- Email: [email protected]
Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event
HLC 2888 No. of Pages 8
Heart, Lung and Circulation (2019) xx, 1–8 1443-9506/04/$36.00 https://doi.org/10.1016/j.hlc.2019.03.020
ORIGINAL ARTICLE
Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event Jia-Li Feng, BMed, MMed, PhD School of Population and Global Health, Faculty of Medicine, Dentistry and Health Sciences, The University of Western Australia, Perth, WA, Australia Received 4 January 2019; received in revised form 2 February 2019; accepted 27 March 2019; online published-ahead-of-print xxx
Background
Sudden cardiac death (SCD) still accounts for the majority of deaths from the four major cardiovascular events (myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF), and stroke) despite substantial progress on prevention.
Methods
Four separate cohorts (one for each of the four major cardiovascular conditions) were captured through person-linked hospital morbidity and mortality data collections between 2000 and 2009 and followed-up for 11.5 years. The incidence rate for each cohort was total SCD cases divided by sum of follow-up time for each individual alive. Kaplan–Meier survival curve was used to calculate unadjusted risk of SCD. Predictors of SCD were identified by fitting multivariable adjusted Cox regression models in each of the cohorts.
Results
There were 1,174 cases of SCD from 53,614 total CVD events across the cohorts (35.6% for MI, 15.6% for HF, 22.4% for AF, 26.4% for stroke). The incidence rate and unadjusted risk of SCD were both highest after incident hospitalisation for HF, followed by MI, stroke and AF. The elevated risk of SCD was independently associated with MI, HF, arrhythmias, peripheral artery disease, diabetes, chronic kidney disease, and prior coronary heart disease (hazard ratios ranging from 1.1 to 2.8). Early revascularisation is protective in 28-day survivors after an incident MI event.
Conclusions
An appreciable incidence of SCD following an incident event of MI, HF, AF and stroke deserves greater prevention efforts. Major medical conditions such as MI, HF, peripheral artery disease, and arrhythmias are risk markers of SCD and coronary revascularisation is protective.
Keywords
Incidence Predictor Sudden cardiac death Major cardiovascular event
Introduction Four major cardiovascular events, myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF), and stroke annually affect around 1.9% of the total general population in Australia and 2.6% in the United States (US) [1–4]. Sudden cardiac death (SCD) still accounts for the majority of deaths from the four major cardiovascular events in spite of substantial progress on prevention [5]. Few observational
studies are available to report the incidence rates of SCD after AF or stroke and limited data show long-term incidence rates of SCD after MI or HF. The observational communitybased Olmsted County study showed a 30-day incidence of SCD (1.2%) and a 5-year unadjusted risk of 6.9% in 3,000 residents after MI from 1979 to 2005 [6]. A single-centre study in China reported the 5-year SCD unadjusted risk of 5.0% in 40 days survivors of MI with an ejection fraction 35% [7]. Another single-centre study in the US of HF patients with
Email: [email protected]
© 2019 Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia and New Zealand (CSANZ). Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
2
J.-L. Feng
preserved ejection fraction who underwent a cardiac procedure reported the unadjusted risk of SCD were 1.0% at 1 year and 3.0% at 5 years [8]. The impact of long-term conditions on the prognosis of these four major cardiovascular events is poorly understood. Thus identifying predictors of SCD as an indication of the prognosis becomes a critical first step for health care planning and future reduction of SCD risk. A Charlson Comorbidity Index (CCI) score of three or more and ejection fraction 25% were risk markers of SCD while coronary revascularisation reduced the risk of SCD in post-MI patients [6]. The single-centre study from the US mentioned above identified predictors of SCD such as diabetes and MI following an admission of HF with preserved ejection fraction [8]. However, few large-scale observational data are available investigating predictors of SCD in patients with AF, although randomised controlled trials suggest diabetes, hypertension, coronary artery disease were predictive of SCD in a population of patients with AF [9–11]. No study has specifically investigated the predictors of SCD following stroke [12]. Although the four major cardiovascular diseases (CVDs) appear to share some common cardiovascular risk factors such as diabetes and hypertension, there is a lack of investigation of such common medical conditions as predictors of SCD following the four major cardiovascular events altogether. The result of this investigation ideally would, first, provide a whole picture of the incidence burden of SCD in the survivors of the four major cardiovascular events. Second, it would broaden our understanding of SCD predictors and facilitate optimisation of treatment after the four major cardiovascular events for a further reduction in SCD incidence. Western Australia’s (WA) data linkage provides a unique source of examining the incidence and predictors of SCD in a whole-population setting following hospitalisation for incident MI, HF, AF or stroke patients over a longterm follow-up.
