Initial versus confirmatory thyroid stimulating hormone (TSH) levels: is there a correlation?

ABSTRACT

Newborn screening for congenital hypothyroidism (CH) in the Philippines began in 1996. The screening method used is the fluoroimmunometric assay of thyroid stimulating hormone (TSH) from dried blood spot. In the past five years (June 1996--Sept 2001), 176,548 newborns have been screened. Of these, 237 had elevated TSH levels and 51 (22%) were confirmed to have CH. One hundred forty-six (61%) had normal TSH levels on confirmatory testing; five (2%) expired; 25 (11%) were lost to follow-up, while 10 (4%) were being recalled at the time of this study. Thirty-three out of 51(65%) CH patients are female. Only 38 of 51 patient charts were available for data analysis. Thirteen of 51 CH patients were lost to follow-up after confirmation of the disorder. The mean age at which levo-thyroxine was initiated is 1 1/2 months at a modal dosage of 25 microg OD. The initial TSH levels as determined by the Philippine Newborn Screening Laboratory directly correlates with the confirmatory TSH levels done in other endocrine laboratories (Spearman's rho=0.57, p value=0.0002, at a=0.05). However, the time of heel prick on the newborn was independent of the TSH levels, (Spearman's rho=-0.16, p value=0.377 at a=0.05) hence there was no significant difference with respect to the initial TSH level of blood samples taken at 48 hours, less than one week, one to two weeks; or even more than two weeks after birth (Kruskall Wallis test, p value=0.064 at a=0.05). Using Fisher's exact test, there is no sufficient evidence to say that there is an association between gender and the incidence of CH among screened newborns whose TSH levels were initially elevated (p 2-tailed=0.183, p 1-tailed=0.113 at a=0.05).