American Journal of Emergency Medicine (2012) 30, 732-736

Original Contribution

The 12-lead electrocardiogram in patients with subarachnoid hemorrhage: early risk prognostication^?,??

Chien-Cheng Huang MD ^a,b, Chi-Hung Huang MD ^b,c, Hung-Yi Kuo MD ^a,b,

Chia-Meng Chan MD ^a,b, Jiann-Hwa Chen MD, MPH ^a,b, Wei-Lung Chen MD, PhD ^a,b,?

^aDepartment of Emergency Medicine, Cathay General Hospital, Taipei 106, Taiwan

^bSchool of Medicine, Fu-Jen Catholic University, Taipei 242, Taiwan

^cDepartment of Cardiology, Cathay General Hospital, Taipei 106, Taiwan

Received 22 March 2011; revised 5 May 2011; accepted 7 May 2011

Abstract

Objective: The aim of this study was to investigate if the electrocardiographic abnormalities assessed early in the emergency department (ED) are associated with the in-hospital mortality of the patients with spontaneous subarachnoid hemorrhage (SAH).

Methods: We studied prospectively a cohort of 222 adult patients with spontaneous SAH in an ED. A 12-lead ECG was performed for these patients in the ED. The patients were stratified into nonsurvivors and survivors based on the in-hospital mortality. The clinical characteristics, heart rate, Corrected QT interval (QTc) and 7 predefined morphologic abnormalities were compared between these 2 groups of patients. Results: Compared with the survivors (n = 178), the nonsurvivors (n = 44) had significantly slower heart rate (75 +- 23 vs 83 +- 16, P = .018) and more prolonged QTc (492 +- 58 vs 458 +- 40, P = .001). There were significantly higher frequency of occurrence of ECG morphologic abnormalities (66% vs 37%, P = .001) and nonspecific ST- or T-wave changes (NSSTTCs; 32% vs 12%, P = .015) in the nonsurvivors compared with those in the survivors. Multiple logistic regression model identified QTc (odds ratio, 1.0; 95% confidence interval, 1.0-1.0; P = .005) and NSSTTC (odds ratio, 3.3; 95% confidence interval, 1.0-10.7; P = .047) as the significant ECG variables associated with in-hospital mortality.

Conclusions: The occurrence of NSSTTC and prolonged QTc assessed early in the ED are independently associated with the in-hospital mortality in adult patients with spontaneous SAH.

(C) 2012

^? Competing interests: None of the authors have any conflicts to disclose.

^?? Authors’ contributions: The study was designed by CCH and

WLC. Acquisition of the data was performed by CCH, HYK, CMC, JHC, and WLC. Analysis and interpretation of data was done by CHH and WLC. The statistical analysis was done by JHC and WLC. The manuscript was drafted by CCH. Critical revision of the manuscript for important intellectual content was done by WLC. Final approval of the manuscript was done by CCH, CHH, HYK, CMC, JHC, and WLC.

* Corresponding author. Tel.: +886 2 27082121×3761; fax: +886 2

27021428.

E-mail address: [email protected] (W.-L. Chen).

Introduction

Certain intracranial events can induce electrocardiographic (ECG) abnormalities that most often include morphologic changes and Rhythm disturbances [1-3]. Subarachnoid hemor- rhage (SAH) is perhaps the most notorious intracranial event that manifests with such ECG changes, and diverse ECG changes have been reported to occur in 25% to 90% of patients with SAH [4-8]. Previous studies have suggested that patients with more severe SAH are more likely to develop Cardiac abnormalities and are further associated with poor neurologic

0735-6757/$ - see front matter (C) 2012 doi:10.1016/j.ajem.2011.05.003

outcome [5,9-11]. Moreover, Kawasaki et al [12] have reported that such ECG abnormalities can be predictors of mortality after SAH in a retrospective study. By assessing the preoperative 12-lead ECG in a substudy of aneurysm surgery trial, Coghlan et al [13] further demonstrated that there were independent associations between ECG abnormalities and 3-month mortality in patients with SAH.