Methods Data Source and Study Cohort Person-linked mortality and morbidity data for this study were obtained from the Mortality Register and Hospital Morbidity Data Collection and linked by WA Data Linkage System using probabilistic matching [13]. The linked dataset included death records for CVD and all public and private hospitalisations for CVD in WA from 1 January 1985 to 30 June 2011. Variables examined in the linked dataset included date of death, underlying and associated causes of death, indication of post-mortem, demographic fields (including age, sex, and Indigenous status), hospital admission and discharge dates, hospital discharge diagnosis fields (principal and 20 additional), and 11 fields of in-hospital procedures. Ethical approval for this study was obtained from the Human Research Ethics Committees of the WA Department of Health (#RA/4/1/1491) and The University of Western Australia (#2014/55).
The study included four cohorts of patients with a principal discharge diagnosis of incident MI, HF, AF or stroke from 1 January 2000 to 31 December 2009. The four cohorts were identified separately from each other, and were not mutually exclusive. The ICD codes used to identify the cohorts are listed in the Supplementary Table 1. The cohort of incident MI included patients with MI recorded in the principal discharge diagnosis field and no hospitalisation history of MI recorded in any discharge diagnosis fields in the preceding 15 years. Following the same practice, incident HF, AF, and stroke patients were identified separately. Validation studies on the coding of MI, HF, AF, and stroke in the Hospital Morbidity Data showed 92.2% of positive predictive values for MI, 99.5% for HF, 84.6% for AF, 85.0% for stroke [14–17]. Lower sensitivity for MI was noted from the Hospital Morbidity Data in the very old patients [14], thus patients 85years and -older were excluded from all four cohorts. In addition, patients aged <35 years were excluded as they are likely to have a lower rate of SCD and different predictors compared to those 35 years old [18]. Of note is that patients surviving <28 days following the admission date of incident MI were excluded from this study (<5% of the MI cohort) because a fatal first MI is less predictable thus providing few prevention efforts [19]. There were a total of 53,614 incident events of MI, HF, AF and stroke, representing 50,877 individuals, with 5.1% of individuals appearing in more than one incident cohort. As SCD incidence rate and predictors were identified separately in each of the four cohorts, individuals appearing in more than one cohort were retained in analyses.
Potential Predictors for SCD Potential predictors for SCD included common medical conditions, cardiac procedures, sex, age, and Indigenous status. The medical conditions included: hypertension, MI, HF, AF, arrhythmias (ventricular tachycardia, ventricular fibrillation, and cardiac arrest), diabetes, chronic kidney disease, peripheral arterial disease, stroke, chronic obstructive pulmonary disease, and coronary heart disease (CHD). Each of the four cohorts was not considered as a potential predictor in its respective cohort. For example, MI was not considered as a potential predictor in the cohort of incident MI cases. Potential predictors of SCD were identified from the 20 additional discharge diagnosis fields during the incident admission and any discharge diagnosis fields in a fixed 15-year lookback period from the discharge date of the incident admission. CHD was only considered in the incident MI cohort because in the other three cohorts, MI as the potential predictor was examined and this was identified from the 15-year prior hospitalisation history only. Coronary procedures were investigated to determine whether they are predictors of SCD in the incident MI cohort only, as coronary revascularisation is most commonly used in patients with CHD. Coronary procedures included coronary revascularisation (percutaneous coronary intervention (PCI) and coronary artery bypass grafting surgery (CABG)). Implantable cardioverter-defibrillator is an important treatment approach for SCD prevention and was, therefore, considered
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
3
Incidence and Predictors of Sudden Cardiac Death
Table 1 Characteristics of individuals with incident hospitalisation for myocardial infarction, heart failure, atrial fibrillation, and stroke, aged 35–84 years in Western Australia.
