ESPE Abstracts

Background: The polyethylene glycol (PEG) method, which utilizes the addition of PEG to precipitate immunoglobulin fractions in order to measure free thyroid stimulating hormone (TSH), has been implemented as a tool for investigating incongruities in TSH measurement.

Aim of the study: To investigate the practical application of PEG-TSH testing in pediatric scenarios with a discrepancy between elevated TSH and normal free thyroxine (FT4) levels.

Methods: The hospital’s electronic laboratory database was queried for TSH tests performed between January 1st, 2015 and March 24th, 2023. Of those, PEG-TSH were identified and data of patients who underwent this assay was retrieved from medical records. Clinical and biochemical characteristics including PEG-TSH guided management were extracted.

Results: During the study period, 2949 TSH tests were performed in 891 children and adolescents for various indications. Of those, 61 (2.1%) were PEG-TSH results, performed in 38 patients (4.3%): 16 patients (42.1%) with congenital hypothyroidism (50% sublingual thyroid, 47% thyroid agenesis and 3% intact gland), 16 (42.1%) with subclinical hypothyroidism and 6 (15.8%) with Hashimoto thyroiditis. Patients were 1.3 months to 19 years old at testing with mean age of 7.1±5.3 years, older age characterized patients with Hashimoto thyroiditis (p < 0.001). Sex- and age- adjusted anthropometric parameters of patients were within the norms and did not differ between the groups: median height/length z-score of 0.34 [IQR -0.48, 0.91], and average BMI/weight-to-length ratio z-score of 0.18±0.85. All patients with congenital hypothyroidism were treated with L-thyroxine (3.41±0.96 mcg/kg/dy), while only 50% with Hashimoto (1.43±0.66 mcg/kg/dy) and no individuals with subclinical hypothyroidism were treated. Both TSH and PEG-TSH levels of patients with congenital hypothyroidism were higher than in the two other groups (P= 0.021 and P= 0.009, respectively), with no differences in FT4 levels between the groups. Spearman’s correlation analysis revealed a strong association between TSH levels and PEG-TSH levels (r= 0.871, p < 0.001). Nearly 50% of PEG-TSH results guided the clinical decision of decreasing dose or not initiating L-thyroxine treatment.

Conclusion: Our study describes the real-life utilization of PEG-TSH assay in the spectrum of pediatric thyroid dysfunction. While use of PEG-TSH was infrequent, our findings support its value in avoiding unnecessary thyroid hormone treatment in equivocal cases. If the PEG-TSH level falls within the normal range, it may be suitable to manage the clinical condition based on the FT4 levels and age-appropriate clinical parameters.