Admission hyperglycemia is associated with poor outcome after emergent coronary bypass grafting surgery

    Admission Hyperglycemia is Associated with Poor Outcome after Emergent Coronary Bypass Grafting Surgery Robert H. Thiele M.D., Christoph Hucklenbruch M.D., Jennie Z. Ma PhD, Douglas Colquhoun M.D., Zhiyi Zuo M.D. PhD, Edward C. Nemergut M.D., Jacob Raphael M.D. PII: DOI: Reference:

S0883-9441(15)00464-5 doi: 10.1016/j.jcrc.2015.09.004 YJCRC 51941

To appear in:

Journal of Critical Care

Please cite this article as: Thiele Robert H., Hucklenbruch Christoph, Ma Jennie Z., Colquhoun Douglas, Zuo Zhiyi, Nemergut Edward C., Raphael Jacob, Admission Hyperglycemia is Associated with Poor Outcome after Emergent Coronary Bypass Grafting Surgery, Journal of Critical Care (2015), doi: 10.1016/j.jcrc.2015.09.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Admission Hyperglycemia is Associated with Poor Outcome

RI P

T

after Emergent Coronary Bypass Grafting Surgery

Robert H. Thiele1, Christoph Hucklenbruch1,3, Jennie Z. Ma2, Douglas

SC

Colquhoun1, Zhiyi Zuo1, Edward C. Nemergut1 and Jacob Raphael1.

Departments of Anesthesiology and 2Biostatistics and Epidemiology, University

NU

1

of Virginia Health System, Charlottesville, VA, USA 3

MA

Department of Anesthesiology, University of Muenster, Muenster, Germany.

AC

CE

PT

ED

Corresponding Author: Jacob Raphael, M.D. Associate Professor of Anesthesiology Department of Anesthesiology University of Virginia Health System PO Box 800710, Charlottesville, Virginia 22908 USA Email: [email protected] Tel: 434-924-9508 Fax: 434-982-0019 From the Department of Anesthesiology, University of Virginia Health System Charlottesville, Virginia 22908, USA

1

ACCEPTED MANUSCRIPT Abstract Purpose: Hyperglycemia during or after cardiac surgery is a common finding that

T

is associated with poor outcome. Very little data, however, is available regarding

RI P

a correlation between admission blood glucose and outcomes after coronary

SC

bypass surgery (CABG). Thus the goal of the current study was to examine the relationship between admission blood glucose and outcome after emergency

NU

CABG surgery.

Materials and Methods: a retrospective analysis to evaluate whether admission

MA

hyperglycemia is associated with increased morbidity or mortality was performed in patients after emergency CABG surgery. The records of all the patients

ED

undergoing emergency CABG surgery between January 1999 and December

PT

2010 at the University of Virginia Health System were reviewed. Postoperative inhospital mortality and complications were considered as study end points.

CE

Results: 240 patients met final inclusion criteria. Overall mortality was 14.1%.

AC

The median admission blood glucose in patients who died 7.4 (interquartile range, 5.9-10.1) mmol/L was significantly higher compared to survivors 6.1 (interquartile range, 5.4- 7.2), (P<0.01). Furthermore, 59% of the patients who died had admission blood glucose levels >6.6 mmol/L whereas only 35% of the patients who survived had similar blood glucose levels (P=0.01). On multivariable analysis, admission blood glucose was identified as an independent risk factor for death following emergency CABG (P=0.01; OR, 1.16; 95% CI, 1.04 - 1.29). Admission blood glucose was further identified as independently associated with increased risk for a composite outcome of death, postoperative renal failure or

2

ACCEPTED MANUSCRIPT stroke (P=0.01; OR, 1.14; 95% CI, 1.03 - 1.27). Conclusions: Our study shows for the first time that admission blood glucose is

RI P

T

correlated with increased morbidity and mortality among patients undergoing

AC

CE

PT

ED

MA

NU

SC

emergency CABG surgery.

3

ACCEPTED MANUSCRIPT Introduction

T

Hyperglycemia is commonly observed during and after cardiac surgery in

RI P

both diabetic and non-diabetic patients. The use of cardiopulmonary bypass (CPB) causes insulin resistance and further exacerbates the hyperglycemic

SC

response1. Both intraoperative and postoperative hyperglycemia have been associated with increased morbidity and mortality after coronary artery bypass

NU

grafting (CABG) surgery2-6. Although the treatment target for glucose remains

reduce

glucose

MA

controversial7-11, protocols that control hyperglycemia, avoid hypoglycemia, and variability

may

associated

with

improved

patient

ED

outcomes12,13.

be

In laboratory studies, hyperglycemia provokes numerous deleterious

PT

effects on myocardium that is subjected to ischemia and reperfusion. In both

CE

diabetic and hyperglycemic dogs, myocardial infarct size is strongly correlated with blood glucose concentration14. Hyperglycemia also abolishes ischemic14,15

AC

and anesthetic16,17 preconditioning and exacerbates reperfusion injury in a human cardiomyocyte

model

of

ischemia

and

reperfusion18.

Moreover,

since

hyperglycemia provokes coronary endothelial dysfunction 19,20, it may further increase the risk for myocardial ischemic events. In addition, diabetic patients that are treated with sulfonylurea-type anti-diabetic medications may also be more susceptible to myocardial ischemia and reperfusion injury due to the antagonistic effect of these drugs on the myocardial KATP channels - an important mediator of myocardial preconditioning21.

