Neonatal Thyroid Screening TSH Interpreter

Newborn blood spot TSH read against the cut-offs of the programme that produced it — with the unit trap that matters most: a blood spot TSH is a whole-blood concentration and is about two thirds of the serum TSH it is so often compared with.

Last updated 16 Sep 2026

Newborn blood spot TSH

Blood spot TSH + programme + timing → interpretation

Blood spot TSH (mU/L whole blood)The number on the newborn screening report, in mU/L of WHOLE BLOOD. This is not a serum TSH and must not be read against a serum reference interval. On a published regression, blood spot TSH = 0.68 × serum TSH + 0.07, so a blood spot result is roughly two thirds of the serum value — a blood spot 8.0 mU/L corresponds to a serum TSH around 11.7. The blood spot reference interval for a neonate sampled on day 5 to 8 was 0.40 to 4.10 mU/L in the same study.

Which screening programme’s cut-offs?There is no universal cut-off, so the programme has to be named. The UK uses an analytical cut-off of 6.0 mU/L whole blood, an action cut-off of 8.0 and a referral threshold of 20.0. A review of newborn screening records that ‘the cutoff for reporting an elevated TSH is a level above 20 to 25 mU/mL in most screening programs’, while some US locations have taken theirs as low as 5 mU/mL and one Chinese survey found domestic cut-offs ranging from 4.82 to 18.0 µIU/mL. A US reader should take the figure from their own state programme rather than from this page.

Which sample is this?The UK pathway samples on day 5, counting the day of birth as day 0; a review of screening practice gives the ideal window as 3 to 5 days after birth, to let the physiological postnatal TSH surge settle. A borderline first result is repeated on a sample taken 7 to 10 days after the initial one, and on that second sample the UK positive threshold drops to 8.0 mU/L.

Gestation at birthBabies born under 32 weeks are retested in the UK at 28 days of postnatal age or at discharge home, whichever is sooner, whatever the first result showed — because ‘physiological evidence suggests that 28 days is the postnatal age by which maturation of thyroid function has occurred in most very preterm infants’. In a two-screen programme, 43.7% of the congenital hypothyroidism cases found on the second screen had a first-screen TSH of only 5.0 to 9.9 mU/L, and preterm birth and neonatal unit admission were far commoner in that group.

Borderline — repeat the sampleExample

Blood spot TSH 12 mU/L whole blood, UK NHS pathway, first screening sample on day 5, born at 32 weeks or more

The UK NHS newborn blood spot pathway

All of this is from the UK laboratory guide to newborn blood spot screening for congenital hypothyroidism, last updated 8 October 2024. Note that the positive threshold on the second sample (8.0) is the same figure that was merely borderline on the first, which is the pathway’s way of saying that a raised TSH which has not settled in 7 to 10 days is not transient.
Blood spot TSH (mU/L whole blood)	Action
Below 6.0 (analytical cut-off)	Single measurement; negative
6.0 or above	Retested in duplicate, giving a triplicate mean
Triplicate mean below 8.0 (action cut-off)	Negative screening result
Triplicate mean 8.0 to below 20.0	Repeat dried blood spot 7 to 10 days after the initial sample, assayed in duplicate
Average 8.0 or above on the second sample	Positive screening result
Triplicate mean 20.0 or above	Positive; reported as ‘CHT suspected’
Sample timing	Day 5, counting the day of birth as day 0
Born under 32 weeks (31+6 or less)	Repeat at 28 days of postnatal age or at discharge home, whichever is sooner
After blood transfusion	May be falsely negative; repeat after at least 72 hours

Cut-offs are programme-specific, not universal

A blood spot TSH of 12 mU/L is borderline in the UK, positive on a 10 mU/L programme and negative where the reporting cut-off is above 20. The number on the report means nothing without the programme that produced it, which is why this page asks for it rather than assuming one.
Programme or source	Cut-off	Units
UK NHS newborn blood spot	Analytical 6.0, action 8.0, referral 20.0	mU/L whole blood
UK national recommendation applied in 2011 to 2012	10	mU/L whole blood
‘Most screening programs’ per a screening review	Above 20 to 25	mU/mL dried blood spot
Lombardy two-screen programme	10 first screen, 5 second screen	mU/L whole blood
Some US locations, per the same review	Reduced from 15 to 5	mU/mL
Reported domestic range within China	4.82 to 18.0	µIU/mL
One Chinese programme’s ROC-optimised cut-off	9.5 (sensitivity 100%, specificity 98.1%)	µIU/mL

