ICU患者的低血糖与死亡强烈相关

9月19日《新英格兰医学杂志》在线发表的NICE-SUGAR研究的亚组分析结果显示，重症监护病房(ICU)患者的低血糖与死亡率增加强烈相关(N. Engl. J. Med. 2012;367:1108-18 [doi: 10.1056/NEJMoa1204942])。

此前发表的NICE-SUGAR研究数据显示，与常规血糖控制组相比，强化血糖控制组的死亡风险增加14%(N. Engl. J. Med. 2009;360:1283-97)。

在本次分析中，研究者在6,026例随机接受强化血糖控制(目标血糖范围81~108 mg/dl)或常规血糖控制(目标血糖值为180 mg/dl)的患者中探讨了中度低血糖(血糖41~70 mg/dl)和重度低血糖(≤40 mg/dl)与死亡之间的相关性。

结果显示，中度低血糖发生率为45%，重度低血糖发生率为3.7%。根据治疗组分析发现，强化控制组和常规控制组的中度低血糖发生率分别为74%和16%，重度低血糖发生率分别为6.9%和0.5%。

总死亡率为26.2%。在无低血糖的3,089例患者中，死亡率为23.5%；在发生中度低血糖的2,714例患者和发生重度低血糖的223例患者中，死亡率分别为28.5%和35.4%，校正风险比分别为1.41和2.10，具有高度显著性。在低血糖持续1天以上的患者中观察到的这种相关性，强于在低血糖持续时间较短的患者观察到的相关性。

Simon Finfer博士

在校正首次低血糖发作后发生的事件后，仍发现中度低血糖与死亡风险增加40%相关，重度低血糖与死亡风险增加1倍相关。

研究者还发现，血糖控制的强度与低血糖和死亡发生率之间存在剂量反应关系。在各个亚组(如糖尿病患者和非糖尿病患者)中观察到的低血糖与死亡之间的相关性均具有鲁棒性。低血糖与分布性(血管舒张性)休克所致死亡之间的相关性最强烈。在少数患者中，低血糖似乎是死亡风险增加的标志物，而非原因。在这些患者中，低血糖可能预示着极重度的潜在疾病过程和即将发生死亡。

所有这些结果均提示低血糖与不良预后之间存在因果关系，但由于该研究的设计问题，无法证实这种因果关系。研究者表示，在控制危重症患者的血糖浓度时，不仅应注重控制高血糖，还应注重避免中度和重度低血糖的发生。

该研究获澳大利亚国立卫生和医学研究理事会、新西兰卫生研究理事会和加拿大卫生研究院支持。研究者声明与多家药企存在联系。

随刊述评：强化血糖控制目前不可接受

华盛顿大学西雅图分校的内科教授Irl B. Hirsch博士表示，对ICU患者进行强化血糖控制以达到极低血糖浓度的做法目前看来是不合理的。上述研究中常规血糖控制组的血糖维持水平较为安全，并且与现行指南建议的140~180 mg/dl水平一致(Diabetes Care 2009;32:1119-31)。每个ICU都应持续评价治疗的质量和使用胰岛素治疗方案的适当性(N. Engl. J. Med. 2012 Sept. 19 [doi: 10.1056/NEJMe1208208])。

爱思唯尔版权所有  未经授权请勿转载

By: MARY ANN MOON, Clinical Endocrinology News Digital Network

Hypoglycemia in patients in the intensive care unit is strongly associated with increased mortality, according to a subanalysis of the NICE-SUGAR trial published online Sept. 19 in the New England Journal of Medicine.

Researchers performed a more detailed, post hoc analysis of data collected in the previously published NICE-SUGAR (Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation) trial, in which ICU patients assigned to intensive glucose control had a significant, 14% increased risk of death, compared with those assigned to conventional treatment (N. Engl. J. Med. 2009;360:1283-97).

When dealing with patients in the ICU, intensive glucose control can lead to hypoglycemia and even death.

In this analysis, the investigators confirmed that both moderate and severe hypoglycemia were significantly more common in ICU patients randomly assigned to receive intensive glucose control than in those assigned to receive conventional glucose control, and that hypoglycemia – in either group – was significantly linked to mortality.

