胰岛素两种使用方法在危重症患者接受肠外营养时的比较
2010-03-12 18:13:59   来源：丁香园   作者：  评论：0 点击：

目的：高血糖是应激的一种常见反应，常导致不良后果，包括增加院内血液感染，住院时间延长，肾移植治疗需求增加，机械通气增加，死亡率上升。文献证实危重症患者需严格控制血糖水平，研究者也发现严格控制血糖会导致低血糖反应。研究者正在寻找接受肠外营养（PN）患者采用何种胰岛素注射方法更好地控制血糖水平。本研究的目的是比较接受PN的危重症患者两种静脉内使用胰岛素方法控制血糖效果，一种是持续胰岛素输注（CII），另一种是将胰岛素加入肠外营养中（IPN）。

　　方法：一所市级综合性中心医院中37例（31例外科患者，6例内科患者）接受PN患者随机分为CII组（21例）或IPN组（16例），目标血糖水平为80～120mg/dL。两组每天测定空腹血糖及四个关键点血糖。低血糖事件定义为<60mg/dL，高血糖事件定义为>200mg/dL。两组之间采用Mann-Whitney U检验评估血糖控制情况。两组之间低血糖和高血糖关系采用卡方检验。

　　结果：CII组和IPN组基线空腹血糖无差异（分别为120mg/dL和122mg/dL，p=NS），使用PN一天后两组血糖均升高（分别为136mg/dL和154mg/dL，p=NS）。使用PN第2天IPN组平均空腹血糖水平较CII组明显高（分别为149mg/dL和107mg/dL，p=0.003），第3天为（分别为140mg/dL和102mg/dL，p<0.0001），第4天（分别为123mg/dL和98mg/dL，p<0.05）。IPN组所有血糖均值（空腹及各时点血糖水平）同样明显高于CII组，第2天为（分别为154mg/dL和115mg/dL，p=0.006），第3天为（分别为141mg/dL和109mg/dL，p<0.0001）。第4～7天血糖控制水平两组之间无显著差异。CII组低血糖事件发生率较IPN组稍高（分别为8例和2例，p=0.08），高血糖事件发生率也稍高（分别为11例和6例，p=NS）。

　　结论：根据本研究结果，CII方法达到目标血糖较IPN方法快，但是最近的文献建议目标血糖提高到150mg/dL。扩大目标血糖范围可能增加PN中加入胰岛素方法的使用。尚需更多的大型研究观察何种胰岛素使用方法能够更好地达到新的目标血糖，并将影响预后的低血糖和高血糖事件控制到最少。

Clinical Nutrition Week 2010 Nutrition Practice Abstracts

Abstract of Distinction. Also appeared in Symposium W20: Glucose Control in Adult and Pediatric Critical Care Patients:

Nutr Clin Pract. 2010 Feb;25(1):93-4.

P3 - A Comparison of Two Methods of Insulin Administration in Critically Ill Patients Receiving Parenteral Nutrition

Yimin Chen, MS, RD, CNSD1; Jenny Lewandowski, MS, RD2; Matthew Sperry, MD3; Kathryn Keim, PhD, RD1; Diane Sowa, MBA, RD1; Sarah Peterson, MS, RD, CNSC1.

1Food and Nutrition, Rush University Medical Center, Chicago, IL; 2Radiant Research, Chicago, IL; 3Intermountain Medical Group, Provo, UT.

Introduction: While hyperglycemia may be a normal response to stress, it is associated with adverse patient outcomes including increased nosocomial bloodstream infections, length of stay, need for renal replacement therapy, need for mechanical ventilation, and mortality. Although the benefits of tight glycemic control in critically ill patients have been confirmed in the literature, hypoglycemia is a detrimental complication with intensive glucose control that has also been observed by investigators. Research has yet to be conducted to explore which method of insulin administration results in optimal glycemic control in patients receiving parenteral nutrition (PN). The objective of this study was to compare glycemic control between two methods of intravenous insulin administration in critically ill patients receiving PN: 1) continuous insulin infusion (CII); 2) addition of insulin to the parenteral nutrition (IPN). Methods: Thirty-seven surgical (n = 31) and medical (n = 6) intensive care unit patients in a tertiary care urban academic medical center receiving PN were prospectively identified and randomized to either CII (n = 21) or IPN (n = 16) group with a goal glycemic control of 80 -120 mg/dL. Blood glucose was monitored via morning blood draw and four point-of-care regimen per day for both groups. Hypoglycemic events were defined as <60 mg/dL; hyperglycemic events were defined as >200 mg/dL. Mann-Whitney U was performed to assess the glycemic control between study groups. Chi-square tests were conducted to determine the association of hypo-and hyperglycemic events between study groups. Results: Baseline morning blood glucose was similar between the CII and IPN groups (120 vs. 122 mg/dL, respectively; p = NS), and increased in both groups on day 1 after PN was initiated (136 and 154 mg/dL, respectively; p = NS). The median morning blood glucose was significantly higher in the IPN group when compared with the CII group on day 2 (149 vs. 107 mg/dL, respectively; p = 0.003), day 3 (140 vs. 102 mg/dL, respectively; p < 0.0001), and day 4 (123 vs. 98 mg/dL, respectively; p < 0.05) of PN infusion (Figure 1). The median combined blood glucose (morning and point-of-care blood glucose levels) was also significantly higher in the IPN group when compared with the CII group on day 2 (154 vs. 115 mg/dL, respectively; p = 0.006) and day 3 (141 vs. 109 mg/dL, respectively; p < 0.0001) of PN infusion. Blood glucose control for all subsequent PN days (days 4 -7) was not significantly different between the two groups (Figure 2). There was a trend towards more hypoglycemic events in the CII group when compared with the IPN group (8 vs. 2, respectively; p = 0.08), as well as hyperglycemic events (11 vs. 6, respectively; p = NS). Conclusions: Based on the results of this study, the CII method reached goal glycemic control sooner than the IPN method; however, investigators from recent literature suggest increasing goal glycemic control to 150 mg/dL. Liberalizing blood glucose goals may allow adequate glycemic control with the addition of insulin in PN. More research is necessary to determine which method of insulin administration is best to achieve new goal glycemic control, while minimizing hypo- and hyperglycemic events that may result in detrimental outcomes in a larger sample size.