Interpreting Minor Variations in Thyroid Function or Echostructure




Overt thyroid dysfunction is documented by serum thyrotropin or T4 concentrations are often ordered for nonspecific complaints and will by definition fall outside of the 95% reference range 5% of the time. In addition, most laboratories quote adult ranges, which are not necessarily applicable to young children, and regression toward the mean is common, justifying that the test be repeated before embarking on treatment. On the other hand, neck ultrasounds are frequently performed for diffuse goiter or non-thyroid conditions. Yet, an ultrasound is not required to make a diagnosis of Hashimoto thyroiditis and small cysts and nodules discovered incidentally often lead to unjustified concerns about neoplasia.


Key points








  • The results of 5% of any laboratory test are outside the reference ranges.



  • Without intervention, these results are often within the reference range when repeated.



  • Diagnosing Hashimoto thyroiditis does not require ultrasonography imaging.






Minor variations in thyroid function


Introduction


The measurement of serum hormone concentrations on automated analyzers in clinical biochemistry laboratories allows rapid turnaround times, which may have contributed to practitioners requesting these ever more frequently. Because of the high prevalence and nonspecific clinical presentation of thyroid diseases, evaluation of thyroid function figures prominently on the list of blood tests requested. At our institution, a mother-child tertiary care center, serum thyrotropin (TSH) is measured in 55 samples every day, almost 10 times more than serum cortisol (A. Djemli, personal communication, 2013). Ordering clinicians should be aware of intra-assay and interassay variability when interpreting these results. Beyond that, most practitioners are aware that serum cortisol has marked circadian variation, but may not know about the less striking but nevertheless potentially significant circadian rhythm of serum TSH (discussed later). The changes in normal thyroid hormone parameters during growth require better appreciation, but ethical limitations in drawing blood from normal children has hampered the establishment of age-related references ranges.


Isolated Hyperthyrotropinemia


Subclinical hypothyroidism is often defined by an increased serum TSH level with a normal serum free thyroxine (fT 4 ) level. We argue that the descriptive term isolated hyperthyrotropinemia is more appropriate and that subclinical hypothyroidism should only be used for individuals with high TSH and low fT 4 levels but neither sign nor symptom of hypothyroidism. In addition, serum TSH levels in normal individuals decrease progressively with age ( Fig. 1 ) and the use of adult reference intervals results in many young children being labeled as having isolated hyperthyrotropinemia.




Fig. 1


Serum TSH levels in normal infants as a function of age in days. Scatter plot of individual values ( dots ), regression ( full black line ), lowest fit ( full red line ) and 5th to 95th confidence intervals ( striped red lines ).

( Adapted from Djemli A, Van Vliet G, Belgoudi J, et al. Reference intervals for free thyroxine, total triiodothyronine, thyrotropin and thyroglobulin for Quebec newborns, children and teenagers. Clin Biochem 2004;37(4):328–30; with permission.)


Aside from age-related changes, there is a nocturnal surge in TSH. This surge may lead to an erroneous interpretation of TSH level being abnormal: a 10-year-old girl was evaluated in our emergency room for an anxiety attack after seeing a horror movie. Because of tachycardia, a sample was drawn at 2:00 am to rule out hyperthyroidism and serum TSH level was 9.33 mU/L (with a normal fT 4 level of 9.78 pmol/L); there was no goiter on examination, a repeat serum TSH test at 2:00 p m on the same day was normal at 1.79 mU/L and there were no detectable antibodies against thyroperoxidase or thyroglobulin in serum. Fig. 2 shows the mean ± standard error serum TSH level in normal children over a 24-hour period: to extrapolate to ±2 SD, the normal range of serum TSH at 2:00 am extends to about 10 mU/L.




Fig. 2


Mean (± standard error) serum TSH levels in normal children over a 24-hour period.

( From Rose SR, Nisula BC. Circadian variation of thyrotropin in childhood. J Clin Endocrinol Metab 1989;68(6):1088; with permission.)


Even less well known is the observation of seasonal variation: in the Northern Hemisphere, mean serum TSH is 10% to 15% higher in the winter and, in countries with wide temperature differences, this may lead to overdiagnosing so-called subclinical hypothyroidism.


