| 夏加庚,樊友武,李启贤,辛 恒,钱正庭.基于列线图预测模型的重症颅脑损伤继发应激性溃疡的影响因素分析[J].,2024,(12):2361-2366 |
| 基于列线图预测模型的重症颅脑损伤继发应激性溃疡的影响因素分析 |
| Analysis of Influencing Factors of Stress Ulcer Secondary to Severe Craniocerebral Injury Based on Nomogram Prediction Model |
| 投稿时间：2024-01-21 修订日期：2024-02-18 |
| DOI：10.13241/j.cnki.pmb.2024.12.032 |
| 中文关键词: 重症颅脑损伤 应激性溃疡 影响因素 预测价值 |
| 英文关键词: Severe craniocerebral injury Stress ulcer Influencing factors Predictive value |
| 基金项目:江苏省中医药科技发展计划项目（YB2020036） |
|
| 摘要点击次数: 271 |
| 全文下载次数: 173 |
| 中文摘要: |
| 摘要 目的：探讨重症颅脑损伤（sTBI）继发应激性溃疡影响因素，并构建列线图预测模型。方法：选取2022年5月至2023年5月我院诊治的sTBI患者116例为研究对象，根据是否继发应激性溃疡分为发生组（n=35）和未发生组（n=81）。收集两组患者的临床资料，采用单因素和多因素Logistic回归模型分析sTBI患者继发应激性溃疡的影响因素，构建列线图预测模型，采用受试者工作特征（ROC）曲线分析对sTBI患者继发应激性溃疡的预测价值。结果：sTBI继发应激性溃疡与入院时格拉斯哥昏迷评分法（GCS）评分、随机血糖水平、血红蛋白（Hb）、受伤至入院时间、胃泌素（GAS）等因素有关（P<0.05）。多因素Logistic回归模型分析结果显示，随机血糖水平升高、GAS水平升高为sTBI患者继发应激性溃疡的独立危险因素（P<0.05）；入院时GCS评分6~8分、Hb水平升高则为独立保护因素（P<0.05）。ROC分析结果显示，入院GCS评分、随机血糖水平、Hb、GAS及列线图预测模型预测sTBI患者继发应激性溃疡的曲线下面积（AUC）分别为0.773、0.745、0.781、0.695、0.960，且列线图预测模型的AUC均高于各单项指标单独检测，提示列线图预测模型对sTBI患者继发应激性溃疡的预测价值更高。Bootstrap法（B=1000）内部验证显示，修正偏差后的预测曲线与理想线基本重合，一致性指数（C-index）为0.921，表明该模型具有较好的预测能力。决策分析显示，其阈值概率范围为0.03~0.93，且表现出净收益率大于0的特点，超过了两条无效线。结论：sTBI继发应激性溃疡与入院GCS评分、随机血糖水平、Hb、GAS密切相关，通过上述指标构建的列线图预测模型，可以提升预测sTBI患者继发应激性溃疡的发生风险，有助于临床早期干预和治疗。 |
| 英文摘要: |
| ABSTRACT Objective: To explore the influencing factors of stress ulcer secondary to severe craniocerebral injury (sTBI), and to construct a nomogram prediction model. Methods: 116 sTBI patients diagnosed and treated in our hospital from May 2022 to May 2023 were selected as the research objects, patients were divided into occurrence group (n=35) and non-occurrence group (n=81) according to whether secondary stress ulcer occurred. The clinical data in two groups were collected, the influencing factors of secondary stress ulcer in sTBI patients were analyzed univariate and multivariate Logistic regression models, and a nomogram prediction model was constructed, the predictive value of secondary stress ulcer in sTBI patients was analyzed by receiver operating characteristic (ROC) curve. Results: Stress ulcer secondary to sTBI was related to admission Glasgow Coma Scale Method (GCS) score, random blood glucose level, hemoglobin (Hb), time from injury to admission, gastrin (GAS) and other factors (P<0.05). Multivariate Logistic regression model analysis showed that, elevated random blood glucose level and elevated GAS level were independent risk factors for secondary stress ulcer in sTBI patients (P<0.05). Admission GCS score of 6~8 points and elevated Hb level were independent protective factors (P<0.05). The results of ROC analysis showed that, the area under the curve (AUC) of admission GCS score, random blood glucose level, Hb, GAS and nomogram prediction model for predicting secondary stress ulcer in sTBI patients were 0.773, 0.745, 0.781, 0.695 and 0.960 respectively, and the AUC of nomogram prediction model was higher than that of each single index, suggesting that the nomogram prediction model had higher predictive value for secondary stress ulcer in sTBI patients. The internal verification of Bootstrap method (B=1000) shows that the prediction curve after correcting the deviation basically coincides with the ideal line, and the concordance index (C-index) was 0.921, indicating that the model has good prediction ability. The decision analysis shows that the threshold probability range was 0.03~0.93, and the net return rate was greater than 0, which exceeds two invalid lines. Conclusion: The secondary stress ulcer of sTBI is closely relate to admission GCS score, random blood glucose level, Hb and GAS, the nomogram prediction model construct by the above indices, which can increase the risk of predicting secondary stress ulcer in sTBI patients, and is helpful for early clinical intervention and treatment. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
