In-Hospital Electrical Storm in Acute Myocardial Infarction　― Clinical Background and Mechanism of the Electrical Instability ―

Abstract

Background: Recurrent ventricular tachycardia (VT) and fibrillation (VF), the so-called “electrical storm” (ES) occurs at various stages of acute myocardial infarction (AMI), but its incidence, background, and short-term prognosis remain unclear.

Methods and Results: A retrospective observational study was performed using the registry database of the Tokyo CCU Network. The individual data of 6,003 patients with AMI during 2011–2012 was corrected. ES was defined as more than 3 episodes of sustained VT/VF during a 24-h period as first documented after hospitalization. ES occurred in 55 patients after admission (0.9%). The ES(+) group had more severe heart failure (Killip class >III), more extensive MI (peak-CK), greater inflammatory reaction (CRP), history of diabetes, and more frequent application of hemodialysis as compared with the ES(−) group (n=5,865). When the ES patients were divided into Early-ES (n=37: ES occurred ≤48 h after the onset of MI) and Late-ES (n=15 >48 h after onset of MI) groups, logistic regression analysis revealed that Early-ES was associated with severity of MI, whereas Late-ES was related to systemic disorders, including inflammation, renal dysfunction, or diabetes. Late-ES was an independent predictor of in-hospital death.

Conclusions: In-hospital ES was a rare clinical manifestation of AMI. The features and background of the ES varied as time elapsed after admission for MI.

Recurrent sustained ventricular tachyarrhythmia (VTA), including ventricular tachycardia (VT) and ventricular fibrillation (VF), rarely occurs in the acute or subacute phase of myocardial infarction (MI), but the exact incidence of this phenomenon observed in the coronary care unit (CCU) remains unknown. It is known as an “electrical storm” (ES).

To date, VTA, including a single episode of VT or VF, seen in the acute phase of MI has been associated with a severe clinical condition derived from broad myocardial injury,^1–7 spontaneous or intentional coronary reperfusion,^8,9 electrolyte imbalance,^5,10 cardiogenic shock^4,6 and high sympathetic nervous activity.^11,12 Clinical documentation of VTA is also a marker of poor in-hospital prognosis.^2–7 The prevalence, clinical background, prognosis, and effective therapy for ES associated with MI have been reported only by several small cohort studies,^12–17 so the precise clinical features remain unclear. We conducted a large-scale study, using the registry data of the Tokyo CCU Network, with a primary aim of evaluating the clinical characteristics and independent predictors of ES occurring during the acute or subacute phase of MI.

The mechanism and clinical features of lethal VTAs have been shown to differ during the different stages post-MI (i.e., the acute and subacute phases of MI both in animal experiments^18,19 and clinical practice^2,4–6,20). The clinical significance of specific VTAs such as VF, VT, and non-sustained VT in terms of the influence on patients’ prognosis varies over time after the onset of MI.^1–7 Thus, the second aim of this study was to compare the clinical characteristics, background, and prognosis between patients with ES during the early and late phase after acute MI (AMI).

Methods

Study Population

The Tokyo CCU Network database has been conducted as an ongoing multicenter registry collecting the clinical information of cases of AMI in patients admitted to the CCU or intensive care unit (ICU) of leading hospitals capable of advanced cardiac care in the Tokyo metropolitan area (67 facilities).²¹ In 2011, we had 4,683 patient records from network hospitals. The detailed data of the patients were provided by all hospitals for 2,811 patients (76% males, average age 70±12 years). In 2012, 4,773 patients with AMI were registered and of them, detailed individual data were provided for 3,192 patients. We carefully confirmed whether the patients had experienced any sustained VT, including pulseless VT and/or VF, in the acute and/or subacute phase; thus, 419 patients with MI had undergone bystander automatic electrical defibrillator applications (84 patients), received DC cardioversions by ambulance crew (attendant) (60 patients), or had documentation of sustained VTA after hospitalization (275 patients). Sustained VTA is defined as a ventricular rhythm >100 beats/min lasting ≥30 s or requiring termination because of hemodynamic instability. We then collected more detailed information, including the time, number, and content of the lethal VTAs, the number of DC cardioversions, place where either VT or VF was first documented, time interval between the onset of MI and the first sustained VT/VF episode, time interval between the coronary intervention and the first VT/VF episode, and various treatments for complications.

After perusal of the individuals’ data, there were 55 patients (46 males, average age 65±13 years) who experienced ES defined as >3 episodes of an individual documentation of sustained VTA during a 24-h period,^22,23 as first documented after hospitalization. Next, we analyzed the background and prognosis of the patients with ES and compared them with patients who did not experience ES during their hospitalization (n=5,865). The 83 patients with ES documented before hospitalization were also excluded from this study.

Of the 55 patients with documented in-hospital ES, the precise time of the onset of the MI could be obtained for 52 patients. According to the interval between the onset of MI and the ES, we divided the patients into 2 groups: 0–48 h after the onset of MI (Early-ES group: 37 patients) and ≥48 h after the onset of MI (Late-ES group: 15 patients). The clinical features, background and prognosis were compared with those of the control group for each group to clarify the independent predictors of ES in each group.