4

ACCEPTED MANUSCRIPT Given these data, the current American College of Cardiology/American Heart Association guidelines advise strict glucose control in patients hospitalized

23

RI P

critically ill patients

T

with acute myocardial infarction22 and similar recommendations also exist for . Furthermore, the Joint Commission on Accreditation of

SC

Healthcare Organizations includes postoperative glucose control for cardiac surgery patients as a core quality-of-care measure for all USA hospitals that

NU

participate in the Medicare program. The American Diabetes Association 24 and

MA

the American College of Endocrinology25,26 currently recommend target-driven glucose control in all hospitalized patients, regardless of the diagnosis. clinical

studies

that

investigated

a

relationship

between

ED

Most

hyperglycemia and outcome after CABG surgery, have focused on the

PT

intraoperative and postoperative periods, where, in theory, the clinician may have

CE

the greatest opportunity to impact outcome2,3,7,27. Much less is known, however, on the potential interaction between preoperative (or admission) hyperglycemia

AC

and outcome in patients undergoing CABG surgery28. Given the strong relationship between admission hyperglycemia and in-hospital mortality from acute myocardial infarction29-33, we sought to identify any interaction between admission blood glucose and morbidity or mortality in patients undergoing emergency CABG surgery with cardiopulmonary bypass (CPB). We chose to focus on patients undergoing emergency operations given their very high-risk nature, in addition to the fact that in emergency situations the physician may not always have an opportunity to effectively treat hyperglycemia before surgery.

5

ACCEPTED MANUSCRIPT

Materials and Methods

T

Study population:

RI P

The study was approved by the University of Virginia Institutional Review Board (IRB-HSR #14160). The requirement for an informed consent was waived

SC

by the IRB. We have retrospectively reviewed the records of all the patients that

NU

underwent emergency CABG surgery at the University of Virginia Health System between January 1999 and December 2010. Emergency CABG was defined

MA

based on the Society of Thoracic Surgeons national database definitions (STS National Database, Adult Cardiac Surgery Section, http://www.sts.org). By

ED

definition, these patients had ongoing ischemia with or without mechanical

PT

dysfunction with shock. Patients who underwent a salvage operation, defined as those undergoing cardiopulmonary resuscitation en route to the operating room

CE

or before anesthesia induction, were included in the study population. Patients

AC

undergoing combined valve-coronary procedures or combined aortic-coronary procedures, and patients that did not have admission blood glucose data (blood glucose level at admission to the hospital) were excluded from the study. Furthermore, given the fact that we anticipated that the use of cardiopulmonary bypass and aortic cross clamping may have an impact on outcome, patients undergoing off-pump coronary bypass surgery were also excluded from the analysis. All patients had a standard median sternotomy. After harvesting the internal mammary artery the patients were fully anticoagulated with heparin (300

6

ACCEPTED MANUSCRIPT U/kg) followed by supplemental doses of 100 U/kg to maintain an activated clotting time greater than 480 seconds. Upon initiation of cardiopulmonary bypass

RI P

T

all patients were allowed to cool to a target core temperature of 34°C. At surgeon discretion, anterograde and/or retrograde high-potassium blood cardioplegia was

SC

used. Intra aortic balloon pump (IABP) was employed to assist in separation from CPB when pharmacologic hemodynamic support was inadequate. Induction and

NU

maintenance of anesthesia was based on the patient’s hemodynamic status and the attending anesthesiologist’s discretion. Insulin was administered based on

MA

serum glucose measurements per the anesthesiologist’s discretion.

ED

For each patient, an extensive set of perioperative data, including demographics, preoperative risk factors and co-morbidities, preoperative

PT

laboratory results (including admission blood glucose), intraoperative variables,

CE

postoperative outcomes and complications were collected prospectively through the registry maintained by the Society of Thoracic Surgeons and the STS

AC

predictive score for mortality calculated. Additional data was obtained through a review of the patients’ medical record. All data was further verified by reviewing the relevant patient perioperative information in the STS national database by two independent investigators (R.H.T and C.H.) prior to analysis. Admission blood glucose was defined as the first blood glucose concentration recorded at hospital admission. If a patient was transferred to our institution from another hospital, admission blood glucose was defined as the first blood glucose concentration recorded at admission to the outside hospital. Three-day blood glucose (3BG) is the arithmetic mean of the maximal blood

7

ACCEPTED MANUSCRIPT glucose levels recorded during the first three postoperative days for each patient. This glucose measurement has been previously reported to be strongly 34

. All glucose

RI P

T

associated with post-operative outcome after cardiac surgery

measurements that were included in the analysis were performed by arterial

SC

blood gas analyzers or the hospital’s central laboratory using the hexokinase

MA

analyzers has been well validated36.

NU

method 35. The accuracy of blood glucose measurements using arterial blood gas

Study end points:

ED

The primary end point was postoperative in-hospital mortality. Secondary endpoints included postoperative myocardial infarction, stroke, postoperative

PT

kidney injury (either new-onset or deteriorating renal dysfunction), ventilator time

CE

greater than 48 hours, new atrial fibrillation, length of postoperative ICU stay, postoperative hospital stay and a composite of death, postoperative stroke or

AC

postoperative kidney injury. Postoperative kidney injury was defined according to the Acute Kidney Injury Network classification37. Postoperative myocardial infarction was defined based on the third universal definition of myocardial infarction

38

. Definitions for other various were standard definitions as set by the

STS (Table 1).

8

ACCEPTED MANUSCRIPT Statistical analysis: Hyperglycemia on admission was defined as glucose level above 6.6

RI P

T

mmol/L 39-41. We used descriptive statistics to delineate the characteristics of the cohort by hyperglycemic status. Normally distributed continuous variables were

SC

compared using the unpaired t test and non-normally distributed variables were evaluated with the Wilcoxon rank test. Categorical variables were compared

NU

using the chi-square test and Fisher's exact test as appropriate.

MA

A univariable logistic regression followed by a multiple logistic regression model were performed in order to investigate the effect of admission glucose on

ED

the study outcomes. Other covariates included age (in 10 years increments),

PT

gender, pre-operative diagnosis of diabetes mellitus, preoperative diagnosis of renal failure, cardiogenic shock on time of admission to the operating room,

CE

cardiopulmonary bypass time, aortic cross clamp time (in 10 minutes increments)

AC

and the three-day blood glucose (3BG). To further assess the predictability of admission glucose on postoperative mortality, a receiver operator characteristic (ROC) curve for admission glucose level was plotted. The 95% confidence intervals and the area under the curve (AUC) were calculated. In addition, a comparison was made with other independent risk factors (age, chronic renal failure and CPB time). Furthermore, the survival probability in patients with admission blood glucose above or below 6.6 mmol/L was estimated by the Kaplan-Meier method and the survival curves between the two groups were compared by the Log-rank test. Data are expressed as mean ± standard deviation for normally distributed parameters or median (interquartile range) for 9

ACCEPTED MANUSCRIPT non-normally distributed parameters. A P value < 0.05 was considered significant with a two-sided test. All the statistical analyses were performed using SAS (SAS

Institute

SC NU MA ED PT CE AC

10

Inc., Cary,

T

9.1

RI P

Statistical Software for Windows

NC).

