CAPENEWS Archive - Thyroid

-Contributed by Dr. Anju Virmani, New Delhi and Dr. Vijayakumar, Calicut

Nikhil Tandon (From CAPENEWS issue # 2, July 1997)

Thyroid disorders are a common endocrine problem with myriad clinical presentations in childhood. With increasing ease of availability of thyroid function tests, the treating physician is often confronted with reports which are either are not in consonance with the clinical profile or appear to be internally inconsistent. Though often there may be technical problems in assay performance there are many other variables which may confound the interpretation of test results. In this write-up an attempt is made to address some of these variables which need to be kept in mind while analyzing thyroid function tests. Prior to that it is critical to familiarize oneself with the classical profile observed in hypothyroidism and hyperthyroidism.

Like that of most endocrine glands, thyroid function needs to be assessed by not only measuring hormones produced by the gland, i.e. T3 and T4, but also the pituitary hormone, i.e. TSH. In primary hypothyroidism, the first abnormality observed is elevated serum TSH. Later, when the diseased gland is unable to respond to the stimulus provided by elevated levels of circulating TSH, serum T4 concentration tends to fall. The main source of T3 is peripheral (tissue) conversion of T4, so it is only much later that serum T3 levels decline. Hence, in the final stage of primary hypothyroidism the profile is: low T3, low T4 and high TSH. In secondary hypothyroidism, T3 and T4 levels are low, but elevation of TSH would obviously be absent. In hyperthyroidism, which is usually because of an autonomously functioning thyroid gland, serum T3 and T4 levels are elevated and due to the negative feedback imposed by their higher circulating levels, TSH is suppressed. Only in very rare situations like a TSH secreting pituitary tumor (and some rarer conditions discussed later) that in a clinically apparent hyperthyroid state, elevated T3 and T4 values are accompanied by non-suppressed or even slightly raised TSH values.


TOTAL T4: Thyroid hormones circulate largely bound to carrier proteins (thyroid binding globulin: TBG, transthyretin and albumin). Most laboratories measure total T4 and total T3, which are dependent on the amount of carrier protein. However, it is the much smaller free hormone fraction which is metabolically active and truly reflects thyroid status. In the following states, there may be abnormal levels of total T4 and T3 without any abnormality of free hormone levels, i.e. normal thyroid functional state:

  1. Hereditary abnormalities of binding proteins: TBG deficiency or excess; abnormal albumin; abnormal transthyretin
  2. Acquired deficiency of binding proteins: nephrotic syndrome (protein loss), severe liver disease (impaired production of proteins); androgens or anabolic steroids
  3. Altered T4 binding to TBG due to drugs: salicylates, phenytoin, phenylbutazone
  4. Antibodies to T4.
Newer developments in immunoassay methods for free T4 and free T3 have overcome many of these problems.

TOTAL T3: In suspected thyroidal illness, one usually measures serum T4 since there is little to be gained by routinely measuring T3. As already mentioned, T3 concentrations are low normal in about 25% patients with hypothyroidism. However the two main indications for measuring total/ free T3 are:

  1. suspected T3-toxicosis
  2. if the patient is on certain drugs which inhibit the peripheral conversion of T4 to T3: dexamethasone, propranalol, propylthiouracil, amiodarone, iodine containing radiographic contrast media (eg iopanoic acid).

TSH: With the recent availability of second generation immunoradiometric assays for TSH, many clinicians tend to use these as the only hormone test ordered. It is important to realize that though a reliably performed second generation TSH IRMA in experienced hands may be the best screening test for thyroid disease, a diagnosis cannot be made on the basis of abnormal TSH alone. Even though third generation IRMAs may be able to differentiate between low and very low TSH levels, the latter being expected in true thyrotoxicosis, it is still prudent to estimate T4 levels. This not only confirms the diagnosis, but also establishes the severity of the thyrotoxic state. In certain situations TSH may give a misleading impression:

  1. Normal TSH: Patient hypo- or hyper-thyroid: hypothalamic-pituitary disease can cause this unexpected result. As mentioned earlier, secondary hypothyroidism or a TSH secreting pituitary tumor may be associated with grossly deranged T3 and T4 values but normal TSH. This may be especially relevant while evaluating a child who has short stature due to a hypothalamic- pituitary disease. In such a condition, an isolated TSH could be normal, but would be inappropriately low T4 and T3 levels.

  2. High TSH: Patient euthyroid: In sub-clinical hypothyroidism, characterized by an isolated raising of TSH, the patient appears clinically euthyroid, and this is mirrored by normal T4 and T3. This situation is important to recognize because it is not inevitable for such persons to progress to a clinically overt hypothyroid state. It is helpful to perform a measurement of antimicrosomal antibodies in them, since it has been observed that in those who have both increased TSH and positive antibody titer, the tendency to progress to overt hypothyroidism is more: a rate of about 5% a year, with the curve leveling off in about 10 years. The risk of thyroid failure is lower if only high TSH or antibodies are discovered. The implication is that those with both high TSH and positive antibody titer need close clinical and biochemical follow-up, rather than being unnecessarily condemned to lifelong thyroxin replacement right away.

  3. High TSH: patient hypothyroid: A practical problem is the finding of high T4 with high TSH in patients of primary hypothyroidism on replacement thyroxin therapy. This is usually seen in a patient who has not been taking thyroxin regularly in the last few weeks, and the sample has been taken soon after ingestion of thyroxin tablets. This is reflected in high serum concentration of T4, with the inadequacy of replacement evidenced by the still elevated TSH value. Ideally a gap of about 12 hours between ingestion of thyroxin and sampling for thyroid function tests is advocated to avoid confusion; and in such a situation the patient should be closely asked about missed doses.

