Reference intervals for thyroid stimulating hormone and free thyroxine derived from neonates undergoing routine screening for congenital hypothyroidism at a university teaching hospital in Nairobi, Kenya: a cross sectional study

BACKGROUND In order to accurately interpret neonatal thyroid function tests (TFTs), it is necessary to have population specific reference intervals (RIs) as there is significant variation across different populations possibly due to genetic, environmental or analytical issues. Despite the importance of RIs, globally there are very few publications on RIs for neonatal TFTs primarily due to ethical and technical issues surrounding recruitment of neonates for a prospective study. To the best of our knowledge, this is the first report from Africa on neonatal RIs for TFTs. METHODS We used hospital based data largely derived from neonates attending the wellness clinic at the Aga Khan University Hospital Nairobi (AKUHN) where screening for congenital hypothyroidism is routinely done. Specifically we derived age and gender stratified RIs for free thyroxine (fT4) and thyroid stimulating hormone (TSH) which had been analyzed on a Roche e601 analyzer from 2011 to 2013. Determination of reference intervals was done using a non-parametric method. RESULTS A total of 1639 and 1329 non duplicate TSH and fT4 values respectively were used to derive RIs. There was a decline in TSH and fT4 levels with increase in age. Compared to the Roche RIs, the derived RIs for TSH in neonates aged 0-6 days and those aged 7-30 days had lower upper limits and narrower RIs. The fT4 lower limits for neonates less than 7 days and those aged 7-30 days were higher than those proposed by Roche. There was a significant difference in TSH RIs between male and female neonates aged less than 15 days. No gender differences were seen for all other age stratifications for both TSH and fT4. Appropriate age and gender specific RIs were subsequently determined. CONCLUSION The AKUHN derived RIs for fT4 and TSH revealed similar age related trends to what has been published. However, the differences seen in upper and lower limits across different age stratifications when compared to the Roche RIs highlight the need for population specific RIs for TFTs especially when setting up a screening programme for congenital hypothyroidism. We subsequently recommend the adoption of the derived RIs by the AKUHN laboratory and hope that the RIs obtained can serve as a reference for the African population.