By far the largest numbers of children referred for isolated hyperthyrotropinemia are those with exogenous obesity. Although excess weight gain caused by hypothyroidism in growing children can easily be distinguished from developing exogenous obesity on clinical grounds alone by assessing height velocity (decreasing in the former, increasing in the latter), many practitioners nevertheless request a serum TSH level, which is occasionally slightly increased. The mechanisms for this association are unclear but serum TSH level decreases if weight loss is achieved and thyroxine treatment in this situation is futile.


Isolated hyperthyrotropinemia in children is often transient, with spontaneous normalization without treatment. In the absence of autoimmunity, a common cause of permanent hyperthyrotropinemia seems to be monoallelic mutations inactivating the TSH receptor. Because the increase of serum TSH level is an appropriate physiologic response to the relative TSH resistance, which results in normal serum fT 4 concentrations and does not worsen over time, it can be left untreated and serial monitoring of serum TSH is unwarranted, with considerable benefits to the patients and savings to the health care system. A first step in evaluating these children is to obtain a serum TSH level on all first-degree relatives, about 40% of whom also have hyperthyrotropinemia.


Persistent hyperthyrotropinemia of neonatal onset is also observed in some patients with pseudohypoparathyroidism, the dysmorphic features becoming more evident as the child ages, or with the brain-lung-thyroid syndrome, in which the respiratory and neurologic phenotypes typically dominate. A more complete discussion of the possible causes of isolated, mild neonatal hyperthyrotropinemia and of its likely innocuous nature can be found in our recent review on congenital hypothyroidism.


Isolated Hypothyroxinemia


Isolated hypothyroxinemia is commonly observed in premature neonates. Numerous studies have found the degree of hypothyroxinemia to be correlated with adverse short-term and long-term outcomes. However, correlation does not imply causation, and a landmark double-blind randomized placebo-controlled trial (RCT) showed that thyroxine treatment did not affect outcome. A possible benefit in neonates born before 27 weeks in post-hoc subgroup analyses may be an effect of iodine, not of thyroxine (which contains 66% of iodine by weight). Very premature infants are in negative iodine balance and the results of an RCT comparing the effects of thyroxine and iodine on the neurodevelopmental outcome of such neonates are awaited.


Children with Prader-Willi syndrome often present mild hypothyroxinemia: this is not present on the neonatal screening specimen but develops at older ages, consistent with evolving hypothalamic dysfunction. On stimulation testing with TSH-releasing hormone, the typical hypothalamic profile is only rarely seen. Thus, routine thyroxine administration is not currently recommended for these individuals.


In children receiving growth hormone (GH) replacement for GH deficiency or pharmacologic doses of GH for other conditions, such as Turner syndrome, a slight decrease in fT 4 level is frequently observed. This decrease results from stimulation of deiodination of T 4 to triiodothyronine (T 3 ) by GH. This decrease does not indicate hypothyroidism but these children are nevertheless often given thyroxine. In these cases, previously detectable serum TSH becoming undetectable suggests unjustified or excessive thyroxine treatment, which may be clinically relevant because such treatment advances skeletal maturation more than height itself, therefore potentially decreasing adult height.


Nonthyroidal Illnesses


T 4 , a biologically inactive prohormone, can be either activated to T 3 through the actions of deiodinases type 1 and 2 or inactivated to reverse T 3 through deiodinase type 3. Both chronic and acute systemic diseases may result in preferential inactivation of T 4 , resulting in the low-T 3 syndrome. When prolonged, severe disease may eventually result in low fT 4 and TSH levels as well, suggesting true central hypothyroidism; this situation is usually seen in patients hospitalized in an intensive care unit (ICU) and is associated with a poor prognosis for survival, but there is no evidence that administering T 4 or T 3 influences outcome. A common cause of suppressed serum TSH in ICUs is the use of dopamine infusions : in 64% of extremely low birth weight neonates, for instance, a dopamine infusion is administered during the first week of life. On recovery from severe nonthyroidal illness, serum TSH levels may be transiently and slightly increased.