  4. Low TSH: Patient euthyroid: Patients recovering from hyperthyroidism may show a suppressed TSH for a variable length of time. So in those with recently treated hyperthyroidism, a low TSH may erroneously suggest a continuing hyperthyroid state. The same phenomenon of low TSH may also be observed in ophthalmic Graves' disease, where the patient is clinically euthyroid and has normal T4 and T3. Once again the low TSH may mislead the treating physician to consider anti-thyroid treatment: only concomitantly performed T4 and T3 values can help resolve the management problem. An interesting subset of euthyroid persons who have suppressed TSH are relatives of patients with Graves' disease. This is an unexplained phenomenon, which is helpful to keep in mind.

Keeping the above information in mind, one can plan the diagnostic approach to a patient with suspected thyroid disorder: In a child suspected to be hypothyroid, the biochemical investigations of choice are T4 and TSH. If low T4 and high TSH are seen, the diagnosis of primary hypothyroidism is clinched. If with low T4, TSH is low or normal, one must consider: hypothalamic disease, hypopituitrism, phenytoin therapy, and high doses of salicylates (rarely if ever used in pediatric practice). If a child presents with symptoms and signs of thyrotoxicosis, once again serum T4 and TSH are usually advised. If T4 is high and TSH suppressed, a diagnosis of primary thyrotoxicosis is made. If despite a low TSH, the T4 is normal, T3 needs to be performed. If T3 is elevated, the diagnosis is T3-toxicosis. However, if with an elevated T4, TSH is high or normal, these possibilities must be kept in mind: TSH secreting pituitary tumor, thyroid hormone resistance syndrome, antibodies against T4 which interfere with the assay, and finally drugs which interfere with the conversion of T4 to T3. It is important to remember that these are all very uncommon situations, and perhaps a repeat assay performed reliably may resolve the diagnostic dilemma without needing to invoke such esoteric diagnoses.

Finally, any combination of T3, T4 and TSH can sometimes be observed in the "sick euthyroid syndrome", though the most common finding is low T3 due to impaired conversion of T4 to T3. The TSH is usually normal in non-thyroidal illnesses. Hence thyroid function tests in an otherwise sick child with equivocal clinical features of thyroid disease need to be interpreted with great caution. Preferably these results should be verified by a repeat assay performed when the child has recovered from the illness.

In conclusion, there are many variables which can influence the biochemical assessment of thyroid hormones. Though they may not pose a frequent problem in clinical practice, it is important to keep them in mind when there is discordance between the clinical presentation and the laboratory values.

P Dougall (from CAPENEWS issue 11, April 2000)

Thyroid scanning is a simple non-invasive technique which not only delineates the anatomy but also gives functional information. The commonly used isotopes for scanning are Technetium-99m (Tc99m), Iodine - 131 (I-131) and Iodine - 123 (I-123). I-131 gives a higher radiation dose and is not suitable for children. Only in very special circumstances, where both trapping and organification are to be measured, it may be used in older children. With I-131 it is possible to measure the 2, 6 and 24 hour I-131 uptake, which may be helpful in differentiating Grave's disease from various types of thyroiditis, in a patient presenting with a thyroid profile suggestive of thyrotoxicosis. In 90% of cases the 20 minute Tc99m uptake is good enough to provide the same information. I-123 being a cyclotron produced isotope, is very expensive and not available in India, but it combines the useful properties of both I-131 and Tc99m. Tc99m is the most commonly used isotope the world over, as it is cheap, readily available and gives low radiation doses as compared to I-131 (0.1 rad / mCi vs 0.68 rad / mCi, to the thyroid). Iodine is essential for thyroid hormone synthesis and has a natural affinity for the thyroid - it is first trapped and then organified by the thyroid cells. Tc99m is trapped by the thyroid like iodine but not organified. After an intravenous injection, maximum accumulation of Tc99m occurs between 20 - 30 min. The conventional protocol for thyroid scan is to give a simple intravenous injection of Tc99m, acquire a flow study on the gamma camera at 2 secs / frame, followed by a 3 min. blood pool and 20 min. high resolution image. A 20 min. uptake is also calculated by the computer which gives an idea about the overall functional status of the gland. The entire procedure is over in about 35 min. Combined with FT3, FT4 and TSH measurement, a complete assessment of structure and function can be made using the Tc99m scan. The biggest advantage is that no patient preparation is required and the results are not operator dependent, unlike ultrasonography.

The usual indications for thyroid scanning in children include assessment of a neck swelling, abnormal descent and development of thyroid, thyroid nodule, toxic and non toxic multinodular goiter, a solitary thyroid nodule - autonomously functioning, or a non functional nodule (benign / malignant).

In neonatal hypothyroidism, several questions may be answered by the thyroid scan. If the gland is found to be ectopic, a permanent form of hypothyroidism has been established. Absence of uptake with the finding of no thyroid tissue on ultrasound confirms thyroid aplasia and permanent hypothyroidism. Absence of uptake with normal gland tissue occurs with defects in iodine trapping or in the presence of maternal thyroid blocking antibodies. A normal or enlarged gland on the scan suggests hereditary defects in thyroxin synthesis, or transient hypothyroidism. If the former, genetic counseling would be needed; in case of the latter, re-evaluation at the age of 3 years after stopping thyroxin treatment. Hypothyroidism diagnosed in later childhood could also be due to ectopia or dyshormogenesis.

