Article, Endocrinology

Association of blood glucose at admission with outcomes in patients treated with therapeutic hypothermia after cardiac arrest

a b s t r a c t

Background: It is well known that hyperglycemia is associated with poor outcomes in critically ill patients. We investigated the association between blood glucose level at admission and the outcomes of patients treated with Therapeutic hypothermia after cardiac arrest.

Methods: A total of 883 cardiac arrest patients who were treated with TH were analyzed from the Korean Hypothermia Network retrospective registry. We examined the association of blood glucose at admission with survival and neurologic outcomes at hospital discharge. Favorable neurologic outcomes were defined as Cerebral performance category scores of 1 and 2.

Results: The mean age of the sample was 56.7 +- 16.2 years, 69.5% of subjects were male, and the mean blood glucose at admission was 14.1 +- 7.0 mmol/L. After adjustment for sex, age, history of diabetes mellitus, hypertension, renal disease and stroke, time from arrest to return of spontaneous circulation, initial rhythm, witness status, bystander cardiopulmonary resuscitation, cause of arrest, and Cumulative dose of adrenaline, the associations between glucose and outcomes were as follows: for favorable neurologic outcomes, an odds ratio of 0.955 (95% confidence interval, 0.918-0.994); and for survival, an odds ratio of 0.974 (95% confidence interval, 0.952-0.996).

Conclusion: These results show that blood glucose level at admission is associated with survival and favorable neurologic outcomes at hospital discharge in patients treated with TH after cardiac arrest. Blood glucose level at admission could be a surrogate marker of ischemic insult severity during cardiac arrest. However, randomized, controlled evidence is needed to address the significance of tight glucose control during TH after cardiac arrest.

(C) 2014


Hyperglycemia is commonly seen in critically ill patients requiring intensive care. The exact mechanism of the development of hyperglycemia in critical illness is not fully understood. However, it has been proposed to be caused by the complex consequences of many factors, including increased cortisol, catecholamine, glucagon, growth hormone, gluconeogenesis, and glycogenolysis [1]. This interplay results in excessive hepatic glucose production and insulin resistance, leading to an elevation in blood glucose. In particular, ischemia-reperfusion injuries frequently result in metabolic derange- ments, such as hyperglycemia [2]. Hyperglycemia after cardiac arrest

* Corresponding author. Department of Emergency Medicine, College of Medicine, The Catholic University of Korea, Seoul St Mary’s Hospital, 222 Banpo-Daero, Seocho-Gu, Seoul, 137-701, Republic of Korea. Tel.: +82 2 2258 1988; fax: +82 2 2258 1997.

E-mail address: [email protected] (C.S. Youn).

is likely to be the result of an interaction between the patient’s characteristics and acute stress.

Focus on the association of hyperglycemia with outcomes in critically ill patients has generated interest in the importance of hyperglycemia in other Disease states, including clinical outcomes. Several studies of critically ill patients have demonstrated a strong association between hyperglycemic stress and poor clinical outcomes, including mortality, morbidity, length of stay, infections, and Overall complications [3-5]. However, this evidence does not prove that hyperglycemia causes poor clinical outcomes because hyperglycemia could merely be a marker of severe illness. In particular, the relationship of hyperglycemia with outcomes after cardiac arrest is not as well understood.

Several clinical trials have suggested that aggressive treatment of hyperglycemia could improve morbidity and mortality in critically ill patients [5,6]. However, recent studies have shown that tight glucose control through intensive insulin therapy might not necessarily improve outcomes in the critically ill subpopulation of cardiac arrest

0735-6757/(C) 2014

post-cardiac arrest care“>patients [7,8]. Recognizing the need for specific blood glucose control guidelines for post-cardiac arrest patients, the International Liaison Committee on Resuscitation recently recommended a more moderate target for blood glucose concentration of up to 144 mg/dL (8 mmol/L) [9]. Better understanding of blood glucose regulation after cardiac arrest would be an important step toward the creation of more specific glucose control guidelines to improve outcomes. Accordingly, our objective was to examine the association between blood glucose level at admission and outcomes in patients treated with Therapeutic hypothermia after cardiac arrest, with and without a history of preexisting diabetes.



This was a multicenter, retrospective, observational, registry- based study. The study was one of the first research projects using the Korean Hypothermia Network (KORHN) registry data. The KORHN, a multicenter clinical research consortium for TH in South Korea, was organized in 2011, and it conducted this multicenter, retrospective, registry project in 2012. The KORHN investigators collected post- cardiac arrest TH data from 24 teaching hospitals around South Korea from 2007 to 2012. Adults with (>=18 years old) out-of-hospital cardiac arrest (OHCA), treated with TH after return of spontaneous circulation (ROSC), were included in the study. Traumatic and Inhospital CArdiac arrest patients were excluded. The institutional review board of the each institution approved the study protocol before data collection. Informed consent was waived because of the retrospective nature of the study.

