传统收集数据的多中心回顾性分析。目的:明确创伤性脊髓损伤(TSCI)后低蛋白血症的潜在原因,为预测个体发生低蛋白血症的可能性提供诊断模型。低蛋白血症是脊髓损伤(SCI)的并发症,是老年SCI患者呼吸衰竭的独立危险因素,也是宫颈SCI患者预后的预测因素。很少有基于列线图的研究使用临床指标来预测TSCI后低蛋白血症的可能性。这项多中心回顾性临床分析包括广西医科大学第一附属医院收治的TSCI患者,梧州共仁医院,2016年至2020年,大化瑶族自治县人民医院。以广西医科大学第一附属医院的患者数据作为训练集,其他2家医院的验证集被用作验证集.所有的病史,诊断程序,并记录影像学检查结果.为了预测TSCI患者是否可能发生低蛋白血症,进行了最小绝对收缩和选择算子回归分析以创建列线图.通过使用决策曲线分析对结果进行分析,对模型进行了验证,校正曲线,C指数,和接收器工作特性曲线。排除数据缺失的患者后,这项研究包括534名患者。男性/女性性别,年龄≥60岁,宫颈SCI,肺炎,胸腔积液,尿路感染(UTI),低钠血症,发烧,低血压,气管切开术是低蛋白血症的独立危险因素。使用这些因素构建了简单且易于复制的临床预测列线图。曲线下面积在训练集中为0.728,在验证集中为0.881。列线图的预测能力令人满意。使用男性/女性的危险因素可以预测TSCI后的低白蛋白血症,年龄≥60岁,宫颈SCI,肺炎,胸腔积液,UTI,低钠血症,发烧,低血压,气管造口术.
A multicenter retrospective analysis of conventionally collected data. To identify the potential causes of hypoproteinemia after traumatic spinal cord injury (TSCI) and provide a diagnostic model for predicting an individual likelihood of developing hypoproteinemia. Hypoproteinemia is a complication of spinal cord injury (SCI), an independent risk factor for respiratory failure in elderly patients with SCI, and a predictor of outcomes in patients with cervical SCI. Few nomogram-based studies have used clinical indicators to predict the likelihood of hypoproteinemia following TSCI. This multicenter retrospective clinical analysis included patients with TSCI admitted to the First Affiliated Hospital of Guangxi Medical University, Wuzhou GongRen Hospital, and Dahua Yao Autonomous County People Hospital between 2016 and 2020. The data of patients from the First Affiliated Hospital of Guangxi Medical University were used as the training set, and those from the other 2 hospitals were used as the validation set. All patient histories, diagnostic procedures, and imaging findings were recorded. To predict whether patients with TSCI may develop hypoproteinemia, a least absolute shrinkage and selection operator regression analysis was conducted to create a nomogram. The model was validated by analyzing the consequences using decision curve analysis, calibration curves, the C-index, and receiver operating characteristic curves. After excluding patients with missing data, 534 patients were included in this
study. Male/female sex, age ≥ 60 years, cervical SCI, pneumonia, pleural effusion, urinary tract infection (UTI), hyponatremia, fever, hypotension, and tracheostomy were identified as independent risk factors of hypoalbuminemia. A simple and easy-to-replicate clinical prediction nomogram was constructed using these factors. The area under the curve was 0.728 in the training set and 0.881 in the validation set. The predictive power of the nomogram was satisfactory. Hypoalbuminemia after TSCI may be predicted using the risk factors of male/female sex, age ≥ 60 years, cervical SCI, pneumonia, pleural effusion, UTI, hyponatremia, fever, hypotension, and tracheostomy.