Statistical Analysis

Categorical variables were tested using the Chi-square test or, if not applicable, Fisher’s exact test. Continuous variables are presented as median values (1st quantile–3rd quantile values) or mean value ± standard deviation. These variables were tested using Student’s t-test or the Mann-Whitney test. To identify the significant predictors of the in-hospital emergence of ES, univariate and multivariate logistic regression analyses were performed with 2 models: model 1 analyzed the independent predictors using the overall patients with ES as compared with control patients, and model 2 analyzed the predictors for each of the Early-ES and Late-ES groups, which were divided by the overall ES, using a multivariate logistic regression analysis. The variables selected were age, sex, coronary risk factors, Killip classification on admission, history of MI, responsible coronary arteries for MI, multivessel disease, degree of coronary stenosis, percutaneous coronary intervention (PCI), time interval between the onset of MI and PCI, coronary flow after PCI (TIMI), rate of coronary arterial bypass grafting (CABG), various laboratory data on admission, temporal pacing for bradyarrhythmias, and hemodialysis. To identify the significant predictors of in-hospital death of Early-ES or Late-ES, univariate and multivariate logistic regression analyses were also performed. Only the variables with a P≤0.05 in the univariate analysis were selected as explanatory variables for the subsequent multivariate analysis. All probability values were 2-tailed, and values of P<0.05 were considered statistically significant. The statistical analyses were performed using R language (R Core Team 2016, R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria).

Results

Clinical Background of ES Occurring During Hospitalization for AMI

Univariate Analysis The demographic and various clinical parameters for both ES(+) group (n=55) and ES(−) group (n=5,865) are comparatively shown in Table 1. The ES(+) group had more male patients than the ES(−) group.

Table 1. Characteristics of the ES(+) Group as Compared With Control Subjects (ES(−) Group)

Parameters	ES(+) group (n=55)	ES(−) group (n=5,865)	P value
Age	64.0 (56.00–75.00)	69.0 (59.00–78.00)	0.062
Sex (male %)	48 (87.3%)	4,402 (75.1%)	0.040
PR	90.0 (73.50–104.20)	78.0 (66.00–91.00)	0.007
Systolic BP	120.0 (95.00–133.00)	133.0 (113.00–155.00)	<0.001
Diastolic BP	68.0 (56.00–82.50)	78.0 (65.00–92.00)	0.003
MI site			0.152
Anterior	32 (59.3%)	2,672 (49.4%)
Inferoposterior	22 (40.7%)	2,737 (50.6%)
Coronary risk factors
Hypertension	33 (62.3%)	3,396 (62.6%)	1.000
Diabetes	23 (43.4%)	1,826 (33.6%)	0.145
Dyslipidemia	20 (37.7%)	2,493 (45.9%)	0.269
Hyperuricemia	2 (3.8%)	288 (5.3%)	1.000
Smoking	15 (28.3%)	2,074 (38.2%)	0.156
BMI	23.90 (21.630–25.950)	23.53 (21.260–25.950)	0.835
Killip classification			<0.001
I	18 (36.0%)	3,990 (75.5%)	I & II vs. III & IV
II	6 (12.0%)	642 (12.1%)
III	10 (20.0%)	285 (5.4%)
IV	16 (32.0%)	367 (6.9%)
Previous history of MI			0.789
None	42 (84.0%)	4,591 (84.7%)
Once	6 (12.0%)	513 (9.5%)
More than twice	2 (4.0%)	315 (5.8%)
STEMI or NSTEMI
STEMI	50 (89.8%)	4,780 (79.8%)	0.076
NSTEMI	5 (10.2%)	1,085 (20.8%)
Culprit coronary artery	n=50	n=5,129	0.143
LMT	5 (10.0%)	125 (2.4%)
LAD	21 (42.0%)	2,393 (46.6%)
LCX	7 (14.0%)	769 (15.0%)
RCA	17 (34.0%)	1,798 (35.0%)
Other lesions	0 (0%)	39 (0.8%)
Presence of other coronary lesions	33 (67.3%)	2,426 (48.9%)	0.014
PCI performance ratio	44 (91.7%)	4,428 (84.6%)	0.227
MI-PCI time interval (h)			0.084
NA	4 (8.3%)	808 (15.4%)
<3	21 (43.8%)	2,129 (40.7%)
3–12	11 (22.9%)	1,470 (28.1%)
12–24	8 (16.7%)	342 (6.5%)
>24	4 (8.3%)	487 (9.3%)
Post-PCI TIMI			<0.001
0	1 (2.3%)	78 (1.8%)
1	4 (9.1%)	39 (0.9%)
2	6 (13.6%)	192 (4.3%)
3	33 (75.0%)	4107 (93.0%)
CABG performance ratio	3 (5.6%)	281 (5.0%)	0.753
Laboratory data on admission
Peak-CK (U/L)	4,117.0 (1,644.00–7,412.00)	1,469.0 (612.50–3,197.00)	<0.001
WBC (number/μL)	10,770.0 (7,400.00–14,700.000)	9,000.0 (6,900.00–11,540.00)	0.002
CRP (mg/dL)	0.635 (0.225–2.700)	0.240 (0.100–0.900)	<0.001
BNP (pg/mL)	247.5 (110.60–894.00)	120.0 (36.85–409.90)	0.003
Creatinine (mg/dL)	1.20 (0.880–1.790)	0.86 (0.700–1.100)	<0.001
Uric acid (mg/dL)	6.85 (5.700–8.175)	5.80 (4.800–6.900)	0.001
Kallium (mEq/L)	4.10 (3.800–4.500)	4.10 (3.800–4.400)	0.460
Glucose (mg/dL)	194.0 (146.80–267.00)	150.0 (122.00–202.00)	<0.001
Total cholesterol (mg/dL)	171.0 (141.50–197.50)	186.0 (157.00–217.00)	0.032
LDL-C (mg/dL)	97.0 (70.00–120.50)	115.0 (91.00–141.00)	0.003
Triglyceride (mg/dL)	85.5 (60.00–132.00)	97.0 (62.00–148.00)	0.405
Treatment for MI
Circulatory support IABP and/or PCPS	37 (67.3%)	1015 (14.3%)	<0.001
Mechanical ventilation	36 (65.5%)	772 (13.1%)	<0.001
Temporary pacing	19 (38.8%)	551 (10.8%)	<0.001
HD or CHDF	14 (29.2%)	275 (5.1%)	<0.001
In-hospital mortality	23 (41.8%)	383 (6.5%)	<0.001