Mean Age, years (SD)
Myocardial infarction
Heart failure
Atrial fibrillation
Stroke
(n = 19,145)
(n = 8,294)
(n = 12,015)
(n = 14,160)
64.4 (12.3)
71.7 (10.9)
66.5 (11.8)
69.2 (12.0)
Age groups 35–54 years
25.2
9.7
17.9
15.1
55–69 years
37.6
25.8
37.7
28.8
70–84 years
37.2
64.5
44.4
56.1
70.8 4.7
57.4 5.7
59.1 1.7
55.6 4.1
Male Indigenous status Medical conditions§ myocardial infarction
–
17.9
6.2
9.0
heart failure
16.6
–
13.1
11.4
diabetes mellitus
24.9
34.6
13.5
25.8
chronic kidney disease
12.6
25.2
7.5
12.7
hypertension atrial fibrillation
55.4 14.9
62.8 38.1
37.4 –
66.0 22.7
arrhythmias*
5.5
4.7
2.0
2.2
strokez
4.1
7.6
2.4
–
peripheral artery disease
9.1
15.1
5.1
10.5
chronic obstructive pulmonary disease
10.9
24.8
9.8
13.2
21.8
–
–
–
5.2
7.8
4.7
3.9
0.2
0.8
0.3
0.2
coronary heart disease (excluding MI)
y
Prior procedures coronary revascularisation£ implantable cardioverter-defibrillator m
Procedures within 28 days of incident admission percutaneous coronary intervention
45.4
1.3
0.5
0.1
coronary artery bypass grafting
6.5
0.8
0.1
0.01
implantable cardioverter-defibrillator
0.5
1.5
0.2
0.03
Data are percentages (%) unless otherwise indicated. Abbreviations: SD, standard deviation; MI, myocardial infarction. § *
includes conditions at incident admission and prior medical history unless otherwise indicated; cases could have multiple medical conditions.
includes ventricular tachycardia, ventricular fibrillation, and cardiac arrest.
z
includes subarachnoid haemorrhage, intracerebral haemorrhage, occlusion and stenosis of precerebral arteries with cerebral infarction, and occlusion of cerebral arteries with cerebral infarction.
y
includes stable and unstable angina pectoris and chronic coronary heart disease, identified from prior medical history only.
£
includes percutaneous coronary intervention and coronary artery bypass grafting surgery.
m
identified within 28 days of the incident admission (inclusive).
as a potential protective factor. The procedures were included if they occurred in any of the 11 procedure fields within 28 days (inclusive) and prior procedures identified using a fixed 15year lookback period from the incident MI admission. Demographic factors were identified from the incident admission (for age and sex) and all admissions (for Indigenous status).
and results of post-mortem if carried-out. Follow-up was censored at the date of death or at 30 June 2011 (the end of the study period), whichever came first. Twenty-two (22) patients who died during follow-up and had a missing cause of death were excluded from all analyses (MI cohort (n = 7), HF (n = 6), AF (n = 2), stroke (n = 7)).
