On Arrival High Blood Glucose Level is Associated With Detrimental and Fatal Hospitalization Outcomes for Acute Coronary Syndrome

Background High blood glucose level is frequently encountered in acute coronary syndrome. We investigated the effects of high blood glucose measured on arrival on hospitalization adverse events in acute coronary syndrome. Our study patients were Javanese in ethnicity, which constitute half of population in Indonesia. We hypothesized that elevated blood glucose has detrimental effects on hospitalization for acute coronary syndrome. Methods We designed an observasional cohort study and recruited 148 consecutive patients with acute coronary syndrome. Venous blood was collected on hospital arrival. High blood glucose level was determined as plasma glucose > 140 mg/dL. Adverse hospitalization events were recorded, i.e. mortality, acute heart failure, cardiogenic shock and heart rhythm disorders. Echocardiography examination was performed to determine left ventricular function. Results The prevalence of on arrival high blood glucose among Javanese patients with acute coronary syndrome was considerably high (36%). On arrival high blood glucose was associated with acute heart failure (P < 0.001) and shock cardiogenic (P = 0.02). Heart rhythm disorders were higher in high blood glucose patients (P = 0.004). Left ventricular dysfunction was more prevalent in high blood glucose patients (P = 0.001) and ejection fraction was lower (P = 0.001). On arrival high blood glucose was independently associated with hospitalization adverse events (adjusted odds ratio = 2.3, 95% confidence interval: 1.1-4.9, P = 0.03) and hospital mortality (adjusted odds ratio = 6.9, 95% confidence interval: 1.2-38.6, P = 0.03). Conclusions Our study suggests that on arrival high blood glucose among Javanese patients with acute coronary syndrome is considerably high and is associated with detrimental and fatal hospitalization outcomes.


Introduction
Elevation of blood glucose level is encountered in ranges from 25% to more than 50% of patients admitted with acute myocardial infarction and is a predictor for death and complications during hospitalization [1][2][3]. In patients with acute myocardial infarction and unstable angina, both conditions are recognized as acute coronary syndrome, high blood glucose on admission is as much as 60% [4,5] and is associated with high rate of left ventricular failure and cardiac death [6].
A decline of insulin sensitivity and a reduction of glucose utilization during acute coronary syndrome contribute to elevated blood glucose level [7]. Rising production of stress hormones, such as catecholamine and corticosteroids, at some stage of acute myocardial ischemic event give rise to stress induced changes in glucose metabolism [8]. High blood glucose level unfavorably affects outcome of acute coronary syndrome through several collective effects on myocardial survival and function [9].
In the present study, we prospectively evaluated the impacts of high blood glucose level measured on arrival on hospitalized patients with acute coronary syndrome, with and without previous diabetes mellitus. Our study patients were Javanese in ethnicity, which constitute half of population in Indonesia [10]. Recent report showed the prevalence of diabetes in Indonesia has been considerably escalating in both rural and urban areas [11]. We hypothesized that the prevalence of elevated blood glucose in our study patients was considerable and its effects on acute phase of intensive hospitalization for acute coronary syndrome was detrimental.

Study patients
In this observational cohort study, we recruited consecutive patients, Javanese in ethnicity, admitted to Intensive Coronary Care Unit (ICCU) of Dr. Sardjito Hospital, the referral centre for cardiac care in The Special Region of Yogyakarta, Island of Java, Indonesia, for having acute coronary syndrome. The time of recruitment was between September 2008 and May 2009. We included male and female patients between 35 and 80 years old with the onset of pain symptom less than 24 hours before admission. We excluded those with known end-stage chronic kidney disease, chronic heart failure (NYHA class more than II), valvular heart disease, malignancy, acute co morbidity such as acute stroke, acute infection and sepsis, chronic infl ammatory disease and venous thromboembolisme. Informed consent was obtained from patients or their families. The procedure of this study was approved by Ethics Committee of Faculty of Medicine Universitas Gadjah Mada, Yogyakarta, Indonesia.
Characteristic data of study patients were acquired during hospitalization. Demography and medical history were attained through medical anamnesis. Clinical presentation on-arrival was assessed by on-duty physician and noted in the patient records.
The diagnosis of acute coronary syndrome comprised acute myocardial infarction (AMI) and unstable angina. AMI was determined as angina type chest pain lasting more than 20 minutes, electrocardiography examination revealed changes in ST segment, either ST elevation or ST depression and elevated cardiac enzyme (i.e. troponin I ≥ 0.6 ng/ mL). Patients with acute coronary syndrome but without any proof of acute infarction (i.e. troponin I < 0.6 ng/mL ) were diagnosed with unstable angina.