ACCEPTED MANUSCRIPT Results:

T

We identified 294 patients that underwent emergency coronary artery

RI P

bypass grafting between January 1999 and December 2010. Among them, 35 patients underwent off-pump CABG (n=10) or a combined CABG-valve

SC

procedure (n=25) and 19 patients did not have admission blood glucose data.

NU

Thus, 240 patients were included in the final analysis. Admission glucose levels are presented in Figure 1. Patient demographics, co-morbidities and preoperative

MA

clinical and laboratory characteristics are summarized in Table 2. All patients had an admission diagnosis of acute coronary syndrome, of which 177 (73.8%)

ED

presented with ST-elevation myocardial infarction (STEMI). Eighty patients

were not diabetics.

PT

(33.3%) had a diagnosis of diabetes mellitus, whereas, 160 (66.7%) patients

CE

Overall postoperative in-hospital mortality was 14.1% (34 patients: 13

AC

diabetics and 21 non diabetics). Admission blood glucose level in the patients who died was significantly higher than in the survivors 7.4 (interquartile range, 5.9-10.1) mmol/L vs. 6.1 (interquartile range, 5.4-7.2), mmol/L, (P<0.01, Figure 2). Furthermore, 59% (20 patients out of 34) of the patients who died had admission blood glucose levels higher than 6.6 mmol/L, whereas only 35% (72 patients out of 206) of the patients in the survival group had admission blood glucose levels higher than 6.6 mmol/L (P=0.01). Other patient preoperative characteristics associated with mortality included female gender (P=0.02), increasing age (P=0.03), diabetes mellitus (P=0.02), a history of congestive heart failure (CHF, P=0.03), cardiogenic shock on admission to the operating room 11

ACCEPTED MANUSCRIPT (P=0.01), the need for hemodynamic support by intra-aortic balloon pump (IABP, P=0.03), preoperative diagnosis of renal failure (P<0.01), and lower hematocrit

RI P

T

(P<0.01). We did not find a correlation between cardiogenic shock on presentation and admission hyperglycemia (data not shown). Furthermore, on 13

, a tight

SC

January 1, 2002, following the publication of van den Berghe et al

glucose control protocol (target glucose 4.4-6.1 mmol/L) was adopted by all

NU

surgical intensive care units in our institution. Despite this change in patient

MA

management, the overall mortality rate, in our cohort, was not different before or after January of 2002, neither in diabetics nor in non-diabetics or the entire

ED

patient cohort. Since preoperative diagnosis of diabetes was found to be associated with increased mortality, we have decided to test whether the diabetic

PT

patient group had an undue influence on mortality. Thus, we have performed a

CE

repeated analysis in the non-diabetic patient group only. Our findings were similar to the analysis of the complete patient cohort. In the non-diabetic patients,

AC

those who died had significantly higher admission glucose levels compared to those who survived 7.3 (interquartile range, 5.4-9.6) mmol/L vs. 5.8 (interquartile range, 5.2-7.1), mmol/L (P=0.01). Furthermore, 12 out of the 21 non-diabetic patients who died (57.1%) had admission blood glucose greater than 6.6 mmol/L whereas only 46 out of the 139 non-diabetics who survived (33.1%) had admission blood glucose level higher than 6.6 mmol/L (P=0.02). Intraoperative variables are summarized in Table 3. There was no difference in CPB time (90±7 minutes in the patients who died vs. 84±5 minutes in the survivors, P=0.07), aortic cross-clamp time (66±4 minutes in the patients

12

ACCEPTED MANUSCRIPT who died vs. 63±5 minutes in those who survived, P=0.16) or in the number of grafts (2.9±0.2 vs. 2.8±0.3, P=0.34).

RI P

T

Postoperative patient characteristics and the incidence of postoperative complications in patients who died vs. those who survived are summarized in

SC

Table 4. The 3BG values were higher in the patients who died compared to those who survived: 8.9 (interquartile range, 7.1-10.2) mmol/L vs. 7.5 (interquartile

NU

range, 6.8-8.4) mmol/L, P<0.01). In addition, patients who died suffered from a

MA

higher incidence of postoperative kidney injury (P<0.01), postoperative stroke (P<0.01), the need for red blood cells (RBC) transfusions (P=0.02) and

ED

ventilation time > 48 hours (P<0.01).

On multivariable analysis (Table 5), admission blood glucose was

PT

identified as an independent risk factor for mortality after emergency CABG

CE

(P=0.01; OR, 1.16; 95% CI, 1.04 – 1.29), with an increase in the risk for mortality of 16% for every 0.55 mmol/L increase in blood glucose levels above 6.6 mmol/L.

