Pediatric Critical Care and Pediatric Anesthesiology, BC Children's Hospital, Vancouver, BC, Canada.
Departments of Pediatrics and Anesthesiology, Pharmacology & Therapeutics, University of British Columbia, Vancouver, BC, Canada.
Pediatr Crit Care Med. 2018 Aug;19(8):e394-e402. doi: 10.1097/PCC.0000000000001598.
We evaluated adapting the quick Sequential (Sepsis-Related) Organ Failure Assessment score (fast respiratory rate, altered mental status, low blood pressure) for pediatric use by selecting thresholds from three commonly used definitions: Pediatric Logistic Organ Dysfunction 2, Pediatric Advanced Life Support, and International Pediatric Sepsis Consensus Conference. We examined their respective performance in identifying children who had a discharge diagnosis of infection at high risk of mortality using PICU registry data, with additional focus on the influence of age on performance.
Analysis of retrospective data obtained from the Virtual Pediatric Systems PICU database. The performance in predicting observed mortality was assessed for the three candidate approaches using receiver operating characteristics analysis, including age group effects.
The Virtual Pediatric Systems database contains data on diagnosis, clinical markers, and outcomes in prospectively collected clinical records from 130 participating PICUs in the United States and Canada.
Children who had a discharge diagnosis of infection in a participating PICU between 2009 and 2014, for which all required data were available.
None.
Data from 40,228 children revealed an overall mortality of 4.22%. Area under the receiver operating characteristics curve (95% CI) was 0.760 (0.749-0.771) for Pediatric Logistic Organ Dysfunction 2 with mechanical ventilation, 0.700 (0.689-0.712) for Pediatric Advanced Life Support, and 0.709 (0.696-0.721) for International Pediatric Sepsis Consensus Conference. When split by age group, the performance of Pediatric Logistic Organ Dysfunction 2 with mechanical ventilation was lowest in the youngest neonates (under 1 wk old), with an area under the receiver operating characteristics curve (95% CI) of 0.724 (0.656-0.791), and in the teenagers (13-18 yr), with an area under the receiver operating characteristics curve of 0.710 (0.682-0.738), yet it still outperformed Pediatric Advanced Life Support and International Pediatric Sepsis Consensus Conference in both groups.
Among critically ill children who had a discharge diagnosis of infection in the PICU, quick Sequential (Sepsis-Related) Organ Failure Assessment score performs best when using the Pediatric Logistic Organ Dysfunction 2 age thresholds with mechanical ventilation, while all definitions performed worse at extremes of pediatric age. Thus, mortality risk varies with vital sign thresholds, and although Pediatric Logistic Organ Dysfunction 2 with mechanical ventilation performed marginally better, it is unlikely to be of use to clinicians. More work is needed to develop a robust and relevant pediatric sepsis risk score.
我们通过从三个常用定义中选择阈值来评估快速序贯（脓毒症相关）器官衰竭评估评分（快速呼吸频率、意识状态改变、低血压）在儿科的适用性：儿科逻辑器官功能障碍 2、儿科高级生命支持和国际儿科脓毒症共识会议。我们使用 PIC 病房登记数据检查了它们各自在识别具有高死亡率风险的感染出院诊断儿童方面的表现，并特别关注年龄对性能的影响。
对来自虚拟儿科系统 PIC 病房数据库的回顾性数据进行分析。使用接收者操作特征分析评估了三种候选方法在预测观察死亡率方面的表现，包括年龄组的影响。
虚拟儿科系统数据库包含了美国和加拿大 130 个参与 PIC 病房前瞻性收集的临床记录中的诊断、临床标志物和结果数据。
2009 年至 2014 年间在参与 PIC 病房接受感染出院诊断的儿童，所有必需数据均可用。
无。
来自 40228 名儿童的数据显示总体死亡率为 4.22%。接收者操作特征曲线下面积（95%CI）为儿科逻辑器官功能障碍 2 伴机械通气 0.760（0.749-0.771），儿科高级生命支持 0.700（0.689-0.712），国际儿科脓毒症共识会议 0.709（0.696-0.721）。按年龄组划分时，儿科逻辑器官功能障碍 2 伴机械通气在最年幼的新生儿（1 周以下）中的表现最低，其接收者操作特征曲线下面积（95%CI）为 0.724（0.656-0.791），在青少年（13-18 岁）中为 0.710（0.682-0.738），但在这两个年龄段均优于儿科高级生命支持和国际儿科脓毒症共识会议。
在患有感染性 PIC 病房出院诊断的危重症儿童中，使用儿科逻辑器官功能障碍 2 伴机械通气的快速序贯（脓毒症相关）器官衰竭评估评分时性能最佳，而所有定义在儿科年龄极端时表现更差。因此，死亡率风险随生命体征阈值而变化，尽管儿科逻辑器官功能障碍 2 伴机械通气的表现略好，但对临床医生来说不太可能有用。需要做更多的工作来开发一个强大且相关的儿科脓毒症风险评分。