Apart from goiter, a cyst in the thyroglossal duct is the most common mid-line anterior neck lesion detected in children. It seldom contains sufficient follicular tissue to be seen on a scintiscan, so it can be easily distinguished from a maldescended thyroid. Management differs: if the thyroid is seen in its normal anatomic location and the cyst lies outside, it can be easily excised; while if the midline neck swelling contains functioning thyroid tissue, it has to be left alone. Often, a sublingual thyroid has been picked up on the scan.

Of 'cold' i.e. nonfunctioning nodules, 16 - 20% are malignant. On the other hand, only 1-3% of 'hot' nodules are malignant. This may be helpful in ordering repeat FNAC, or selecting the site for FNAC to improve the diagnostic accuracy. Though relatively uncommon in children, thyroid malignancies can occur in adolescents and the initial diagnosis can often be made with the help of a thyroid scan. Post surgical evaluation for remnant thyroid tissue or distant metastasis can be very effectively achieved using the I-131 whole body scan. If detected, the remnant thyroid can be ablated using therapeutic doses of I-131 and thyroid metastasis from well differentiated cancers treated with high doses of I-131. Combined with serum thyroglobulin measurements, the whole body I-131 scan approaches 100% sensitivity and specificity for picking up metastasis.

Neonatal thyrotoxicosis is very uncommon. The vast majority of thyrotoxicosis in children is caused by Graves disease, whose peak incidence is between 11-15 years of age. Childhood disease is still uncommon, accounting for less than 5% of all cases of Graves disease. Painless thyroiditis is virtually unrecognised in children and subacute thyroiditis is rare. Autonomously functioning adenomas are also uncommon, but may be part of McCune Albright syndrome. Measurements of I-131 uptake and thyroid scintiscans may occasionally be needed in distinguishing other forms of thyrotoxicosis. Thyroid radio-iodine uptake is low in those few children with silent thyroiditis, subacute thyroiditis, and factitious thyrotoxicosis. A thyroid scan has a clear indication in the setting of single nodule with thyrotoxicosis to differentiate between an autonomously functioning nodule and an incidental nodule with concomitant Graves disease. A thyroid scan is always performed in a child with Graves disease before planning I-131 therapy for thyrotoxicosis.

In conclusion, the thyroid scan is a simple, non-invasive and cost effective way of assessing the underlying pathophysiology and thyroid structure in children, not needing any prior patient preparation or much patient cooperation, and not observer dependent to interpret. Being readily available, it should be used routinely in assessing developmental anomalies and establishing the underlying causes of neonatal hypothyroidism.

Dr Venkatesh Rangarajan (from CAPENEWS 5.2, August 2001)

Neonatal thyroid diseases manifesting as hyperthyroidism or hypothyroidism have been described. Detection and treatment of congenital hypothyroidism (CH) are more important as the consequences are long term and serious. Though there is hardly any data from India (1), indications are that our incidence of CH would be similar to, if not more than, the worldwide incidence of 1: 4000 births. Screening is necessary because symptoms and signs are almost always absent in the newborn, and by the time they appear a few weeks or months later, irreversible brain damage has already occurred. There is no longer any doubt that neonatal thyroid screening is highly cost effective. In spite of this, unfortunately so far in India it is offered mainly to newborns of mothers with thyroid illness, on thyroxin or anti-thyroid drugs, and occasionally to those with a strong family history of thyroid diseases, or to newborns with congenital malformations.

Today our pediatricians offer several expensive vaccines, outside of the basic EPI, to individual patients who can afford the cost. In the same way, there is need to begin screening as many newborns as possible, without waiting for the formulation of government policies for universal screening.

Currently there are centers offering thyroid function tests (TFT), which include TSH (by IRMA, ultra-sensitive RIA, or chemiluminence), total T4 (tT4), total T3 (tT3), free T4 (fT4), and free T3 (fT3), at very competitive rates due to centralized sample processing. Availability of testing facilities is no longer a problem in most parts of India.

In industrialized countries, heel prick capillary blood spotted on filter paper is the commonest type of sample for neonatal screening. Miyagi K et al (2) reported their experience with TSH in dried blood filter paper samples of 242 normal and 42 abnormal babies and concluded that the bioluminescence TSH assay could be used for detecting both primary and central hypo-thyroidism as well as hyperthyroidism in neonatal screening. In the US (3), tT4 is measured in all infants, with supplemental TSH in the same samples for those with T4 below the geometric mean value – 2.3 SD in a log normal distribution. Travet et al (4) measured fT4 in 10,000 screening samples and concluded that in screening programs using TSH, fT4 assay is an attractive improvement over tT4, whether performed as the first screening test or as a confirmatory test. Galloway (5) reported an audit from England on the organization of neonatal screening for CH, mentioning their shortcoming in timely reporting of assay results within 28 days and other difficulties.

In India, cord blood collection and testing for TFT may be simpler and more practical. The sample is sufficient for performing the entire spectrum of TFT. If cord blood has not been collected, sampling should be avoided in the first 48-72 hours, in order to avoid the TSH surge. The choice of the sample (dried filter paper spot or cord blood), test (primary TSH, primary tT4, primary fT4, or combination of TSH-T4), local availability, costs (which depend on the area and number of tests done), and interpretation are some of the issues which professional organizations and bodies in India must address when planning to start screening.

