Neurological Complications Acquired During Pediatric Critical Illness: Exploratory "Mixed Graphical Modeling" Analysis Using Serum Biomarker Levels.
Neurological Complications Acquired During Pediatric Critical Illness: Exploratory "Mixed Graphical Modeling" Analysis Using Serum Biomarker Levels.
复制标题
DOI：
10.1097/pcc.0000000000002776
复制
发表时间：
2021-10-01
期刊：
影响因子：
--
通讯作者：
Au AK
中科院分区：
文献类型：
--
作者：
Raghu VK;Horvat CM;Kochanek PM;Fink EL;Clark RSB;Benos PV;Au AK
神经并发症，包括在入院时不存在的神经疾病的急性发展，但在疾病过程中发展，由于镇静、神经肌肉阻断和年龄较小，在儿科重症监护病房(PICU)可能很难检测到。我们评估了血清生物标记物和临床变量与神经并发症发生的直接关系。分析是使用混合图形模型进行的，这是一种机器学习方法，允许从连续和离散数据推断因果关系。对先前一项前瞻性观察研究的二次分析。儿科ICU，单一的四级护理中心。住进PICU的人，年龄18岁，通过留置导管进入血管内。没有。这项分析包括101名患者。用酶联免疫吸附试验检测血清中胶质纤维酸性蛋白(GFAP)、泛素C末端水解酶-L1(UCH-L1)和α-II血影蛋白分解产物150(SBDP150)的含量。血清神经元特异性烯醇化酶(NSE)、髓鞘碱性蛋白(MBP)和S100B在这些模型中的应用已有报道。记录人口学数据、所选临床疗法的使用、住院时间和辅助神经学测试(头部CT、脑MRI、EEG)结果。将MGM-FCI-MAX算法应用于数据集。101例患者中有13例在危重期间出现神经系统并发症。混合图形模型确定神经生物标记物NSE和UCH-L1的峰值水平，以及星形胶质细胞生物标记物GFAP是神经并发症发生的直接原因决定因素；相反，包括年龄、性别、住院时间和初级神经学诊断在内的临床变量不是直接原因决定因素。除了临床数据外，还包括生物标记物的图形模型是评估危重儿童神经系统并发症发生的直接关系的有希望的方法。未来的工作需要进一步验证和完善这些模型，以确定它们是否可以用于预测哪些患者有早期神经并发症的风险。
Neurological complications, consisting of the acute development of a neurological disorder that is not present on admission, but develops during the course of illness, can be difficult to detect in the Pediatric Intensive Care Unit (PICU) due to sedation, neuromuscular blockade, and young age. We evaluated the direct relationships of serum biomarkers and clinical variables to the development of neurological complications. Analysis was performed using mixed graphical models, a machine learning approach that allows inference of cause-effect associations from continuous and discrete data. Secondary analysis of a previous prospective observational study. Pediatric ICU, single quaternary-care center. Individuals admitted to the PICU, <18 years of age, with intravascular access via an indwelling catheter. None. 101 patients were included in this analysis. Serum (days 1-7) was analyzed for glial fibrillary acidic protein (GFAP), ubiquitin C-terminal hydrolase-L1 (UCH-L1) and alpha-II spectrin breakdown product 150 (SBDP150) utilizing enzyme-linked immunosorbent assays. Serum levels of neuron-specific enolase (NSE), myelin basic protein (MBP) and S100B used in these models were reported previously. Demographic data, use of selected clinical therapies, lengths of stay, and ancillary neurological testing (head CT, brain MRI, EEG) results were recorded. The MGM-FCI-MAX algorithm was applied to the dataset. 13/101 patients developed a neurological complication during their critical illness. The mixed graphical model identified peak levels of the neuronal biomarkers NSE and UCH-L1, and the astrocyte biomarker GFAP to be the direct causal determinants for the development of a neurological complication; in contrast, clinical variables including age, sex, length of stay, and primary neurological diagnosis were not direct causal determinants. Graphical models that include biomarkers in addition to clinical data are promising methods to evaluate direct relationships in the development of neurological complications in critically ill children. Future work is required to further validate and refine these models, to determine if they can be used to predict which patients are at risk for/or with early neurological complications.