to T3 by removing an iodine atom.

Thyroid hormone production is tightly regulated by pituitary secretion of thyroid stimulating hormone (TSH), which is in turn regulated by hypothalamic thyrotropin-releasing hormone (stimulatory) and somatostatin (inhibitory). The pituitary senses the level of unbound (free) T3. When that level falls below a critical set point, TSH increases which then stimulates the thyroid gland to produce and release more thyroid hormones. When thyroid hormone levels normalize, they induce feedback inhibition of TSH release, and TSH levels return to normal. Thus a mildly increased TSH (6 to 15 µIU/mL) may be an appropriate response to a transient decrease in thyroid hormone production, and does not always signify an immediate need to start thyroid hormone replacement.

Ups and downs of thyroid function

By far the most common cause of a progressive decline in thyroid function is autoimmune or Hashimoto's thyroiditis. The hallmark of this condition is a progressive rise in TSH (usually to levels above 20 µIU/mL), accompanied by a gradual fall in circulating thyroid hormone levels. In the early stages, thyroid hormone levels may still fall within the low-normal range, but eventually, as the thyroid fails, the free T4 level will become subnormal. In most but not all cases, the rise in TSH stimulates growth of the gland itself, resulting in a goiter palpable on physical examination. Goiter may appear early in the course of disease, while symptoms are absent. Eventually, as thyroid hormone levels fall well below normal, the classic symptoms noted above will appear, although their onset is so insidious they may go unrecognized. In about 90% of cases, elevated levels of anti-thyroid antibodies, particularly thyroid peroxidase (TPO) antibodies, will be present at the time of diagnosis, thus confirming a suspicion of autoimmune thyroiditis.

The other autoimmune problem that alters thyroid function is the production of antibodies directed at the TSH receptor on thyroid follicular cells. Often, these antibodies stimulate the thyroid cells as effectively as TSH, resulting in an increase in both T4 and T3 secretion. However, as T4 and T3 levels rise and TSH production is shut down by negative feedback, production of antibodies continues unabated, resulting in sustained elevation of T4 and T3 and, eventually, many of the classic symptoms of hyperthyroidism. These include inability to sit still and focus, heat intolerance, rapid heart rate, weight loss despite increased appetite, and occasionally excessive stooling. Exophthalmos is not related to elevation of T4 and T3, but to antibodies directed against the extraocular smooth muscles. This type of hyperthyroidism, also known as Graves disease, is at least five times more common in girls than in boys. Most school-aged patients will have a significant decline in their academic performance, due to their inability to focus, concentrate, and sit still.

To test or not to test

Thyroid tests should be ordered mainly in these situations:

  • Any child with an enlarged thyroid on physical examination
  • More than one sign or symptom of hypothyroidism
  • Goiter
  • New onset of fatigue
  • Cold intolerance
  • Acquired growth failure (may be only sign)
  • Constipation, dry skin
  • Follow-up of abnormal newborn screening test result
  • More than one sign or symptom of hyperthyroidism
  • Hyperactivity, impulsive behavior
  • Tachycardia
  • Unexplained weight loss
  • Heat intolerance
  • Declining school performance

Thyroid tests are not likely to be helpful in these situations:

  • Obesity, unless it is of recent onset and there is also slowing of linear growth. Even severe hypothyroidism generally causes only modest (five to 10 lb) weight gain, which is mostly water, not fat.1
  • Mild short stature (3rd percentile or above) with a normal rate of growth
  • Failure to thrive. Hypothyroidism in infants is almost always detected on newborn screening and will not affect weight gain more than linear growth
  • Hair falling out
  • History of thyroid disease in a relative without any thyroid-related signs or symptoms
  • Hyperactivity alone without goiter or any other hyperthyroid symptoms.

Which tests should you order?