I have reproduced pediatric reference ranges for thyroid hormones below (6,7):
Reference ranges (2.5-97.5th percentiles) for Thyroid Hormones in Neonates, Infants, Children & Adolescents

AGE ug/dl nmol/L
Cord blood 4.6-13 59.2-167
1-3 days 11.8-23.2 15.19-198.6
3-10 days 9.9-21.9 127.4-281.9
10-45 days 8.2-16.2 105.5-208.5
45-90 days 6.4-14.0 82.4-180.2
3-12 months 7.8-16.5 100.4-212.4
1-5 years 7.3-16.0 94.0-193.1
5-10 years 6.4-13.3 82.4-171.2
10-15years 5.6-11.7 72.1-150.5

AGE ng/dl pmol/L
0-4 days 2.2-9.3 28-68
> 2 weeks 0.9-2.0 12-26

AGE ng/dl nmol/L
Cord blood 15-75 0.23-1.2
1-3 days 32-216 0.49-3.3
3-10 days 50-250 0.77-3.8
1-12 months 105-280 1.6-4.3
1-5 years 105-269 1.6-4.1
5-10 years 94-241 1.4-3.7
10-15 years 83-213 1.3-3.3


  1. Neonatal thyroid screening for congenital hypothyroidism using filter paper thyroxine technique. Desai MP, Upadhye P, Colaco MP. Indian J Med Res 1994, 100: 36-42.
  2. Evaluation of highly sensitive thyrotropin assay for detecting thyroid diseases in neonatal screening: Preliminary studies. Miyai K, Miyagi T, Ashida N et al. Endocr J 1998, 12, 45: 6,761-6.
  3. Modification of a screening program for neonatal hypothyroidism. J Pediatrics 1978, 92: 274-277.
  4. Free thyroxin measured in dried blood spots: Results for 10,000 euthyroid and 29 hypothyroid newborns. Travet G, Lemonnier F et al. Clin Chemistry 1985, 31: 1829-32.
  5. An audit of the organization of neonatal screening for phenylketonuria and congenital hypothyroidism in the northern region. Galloway, Stevenson J. Public Health 1996, 110: 2,119-21.
  6. Basic and Clinical Endocrinology. Greenspan FS, Baxter JD, 4th ed, Appleton-Lange, 1994.
  7. Establishment of reference ranges for thyrotropin, triiodothyronine, thyroxine and free thyroxin in neonates, infants and children and adolescents. Widemann G, Jonetz-Mentzel L. Eur J Clin Chem Clin Biochem 1993; 31:277-88.

Vaman & Anuradha Khadilkar (from CAPENEWS 6.2, August 2002)


Congenital hypothyroidism is a disorder which manifests with a frequency of approximately 1 in 4000 live born babies world wide. In India the disorder has been reported with a frequency varying between 1 in 2500 to 1 in 5000 (1). In the neonatal period the clinical manifestations are very subtle, hence in the majority of cases the disease is missed at this period. The only reliable method of detection of this potentially preventable cause of mental retardation is by neonatal thyroid screening. The neurological outcome is good in almost all cases if the deficiency is detected and adequate replacement started within the first 6 weeks of life. However if replacement is not started earlier than 6 weeks then irreversible brain damage is almost certain.

Subjects and methods:

TSH was measured on 203 full term babies born consecutively in Jehangir Hospital, Pune, irrespective of the mode of delivery. At the time of delivery, 3 drops of cord blood were taken on a Whatman filter paper DPC, USA and allowed to dry. The dried drop of blood on the filter paper (FP) was analyzed within one week. A hole of a specific diameter was punched into the blood stained part of the FP. TSH was extracted by elution method and estimated by radioimmunoassay. In 36 babies TSH was also measured on the cord blood using the ELISA method, and the values obtained by the two methods compared. Statistical analysis using SPSS for Windows was performed. Students paired t-test was used to compare the results between the serum method and the FP method. Results:

Two hundred and three samples (107 males and 96 females) were studied. One hundred and seventy seven babies were born by vaginal delivery and 26 by Caesarean section.

Table 1: TSH values (mIU/ml) measured by filter paper method in total and sub-groups of patients

Mean Std Dev
Total 12.3* 4.9
Males 12.3 5.0
Females 12.3 4.9
C-Section 11.3 4.1
Vaginal 12.5 5.1

* Minimum: 2.9, Maximum 28.5. Of the 36 paired samples studied there were 14 males and 22 females. Six babies were born by C-section while 30 babies were born vaginally.

Table 2: TSH values (mIU/ml) on paired samples

Mean Std Dev Mini-mum Maxi-mum
Serum 12.6 6.3 3.4 25.0
Paper11.9 5.9 3.3 28.5
Student's t-tests for paired samples: not significant


Neonatal hypothyroid screening programs have been successfully conducted in the developed world for over two decades and have completely eliminated this preventable cause of mental retardation. In India, no such national program is in place. The problems of successfully conducting such a program include

  • inconvenient methods,
  • cost, and
  • lack of reliable laboratories.
We therefore attempted to develop the same filter paper method which is in vogue in countries like the UK. The advantages of this method include:
  • small volume of blood,
  • no special container,
  • no need to separate the serum,
  • no special storage requirement (can be kept at room temperature) and
  • no immediate need for processing.
Another problem with our health care system is the complete lack of a social service network that can reach out to the babies at home. Modern obstetric trends tend to discharge babies early. Physiologically, TSH starts rising soon after birth and does not return to normal until the end of day five of life. Neonatal thyroid screening programs which test only TSH should therefore be done on blood samples collected after day 5, which is often not possible in our setup. Theoretically, it should be possible to use cord blood for the same purpose, however it is important to determine the cut off value based on local population norms.

