Abstract
Congenital hypothyroidism is screened for in the UK using blood spot thyroid-stimulating hormone (TSH) screening at 5–8 d of age. Although standards are set by the UK Newborn Screening Programme Centre, there are variations in TSH cut-offs used. The introduction of repeat screening of preterm babies at 36 weeks' gestational age in 2005 was controversial in its utility and timing. Two cases of preterm babies are presented, who had normal blood spot TSH values on the first test and who became screen positive when re-tested at term. The first with Trisomy 21 was born at 29 + 6 weeks with an initial blood spot TSH of 3.3 mU/L rising to 263 mU/L at term-corrected gestational age (plasma TSH 476.5 mU/L). The second was born at 24 + 6 weeks’ gestational age and on day 7, the heel prick blood spot TSH was <2 mU/L, rising to 6.4 mU/L at 36 weeks corrected gestational age. After a barium enema, the plasma TSH increased to 66.6 mU/L with a free thyroxine of 7.6 pmol/L at day 101. Both cases were treated with thyroxine until death due to complications of prematurity. These cases illustrate the difficulties in screening for congenital hypothyroidism in preterm infants, due to the immaturity of the hypothalamo–pituitary–thyroid axis, and the effect of intercurrent illness and drugs on thyroid function. Despite a reassuring published review of 2200 preterm infants, these cases suggest that it may be unwise not to re-screen ex-preterm infants for congenital hypothyroidism at term.
Background
Congenital hypothyroidism occurs in about 1:4000 live births with most cases being due to agenesis of the thyroid, but with a significant proportion being due to dyshormonogenesis. Untreated congenital hypothyroidism causes mental and growth retardation, which can be prevented by early thyroid hormone replacement. In the UK, all infants are screened using blood spot thyroid stimulating hormone (TSH) measurements at 5–8 d of age. Standards for screening, diagnosis and treatment for infants with hypothyroidism are published by the UK Newborn Screening Programme Centre. 1 According to these standards, babies with blood spot TSH concentrations between 10 and 20 mU/L are recalled for repeat blood spot analysis and values greater than 20 mU/L are regarded as a positive test leading to further immediate investigation. There are variations in the cut-off TSH blood spot screening and whole-blood values that are used around the UK.
In Wales, blood spot TSH concentrations greater than 20 mU/L require immediate referral to paediatricians for investigations whereas values of 5–20 mU/L require repeat blood spot analysis and referral if TSH remains greater than 5 mU/L. Approximately 35,000 babies are screened in Wales annually. In 2005, the UK Newborn Screening Programme Centre produced recommendations, which specified that screening of preterm infants should be repeated at 36 weeks' corrected gestation. This was based on the observation that immaturity of the hypothalamo–pituitary axis in very low birth weight and preterm infants may mask congenital hypothyroidism. 2 Interpretation of thyroid function testing is difficult in preterm babies and there is controversy in the literature and among professionals about the utility and timing of the second screen in preterm infants. 3 Six per cent of the specimens processed in Wales (approximately 2000) are repeat specimens, due to the fact that the babies were preterm. Wales has adopted the UK national guidelines with regard to the testing of preterm infants, i.e. that they are re-tested at term-corrected gestational age. Proposals have been made to reduce this burden of re-testing by being more selective in the population undergoing repeat measures. This is currently under review by the UK Newborn Screening Programme and at the time of writing guidance is awaited. We present two cases of preterm infants, identified at our regional reference laboratory, over a three-year period, with normal blood spot TSH values at first test who became screen positive when re-tested at term.
Case 1
A male infant was born at 29+6 weeks' gestational age, weighing 0.725 kg. At delivery, the baby was intubated, surfactant administered and was ventilated for 12 h. By day 35, he was off continuous positive airway pressure (CPAP) support and breathing spontaneously in air, but was commenced on diuretics for chronic lung disease. At day 58, a karyotype confirmed 47XY + 21. On day 7, the initial screening blood spot TSH concentration was 3.3 mU/L (plasma TSH equivalent = 7.3 mU/L). At term-corrected gestational age (day 69 of life), the repeat screening (whole blood) TSH was 263 mU/L (plasma TSH equivalent = 578.6 mU/L). The plasma TSH at this time was 476.5 mU/L (reference range 0.5–5), with a free thyroxine (T4) of 5.7 pmol/L (reference range 10–29).