Follow-Up
Statistical Analysis
The primary endpoint was SCD. Four criteria used to identify SCD, were detailed in our previous paper [19], based on following elements using administrative data: place of death, proximity to hospital admission, prior and/or concurrent medical history, underlying and associated causes of death,
Baseline characteristics are presented as mean standard deviation or as frequency (%) for continuous and categorical variables respectively. The crude incidence rate of SCD for each study cohort was calculated using the total SCD cases occurring in the cohort as the numerator and the total
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
4
J.-L. Feng
number in each study cohort as the denominator. The incidence rate of SCD for each study cohort was calculated using the total SCD cases as the numerator and the sum of followup time for each patient in that cohort as the denominator. Kaplan–Meier survival curves were used to estimate the unadjusted risk of SCD in all four cohorts. Predictors of SCD were assessed for each study cohort by multivariable Cox proportional hazards regression models including age, sex, Indigenous status, medical conditions, and/or cardiac procedures (only for the MI cohort). The associations of the candidate predictors with SCD are presented as hazard ratios (HRs) (95% confidence interval (CI)) (Table 3). No evidence of violation of the proportional hazard assumption was found by including an interaction term with ln (1 + time) in the model for each potential predictor. Two-tailed values of p < 0.05 were considered statistically significant. All analyses were performed with SAS software version 9.4 (Cary, NC, USA).
Results The mean age in individuals with incident hospitalisation for HF was 7 years higher than 28-day survivors of incident MI, 5 years higher than incident hospitalisation for AF, and 3 years for hospitalisation for incident stroke. Over two-thirds of 28-day survivors with incident MI were men whereas the gender divide was less apparent for the other three cohorts. Individuals with incident AF had the lowest prevalence of all medical conditions examined compared to the other three cohorts (Table 1). Hypertension was the most common medical condition recorded but varied across the four cohorts, ranging from 37.4% in AF cohort to 66.0% in stroke cohort. Diabetes was the second most common medical condition recorded in the cohorts of incident MI, AF, and stroke whereas AF was the second most common in HF cohort. Approximately half of the individuals with incident MI underwent PCI therapy within 28 days of incident admission.
HF, AF, and stroke respectively. Taking into account how many years each individual contributed to the study, the incidence rate of SCD was higher in the cohort of HF, followed by MI, and then stroke and AF (Table 2). The median time to follow-up for SCD was 3.2 to 5.0 years across the cohorts. Figure 1 shows the unadjusted risk of SCD in the four cohorts. Significant difference was found in the risk of SCD incidence between the four cohorts (Log-Rank 523.8, p < 0.0001).
Predictors of SCD Table 3 presents multivariable-adjusted hazard ratios for the association of selected medical conditions with the hazard of SCD for the four cohorts after adjustment for age, sex, and Indigenous status. In the incident MI cohort, most heartrelated medical conditions including HF, hypertension, stroke, peripheral artery disease, diabetes, and chronic kidney disease were associated with 1.1 times to 2.1 times greater hazard of SCD. The hazard of SCD was 40% and 34% lower for individuals who underwent PCI (HR 0.60, 95% CI 0.48 to 0.75) and CABG (HR 0.66, 95% CI 0.48 to 0.90) within 28 days following incident hospitalisation for MI respectively. However, there was no significant association between the presence of implantable cardioverter-defibrillator and SCD (HR 1.19, 95% CI 0.43 to 3.28). In the incident HF cohort, arrhythmias and MI were associated with a two-fold greater hazard of SCD while peripheral artery disease was associated with a 1.4 times greater hazard of SCD. In the incident AF cohort, MI, HF, and arrhythmias were also strong risk markers of SCD. In the incident stroke cohort, MI, peripheral artery disease, and chronic kidney disease were associated with approximately a two-fold elevated hazard of SCD.