An ideal neonatal thyroid screening program would use both T4 and TSH so that along with the common primary hypothyroidism, the rare secondary-tertiary hypothyroidism [frequency of approximately 1 in 100,000 live births (2)] are also diagnosed. However to diagnose secondary-tertiary forms a combination of T4 and TSH is required, increasing the cost of screening. Omission of T4 from the screening test does not miss very many hypothyroid babies because of the rarity of secondary and tertiary hypothyroidism. For a developing country like India it is cost effective and logical to use only TSH for neonatal screening. Thailand has introduced a National Neonatal Thyroid Screening Program on cord blood using TSH and a filter paper method (3).

We found no difference in the TSH values measured by filter paper method and the standard ELISA method. Also it is important to note that normal values of TSH on cord blood are much higher i.e. up to 30 mIU/ml as compared to the adult normal of less than 5 mIU/ml. We also confirmed that our results match the cut off values given by the Thai national program (3).


Cord blood can be used to screen neonates for hypothyroidism. Use of filter paper is a simple, convenient, reliable and reproducible method to carry out a neonatal thyroid screening program. The cuts off values for cord blood TSH are much higher (30 mIu/ml) as compared to values on day 5.


  1. Desai M. Recent trends in the management of thyroid disorders. Indian Pediatrics 1992; 29: 1465-1470.
  2. Delange F, Fisher DA. The thyroid gland. In: Clinical Pediatric Endocrinology, 3rd ed. Ed: CGD Brook, Blackwell Science, 1995; pp 397-433.
  3. Mahachoklertwatana W, Phuapradit W, Siripoonya P, Charoenpol O, Thuvasethakul P, Rajatanavin R. Five year thyrotropin screening for congenital hypothyroidism in Ramathibodi Hospital. J Med Assoc Thai 1999, 82 Suppl 1: S27-32.

(from CAPENEWS 8.3, December 2004)

Dr Annette Gruters, speaking on neonatal screening for congenital hypothyroidism (CH), showed the screening map of the world as of 2003: North America, Europe, Japan and Australia, with regional programs in southern Africa, Middle East, and South America. The frequency varies from 1: 11,000- 30,000 African Americans, through 1: 1,400 in Saudi Arabia, to 1: 800 in Ecuador, though in iodine replete areas it is mostly 1:3,000- 4,000 births. Some points she made were:

  • The problems they have commonly encountered included:
    1. Motivating parents to maintain regularity of therapy, because the child looks so normal.
    2. Sampling errors: due to insufficient sample or timing of sampling:
      1. TSH levels increase with gestation, being very low before 31 weeks. Thus in preterms, repeat screening may be is necessary at gestational age of 30 wks+;
      2. Levels of TSH and T4 are high in 24-36 hours of age: this normal surge could cause confusion if the timing was not mentioned clearly.
    3. Use of betadine for disinfection: the resultant iodine excess causes a TSH increase which later declines: this increases the frequency of false positives.
  • Thyroglobulin (Tg) levels and an ultrasound in all newborns diagnosed to have CH could substitute for a radio-nuclide thyroid scan. Low T4, low Tg, and no thyroid tissue on the ultrasound were characteristic of athyreosis; measurable T4 and Tg with no thyroid tissue on the ultrasound suggested an ectopic gland, while measurable T4 and Tg and a visualized thyroid gland were suggestive of a eutopic gland.
  • IQ depends on when the treatment is started: the earlier the better: with normal IQs if started before the age of 2 wks, even with low doses. Ideally, thyroxin should begin before 3 wks of age, in a dose of 10-15 ug/ kg/ day.
  • Newborn thyroid screening also works as a monitoring device for iodinization programs, by detecting areas of iodine deficiency.
  • Areas with marked iodine excess may encounter the phenomenon of a delayed TSH surge. In Japan, TSH is reassessed in the 3-4th wk of life.
  • Initially, when only thyroid screening was being done, most countries used cord blood. When other disorders were added, sampling time was changed. In our circumstances, cord blood screening is the most practical way to start.

Dr Meena Desai discussed the problems and challenges in the CH screening program in Mumbai: the slide of a postbox filled with postcards sent by her team for recall of suspected infants but never picked by the slum dwellers they were addressed to said it very graphically. She showed a similar incidence whether cord blood screening used primary TSH (cut-off > 25 uU/ ml; 1982-1984, found incidence 1: 2,481) or primary T4 (cut off < 6.5 ug/dl; 1986-88, incidence 1:2,800).

Dr Raghupathy, discussing the CH experience at Vellore, said that all babies born at the Christian Medical College have been screened since July 2001, using cord blood. In all those with TSH > 25 uU/ ml, a repeat test is done at age > 72 hours. Thyroid scans are done only with I-131, because technetium scans are known to miss some cases. Over 3 years (2001- July 2004) 22,853 babies were screened: 825 had TSH >25 uU/ ml: repeat tests were done in 741: and 17 (5 of whom were born of consanguineous marriages) diagnosed to have CH: an incidence of 1:1,344. He juxtaposed this incidence with the figures from Birmingham, UK: 1: 5,400 Indians, 1: 781 Pakistanis (Thyroid 2002: 12:591-598). He said they were lucky that a baby was diagnosed early in their program, otherwise it might have been difficult to persuade hospital authorities to continue screening!