BP, blood pressure on admission; CABG, coronary artery bypass grafting; CHDF, continuous hemodiafiltration; CK, creatine phosphokinase; CRP, C-reactive protein; ES, electrical storm; HD, hemodialysis; IABP, intra-aortic balloon pumping; LAD, left anterior descending artery; LCX, left circumflex artery; LMT, left main trunk; MI, myocardial infarction; PCI, percutaneous coronary intervention; PCPS, percutaneous cardiopulmonary support; PR, pulse rate on admission; RCA, right coronary artery; STEMI, ST-elevation MI; NSTEMI, non-ST-elevation MI.

With regards to hemodynamic variables on admission, pulse rate was faster and systolic and diastolic blood pressures were lower in the ES(+) group as compared with the ES(−) group.

The prevalence of various coronary risk factors did not differ between the 2 groups. The distribution of the Killip classification on admission significantly differed between groups, as there was a significantly greater proportion of patients with Killip III and IV classification in the ES(+) group than ES(−) group (52.0% vs. 12.3%).

Coronary angiography was performed in 50/55 patients (91%) in the ES(+) group and 5,124/5,865 patients (87%) in the ES(−) group, respectively. The culprit coronary lesions had a similar distribution in both groups, even though the prevalence of an left main trunk lesion tend to be greater in the ES(+) group. The ratio of patients who had at least one other coronary significant lesion in addition to the culprit lesion was greater in the ES(+) group than the ES(−) group (33/55 patients: 67.3% and 2,426/5,865 patients: 48.9%, respectively, P=0.009).

In total, 80% (44/55 patients) of the patients underwent PCI in the ES(+) group, while 75.5% (4,428/5,865 patients) did in the ES(−) group (NS). The time interval between the onset of MI and PCI was similar between the 2 groups. The TIMI classification after PCI showed a worse residual ischemia in the ES(+) group than ES(−) group.

With regards to the laboratory data on admission, the peak creatine phosphokinase (peak-CK), known to reflect the size of the MI, was significantly greater in the ES(+) group. It is noteworthy that the serum C-reactive protein (CRP) level on admission was significantly larger in the ES(+) group than ES(−) group. The serum levels of creatinine, glucose, and uric acid were also larger in the ES(+) group.

With regards to the invasive treatments applied in clinical practice to improve lethal conditions of the patients, the rate of application of circulatory support devices, including intra-aortic balloon pump (IABP) and/or percutaneous cardiopulmonary support (PCPS) was significantly greater in the ES(+) group than ES(−) group. The application of mechanical ventilation and hemodialysis (HD) treatment was more frequent in the ES(+) group. Interestingly, temporary pacing was indicated with a greater ratio in the ES(+) group than in the ES(−) group (38.8% vs. 10.8%: P<0.001).

In summary, ES was associated with a greater infarction size, more severe coronary atherosclerotic lesions, and more deteriorated hemodynamic status, and thus a severe clinical condition related broadly to MI was underlying the emergence of the ES, rendering it necessary to perform advanced cardiac care such as circulatory support. In addition, the presence of inflammation, chronic kidney disease or acute kidney injury, or a high-glucose level may aggravate the various etiologic factors leading to a recurrent form of the arrhythmia. Consequently, the in-hospital mortality was significantly higher in the ES(+) group than in the ES(−) group (41.8% vs. 6.5%; P<0.001).

Multivariate Analysis To determine the independent risk factors, those significantly associated with ES in the univariate models were included in a stepwise regression analysis with backward elimination. The selected variables for the final multivariate analysis were sex, Killip classification, peak-CK, serum CRP level on admission, and application of cardiac pacing and HD (HD or continuous hemodiafiltration [CHDF]).

After the logistic regression analysis, the independent predictors of in-hospital ES were Killip classification (odds ratio [OR]: 5.601, P<0.001), clinical use of temporary pacing for bradycardic events (OR: 3.980, P<0.001), CRP (OR: 1.073, P=0.039), peak-CK (OR: 1.005, P=0.049), and finally the clinical use of HD (OR: 2.360, P=0.049) (Table 2).