Why a blood spot TSH is not a serum TSH

The last two rows are this page’s arithmetic on the published regression, not figures from the guideline, and they are shown because they explain why the screening cut-offs look low to anyone used to serum TSH. TSH is confined to the plasma fraction, so a whole-blood measurement is diluted by the red cells; the practical consequence is that the confirmatory serum TSH after a positive screen reads higher than the screen did, and that is expected rather than a deterioration.
Quantity	Figure
Published relationship	blood spot TSH = 0.68 × serum TSH + 0.07
Blood spot reference interval, neonate day 5 to 8	0.40 to 4.10 mU/L
Blood spot reference interval, adults (n = 109)	0.49 to 3.07 mU/L
UK action cut-off of 8.0 mU/L blood spot, expressed as serum	About 11.7 mU/L
UK referral threshold of 20.0 mU/L blood spot, expressed as serum	About 29.3 mU/L
Method behind the regression	PerkinElmer Neonatal hTSH on a Genetic Screening Processor, against serum TSH on the Abbott Architect

What a normal TSH screen does not exclude

The first row is the structural blind spot and the most important line on this page. A screen that measures TSH cannot find a disease whose defining feature is a TSH that is too low. If a baby has hypoglycaemia, prolonged jaundice, midline defects, micropenis or a suggestive family history, measure a serum free T4 regardless of what the blood spot said.
Missed by a TSH-based screen	Why
Central (secondary) congenital hypothyroidism	The programme ‘is designed to detect only primary CHT’; secondary disease due to generalised pituitary failure or isolated TSH deficiency ‘will not be detected’, because the TSH is low, not high
Late-rising TSH in the very preterm baby	Thyroid axis maturation is incomplete; the UK retests everyone born under 32 weeks at 28 days or discharge
Congenital hypothyroidism after transfusion	Transfusion ‘could cause a false negative result’; repeat after at least 72 hours
Cases with a first-screen TSH in the 5 to 10 range	43.7% of the cases detected on the second screen of a two-screen programme sat in that band on the first
Thyroid-binding protein abnormalities and thyroid hormone resistance	Not a TSH-detectable pattern; a free T4 and the clinical picture are needed

A whole-blood number, a programme-specific threshold, and a disease the test cannot see

Three things go wrong with a newborn blood spot TSH, and none of them is the assay. The first is the unit. A blood spot TSH is reported in milliunits per litre of whole blood, and a serum TSH is milliunits per litre of serum. Those are different quantities, because TSH lives in the plasma fraction and a whole-blood measurement is diluted by the red cells. The relationship has been measured: blood spot TSH equals 0.68 times the serum TSH plus 0.07, on a PerkinElmer neonatal kit against an Abbott Architect serum assay. So the UK action cut-off of 8.0 mU/L of whole blood corresponds to a serum TSH somewhere around 11.7, and the referral threshold of 20.0 to something near 29. A clinician who reads a blood spot 8.0 against a serum reference interval topping out at 4 or 5 will think it is mildly raised; a clinician who compares a confirmatory serum TSH with the screening figure will think the baby has deteriorated between the two samples. Neither is true.

The second is that the cut-off is a property of the programme, not of the analyte. The UK uses an analytical cut-off of 6.0, an action cut-off of 8.0 and a referral threshold of 20.0 mU/L of whole blood, and samples on day 5 with the day of birth counted as day 0. A review of newborn screening puts the reporting cut-off in most programmes above 20 to 25. Lombardy screens twice, at 10 and then at 5. Some US locations have taken their cut-off down from 15 to 5. A survey of Chinese programmes found cut-offs ranging from 4.82 to 18.0. A blood spot TSH of 12 is therefore positive, borderline or negative depending on which laboratory produced it, and the trade-off is measurable: against the 10 mU/L standard applied in the UK in 2011 to 2012, moving to 8 gave a relative negative likelihood ratio of 0.11, and moving to 6 gave 0.11 with a relative positive likelihood ratio of 0.52 — better at excluding, worse at confirming.

The third is the one that costs babies. This is a TSH screen, so it finds primary congenital hypothyroidism, in which a failing thyroid drives the TSH up. It is structurally blind to central hypothyroidism, in which the pituitary or hypothalamus fails and the TSH is low or normal alongside a low free T4. The UK laboratory guide says so in terms: secondary congenital hypothyroidism due to generalised pituitary failure or isolated TSH deficiency will not be detected, and the programme is designed to detect only primary disease. A normal screen is therefore not a normal thyroid axis. Hypoglycaemia, prolonged jaundice, micropenis, midline facial defects, nystagmus or a family history of hypopituitarism are indications for a serum free T4 whatever the card said.

Two further blind spots are handled by policy rather than by interpretation. Very preterm babies have an immature axis and a TSH that can rise later, so the UK retests everyone born at under 32 weeks at 28 days of postnatal age or at discharge, whichever comes first, on the stated physiological ground that 28 days is when maturation has occurred in most very preterm infants. The evidence that this is not theoretical is that in a two-screen programme, 43.7% of the cases picked up on the second screen had a first-screen TSH of only 5.0 to 9.9 mU/L, and preterm birth and neonatal unit admission were both far commoner in that group. And a transfused baby’s card may be falsely negative, so the programme asks for a repeat at least 72 hours afterwards. Both are reasons why a negative screen is a statement about one sample rather than about the child.

Frequently asked questions

Is a blood spot TSH the same as a serum TSH?