Disorders of Thyroid Hormone Metabolism or Transport into the Cells


Over the past decade, the genetic mechanisms underlying 2 seemingly rare syndromes involving disordered thyroid hormone metabolism or transport into the cells have been identified. Deficiency of selenium insertion protein SECISBP2 (also called SBP2), an autosomal recessive condition, may present with a subtle phenotype of an isolated slowing of linear growth and of skeletal maturation. However, consistent with a global effect on selenoproteins, a more severe multisystem disorder can also be observed. Serum selenium level is low and may be the clue to the diagnosis. In contrast, male patients hemizygous for mutations in the gene encoding the monocarboxylate transporter 8 (MCT8, which transports T 3 into certain brain cells) are diagnosed from a dramatic neurologic phenotype known as the Allan-Herndon-Dudley syndrome. The serum levels of thyroid hormones and TSH reflect the different mechanisms of these two conditions: high T 4 and low T 3 in the former, low T 4 and high T 3 in the latter, with normal or slightly increased TSH in both; these abnormalities are often subtle. In boys hemizygous for MCT8 mutations, treatment with diiodothyropropionic acid (a thyroid hormone analogue that does not require MCT8 to enter tissues) normalizes thyroid function and ameliorates metabolism but has no effect on the severe neurologic phenotype. In SBP2 deficiency, T 3 treatment induces transient growth acceleration in children with low serum T 3 levels, whereas selenium replacement increases serum selenium but not selenoproteins synthesis. Antioxidant treatment may be tried in severe cases.


Isolated Serum Thyrotropin Suppression (with Normal or Low Serum Free Thyroxine Level)


In adolescents with anorexia nervosa, a chronically negative caloric balance leads to the low-T 3 syndrome. As in ICU patients, this phenomenon is energy-preserving and should not be treated. In addition, the numerous other endocrine dysfunctions found in anorexia nervosa are thought to have a hypothalamic origin and, consistent with this concept, an isolated decrease in serum TSH level can also be observed. The abnormalities in thyroid function in bulimia nervosa are less well characterized.


Subclinical Hypothyroidism


Worldwide, the commonest cause of subclinical hypothyroidism is iodine deficiency. Although its severe form has been eradicated in many parts of the world where it was previously endemic, its more moderate form is reappearing in some industrialized countries. At the population level, this likely reflects lapses in salt iodization programs but, at the individual level, a dietary history should be obtained to uncover unusual habits.


Among the many challenges pediatricians face in the care of children with Down syndrome is the interpretation of their thyroid function tests. Recent population-based studies do not support earlier claims that overt congenital hypothyroidism is much more common in Down syndrome than in the general population. On the screening specimen, there is a slight shift of the T 4 distribution to the left and of TSH to the right in newborns with Down syndrome compared with normal newborns. These shifts are so mild that they do not result in an increased recall rate after neonatal screening when a TSH cutoff of 15 mU/L is used. They persist throughout growth and are associated with decreased thyroid volume on ultrasonography. This state of relative TSH resistance does not require treatment: another landmark double-blind RCT of thyroxine versus placebo in the Netherlands only found a slight difference favoring thyroxine in one of the motor development subscales at 2 years but this did not translate into higher motor or cognitive scores at age 11 years.


Superimposed on their intrinsic congenital condition and persistent shift in TSH and T 4 distributions, older children of both sexes with Down syndrome (without the female predominance seen in the normal population) have a greater propensity to autoimmune thyroid disease, most commonly Hashimoto thyroiditis. Because some symptoms of hypothyroidism, such as constipation, overlap with characteristics of Down syndrome, regular screening of these children with serum TSH has been advocated. However, this should not distract clinicians from the fact that, as in normal children, slowing of linear growth, which can be observed on Down syndrome–specific growth charts, is the key to making a clinical diagnosis of hypothyroidism in a timely fashion.


Subclinical Hyperthyroidism


Iatrogenic


In a population attending a health fair in Colorado in 1995, 535 of 25,862 adult subjects were found to have subclinical hyperthyroidism. Of subjects taking a thyroid medication, 20.7% had subclinical hyperthyroidism, whereas this was only 0.9% of those who were not taking such medication, suggesting overprescription. This finding was recently confirmed in the general population in the United Kingdom. The overprescription of thyroxine to children has not been specifically studied. However, 38% of children diagnosed as having congenital hypothyroidism (1 in 2300) in the United States have stopped treatment by the age of 4 years, without medical surveillance but without any reports of increased prevalence of recurring hypothyroidism in the ensuing years. By contrast, in the early days of screening, when prevalence was estimated as 1 in 4000 newborns, permanence of hypothyroidism was documented in almost all children. Taken together, this suggests that current screening algorithms have become too sensitive, leading to overdiagnosis and overtreatment.