Dr Vaman Khadilkar described CH experience at Pune, in cord blood on filter paper, using a TSH (done by radioimmunoassay) cut-off level of 30 uU/ ml. He emphasized the advantages of cord blood: no need to prick the newborn, 100% coverage, sufficient sample, and high acceptance by parents. He pointed out Thailand has the largest cord blood screening program in the world; the cut-off used is > 35 uU/ ml. Dr Gruters commented that in the US, cord blood was used in the 1970s, and the TSH cut-off used was 50 uU/ ml; in Germany now blood was drawn on day 3 of life, and with TSH tested by an immunoflurometric assay, their cut-off was 15 uU/ ml.

Dr Grueters discussing the genetics of neonatal thyroid disease, pointed out that

  • With TPO gene mutations, only 20-30% patients have goiter at birth, though 100% develop it later in life.
  • TSH receptor mutations are autosomal recessive, and can cause varying degrees of hypoplasia or aplasia.
  • PAX8 mutation result in cystic mal-formations of the thyroid gland.

Thyroid Function in the Newborn and Neonate: Relevance in Clinical Practice. CB Sridhar, in discussion with Prof Odell.

Sridhar: Prof Odell, please discuss thyroid function in the newborn and neonate.

Odell: Fetal thyroid function begins at the end of the first trimester. Thereafter there is a steady increase in fetal thyroid binding globulin, total T4 and T3. The gradient favors maternal to fetal transfer, which is abolished by the 35th week. Thyroid hormone production rates per unit body weight are greater in neonates and children than in adults. In the cord blood mean T4 is 150nmol/l (12ug/dl), free T4 concentrations are slightly less than maternal values; T3 levels are low (50 ng/dl) and reverse T3 levels are high. After delivery, neonatal TSH surges rapidly to a peak at 30 minutes of extra uterine life, and returns to baseline by about 48 hours. The relative decrease of environmental temperature following delivery is believed to be the cause for this. Anti- TPO and thyroglobulin antibodies do not transfer thyroid disease from mother to fetus. Thyroid stimulating antibodies (TSA) may be transferred through the placenta and cause transient neonatal hyperthyroidism, which subsides in 4 weeks.

CBS: Is the concern of mothers on thyroxin replacement about transfer of hormones in breast milk justified? WDO: Thyroxin is secreted in breast milk, but in euthyroid mothers the amount is small and is balanced by a mild decrease in the infant's thyroid secretion, so the infant maintains euthyroidism. Mothers who are thyrotoxic or on excess thyroid hormone replacement may transfer excess amounts of thyroxin to the infant and produce hyperthyroidism. Therefore it is important to maintain euthyroidism in the nursing mother.

CBS: How should maternal hyper-thyroidism be managed? What is its impact on the neonate? WDO: As I just said, transient neonatal hyperthyroidism can develop when mothers with hyperthyroidism transfer TSA through the placenta. This subsides in 4-6 weeks. Anti-thyroid drugs (neomercazole, propylthiouracil) are also secreted in breast milk. Generally, we advise women taking these drugs not to nurse their infants. These drugs also readily cross the placenta during pregnancy and are concentrated by the fetal thyroid, so the lowest possible dose should be used to treat hyperthyroidism during pregnancy. PTU is generally preferred because it has lesser transplacental passage. Because of the fall in TSA in pregnancy, the dose of anti-thyroid drugs required is less than in the non pregnant state.

Ram Murty Shastry, Goa

A successful and novel strategy to rapidly normalize serum free thyroxin level in primary congenital hypothyroidism. (Paul Hofman et al, Starship Children's Hospital, Auckland, New Zealand)

36 infants (25 girls) with CH were started on L-thyroxin at 15, 12 or 10 ugm/kg/day for athyreosis (n=5), ectopia (n=20) and dyshormonogenesis (n=10), respectively. Serum TSH, FT4 and FT3 levels were monitored weekly for 4 weeks and L-thyroxin dose adjusted to maintain FT4 in the normal neonatal range.

29 infants had initial TSH > 100 mIU/L and 7 had TSH > 40 mIU/L. Initial FT4 values were < 11pmol/L in 31 of them. At the end of 1st, 2nd, 3rd and 4th weeks of treatment, 41%, 38%, 12% and 11%, respectively required dosage adjustment. At the end of 1st week of treatment, 80% children had FT4 levels in the normal/above normal range and remaining 20% normalized in the 2nd week. In each week of treatment, 16%, 14%, 3% and 3% infants had FT4 values above the upper limit of normal without thyrotoxic symptoms. The TSH was corrected to the normal range in 33%, 48%, 15% and 5% at the end of 1st, 2nd, 3rd and 4th weeks of treatment.

Conclusion: This approach successfully normalized FT4 in the majority of subjects within a week without marked or persistent hyperthyroidism. The authors have proposed initial doses (in ugm/kg/day) of 15 for athyreosis, 10 for ectopia and 8-10 for dyshormonogenesis. Frequent dosage adjustment in the first month will reduce the impact of incorrect dosing.

