Objective To explore the changes in hypothalamic-pituitary-thyroid（ HPT） axis hormones and inflammatory factor levels in non-suicidal self-injury（ NSSI） adolescents， and investigate the risk relationship with NSSI in adolescents. Methods A total of 62 depressive disorder adolescents who visited the Outpatient Department of Inner Mongolia Mental Health Center from March 2024 to May 2025 were selected as research subjects. Patients were divided into a depression-only group（ n=19） and a depression-with-NSSI group （n=43） based on the presence of NSSI. Levels of HPT axis hormones and inflammatory factors were measured by collecting blood samples from the elbow vein. A binary Logistic regression model was used to construct a risk prediction model for adolescent NSSI based on HPT axis and inflammatory factors， and a receiver operating characteristic（ ROC） curve was plotted. The predictive performance of the diagnostic model was evaluated using the area under the curve（ AUC）， optimal cutoff value， sensitivity and specificity. Results Compared with depression-only group， depression-with-NSSI group exhibited decreased thyroid-stimulating hormone（ TSH） levels［ （1.50±0.79） vs.（ 2.41±1.40）； t=3.267， P=0.002］， and increased neutrophil-to-lymphocyte ratio （NLR） levels［ （2.29±1.29） vs.（ 1.39±0.44）； t=-2.942， P=0.005］， with statistically significant differences. Binary Logistic regression analysis revealed that decreased TSH［ OR=0.415， 95%CI（ 0.212， 0.813）， P=0.010］ and increased NLR［ OR=16.941， 95%CI（ 1.586， 180.996）， P=0.019］ were statistically associated with an increased risk of adolescent NSSI. ROC curve analysis revealed that the AUC for predicting adolescent NSSI risk were 0.690 for TSH and 0.715 for NLR， with a combined AUC of 0.794（ P＜0.05）. Conclusions Both TSH and the inflammatory factor NLR are independent predictors of NSSI in adolescents， and have predictive effects on the risk of NSSI in adolescents.