The present study shows that the recommended timing for screening, ideally between days 3–6 after delivery, is not complied by all health care units. This supports the need for action, particularly the establishment of clearer written guidelines and standard procedures.
In some units the CH screening is postponed in children fed with no protein intake. This represents no reason for delay since the absence of protein intake does not compromise the assessment. Of notice, this procedure was already observed in 2011 and no improvement has occurred, which highlights the need to reinforce information on best practice guidelines.
Early CH diagnosis and treatment may optimize children’s psychomotor development and prevent neurological disability. Diverse approaches to the screening are used worldwide: TSH determination, T4 measurement or a combined strategy of primary TSH with TT4 as a second-tier test (the ideal screening approach, though the most expensive) [7]. Portugal adopted the latter.
Particular groups of neonates are described to have increased risk of CH; these include preterm and VLBW babies, in which a second and a third sample collection are recommended by several Paediatric Endocrinology Societies [14–19]. The reason for these additional measurements relies on the expected lag in the postnatal TSH rise (absence or delay of TSH surge) and on potential hypothyroxinemia, due to hypothalamic-pituitary-thyroid axis immaturity. Both are strongly influenced by the gestational age: the lower the gestational age, the lower and delayed the TSH, TT4 and FT4 peaks [18, 20, 21–27]. Several studies have reported that screening solely based on an increase of TSH levels likely results in a failure to detect CH in most preterm/VLBW babies. As such, repetition of the diagnosis and inclusion of T4 measurement likely justifies the higher rate of diagnosis CH and/or a CH with delayed TSH rise in this population [1, 2, 4, 6, 28, 29]. The guidelines for additional screening in preterm/VLBW babies are presently in place in all NCU and GPU, which represents an improvement of the situation from 2011 (when retesting was below 80%). The standard screening time is between day 3 and 6, the first recall blood sample two weeks after birth (day 14–15) and the second recall at the fourth week. It is of relevance to refer that in extreme preterm babies, the third sample should not be collected prior to 30 weeks (desirably at 32 weeks), which deserves particular attention in revised guidelines.
The environment of NCUs is very adverse to thyroid function, even for full term babies. Babies seriously ill, using dopamine, caffeine and corticosteroids are likely to present impaired thyroid function [15, 28, 30]. Dopamine inhibits TSH and T4 secretion [28, 31–33]; caffeine may increase the expression of the gene encoding for TSH [30]; corticosteroids may lead to reduction of T3 and TSH levels and elevation of reverse T3 [33, 34]; many other drugs may affect thyroid function (e.g. aminophylline, heparin, amiodarone). In nonthyroidal illness (also called sick euthyroid syndrome) babies are described to have low serum T3 levels, normal to low TT4, elevated reverse T3 and normal TSH [34, 35]. Therefore, babies on medication should repeat the screening [24, 37, 38], a procedure not widely accomplished for Portuguese babies, as observed here.
Other groups of neonates are at risk of CH, but guidelines are far from consensual: in our sample 31% NCUs retested babies with Down syndrome and 15% of them retested babies before a blood or exchange transfusion. These two clinical situations are not included in the current Portuguese guidelines. Other risk categories of neonates include twins (especially same-sex twins) and preterm babies older than 30 weeks (especially if small for gestational age), clinical conditions that may require repetition of the metabolic screening [15, 22]; this deserves to be analysed in terms of the cost-benefit if to be universally applied.
One special word with respect to iodine. Preterm babies are very prone to thyroid dysfunction due to excessive iodine [7, 39–41]. If a very immature thyroid is exposed to a great amount of iodine, it will stop functioning (Wolff-Chaikoff effect) to avoid overproduction of thyroid hormones. This arrest may lead to transitory hypothyroidism that may persist until maturation of the gland (36–40 weeks), especially in iodine deficient populations as it is the case of Portugal [42]. In fact, due to the increased permeability of the skin and decreased renal clearance, a single use of iodinated disinfectants in maternal caesarean, vaginal douching, neonatal skin disinfection or iodinated contrast media may lead to hypothyroidism [18, 43]. The most popular iodinated disinfectant used in caesareans has 105 mcg of iodine/mL. Since new-borns need around 150–180 mcg/day of iodine, a single use may lead to neonatal iodine overload, even in full term babies. For this reason, there is a strong recommendation to exclude iodine-based disinfectants from maternal, obstetrics and neonatal wards, which unfortunately is still not done in many of the units in our study.
A final note on the timing between sample collection and assessment. While collection is mostly done within the recommended timing, the results of the test may be compromisingly delayed due to administrative procedures of Guthrie cards mailing. The reasons for this delay reside on mailing just on working days (and most likely not on all working days), use of ordinary rather than express mail or even requesting that mailing is done by the parents. Guidelines to optimize the time between collection and assessment outcome are needed.
This study has limitations. Questionnaires were sent to 91% of the public national NCUs and to all CPUs of a single geographic region (Guimarães), with a response rate under 50% (42% in case of NCUs and 48% for GPUs). It also did not include private NCUs. As such it may not be generalizable to the whole country. Selection and reporting bias cannot be excluded; those who responded may be the best performers. Still, the information obtained alerts for the need to improve communication and to revise and update current guidelines with respect to new-born screening.