Laboratory examination
Venous blood was collected immediately upon hospital arrival, before starting specifi c treatment such as revascularization (i.e. thrombolysis or primary percutaneous coronary intervention (PCI)) and heparinization. Routine complete blood cell count, blood chemistry measurement (glucose, creatinine and lipid profi les) and troponin I level were conducted in our hospital central laboratory.
Plasma glucose level was measured using glucose oxidase method. High blood glucose level was determined as plasma glucose > 140 mg/dL in both patients with and without known diabetes mellitus. The cut-off point of 140 mg/dL was in accordance with American Heart Association (AHA) scientifi c statement [12]. Accordingly, patients were divided into two groups, those with blood glucose level above 140 mg/dL in high blood glucose group and those with blood glucose level below and equal to 140 mg/dL in normal blood glucose group.

Management of patients
Since this study is observational, we did not take part in managing patients. The standard treatment for managing acute coronary syndrome, i.e. revascularization, heparinization, antiplatelet and pain management, was applied by intensivist cardiologists. For management of high blood glucose level, the diabetologists were consulted. As a procedure, continuous intravenous insulin and glucose treatment were administered if blood glucose level ≥ 200 mg/dL, both in diabetic and non diabetic patients. Blood glucose level was monitored every one hour until targeted blood glucose level was reached (approximately 110 mg/dL). Once it was achieved (usually within 24 hours), intravenous insulin was maintained in low dose or switched into subcutaneous basal and postprandial dose.

Hospitalization events
We recorded adverse events during intensive hospitalization in ICCU. Hospital mortality, haemodynamic disturbances (acute heart failure and cardiogenic shock) and heart rhythm disorders were documented. Hospital mortality was determined death during hospitalization from any cause. Acute heart failure was diagnosed based on clinical sign and symptom of pulmonary congestion, the use of intravenous loop diuretics (furosemide) or the use of positive inotropic agents. On arrival heart failure was assessed if the condition occurred in patients on hospital arrival and subsequent heart failure was assessed if the condition started or lasted after 24 hour hospitalization. Cardiogenic shock was diagnosed based on systolic blood pressure < 90 mmHg and the use of vasopressor agents. Heart rhythm disorders were recorded through continuous electronic monitor which is a standard procedure in ICCU. Intensivist cardiologists evaluated these events and treated the patients accordingly.
To determine left ventricular function, an echocardiography examination was performed. It was conducted within day three until day seven hospitalization. M-mode and twodimensional echocardiography as well as Doppler echocardiography were performed. Left ventricular systolic function was determined based on the value of left ventricular ejection fraction (LVEF). Normal LVEF is ≥ 50%. Accordingly, we used cut-off point of LVEF 50% to determine left ventricular systolic dysfunction.

Statistical analysis
We performed statistical analysis using SPSS version 13.0 (SPSS, Chicago, IL, USA). Normality test was conducted for continuous data with Kolmogorov-Smirnov test. Between groups, the mean values of these data were compared using Student T or Mann Whitney tests. The proportion of categorical data were compared using chi squared test or Fisher exact test. The correlation between continuous variables was tested using Spearman test. Univariable analysis was used to asses the association between blood glucose and other predictors with hospitalization events. Predictors signifi cantly associated with hospitalization events then entered into multivariable analysis with stepwise multivariate logistic regression method to analyse the independent association. Subgroup analysis was conducted for diabetic status. A statistical signifi cance was considered when P < 0.05.

Patient characteristics
We enrolled 148 consecutive patients. Fifty three patients had high blood glucose level on arrival, which comprised 36% of all study patients and 45% of patients with acute myocardial infarction. Those with known diabetes mellitus were likely to have high blood glucose on arrival (21 of 28 diabetic patients). Nevertheless, thirty two patients (60%) with high blood glucose did not report previous diabetes mellitus. Female patients tended to have high blood glucose level on arrival (16 of 30 female patients) although male patients dominated the proportion (70%). Patients with high blood glucose had higher proportion to suffer from acute myocardial infarction (AMI) than from unstable angina (87% versus 7%, P = 0.001). Table 1 summarized the characteristic of study patients.
On arrival high blood glucose did not associate with blood pressure, either systolic or diastolic. Heart rate was slightly higher in patients with high blood glucose (P = 0.01). On arrival high blood glucose had higher white blood cell (WBC) count (P = 0.001) and troponin I level (P = 0.007) than normal glucose. As shown in table 2, Spearman test revealed that blood glucose level correlate with heart rate (r = 0.26, P = 0.001), WBC count (r = 0.36, P < 0.001) and troponin I (r = 0.33, P < 0.001).
Almost 50 % of on arrival high blood glucose patients were treated with intravenous insulin to control hyperglycemia. All of them had blood glucose level ≥ 200 mg/dL when insulin therapy was commenced. Revascularization therapy for AMI with elevated ST segment was similar between groups as well as heparinization. Antiplatelets, beta blockers and ACE inhibitors were comparable between groups. List of medication during hospitalization was shown in table 3.

