Glucose Control


Data supporting a strong association between hyperglycemia and poor outcomes are abundant, and include patients with myocardial infarction {10338454,12446057,10711923}, traumatic injuries (Yendamuri Journal of Trauma 2003;55(1):33, Sperry Journal of Trauma 2007;63(3):487), and postoperative wound infections (Estrada Annals of Thoracic Surgery 2003;75(5):1392, Guvener Endocrine Journal 2002;49(5):531, Latham Inf Control & Hosp Epidem 2001;22(10):607, Golden Diabetes Care 1999;22(9):1408). Similarly, hyperglycemia has been associated with poor outcomes in patients suffering from stroke (Williams Neurology 2002;59(1):67, Capes Stroke 2001;32(10):2426, Kagansky Archives Neurol 2001;58(8):1209, Parsons Annals Neurol 2002;52(1):20, Bruno Neurol 2002;59(5):669), traumatic brain injury (Jeremitsky Journal of Trauma 2005;58(1):47), and subarachnoid hemorrhage (Lanzino J Neurosurg 1993;79(6):885, Claassen Stroke 2002;33(5):1225, Claassen Critical Care Medicine 2004;32(3):832, Juvela J Neurosurg 2005;102(6):998)

Glucose Control in the OR

Hyperglycemia and Outcome

Intraoperative hyperglycemia is an independent risk factor for perioperative complications in cardiac surgery patients (Gandhi Mayo Clin Proc 2005;80(7):862).

Effects of Treatment

Aggressive treatment of hyperglycemia in cardiac surgery patients does not improve mortality and in fact increases the incidence of stroke (p = 0.020) and death (p = 0.061) [Gandhi GY et. al. Annals of Internal Medicine 146: 233, 2007]. Thus, in instances where hyperglycemia has been correlated with adverse outcomes, it should not be assumed that aggressive treatment is necessarily beneficial.

Glucose Control in the ICU

Prospective Studies

Van den Bergh 2001 (SICU)

In 2001, Van den Bergh et al. conducted a study of 1548 predominantly surgical patients and demonstrated a statistically significant mortality reduction of 32% when patients were placed on a strict glucose control regimen (goal 80 – 110 mg/dL) as compared to “traditional” glucose control (van den BergheNEJM 2001;345(19):1359). This often cited study led many institutions to initiate a similar glucose control regimen in their intensive care units. Of note, most of Van den Bergh’s patients had undergone recent cardiac surgery). When Van den Bergh et al. reviewed a subset of neurologically injured patients from this study, they found that strict glucose control improved outcome scores but led to an increase in mortality (Van den Berghe Neurology 2005;64(8):1348).

Finney 2003 (Critically Ill Patients)

A prospective, observational study of 531 heterogeneous critical care patients found a statistically significant relationship between likelihood of death (odds ratio) and amount of insulin administered in each subgroup (categorized based on average glucose), but did not find a relationship between mortality and glucose level (Finney JAMA 2003;290(15):2041). These findings are compatible with the idea that elevated glucose is simply a marker for injury and that treating it does not provide any recognizable benefit.

Van den Bergh 2006 (MICU)

Van den Bergh et al. were unable to demonstrate a benefit of strict glucose control in the medical ICU (Van den Berghe NEJM 2006;354(5):449).

GluControl 2007

Study of 1101 critically-ill patients showed no reduction in mortality but an increased incidence of severe hypoglycemia when tight glucose control was used (Devos Intensive Care Med 2007;33(S189):189).

VISEP Study 2008 (Sepsis)

In the pre-maturely terminated (lack of benefit and in increased incidence of severe (< 40 mg/dL) hypoglycemia) Efficacy of Volume Substitution and Insulin Therapy in Severe Sepsis (VISEP) study, strict glucose control had no effect on mortality but increased the incidence of severe hypoglycemia and serious adverse events. Additionally, Cox regression analysis identified hypoglycemia as an independent risk factor for death from any cause (Brunkhorst NEJM 2008;358(2):125).

NICE-SUGAR (Medical and Surgical ICU)

Multicenter, international, randomized trial of strict (81-108 mg/dL) vs. conventional (144-180 mg/gL) glucose control in 6104 patients expected to remain in the ICU for 3 days or more. Odds ratio for mortality at 90 days was 1.14 (95% CI 1.07-1.31, p = 0.02) in the intense glucose control group [NEJM 360: 1283, 2009].


There are some who believe that the hyperglycemic response to stress may be adaptive and beneficial, and thus to attenuate it may be harmful (Preiser J-C. Critical Care 2008;12:116). Indeed, a review of 12 glucose control studies by Wilson et al. suggested that the 80 – 110 mg/dL range may not be ideal as it is based on a trial of predominantly post-surgical patients (Wilson Diabetes Care 2007;30(4):1005), all of whom received 200-300 g of IV glucose daily in the ICU (Van den Berghe NEJM 2001;345(19):1359-1367; Brunkhorst NEJM 2008;358(2):125). While Wilson et al. acknowledge that it is difficult to make comparisons across studies, a retrospective study of 1600 consecutive critically ill patients by Krinsley et al. showed a mortality reduction when glucose was kept below 140 mg/dL with no appreciable change in hypoglycemia, suggesting that the upper limit of glucose with regards to mortality may be need to be revised (Krinsley Mayo Clin Proc 2003;78(12):1471).