Discussion
Incidence of SCD Over a maximum follow-up of 11.5 years, the crude incidence rates of SCD were 2.3%, 4.2%, 1.5%, 1.4% in the cohorts MI,
This study showed that the incidence rate of SCD was highest following incident hospitalisation for HF, followed by MI, stroke and then AF. Among the four cohorts, MI and
Table 2 Incidence rates of sudden cardiac death (SCD) following incident hospitalised myocardial infarction, heart failure, atrial fibrillation, and stroke, in individuals aged 35–84 years in Western Australia. Study cohorts
Myocardial infarction (n = 19,145)
Heart failure (n = 8,294)
Atrial fibrillation (n = 12,015)
Stroke (n = 14,160)
Number of sudden cardiac death
439
350
183
202
Person-years
101,051
32,134
64,479
59,381
Crude incidence rate (%)
2.3 (439/19,145)
4.2 (350/8,294)
1.5 (183/12,014)
1.4 (202/14,160)
Incidence rate (per 1,000 person-years) Unadjusted risk at 1 year (%, 95% CI)
4.3 (439/10,1051) 0.7 (0.6–0.8)
10.9 (350/32,134) 1.6 (1.3–1.9)
2.8 (183/64,479) 0.4 (0.3–0.5)
3.4 (202/59,381) 0.4 (0.3–0.5)
Unadjusted risk at 5 year (%, 95% CI)
2.1 (1.9–2.4)
5.0 (4.4–5.6)
1.4 (1.2–1.6)
1.7 (1.4–2.0)
Abbreviation: CI, confidence interval.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
5
Incidence and Predictors of Sudden Cardiac Death
Figure 1 Kaplan–Meier cumulative mortality curves for sudden cardiac death following incident hospitalisation for myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF), and stroke (Str) in individuals aged 35–84 years in Western Australia.
peripheral artery disease are recognised as the two most common predictors of SCD, followed by arrhythmias and HF, and then chronic kidney disease and diabetes. In the incident MI cohort, there was a reduced hazard for SCD associated with receipt of coronary revascularisation procedures at the time of MI. These findings suggest that patients potentially benefit from optimised treatment of these medical conditions and preventing SCD. For patients following incident hospitalisation for HF, treating cardiac related conditions would be particularly important as they have the highest risk of succumbing to SCD. Several international studies have shown a range of incidence of SCD following MI, HF, or AF. The cohort of 28-day survivors of incident MI (4.3 per 1,000 person-years) is at the lower end of the range of incidence rate of SCD (4 to 18 per 1,000 person-years) compared to other studies on MI [6,7,20]. Several factors are likely to account for the difference. The cohort of incident MI includes 28-day survivors only, this differs from the above three studies where the full cohort or surviving at least 40 days of MI were included, probably accounting for the observed difference in incident rate. As advanced management of MI has improved the prognosis of these patients [21], different lengths of follow-up time (maximum 11.5 years versus 5 years) may explain the differing results between this
study and the study in the FuWei Hospital [7]. Also of note is that the cohort was the incident MI cohort who had no prior hospitalisation for MI in the preceding 15 years. This differs from two of studies mentioned above where MI patients and those with prior history of CHD were included [7,20]. Compared to the findings of Eisen and Marijon in a cohort of established AF (electrocardiographic evidence of AF within 12 months) or aged at least 65 years old respectively, a relatively lower risk of SCD was found in the incident AF cohort (1.2% versus 0.4% at 1 year; 3.8% versus 0.8% at 3 years respectively) [9,10]. Study cohorts with different types of HF (reduced left ventricular ejection fraction versus preserved) may lead to a difference in observed risk of SCD due to the inverse association between mortality and left ventricular ejection fraction [22,23]. A meta-analysis showed patients with reduced ejection fraction have a higher risk of mortality than those with preserved ejection fraction (75% of patients with reduced ejection fraction) [24]. As suggested by a previous study on the same WA population [25], approximately half of our HF cohort had CHD comprising patients with both reduced and preserved ejection fraction. All these may explain why the HF cohort had a slightly higher risk of SCD at both 1 and 5 years than the US study where only HF patients with preserved ejection fraction were included [8].