High blood glucose was associated with hemodynamic disturbances
Hemodynamic disturbances frequently complicate acute cardiac ischemic events. Patients presented with high blood glucose on arrival had signifi cantly higher proportion of on arrival and subsequent acute heart failure as compared with those with normal blood glucose (43% versus 16%, P < 0.001 and 60% versus 20%, P < 0.001, respectively). Furthermore, shock cardiogenic occurred more frequently in those with on arrival high blood glucose (10% versus 1%, P = 0.02).

High blood glucose was associated with rhythm disorders
Rhythm disorders following acute cardiac ischemic event could be devastating. Monitoring heart rhythm is an essential part of managing the patients. Patients with on arrival high blood glucose had signifi cantly higher proportion of rhythm disorders as compared with those with normal glucose (40% versus 18%, P = 0.004). The majority (13 of 21 patients) of these rhythm disorders was originated from ventricle, i.e. ventricular extra systole or ventricular tachycardia/fi brillation. Fatal rhythm disorder leading to sudden death occurred in 2 patients with on arrival high blood glucose (Table 4).

High blood glucose was associated with left ventricular dysfunction
Fifty nine patients underwent echocardiography examination during hospitalization. Cardiologists conducted the examination blindly to blood glucose level. Left ventricular systolic dysfunction developed more frequently in patients with on arrival high blood glucose than those with normal blood glucose (80% versus 37%, P = 0.001; Fig. 1). Furthermore, the higher the blood glucose value, the lower the left ventricular ejection fraction (Spearman correlation:

Discussion
Our study showed that high blood glucose level on hospital arrival was associated with detrimental effects during hospitalization. Among patients with acute coronary syndrome, on arrival high blood glucose associated with hemodynamic disturbances, rhythm disorder, left ventricular dysfunction and hospital mortality.
The frequency of high blood glucose in our study was 36% among acute coronary syndrome and 45% among AMI. Acute myocardial ischemic event due to coronary inadequate blood fl ow may lead to myocardial infarction. This subset of patients tends to have complication during hospitalization. Myocardial infarction is a stress condition that caused elevation of stress hormones and subsequent elevation of blood glucose [8]. Our result showed that almost 90% of patients with high blood glucose were diagnosed with AMI. Furthermore high blood glucose associated with elevated WBC count, marker for infl ammation, and troponin I level, marker for injured myocardial. Positif correlation was observed be- Table 4. High Blood Glucose and Hospitalization Events AHF: acute heart failure.  This indicated that the elevated blood glucose level was in conjunction with the severity of myocardial damage. Of those with on arrival high blood glucose, 40% had previous diabetes mellitus. Unsurprisingly, diabetic patients had their blood glucose raised when they presented with acute coronary syndrome. This refl ected the previous glucose dysmetabolism and glucose handling abnormalities which were exacerbated by stress response mediated by catecholamine and corticosteroid hormones [13].
Patients with hemodynamic disturbances signifi cantly have high blood glucose on arrival. Acute heart failure occurred on arrival, when no treatment was administered yet, was associated with high blood glucose. After 24 hour proper treatment, subsequent heart failure was also associated with on arrival high blood glucose. Furthermore, patients with on arrival high blood glucose developed left ventricular systolic dysfunction and lower ventricular ejection fraction more frequently during hospitalization. Heart failure and ventricular dysfunction augment the autonomic nervous system activity with subsequent release of catecholamine and corticosteroids. This stress response occurs early and lasts for the fi rst fi ve days, during which increased gluconeogenesis and lipolysis lead to increased circulating glucose, free-fatty acids and lactic acid [14]. These metabolic disturbances change metabolic substrate for myocardium which becomes intoxicated [15]. Our study showed higher proportion of rhythm disorder, particularly originated from ventricular wall, frequently occurred in those with high blood glucose. This arrhythmia might have been a refl ection of cathecholamine response and intoxicated myocardium [15].   Our study showed that high blood glucose on arrival independently associated with hospital mortality. This fi nding encompassed both previously known diabetic and non diabetic patients. Previous reports showed that nondiabetic patients with high blood glucose at admission have similar mortality rate to those with established diabetes [16]. In patients with revascularization therapy, admission hyperglycemia was strongly associated with in hospital mortality regardless a previous diagnosis of diabetes [17].

Hospitalization Events
In conclusion, our study suggests that on arrival high blood glucose level among Javanese patients with acute coronary syndrome is considerably high and this is associated with detrimental and fatal hospitalization outcomes.