Characteristics of non-suicidal self-injury and its associations with thyroid and sex hormones in adolescents with major depressive disorder and bipolar disorder

BACKGROUND: Non-suicidal self-injury (NSSI) is prevalent among adolescents with mood disorders and represents an important clinical risk factor for future suicidal behaviors. Although thyroid and sex hormones have been implicated in emotional and behavioral dysregulation, whether their associations with NSSI differ between major depressive disorder (MDD) and bipolar disorder (BD) remains unclear. This study examined NSSI characteristics and their associations with thyroid and sex hormones in adolescents with MDD and BD. METHODS: This retrospective study included 471 adolescents diagnosed with MDD or BD. Demographic characteristics and hormone levels were compared within each diagnostic group. Sex-stratified binary logistic regression analyses were conducted to identify factors associated with NSSI. Additionally, adolescents with NSSI were compared between the MDD and BD groups. RESULTS: The prevalence of NSSI was 58.3% in adolescents with BD and 66.0% in those with MDD. Females showed significantly higher rates of NSSI than males in BD (p < 0.001) and MDD (p < 0.001) groups. Among adolescents with BD, those with NSSI had an earlier age at onset (p = 0.007) and lower testosterone levels (p < 0.001) than those without NSSI. Sex-stratified logistic regression showed that lower thyroid-stimulating hormone (TSH) levels (OR = 0.789, 95% CI:0.633-0.983) and earlier age at onset (OR = 0.801, 95% CI: 0.648-0.990) were significantly associated with NSSI in female patients, whereas higher testosterone levels were associated with NSSI in male patients (OR = 1.226, 95% CI: 1.012-1.484). In adolescents with MDD, those with NSSI showed lower testosterone (p = 0.040) and TSH levels (p = 0.040) and an earlier age at onset (p < 0.001) than those without NSSI; In sex-stratified logistic regression analyses, earlier age at onset remained significantly associated with NSSI in both male (OR = 0.577, 95% CI: 0.356-0.935) and female patients (OR = 0.744, 95% CI: 0.623-0.888), whereas none of the hormone indices were independently associated with NSSI in the sex-stratified analyses (p > 0.05). Further comparisons showed that, among adolescents with NSSI, there were no demographic or hormonal differences between the BD and MDD groups (p > 0.05). CONCLUSIONS: Testosterone and TSH showed sex-specific associations with NSSI in adolescents with BD, whereas earlier age at onset was more consistently associated with NSSI in adolescents with MDD. These findings highlight the importance of considering both diagnostic subtype and sex when assessing NSSI risk in adolescents. However, given the limitations of the present study, further validation is required.