AC

Other factors associated with an increased risk for mortality included: cardiopulmonary bypass time (P<0.01; OR, 1.21; 95% CI, 1.11-1.31), preoperative diagnosis of renal failure (P=0.02; OR, 3.51; 95% CI, 1.18-10.39), age (P=0.04; OR, 1.58; 95% CI, 1.0-2.5) and 3BG (P<0.01; OR, 1.31; 95% CI, 1.1 – 1.56). Figure 3 demonstrates the relationship between sensitivity and specificity in determining the predictive value of admission blood glucose and the other independent risk factors for postoperative in-hospital mortality. The area under the curve (AUC) for admission glucose was 0.65 (95% CI 0.54-0.77), while the

13

ACCEPTED MANUSCRIPT AUC for age, preoperative renal failure and CPB time was 0.62 (95% CI 0.530.72), 0.64 (95% CI 0.55-0.73) and 0.81(95% CI 0.74-0.88), respectively. Kaplan-

T

Meier curves (Figure 4) plotting postoperative in-hospital mortality over time

RI P

showed a significant survival benefit to patients with admission glucose ≤ 6.6

SC

mmol/L compared to those with admission glucose >6.6 mmol/L (P=0.03). Patients who experienced at least one event of the composite outcome

NU

had a higher admission blood glucose level compared to those who did not

MA

experience any of the composite events 7.8 (interquartile range, 6.3-9.4] mmol/L vs. 6.2 (interquartile range, 4.8-7.5) mmol/L, P=0.02), as well as higher 3BG

ED

levels 8.6 (interquartile range, 7.4-10.6) mmol/L vs. 6.7 (interquartile range, 5.68) mmol/L, (P=0.03).

PT

On multivariable analysis (Table 6), admission blood glucose (P=0.02;

CE

OR, 1.24; 95% CI, 1.07-1.27), 3BG (P=0.01; OR, 1.19; 95% CI, 1.05-1.36), preoperative diagnosis of renal failure (P<0.01; OR, 2.72; 95% CI, 6.18-50.81)

AC

and cardiopulmonary bypass time (P<0.01; OR 1.16; 95% CI, 1.09-1.24) were found as independent predictors of increased risk for experiencing one or more of the composite events.

14

ACCEPTED MANUSCRIPT

Discussion

T

Our data suggest that there is a significant association between admission

RI P

glucose and both morbidity and mortality in patients undergoing emergency CABG surgery with cardiopulmonary bypass. This association is independent of risk

factors

such

as

age,

preoperative

SC

traditional

renal

failure

and

NU

cardiopulmonary bypass time. Admission blood glucose is also strongly correlated with the likelihood of experiencing at least one event of the composite

MA

of postoperative in-hospital mortality, stroke or acute kidney injury. This is an important and unique finding as most investigators have focused on the

ED

association between intraoperative or postoperative hyperglycemia and mortality after CABG surgery2,34,42,43. In addition, most therapeutic interventions have

PT

implemented changes in intraoperative3,7,12 or postoperative care in the ICU13.

7-9,11,27

. Moreover, none of these studies evaluated the impact of

AC

controversial

CE

However, despite extensive research, the treatment target for glucose remains

admission blood glucose on postoperative outcome. Our findings suggest that a single admission blood glucose level > 6.6 mmol/L is associated with increased mortality. Hence, admission blood glucose could predict outcome after emergency coronary artery bypass grafting. Our data also demonstrates that the predictive power of admission blood glucose for mortality in this setting is similar to other independent preoperative risk factors for cardiac surgery patients. In a large cohort of 8648 CABG patients, Ranucci et al have reported that age, preoperative renal failure and preoperative left ventricular ejection fraction were found to be the most significant factors in predicting 15

ACCEPTED MANUSCRIPT mortality after cardiac surgery. The authors reported an AUC of 0.66, 0.57 and 0.65 for age, left ventricular ejection fraction and serum creatinine, respectively 44.

RI P

T

These findings are very similar to the AUC that are reported in this study. Our results are also in agreement with studies that have found analogous 39-41,45

. Nevertheless, this study was not aimed to

SC

findings in the face of acute MI

determine the optimal target blood glucose level during or after cardiac surgery.

NU

Although we found that admission blood glucose level > 6.6 mmol/L was clearly

MA

associated with increased post-operative morbidity and mortality, there is still controversy regarding the optimal blood glucose level during cardiac operations.

therapy

ED

While several studies advocate tight glucose control using intensive insulin 13,43,46,47

, other groups have shown that this approach is associated with

PT

a higher rate of complications and mortality, hence, more moderate approach

CE

may be superior 7-11,27. Although the clinical benefits of tight glucose control using intensive insulin treatment protocols are still under debate, current STS

mmol/L48.

AC

recommendations are to maintain perioperative glucose levels below 10

Our findings do not correlate with several trials that established a correlation between perioperative glucose levels and postoperative outcome2-4. However, the major significant difference between the current study and these trials is that we have focused on admission blood glucose as a predictor of morbidity and mortality, while the other studies did not evaluate the impact of this parameter on postoperative outcomes.

16

ACCEPTED MANUSCRIPT On January 1, 2002, following the publication by van den Berghe et al13, a tight glucose control protocol (target glucose 4.4-6.1 mmol/L) was adopted by all

RI P

T

surgical intensive care units in our institution. Although the approach to postoperative glucose control has changed, the mortality rate in our high-risk

SC

cohort was not different before or after the initiation of the tight glucose control protocol. While this does not mean that treatment of elevated glucose with insulin

NU

does not impact outcome, it may further emphasize admission blood glucose as

MA

a powerful predictor of morbidity and mortality after emergency CABG surgery. The overall mortality in our study of 14.1% was not unexpected given the

ED

very high-risk patient population in our cohort. In addition, several other groups

CE

CABG surgeries49-51

PT

recently reported similar mortality rates for patients undergoing emergency

AC

Is glucose a marker of illness severity or a cause of adverse outcome? Glucose may be a biomarker that reflects illness severity in critically ill patients. Hyperglycemia is frequently observed during acute illness or injury (known as “stress diabetes”). This is at least in part due to elevated hepatic glucose production, release of counter regulatory hormones, and peripheral insulin resistance – all of which can exacerbate the level of hyperglycemia 46,52,53. In the context of critical illness, hyperglycemia causes mitochondrial dysfunction and disturbances in neuronal, endothelial, and immune function, leading to endorgan injury, prolonged mechanical ventilation, and sepsis54,55. Furthermore, impaired myocardial performance following acute myocardial ischemia may result 17

ACCEPTED MANUSCRIPT in activation of compensatory mechanisms, including activation of the sympathetic nervous system, which can further contribute to development of

T

hyperglycemia56,57. As such, the association between admission glucose and

RI P

increased morbidity and mortality may simply reflect the tendency among “sicker”

SC

patients to experience hyperglycemia.