Table 2. Independent Predictors of In-Hospital Emergence of Electrical Storm

Parameters	Univariate analysis			Multivariate analysis
	OR	CI	P value	OR	CI	P value
Male sex	2.277	1.028–5.044	0.043	1.921	0.767–4.811	0.164
Killip classification (I & II vs. III & IV)	7.696	4.303–13.485	<0.001	5.601	2.721–11.532	<0.001
Peak-CK (per 100)	1.005	1.002–1.009	0.002	1.004	1.000–1.007	0.049
CRP	1.057	1.028–1.087	<0.001	1.073	1.004–1.148	0.039
Use of HD or CHDF	7.614	4.039–14.355	<0.001	2.360	1.003–3.553	0.049
Use of temporary pacing	5.256	2.939–9.401	<0.001	3.980	1.957–8.095	<0.001

CI, confidence interval; OR, odds ratio. Other abbreviations as in Table 1.

Difference in the Clinical Background and Features of Early- and Late-ES

As shown in Figure 1, the documented VTAs characteristically differed between the ES observed during the early phase of hospitalization (within 48 h after the onset of MI) (Early-ES) and that during the late phase (>48 h after the onset of MI) (Late-ES). Early-ES mainly comprised VF, whereas the VT was the mainstream of the arrhythmias in Late-ES (Figure 1A). Most cases of Early-ES occurred in the cardiac catheterization laboratory (Figure 1B). In contrast, Late-ES were first documented in the CCU or even in the general ward. As such, according to the timing of the ES from the onset of MI, the clinical features and environmental factors varied appreciably as time elapsed.

Figure 1.

Characteristic differences between the Early-ES and Late-ES groups. (A) Main VTAs comprising the ES. Early-ES mainly comprised VF, whereas VT was the mainstream arrhythmia in Late-ES. (B) Location where first sustained VTA was documented: mainly in the catheter laboratory for Early-ES patients but either in the ICU or general ward for the Late-ES group. ES, electrical storm; ICU, intensive care unit; VF, ventricular fibrillation; VT, ventricular tachycardia; VTA, ventricular tachyarrhythmia.

Therefore, we divided the ES(+) group into 2 groups: Early-ES group (37 patients, 33 male, average age 63.7±12.7 years) and Late-ES group (15 patients, 13 male, average age 68.4±11.9 years). Late-ES was associated with more severe ES because the number of DC shock deliveries was greater than that for Early-ES [4.0 (3.0–5.0) vs. 5.0 (3.5–9.5) times: median numbers (1st quartile–3rd quartile)] (Figure 2A). Late-ES was also associated with a poor prognosis as compared with Early-ES (in-hospital mortality: 11/15 (73.3%) vs. 11/37 (29.7%) P<0.001) (Figure 2B); however, even the in-hospital mortality in the Early-ES group was significantly greater than that in the control group (29.7% vs. 6.8% P<0.001, Table 3).

Figure 2.

Severity of the electrical storm (ES) and the patients’ prognosis compared between groups. (A) Box-and-whisker plot showing Late-ES associated with more concentrated clusters of arrhythmias, as the median number (1st quantile–3rd quantile) of DC shock deliveries was greater than for Early-ES. (B) Late-ES also associated with a poor prognosis as compared with Early-ES.

Table 3. Characteristics of Patients With Early-ES or Late-ES: Univariate and Multivariate Analyses of the Predictors in Both Groups

Variables	Control group (n=5,862)	Early-ES group (n=37) <48 h after onset of MI	Univariate analysis	Multivariate analysis			Late-ES group (n=15) ≥48 h after onset of MI	Univariate analysis	Multivariate analysis
			P value	OR	CI	P value		P value	OR	CI	P value
Age	69.0 (59.00– 78.00)	63.0 (55.00– 75.00)	0.043				69.0 (60.00– 76.00)	0.925
Sex (male %)	4,402 (75.1%)	33 (89.2%)	0.066				13 (86.7%)	0.265
Infarction site			0.573					0.112
Anterior	2,672 (49.4%)	20 (54.1%)					10 (71.4%)
Inferoposterior	2,737 (50.6%)	17 (45.9%)					4 (28.6%)
Coronary risk factors
Diabetes	1,826 (33.6%)	10 (29.4%)	0.603				12 (75%)	0.002	7.86	1.495– 41.300	0.015
Killip classification		Prevalence of class III+IV*	<0.001	7.945	3.522– 17.923	<0.001	Prevalence of class III+IV*	0.003	2.808	0.743– 10.609	0.128
I	3,990 (75.5%)	11 (34.4%)					6 (40%)
II	642 (12.1%)	3 (9.4%)	0.419				2 (13.3%)	0.373
III	285 (5.4%)	6 (18.8%)	<0.001				4 (26.7%)	0.001
IV	367 (6.9%)	12 (37.5%)	<0.001				3 (20%)	0.017
History of MI
None	4,591 (84.7%)	27 (81.8%)					13 (86.7%)
Once	513 (9.5%)	5 (15.2%)	0.302				1 (6.7%)	0.719
≥Twice	315 (5.8%)	1 (3.0%)	0.546				1 (6.7%)	0.912
NSTEMI/ STEMI+NSTEMI	1,085 (20.8%)	1 (2.9%)	0.033	5.333	0.710– 40.083	0.104	3 (25.0%)	0.724
Presence of other coronary lesions	2,426 (48.9%)	21 (63.6%)	0.097				11 (73.3%)	0.071
MI-PCI time interval (h)
NA	808 (15.4%)	1 (3.2%)					2 (14.3%)
<3	2,129 (40.7%)	18 (58.1%)	0.062				2 (14.3%)	0.333
3–12	1,470 (28.1%)	8 (25.8%)	0.163				3 (21.4%)	0.833
12–24	342 (6.5%)	2 (6.5%)	0.205				5 (35.7%)	0.034
>24	487 (9.3%)	2 (6.5%)	0.328				2 (14.3%)	0.613
Laboratory data on admission
Peak-CK (per 100)	1469.0 (612.50– 3,197.00)	5847.0 (3,376.00– 9,004.00)	0.002	1.009	1.003– 1.016	0.006	2065.0 (1,434.00– 3,936.00)	0.342
CRP	0.240 (0.100– 0.900)	0.300 (0.107– 1.448)	0.962				2.130 (0.900– 14.440)	<0.001	1.093	1.040– 1.148	<0.001
Glucose	150.0 (122.00– 202.00)	185.5 (142.50– 248.50)	0.014				225.0 (161.00– 309.80)	0.010
Creatinine	0.860 (0.700– 1.100)	1.005 (0.818– 1.432)	0.022				1.300 (1.168– 2.420)	0.004
Treatment of MI
Circulatory support
IABP	900 (16.2%)	11 (33.3%)	<0.001				6 (37.5%)	0.002
PCPS	16 (0.3%)	0 (0%)	0.991				1 (6.2%)	<0.001
IABP+PCPS	109 (2.0%)	12 (36.4%)	<0.001				5 (31.2%)	<0.001
Mechanical ventilation	590 (11.0%)	20 (60.6%)	<0.001				10 (62.5%)	<0.001
Temporary pacing	551 (10.8%)	14 (43.8%)	<0.001	4.358	1.970– 9.641	<0.001	3 (21.4%)	0.211
HD or CHDF	271 (5.1%)	6 (20.0%)	0.001				7 (46.7%)	<0.001	10.305	2.719– 39.050	<0.001
In-hospital mortality	383 (6.5%)	11 (29.7%)	<0.001				11 (73.3%)	<0.001