Moreover, the findings show that there is a dose-response relationship between the intensity of glucose control and the rates of hypoglycemia and of death, that the association is consistent across several subgroups of patients, and that the link is strongest for death caused by distributive (vasodilative) shock. All of these findings point to a causal relationship between hypoglycemia and adverse outcomes, but causality cannot be proved because of the design of this study, said Dr. Simon Finfer, of the NICE-SUGAR writing committee and Royal North Shore Hospital and the George Institute for Global Health, University of Sydney, and his associates.

Nevertheless, "it would seem rather prudent to ensure that strategies for managing blood glucose concentration in critically ill patients focus not only on the control of hyperglycemia but also on avoidance of both moderate and severe hypoglycemia," they noted.

Hyperglycemia occurs frequently in ICU patients and is associated with increased morbidity and mortality. A study published in 2001 suggested that intensive glucose control, using insulin if necessary, reduced that morbidity and mortality (N. Engl. J. Med. 2001;345:1359-67), and the practice became widespread.

From that point until the publication of the NICE-SUGAR study in 2009, glucose control in hospital patients was primarily aimed at avoiding hyperglycemia. But NICE-SUGAR and other studies prompted a revision of guidelines for inpatient glycemic control with an increased focus on avoiding hypoglycemia (Diabetes Care 2009;32:1119-31).

For the current analysis, the NICE-SUGAR investigators looked at the associations between moderate hypoglycemia (blood glucose 41-70 mg/dL) and severe hypoglycemia (40 mg/dL or below) and death in 6,026 of the study subjects, who had been randomized to receive either intensive blood glucose control, with a target blood glucose range of 81-108 mg/dL, or conventional control, with a target of 180 mg/dL.

Moderate hypoglycemia developed 45% of patients, and severe hypoglycemia occurred in 3.7% of patients. Broken down by treatment group, moderate hypoglycemia occurred in 74% of the intensive-control group and 16% of the conventional-control group; severe hypoglycemia developed in 6.9% and 0.5%, respectively.

Overall mortality was 26.2%. Mortality in the 3,089 patients with no hypoglycemia was 23.5%. That rate was 28.5% in the 2,714 patients who developed moderate hypoglycemia and 35.4% in the 223 who developed severe hypoglycemia, for highly significant adjusted hazard ratios of 1.41 and 2.10, respectively. The association was stronger in patients whose hypoglycemia lasted more than 1 day than in those with shorter durations of hypoglycemia.

"Even after adjustment for events occurring after the first episode of hypoglycemia, moderate hypoglycemia was associated with an increase in the risk of death of 40%, and severe hypoglycemia with a doubling of the risk," Dr. Finfer and his colleagues said (N. Engl. J. Med. 2012;367:1108-18 [doi: 10.1056/NEJMoa1204942]).

The link between hypoglycemia and death also was robust across subgroups of patients, including patients who had underlying diabetes and those who did not have underlying diabetes.

In a minority of patients, hypoglycemia appears to be a marker, rather than a cause, of an increased risk of death. In these cases it probably signals very severe underlying disease processes and impending death, the investigators said.

This study was supported by the Australian National Health and Medical Research Council, the Health Research Council of New Zealand, and the Canadian Institutes of Health Research. Dr. Finfer reported ties to Edwards, and his associates reported ties to Fresenius Kabi, AstraZeneca, and Eisai.

View on The News
Intensive Glucose Control Now Unacceptable

Intensive glucose control aimed at achieving very low blood glucose levels in ICU patients should now be considered unacceptable, "and older, nonchalant attitudes need to be abandoned," said Dr. Irl B. Hirsch.

Instead, "maintaining blood glucose at levels similar to those in the conventional-control group of the NICE-SUGAR population is safe," and is consistent with existing guidelines recommending 140-180 mg/dL (Diabetes Care 2009;32:1119-31). "Continued assessment of the quality of care and the appropriate use of insulin protocol should be the standard for every ICU," he said.

DR. HIRSCH is professor of medicine at the University of Washington, Seattle. He reported ties to Johnson & Johnson, Roche, Abbott, CellNovo, Sanofi, and Halozyme. These remarks were taken from his editorial comment accompanying Dr. Finfer’s report (N. Engl. J. Med. 2012 Sept. 19 [doi: 10.1056/NEJMe1208208]).