Wilson et al. further point out that in addition to differences in target glucose levels, there is no standardized dosing regimen across studies: different infusion algorithms lead to substantial variation in treatment efficacy as well as the incidence of hypoglycemia. Additionally, not all regimens take into account insulin sensitivity (Bode Endocrine Practice 2004;10 Suppl 2:71-80), which can also affect efficacy. Finally, it may be that glucose variability is just as important as mean values in determining outcome (Egi Anesthesiology 2006;105(2):244), a possibility that is neglected by most current protocols. Subramaniam et al found that preoperative HbA1C identifies risk for postoperative glycemic variability and patients with high postoperative glycemic is associated with major adverse events after cardiac surgery (Anesth Analg 2014;118:277–87).

Special Cases

Neurologic Injury

Animal data suggest that hyperglycemia prior to or during ischemic injury has a negative impact on outcome. Hyperglycemia has also been shown to adversely effect animal spinal cord injury [Anesthesiology 70: 64, 1989]. A retrospective study of 960 stroke patients showed that admission hyperglycemia >= 130 mg/dL was associated with a higher mortality rate (odds ratio = 3.15, p = 0.004). Persistent hyperglycemia for 48 hours was associated with even higher mortality rate (OR = 6.54, p < 0.001). Glycemic control was associated with a 4.6-fold decrease in mortality risk (p < 0.001) [Acad Emerg Med 13: 174, 2006]. Hyperglycemia was also shown to negatively effect outcomes in a retrospective study of 77 severe brain injury patients [J Trauma 58: 47, 2005].

Subanalysis of Van den Bergh’s large randomized, controlled trial of SICU patients with tight (< 110 mg/dL) control showed the following – in 63 patients with isolated brain injury, intensive insulin therapy reduced ventilation dependency (p = 0.0007) and the risk of critical illness associated polyneuropathy (CIPNP, p < 0.0001). Prevention of CIPNP explained the ability of intensive insulin therapy to reduce the risk of prolonged mechanical ventilation (OR 3.75, p = 0.005). In isolated brain injury patients, intensive insulin therapy reduced mean (p = 0.003) and maximal (p < 0.0001) ICP while identical CPPs were obtained with eightfold less vasopressors (p = 0.01). Seizures (p < 0.0001) occurred less frequently. At 12 months follow-up, more brain-injured survivors in the intensive insulin group were able to care for most of their own needs (p = 0.05) [Neurology 64: 1348, 2005].

Vespa et al. recently studied parenchymal glucose values by placing intracerebral microdialysis catheters in 47 patients with severe traumatic brain injuries (33 of whom received intensive insulin therapy with a target glucose 120 – 150 mg/dL, 14 of whom did not) and found that markedly low levels of intraparenchymal glucose could occur in the intensive insulin therapy group despite relatively normal systemic blood glucose levels (Vespa Critical Care Medicine 2006;34(3):850). While upregulation of the blood brain barrier GLUT1 transporter in hypoglycemic states is well established in animal models (Kumagai Diabetes 1995;44(12):1399, Boado J Neurochem 1993;60(6):2290, Boado J Neurochem 2002;80(3):552), evidence of downregulation of the GLUT1 in animal models is mixed (Cornford Neurochem Res 1995;20(7):869, Simpson J Neurochem 1999;72(1):238). Based on Vespa et al.’s data, it seems possible that patients who had chronically elevated serum glucose levels have less native GLUT1 and therefore low CNS glucose levels following rapid normalization of chronic hyperglycemia. If this were the case, response to intensive insulin therapy would depend heavily on the level of pre-SAH glucose control. In patients who have poorly controlled diabetes prior to their SAH, tight glucose control might lead to local hypoglycemia that would not be detected by standard laboratory measures.

Bilotta 2007 (SAH patients)

While the prognostic ramifications of hyperglycemia in subarachnoid hemorrhage patients are known, a benefit of glucose control has not been demonstrated. Bilotta et. al. conducted a prospective, randomized, controlled trial of strict glucose control in 78 patients with subarachnoid hemorrhage did not show a mortality benefit, but the authors did not report glucose values for each group, making it difficult to interpret the results (Bilotta Journal of Neurosurgical Anesth 2007;19(3):156).

Hyperglycemia in CNS Injury

  • Multiple animal studies show worsened neurologic outcome in both brain and spine
  • Retrospective study of 960 emergency stroke patients shows BG > 130 increases mortality (OR 3.15), especially for 48 hours after, and correction decreases mortality (OR 4.6) [Acad Emerg Med 13: 174, 2006]
  • Subgroup analysis of 63 TBI patients showed: keeping glucose < 110 mg/dL lowered ventilator time (most likely by decreasing CIPNP), ICP, vasopressor requirements, and seizure incidence. Outcomes were also improved [Neurology 64: 1348, 2005]
  • IVF in the first 24 hours after injury should be isotonic (0.9%) and dextrose free
  • In order to prevent cerebral edema, do not lower glucose by more than 75 mg/dL/hr

When severe hyperglycemia is present, do not lower glucose by more than 75-100 mg/dL/hr, otherwise brain edema may occur.



Balachundhar Subramaniam, Adam Lerner, Victor Novack, Kamal Khabbaz, Maya Paryente-Wiesmann, Philip Hess, Daniel Talmor
Increased glycemic variability in patients with elevated preoperative HbA1C predicts adverse outcomes following coronary artery bypass grafting surgery.
Anesth. Analg.: 2014, 118(2);277-87
[PubMed:24445629] [] [DOI] (I p)