Postcardiac arrest care

Patients successfully resuscitated from cardiac arrest were indicated with post-cardiac arrest care. The KORHN post-cardiac arrest care protocol was made before starting retrospective registry project. The contents of KORHN post-cardiac arrest care protocol are as follows: inclusion and exclusion criteria of TH, method of induction, maintenance and rewarming of TH, indication of emergency coronary angiography, hemodynamic optimization, Ventilator management, glucose and seizure control, and finally neurologic prognostication. A target core temperature of 32?C to 34?C was recommended and maintained for 12 to 24 hours, followed by gradual rewarming. Emergency coronary angiography was recommended when ST- elevation myocardial infarction was suspected. Various clinical parameters were used to achieve hemodynamic optimization, ventilator management, and glucose management: SaO2 of 94% to 96%, PaCO2 of 35 to 45 mm Hg, mean arterial pressure of greater than or equal to 70 mm Hg, urine output of greater than or equal to

0.5 mL/kg per hour, and glucose of 144 to 180 mg/dL. However, our protocol was slightly transformed according to the clinical practice at the participating site, and, ultimately, treatment was made at the discretion of the attending physicians of individual site.

Data collection

The KORHN database contains demographics, resuscitation vari- ables, and postresuscitation variables. The following demographic and resuscitation variables were collected for each patient: age, sex, underlying disease (eg, hypertension, diabetes mellitus, coronary artery disease, renal disease, pulmonary disease, and stroke), cause of arrest, witnessed collapse, bystander cardiopulmonary resuscitation (CPR), initially presenting rhythm after arrest, resuscitation time, discharge status, and Cerebral Performance Category scale score at the time of hospital discharge. Definitions were based on Utstein style guidelines [10]. blood glucose measurement on admission was

defined as highest value of peripheral blood glucose checked within the 1 hour after ROSC. A hypoglycemic event was defined as blood glucose less than 4.4 mmol/L, checked during TH. Outcomes were assessed immediately before hospital discharge. Neurologic outcomes were categorized according to the Glasgow-Pittsburgh CPC and were dichotomized as favorable neurologic outcomes (CPC 1 and 2) or unfavorable neurologic outcomes (CPC 3-5).

The data form, standard definitions of 87 variables, and a registration manual were developed by literature review and a consensus of the study investigators. The registry data were collected by medical chart or electronic medical record reviews. The collected data from each hospital were verified for completeness by the principle investigator at each site and were recorded on a Web-based data registration system ( by the clinical research coordinator of the site. A data manager and 3 clinical research associates monitored and reviewed the Data quality regularly. The site principal investigators or site clinical research coordinators were contacted through the query function in the system or directly by telephone to clarify data.

Statistical analysis

Categorical variables are presented as counts and percentages and were compared with the ?2 test or Fisher exact test, when appropriate; continuous variables are presented as means +- SDs and were compared using Student t test. Multivariate binary logistic regression analysis was used to assess independent predictors of mortality and neurologic outcomes. All variables with a significance level less than 0.1 by univariate analysis were included in a multivariate logistic regression model. All of the statistical analyses were performed using SPSS software, version 17.0 (SPSS, Chicago, IL). P b .05 was considered statistically significant for all comparisons.


Of a total of 930 OHCA patients entered in the registry, 883 were included in the analysis, of whom 528 (59.8%) survived and were discharged from the hospital; 239 (26.3%) of these patients achieved favorable neurologic outcomes. The demographic data of the enrolled patients are summarized in Table 1. The mean age was 56.7 +- 16.2 years, and 69.5% of the subjects were male. One hundred ninety-eight patients had a history of diabetes, and the mean blood glucose at admission was 14.1 +- 7.0 mmol/L.

