The prospective study was conducted in a tertiary care centre in Tamil Nadu, India over a period of 18 months from April 2017 to September 2018, which studied a total of 263 patients. They were included based upon the age (more than 18 years) and an acute onset of stroke either haemorrhagic or ischaemic. Among them, 161 patients were excluded due to the presence of either known coronary artery disease, previous documented abnormal echocardiographic findings or had electrolyte abnormalities while 2 patients were excluded as they were unwilling to be part of the study. Thus a total of 100 patients were included in the study. The study did not include patients with subarachnoid haemorrhage.

This study was approved by the Institutional Ethics Committee (approval letter: CSP–MED/17/APR/35/53) and written informed consent was obtained from all the patients enrolled in the study. A detailed history including their presenting complaints, duration and co-morbidities was obtained. A thorough clinical examination was performed for all the patients. The observations were recorded in a pre-structured proforma. Their baseline investigations including the haemogram, sugars, lipid profile, renal and liver function tests, urine analysis, cardiac biomarkers like CPK MB, Trop I and BNP were carried on and documented. The patients underwent neuroimaging (CT/MRI brain) and based on the radiological findings were grouped as ischaemic or the haemorrhagic subtypes. All the patients underwent electrocardiogram within 2 hours of hospital admission and echocardiogram. The findings in relation to the type of stroke and the patients’ outcome were documented and analysed.

The data was analysed using SPSS Software. The categorical variables were presented as frequency of occurrence and were analysed by Chi-square or Fischer-exact test. All tests were two tailed and a p-value <0.05 was considered statistically significant.

A total of 100 acute stroke patients were included in the study among which 78 patients had cerebral infarct while 22 patients had haemorrhage as the cause for acute cerebrovascular event. The clinical and demographic data of patients from both the groups are presented in [Table/Fig-1]. Considering the age distribution, younger age was common among the patients with haemorrhage while infarctions were common among the geriatric population. Both the groups had slight male predominance. The occurrence of pre-existing co-morbidities like systemic hypertension, dyslipidemia were similar in both the groups, with a slightly higher occurrence of diabetes among the infarct group. Addiction to smoking, systolic and diastolic blood pressure was significantly higher among the patients with haemorrhagic stroke. However, there was no significant difference between the groups in terms of BMI and alcohol consumption.

The [Table/Fig-2] compares the ECG findings among the two study group patients. T wave inversions were observed among 54 patients with stroke and were the most frequently occurring single ECG abnormality in stroke. No significant relationship was demonstrable between the type of stroke and the pattern of ECG abnormalities.

The [Table/Fig-3] correlates the ECG changes and the cardiac biomarkers and echocardiographic findings. ST segment depression was the single abnormal ECG finding which correlated significantly with elevated Troponin T, Creatinine Phosphokinase (CPK-MB), and Brain Natriuretic Peptide (BNP) as well as the presence of abnormal findings on 2d Echocardiogram.

The [Table/Fig-4] presents the impact of various parameters on the outcome of patients. The higher age at presentation, higher systolic and diastolic blood pressures, low haemoglobin, higher BNP levels were associated with poor outcome (higher mortality) during the hospital stay of the patients. Similarly, the presence of ST segment depression elevated CPK-MB and the occurrence of abnormalities in 2d echocardiogram were strongly associated with increased mortality of the patients.

Among the 100 study patients, 67 were males and 33 were females with a male to female ratio of 2.03:1. This is similar to the observations made by Purshotahamam S et al., (2.4:1), Anand K et al., (1.7:1) and Nagaraja D et al., (2:1) [9-11]. Women have a lesser incidence of stroke compared to the males which could be partly attributed to the genetic susceptibility, lower blood pressure values and the effects of estrogen on cerebral circulation. The study comprised 78% of ischaemic stroke patients which was slightly higher compared to the observations of Kuruvilla T and Bharucha NE, and Kumar HH et al., which were 57.3% and 56%, respectively [12,13]. Haemorrhagic stroke was seen among 22% patients in the present study which was lower than the previous studies [12,13]. Atherosclerosis of the small and medium sized cerebral arteries is the most common cause of ischaemic stroke [14]. Haemorrhagic strokes though less common than ischaemic strokes cause significantly higher mortality and morbidity than the ischaemic strokes [15].

Acute coronary syndrome or myocardial damage due to either atherosclerotic or non-atherosclerotic aetiology will be associated with elevated cardiac biomarkers, ECG changes along with or without echocardiographic abnormalities. However, the cardiac events due to non-atherosclerotic causes were clinically ruled out while including the patients in the study. Further, the atherosclerotic aetiology was substantiated by carotid vertebral Doppler done in all these patients which showed plaques in the carotids.

