| 摘要: | 研究背景與目的：文獻指出住院病人營養不良的發生率約20-40 %，入住加護病房的重症病人營養不良的比率更高，營養狀況不良與否攸關病人的預後，臨床上常使用營養篩選工具與客觀數據來判斷病人營養狀況，本研究的目的是為了要探討主觀整體營養評估與客觀數值及臨床結果的相關性。研究方法：本研究在永康奇美醫學中心一般外科加護病房進行，以回溯病歷統計方式從2010年1月至2012年3月共889位外科重症病人使用主觀整體營養評估(SGA, Subjective Global Assessment)篩選工具，與臨床客觀資料如體位測量：身高(BH, body height)、體重(BW, body weight)、身體質量指數(BMI, body mass index)；血液生化值如白蛋白(Albumin)、前白蛋白(Pre-albumin)、血中尿素氮(BUN, blood urea nitrogen)、血中肌酸酐(Cr, creatitine)、總淋巴球計數(TLC, total lymphocyte count)，及入住加護病房時疾病嚴重度分數包括急性生理和慢性健康評估系統(APACHE II, Acute Physiology and Chronic Health Evaluation II)、治療介入系統指標(TISS, therapeutic intervention scoring system)、昏迷指數(GCS, Glasgow Coma Scale)並收集病人加護病房住院天數與總住院天數及死亡率。病人經SGA評估後，分為營養狀況良好(SGA:A級)及營養狀況不良(SGA:B級+C級)，兩組進行比較。統計分析方法方面，連續變項以T檢定(student T test)，類別變項以卡方檢定( Chi-square test)來比較兩組的差異性， 並使用propensity score matching研究方法降低性別及年齡上所造成的選擇性偏差。研究結果：在889位外科重症病人中營養狀況良好(SGA A)有423位，營養狀況不良(SGA B或C)有466位，死亡110人(死亡率為12.4 %)。在單變項分析中，兩組無論在年齡(51.06 ± 18.56 years vs 66.24 ± 14.68 years，p<0.001)、身高(164.7 ± 8.06 cm vs 161.2 ± 8.50 cm，p<0.001)、體重(68.78 ± 14.76 kg vs 57.38 ± 11.20 kg，p<0.001)、BMI(25.35 ± 5.12 kg/m2 vs 22.03 ± 3.64 kg/m2，p<0.001)、BUN(17.69 ± 13.79 mg/ dl vs 27.03 ± 22.10 mg/ dl，p<0.001)、Cr<1.5mg/dl(82.98 % vs 67.38%，p<0.001)、Albumin<3.5mg/dl(74.47% vs 94.85 %，p<0.001)、Pre-albumin<20mg/dl (54.14 % vs 85.62 %，p<0.001)、APACHE II (10.91± 7.45 vs 14.31 ± 8.13，p<0.001)、TISS (26.67± 9.17 vs 28.23 ± 6.93，p=0.0046)，加護病房平均天數(6.89 ± 6.25 days vs 8.56 ± 9.36 days，p=.00016)，總住院平均天數(21.63 ± 17.00 days vs 29.98 ± 21.69 days ，p<0.001)都有顯著的差異。使用propensity score1:1配對的結果，多變項分析顯示BMI<18.5 kg/m2 ( OR為6.42，95 % CI為2.31-17.88)，Albumin<3.5 mg/ dl (OR為2.05，95 % CI為1.04-4.01)，Pre-albumin<20 mg/ dl (OR為1.02 ，95 %CI為1.01-1.04)，TLC<900 cell/mm3 (OR為2.05，95 % CI為1.52-4.29) 在外科重症病人發生營養不良的風險上有顯著的差異。研究結論：營養不良在外科重症病人中有相當高的盛行率，它不但造成較高的死亡率，同時也延長病人的平均住院天數。臨床上除了使用主觀整体營養評估，若無法使用營養評估工具評估時也可藉由測量血中白蛋白、前白蛋白、總淋巴球計數來評估病人營養狀態，且當病人血液動力學穩定後，照護團隊能及早進行病人營養支持介入與擬訂營養治療方針，或許可改善病人臨床預後。
Background and Objective: Overall 20-40%of hospital patients are malnourished. The incidence of malnutrition among critically ill patients was even higher. Nutrition screening tools and objective parameters such as serum biomarker, body mass index, were often used to determine the nutritional status of patients. The purpose of this study was to investigate the association between the nutrition assessment tool using subjective global assessment (SGA) from the objective parameters among the critically ill surgical patients.Methods: This study was conducted in the surgical intensive care unit of Chi Mei Medical Center. We retrospectively collected patients consecutively from January, 2010 to March, 2012. SGA and objective parameters including body height, body