Thyroid testing is one of those situations where less is more. More tests increase the chances of finding a trivial abnormality but are not more likely to uncover actual thyroid disease. In general, only two tests are needed for initial thyroid screening:

  • TSH. This is the essential screening test, because a result that falls within the normal range (between 0.5 and 5.5 µIU/mL in most laboratories) excludes primary hypo- or hyperthyroidism. The problem with TSH assays is the interpretation of mildly elevated or slightly decreased levels.
  • Free T4. This test has largely replaced total T4, since it measures only the biologically active fraction (0.03%) and is affected very little by changes in thyroid binding proteins. In contrast, a low total T4 can be seen in children (mostly males) who are born deficient in thyroid binding globulin (TBG) but have normal free T4 levels and are thus euthyroid. A teenage girl on birth control pills will often have an elevated total T4 due to an increased concentration of TBG but the free T4 will be normal. There are two methods for measuring free T4: the more accurate but more expensive and time-consuming "equilibrium dialysis" method, and the faster and less expensive "direct" or "non-dialysis" method. For screening purposes, the direct free T4 assay is quite satisfactory.

Which tests NOT to order

The following tests are not relevant or not worth ordering except under certain cirmstances:

  • T3 uptake. This test does not measure thyroid function at all, but thyroid-binding protein saturation. A high T3 uptake means binding proteins are more saturated than normal. When combined with a total T4, this test allows the "free thyroid index" to be calculated, thus correcting for low or high thyroid binding proteins. For example, a child with TBG deficiency has a low T4 and a high T3 uptake, so the product of the two, the free thyroid index, is normal. Since free T4 can be measured directly and inexpensively, there is no reason at all to order a T3 uptake. A somewhat low or high T3 uptake is the presence of a normal TSH level is not clinically relevant.
  • Total T3. Since T3 is the more active thyroid hormone, this test, in theory, should always be high in hyperthyroidism and low in hypothyroidism, like the total T4. In practice, it is not infrequently somewhat above the reference range when the more reliable thyroid tests (free T4, TSH) are normal, and thus gives misleading information. In addition, it is not consistently low in hypothyroidism. The only value this test has is for patients with clear-cut hyperthyroidism, in whom total T3 is sometimes very elevated when total and free T4 are normal to mildly elevated-a situation referred to as T3 toxicosis. Endocrinologists use this test to assess the initial severity of hyperthyroidism and its response to treatment, but primary care physicians should not order this test unless the child appears clearly hyperthyroid.
  • Free T3. This test works on the same principle as the free T4, but as a screening test is no more helpful than the total T3. Don't waste money on it.
  • Thyroid antibodies: thyroid peroxidase (TPO) and thyroglobulin (TG). These tests are of value when a child is found to have an elevated TSH level, as a very high level of antibodies strongly suggests a diagnosis of autoimmune thyroiditis. However, this test is positive in a relatively high proportion of the general population (10% to 11% in a recent national survey, with prevalence increased in females and with increasing age).2 It follows that a large number of people who have these antibodies are not hypothyroid and will never become hypothyroid. If the free T4 and TSH are normal, is it helpful to know that a child has positive antibodies? Although a few endocrinologists treat such patients on the theory that they have mild hypothyroidism despite the normal TSH, there are no controlled studies showing benefit and few pediatric endocrinologists would treat. Relative to the normal population, such children have a greater chance of becoming hypothyroid in the future, but that could be years down the road. In my view, annual thyroid testing for a healthy child with thyroid antibodies but no goiter and a normal T4 and TSH is not worthwhile. Others may disagree.

When tests are abnormal: Some common scenarios

The following six cases discuss what to do in some common situations when thyroid test results are abnormal.

Case 1.You are evaluating a child for short stature (or fatigue, or a positive family history of thyroid disease), and find that the free T4 is mid-normal, but the TSH is 0.3 (lab normal range 0.5-5.5). What should you do?

  • Discussion. A TSH which is low but greater than 0.1 with a normal total or free T4 has no clinical significance. This result does not account for any of the symptoms that may have prompted the thyroid testing, and should be viewed as a variation of normal. If the TSH is less than 0.1 and the free T4 is elevated, you probably have a hyperthyroid child, but even in early/mild hyperthyroidism, the TSH is invariably less than 0.1 and often less than 0.01. A low TSH is also not indicative of pituitary TSH deficiency. In such cases, the free T4 will be low and the TSH generally normal. Tell the child's parents that the thyroid tests are fine, and do not repeat them unless clinical symptoms of hyperthyroidism appear. I have never seen a case like this evolve into clinical hyperthyroidism.