Neonates born to mothers with Graves disease


Graves disease is often quoted as affecting 1 in 500 pregnant women. However, clinically diagnosed hyperthyroidism in the fetus or in the neonate is exceptional: the computerized database of our service contains 9 such entries over 9 years and our institution provides maternal and neonatal care to a population with about 50,000 births per year, thus validating the often-quoted prevalence of neonatal hyperthyroidism of 1 in 50,000 births. Overt congenital hyperthyroidism is a serious condition in which fetal mortality is still observed, leading to the recommendation that pregnant women with a present or past history of Graves disease should have the titer of TSH receptor antibodies measured in their serum in midpregnancy. However, the justified fear of overt congenital hyperthyroidism should not lead to overdiagnosis and overtreatment. In a recent French series, biochemical hyperthyroidism was present in 7 of 23 neonates born to TSH receptor antibody–positive women but none developed clinical symptoms. Thus, we argue that clinical surveillance (especially of weight gain and heart rate) suffices, being cognizant that, if the mother is treated with antithyroid drugs, hyperthyroidism in her newborn child may be masked during the first few days and become evident later, because of the shorter half-lives of drugs compared with that of TSH receptor–stimulating antibodies.


Hyperfunctioning thyroid nodules


Autonomous thyroid adenomas (or so-called hot nodules) are rare in children and the associated hyperthyroidism is generally mild, with the commonest manifestation being tachycardia. Serum TSH concentrations may be undetectable in spite of normal fT 4 levels, in which case it is justified to measure serum T 3 (to diagnose T 3 toxicosis).


Interference from Cardiovascular, Antiepileptic, and Psychotropic Medications


The most clinically significant interference between nonthyroid medications and thyroid function, TSH suppression by dopamine infusions, was discussed earlier. In addition, several other classes of drugs may interfere with thyroid function testing. Among these, the most widely used are antiepileptic drugs. For instance, serum fT4 level is lower and TSH level higher in epileptic children on valproate than in those on carbamazepine. Newer antipsychotic drugs often induce major weight gain, which may affect serum TSH, as discussed earlier. Less used in children than in adults is the antiarrhythmic drug amiodarone, which has an extremely rich iodine content and may induce both overt hypothyroidism and hyperthyroidism but also more subtle abnormalities in thyroid function tests. Likewise, interferon therapy may induce both overt and subclinical thyroid dysfunction.




Minor variations in thyroid function


Introduction


The measurement of serum hormone concentrations on automated analyzers in clinical biochemistry laboratories allows rapid turnaround times, which may have contributed to practitioners requesting these ever more frequently. Because of the high prevalence and nonspecific clinical presentation of thyroid diseases, evaluation of thyroid function figures prominently on the list of blood tests requested. At our institution, a mother-child tertiary care center, serum thyrotropin (TSH) is measured in 55 samples every day, almost 10 times more than serum cortisol (A. Djemli, personal communication, 2013). Ordering clinicians should be aware of intra-assay and interassay variability when interpreting these results. Beyond that, most practitioners are aware that serum cortisol has marked circadian variation, but may not know about the less striking but nevertheless potentially significant circadian rhythm of serum TSH (discussed later). The changes in normal thyroid hormone parameters during growth require better appreciation, but ethical limitations in drawing blood from normal children has hampered the establishment of age-related references ranges.


Isolated Hyperthyrotropinemia


Subclinical hypothyroidism is often defined by an increased serum TSH level with a normal serum free thyroxine (fT 4 ) level. We argue that the descriptive term isolated hyperthyrotropinemia is more appropriate and that subclinical hypothyroidism should only be used for individuals with high TSH and low fT 4 levels but neither sign nor symptom of hypothyroidism. In addition, serum TSH levels in normal individuals decrease progressively with age ( Fig. 1 ) and the use of adult reference intervals results in many young children being labeled as having isolated hyperthyrotropinemia.




Fig. 1


Serum TSH levels in normal infants as a function of age in days. Scatter plot of individual values ( dots ), regression ( full black line ), lowest fit ( full red line ) and 5th to 95th confidence intervals ( striped red lines ).

