Endocrine Abstracts (2008) 17 P30

Diagnostic accuracy of four modalities: venous thyroid function tests, thyroglobulin, ultrasound and 99mpertechnetate scanning (Quattro mode): in the investigation of newborns referred with TSH elevation

D Neumann1, J Jones1, R Perry1, H Kaur1, S Maroo1, M Attaie1, E Thompson1, S Butler1, G Irwin1, A Watt1, J Pohlenz2 & M Donaldson1


1Royal Hospital for Sick Children, Glasgow, UK; 2Johannes Guttenberg University Medical School, Mainz, Germany.


Introduction: Thyroid imaging in the newborns referred with TSH elevation is helpful in establishing a lifelong diagnosis of congenital hypothyroidism (CH), in defining the cause and hence assisting with genetic counselling and in gauging thyroxine requirement.

Study aims: To compare diagnostic success when using one, two, three or all four (Quattro mode) of the following modalities – pre-treatment venous thyroid function tests (TFTs), quantitative thyroglobulin (Tg), ultrasound scan (USS) and radio-isotope scan (RIS).

Results: Data on 46 newborns (29 F) born between January 2004 and May 2008 and attending our unit for diagnostic imaging were assessed retrospectively, with radiologists blinded as to initial interpretation of scans. Final diagnosis in the 42 children with definite CH was dyshormonogenesis in 10 (24%) including twins with thyroglobulin gene defect causing obstructive goitre, and thyroid dysgenesis in 32 (76%). Causes of dysgenesis included thyroid ectopia (14), with maternal ectopia in 1, hypoplastic gland in situ (4), due to TSH-R defect in 1 and athyreosis (2). Five infants had a form of severe thyroid dysgenesis with measurable Tg in all five, no thyroid tissue on both scans in 2 and remnants of tissue in situ on USS but no uptake on RIS in 3. Dysgenesis was unclassified in 7.

A definitive diagnosis could be made in 100% of cases when all components of Quattro mode were performed. The diagnostic ‘success’ rate using other combinations was: TFTs alone 0%; TFTs and Tg 2.9%; TFTs and USS 40%; TFTs and RIS 75%; TFTs and dual scanning 86%; TFTs, USS and Tg 43%; TFTs, RIS and Tg 75%. Incorrect causes suggested by these combinations were: TFTs alone 17%; TFTs and Tg 3.8%; TFTs and USS 37%; TFTs and RIS 11%; TFTs and dual scanning 0%; TFTs, Tg and USS 0%; TFTs, Tg and RIS 11%.

Conclusion: Our data show that USS, whilst a valuable diagnostic modality, should be used in conjunction with Tg and, ideally, RIS. The use of Quattro mode has led us to reclassify thyroid dysgenesis, adding two novel descriptions and rendering true athyreosis with undetectable Tg rare. Quattro mode also allows accurate targeting of subsequent mutation analysis. An updated and internationally accepted classification of the causes of CH is now required.