Case 2. What if, on screening, the free T4 is normal but the TSH is 0.05 and the child has no clinical signs of hyperthyroidism?

  • Discussion. This situation, which some would call "subclinical hyperthyroidism" is less common than the child with a TSH of 0.1 to 0.5. Such children do sometimes progress over a period of months to overt hyperthyroidism, particularly if they have a goiter. However, we have recently seen an increase in such referrals, and have found that in many cases, if one waits two to three months and repeats the tests, the TSH normalizes. On the other hand, if one repeats them in a week, as is often done, the abnormality in TSH tends to persist. While one could refer such children right away, another strategy is to tell parents the tests should be repeated in two to three months. You might add a T3 to the next blood draw and TPO antibodies. If the second test has normalized, the TSH suppression was transient and no further testing is necessary. If the abnormality persists, non-urgent referral is recommended. A few such patients will be in the mildly hyperthyroid phase of Hashimoto's thyroiditis, which is more likely if the TPO antibody is elevated.3

Case 3.A patient who is being screened for thyroid disease due to obesity has a normal free T4 of 1.2, but the TSH is slightly increased to 7.0. There is no goiter. How aggressive should one be in further testing and treatment?

  • Discussion. This situation is extremely common and also rather controversial. Some refer to this situation as "subclinical hypothyroidism," the expectation being that many of these children will become overtly hypothyroid if one follows them over time. There are no large studies in the pediatric literature that allow one to estimate how often this occurs. In one study of 18 patients with both increased antibodies and mild to moderate TSH elevation followed for a mean of four years, seven had normalization of TSH, 10 had persistence of TSH elevation with normal T4, and only one developed overt hypothyroidism.4 In one adult study, progression to overt hypothyroidism was found to be 10 times more common when the TSH was 10.0 to 14.9 vs 5.5 to 9.9, and 40 times more common when the TSH was 15 to 19.9.5 Additional risk factors for progression were a goiter, an elevated level of TPO antibodies, and a borderline low free T4. A recent study in adults found that mean TSH levels are a bit higher in obese than normal weight individuals, accounting for the high frequency with which a slightly elevated TSH is seen in obese patients.6 There is no evidence that these patients are actually hypothyroid or that they would lose weight if started on thyroid hormone replacement.

Pediatric endocrinologists are split on whether to treat children with normal free T4, TSH of 5.5 to 10, and no goiter. Some believe that any slightly elevated TSH that is present on more than one test should be treated aggressively, while others believe this is usually a normal variation. My own experience is that few of these patients become overtly hypothyroid, at least in the short term (one to two years). I ask referring physicians to repeat the free T4 and TSH in six to 12 months, or sooner if a goiter appears or there are classic hypothyroid symptoms. About half the time, TSH levels fall back into the normal range, and half the time they remain in the 5.5 to 10 range. In only a few cases do TSH levels increase to more than 10, and even then overt hypothyroid symptoms are rare, but at that point the child should be seen. Parents need to be informed there are generally no signs or symptoms (obesity, fatigue, short stature) that can be attributed to a slightly elevated TSH when the free T4 is normal. Even though thyroid hormone replacement is safe and not very expensive, most parents would prefer to wait to see if the child develops more clear-cut hypothyroidism before starting treatment.

Case 4.A 10-year-old boy with hyperactivity is screened for thyroid disease and is found to have a slightly low free T4 of 0.82 (normal is 0.9 to 1.6) with a normal TSH of 2.7. How should one interpret such tests and is treatment or additional evaluation warranted?

  • Discussion. First, it is important to note that the child does not have primary hypothyroidism, or the TSH would be elevated. Secondary (also known as pituitary or central) hypothyroidism is far less common than primary hypothyroidism. It is most often is seen in the setting of known or suspected hypothalamic/pituitary disease, as in the child who has had surgery in or radiation to that region. Obviously the results in no way explain this child's hyperactivity. So what are the chances that you have stumbled upon a child with undiagnosed pituitary disease by finding a slightly low free T4? In my experience, quite low. Sometimes the abnormality persists, but often by the time I have seen the child, the free T4 has normalized. When the low free T4 persists, often it simply means that the child's free T4 is normal for him or her but falls outside the population normal range, which encompasses 95% of normal individuals.