*Prevalence of class III and class IV in this group was compared with that of the control group. Abbreviations as in Tables 1,2.

Univariate and Multivariate Analyses for the Predictors of Early- and Late-ES

Table 3 shows the results of the univariate and multivariate analyses for the predictors of both categories of ES. In the univariate analysis, the Killip classification, glucose level, and clinical use of circulatory support, mechanical ventilation, and HD were shown to be the common predictors of both Early- and Late-ES. In the logistic regression analysis, the significant predictors of Early-ES were worse Killip class, use of temporary pacing, and peak-CK, whereas those of Late-ES were CRP level on the admission, use of HD, and diabetes.

Independent Predictors of In-Hospital Mortality (Table 4)

Of a total of 5,920 patients who were enrolled in this study, 406 died in the admitted hospitals (in-hospital mortality 6.8%). After logistic regression analysis, there were many independent prognosticators of in-hospital mortality, including older age, anterior infarction, greater peak-CK, WBC, higher creatinine level on admission, higher glucose level, lower total cholesterol level, current smoker, higher Killip class, other coronary lesions, multiple organ failure, temporary pacing application, and finally emergence of ES. The most powerful predictor was age, then the peak-CK. The third most important predictor was the emergence of ES. When the ES(+) group was divided into the Early-ES and Late-ES groups, Late-ES remained as a significant prognosticator. In contrast, Early-ES was eliminated from the significant variables (Table 4).

Table 4. Independent Predictors of In-Hospital Death

Parameters	OR	95% CI	P value
Age	1.070	1.044–1.097	<0.001 (1.063E-07)
Infarction site (anterior/inferoposterior)	2.584	1.515–4.406	<0.001 (4.888E-04)
Peak-CK (per 100)	1.012	1.007–1.017	<0.001 (2.396E-06)
WBC (per 1,000)	1.038	1.013–1.063	0.003
Creatinine	1.225	1.087–1.380	<0.001 (8.521E-04)
Glucose	1.002	1.000–1.004	0.036
Total cholesterol	0.994	0.989–0.999	0.016
Current smoker	1.850	1.113–3.077	0.018
Killip classification (I & II vs. III & IV)	2.093	1.167–3.752	0.013
Coronary lesions other than culprit	0.975	1.175–3.320	0.010
Multiple organ failure	8.256	1.743–49.168	0.009
Temporary pacing application	1.868	1.028–3.395	0.040
Electrical storm	10.053	3.142–32.172	<0.001 (1.007E-04)
Early	4.204	0.981–18.008	0.053
Late	48.444	6.654–352.674	<0.001 (1.275E-04)

CI, confidence interval; OR, odds ratio.

Discussion

The main findings of the present study were as follows. (1) The incidence of in-hospital ES was 0.92% among the patients admitted to Tokyo CCU Network hospitals because of AMI. (2) The appearance of ES was significantly related to the severity of the MI (peak-CK) and presence of heart failure (Killip classification) on admission, leading to more frequent use of circulatory support and mechanical ventilation. (3) Elevated CRP on admission, use of HD, and history of diabetes were also associated with ES. (4) Finally, the ES itself that appeared in the late phase of AMI was an independent predictor of in-hospital death.