Alternatively, hyperglycemia may directly contribute to organ injury and

NU

dysfunction by several independent mechanisms. First, hyperglycemia inhibits

MA

production of nitric oxide and increases the production of reactive oxygen species in endothelial and vascular smooth muscle cells, thus impairing endothelial Second,

hyperglycemia

may

be

the

result

of

insulin

ED

function58-60.

deficiency/resistance, which is associated with increased lipolysis and excess

PT

circulating free fatty acids. These are highly toxic to ischemic myocardium and

CE

may cause damage to the cardiomyocyte and endothelial cell membranes resulting in calcium overload, and arrhythmias61. Finally, hyperglycemic patients

AC

also have impaired platelet function characterized by increased levels of plasminogen activator inhibitor-1 and adhesion molecules62,63. This enhances platelet adhesiveness and hyperaggregability and could predispose to coronary thrombosis and graft failure, which can lead to myocardial ischemia.

18

ACCEPTED MANUSCRIPT Study limitations.

T

Our study consists of several limitations: First, it is retrospective, which

RI P

introduces potential differences in baseline patient population characteristics, potential differences in medical practice and potential biases in regard to

SC

treatments.

NU

Second, we used admission blood glucose values, which might not have been taken in the fasting state. We assumed, however, that non-fasting patients

MA

have been evenly distributed among patients who survived and those who died. Third, hemoglobin A1C data was missing in most of the non-diabetic

ED

patients in our study population, since the measurement of glycosylated

PT

hemoglobin in non-diabetics was not the routine practice during the study period. Therefore, we were unable to comment on the differential impact of chronic vs.

AC

CE

acutely elevated blood glucose levels.

This study was not designed to evaluate the optimal target blood glucose levels for cardiac surgical patients and indeed different glucose targets may be indicated for different patient populations (i.e. diabetics vs. non-diabetics, emergency vs. elective). Further, while in our institution algorithm-based preoperative continuous insulin therapy, to maintain perioperative glucose levels between 6.7-8.7 mmol/L, has been recently implemented, the clinician should not interpret our results to suggest that preoperative hyperglycemia should be rapidly

19

ACCEPTED MANUSCRIPT corrected before CABG surgery. Such recommendations can only come from carefully designed, properly powered, randomized, controlled, trials.

RI P

T

In summary, in our study population of patients that underwent emergency CABG surgery, admission blood glucose >6.6 mmol/L was found to be a strong

SC

predictor of increased postoperative in-hospital mortality and morbidity. These findings call attention to a new prognostic marker that could be used for early risk

NU

stratification and management of patients that require emergency CABG. Further

MA

research is needed to determine whether hyperglycemia on admission is a marker reflecting illness severity or a direct cause of adverse outcome, and

AC

CE

PT

these high-risk patients.

ED

whether new glucose level targets would confer benefit and improve survival in

20

ACCEPTED MANUSCRIPT References:

AC

CE

PT

ED

MA

NU

SC

RI P

T

1. Knapik P, Nadziakiewicz P, Urbanska E, Saucha W, Herdynska M, Zembala M. Cardiopulmonary bypass increases postoperative glycemia and insulin consumption after coronary surgery. Ann Thorac Surg 2009;87:1859-65. 2. Doenst T, Wijeysundera D, Karkouti K, et al. Hyperglycemia during cardiopulmonary bypass is an independent risk factor for mortality in patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 2005;130:1144. 3. Ouattara A, Lecomte P, Le Manach Y, et al. Poor intraoperative blood glucose control is associated with a worsened hospital outcome after cardiac surgery in diabetic patients. Anesthesiology 2005;103:687-94. 4. Jones KW, Cain AS, Mitchell JH, et al. Hyperglycemia predicts mortality after CABG: postoperative hyperglycemia predicts dramatic increases in mortality after coronary artery bypass graft surgery. Journal of diabetes and its complications 2008;22:365-70. 5. Haga KK, McClymont KL, Clarke S, et al. The effect of tight glycaemic control, during and after cardiac surgery, on patient mortality and morbidity: A systematic review and meta-analysis. Journal of cardiothoracic surgery 2011;6:3. 6. Giakoumidakis K, Nenekidis I, Brokalaki H. The correlation between perioperative hyperglycemia and mortality in cardiac surgery patients: a systematic review. European journal of cardiovascular nursing : journal of the Working Group on Cardiovascular Nursing of the European Society of Cardiology 2012;11:105-13. 7. Gandhi GY, Nuttall GA, Abel MD, et al. Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery: a randomized trial. Annals of internal medicine 2007;146:233-43. 8. Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. The New England journal of medicine 2009;360:1283-97. 9. Bhamidipati CM, LaPar DJ, Stukenborg GJ, et al. Superiority of moderate control of hyperglycemia to tight control in patients undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 2011;141:543-51. 10. LaPar DJ, Isbell JM, Kern JA, Ailawadi G, Kron IL. Surgical Care Improvement Project measure for postoperative glucose control should not be used as a measure of quality after cardiac surgery. J Thorac Cardiovasc Surg 2014;147:1041-8. 11. Desai SP, Henry LL, Holmes SD, et al. Strict versus liberal target range for perioperative glucose in patients undergoing coronary artery bypass grafting: a prospective randomized controlled trial. J Thorac Cardiovasc Surg 2012;143:318-25. 12. Lazar HL, Chipkin SR, Fitzgerald CA, Bao Y, Cabral H, Apstein CS. Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation 2004;109:1497-502. 13. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. The New England journal of medicine 2001;345:1359-67. 14. Kersten JR, Toller WG, Gross ER, Pagel PS, Warltier DC. Diabetes abolishes ischemic preconditioning: role of glucose, insulin, and osmolality. Am J Physiol Heart Circ Physiol 2000;278:H1218-24.