One situation in which a low free T4 with normal TSH may be meaningful is in the child with significant short stature and a decreased rate of growth. Since pituitary deficiencies are often multiple, such a child could have a deficiency of both TSH and growth hormone-particularly if the insulin-like growth factor-1 level is also low. Certain drugs may also cause a low free T4 without a compensatory rise in TSH. The drugs which do this most often are carbamazapine (Tegretol) and the structurally similar anticonvulsant oxcarbazepine (Trileptal).

If you see a child with this problem, I suggest waiting a few months and repeating the test. If a direct free T4 was ordered the first time, order a free T4 by equilibrium dialysis the second time. If both are low, the child can be referred to a pediatric endocrinologist, but unless there is either a poor growth rate or a history of central nervous system (CNS) disease or new CNS symptoms, the chance of pinpointing a cause (e.g. a tumor near the pituitary) is quite low. Children with a low free T4 level who are taking the seizure medications just mentioned may not have true hypothyroidism; the test result may be due to displacement of T4 from thyroid-binding globulin. Most endocrinologists will treat to normalize the free T4 but will try to withdraw treatment if the child is taken off the seizure medication.

Case 5.A 3-week-old child who had a birth weight of 7 lb 6 oz is sleeping a lot, feeding poorly, and not gaining weight well. The child passed his newborn thyroid screening but you want to rule out hypothyroidism anyway. The free T4 comes back 2.1, the total T4 is 15.5 (both slightly elevated), and the TSH is 8.0 (also slightly elevated). What does this mean? Can the child be hypo- and hyperthyroid at the same time?

  • Discussion. Thyroid tests done in the perinatal period need to be interpreted knowing that T4, free T4, and TSH levels all have a normal range at that age that is somewhat higher than what one sees in older children. For example, some labs report an upper range of normal for free T4 of 2.3 and for TSH of 16 in the first month of life.7 For children older than 5 years, the upper ends of the normal range by the same methods are 1.6 for free T4 and 6 for TSH. As long as the TSH is not suppressed to less than 0.1, you can be confident the infant is not hyperthyroid. Hyperthyroidism at this age is rare and generally seen only in infants whose mothers have hyperthyroidism. In addition, a TSH of eight is not nearly high enough to diagnose congenital hypothyroidism. If parents need reassurance the child does not have a thyroid problem, repeating a free T4 and TSH in a month will generally show that the levels, if not completely normal, are trending towards the normal range.

Case 6. You find an enlarged, symmetric, and non-tender thyroid on a routine exam and obtain a free T4 and TSH, which are both normal. Should this child be referred?

  • Discussion. First, the parents should be told that the situation is not urgent, as it is very unlikely that treatment will be offered. There is little evidence that euthyroid goiters, which are fairly common, will shrink if thyroid hormone is given. Some physicians order thyroid ultrasounds routinely, but too often, it seems, they will detect very small non-palpable nodules (less than one cm in diameter) which are too small to biopsy but will generate lots of worry. It is best to let the specialist decide of an ultrasound is warranted. Few euthyroid goiters are due to thyroiditis, and most probably represent benign variations of normal. Rarely, what seems to be an asymmetric goiter may end up being a thyroid tumor. My strategy is to see these children at least once when the parent or physician requests it, measure the gland, make sure there are no nodules or masses by palpation (which would require an ultrasound) perhaps repeat the free T4 and TSH if either was borderline, and then refer the child back to the PCP for monitoring. If the gland should increase in size over time or if symptoms of hypo- or hyperthyroidism develop (not common), the free T4 and TSH should be repeated.

When is it time for an urgent referral?

By urgent referral, I mean not simply asking the parents to call and schedule an appointment but faxing the test results to your local pediatric endocrinologist in the hope that the child can be seen on an expedited basis. This list, you will notice, is short.

  • The TSH is greater than 20, particularly if the free T4 is low or the child has a goiter.
  • The total or free T4 are well above the normal range (T4 greater than 15, free T4 greater than 3.0), accompanied by a TSH less than 0.1, and usually (but not always) symptoms of hyperthyroidism.

The take-home message is...