When the ES was categorized into 2 groups (i.e., Early-ES (≤48 h after the onset of MI) and Late-ES (>48 h), Early-ES emerged in the cardiac catheterization laboratory in the majority of the patients and was more likely to appear as VF than VT. In contrast, Late-ES occurred in either the CCU or even the general ward and appeared more as VT than VF, and Late-ES was particularly of a more recurrent form as shown by the greater number of DC shock deliveries. Early-ES was significantly associated with the severity of heart failure (Killip classification) and the use of temporary pacing, whereas Late-ES was closely related to the presence of inflammation (CRP titer), diabetes, and use of HD.

Clinical Background of Early-ES

In the present evaluation, Early-ES was associated with a hemodynamically deteriorated condition (Killip III or IV), leading to greater application of circulatory support and mechanical ventilation. Thus, there might have been a significant number of patients with so-called secondary (non-primary) VF included in this population. Secondary VF is commonly defined as VF occurring in the presence of heart failure and cardiogenic shock and has been shown to be associated with an extremely higher in-hospital mortality (up to approximately 50% of the 1-month mortality) as compared with control subjects.^5,6

Another important mechanism of lethal VTA is reperfusion arrhythmias in this stage of the MI. Correspondingly, a majority of the patients in the Early-ES group experienced the ES in the cardiac catheterization laboratory; stated differently, their arrhythmias occurred in the peri-procedural (PCI) period. The incidence of VF during the reperfusion period has been 1.5–5.9% in previous reports.^9,24–26 Although the incidence of ES associated with coronary reperfusion remains unknown, it might be less than 1.0%, which is consistent with our report. It has also been shown that reperfusion VF accounts for more than 20% of the VF occurring during the first 48 h after the onset of MI.⁹ In contrast to secondary VF, the prognosis of reperfusion VF is not so unfavorable, as in-hospital mortality was 3.0–16.7% in the patients with reperfusion VF.⁹ The fact that the mortality rate of our Early-ES group was relatively low, as compared with the Late-ES group, may reflect that a transient mechanism may have contributed to some degree to the appearance of Early-ES.

There are several factors related to temporary pacing that may cause VTAs. Temporary pacing applied during the acute phase of MI has been shown to induce a significant number of device malfunctions, including both pacing and sensing failures.²⁷ Such pacemaker failures may cause marked bradycardia or cardiac arrest, and R-on-T phenomenon, leading to inadvertent occurrence of VTAs.²⁸ It was also shown that VTAs are likely to be induced during the insertion of pacing catheters, probably because of a tension effect on the attached endocardium during catheter manipulation.^27,29 Finally, it was reported that high-degree atrioventricular block observed in the acute or subacute phase of MI is the most powerful predictor of cardiac death.³⁰ Thus, the use of temporary pacing may become an independent predictor for the appearance of Early-ES and even for in-hospital death (Table 4).

Clinical Background of Late-ES

Late-ES occurred in the patients with various comorbidities, including inflammatory reactions, chronic kidney disease and acute kidney injury, and diabetes.

AMI has been shown to trigger an inflammatory reaction, which facilitates myocardial injury.³¹ It has been also shown that inflammatory markers such as CRP correlate with the extent of MI and the patient’s outcome after AMI.^32,33 Doi et al demonstrated in their randomized controlled study that early administration of n-3 polyunsaturated fatty acids (i.e., eicosapentaenoic acid (EPA), which is known to possess anti-inflammatory effects) significantly reduced the incidence of ventricular arrhythmias, which occurred >48 h after the onset of MI together with lower CRP level on the 3rd–4th day after the onset of MI, compared with the control patients.³⁴ In the present study, CRP measured at the time of admission was one of the independent predictors of ES, particularly Late-ES, but not for Early-ES.

It was shown by several previous reports that advanced renal dysfunction is an independent predictor of a poor prognosis after AMI.^35,36 It was also revealed that the presence of chronic kidney disease is significantly associated with the emergence of VTAs after AMI.^4,35 Thus it is thought to be reasonable that the use of HD was a significant marker of ES. In addition, a past history of diabetes seemed to be associated with Late-ES in our study. Diabetes is a well-known pathological condition leading to amplification of the inflammatory process and aggravation of kidney function.

Study Limitations

There were some limitations to this study. Because the in-hospital emergence of ES is a very rare clinical manifestation in AMI (0.9%), the number of patients with ES was low despite the total number of registered patients with an MI having been more than 6,000. Particularly, the Late-ES group had only 15 patients. Nevertheless, the characteristic differences in the background, clinical features, and prognosis among the Early-ES group, Late-ES group, and control patients were clearly recognized in this study using reasonable methods of statistical analysis.

Because Late-ES was associated with severe systemic illness, a high CRP value may indicate the state of complicating infectious disease such as pneumonia rather than exclusively reflecting the changes associated with MI. However, in this registry, the reported causes of death in the Late-ES groups were cardiogenic shock in 4 patients, lethal arrhythmias in 3, reattack of MI in 2, multiple organ failure in 1, and interstitial pneumonia in 1. Thus it is inferred that the contribution of the coexisting infectious disease, if any, was quite small to the elevation of CRP.

There were some unavailable and missing data in the individual data sheets. Because the proportion of defects could not be ignored in the left ventricular ejection fraction, in-hospital documentation of atrial fibrillation, and currently prescribed medications before admission, those variables were excluded from this analysis. Finally, the long-term outcome could not be corrected in this registry.

