关键词: 重型颅脑损伤, 应激性溃疡, 格拉斯哥昏迷评分, 列线图预测模型

Abstract: Objective To investigate the influencing factors of stress ulcers(SU) in patients with severe traumatic brain injury(STBI),establish an effective column chart prediction model for predicting SU in STBI patients,and verify it. Methods From December 2021 to December 2024,276 patients with STBI admitted to our hospital were regarded as the research subjects.They were assigned into a modeling set of 193 cases and a validation set of 83 cases randomly according to a 7:3 ratio.The modeling set was assigned into an occurrence group(62 cases) and a non occurrence group(131 cases) based on whether the patients had SU.Logistic regression analysis was applied to screen risk factors for SU in STBI patients.The R software and RMS package were applied to construct a column chart risk prediction model.ROC curve was applied to evaluate the predictive ability of the model.Non parametric repeated sampling Bootstrap method was used for internal validation,and the calibration curve was used to evaluate the calibration ability of the model. Results Compared with the non occurrence group,the GCS score,white blood cell count,absolute value of neutrophil and postoperative systolic blood pressure in the occurrence group were prominently lower(P＜0.05),while the proportions of mechanical ventilation time>48 h and diabetes were prominently higher(P＜0.05).GCS score(OR=0.378),white blood cell count(OR=0.155),absolute value of neutrophil(OR=0.038),postoperative systolic blood pressure(OR=0.813),and diabetes(OR=2.558) were the influencing factors for the occurrence of SU in patients with STBI(P＜0.05).A column chart prediction model for SU in STBI patients was constructed and validated,and the column chart prediction model had good discrimination and consistency. Conclusion GCS score,white blood cell count,absolute value of neutrophil,postoperative systolic blood pressure,and diabetes are the influencing factors for the occurrence of SU in STBI patients.Based on the above factors,a nomogram prediction model is constructed,and the effectiveness of this model in predicting the risk of SU in STBI patients was further validated.

Key words: Severe traumatic brain injury, Stress ulcer, Glasgow Coma Scale, Column chart prediction model