A 73% of the present study population had ECG changes. This finding is similar to the observations made in the previous studies by Goldstein DS, (92%), Purushothaman S et al., (78%) and Bozluolcay M et al., (62%) [5,6,9]. The most common ECG abnormalities associated with acute stroke in the present study were T wave inversion, ST segment depression and ST segment elevation. There were other observations made in the ECG that included sinus tachycardia, sinus bradycardia, multifocal atrial tachycardia, left and right bundle branch blocks, ventricular arrhythmias and left ventricular hypertrophy. Most of these ECG changes could simulate an acute coronary event or myocardial injury. The frequency of ECG changes was similar to the observations made in previous studies which showed ischaemic changes in ECG along with QTc prolongation in more than 90% patients with ischaemic/haemorrhagic stroke while 8-40% decrease in frequency of ECG changes happened among patients with known cardiac ailment [16-18]. Dogan A et al., observed ischaemia like changes in ECG in 65% of patients, 26% had QTc prolongation and arrhythmias in 44% in their study on ischaemic stroke patients without history of primary heart disease [4]. Similarly, Lindgren A et al., observed transient ST-T changes among 54% ischaemic stroke patients with no primary heart disease [19].

The ECG changes are highly variable over time especially in the setting of acute stroke and standardising them cannot be done until continuous monitoring of the ECG is attempted. This explains the varied ECG changes among various studies [5,6,12]. Most of the ECG abnormalities described in literature after an acute cerebral event were associated with sub arachnoid haemorrhage. However, the present authors confined to ischaemic and haemorrhagic (intracerebral) stroke in the study.

Abnormal T wave was the most common abnormality observed among the ischaemic stroke patients. T wave inversions due to acute myocardial infarctions are classically narrow and symmetrical while those due to non-cardiac causes have low amplitude and are abnormally inverted [3]. ST-T changes in the ECG were observed among 59% of the study population; however, the frequency of these changes was higher among the ischaemic stroke patients compared to the haemorrhagic stroke patients (63.3% vs 57.69%). These observations are in disagreement with those made by Mc Dermott MM et al., and Ebrahim K et al., [7,20]. The ECG changes in acute stroke are transient which probably explains the Difference in findings. Fure B et al., observed ST depression and T wave inversions (24.5% and 17.8%) as the most frequent ECG changes while in our study, T wave inversions followed by the ST segment depression were the commonest findings (54% and 16%) [8].

Several studies have observed the prognostic importance of the ECG changes particularly the ST segment and QT intervals in ischaemic stroke patients [4,6]. Prolonged QTc interval, concurrent myocardial infarction and ventricular arrhythmias have been strongly associated with increased mortality among the stroke patients [21]. However, in the index study, ST segment depression showed significant correlation with the presence of abnormal echocardiogram and elevated cardiac enzymes viz., Troponin T, Creatinine phosphokinase-MB, and also Brain natriuretic peptide. ST segment depression was also associated significantly with poorer in-hospital patient outcome.

In a prospective study by Jensen JK et al., Troponin T and CPK-MB were found elevated among 10% and 9% of acute stroke patients respectively while 3% patients had both Troponin T and CPK-MB elevated along with electrocardiographic changes [22,23]. This is in contrast to the observations made in the present study wherein 32 patients had elevated cardiac enzymes and all of them had ECG changes; 42.3% patients in the ischaemic group and 31.82% haemorrhagic stroke patients had both ECG and echocardiographic abnormalities.

It is well known that Left Ventricular Hypertrophy (LVH) on echocardiogram is an independent risk factor for hypertension related target organ damage [24,25]. In stroke patients, the presence of LVH increases the risk of recurrent cerebrovascular accidents two folds [26]. It takes longer duration for LVH to develop; hence it may not be a new sign in acute stroke patients. Systemic hypertension has a strong bearing on the occurrence of CVA as evidenced by the presence of left ventricular hypertrophy among 17% of the study population.

Stroke is one of the major causes of mortality and morbidity. This study highlighted many independent risk factors for mortality among patients with acute stroke. Higher age group, lower haemoglobin levels, elevated BNP levels, presence of abnormal findings on echocardiogram and CPK-MB elevations were significantly associated with higher rates of mortality among these patients. The mortality rates were not significantly altered by the presence of traditional risk factors like the male sex, presence of diabetes mellitus, systemic hypertension. The mortality rate was 9.1% among the haemorrhagic stroke patients while it was 5.1% among the ischaemic stroke patients. Though the rate seemed slightly higher in the haemorrhagic group, it was not statistically significant.

Serial ECGs were not recorded to document the occurrence of transient changes in the ECGs of patients with acute stroke. A control group was not studied where the presence of abnormal ECG patterns among normal population could have added more significance.

Acute cerebrovascular accidents cause autonomic imbalance in the myocardium which causes many electrocardiographic abnormalities and predisposes these patients to increased mortality risk. Hence, irrespective of the presence of co-morbidities, ECG changes particularly ST segment depression and the presence of elevated CPK-MB serve as independent predictors of patients’ outcome in acute stroke, especially in resource limited settings.