Juvenile hyperthyroidism: an experience (A Bhansali et al, Indian Pediatrics, April 2006)

The authors have retrospectively analyzed 56 patients with juvenile hyperthyroidism: 38 females and 18 males: mean (+SD) age of 14.9+3.4 yr (range3-18 yr). Common presenting symptoms were weight loss (82.1%), excessive sweating (78.6%), heat intolerance (76.8%), increased appetite (73.2%), diarrhea (48.2%) and accelerated linear growth (7.1%). Goiter was present in 98.2%: mainly diffuse (94.5%), and multinodular in 4.8%. All patients were treated with carbimazole (0.6-0.8mg/kg/body weight) till they were euthyroid, titrated at 3 monthly intervals, with the maintenance dose (5mg) continued for 12 months. Mean (+SD) duration of follow up was 4.9+3yr (range, 1.6-16yr) and mean duration of treatment was 34.4+22.6 mo (range 12-120 mo). Remission was achieved in 47.6% patients in 1-33 mo (mean 5.2+4.7 mo). 1/56 patients developed agranulocytosis after 3 yr of carbi-mazole therapy and was given radio-iodine ablation. The overall incidence of drug side-effects was 1.8%, much lower than in other pediatric series. The authors concluded that carbimazole is cheap, effective, easily available and rarely associated with side effects.

Comments: The optimal treatment of juvenile Grave's disease in children and adolescents remains controversial, though most children can be maintained on a once daily, low dose of carbimazole for 1-3 yr, without much adverse effects and with prolonged remission on withdrawal in many cases. Even in children who do relapse after drug withdrawal, many prefer to maintain euthyroidism with low once daily carbimazole for many years, reserving thyroid ablation for those in whom treatment with ATD has been unsuccessful.

Intellectual and motor development of young adults with congenital hypothyroidism diagnosed by neonatal screening is reported by MJE Kempers et al from Emma Children's Hospital, Amsterdam (JCEM vol 91, no 2, 2006). The authors examined the cognitive and motor functioning in young adults with CH, born in the first 2 years after introduction of the Dutch neonatal screening program. 70 patients (mean age 21.5yrs) were tested; 49 of them were previously tested at 9.5yrs. The median age at the start of treatment was 28 days (range, 4-293 d). CH was classified as mild, moderate and severe, according to pre-treatment T4 levels. Patients, particularly those with severe CH, had significantly higher (ie worse) motor scores compared with controls, and lower full scale, verbal and performance IQ scores than the normal population. Among patients with severe CH, 37% had full scale IQ < 85 and 49% had subnormal motor score. Initial T4 concentration and severity of CH was associated with full scale IQ and motor score; but no correlation with the starting day of therapy. No significant change in IQ from childhood to adulthood was found, and for the majority of patients, motor score classification remained the same at 9.5yrs and 21.5yrs. The authors concluded that cognitive and motor deficits in CH patients who started treatment at a median age of 28 days after birth, persist into adulthood, predominantly in severely affected CH group. The outcomes could not be correlated with age of starting therapy.

Comments: There are several caveats to the interpretation of this paper-

  1. The median age of starting treatment was very late as compared to current practice. In fact, the range was as bad as 18-293 days in the mild group of CH.
  2. The kind of dose regimens used (because those were the earliest years of newborn screen programs) may have resulted in lower and slower achievement of blood and CNS thyroid hormone levels than in currently used regimens.
However, the value of this paper lies in the long term nature of follow-up, and the knowledge that motor as well as IQ deficiencies noted by numerous studies in childhood in children who are the most severely affected but are treated by newborn screen, also persist in young adulthood.

ESICON 2005: SOME POINTS (AP Weetman, Sheffield, UK)Back to Top

THYROID HORMONE REPLACEMENT: The commonest reason for a high TSH in spite of an adequate dose (>1.6 mcg/kg/day) is poor adherence. Once one keeps in mind that one third of renal transplant loss and 40% glaucoma caused blindness is due to poor compliance to therapy, it is not difficult to understand that patients often forget their thyroxin. Tricks to try:

  1. Ask the patient to make a weekly replenishment bottle, in which the week's supply of thyroxin is taken out. Whatever remains at the end of the week can be taken together as a single dose.
  2. Ensure that interfering drugs (iron, aluminium hydroxide, rifampicin, carbamazepine, etc.) are being 6-8 hours away from the thyroxin dose, as far as possible.
  3. Rule out malabsorption: absorption can be confirmed by administering the patient 1000 mcg thyroxin every week for 4 weeks, and then re-testing TSH.
  4. Yamamoto (2003) reported 3 patients whose TSH dropped only after the thyroxin tablets were pulverized, remaining high when the tablets were taken whole. If TSH does not come down, ask patients to pulverize the tablets.
THYROTOXICOSIS: In thyrotoxicosis, block replacement, i.e. adding thyroxin while continuing neomercazole once the TSH starts rising, works better than conventional therapy, with remission in 6 months cf. 18 months, less fluctuations in thyroid status, and better compliance. It cannot be used during pregnancy. The incidence of hepatotoxicity due to anti-thyroid drugs is < 0.5%. Baseline abnormalities in liver function may be due to the thyrotoxicosis itself. If planning surgery for thyrotoxicosis, addition of colestyramine (4 gm tds in adults) reduces free T4 by 61% at 2 wk, cf. 43% with anti-thyroid drugs alone. Radio-iodine therapy worsens ophthalmopathy, so pre-treatment steroids may be advisable.