( Adapted from Djemli A, Van Vliet G, Belgoudi J, et al. Reference intervals for free thyroxine, total triiodothyronine, thyrotropin and thyroglobulin for Quebec newborns, children and teenagers. Clin Biochem 2004;37(4):328–30; with permission.)


Aside from age-related changes, there is a nocturnal surge in TSH. This surge may lead to an erroneous interpretation of TSH level being abnormal: a 10-year-old girl was evaluated in our emergency room for an anxiety attack after seeing a horror movie. Because of tachycardia, a sample was drawn at 2:00 am to rule out hyperthyroidism and serum TSH level was 9.33 mU/L (with a normal fT 4 level of 9.78 pmol/L); there was no goiter on examination, a repeat serum TSH test at 2:00 p m on the same day was normal at 1.79 mU/L and there were no detectable antibodies against thyroperoxidase or thyroglobulin in serum. Fig. 2 shows the mean ± standard error serum TSH level in normal children over a 24-hour period: to extrapolate to ±2 SD, the normal range of serum TSH at 2:00 am extends to about 10 mU/L.




Fig. 2


Mean (± standard error) serum TSH levels in normal children over a 24-hour period.

( From Rose SR, Nisula BC. Circadian variation of thyrotropin in childhood. J Clin Endocrinol Metab 1989;68(6):1088; with permission.)


Even less well known is the observation of seasonal variation: in the Northern Hemisphere, mean serum TSH is 10% to 15% higher in the winter and, in countries with wide temperature differences, this may lead to overdiagnosing so-called subclinical hypothyroidism.


By far the largest numbers of children referred for isolated hyperthyrotropinemia are those with exogenous obesity. Although excess weight gain caused by hypothyroidism in growing children can easily be distinguished from developing exogenous obesity on clinical grounds alone by assessing height velocity (decreasing in the former, increasing in the latter), many practitioners nevertheless request a serum TSH level, which is occasionally slightly increased. The mechanisms for this association are unclear but serum TSH level decreases if weight loss is achieved and thyroxine treatment in this situation is futile.


Isolated hyperthyrotropinemia in children is often transient, with spontaneous normalization without treatment. In the absence of autoimmunity, a common cause of permanent hyperthyrotropinemia seems to be monoallelic mutations inactivating the TSH receptor. Because the increase of serum TSH level is an appropriate physiologic response to the relative TSH resistance, which results in normal serum fT 4 concentrations and does not worsen over time, it can be left untreated and serial monitoring of serum TSH is unwarranted, with considerable benefits to the patients and savings to the health care system. A first step in evaluating these children is to obtain a serum TSH level on all first-degree relatives, about 40% of whom also have hyperthyrotropinemia.


Persistent hyperthyrotropinemia of neonatal onset is also observed in some patients with pseudohypoparathyroidism, the dysmorphic features becoming more evident as the child ages, or with the brain-lung-thyroid syndrome, in which the respiratory and neurologic phenotypes typically dominate. A more complete discussion of the possible causes of isolated, mild neonatal hyperthyrotropinemia and of its likely innocuous nature can be found in our recent review on congenital hypothyroidism.


Isolated Hypothyroxinemia


Isolated hypothyroxinemia is commonly observed in premature neonates. Numerous studies have found the degree of hypothyroxinemia to be correlated with adverse short-term and long-term outcomes. However, correlation does not imply causation, and a landmark double-blind randomized placebo-controlled trial (RCT) showed that thyroxine treatment did not affect outcome. A possible benefit in neonates born before 27 weeks in post-hoc subgroup analyses may be an effect of iodine, not of thyroxine (which contains 66% of iodine by weight). Very premature infants are in negative iodine balance and the results of an RCT comparing the effects of thyroxine and iodine on the neurodevelopmental outcome of such neonates are awaited.


Children with Prader-Willi syndrome often present mild hypothyroxinemia: this is not present on the neonatal screening specimen but develops at older ages, consistent with evolving hypothalamic dysfunction. On stimulation testing with TSH-releasing hormone, the typical hypothalamic profile is only rarely seen. Thus, routine thyroxine administration is not currently recommended for these individuals.


