Background The recommended screening cut-off for congenital hypothyroidisim (CHT) in the UK is a blood spot Thyroid Stimulating Hormone (TSH) of 10 μ/l but centres use values ranging from 5–10 μ/l. UK guidelines (2005) recommend that the preterm infant (<36 w) should have a second blood spot sample at the equivalent of 36w gestation because of the potential inability of the preterm infant to mount an appropriate TSH response.
Objective To determine the value of the second blood spot screen in preterm infants.
Methods Regional whole blood TSH data in preterm infants (<35 weeks gestation) from a 5 year period (2005–2010) were assessed. The number of infants who had values greater than the local screening threshold (6 μ/l) were identified.
Results A first TSH sample was obtained from 5518 infants (median gestational age 32w, range 22–35), with a second sample obtained from 5134 (32w, range 23–35). In 38 infants, TSH concentration fell from above to below the screening threshold and in six infants values rose to 6–10 μ/l. TSH concentrations fell below 6 μ/l in five of these infants and the remaining baby had a serum TSH of 6.8 μ/l with a normal free thyroxine. Five infants had raised TSH concentrations (>6 μ/l) on both occasions (birth weight 830–2810 g).Three of the five infants had unequivocal hypothyroidism (serum TSH >80 μ/l on second screen). The initial TSH concentration on one of these three infants was 8.5 μ/l (subsequent serum TSH 134.7 μ/l, free thyroxine 10.4 pmol/l, birth weight 2810 g). The remaining two infants (birth weight 830 g and 2760) had an initial TSH of 6.8 and 11 μ with serum values of 19.7 and 8.3 μ/l respectively.
Conclusions A second screening test will detect babies with CHT when a 10 μ/l cut-off is used. Preterm infants with CHT and a respectable birthweight (>2000g) can have a bloodspot TSH less than 10 μ/l on first screen. However a ‘once-only’ approach using our local cut-off of 6 μ/l may make a second test redundant. The incidence of CHT in preterm infants (<36w) is higher (∼1 in 1100 to 1 in 1800) than the incidence in newborns as a whole (∼1 in 3500).