21

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

T

15. Kersten JR, Schmeling TJ, Orth KG, Pagel PS, Warltier DC. Acute hyperglycemia abolishes ischemic preconditioning in vivo. Am J Physiol 1998;275:H721-5. 16. Kehl F, Krolikowski JG, Mraovic B, Pagel PS, Warltier DC, Kersten JR. Hyperglycemia prevents isoflurane-induced preconditioning against myocardial infarction. Anesthesiology 2002;96:183-8. 17. Raphael J, Gozal Y, Navot N, Zuo Z. Activation of Adenosine Triphosphateregulated Potassium Channels during Reperfusion Restores Isoflurane Postconditioninginduced Cardiac Protection in Acutely Hyperglycemic Rabbits. Anesthesiology 2015;122:1299-311. 18. Verma S, Maitland A, Weisel RD, et al. Hyperglycemia exaggerates ischemiareperfusion-induced cardiomyocyte injury: reversal with endothelin antagonism. J Thorac Cardiovasc Surg 2002;123:1120-4. 19. Gross ER, LaDisa JF, Jr., Weihrauch D, et al. Reactive oxygen species modulate coronary wall shear stress and endothelial function during hyperglycemia. Am J Physiol Heart Circ Physiol 2003;284:H1552-9. 20. Koltai MZ, Hadhazy P, Posa I, et al. Characteristics of coronary endothelial dysfunction in experimental diabetes. Cardiovasc Res 1997;34:157-63. 21. Jorgensen CH, Gislason GH, Bretler D, et al. Glyburide increases risk in patients with diabetes mellitus after emergent percutaneous intervention for myocardial infarction--a nationwide study. International journal of cardiology 2011;152:327-31. 22. Jneid H, Anderson JL, Wright RS, et al. 2012 ACCF/AHA focused update of the guideline for the management of patients with unstable angina/non-ST-elevation myocardial infarction (updating the 2007 guideline and replacing the 2011 focused update): a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Journal of the American College of Cardiology 2012;60:645-81. 23. Finfer S, Wernerman J, Preiser JC, et al. Clinical review: Consensus recommendations on measurement of blood glucose and reporting glycemic control in critically ill adults. Critical care 2013;17:229. 24. American Diabetes A. Standards of medical care in diabetes--2013. Diabetes care 2013;36 Suppl 1:S11-66. 25. Umpierrez GE, Hellman R, Korytkowski MT, et al. Management of hyperglycemia in hospitalized patients in non-critical care setting: an endocrine society clinical practice guideline. The Journal of clinical endocrinology and metabolism 2012;97:16-38. 26. Handelsman Y, Bloomgarden ZT, Grunberger G, et al. American association of clinical endocrinologists and american college of endocrinology - clinical practice guidelines for developing a diabetes mellitus comprehensive care plan - 2015. Endocrine practice : official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists 2015;21:1-87. 27. Duncan AE, Abd-Elsayed A, Maheshwari A, Xu M, Soltesz E, Koch CG. Role of intraoperative and postoperative blood glucose concentrations in predicting outcomes after cardiac surgery. Anesthesiology 2010;112:860-71.

22

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

T

28. Zindrou D, Taylor KM, Bagger JP. Admission plasma glucose: an independent risk factor in nondiabetic women after coronary artery bypass grafting. Diabetes care 2001;24:1634-9. 29. Hoebers LP, Damman P, Claessen BE, et al. Predictive value of plasma glucose level on admission for short and long term mortality in patients with ST-elevation myocardial infarction treated with primary percutaneous coronary intervention. Am J Cardiol 2012;109:53-9. 30. Kosiborod M, Inzucchi SE, Krumholz HM, et al. Glucose normalization and outcomes in patients with acute myocardial infarction. Arch Intern Med 2009;169:43846. 31. Stranders I, Diamant M, van Gelder RE, et al. Admission blood glucose level as risk indicator of death after myocardial infarction in patients with and without diabetes mellitus. Arch Intern Med 2004;164:982-8. 32. Ekmekci A, Cicek G, Uluganyan M, et al. Admission hyperglycemia predicts inhospital mortality and major adverse cardiac events after primary percutaneous coronary intervention in patients without diabetes mellitus. Angiology 2014;65:154-9. 33. Mudespacher D, Radovanovic D, Camenzind E, et al. Admission glycaemia and outcome in patients with acute coronary syndrome. Diab Vasc Dis Res 2007;4:346-52. 34. Furnary AP, Wu Y. Clinical effects of hyperglycemia in the cardiac surgery population: the Portland Diabetic Project. Endocrine practice : official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists 2006;12 Suppl 3:22-6. 35. Bondar RJ, Mead DC. Evaluation of glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides in the hexokinase method for determining glucose in serum. Clinical chemistry 1974;20:586-90. 36. Claerhout H, De Prins M, Mesotten D, et al. Performance of strip-based glucose meters and cassette-based blood gas analyzer for monitoring glucose levels in a surgical intensive care setting. Clinical chemistry and laboratory medicine : CCLM / FESCC 2015. 37. Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Critical care 2007;11:R31. 38. Thygesen K, Alpert JS, Jaffe AS, et al. Third universal definition of myocardial infarction. Circulation 2012;126:2020-35. 39. Diaz R, Goyal A, Mehta SR, et al. Glucose-insulin-potassium therapy in patients with ST-segment elevation myocardial infarction. JAMA 2007;298:2399-405. 40. Kosiborod M. Blood glucose and its prognostic implications in patients hospitalised with acute myocardial infarction. Diab Vasc Dis Res 2008;5:269-75. 41. Deedwania P, Kosiborod M, Barrett E, et al. Hyperglycemia and acute coronary syndrome: a scientific statement from the American Heart Association Diabetes Committee of the Council on Nutrition, Physical Activity, and Metabolism. Circulation 2008;117:1610-9. 42. Gandhi GY, Nuttall GA, Abel MD, et al. Intraoperative hyperglycemia and perioperative outcomes in cardiac surgery patients. Mayo Clinic proceedings 2005;80:862-6.