In children receiving growth hormone (GH) replacement for GH deficiency or pharmacologic doses of GH for other conditions, such as Turner syndrome, a slight decrease in fT 4 level is frequently observed. This decrease results from stimulation of deiodination of T 4 to triiodothyronine (T 3 ) by GH. This decrease does not indicate hypothyroidism but these children are nevertheless often given thyroxine. In these cases, previously detectable serum TSH becoming undetectable suggests unjustified or excessive thyroxine treatment, which may be clinically relevant because such treatment advances skeletal maturation more than height itself, therefore potentially decreasing adult height.


Nonthyroidal Illnesses


T 4 , a biologically inactive prohormone, can be either activated to T 3 through the actions of deiodinases type 1 and 2 or inactivated to reverse T 3 through deiodinase type 3. Both chronic and acute systemic diseases may result in preferential inactivation of T 4 , resulting in the low-T 3 syndrome. When prolonged, severe disease may eventually result in low fT 4 and TSH levels as well, suggesting true central hypothyroidism; this situation is usually seen in patients hospitalized in an intensive care unit (ICU) and is associated with a poor prognosis for survival, but there is no evidence that administering T 4 or T 3 influences outcome. A common cause of suppressed serum TSH in ICUs is the use of dopamine infusions : in 64% of extremely low birth weight neonates, for instance, a dopamine infusion is administered during the first week of life. On recovery from severe nonthyroidal illness, serum TSH levels may be transiently and slightly increased.


Disorders of Thyroid Hormone Metabolism or Transport into the Cells


Over the past decade, the genetic mechanisms underlying 2 seemingly rare syndromes involving disordered thyroid hormone metabolism or transport into the cells have been identified. Deficiency of selenium insertion protein SECISBP2 (also called SBP2), an autosomal recessive condition, may present with a subtle phenotype of an isolated slowing of linear growth and of skeletal maturation. However, consistent with a global effect on selenoproteins, a more severe multisystem disorder can also be observed. Serum selenium level is low and may be the clue to the diagnosis. In contrast, male patients hemizygous for mutations in the gene encoding the monocarboxylate transporter 8 (MCT8, which transports T 3 into certain brain cells) are diagnosed from a dramatic neurologic phenotype known as the Allan-Herndon-Dudley syndrome. The serum levels of thyroid hormones and TSH reflect the different mechanisms of these two conditions: high T 4 and low T 3 in the former, low T 4 and high T 3 in the latter, with normal or slightly increased TSH in both; these abnormalities are often subtle. In boys hemizygous for MCT8 mutations, treatment with diiodothyropropionic acid (a thyroid hormone analogue that does not require MCT8 to enter tissues) normalizes thyroid function and ameliorates metabolism but has no effect on the severe neurologic phenotype. In SBP2 deficiency, T 3 treatment induces transient growth acceleration in children with low serum T 3 levels, whereas selenium replacement increases serum selenium but not selenoproteins synthesis. Antioxidant treatment may be tried in severe cases.


Isolated Serum Thyrotropin Suppression (with Normal or Low Serum Free Thyroxine Level)


In adolescents with anorexia nervosa, a chronically negative caloric balance leads to the low-T 3 syndrome. As in ICU patients, this phenomenon is energy-preserving and should not be treated. In addition, the numerous other endocrine dysfunctions found in anorexia nervosa are thought to have a hypothalamic origin and, consistent with this concept, an isolated decrease in serum TSH level can also be observed. The abnormalities in thyroid function in bulimia nervosa are less well characterized.


Subclinical Hypothyroidism


Worldwide, the commonest cause of subclinical hypothyroidism is iodine deficiency. Although its severe form has been eradicated in many parts of the world where it was previously endemic, its more moderate form is reappearing in some industrialized countries. At the population level, this likely reflects lapses in salt iodization programs but, at the individual level, a dietary history should be obtained to uncover unusual habits.


Among the many challenges pediatricians face in the care of children with Down syndrome is the interpretation of their thyroid function tests. Recent population-based studies do not support earlier claims that overt congenital hypothyroidism is much more common in Down syndrome than in the general population. On the screening specimen, there is a slight shift of the T 4 distribution to the left and of TSH to the right in newborns with Down syndrome compared with normal newborns. These shifts are so mild that they do not result in an increased recall rate after neonatal screening when a TSH cutoff of 15 mU/L is used. They persist throughout growth and are associated with decreased thyroid volume on ultrasonography. This state of relative TSH resistance does not require treatment: another landmark double-blind RCT of thyroxine versus placebo in the Netherlands only found a slight difference favoring thyroxine in one of the motor development subscales at 2 years but this did not translate into higher motor or cognitive scores at age 11 years.


