**Background:** Thyroid hormone (TH) plays a critical role in skeletal homeostasis by regulating chondrocytes, osteoblasts, and osteoclasts. Both hyperthyroidism and hypothyroidism can lead to bone mineral density (BMD) loss through dysregulation of endochondral bone formation and osteoclastogenesis. However, the influence of abnormal TH and TSH levels on BMD in children and adolescents with chronic conditions remains controversial. Pediatric reference intervals for TH and TSH differ from adult ranges and vary with age, making accurate identification of abnormal thyroid function essential. This study aimed to examine the association between thyroid function and BMD in children with various underlying conditions and to identify which thyroid parameter most sensitively reflects low bone mass.
**Methods:** The authors retrospectively reviewed 619 children and adolescents aged 10–18 years previously diagnosed with hemato-oncologic (79.6%), rheumatoid (10.3%), gastrointestinal (3.7%), endocrinologic (3.1%), and miscellaneous (3.2%) conditions at a single tertiary care center, all at least 1 year post-treatment completion. Serum T3, FT4, and TSH were measured via direct chemiluminescent immunoassay on the same morning as BMD assessment. BMD of the lumbar spine (LSBMD), right femur (RFBMD), and left femur (LFBMD) was measured using dual-energy X-ray absorptiometry (DXA). Thyroid parameters were subcategorized into normal/abnormal and low/normal/high based on age-matched reference ranges (T3: 0.83–2.13 ng/mL; FT4: 0.8–2 ng/dL; TSH: 0.6–8 µU/mL for ages 10–15 years, 0.6–6 µU/mL for ages 16–18 years). Statistical analyses included Mann–Whitney U tests, one-way ANOVA, Pearson correlation, univariate and multivariate linear regression, and multiple logistic regression adjusted for age, sex, BMI, and underlying conditions.
**Key Results:** The mean age was 13.22 ± 3.23 years, mean BMI z-score was 0.03 ± 1.52, and mean LSBMD z-score was 0.49 ± 1.28. Approximately 37.7% of patients had an LSBMD z-score < -1.0. Mean T3, FT4, and TSH levels were 1.25 ± 0.29 ng/mL, 1.28 ± 0.19 ng/dL, and 2.76 ± 1.87 µU/mL, respectively. Both lumbar (p = 0.018) and femoral (left p = 0.036, right p = 0.007) BMD z-scores were significantly lower in the abnormal TSH group compared with the normal TSH group. No significant BMD differences were observed between abnormal and normal T3 or FT4 groups. In univariate regression, height (β = 0.42, p < 0.001), weight (β = 0.41, p < 0.001), BMI (β = 0.31, p < 0.001), RFBMD (β = 0.73, p < 0.001), and LFBMD (β = 0.74, p < 0.001) z-scores were significantly associated with LSBMD z-scores, while T3, FT4, and TSH levels were not. In multivariate regression, only BMI z-score (β = 0.32, p < 0.001) remained independently associated. However, in multiple logistic regression, the adjusted odds ratios (ORs) for abnormal TSH were significant across all LSBMD z-score thresholds and increased with severity: OR = 2.25 (95% CI: 1.24–4.08, p = 0.008) for LSBMD z-score < -1.0; OR = 2.65 (95% CI: 1.29–5.46, p = 0.008) for < -2.0; and OR = 6.21 (95% CI: 1.8–21.41, p = 0.004) for < -3.0. Abnormal T3 and FT4 levels were not significantly associated with decreased LSBMD z-scores after adjustment.
**Clinical Implications:** This study provides evidence that abnormal TSH levels—whether elevated or decreased—are more strongly associated with low BMD in children and adolescents with chronic conditions than T3 or FT4 levels. The finding that the odds of low BMD increase with the severity of bone mass deficit (six-fold for severely low BMD) suggests TSH may serve as a sensitive early indicator of secondary osteoporosis risk. The authors recommend that pediatric healthcare providers monitor thyroid function alongside skeletal assessment when TSH falls outside age-specific reference ranges, even in the absence of overt thyroid disease. Limitations include the cross-sectional design, uneven disease category distribution, small proportion of abnormal thyroid function (9.5% for TSH, 7.4% for T3, 0.5% for FT4), and lack of total body less head BMD measurements. Larger prospective studies are needed to further clarify the role of TSH in pediatric bone fragility.