Because studies have reported that the ECG abnormalities usually appeared early after brain injury and disappeared within 1 day [8,14,15], the timing of ECG recording may influence the conclusion about the prevalence of ECG abnormalities and their association with the outcome. However, little information is known about the relation between the ECG abnormalities assessed in the early stage of SAH and the prognosis of the patients. Therefore, the aim of this study was to investigate whether the ECG abnormalities assessed early in the emergency department (ED) are associated with the in-hospital mortality in patients with spontaneous SAH.

Materials and methods

Study design

This was a prospective cohort study that investigated whether the ECG abnormalities occurring in adult patients with spontaneous SAH in the ED are independently associated with the in-hospital mortality of the patients. The study protocol was approved by the institutional review board of the hospital. Written informed consent was obtained from the patients themselves or their next of kin before enrolling in the study.

Study setting and selection of participants

This study was conducted in a 700-bed university- affiliated medical center with a 40-bed ED staffed with board-certified emergency physicians that provides care for approximately 55 000 patients per year. From October 2003 to September 2010, adult patients who were admitted within 12 hours of the first clinical symptoms with nonTraumatic SAH diagnosed by Computed tomographic scans of brain, or xanthochromia of the cerebrospinal fluid if the computed tomography was nondiagnostic were eligibly enrolled. The exclusion criteria of this study included the following: (1) known arrhythmia (reviewing the past medical chart or by the statements of the patients themselves), (2) cardiac pacing, (3) aged less than 18 years, or (4) referred from other hospital (the ECG abnormalities of the patient seen in our ED were not in the early stage any more).

Study protocol

After obtaining the written informed consent, a standard 12-lead ECG recording was performed immediately after the

diagnosis was made. The patient’s ECG was interpreted by the same cardiologist who was blinded to the outcome of the patient. The following measurements were made for each ECG: heart rate (ventricular rate) and heart rate-corrected QT interval (QTc) using the Bazett formula. The morpho- logic abnormalities were defined as the presence of 1 or more of the following 7 variables, which are commonly noted after SAH [2,12,13], in at least 2 leads: (1) abnormal Q or QS wave (>=30 milliseconds or a pathologic R wave in V₁ to V₂),

(2) ST elevation (ST elevation >=0.1 mV), (3) ST depression (ST depression >=0.1mV, 80 milliseconds post-J point), (4) peaked upright T wave (prominent peaked T wave), (5) T-wave inversions (pathologic T-wave inversion), (6) giant T-wave inversions (T-wave inversions N10 mV in depth), or (7) nonspecific ST- or T-wave changes (NSSTTC; ST- or T-wave abnormalities not meeting the above criteria).

The following demographic data and clinical variables were recorded at the same time for all patients: age, sex, vital signs, laboratory data, Hunt-Hess scale (class I, asymptom- atic or mild headache; class II, moderate or Severe headache, Nuchal rigidity, can have oculomotor palsy; class III, confused, drowsiness, or mild focal signs; class IV, stupor or hemiparesis; class V, coma, moribund, and/or extensor posturing) [10], World Federation of Neurological Surgeons (WFNS) class (class I, Glasgow coma scale [GCS] 15, no motor deficit; class II, GCS 13-14, no motor deficit; class III, GCS 13-14, presence of motor deficit; class IV: GCS 7-12; class V: GCS 3-6) [16], underlying diseases, comedication that can affect the heart rate, the time of symptoms onset to arrival on the ED, and the time from ED visit to ECG recording. After hospital discharge, the in-patient medical record was reviewed to complete the data collection: length of hospital stay and outcome (in-hospital mortality). Patients discharged from the hospital in less than 28 days or who remained alive for more than 28 days were classified as “survivors” in this study; otherwise, the patients were referred to as “nonsurvivors.”