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
6
J.-L. Feng
Table 3 Multivariable-adjusted hazard ratios{ and 95% confidence intervals for sudden cardiac death following incident hospitalisation for myocardial infarction, heart failure, atrial fibrillation, and stroke in individuals aged 35–84 years in Western Australia. Myocardial infarction
Heart failure
Atrial fibrillation
Stroke
(n = 19,145)
(n = 8,294)
(n = 12,015)
(n = 14,160)
Age£
1.05 (1.04, 1.06)
1.02 (1.01, 1.04)
1.06 (1.04, 1.07)
1.05 (1.03, 1.06)
Sex (female as reference)
1.23 (1.00, 1.51)
1.22 (0.98. 1.52)
1.04 (0.77, 1.40)
1.34 (1.00, 1.79)
myocardial infarction heart failure
– 2.11 (1.69, 2.63)
1.82 (1.43, 2.32) –
1.94 (1.29, 2.92) 2.59 (1.86, 3.59)
1.86 (1.28, 2.71) 1.24 (0.83, 1.86)
diabetes
1.41 (1.15, 1.74)
1.01 (0.79, 1.28)
1.47 (1.02, 2.12)
1.06 (0.78, 1.45)
chronic kidney disease
1.66 (1.32, 2.09)
1.23 (0.96, 1.59)
1.37 (0.89, 2.11)
2.00 (1.41, 2.83)
hypertension
1.29 (1.02, 1.62)
0.87 (0.69, 1.10)
1.09 (0.79, 1.51)
1.29 (0.92, 1.81)
atrial fibrillation
1.13 (0.90, 1.43)
0.91 (0.73, 1.14)
–
1.37 (1.02, 1.88)
arrhythmias*
1.22 (0.84, 1.78)
2.78 (1.96, 3.92)
2.19 (1.17, 4.12)
1.49 (0.76, 2.90)
strokez
1.41 (1.03, 1.94)
0.90 (0.58, 1.40)
0.99 (0.46, 2.13)
–
peripheral artery disease chronic obstructive pulmonary disease
1.42 (1.11, 1.83) 1.10 (0.86, 1.42)
1.36 (1.03, 1.82) 1.25 (0.99, 1.59)
1.44 (0.90, 2.29) 1.27 (0.85, 1.88)
2.05 (1.45, 2.89) 1.42 (1.01, 2.04)
1.61 (1.30, 2.00)
–
–
–
§
Medical conditions
coronary heart disease
y
V
Procedures
{
percutaneous coronary intervention
0.60 (0.48, 0.75)
–
–
–
coronary artery bypass grafting
0.66 (0.48, 0.90)
–
–
–
implantable cardioverter-defibrillator
1.19 (0.43, 3.28)
–
–
–
adjusted for age, age2, sex, Indigenous status, and all medical conditions stated in this table.
£
hazard ratios at mean age.
§
includes concurrent conditions at incident admission and prior medical history unless otherwise indicated.
*
includes ventricular tachycardia, ventricular fibrillation, and cardiac arrest.
z
includes subarachnoid haemorrhage, intracerebral haemorrhage, occlusion and stenosis of precerebral arteries with cerebral infarction, and occlusion of cerebral arteries with cerebral infarction.
y
includes stable and unstable angina and chronic coronary heart disease, identified from prior medical history only.
V
identified within 28 days of the discharge date of the incident admission (inclusive) from any of the 11 procedure fields.