The relationships between variables and patient outcomes are presented in Table 2. Compared with nonsurvivors, the survivors were

Table 1

Baseline characteristics of the patients included in this study

N = 883

Male sex, n (%) 614 (69.5)

Age, y (mean +- SD) 56.7 +- 16.2

Premorbid disease, n (%)

Diabetes mellitus 198 (22.4)

Hypertension 301 (34.1)

Coronary artery disease 105 (11.9)

Renal disease 56 (6.3)

Stroke 44 (5.0)

Resuscitation variables

Witnessed, n (%) 594 (67.4)

Bystander CPR, n (%) 263 (31.1)

Shockable rhythm, n (%) 267 (31.3)

cardiac etiology, n (%) 538 (60.9)

Time from collapse to ROSC, min (mean +- SD) 34.1 +- 18.2 Cumulative adrenaline dose during resuscitation, mg (mean +- SD) 3.9 +- 4.0

Hypoglycemic event, n (%) 204 (23.1)

Glucose, mmol/L (mean +- SD) 14.1 +- 7.0

Table 2

Univariate logistic regression analysis for independent factors associated with outcomes

Table 3

Multivariate logistic regression analysis for independent factors associated with favorable neurologic outcomes and survival

For survival discharge For favorable neurologic


OR (95% CI) P OR (95% CI) P

Male sex 1.844 (1.379-2.466)


1.717 (1.217-2.423)


Age 0.976 (0.968-0.985)

Premorbid disease


0.958 (0.948-0.967)


Diabetes mellitus 0.597 (0.434-0.821)


0.426 (0.281-0.646)



0.846 (0.638-1.123)


0.561 (0.402-0.782)


Coronary artery disease

1.158 (0.760-1.763)


0.786 (0.505-1.222)


Renal disease

0.519 (0.301-0.896)


0.194 (0.069-0.542)



0.658 (0.359-1.207)


0.331 (0.129-0.851)


Resuscitation variables Witnessed

2.264 (1.698-3.018)


2.699 (1.876-3.884)


Bystander CPR

1.240 (0.920-1.671)


2.028 (1.475-2.787)


Shockable rhythm

3.971 (2.828-5.578)


8.869 (6.297-12.492)


Cardiac etiology

2.887 (2.180-3.824)


7.659 (4.986-11.765)


Time from collapse

0.979 (0.972-0.987)


0.961 (0.95-0.972)



Cumulative adrenaline

0.968 (0.934-1.003)


0.947 (0.901-0.994)


dose during resuscitation

Hypoglycemic event

0.754 (0.550-1.035)


0.561 (0.382-0.826)


Glucose, mmol/L

0.965 (0.945-0.986)


0.958 (0.934-0.983)


younger. The proportion of patients with diabetes mellitus was lower in the survivor group. Survivors had higher proportions of witnessed collapse, shockable rhythm, and Noncardiac etiology. The mean blood glucose level was significantly lower in the survival group (13.4 +- 6.5 vs 15.1 +- 7.6 mmol/L; P = .001), and the mean blood glucose level was significantly lower in the favorable neurologic outcome group (12.8 +- 5.0 vs 14.6 +- 7.6 mmol/L; P b .001) (Fig. 1).

Patients with favorable neurologic outcomes were significantly younger. Patients with diabetes mellitus, hypertension, and renal disease were more likely to have unfavorable neurologic outcomes. Favorable neurologic outcomes were significantly associated with resuscitation variables, such as higher proportions of witnessed collapse, shockable rhythm, bystander CPR, and cardiac etiology; a lower dose of adrenaline administered during resuscitation; and a shorter time from collapse to ROSC. Furthermore, patients with favorable neurologic outcomes had lower proportions of hypoglyce- mic events and lower blood glucose levels.

Fig. 1. Box and whiskers plot shows comparing blood glucose value between the groups. The ends of the box indicate the 25th and 75th percentiles values, and the middle line is the mean. The whiskers indicate their 95% confidence interval.

OR (95% CI) P

For survival at discharge Male sex

1.669 (1.195-2.332)



0.976 (0.966-0.986)


Diabetes mellitus

0.827 (0.558-1.227)


Renal disease

0.809 (0.423-1.548)



1.716 (1.227-2.399)


Shockable rhythm

2.363 (1.581-3.533)


Cardiac etiology

2.088 (1.481-2.944)


Time from collapse to ROSC

0.977 (0.969-0.986)


Glucose, mmol/L

0.974 (0.952-0.996)


For favorable neurologic outcome

Male sex

1.425 (0.855-2.375)



0.944 (0.929-0.960)


Diabetes mellitus

0.790 (0.410-1.524)



0.952 (0.544-1.667)


Renal disease

0.413 (0.109-1.565)



0.620 (0.154-2.506)



1.720 (1.006-2.938)


Bystander CPR

0.888 (0.537-1.468)


Shockable rhythm

2.545 (1.213-6.973)


Cardiac etiology

6.389 (3.379-12.081)


Time from collapse to ROSC

0.932 (0.913-0.952)


Cumulative adrenaline dose during resuscitation

1.047 (0.973-1.126)


Hypoglycemic event

0.521 (0.279-0.970)


Glucose, mmol/L

0.955 (0.918-0.994)


