前白蛋白并非评估营养支持是否充足的敏感标志物
2010-03-12 18:13:27   来源：   作者：  评论：0 点击：

目的：前白蛋白（PAB）通常作为评估体内蛋白状态的标志物，因此也可以用于监测患者对营养支持治疗的反应。采用PAB评估体内蛋白状态可能受炎症反应影响，因为炎症时肝脏优先使用PAB合成急性期蛋白，如C反应蛋白（CRP）。本研究的目的是观察除了妇产科和精神科以外所有病区接受肠内营养（EN）年龄大于18岁的成人住院患者的PAB是否能够反映体内热量和蛋白质变化或炎症时变化。

　　方法：一项回顾性分析观察了154例接受三天以上EN且至少两次测量PAB的成人患者。根据实际或体重校正后患者个体需求计算热量（30～35千卡/公斤体重）。根据体重和临床情况计算蛋白质需求量（0.9～2.5克/公斤体重）。基线和每周两次一共30天进行评估，热量和蛋白质摄入与PAB变化比较。PAB低于18mg/dL时测量C反应蛋白浓度。本研究通过伦理委员会批准。统计采用SPSS软件进行。

　　结果：每天平均热量和蛋白质需求量分别为1966±353千卡和109.8±31.4克。57%的需求热量和56%的需求蛋白质通过输注给予。患者根据输注热量和蛋白质百分比进行分组。第一组（n=52）输注需求热量百分比为4～50%，第二组（n=54）为51～68%，第三组（n=48）为69～114%。第一组（n=54）输注需求蛋白质百分比为4～48%。第二组（n=52）为49～65%，第三组（n=48）为66～143%。根据热量和蛋白质百分比分组进行单向方差分析。无论按输注热量百分比还是按输注蛋白质百分比，三组PAB变化无显著差异，分别为F（2，151）=1.005，p=0.37和F（2，151）=1.906，p=0.15。通过分析C反应判断是否存在炎症。CRP变化与PAB变化无相关性（r=-0.544，p<0.001）。校正CRP后可采用多元线性回归模型评估输注热量百分比或输注蛋白质百分比预测PAB变化能力。校正输注热量百分比（B=-0.051，p<0.001）或输注蛋白质百分比（B=-0.051，p<0.001）后，仅有CRP变化能够显著预测PAB水平变化，解释29.6%PAB变化差异。

　　结论：这些结果显示PAB并非评估营养支持是否充足的敏感标志物。CRP变化是能够预测PAB水平变化的唯一变量，提示炎症状态变化，而不是营养摄入变化，可引起PAB水平升高。

Clinical Nutrition Week 2010 Nutrition Practice Abstracts

Abstracts of Distinction

Nutr Clin Pract. 2010 Feb;25(1):93.

P2 -The Use of Prealbumin and C-Reactive Protein for Monitoring Nutrition Support in Adult Patients Receiving Enteral Nutrition in an Urban Medical Center.

Cassie Davis, MS RD; Diane Sowa, MBA RD; Kathryn Keim, PhD RD; Kelly Kinnare, MS RD CNSD.

Food and Nutrition, Rush University Medical Center, Chicago, IL.

Introduction: Prealbumin (PAB) is commonly used as a marker to assess protein status and is therefore also used to monitor a patient's response to nutrition support. The ability of PAB to adequately assess protein status may be influenced by the presence and severity of the inflammatory response because the liver preferentially synthesizes acute phase proteins such as C-reactive protein (CRP) at the expense of PAB in the presence of inflammation. The purpose of this study was to determine whether changes in PAB are reflective of the delivery of adequate calories and protein or of changes in inflammatory status in hospitalized adults, greater than 18 years of age, receiving enteral nutrition (EN) on all patient care units, except Maternity and Psychiatric units. Methods: A retrospective review was conducted on 154 adult patients who received EN for more than three days and had at least two measures of PAB. Calorie requirements were calculated (30-35 kcal/kg) based on actual or adjusted body weight and individualized to patients' needs. Protein requirements were calculated (0.9 -2.5 g/kg) based on body weight and clinical condition. Calorie and protein intake was compared to changes in PAB, assessed at baseline and twice a week for up to 30 days. C-reactive protein was assessed when PAB was less than 18 mg/ dL. Approval for the study was obtained from the Institutional Review Board for Human Subjects. SPSS for Windows (version 15.0, 2006, SPSS Inc, Chicago, IL) was used for statistical analysis. Results: Mean calorie and protein requirements were 1966 +/-353 kcal/day and 109.8 +/-31.4 g protein/day, respectively. Fifty-seven percent of calorie needs and 56% of protein needs were delivered. Subjects were divided into tertiles based on percent calories and protein delivered. Percent of calorie requirements delivered for the first tertile (n=52) was 4-50%, second tertile (n=54) was 51-68%, and third tertile (n=48) was 69-114%. Percent of protein requirements delivered for the first tertile (n=54) was 4-48%, the second tertile (n=52) was 49-65%, and the third tertile (n=48) was 66-143%. One-way ANOVA were conducted for both calories and protein based on tertiles of percent calories and protein delivered. There was no significant difference in change in PAB among the three tertiles for either percent calories delivered: F (2, 151)=1.005, p=0.37 or protein delivered: F (2, 151)=1.906, p=0.15. C-reactive protein was analyzed to account for the presence or absence of inflammation. Change in CRP was negatively correlated with change in PAB (r =-0.544, p<0.001). Two multiple linear regression models were fit to assess the ability of either percent calories delivered or percent protein delivered to predict changes in PAB while adjusting for CRP. Only change in CRP was able to significantly predict change in PAB levels, explaining 29.6% of the variance (R2=0.296) in change in PAB consistently, adjusting for either percent calories delivered (B=-0.051, p<0.001) or percent protein delivered (B=-0.051, p<0.001). Conclusions: These results indicate that PAB is not a sensitive marker for evaluating the adequacy of nutrition support. Change in CRP was the only variable that was able to significantly predict changes in PAB levels, suggesting that a change in inflammatory status, rather than nutrient intake, was responsible for the increases seen in PAB levels.