Myocardial infarction, peripheral artery disease, arrhythmias, and HF are identified as common predictors of SCD after the hospitalisation of four major cardiovascular events. Consistent with literature reporting strong evidence that acute ischaemia and infarction increase the risk of arrhythmic SCD, the present study observes both history of MI and arrhythmias as strongly predictive of SCD [26]. In addition, the findings of prior CHD (excluding MI) and HF which are associated with an increased risk of SCD are in line with prior studies [22,27]. This study also adds to the literature to report arrhythmias (including ventricular tachycardia, ventricular fibrillation, and cardiac arrest) is a strong risk indication in patients following hospitalisation for incident AF. This is supported by two possible mechanisms available to explain the pathways, namely where AF may reduce myocardial threshold for ventricular tachycardia induction and may elevate electrical instability [28,29]. One case study and one animal study have shown AF facilitates provocation of lethal arrhythmias although the pathophysiologic substrates are not completely clear among AF, fatal arrhythmias, and SCD [28,30]. Another finding of the present study is that
peripheral artery disease doubled the risk of SCD after stroke, indicating the presence of peripheral artery disease is a powerful prognosis marker of fatal events [31], particularly SCD. Polyvascular disease with inflammation and ischaemia of thrombotic origin may explain the relationship between peripheral artery disease and SCD [32]. This study extends the accumulating evidence showing chronic kidney disease as an independent marker for longterm cardiac mortality [33,34], including SCD, which is suggested by the mechanism of chronic kidney disease with a worsening kidney function increasing the risk of SCD, although the complete mechanism requires more investigation. The significant association between chronic obstructive pulmonary disease and SCD after stroke is observed, in agreement with studies from the Netherlands which have shown that chronic obstructive pulmonary disease increases the risk of SCD independent of cardiovascular risk factors [35,36]. The interplay between chronic obstructive pulmonary disease and SCD following stroke is of interest as chronic obstructive pulmonary disease is often encountered as a comorbidity of stroke.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
7
Incidence and Predictors of Sudden Cardiac Death
Consistent with the literature [6,7,37], coronary revascularisation is found to be associated with a lower risk of SCD, suggesting this procedure can interrupt the pathway to SCD in the post-MI population. The favourable risk of SCD driven by the coronary revascularisation may be overestimated due to selection bias where patients underwent the procedure. The reasons for non-revascularisation are multifactorial and may relate to the severity of MI, comorbidities, old age, and patient preferences [38], warranting further investigation of the indication for revascularisation. From a prevention perspective, this study provides a unique context of comparing the common medical conditions as predictors of SCD in patients surviving after the four major cardiovascular events. The risk of SCD is found to be varying widely, higher in the HF cohort and lower in the AF cohort. The results underscore the common medical conditions such as MI, HF, arrhythmias, peripheral artery disease, chronic kidney disease, and diabetes are predictive of SCD although some of them were not always predicting SCD through all the cohorts. The findings suggest two potential challenges of offering optimal care for patients. First, current management models for primary-disease-oriented care may not effectively address the health care needs for the four major CVD events for which, currently, clinical guidelines mainly focus treatment [39–42]. Second, polyvascular disease may mediate the relationship between chronic kidney disease, chronic obstructive pulmonary disease and SCD and, therefore, increasing the challenge of successful treatment as this adds to the overall complexity of care [43]. A coordinated comprehensive system of care delivery may be required to address the potentially multiple comorbid conditions for a patient regarding SCD prevention.
Strengths and Limitations The strengths of this study are the utilisation of high-quality administrative health data to capture the study cohort of all incident hospitalised CVD events occurring in a population setting, and the availability of providing long-term follow-up data for this cohort. This study design of using incident cases of major CVD events indicated the natural history of SCD and excluded the potential over-representation derived from prevalent patients with a long course of disease. Application of 15-year look-back period enabled determination of prior hospitalisations of MI, HF, AF and stroke, maximising the ability to identify the incident cohorts. A fixed 15-year lookback period for prior medical conditions of hospitalisation also minimises the under-ascertainment of the status of the medical conditions as potential predictors reported in this study. A number of possible limitations of this study must also be considered. Similar to other studies, identification of SCD is difficult and susceptible to error. Multiple sources were applied for SCD cases ascertainment, although some cases may have been missed and no validation of the cases has taken place. Other factors such as smoking, alcohol use, body mass index, and left ventricular function were not either well recorded or available in the administrative health
data. Thus, whether these factors confound the observed association remains unknown. The results from the cohort of MI may not be generalisable to the whole incident MI population as patients who died within 28 days following the admission date of incident hospitalisation for MI were not included.