weight, body mass index, serum albumin, pre-albumin, creatinine, blood urea nitrogen, total lymphocyte count were measured. Acute physiology and chronic health evaluation II score (APACHE II score), therapeutic intervention scoring system (TISS score), Glasgow Coma Score (GCS), length of ICU and hospital stay were recorded. Cases were divided into well- nourished (SGA A) and malnourished (SGA B+C) groups, the difference between two groups were compared.All of the continuous data are expressed as the mean ± SD; categorical variables are reported as percentages. The statistical comparative analysis was performed using the Chi-square test for qualitative data. Student t-test was used for quantitative data. For reducing the selection bias from age and gender, the SGA group A and SGA group B+C was matched by age and gender. This study was applied the propensity score matching on 1:1 case-control. Statistical difference was defined as p<0.05.Results: Of all the 889 patients, 423 patients were well-nourished and 466 patients were malnourished. The overall mortality was 12.4%. As comparing the well-nourished to malnourished patients, univariate analysis showed significant difference in age (51.06 ± 18.56 years vs 66.24 ± 14.68 years, p<0.001), body height (164.7 ± 8.06 cm vs 161.2 ± 8.50 cm, p<0.001), body weight (68.78 ± 14.76 kg vs 57.38 ± 11.20 kg, p<0.001), BMI(25.35 ± 5.12 kg/m2 vs 22.03 ± 3.64 kg/m2, p<0.001), BUN(17.69 ± 13.79 mg/ dl vs 27.03 ± 22.10 mg/ dl, p<0.001), Cr<1.5mg/dl(82.98 % vs 67.38%, p<0.001), Albumin<3.5mg/dl(74.47% vs 94.85 %, p<0.001), Pre-albumin<20mg/dl (54.14 % vs 85.62 %, p<0.001), APACHE II (10.91± 7.45 vs 14.31 ± 8.13, p<0.001), TISS (26.67± 9.17 vs 28.23 ± 6.93, p=0.0046), length of ICU stay(6.89 ± 6.25 days vs 8.56 ± 9.36 days, p=.00016), length of hospital stay (21.63 ± 17.00 days vs 29.98 ± 21.69 days, p<0.001)。Using propensity score matching on 1:1 case-control revealed BMI<18.5 kg/m2 ( OR 6.42，95 % CI 2.31-17.88), Albumin<3.5 mg/ dl (OR 2.05, 95 % CI 1.04-4.01), Pre-albumin<20 mg/ dl (OR 1.02, 95 %CI 1.01-1.04), TLC<900 cell/mm3 (OR 2.05, 95 % CI 1.52-4.29) were significantly associated with nutritional assessment tool using SGA in critically ill surgical patients.Conclusion: The prevalence of malnutrition in critically ill surgical patients is high. It not only causes a higher mortality rate, but also prolongs the ICU and hospital stay. In addition to using SGA in assessing the nutritional status of critically ill surgical patients, measuring serum albumin, pre-albumin, and total lymphocyte count could also be applied. Early recognizing malnourished patients and prompt nutritional intervention may improve patient outcomes in surgical ICU. |