In the multivariate logistic analysis, the odds ratios (ORs) for surviving with a favorable neurologic outcome, adjusted for early glucose normalization and other confounding variables, are displayed in Table 3. After adjustment for sex, age, history of diabetes mellitus, hypertension, renal disease and stroke, time from arrest to ROSC, initial rhythm, witness status, bystander CPR, cause of arrest, and adrenaline, the associations between glucose level and outcomes were as follows: for favorable neurologic outcomes, the OR was 0.955 (95% confidence interval [CI], 0.918-0.994); and for survival, the OR was 0.974 (95% CI, 0.952-0.996) (Fig. 2). These findings could mean that there was 2.8% greater mortality for every 1 mmol/L of blood glucose and 4.5% more unfavorable neurologic outcomes.


The current study showed that blood glucose levels at admission were associated with survival and neurologic outcomes at hospital discharge in patients treated with TH after cardiac arrest. The subjects were enrolled from a retrospective, multicenter registry in South Korea.

Some retrospective studies have shown that higher glucose levels are associated with increased mortality and worse neurologic outcomes [8,11,12]. Of these studies, only 1 examined patients with TH [8], and this study described the relationship between glucose levels 12 hours after cardiac arrest and outcomes. Our study was the largest study to our knowledge evaluating the relationships between blood glucose levels on admission and the outcomes of patients with TH after cardiac arrest.

transient hyperglycemia during severe illness is currently called stress hyperglycemia [13]. According to the American Diabetes Association, these patients can be classified as follows: medical history of diabetes, unrecognized diabetes, and hospital-related hyperglycemia [14]. However, this classification requires hospital confirmation of diabetes, which was not available in this retrospective registry data. Therefore, we performed an analysis of the impact of blood glucose level at admission, regardless of history of diabetes.

High blood glucose levels at admission have been associated with aggravated ischemic brain injury in both diabetic and nonDiabetic patients and even in patients undergoing thrombolytic therapy

Fig. 2. Forest plot shows small circles for the adjusted ORs of survival at discharge and favorable neurologic outcomes after cardiac arrest treated with TH and horizontal lines for their 95% confidence interval.

[15,16]. Moreover, experimental research has shown that glucose exacerbated ischemic brain injury in models of both focal and global cerebral ischemia-reperfusion [17-20]. Increased oxidative stress and protein glycosylation are possible mechanisms. Another important factor is the blood insulin level because of its protective effect in Experimental models. Recently, Molnar et al [21] reported that hyperglycemia increased S-100? protein more than normoglycemia in a pig model of cardiac arrest without elevation of inflammatory cytokines. Thus, the exact mechanism by which hyperglycemia worsens ischemic brain injury is currently unknown.

Several explanations could exist for the associations between Elevated blood glucose levels at admission and unfavorable outcomes after cardiac arrest. Although the exact mechanism is currently unknown, hyperglycemia can be directly toxic and cause ischemic brain injury, or it can aggravate ischemic-reperfusion injury. Second, hyperglycemia could be a surrogate marker of the extent of ischemic insult during cardiac arrest. Third, patients with elevated glucose levels are relatively deficient in insulin. Fourth, patients with unrecognized diabetes might have been included in this study. A history of diabetes was associated with adverse outcomes after OHCA [22].

There were several limitations of this study. Diabetes mellitus was defined as known diabetic status on admission. Many cardiac arrest patients are known to have undetected DM, and they would not have been excluded from our study. A major limitation of this study was the lack of identification data to distinguish nondiabetes from undetected diabetes. We did not routinely measure HbA1c levels or test for diabetes during admission. Another limitation of our study was that although blood glucose levels at admission would have been responsive to the acute stress associated with cardiac arrest, many other factors, such as prior meals or diurnal variation, could have contributed to the variability in blood glucose levels. Finally, blood glucose level at admission was only investigated, and we could not identify the impact of glucose-lowering therapy, especially insulin, on patient outcomes because data on glucose levels during the course of TH were not available in our registry. The treatment strategy of high blood glucose was at the discretion of the attending physician or the protocol of the individual hospital. Larger, randomized trials are needed to investigate the relationship between blood glucose

lowering therapy and mortality among patients treated with TH after cardiac arrest.


These results showed that blood glucose level at admission was associated with survival and favorable neurologic outcomes at hospital discharge in patients treated with TH after cardiac arrest. Blood glucose level at admission could be a surrogate marker of ischemic insult severity during cardiac arrest. However, randomized, controlled evidence is needed to address the significance of tight glucose control during TH after cardiac arrest.


