Mild TSH elevation is independently associated with height in children

Abstract

Abstract Objectives To investigate whether mildly elevated thyroid stimulating hormone (TSH) (<10 mIU/L) is associated with height in children independently of the growth hormone–insulin-like growth factor 1 (GH–IGF-1) axis, and to assess sex differences. Methods This retrospective cross-sectional study included 2,447 children aged 2–10 years presenting for slow growth evaluation (2022–2024). Multivariable linear regression was used with adjustment for age, sex, and IGF-1 SDS; restricted cubic splines (RCS) were used to assess the dose–response relationship. Sex-stratified and sensitivity analyses were performed. Results Mildly elevated TSH was present in 96 (3.92 %) children. Overall, TSH was inversely associated with height SDS (β=−0.03, p<0.001) independently of IGF-1; RCS confirmed linearity. The association was significant in boys (β=−0.03, p=0.002) and borderline in girls. Boys with elevated TSH had lower height SDS (β=−0.18, p=0.025). Sensitivity analyses: the association persisted after excluding low IGF-1 (β=−0.02, p=0.004) but not in children with short stature (height SDS ≤−2). Conclusions Mildly elevated TSH is independently associated with reduced height in children aged 2–10 years, independent of IGF-1. The association was significant in boys and borderline in girls, with no significant interaction. The absence of bone age and parental height data is a major limitation. These findings support growth monitoring rather than routine levothyroxine therapy.