No. A blood spot TSH is a whole-blood concentration and a serum TSH is a serum concentration, and TSH is confined to the plasma fraction, so the whole-blood figure is the smaller of the two. A published regression gives blood spot TSH = 0.68 × serum TSH + 0.07. The two must not be read against each other’s reference intervals, and a confirmatory serum TSH that is higher than the screening result is expected rather than a deterioration.

What blood spot TSH is a positive newborn screen in the UK?

A triplicate mean of 20.0 mU/L whole blood or above on the initial sample is reported as ‘CHT suspected’. Below 8.0 mU/L is a negative screen. Between 8.0 and 20.0 is borderline and triggers a repeat dried blood spot taken 7 to 10 days after the first, on which an average of 8.0 mU/L or above is itself a positive result. The analytical cut-off, above which samples are retested in duplicate, is 6.0 mU/L.

When is the newborn blood spot taken?

On day 5 in the UK, counting the day of birth as day 0. A review of screening practice gives the ideal window as 3 to 5 days after birth, to let the physiological postnatal TSH surge settle and reduce false positives. Babies born at under 32 weeks are retested at 28 days of postnatal age or at discharge home, whichever is sooner.

Why are preterm babies screened twice?

Because the thyroid axis is immature and the TSH can rise later. The UK programme retests all babies born at under 32 weeks at 28 days or at discharge, on the stated ground that 28 days is the postnatal age by which thyroid function has matured in most very preterm infants. In a two-screen programme, 43.7% of the cases detected on the second screen had a first-screen TSH of only 5.0 to 9.9 mU/L, and preterm birth and neonatal unit admission were much commoner in that low group.

Does a normal newborn screen exclude congenital hypothyroidism?

It does not exclude central or secondary congenital hypothyroidism, because the screen measures TSH and central disease presents with a TSH that is low or normal alongside a low free T4. The UK laboratory guide states that secondary disease due to generalised pituitary failure or isolated TSH deficiency will not be detected and that the programme is designed to detect only primary disease. If the clinical picture suggests hypopituitarism, measure a serum free T4 regardless of the screen.

Why do screening cut-offs differ between countries?

Because the cut-off is a policy choice about the balance between missed cases and false positives. The UK acts at 8.0 and refers at 20.0 mU/L whole blood; most programmes report above 20 to 25; Lombardy screens at 10 and then at 5; some US locations use 5; a survey of Chinese programmes found 4.82 to 18.0. The trade-off has been quantified: against a 10 mU/L standard, moving to 8 improved exclusion (relative negative likelihood ratio 0.11) and moving to 6 worsened confirmation (relative positive likelihood ratio 0.52).

Related calculators

References

UK National Screening Committee / NHS England. A Laboratory Guide to Newborn Blood Spot Screening in the UK for Congenital Hypothyroidism. GOV.UK; last updated 8 October 2024 — the 6.0 analytical, 8.0 action and 20.0 referral cut-offs in mU/L whole blood, the day 5 sample, the 7-to-10-day repeat, the under-32-weeks retest at 28 days, the transfusion caveat, and the statement that only primary CHT is detected.
Morris TG, Layley J, Geevarghese R, Steele L, Ssali J, Mirzazadeh M. Adult and neonate TSH blood spot reference intervals. Ann Clin Biochem. 2022 — the regression blood spot TSH = 0.68 × serum TSH + 0.07, and blood spot reference intervals of 0.40 to 4.10 mU/L for neonates on day 5 to 8 and 0.49 to 3.07 mU/L for adults.
Newborn screening for primary congenital hypothyroidism: estimating test performance at different TSH thresholds. J Clin Endocrinol Metab. 2018;103(10):3720 — 813,087 UK live births; sensitivity 96.76%, specificity 99.97%, PPV 66.88%, and the relative likelihood ratios for 6 and 8 mU/L against the 10 mU/L standard.
Newborn screening for congenital hypothyroidism: the benefit of using differential TSH cutoffs in a 2-screen program. J Clin Endocrinol Metab. 2021;106(1):e338 — 10 mU/L first screen and 5 mU/L second screen in Lombardy, with 43.7% of second-screen cases having a first-screen TSH of 5.0 to 9.9 mU/L.
Newborn screening for congenital hypothyroidism. J Clin Res Pediatr Endocrinol — the reporting cut-off above 20 to 25 mU/mL in most programmes, the 3-to-5-day sampling window, the reduction from 15 to 5 mU/mL in some US locations, and the prognostic difference between a TSH above 50 and one between 20 and 49.
Exploring reference ranges for thyroid-stimulating hormone in neonatal screening tests for preterm infants. Front Endocrinol. 2025;16:1616748 — regional cut-off variation of 4.82 to 18.0 µIU/mL, an ROC-optimised cut-off of 9.5 µIU/mL, and the reported 2024 European Society for Paediatric Endocrinology recommendation of secondary screening at 2 to 4 weeks in preterm and low-birth-weight infants.

Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.