Conclusions

This metropolitan registry study demonstrated that ES was significantly related to the severity of MI and the presence of heart failure on admission, particularly for Early-ES. In contrast, the level of the inflammatory reaction and presence of renal dysfunction and diabetes were more important predisposing factors for Late-ES. To improve the prognosis of patients with ES associated with AMI, close attention should be paid not only to the cardiac conditions, including cardiac function and electrophysiological data, but also the systemic condition, including the status of comorbidities.

Acknowledgments

The authors express their special gratitude to all members of the Tokyo CCU Network and Ms. Nobuko Yoshida (Tokyo CCU Network office) for data collection and to Mr. John Martin for his linguistic assistance.

References

• 1.   Gheeraert PJ, Henriques JP, De Buyzere ML, Voet J, Calle P, Taeymans Y, et al. Out-of-hospital ventricular fibrillation in patients with acute myocardial infarction: Coronary angiographic determinants. J Am Coll Cardiol 2000; 35: 144–150.
• 2.   Cheema AN, Sheu K, Parker M, Kadish AH, Goldberger JJ. Nonsustained ventricular tachycardia in the setting of acute myocardial infarction: Tachycardia characteristics and their prognostic implications. Circulation 1998; 98: 2030–2036.
• 3.   Mont L, Cinca J, Blanch P, Blanco J, Figueras J, Brotons C, et al. Predisposing factors and prognostic value of sustained monomorphic ventricular tachycardia in the early phase of acute myocardial infarction. J Am Coll Cardiol 1996; 28: 1670–1676.
• 4.   Piccini JP, Berger JS, Brown DL. Early sustained ventricular arrhythmias complicating acute myocardial infarction. Am J Med 2008; 121: 797–804.
• 5.   Volpi A, Cavalli A, Santoro L, Negri E. Incidence and prognosis of early primary ventricular fibrillation in acute myocardial infarction: Results of the Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico (GISSI-2) database. Am J Cardiol 1998; 82: 265–271.
• 6.   Henkel DM, Witt BJ, Gersh BJ, Jacobsen SJ, Weston SA, Meverden RA, et al. Ventricular arrhythmias after acute myocardial infarction: A 20-year community study. Am Heart J 2006; 151: 806–812.
• 7.   Goldberg RJ, Yarzebski J, Spencer FA, Zevallos JC, Lessard D, Gore JM. Thirty-year trends (1975–2005) in the magnitude, patient characteristics, and hospital outcomes of patients with acute myocardial infarction complicated by ventricular fibrillation. Am J Cardiol 2008; 102: 1595–1601.
• 8.   Solomon SD, Ridker PM, Antman EM. Ventricular arrhythmias in trials of thrombolytic therapy for acute myocardial infarction: A meta-analysis. Circulation 1993; 88: 2575–2581.
• 9.   Demidova MM, Carlson J, Erlinge D, Platonov PG. Predictors of ventricular fibrillation at reperfusion in patients with acute ST-elevation myocardial infarction treated by primary percutaneous coronary intervention. Am J Cardiol 2015; 115: 417–422.
• 10.   Nordrehaug JE, von der Lippe G. Hypokalaemia and ventricular fibrillation in acute myocardial infarction. Br Heart J 1983; 50: 525–529.
• 11.   Piccini JP, Hranitzky PM, Kilaru R, Rouleau JL, White HD, Aylward PE, et al. Relation of mortality to failure to prescribe beta blockers acutely in patients with sustained ventricular tachycardia and ventricular fibrillation following acute myocardial infarction (from the VALsartan In Acute myocardial iNfarcTion trial [VALIANT] Registry). Am J Cardiol 2008; 102: 1427–1432.
• 12.   Nademanee K, Taylor R, Bailey WE, Rieders DE, Kosar EM. Treating electrical storm: Sympathetic blockade versus advanced cardiac life support-guided therapy. Circulation 2000; 102: 742–747.
• 13.   Hanson EC, Levine FH, Kay HR, Leinbach RC, Gold HK, Daggett WM, et al. Control of postinfarction ventricular irritability with the intraaortic balloon pump. Circulation 1980; 62(2 Pt 2): I130–I137.
• 14.   Bänsch D, Oyang F, Antz M, Arentz T, Weber R, Val-Mejias JE, et al. Successful catheter ablation of electrical storm after myocardial infarction. Circulation 2003; 108: 3011–3016.
• 15.   Ricci JM, Dukkipati SR, Pica MC, Haines DE, Goldstein JA. Malignant ventricular arrhythmias in patients with acute right ventricular infarction undergoing mechanical reperfusion. Am J Cardiol 2009; 104: 1678–1683.
• 16.   Miwa Y, Ikeda T, Mera H, Miyakoshi M, Hoshida K, Yanagisawa R, et al. Effects of landiolol, an ultra-short-acting β₁-selective blocker, on electrical storm refractory to class III antiarrhythmic drugs. Circ J 2010; 74: 856–863.
