比較 APACHE II, SOFA, PIRO, qSOFA 及uqSOFA 分數預測加護病房外科敗血症病人死亡及不良事件之敏感性及特異性

研究背景：敗血症及敗血性休克為外科敗血症病人入住加護病房和造成院內死亡的重要原因，其死亡率高達28-36%。過去針對病人疾病嚴重程度建立許多量化的評估工具，但仍希望發展出快速且完整評估敗血症病人不同器官功能的評估工具，可協助臨床醫療人員對敗血症病人做更準確的判斷，盼以此改善敗血症外科病人的存活率。研究目的：以2016年第三次國際共識會議提出的快速版連續性器官衰竭評分(quick Sequential Organ Failure Assessment Score, qSOFA)加上尿量為uqSOFA，探討在預測加護病房外科敗血症病人的死亡和不良事件的敏感性和特異性，並與急性生理與慢性健康評分(Acute Physiologic and Chronic Health Evaluation II Score, APACHE II)、連續性器官衰竭評分 (Sequential Organ Failure Assessment, SOFA) 、基本體質、疾病衝擊、生理反應和器官功能障礙模型(Predisposition, Infection/injury, Response, Organ dysfunction ,PIRO)及qSOFA的預測力進行比較。研究方法：本研究採回溯性方式、描述性相關性研究設計，以結構式問卷回溯加護病房2016年至2017年診斷敗血症或敗血性休克且接受手術治療後103位病人的病歷資料，將APACHE II、SOFA、PIRO及qSOFA使用ROC curve比較不同評估工具預測加護病房外科敗血症病人的死亡和住院天數能力之差異，以邏輯式回歸分析計算qSOFA於死亡和敗血症不良事件之風險值。研究結果：uqSOFA的加護病房外科敗血症病人死亡預測力以首8小時內uqSOFA≧3分的特異性較好(0.72)，但敏感性較差(0.67)。uqSOFA≧2分於不良事件預測力的特異性(0.73)和敏感性(0.74)皆較好。五種疾病嚴重程度評估工具於加護病房外科敗血症病人死亡預測能力以APACHE II(0.84)最好，但兩兩相比並無顯著差異。於加護病房外科敗血症病人不良事件預測能力以SOFA(0.88)最好，但兩兩相比並無顯著差異。結論：本研究發現APACHE II、 PIRO、SOFA、qSOFA及uqSOFA五種評估工具對於敗血症死亡和不良事件發生皆有合適的預測能力。以臨床實用性和便利性為考量的目的，建議可以使用qSOFA及uqSOFA作為加護病房敗血症外科病人的疾病嚴重度評估工具，以便簡化工作流程、節省計算時間和成本、快速獲得結果。特別是使用uqSOFA≧2分高特異性的特點，能對敗血症不良事件中獲得有效提醒，如此可盡早介入治療防範疾病惡化和改善臨床預後結果。

關鍵字

敗血症；疾病嚴重度評估工具；死亡；敗血症不良事件

並列摘要

Background：Sepsis and septic shock are the leading causes of admission to the intensive care unit and in-hospital death. The associated mortality rate may rises as high as 28-36%. Several effective assessment tools of disease severity had been developed; however, the feasibility of using these tools on surgical patients with multiple organ failures needs to be reassured as these tools may help clinical teams in detecting and diagnosing sepsis more accurately, and improve survival rate in surgical patients with sepsis. Purpose：The primary aims are to check the sensitivity and specificity the new tool Urine point quick Sequential Organ Failure Assessment Score (uqSOFA), which is modified from the quick Sequential Organ Failure Assessment Score (qSOFA) with the addition of a urine item, for predicting the mortality and the adverse event, and to compare the predictability of the assessment tools of Acute Physiologic and Chronic Health Evaluation (APACHE II), Sequential Organ Failure Assessment (SOFA), Predisposition, Infection/injury, Response, Organ dysfunction (PIRO) and qSOFA. Method：A retrospective observational cohort study was conducted with a chart review of 103 septic ICU patients who met the inclusion criteria from the year 2016 to the year 2017. Discrimination was assessed using the area under the receiver operating characteristic (ROC) curve. Logistic regression was used to determine the odds ratio of the mortality and the adverse event of sepsis. Result：The score of uqSOFA≧3 demonstrated greater specificity( 0.72) but less sensitivity (0.67) of the predictability of the mortality. The score of uqSOFA≧2 demonstrated better specificity (0.73) and sensitivity (0.74) for predicting the adverse event. APACHE II (0.84)showed the best discriminating capacity in predicting ICU mortality and SOFA(0.88) is the best in predicting the adverse event. However, there were no significant differences among the five tools in predicting the mortality and the adverse event. Conclusion：The assessment tools of APACHE II, SOFA, PIRO, qSOFA, and uqSOFA are able to discriminate mortality and adverse event. More specifically, the tools of qSOFA and uqSOFA have proven to be more useful and effective for assessing disease severity in surgical ICU patients. In particular, a uqSOFA score of ≧2 can be interpreted as a warning sign of adverse events so that intervention can be initiated earlier to prevent clinical deterioration and to improve outcome.