23

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

T

43. Zerr KJ, Furnary AP, Grunkemeier GL, Bookin S, Kanhere V, Starr A. Glucose control lowers the risk of wound infection in diabetics after open heart operations. Ann Thorac Surg 1997;63:356-61. 44. Ranucci M, Castelvecchio S, Menicanti L, Frigiola A, Pelissero G. Risk of assessing mortality risk in elective cardiac operations: age, creatinine, ejection fraction, and the law of parsimony. Circulation 2009;119:3053-61. 45. Kosiborod M, Rathore SS, Inzucchi SE, et al. Admission glucose and mortality in elderly patients hospitalized with acute myocardial infarction: implications for patients with and without recognized diabetes. Circulation 2005;111:3078-86. 46. Van den Berghe G. How does blood glucose control with insulin save lives in intensive care? The Journal of clinical investigation 2004;114:1187-95. 47. Blaha J, Mraz M, Kopecky P, et al. Perioperative tight glucose control reduces postoperative adverse events in non-diabetic cardiac surgery patients. The Journal of clinical endocrinology and metabolism 2015:jc20151959. 48. Lazar HL, McDonnell M, Chipkin SR, et al. The Society of Thoracic Surgeons practice guideline series: Blood glucose management during adult cardiac surgery. Ann Thorac Surg 2009;87:663-9. 49. Biancari F, Onorati F, Rubino AS, et al. Outcome of emergency coronary artery bypass grafting. Journal of cardiothoracic and vascular anesthesia 2015;29:275-82. 50. Khan AN, Sabbagh S, Ittaman S, et al. Outcome of early revascularization surgery in patients with ST-elevation myocardial infarction. Journal of interventional cardiology 2015;28:14-23. 51. Khaladj N, Bobylev D, Peterss S, et al. Immediate surgical coronary revascularisation in patients presenting with acute myocardial infarction. Journal of cardiothoracic surgery 2013;8:167. 52. Mesotten D, Wouters PJ, Peeters RP, et al. Regulation of the somatotropic axis by intensive insulin therapy during protracted critical illness. The Journal of clinical endocrinology and metabolism 2004;89:3105-13. 53. Vanhorebeek I, Langouche L. Molecular mechanisms behind clinical benefits of intensive insulin therapy during critical illness: glucose versus insulin. Best practice & research Clinical anaesthesiology 2009;23:449-59. 54. Vanhorebeek I, Langouche L, Van den Berghe G. Tight blood glucose control: what is the evidence? Critical care medicine 2007;35:S496-502. 55. Thiessen S, Vanhorebeek I, Van den Berghe G. Glycemic control and outcome related to cardiopulmonary bypass. Best practice & research Clinical anaesthesiology 2015;29:177-87. 56. Capes SE, Hunt D, Malmberg K, Gerstein HC. Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet 2000;355:773-8. 57. Oswald GA, Smith CC, Betteridge DJ, Yudkin JS. Determinants and importance of stress hyperglycaemia in non-diabetic patients with myocardial infarction. Br Med J (Clin Res Ed) 1986;293:917-22. 58. Raphael J, Gozal Y, Navot N, Zuo Z. Hyperglycemia inhibits anesthetic-induced postconditioning in the rabbit heart via modulation of phosphatidylinositol-3-kinase/Akt and endothelial nitric oxide synthase signaling. J Cardiovasc Pharmacol 2010;55:348-57.

24

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

T

59. Title LM, Cummings PM, Giddens K, Nassar BA. Oral glucose loading acutely attenuates endothelium-dependent vasodilation in healthy adults without diabetes: an effect prevented by vitamins C and E. Journal of the American College of Cardiology 2000;36:2185-91. 60. Williams SB, Goldfine AB, Timimi FK, et al. Acute hyperglycemia attenuates endothelium-dependent vasodilation in humans in vivo. Circulation 1998;97:1695-701. 61. Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and arrhythmias. Lancet 1994;343:155-8. 62. Davi G, Catalano I, Averna M, et al. Thromboxane biosynthesis and platelet function in type II diabetes mellitus. The New England journal of medicine 1990;322:1769-74. 63. Marfella R, Esposito K, Giunta R, et al. Circulating adhesion molecules in humans: role of hyperglycemia and hyperinsulinemia. Circulation 2000;101:2247-51.

25

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

26

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

27

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

28

AC

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

29

ACCEPTED MANUSCRIPT

Table 1. Definitions of key outcomes Definition

Postoperative inhospital mortality

Patients who die during the hospitalization in which the operation was performed, even if it is more than 30 days after the operation

Postoperative MI

Elevation of cardiac biomarker values > 10Xthe 99th percentile (in patients with normal preoperative cTN levels) in addition to either (i) new pathological Q waves or new LBBB, or (ii) angiographycally documented new graft or new native coronary artery occlusion, or (iii) imaging evidence of new loss of viable myocardium or new regional motion abnormality

Postoperative kidney injury

An increase in serum creatinine level to ≥ 50% or ≥ 0.3 mg/dL from the baseline level or a new requirement for dialysis.

Stroke

Postoperative central neurologic deficit persisting > 72 hours

Atrial fibrillation

New onset of atrial fibrillation/flutter requiring treatment; Does not include recurrence of atrial fibrillation/flutter that was present preoperatively

AC

CE

PT

ED

MA

NU

SC

RI P

T

Outcome

Ventilation > 48 hours

Pulmonary insufficiency requiring ventilatory support that includes (but is not limited to) causes such as acute respiratory distress syndrome and pulmonary edema, and/or any patient receiving ventilation > 48 hours postoperatively

30

ACCEPTED MANUSCRIPT Table 2. Preoperative patient characteristics in patients who died vs. patients who survived.

