血清白蛋白是使用呼吸器的重症病患死亡率的重要決定因子

本研究主要目的是探討血清白蛋白是否為評估加護病房病患死亡率的良好指標；次要的目的是要比較APACHE II及APACHE III與加護病房病患死亡率的相關性。 本實驗屬回溯性研究，收集對象為持續在加護病房3天以上的重症病患。血清白蛋白等生化檢驗值於病患入加護病房72小時內取得，而APACHE II及III分數等重症指標評分則是於進入加護病房24小時內進行，並記錄病患存活狀況。以Student’s t-test及Mann-Whitney rank sum test來比較存活組及非存活組的各項基本資料；以多變項邏輯式迴歸探討加護病房的死亡率與各項因子的危險比對值。 於研究期間共收集160位加護病房的病患，其中存活組83位，非存活組77位。非存活組的血清白蛋白顯著低於存活組，而APACHE II及III則是非存活組顯著高於存活組。於單變項分析時，APACHE II、III及白蛋白對於死亡率皆有顯著的影響。在多變項邏輯式迴歸分析時，調整C-反應蛋白，發現白蛋白對於死亡率仍有顯著的影響，但APACHE II及APACHE III與死亡率的關係則降低。進一步將白蛋白分層，無論是否加入各項重症指標調整時，白蛋白與死亡率皆有顯著相關性。 本研究發現無論是APACHE II或是APACHE III的分數越高，死亡率越高。APACHE II和APACHE III與死亡率的相關性並無顯著差異，但是血清白蛋白卻是評估重症病患死亡率的一個重要決定因子。

The purpose of this study was to determine whether serum albumin level could be considered as an important determinant of mortality in the intensive care unit (ICU). Additional purpose was to compare the degree of association of APACHE II or III score with mortality in mechanically ventilated critically ill patients. This retrospective study selected patients who had to stay in ICU more than three days. Medical charts were examined for anthropometric and biochemical measurements within the first 72 hours of their admission. The admission scores of APACHE II and III were obtained from the ICU database. Patient’s survival status was also recorded. Student t-test and Mann-Whitney rank sum test were used to compare the demographic and clinical outcomes of survivor and non-survivor groups. Multivariate logistic regression analysis was used to explore odds ratio for patients’ clinical parameters associated with mortality at their admission to the ICU. The study population consisted of one hundred and sixty patients including eighty-three survivors and seventy-seven non-survivors in the ICU. Mean serum albumin level in the non-survivor group was significantly lower than the survivor group. The APACHE II and III scores in the non-survivor group were significantly higher than the survivor group. Univariate analysis showed that APACHE II, III, and albumin significantly associated with the mortality. Albumin still maintained the same degree of association with mortality when adjusting for C-reactive protein in multiple logistic regression analysis, while the association between APACHE II or III and mortality decreased. Our results showed that both APACHE II and III have the similar and significant association with mortality in the ICU. Serum albumin level other than APACHE II and III is an important determinant of morality in the critically ill patients.