Statistical analysis

?2 test or Fisher exact test when appropriate was used for the statistical analysis of categorical variables. Continuous variables were presented as mean (SD) and compared using the independent samples t test. For statistical purposes, the Clinical scores used in this study were dichotomized in good and poor groups (WFNS 1-3 vs WFNS 4-5, Hunt-Hess 1-3 vs Hunt-Hess 4-5). The clinical variables and ECG abnormalities with univariate comparison P b .2 between 2 groups were eligible for inclusion in a forward selection multiple logistic regression model to identify the ECG abnormalities assessed early in the ED that were indepen- dently associated with in-hospital mortality of the patients with spontaneous SAH. A P b .05 was considered statistically significant. Statistical analyses were performed using a common statistical package (SPSS 14.0 for Windows; SPSS Inc, Chicago, IL).

Results

Table 2 ECG findings of the patients

Nonsurvivors (n = 44)		Survivors (n = 178)	P
Heart rate (beats/min), mean +- SD QTc (ms), mean +- SD Morphology changes, n (%) Abnormal Q or QS wave ST elevation ST depression Peaked upright T wave T-wave inversion Giant T-wave inversions NSSTTC	75 +- 23	83 +- 16	.018 ?
	492 +- 58	458 +- 40	.001 ?
	6 (14)	11 (6)	.113
	3 (7)	8 (4)	.459
	5 (11)	19 (11)	1.000
	2 (5)	3 (2)	.258
	5 (11)	10 (6)	.184
	1 (2)	4 (2)	1.000
	15 (32)	21 (12)	.001 ?
* P b .05 between 2 groups.

During the 7-year study period, a total of 243 nontrau- matic adult patients with SAH were treated in the ED. Of them, 21 patients who did not meet the inclusion criteria were not included in the present study: 11 patients had known arrhythmia, 4 patients had cardiac pacing, and 6 patients were already diagnosed as having SAH at another hospital. In all, 222 of 243 patients were included in the final analysis. Based on their in-hospital mortality, those 222 patients, aged 19 to 77 years, were stratified into non- survivors (n = 44) or survivors (n = 178).

The basic characteristics of both groups of patients are shown in Table 1. There were no significant differences in age, sex, mean arterial pressure, comedication, underlying diseases, and the time from ED visit to ECG recording between these 2 groups of patients. However, the WFNS class, Hunt-Hess scale, white blood cell count, glucose, and the occurrence of ECG morphologic abnormal- ities were significantly higher, whereas the length of hospital stay was significantly lower, in the nonsurvivors compared with those in the survivors. In addition, the frequency of

ECG morphologic abnormalities for all enrolled patients was noted to be 42% (94/222).

Table 2 demonstrates the heart rate, QTc, and ECG morphologic abnormalities for both groups of patients. We found that the nonsurvivors had significantly slower heart rate and more prolonged QTc compared with those of the survivors. In addition, there was significantly higher frequency of NSSTTC in the nonsurvivors than in the survivors.

Multiple logistic regression models were used to analyze the risk of in-hospital mortality (dependent variable), and the independent variables included in the analysis were heart rate, QTc, abnormal Q or QS wave, T inversion, NSSTTC, WFNS class, Hunt-Hess scale, WBC count, and glucose. The results showed that QTc, NSSTTC, WBC count, and WFNS class were the significant independent variables associated with in-hospital mortality for adult patients with nontrau- matic SAH in the ED. The odds ratio and P value for these 4 variables are listed in Table 3.

Table 1 Demographic and clinical characteristics of the patients with SAH in the ED

Nonsurvivors Survivors P (n = 44) (n = 178)