• 17.   Murata H, Miyauchi Y, Hayashi M, Iwasaki YK, Yodogawa K, Ueno A, et al. Clinical and electrocardiographic characteristics of electrical storms due to monomorphic ventricular tachycardia refractory to intravenous amiodarone. Circ J 2015; 79: 2130–2137.
• 18.   Harris AS, Rojas AG. The initiation of ventricular fibrillation due to coronary occlusion. Exp Med Surg 1943; I-105.
• 19.   Scherlag BJ, Lazzara R. Ischemic arrhythmias: Basic mechanisms. In: Mandel WJ, editor. Cardiac arrhythmias: Their mechanisms, diagnosis, and management Lippincott, 1980; 366–395.
• 20.   Goldberg R, Szklo M, Tonascia J, Kennedy HL. Length of time between hospital admission and ventricular fibrillation or cardiac arrest complicating acute myocardial infarction: Effect on prognosis. Johns Hopkins Med J 1979; 145: 187–191.
• 21.   Miyachi H, Takagi A, Miyauchi K, Yamasaki M, Tanaka H, Yoshikawa M, et al. Current characteristics and management of ST elevation and non-ST elevation myocardial infarction in the Tokyo metropolitan area: From the Tokyo CCU network registered cohort. Heart Vessels 2016; 31: 1740–1751.
• 22.   Hohnloser SH, Al-Khalidi HR, Pratt CM, Brum JM, Tatla DS, Tchou P, et al. Electrical storm in patients with an implantable defibrillator: Incidence, features, and preventive therapy: Insights from a randomized trial. Eur Heart J 2006; 27: 3027–3032.
• 23.   Agone A. Electrical storm: Incidence, prognosis and therapy. J Atr Fibrillation 2015; 8: 1150–1155.
• 24.   Henriques JP, Gheeraert PJ, Ottervanger JP, de Boer MJ, Dambrink JH, Gosselink AT, et al. Ventricular fibrillation in acute myocardial infarction before and during primary PCI. J Cardiol 2005; 105: 262–266.
• 25.   Mehta RH, Harjai KJ, Grines L, Stone GW, Boura J, Cox D, et al; Primary Angioplasty in Myocardial Infarction (PAMI) Investigators. Sustained ventricular tachycardia or fibrillation in the cardiac catheterization laboratory among patients receiving primary percutaneous coronary intervention: Incidence, predictors, and outcomes. J Am Coll Cardiol 2004; 43: 1765–1772.
• 26.   Mehta RH, Starr AZ, Lopes RD, Hochman JS, Widimsky P, Pieper KS, et al; APEX AMI Investigators. Incidence of and outcomes associated with ventricular tachycardia or fibrillation in patients undergoing primary percutaneous coronary intervention. JAMA 2009; 301: 1779–1789.
• 27.   Austin JL, Preis LK, Crampton RS, Beller GA, Martin PR. Analysis of pacemaker malfunction and complications of temporary pacing in the coronary care unit. Am J Cardiol 1982; 49: 301–306.
• 28.   Ren X, Hongo RH. Polymorphic ventricular tachycardia from R-on-T pacing. J Am Coll Cardiol 2009; 53: 218.
• 29.   Sclarovsky S, Zafrir N, Strasberg B, Kracoff O, Lewin RF, Arditi A, et al. Ventricular fibrillation complicating temporary ventricular pacing in acute myocardial infarction: Significance of right ventricular infarction. Am J Cardiol 1981; 48: 1160–1166.
• 30.   Bloch Thomsen PE, Jons C, Raatikainen MJ, Moerch Joergensen R, Hartikainen J, Virtanen V, et al; Cardiac Arrhythmias and Risk Stratification After Acute Myocardial Infarction (CARISMA) Study Group. Long-term recording of cardiac arrhythmias with an implantable cardiac monitor in patients with reduced ejection fraction after acute myocardial infarction: The Cardiac Arrhythmias and Risk Stratification After Acute Myocardial Infarction (CARISMA) study. Circulation 2010; 122: 1258–1264.
• 31.   Frangogiannis NG, Smith CW, Entman ML. The inflammatory response in myocardial infarction. Cardiovasc Res 2002; 53: 31–47.
• 32.   Suleiman M, Aronson D, Reisner SA, Kapeliovich MR, Markiewicz W, Levy Y, et al. Admission C-reactive protein levels and 30-day mortality in patients with acute myocardial infarction. Am J Med 2003; 115: 695–701.
• 33.   Pietilä KO, Harmoinen AP, Jokiniitty J, Pasternack AI. Serum C-reactive protein concentration in acute myocardial infarction and its relationship to mortality during 24 months of follow-up in patients under thrombolytic treatment. Eur Heart J 1996; 17: 1345–1349.
• 34.   Doi M, Nosaka K, Miyoshi T, Iwamoto M, Kajiya M, Okawa K, et al. Early eicosapentaenoic acid treatment after percutaneous coronary intervention reduces acute inflammatory responses and ventricular arrhythmias in patients with acute myocardial infarction: A randomized, controlled study. Int J Cardiol 2014; 176: 577–582.
• 35.   McCullough PA, Soman SS, Shah SS, Smith ST, Marks KR, Yee J, et al. Risks associated with renal dysfunction in patients in the coronary care unit. J Am Coll Cardiol 2000; 36: 679–684.
• 36.   Beattie JN, Soman SS, Sandberg KR, Yee J, Borzak S, Garg M, et al. Determinants of mortality after myocardial infarction in patients with advanced renal dysfunction. Am J Kidney Dis 2001; 37: 1191–1200.