Abstract:Objective: To explore the value of serum YTH structural domain N6-methyladenosine RNA binding protein 2 (YTHDF2), uncoupling protein 2 (UCP2), and interleukin-1 receptor-associated kinase 4 (IRAK4) in evaluating neurological dysfunction in patients with severe craniocerebral injury. Methods: From March 2023 to December 2024, 200 patients with severe craniocerebral injury admitted to our hospital were considered as the observation group. Complying with the neurological function recovery at 2 months after treatment, they were separated into a good recovery group and a poor recovery group. ELISA method was used to detect serum YTHDF2, UCP2, and IRAK4. Multivariate logistic regression was used to analyze the influencing factors of poor neurological function recovery in patients with severe craniocerebral injury. ROC curve was used to analyze the value of serum YTHDF2, UCP2, and IRAK4 alone and their combination for evaluating poor neurological function recovery in patients with severe craniocerebral injury. AUC comparison was conducted using DeLong test. Bootstrap repeated sampling 100 times was used for internal validation. Restrictive cubic spline plot was used to evaluate the effect dose relationship between serum YTHDF2, UCP2, IRAK4 levels and poor neurological function recovery in patients with severe traumatic brain injury. Results: The poor recovery group had higher serum YTHDF2 and IRAK4, and larger proportions of abnormal admission pupil reflex, preoperative GCS score of 3-5 points, and cerebral herniation, and lower UCP2 than the good recovery group (P<0.05). YTHDF2 and IRAK4 were independent risk factors for poor neurological function recovery in patients with severe craniocerebral injury, while UCP2 was a protective factor for poor neurological function recovery in patients with severe craniocerebral injury (P<0.05). The AUC values of serum YTHDF2, UCP2, and IRAK4 in evaluating poor neurological function recovery in patients with severe craniocerebral injury were 0.828, 0.807, and 0.820, respectively. The AUC of combined evaluation was 0.921, with the AUC of combined evaluation being higher than that of individual indicator (Z=2.204, 2.807, 2.304, P<0.05). The internal validation results of Bootstrap found that the calibration curve was basically consistent with the ideal curve, and the Hosmer-Lemeshow test showed χ2=7.124, P=0.918. There was a significant non-linear relationship between serum YTHDF2, UCP2, IRAK4 and poor neurological function recovery in patients with severe traumatic brain injury (P<0.05). Conclusion: High expression of YTHDF2 and IRAK4, and low expression of UCP2 before treatment may be influencing factors for poor neurological function recovery in patients after treatment. The combination of the three can effectively evaluate poor neurological function recovery in patients with severe craniocerebral injury.