Age (y), mean +- SD 54 +- 11 52 +- 13 .786 Sex, male/female, n 15/29 74/104 .395 Vital signs, mean +- SD MAP (mm Hg) 125 +- 9 126 +- 12 .788 WFNS class, n (%) b .001 ? I-III 17 (39) 173 (97) IV-V 27 (61) 5 (3) Hunt-Hess scale, n (%) b.001 ? I-III 17 (39) 169 (95) IV-V 27 (61) 9 (5) Laboratory data, mean +- SD WBC (per mm3) 11605 +- 2760 9644 +- 2321 b.001 ? Glucose (mg/dL) 121 +- 39 102 +- 33 .002 ? Comedication, n (%) ?-Blocker 8 (18) 21 (12) .316 Underlying disease, n (%) Hypertension 9 (20) 30 (17) .658 CAD 5 (11) 11 (6) .324 DM 4 (9) 18 (10) 1.000 Time of E-E (min), 104 +- 33 111 +- 41 .305 mean +- SD ECG MA, n (%) 29 (66) 65 (37) .001 ? Length of hospital 13 +- 7 25 +- 6 b.001 ? stay (d), mean +- SD

MAP indicates mean arterial pressure; CAD, coronary artery disease; DM, diabetes mellitus; MA, morphologic abnormalities; time of E-E, time from ED visit to ECG recording. * P b .05 between 2 groups.

Discussion

Our study demonstrated that the NSSTTC and QTc assessed early in the ED were independently associated with the in-hospital mortality of the patients with spontaneous

Table 3 Statistically independent variables associated with in- hospital mortality in the multiple logistic regression models

OR (95% CI)		P
WFNS class	43.5 (12.9-146.5) a	b.001
WBC	1.0 (1.0-1.0)	.001
QTc	1.0 (1.0-1.0)	.005
NSSTTC	3.3 (1.0-10.7)	.047
OR indicates odds ratio; CI, confidence interval.
a The OR for WFNS class represents the OR for each 1 point
increases from low grade of WFNS (I-III) to high grade of WFNS
(VI-V).

SAH. The occurrence of NSSTTC and prolonged QTc were indicative of poor outcome in these patients. Although the relation between ECG abnormalities and prognosis has been studied in patients with SAH in previous studies, no such study was mainly conducted in the ED to prospectively investigate this relationship in the early stage of SAH. Brouwers et al [4] found that the most pronounced ECG changes occurred during the first 72 hours after SAH. Di Pasquale et al [17] found that 90% of patients had ECG abnormalities during the first 48 hours, suggesting that previous studies in which surveillance was started later in the course of illness might miss significant data. In the present study, we found that the frequency of ECG morphologic abnormalities in the early stage for all enrolled patients was 42%, which was similar to the previous reports that ECG morphologic abnormalities occurred in 25% to 90% patients with SAH [4-6]. We therefore speculated that the morpho- logic abnormalities of ECG may appear in the acute phase and persist for several days in patients with SAH.

Factors that may influence the development of arrhyth- mias in patients with SAH include cerebral vasospasm, hypoxia, electrolyte imbalance, and sudden increase in intracranial pressure triggering a sympathetic or vagal discharge due to compression of brain structures. Studies have further demonstrated that SAH, by causing local cerebral arteriolar spasm, can give rise to ischemic lesions in the hypothalamus and surrounding area [18], which therefore causes sympathetic stimulation of the heart via elevated plasma catecholamine, which in turn is responsi- ble for the ECG abnormalities, especially for arrhythmias, via 1 of 2 pathways: an indirect effect via humoral mediators such as epinephrine and norepinephrine and a direct effect via afferent and efferent connections with the sympathetic and vagal nervous systems [19,20]. Theories about the underlying causes of ECG morphologic abnor- malities in SAH are controversial. Cardiac injury due to elevated myocardial wall stress associated with arrhythmias has been suggested as a causative factor. Yuki et al [21] proposed that Coronary vasospasm and reversible post- ischemic “stunned myocardium” may influence the devel- opment of ECG morphologic changes in patients with SAH. Although the investigations about the mechanisms are still ongoing, many studies have reported that a prolonged QTc and the occurrence of ECG morphologic abnormalities may identify patients who are more likely to have adverse hemodynamic outcomes during their hospital course [13] and that may further contribute to the outcomes. Therefore, the association between ECG abnor- malities in the early stage of SAH and the poor outcome of the patients might have a prognostic value for patients with SAH in the ED.