Do remember that iron and soy protein interfere with the absorption of thyroxine, so they should be given some time apart. I got this interesting tip for administering thyroxine to infants: "I usually tell parents to cut the synthroid in quarters and put the quarter pill in a quarter tespoon of baby fruit. If the synthroid comes back out, you see it and you can just "shovel" it back in. I've never had a baby choke or not be controlled. A vet once taught me to gently stroke over the trachea. This causes a reflex in animals and human babies to swallow!" Tessa Lehbinger

Clippings For CapenewsBack to Top
Ram Murty Shastry, Goa CAPE NEWS 11-3 , APRIL 2007

Iodine Nutritional Status of exclusively breast-fed infants and their mothers in New Delhi. Rishi Gupta, Anju Seth, CS Pandav, MG Karmarkar, S Aneja. JPEM; 2006; 19 (12). Iodine deficiency during the period of brain growth (intrauterine period and first 2 years of life) has a profound effect upon brain development since the synthesis of thyroxin is dependent upon an adequate supply of iodine. Iodine deficiency is an important public health problem in India. However, there are no published data on iodine nutritional status of pregnant/ lactating women and infants.

This cross-sectional study, with the objective of assessing the iodine nutritional status of exclusively breast-fed infants and their mothers, was conducted in the Department of Pediatrics, Kalawati Saran Children's Hospital & Lady Hardinge Medical College, New Delhi, in collaboration with the International Council for Control of Iodine Deficiency Disorders.

Spot urinary iodide (UI) and serum TSH levels were measured in 175 healthy, exclusively breast-fed infants (aged 4-24 weeks) and their mothers (aged 16-40 years). Iodine content of salt used by the participants for domestic consumption was also analyzed.

89.7% of mothers had BMI in the range of 18.5-24.9; of 175 infants, 90% had weight by height >=80% of the expectedvalue, suggesting normal nutritional status in both mothers and infants. Median UI levels in mothers and infants were 124ugm/l and 162ugm/l respectively, with 34% mothers and 21% infants having UI levels < 100ugm/l (indicating iodine deficiency). Of the mothers, 44 had mild deficiency (UI 50-99ugm/l), 14 had moderate (UI < 50ugm/l) and 1 had severe iodine deficiency (UI < 20ugm/l). Of the infants, 23 had mild, 11 moderate and 3 infants severe iodine deficiency, respectively. Serum TSH was elevated in 29% mothers (>5mIU/l) and 2% of infants (>9.1mIU/l). No correlation was observed between individual mother infant UI or serum TSH levels. 96% of the salt samples tested had adequate iodine concentration, ie >15ppm.

Conclusion: The study demonstrated significant iodine deficiency in both mothers and infants despite consumption of adequately iodized salt. The iodine nutritional status of the infants was better compared to the mothers, indicating a preferential iodine supply to the infants over the mothers.


  1. The results of this study emphasize that any program of salt iodization in a population must pay particular attention to these vulnerable groups.
  2. In countries known to be iodine deficient, specific iodine supplementation should be considered for pregnant/ lactating women and non-breast-fed infants to ensure optimal brain development.
  3. No correlation was observed between maternal and infant UI levels. This is understandable: UI excretion, while a good indicator of iodine nutritional status of a community, has been found to have poor sensitivity for individuals

Excerpts from international update Mumbai 2008Back to Top
Cape News , Vol 12-1, April 2008

Congenital hypothyroidism:
Almost all programs of newborn screening are based on detection of elevated TSH concentrations alone or in combination with T4/fT4.TSH in cord blood samples using similar cut offs as after 3 days of life can be used as an alternative approach. Repeat testing at 2-6 weeks of life can detect an additional 10% of infants with CH.Asymptomatic hyperthyrotropinemia occurs in 1:8000 births, with 50% due to perinatal iodine exposure, eg with use of betadine as disinfectant. This condition is usually treated, but can be managed expectantly if fT4 is in the upper half of normal range. Serum TSH in some treated infants with proven CH may remain relatively elevated for a few months despite normal T4, due to a resetting of the feedback threshold for T4 suppression of TSH release. The resetting occurs prematurely but the mechanism is obscure. (A Greuters)

Thyroid function in the premature infant:
VLBW, ELBW infants are predisposed to development of transient primary hypothyroidism and the syndrome of transient hypothyroxinemia of prematurity due to their immature hypothalamic- pituitary- thyroid system and high neonatal morbidity. Transient primary hypothyroidism (low T4, high TSH) varies geographically, relative to iodine intake.

Premature infants require higher iodine levels than term babies to maintain adequate T4 production in extrauterine environment. Fetal thyroid is also inordinately sensitive to inhibitory effects of iodide on hormone synthesis (perinatal iodide exposure, eg betadine use). This will require treatment if the hypothyroidism persists for several weeks. (A Greuters)

Transient hypothyroxinemia of prematurity:
In ELBW, VLBW infants serum TBG, T4, T3, and fT4 are lower, with an obtunded TSH surge. This immaturity is inversely proportional to their gestational ages. Nadir values are reached by 7-10 days of life, with reequilibration with cord values by 3-4 weeks. The concomitant high prevalence of neonatal morbidity leads to characteristic changes of nonthyroidal illness. The impact on brain maturation is not clear. (A Gruters)

Primary hypothyroidism (PH) and Histiocytosis: In an unusual case, PH with goiter was an early presentation of Langerhans cell histiocytosis in a five year old girl, followed by progressive CNS involvement, hypothalamic mass,hypocortisolemia, and GHD, and subsequent death. Provisional diagnosis made by FNAC of thyroid and confirmed by electron microscopy of thyroid tissue (birbeck's granules). Infiltrative disorders might need to be ruled out in primary hypothyroidism with unusual presentations, such as firm goiter at an early age, as in this child. (L Priyambada, V Bhatia)