Objectives: Sepsis associated encephalopathy (SAE) is a common neurological complication of sepsis. Delirium is a common symtom of SAE. The pathophysiology of SAE is still unclear, but several likely mechanisms have been proposed, such as mitochondrial and endothelial dysfunction, neurotransmission disturbances, derangements of calcium homeostasis, cerebral microcirculation dysfunction, and brain hypoperfusion. Near-infrared spectroscopy (NIRS) is a non-invasive measure for regional cerebral oxygen saturation (rSO2), which has attracted more attention these years. Previous studies have reported that abnormal NIRS values were associated with delirium in critically ill patients. Blood pressure management according to NIRS monitoring improved the organ perfusion and prognosis of patients. This study aimed to observe the dynamic changes of rSO2 using NIRS in septic shock patients, and analyze the relationship between them.
Methods: A total of 48 septic patients who admitted to the intensive care unit (ICU) of Xiangya Hospital, Central South University from August 2017 to May 2018, were retrospectively study. Septic shock was diagnosed according to the criteria of sepsis 3.0 defined by the American Association of Critical Care Medicine and the European Society of Critical Care Medicine. NIRS monitoring was performed during the first 6 hours admitted to ICU with sensors placed on the bilateral forehead of patients. The maximum (rSO2max), minimum (rSO2min), mean value, and the variation rate during the first 6 hours of monitor were recorded. The following data were collected upon the first 24 h after admission to the ICU: The baseline data of patients, laboratory examination results (routine blood test, liver and renal function, blood gas analysis, indicators of infection, and coagulation function), scoring system results [Glasgow Coma Scale (GCS), Acute Physiology and Chronic Health Evaluation II (APACHE II) and Sequential Organ Failure Assessment (SOFA)]. Delirium was screened with the Confusion Assessment Method for ICU (CAM-ICU). The length of time on mechanical ventilation (MV), length of ICU-stay, length of hospital-stay, and 28-day mortality were also recorded. The primary outcome was 28-day mortality, and the secondary outcomes were the incidence of delirium, length of ICU-stay, and length of hospital-stay. The differences between survivors and non-survivors, and patients with or without delirium were analyzed, and the risk factors for delirium were assessed. The performance of rSO2-related indexes (rSO2max, rSO2min, the mean value, and the variation rate of rSO2) in predicting 28-day mortality and delirium was analyzed and the cutoff values were determined.
Results: The overall 28-day mortality of septic shock patients was 47.92% (23/48), and the incidence of delirium was 18.75% (9/48). The rSO2min was significantly lower in the non-survivors than the survivors (P=0.042). The variation rate of rSO2 was higher in patients with delirium than those without delirium (P=0.006). The independent risk factors for delirium were rSO2max, the level of direct bilirubin (DBIL), and whether achieved the 6-hour bundle. To predict the 28-day mortality of septic shock patients, the area under the receiver operating characteristic curve (AUROC) for rSO2max, rSO2min, the mean value and the variation rate of rSO2 were 0.616, 0.606, 0.623, and 0.504, respectively. To predict the incidence of delirium, AUROC for rSO2max, rSO2min, the mean value and the variation rate of rSO2 were 0.682, 0.617, 0.580, and 0.501, respectively. The best cutoff value for rSO2max in predicting delirium was 77.5% (sensitivity was 0.444, specificity was 0.897). The best cutoff value for rSO2min in predicting delirium was 65.5% (sensitivity was 0.556, specificity was 0.744).
Conclusions: Cerebral anoxia and hyperoxia, as well as the large fluctuation of cerebral oxygen saturation are important factors that affect the outcomes and the incidence of delirium in septic shock patients, which should be paid attention to in clinical practice. Dynamic monitoring of cerebral oxygen saturation and maintain its stability may be of great significance in patients with septic shock.
目的: 脓毒症相关性脑病(sepsis associated encephalopathy，SAE)是脓毒症的一种常见神经并发症，谵妄是SAE的一种常见表现形式。SAE的病理生理学机制尚不明确，现有研究发现线粒体和内皮细胞功能障碍、神经递质异常、钙失调、脑微循环障碍和脑低灌注等可能参与其中。近红外光谱法(near-infrared spectroscopy，NIRS)是一种无创监测区域脑氧饱和度(regional cerebral oxygen saturation，rSO2)的方法，也是近年来研究的前沿热点。研究显示重症患者NIRS监测rSO2异常与谵妄相关，且基于NIRS监测的治疗有助于改善患者的器官灌注和预后。本研究旨在通过NIRS监测脓毒症休克患者rSO2的动态变化，并分析其与预后之间的关系。方法: 收集2017年8月到2018年5月中南大学湘雅医院重症医学科收治的48例脓毒症休克患者资料，诊断依据欧美危重病医学会指定国际专家制订的脓毒症3.0标准。患者入重症监护室(intensive care unit，ICU)后在双侧前额放置NIRS监测探头对rSO2进行连续动态监测至少6 h，记录rSO2最高值(rSO2max)和最低值(rSO2min)，计算rSO2平均值和rSO2变异度。收集患者的一般资料、入ICU 24 h内的实验室检查结果(血常规、肝肾功能、血气分析、感染指标、凝血功能)、评分结果[格拉斯哥昏迷量表(Glasgow Coma Scale，GCS)评分、急性生理学与慢性健康状况评价II(Acute Physiology and Chronic Health Evaluation II，APACHE II)评分、序贯器官衰竭评价(Sequential Organ Failure Assessment，SOFA)评分等]。谵妄的评估采用ICU意识模糊评估法(Confusion Assessment Method for ICU，CAM-ICU)量表。记录患者机械通气(mechanical ventilation，MV)时间、ICU时间、住院时间和28 d病死率。首要结局指标为28 d病死率，次要结局指标为谵妄发生率、ICU时间和住院时间。分析存活与死亡患者、有谵妄与无谵妄患者之间实验室和临床指标的差异，评估谵妄的危险因素，评价rSO2相关指标(rSO2max、rSO2min、rSO2平均值和rSO2变异度)用于预测28 d病死率和谵妄的能力及其界值。结果: 脓毒症休克患者总的28 d病死率为47.92%(23/48)，谵妄发生率为18.75%(9/48)。死亡患者的rSO2min显著低于存活患者(P=0.042)。谵妄患者的rSO2变异度显著高于无谵妄患者(P=0.006)。谵妄的独立危险因素是rSO2max、直接胆红素(direct bilirubin，DBIL)水平和6 h集束化治疗是否达标。采用rSO2相关指标预测脓毒症休克患者28 d是否死亡，rSO2max、rSO2min、rSO2平均值和rSO2变异度的受试者工作特征曲线下面积(area under the receiver operating characteristic curve，AUROC)分别为0.616、0.606、0.623和0.504；预测脓毒症休克患者是否发生谵妄，rSO2max、rSO2min、rSO2平均值和rSO2变异度的AUROC分别为0.682、0.617、0.580和0.501。rSO2max预测谵妄的最佳界值是77.5%(敏感度为0.444，特异度为0.897)，rSO2min预测谵妄的最佳界值是65.5%(敏感度为0.556，特异度为0.744)。结论: rSO2过低、过高或波动过大均与脓毒症休克不良预后和谵妄发生相关，需要引起临床重视。动态监测脓毒症休克患者的rSO2并维持其平稳，可能对该类患者具有重要临床意义。.
Keywords: delirium; mortality; near-infrared spectroscopy; regional cerebral oxygen saturation; septic shock.