Coghlan et al [13] have reported that bradycardia and NSSTTC were strongly and independently associated with 3- month mortality in a substudy of the intraoperative hypothermia aneurysm trial in patients with mild-to- moderate SAH (preoperative WFNS classes 1-3). Previous

studies have also reported that the incidence of QTc prolongation was 11% to 66% in patients after SAH [4,7,22]. Andreoli et al [8] have reported that cardiac arrhythmias were documented in 91% of patients with SAH in their prospectively placed Holter monitor on all patients admitted due to SAH; 41% of them were found to be serious, including bradycardia and ventricular tachycardia. Di Pasquale et al [17] have found that there was a more prolonged QTc in those patients with SAH who had malignant ventricular arrhythmias attack. Frontera et al

[11] have reported that the ECG morphologic abnormalities on admission could predict the occurrence of cardiac arrhythmia, which were further associated with outcome. In the present study, we also found that the QTc and existence of NSSTTC were independently associated with the in-hospital mortality after adjusting with clinical vari- ables, suggesting that a routine measurement of QTc interval and examination of 12-lead ECG in patients with SAH may help detect predisposition to poor outcome.

We also found that the nonsurvivors had higher glucose level and WBC count compared with the survivors. Hyperglycemia and leukocytosis are common after SAH; they were reported to be associated with serious hospital complications and increased risk of death or severe disability [23,24]. A catecholamine surge and generalized stress response were reported to be partially responsible for the hyperglycemia in the setting of acute neurologic injury [25,26]. The elevations of Proinflammatory cytokines in the peripheral blood and in cerebrospinal fluid were speculated to be the possible mechanism of leukocytosis in patients with central nervous events [27]. In addition, the WBC count assessed in the ED was identified to be an independent variable associated with in-hospital mortality in patients with SAH in the present study. Our results not only were consistent with previous studies [23-25,27] but also might provide clinicians, especially emergency physicians, valu- able information when facing such critically ill patients.

Our study did have some limitations. First, it is problematic to identify if the ECG abnormalities assessed in the present study were the acute changes for those patients (n = 10) who had no history of arrhythmias or ECG morphologic abnormalities because their history could not be obtained or no previous 12-lead ECG could be obtained to compare with. There could be potential biases when those patients were included in the final analysis. Second, a relatively small sample size was enrolled in this study. Further study with a large sample size is needed in the future. Third, the interpretation of ECG morphologic abnormalities, especially for NSSTTC, might be subjective. To minimize the bias, the patient’s ECG has been interpreted by the same investigator (cardiologist) who was blinded to the outcome of the patient. Nevertheless, to the best of our knowledge, the present study was the first prospective study performed in the ED to explore the relation between early ECG abnormalities and the outcome of the patients with SAH.

Conclusions

This study demonstrated that the ECG morphologic abnormalities and QTc interval assessed early in the ED by a standard 12-lead ECG are independently associated with the in-hospital mortality for adult patients with nontraumatic SAH. Prolonged QTc and occurrence of NSSTTC may be indicative of poor outcome in such patients.