Superimposed on their intrinsic congenital condition and persistent shift in TSH and T 4 distributions, older children of both sexes with Down syndrome (without the female predominance seen in the normal population) have a greater propensity to autoimmune thyroid disease, most commonly Hashimoto thyroiditis. Because some symptoms of hypothyroidism, such as constipation, overlap with characteristics of Down syndrome, regular screening of these children with serum TSH has been advocated. However, this should not distract clinicians from the fact that, as in normal children, slowing of linear growth, which can be observed on Down syndrome–specific growth charts, is the key to making a clinical diagnosis of hypothyroidism in a timely fashion.


Subclinical Hyperthyroidism


Iatrogenic


In a population attending a health fair in Colorado in 1995, 535 of 25,862 adult subjects were found to have subclinical hyperthyroidism. Of subjects taking a thyroid medication, 20.7% had subclinical hyperthyroidism, whereas this was only 0.9% of those who were not taking such medication, suggesting overprescription. This finding was recently confirmed in the general population in the United Kingdom. The overprescription of thyroxine to children has not been specifically studied. However, 38% of children diagnosed as having congenital hypothyroidism (1 in 2300) in the United States have stopped treatment by the age of 4 years, without medical surveillance but without any reports of increased prevalence of recurring hypothyroidism in the ensuing years. By contrast, in the early days of screening, when prevalence was estimated as 1 in 4000 newborns, permanence of hypothyroidism was documented in almost all children. Taken together, this suggests that current screening algorithms have become too sensitive, leading to overdiagnosis and overtreatment.


Neonates born to mothers with Graves disease


Graves disease is often quoted as affecting 1 in 500 pregnant women. However, clinically diagnosed hyperthyroidism in the fetus or in the neonate is exceptional: the computerized database of our service contains 9 such entries over 9 years and our institution provides maternal and neonatal care to a population with about 50,000 births per year, thus validating the often-quoted prevalence of neonatal hyperthyroidism of 1 in 50,000 births. Overt congenital hyperthyroidism is a serious condition in which fetal mortality is still observed, leading to the recommendation that pregnant women with a present or past history of Graves disease should have the titer of TSH receptor antibodies measured in their serum in midpregnancy. However, the justified fear of overt congenital hyperthyroidism should not lead to overdiagnosis and overtreatment. In a recent French series, biochemical hyperthyroidism was present in 7 of 23 neonates born to TSH receptor antibody–positive women but none developed clinical symptoms. Thus, we argue that clinical surveillance (especially of weight gain and heart rate) suffices, being cognizant that, if the mother is treated with antithyroid drugs, hyperthyroidism in her newborn child may be masked during the first few days and become evident later, because of the shorter half-lives of drugs compared with that of TSH receptor–stimulating antibodies.


Hyperfunctioning thyroid nodules


Autonomous thyroid adenomas (or so-called hot nodules) are rare in children and the associated hyperthyroidism is generally mild, with the commonest manifestation being tachycardia. Serum TSH concentrations may be undetectable in spite of normal fT 4 levels, in which case it is justified to measure serum T 3 (to diagnose T 3 toxicosis).


Interference from Cardiovascular, Antiepileptic, and Psychotropic Medications


The most clinically significant interference between nonthyroid medications and thyroid function, TSH suppression by dopamine infusions, was discussed earlier. In addition, several other classes of drugs may interfere with thyroid function testing. Among these, the most widely used are antiepileptic drugs. For instance, serum fT4 level is lower and TSH level higher in epileptic children on valproate than in those on carbamazepine. Newer antipsychotic drugs often induce major weight gain, which may affect serum TSH, as discussed earlier. Less used in children than in adults is the antiarrhythmic drug amiodarone, which has an extremely rich iodine content and may induce both overt hypothyroidism and hyperthyroidism but also more subtle abnormalities in thyroid function tests. Likewise, interferon therapy may induce both overt and subclinical thyroid dysfunction.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Oct 2, 2017 | Posted by in PEDIATRICS | Comments Off on Interpreting Minor Variations in Thyroid Function or Echostructure

Full access? Get Clinical Tree

Get Clinical Tree app for offline access