On ultrasound scanning, the thyroid gland appeared normal. A technetium radioisotope scan showed an area of uptake in the neck consistent with a normally located thyroid gland. Thyroxine replacement therapy was commenced at a dose of 10 μg/kg body weight. Following the starting of treatment, the plasma TSH fell to 4.0 mU/L, with a free T4 of 22.3 pmol/L.
On day 99, he was recommenced on CPAP support during an episode of presumed sepsis and further deterioration in pulmonary function led to re-intubation and ventilation. Care was eventually withdrawn when he remained in respiratory failure despite maximal intensive care support.
Case 2
The second case was a male infant born at 24+6 weeks’ gestational age, weighing 0.830 kg, requiring intubation and surfactant administration, and ventilation until day 63. A section of small bowel was removed on day 46 for necrotizing enterocolitis. On day 7, the heel prick blood spot TSH was less than 2 mU/L (plasma TSH equivalent = less than 4.8 mU/L). At 36 weeks' corrected gestational age, the heel prick blood spot was 6.35 mU/L (plasma TSH equivalent = 14.0 mU/L), with the repeat value being 6.05 mU/L (plasma TSH equivalent = 13.3 mU/L). Plasma TSH concentration was 8.67 mU/L, with a free T4 of 14.5 pmol/L.
After a barium enema, the plasma TSH increased to 66.6 mU/L with a free T4 of 7.6 pmol/L at day 101. Subsequent to this, thyroxine was commenced at 10 µg/kg, and after two weeks the TSH increased to 86.9 mU/L. However, after a period of septicaemia, the thyroxine dose was increased to 12 µg/kg/d and the TSH fell to 4.99 mU/L with a free T4 of 20.2 pmol/L.
The outcome for this infant was also withdrawal of care, after developing short gut syndrome, parenteral-nutrition-related liver disease, retinopathy of prematurity and post-fungal meningitic hydrocephalus. In this case, the hyperthyrotropinaemia may have been due to iodine exposure from the gastrograffin contrast used in the barium enema, a known cause of neonatal thyroid suppression. 3
Discussion
The interpretation of thyroid function tests in preterm infants is complex due to immaturity of the hypothalamo–pituitary–thyroid axis, the effect of intercurrent illness on thyroid function and the risk of exposure to drugs that affect neonatal thyroid function such as iodine-containing compounds. Preterm babies are less able to regulate iodine balance. 4 Furthermore, the inhibition of TSH by T4 develops from 27 weeks' gestation onwards, so this feedback loop is in the process of maturing in preterm infants. In term infants, TSH normally peaks within 30 min of delivery, with a subsequent logarithmic decrease in values over the next two to four days. In preterm infants, this surge in TSH is obtunded, and may be practically non-existent in the context of intercurrent illness.
These two cases illustrate the difficulties in screening for congenital hypothyroidism in preterm infants. Case 1 was an infant with Down's syndrome. Although thyroid dysfunction is known to be more prevalent in these children, there is little evidence of an increased frequency of thyroid gland dysgenesis. 5 It has been argued that the increased frequency of biochemical thyroid dysfunction reflects a change in the set-point for pituitary–thyroid signalling, though recent studies have neither shown evidence of bioinactive TSH nor molecular defects in the TSH receptor. 6 The possibility of a chance association of congenital hypothyroidism with Down's syndrome and prematurity cannot be excluded, with a prematurity-induced delay in TSH response to primary hypothyroidism masking the underlying diagnosis when initially screened.
In case 2, it is likely that exposure to iodine, contained in the gastrograffin enema, resulted in an elevation in TSH by the Wolff–Chaikoff effect. This is likely to have been a transient cause of hypothyroidism but unfortunately the infant died before this could be ascertained. Nevertheless, this case demonstrates that premature infants may be exposed to iodine-containing compounds which need to be considered in the differential diagnosis of disordered thyroid function.
A review of 2200 preterm infants did not identify any further cases in which re-testing at 36 weeks had resulted in a requirement for long-term thyroxine treatment, using a cut-off level of 6 mU/L on blood spot testing. 3 We have presented two cases of increased blood spot TSH on repeat screening of preterm infants at 36 weeks' corrected gestational age in Wales. These two cases suggest that it may be unwise to abandon the re-screening of ex-preterm infants at 36 weeks' corrected gestation despite the practical difficulties involved in ensuring this occurs and the reassuring data published from Newcastle upon Tyne. 3
DECLARATIONS