References

1. Davis TP, Lesch M. Electrocardiographic changes associated with acute cerebrovascular disease: a clinical review. Prog Cardiol Dis 1993;36:245-60.
2. Perron AD, Brady WJ. electrocardiographic manifestations of CNS events. Am J Emerg Med 2000;18:715-20.
3. Catanzaro JN, Meraj PM, Zheng S, et al. Electrocardiographic T-wave changes underlying acute cardiac and Cerebral events. Am J Emerg Med 2008;26:716-20.
4. Brouwers PJAM, Wijdicks EFM, Hasan D, et al. Serial electrocardio- graphic recording in aneurismal subarachnoid hemorrhage. Stroke 1989;20:1162-7.
5. Zaroff J, Rordorf G, Newell J, et al. Cardiac outcome in patients with subarachnoid hemorrhage and electrocardiographic abnormalities. Neurosurgery 1999;44:34-9.
6. Sommargren CE, Zaroff JG, Banki NM, et al. Electrocardiographic Repolarization abnormalities in subarachnoid hemorrhage. J Electro- cardiol 2002;35(suppl):257-62.
7. Marion DW, Segal R, Thompson ME. Subarachnoid hemorrhage and the heart. Neurosurgery 1986;18:101-6.
8. Andreoli A, di Pasquale G, Pinelli G, et al. Subarachnoid hemorrhage: frequency and severity of cardiac arrhythmia: a survey of 70 cases studied in the acute phase. Stroke 1987;18:558-64.
9. Sakr YL, Lim N, Amaral AC, et al. Relation of ECG changes to neurological outcome in patients with aneurismal subarachnoid hemorrhage. Int J Cardiol 2004;96:369-73.
10. Hunt W, Hess R. Surgical risk as related to time of intervention in the repair of Intracranial aneurysms. J Neurosurg 1967;28:14-20.
11. Frontera JA, Parra A, Shimbo D, et al. Cardiac arrhythmias after subarachnoid hemorrhage: risk factors and impact on outcome. Cerebrovasc Dis 2008;26:71-8.
12. Kawasaki T, Azuma A, Sawada T, et al. Electrocardiographic score as a predictor of mortality after subarachnoid hemorrhage. Circ J 2002;66:567-70.
13. Coghlan LA, Hindman BJ, Bayman EO, et al. Independent associations between electrocardiographic abnormalities and out- comes in patients with aneurismal subarachnoid hemorrhage. Stroke 2009;40:412-8.
14. Banki NM, Kopelnik A, Dae MW, et al. Acute neurocardiogenic injury after subarachnoid hemorrhage. Circulation 2005;112:3314-9.
15. Kuroiwa T, Morita H, Tanabe H, et al. Significance of ST segment elevation in electrocardiograms in patients with ruptured cerebra aneurysms. Acta Neurochir (Wien) 1995;133:141-6.
16. Drake GD, Hunt WE, Kassell N, et al. Report of World Federation of Neurological Surgeons Committee on a universal subarachnoid hemorrhage grading scale. J Neurosurg 1988;68:985-6.
17. Di Pasquale G, Pinelli G, Andreoli A, et al. Holter detection of cardiac arrhythmias in intracranial subarachnoid hemorrhage. Am J Cardiol 1987;59:596-600.
18. Crompton MR. Hypothalamic lesions following the rupture of cerebral berry aneurysms. Brain 1963;86:301-14.
19. Myers MG, Norris JW, Hachinski VC, et al. Plasma norepinephrine in

stroke. Stroke 1981;12:200-4.

1. Benedict CR, Loach AB. Clinical significance of plasma adrenaline and noradrenaline concentrations in patients with subarachnoid hemorrhage. J Neurol Neurosurg Psychiatry 1978;41:113-7.
2. Yuki K, Kodama Y, Onda J, et al. Coronary vasospasm following subarachnoid hemorrhage as a cause of stunned myocardium: a case report. J Neurosurg 1991;75:308-11.
3. Svigelj V, Grad A, Tekavcic I, et al. Cardiac arrhythmia associated with reversible damage to insula in patients with subarachnoid hemorrhage. Stroke 1994;25:1053-5.
4. Frontera JA, Fernandez A, Claassen J, et al. Hyperglycemia after SAH: predictors, associated complications, and impact on outcome. Stroke 2006;37:199-203.
5. Parkinson D, Stephensen S. Leukocytosis and subarachnoid hemor- rhage. Surg Neurol 1984;21:132-4.
6. Yoshimoto Y, Tanaka Y, Hoya K. Acute systemic inflammatory response syndrome in subarachnoid hemorrhage. Stroke 2001;32: 1989-93.
7. Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycemia and prognosis of stroke in nondiabetic and Diabetic patients: a systematic overview. Stroke 2001;32:2426-32.
8. Vila N, Castillo J, Davalos A, et al. Proinflammatory cytokines and early neurological worsening in ischemic stroke. Stroke 2000;31: 2325-9.