Target patients for referral who are most likely to need evaluation and possibly treatment with either thyroid hormone or antithyroid medication. This will eliminate or greatly decrease the parental anxiety aroused by being told their child has a thyroid problem when none exists. It will also allow those patients who truly need to be seen by a pediatric endocrinologist to be given appointments more promptly-the best of all possible worlds.


1. Glass AR, Kushner J: Obesity, nutrition, and the thyroid. The Endocrinologist 1996;6:392

2. Hollowell JG, Staehling NW, Flanders WD: Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 2005;87:489

3. Nabhan ZM, Kreher NC, Eugster EA: Hashitoxicosis in children: Clinical features and natural history. J Pediatr 2005;146:533

4. Moore DC: Natural course of 'subclinical' hypothyroidism in childhood and adolescence. Arch Pediatr Adol Med 1996;150:293

5. Diez JJ, Iglesias P: Spontaneous subclinical hypothyroidism in patients older than 55 years: An analysis of natural course and risk factors for development of overt thyroid failure. J Clin Endocrinol Metab 2004;89:4890

6. Knudsen N, Laurberg P, Rasmussen LB, et al: Small differences in thyroid function may be important for body mass index and the occurrence of obesity in the population. J Clin Endocrinol Metab 2005;90:4019

7. Soldin SJ, Morales A, Albalos F, et al: Pediatric reference ranges on the Abbott IMx for FSH, LH, TSH. T4, T3, free T4, free T3, and T-uptake. Clin Biochem 1995;28:603


After reviewing this article, the physician should be able to:

  • Understand the basic pathophysiology of thyroid dysfunction
  • Describe the indications and limitations of available tests of thyroid function
  • Recognize what symptoms and test results do and do not warrant referral to a pediatric endocrinologist


This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of CME2, Inc. ("cme2") and Contemporary Pediatrics. cme2 is accredited by the ACCME to provide continuing medical education for physicians.

cme2 designates this educational activity for a maximum of 1.0 AMA PRA Category 1 Credit™. Physicians should only claim credit commensurate with the extent of their participation in the activity.

Target audience: Pediatricians and primary care physicians

Educational objectives

Understand the basic pathophysiology of thyroid dysfunctionj

Describe the indications and limitations of available tests of thyroid function

Recognize what symptoms and test results do and do not warrant referral to a pediatric endocrinologist

To earn CME credit for this activity

Participants should study the article and log on to, where they must pass a post-test and complete an online evaluation of the CME activity. After passing the post-test and completing the online evaluation, a CME certificate will be e-mailed to them. The release date for this activity is January 1, 2007. The expiration date is January 1, 2008.


Editors Toby Hindin, Karen Bardossi, and Lisa Maresca disclose that they do not have any financial relationships with any manufacturer in this area of medicine.

Manuscript reviewers disclose that they do not have any financial relationships with any manufacturer in this area of medicine.

Author Paul Kaplowitz, MD, PhD discloses that he does not have any financial relationships with any manufacturer in this area of medicine.

Resolution of conflict of interest

cme2 has implemented a process to resolve conflicts of interest for each continuing medical education activity, to help ensure content validity, independence, fair balance, and that the content is aligned with the interest of the public. Conflicts, if any, are resolved through a peer review process.

Unapproved/off-label use discussion

Faculty may discuss information about pharmaceutical agents, devices, or diagnostic products that are outside of FDA-approved labeling. This information is intended solely for CME and is not intended to promote off-label use of these medications. If you have questions, contact the medical affairs department of the manufacturer for the most recent prescribing information. Faculty are required to disclose any off-label discussion.

Related Videos
Angela Nash, PhD, APRN, CPNP-PC, PMHS | Image credit: UTHealth Houston
Allison Scott, DNP, CPNP-PC, IBCLC
Joanne M. Howard, MSN, MA, RN, CPNP-PC, PMHS & Anne Craig, MSN, RN, CPNP-PC
Juanita Mora, MD
Natasha Hoyte, MPH, CPNP-PC
Lauren Flagg
Venous thromboembolism, Heparin-induced thrombocytopenia, and direct oral anticoagulants | Image credit: Contemporary Pediatrics
Sally Humphrey, DNP, APRN, CPNP-PC | Image Credit: Contemporary Pediatrics
© 2024 MJH Life Sciences

All rights reserved.