Li Junhong, Huang Jingjing, Pang Tingting, Chen Zikun, Li Jing, Wu Lin, Hu Yuqiang, Chen Wei
Department of Neurology, The First Affiliated Hospital of Guangxi University of Chinese Medicine, Nanning, China.
Guangxi University of Chinese Medicine, Nanning, China.
Front Neurol. 2021 Dec 10;12:710144. doi: 10.3389/fneur.2021.710144. eCollection 2021.
Infections after acute ischemic stroke are common and likely to complicate the clinical course and negatively affect patient outcomes. Despite the development of various risk factors and predictive models for infectious and inflammatory disorders (IAID) after stroke, more objective and easily obtainable predictors remain necessary. This study involves the development and validation of an accessible, accurate nomogram for predicting in-hospital IAID in patients with acute ischemic stroke (AIS). A retrospective cohort of 2,257 patients with AIS confirmed by neurological examination and radiography was assessed. The International Statistical Classification of Diseases and Health related Problem's definition was used for IAID. Data was obtained from two hospitals between January 2016 and March 2020. The incidence of IAID was 19.8 and 20.8% in the derivation and validation cohorts, respectively. Using an absolute shrinkage and selection operator (LASSO) algorithm, four biochemical blood predictors and four clinical indicators were optimized from fifty-five features. Using a multivariable analysis, four predictors, namely age (adjusted odds ratio, 1.05; 95% confidence interval [CI], 1.038-1.062; < 0.001), comatose state (28.033[4.706-536.403], = 0.002), diabetes (0.417[0.27-0.649], < 0.001), and congestive heart failure (CHF) (5.488[2.451-12.912], < 0.001) were found to be risk factors for IAID. Furthermore, neutrophil, monocyte, hemoglobin, and high-sensitivity C-reactive protein were also found to be independently associated with IAID. Consequently, a reliable clinical-lab nomogram was constructed to predict IAID in our study (C-index value = 0.83). The results of the ROC analysis were consistent with the calibration curve analysis. The decision curve demonstrated that the clinical-lab model added more net benefit than either the lab-score or clinical models in differentiating IAID from AIS patients. The clinical-lab nomogram predicted IAID in patients with acute ischemic stroke. As a result, this nomogram can be used for identification of high-risk patients and to further guide clinical decisions.
急性缺血性卒中后的感染很常见，可能会使临床病程复杂化，并对患者预后产生负面影响。尽管已经开发出各种针对卒中后感染性和炎症性疾病（IAID）的危险因素和预测模型，但仍需要更客观且易于获得的预测指标。本研究涉及开发和验证一种可获取、准确的列线图，用于预测急性缺血性卒中（AIS）患者的院内IAID。对2257例经神经学检查和影像学确诊的AIS患者进行回顾性队列研究。IAID采用国际疾病和相关健康问题统计分类的定义。数据取自2016年1月至2020年3月期间的两家医院。在推导队列和验证队列中，IAID的发生率分别为19.8%和20.8%。使用绝对收缩和选择算子（LASSO）算法，从55个特征中优化出4个血液生化预测指标和4个临床指标。通过多变量分析，发现4个预测指标，即年龄（调整比值比，1.05；95%置信区间[CI]，1.038 - 1.062；P < 0.001）、昏迷状态（28.033[4.706 - 536.403]，P = 0.002）、糖尿病（0.417[0.27 - 0.649]，P < 0.001）和充血性心力衰竭（CHF）（5.488[2.451 - 12.912]，P < 0.001）是IAID的危险因素。此外，还发现中性粒细胞、单核细胞、血红蛋白和高敏C反应蛋白也与IAID独立相关。因此，在本研究中构建了一个可靠的临床实验室列线图来预测IAID（C指数值 = 0.83）。ROC分析结果与校准曲线分析一致。决策曲线表明，在区分IAID与AIS患者方面，临床实验室模型比实验室评分或临床模型增加了更多的净效益。该临床实验室列线图可预测急性缺血性卒中患者的IAID。因此，该列线图可用于识别高危患者并进一步指导临床决策。
Zotero 插件