SC

RI P

T

BMI – body mass index; COPD – chronic obstructive pulmonary disease; CHF – congestive heart failure; EF – ejection fraction; IABP – intraaortic balloon pump; IQR – interquartile range; MI – myocardial infarction; STEMI – ST elevation myocardial infarction. Variable Alive Deceased (n=206) (n=34) P Value 63.4(12.1)

NU

Age, y(SD)

68.2(9.7)

0.03

138/68

15/19

0.02

BMI, mean(SD)

28(5.1)

28.6(4.4)

0.53

136(66)

26(76.4)

0.35

38(18.4)

13(38.2)

0.02

30(14.5)

9(26.4)

0.06

62(30.1)

17(50)

0.03

46(22.3)

12(35)

0.09

Cardiogenic shock, n(%) IABP, n(%)

34(16.5) 125(60.6)

13(38.2) 28(82.3)

0.01 0.03

Previous MI, n(%)

133(64.5)

26(76.4)

0.15

STEMI at admission, n(%)

155(75.2)

27(80.6)

0.11

Left main stenosis>50%, n(%)

106(51.4)

14(41.1)

0.26

Reoperation, n(%)

56(27.2)

7(20.6)

0.21

Preoperative diagnosis of renal failure, n(%)

22(10.6)

14(41.1)

<0.01

Peripheral arterial disease, n(%)

25(12.1)

5(14.7)

0.71

Cerebrovascular disease, n(%)

27(13.1)

4(11.7)

0.45

Preoperative β blockers, n(%)

97(47.1)

12(35.3)

0.08

MA

Gender, male/female, n

Hypertension, n(%)

ED

Diabetes mellitus, n(%) COPD, n(%)

PT

CHF, n(%)

AC

CE

EF<35%, n(%)

31

ACCEPTED MANUSCRIPT

77(37.4)

12(35.3)

0.73

Preoperative hematocrit, mean (SD)

39(5.5)

36.2(4.2)

<0.01

7.4(5.9-10.1)

6.2 (5.4-7.2)

RI P

<0.01

20(59)

0.01

2(5.8)

0.31

9.5(4.2)

8.5(4.5)

0.22

7.57 (9.9)

15.45 (15.1)

0.03

72(35)

Salvage operation, n(%)

10(4.9)

Time from admission to surgery, hrs (SD)

AC

CE

PT

ED

STS predictive mortality score, mean (SD)

MA

NU

Patients with admission blood glucose >6.6 mmol/L, n(%)

SC

Admission blood glucose (mmol/L), median (IQR)

T

Preoperative statins, n(%)

32

ACCEPTED MANUSCRIPT Table 3. Intraoperative variables in patients who died vs. patients who survived.

84(5)

Crossclamp time, min(SD)

63(5)

MA

NU

CPB time, min(SD)

2.8(0.3)

AC

CE

PT

ED

Number of grafts, n(SD)

33

Deceased (n=34)

P Value

90(7)

0.07

66(4)

0.16

2.9(0.2)

0.34

RI P

Alive (n=206)

SC

Variable

T

CPB – cardiopulmonary bypass.

ACCEPTED MANUSCRIPT Table 4. Postoperative variables and complications in patients who died vs. those who survived.

Alive (n=206)

2(1)

P Value

0(0)

0.68

7(20.6)

0.97

17(50)

<0.01

5(2.4)

6(17.6)

<0.01

56(192)

75(92)

0.57

Ventilation > 48 hours, n(%)

36(17.4)

15(44.1)

<0.01

Patients receiving blood transfusion, n(%)

134(65)

29(85.3)

0.01

3BG (mmmol/L), median(IQR)

8.9(7.1-10.2)

7.5(6.8-8.4)

<0.01

ICU stay, hrs(SD)

104(181)

143(184)

0.26

NU

Myocardial Infarction, n(%)

Deceased (n=34)

SC

Variable

RI P

T

ICU – intensive care unit; IQR – interquartile range; 3BG – 3 day blood glucose.

42(20.4)

MA

Atrial fibrillation, n(%)

24(11.6)

ED

Acute kidney injury, n(%)

CE

PT

Stroke, n(%)

AC

Ventilation time, hrs(SD)

34

ACCEPTED MANUSCRIPT 10.5(12)

8.25(11.9)

AC

CE

PT

ED

MA

NU

SC

RI P

T

Postoperative hospital stay, d(SD)

35

0.67

ACCEPTED MANUSCRIPT Table 5. Independent predictors of postoperative mortality on multivariable analysis.

Variable

Odds Ratio

95% CI

P Value

1.58

(1.0, 2.5)

0.04

(1.18, 10.39)

0.02

1.21

(1.11, 1.31)

<0.01

1.16

(1.04, 1.29)

0.01

1.31

(1.1, 1.56)

<0.01

SC

Age (10 year increment)

3.51

NU

Preoperative diagnosis of renal failure

MA

CPB time (10 minute increment)

ED

Admission glucose (0.55 mmol/L increment above 6.6 mmol/L)

AC

CE

PT

3BG

RI P

T

CPB – cardiopulmonary bypass; 3BG - Three-day blood glucose

36

ACCEPTED MANUSCRIPT Table 6. Independent predictors for the composite of postoperative in-hospital mortality/stroke/postoperative kidney injury.

RI P

T

CPB – cardiopulmonary bypass; 3BG - Three-day blood glucose

Odds Ratio

MA

AC

CE

PT

3BG

ED

Admission blood glucose

NU

Preoperative diagnosis of renal failure

CPB time (10 minute increment)

95% CI

P Value

2.72

(6.18, 50.81)

<0.01

1.16

(1.09, 1.24)

<0.01

1.24

(1.07, 1.27)

0.02

1.19

(1.05, 1.36)

0.01

SC

Variable

37

ACCEPTED MANUSCRIPT Figure legends:

RI P

T

Figure 1. Patient distribution vs. admission blood glucose levels (in mmol/L).

Figure 2. A box plot diagram of median admission blood glucose levels in

SC

survivors vs. non-survivors. *P<0.01.

NU

Figure 3. Receiver operator characteristic curves for admission blood glucose

MA

(A), age (B), preoperative renal failure (C) and cardiopulmonary bypass time (D)

ED

to predict mortality after emergency CABG surgery with cardiopulmonary bypass.

Figure 4. A Kaplan-Meier survival curve plotting postoperative in-hospital

AC

CE

or > 6.6 mmol/L.

PT

mortality over time in patients with admission blood glucose  6.6 mmol/L

38