  1. McCowen KC, Malhotra A, Bistrian BR. Stress-induced hyperglycemia. Crit Care Clin 2001;17(1):107-24.
  2. Beiser DG, Carr GE, Edelson DP, et al. Derangements in blood glucose following Initial resuscitation from in-hospital cardiac arrest: a report from the national registry of cardiopulmonary resuscitation. Resuscitation 2009;80(6):624-30.
  3. Gale SC, Sicoutris C, Reilly PM, et al. Poor Glycemic control is associated with increased mortality in critically ill trauma patients. Am Surg 2007;73(5):454-60.
  4. Krinsley J, Grissler B. Intensive glycemic management in critically ill patients. Jt Comm J Qual Patient Saf 2005;31(6):308-12.
  5. Krinsley JS. Effect of an intensive glucose management protocol on the mortality of critically ill adult patients. Mayo Clin Proc 2004;79(8):992-1000.
  6. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med 2001;345(19):1359-67.
  7. Oksanen T, Skrifvars MB, Varpula T, et al. Strict versus moderate glucose control after resuscitation from ventricular fibrillation. Intensive Care Med 2007;33(12):2093-100.
  8. Losert H, Sterz F, Roine RO, et al. Strict normoglycaemic blood glucose levels in the therapeutic management of patients within 12 h after cardiac arrest might not be necessary. Resuscitation 2008;76(2):214-20.
  9. Nolan JP, Neumar RW, Adrie C, et al. Post-cardiac arrest syndrome: epidemiology, pathophysiology, treatment, and prognostication: a scientific statement from the International Liaison Committee on Resuscitation; the American Heart Association Emergency Cardiovascular Care Committee; the Council on Cardiovascular Surgery and Anesthesia; the Council on Cardiopulmonary, Perioperative, and Critical Care; the Council on Clinical Cardiology; the Council on Stroke (Part II). Int Emerg Nurs 2010;18(1):8-28.
  10. Jacobs I, Nadkarni V, Arrest tITFoC, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for Resuscitation Registries: a statement for healthcare professionals from a Task Force of the International Liaison Committee on Resuscitation (American Heart

    Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Councils of Southern Africa). Circulation 2004;110(21):3385-97.

    Nurmi J, Boyd J, Anttalainen N, et al. Early increase in blood glucose in patients resuscitated from out-of-hospital ventricular fibrillation predicts poor outcome. Diabetes Care 2012;35(3):510-2.

  11. Mullner M, Sterz F, Binder M, Schreiber W, Deimel A, Laggner AN. Blood glucose concentration after cardiopulmonary resuscitation influences functional neuro- logical recovery in human cardiac arrest survivors. J Cereb Blood Flow Metab 1997;17(4):430-6.
  12. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet 2009;373(9677):1798-807.
  13. Clement S, Braithwaite SS, Magee MF, et al. Management of diabetes and hyperglycemia in hospitals. Diabetes Care 2004;27(2):553-91.
  14. Desilles JP, Meseguer E, Labreuche J, et al. Diabetes mellitus, admission glucose, and outcomes after stroke thrombolysis: a registry and systematic review. Stroke 2013;44(7):1915-23.
  15. Capes SE, Hunt D, Malmberg K, Pathak P, Gerstein HC. Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke 2001;32(10):2426-32.
  16. Farrokhnia N, Ericsson A, Terent A, et al. MEK-inhibitor U0126 in hyperglycaemic focal ischaemic brain injury in the rat. Eur J Clin Invest 2008;38(9):679-85.
  17. Martin A, Rojas S, Chamorro A, et al. Why does Acute hyperglycemia worsen the outcome of transient focal cerebral ischemia? Role of corticosteroids, inflammation, and protein O-glycosylation. Stroke 2006;37(5):1288-95.
  18. Li PA, He QP, Yi-Bing O, et al. Phosphorylation of extracellular signal-regulated kinase after transient cerebral ischemia in hyperglycemic rats. Neurobiol Dis 2001;8(1):127-35.
  19. Li PA, Kristian T, Shamloo M, et al. Effects of preischemic hyperglycemia on brain damage incurred by rats subjected to 2.5 or 5 minutes of forebrain ischemia. Stroke 1996;27(9):1592-601.
  20. Molnar M, Bergquist M, Larsson A, et al. Hyperglycaemia increases S100beta after short Experimental cardiac arrest. Acta Anaesthesiol Scand 2014;58(1):106-13.
  21. Larsson M, Thoren AB, Herlitz J. A history of diabetes is associated with an adverse outcome among patients admitted to hospital alive after an out-of-hospital cardiac arrest. Resuscitation 2005;66(3):303-7.