Conclusions This population-based study using administrative health data reported the incidence of SCD after an incident hospitalisation for MI, HF, AF or stroke event is appreciable, calling for prevention efforts. Medical conditions such as MI, HF, peripheral artery disease, and arrhythmias are predictors of SCD, broadening our understanding of SCD and requiring optimal care for the survivors of the four major cardiovascular events. Early revascularisation is protective in the MI cohort, emphasising timely intervention.
Authorship Statement The author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and my discussed interpretation.
Acknowledgements The author wishes to thank the staff at the Western Australian Data Linkage Brach and the Department of Health Inpatient Data Collection and Registry of Births, Deaths and Marriages for the provision of data, the Victorian Department of Justice and Regulation, and the National Coronial Information System. This work was supported by a project grant from the National Health and Medical Research Council of Australia, Perth, Western Australia, Australia (No.572558). JLF was supported by an International Postgraduate Research Scholarship (IPRS) and Australian Postgraduate Award when she was doing this work. The funding bodies played no part in the study design, analysis or interpretation of results.
Conflicts of Interest None declared.
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.hlc. 2019.03.020.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
8
J.-L. Feng
References [1] Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, et al. Heart disease and stroke statistics—2015 update: a report from the American Heart Association. Circulation 2015;131:e29–322. [2] Australian Institute of Health and Welfare. Cardiovascular disease: Australian facts 2011. Canberra: AIHW;[cited 23 Mar 2018]. Cardiovascular disease series. Cat. no. CVD 53. Available from: 2011, http://www.aihw. gov.au/WorkArea/DownloadAsset.aspx?id=10737418530. [3] Ball J, Thompson DR, Ski CF, Carrington MJ, Gerber T, Stewart S. Estimating the current and future prevalence of atrial fibrillation in the Australian adult population. Med J Aust 2015;202:32–5. [4] Key statistics-Heart Research Australia [cited 23 Mar 2018]. Available from: https://www.heartfoundation.org.au/about-us/what-we-do/ heart-disease-in-australia. [5] Wong CX, Brown A, Lau DH, Chugh SS, Albert CM, Kalman JM, et al. Epidemiology of Sudden Cardiac Death: Global and Regional Perspectives. Heart Lung and Circulation 2019;28:6–14. [6] Adabag AS, Therneau TM, Gersh BJ, Weston SA, Roger VL. Sudden death after myocardial infarction. JAMA 2008;300:2022–9. [7] Fan X, Hua W, Xu Y, Ding L, Niu H, Chen K, et al. Incidence and predictors of sudden cardiac death in patients with reduced left ventricular ejection fraction after myocardial infarction in an era of revascularisation. Heart 2014;100:1242–9. [8] Al-Khatib SM, Shaw LK, O’Connor C, Kong M, Califf RM. Incidence and predictors of sudden cardiac death in patients with diastolic heart failure. J Cardiovasc Electrophysiol 2007;18:1231–5. [9] Eisen A, Ruff CT, Braunwald E, Nordio F, Corbalan R, Dalby A, et al. Sudden Cardiac Death in Patients With Atrial Fibrillation: Insights From the ENGAGE AF-TIMI 48 Trial. J Am Heart Assoc 2016;5. [10] Marijon E, Le Heuzey JY, Connolly S, Yang S, Pogue J, et al. Causes of death and influencing factors in patients with atrial fibrillation: a competing-risk analysis from the randomized evaluation of long-term anticoagulant therapy study. Circulation 2013;128:2192–201. [11] Hagens VE, Rienstra M, Van Veldhuisen DJ, Crijns HJ, et al. Determinants of sudden cardiac death in patients with persistent atrial fibrillation in the rate control versus electrical cardioversion (RACE) study. Am J Cardiol 2006;98:929–32. [12] Soros P, Hachinski V. Cardiovascular and neurological causes of sudden death after ischaemic stroke. Lancet Neurol 2012;11:179–88. [13] Waien SA. Linking large administrative databases: a method for conducting emergency medical services cohort studies using existing data. Acad Emerg Med 1997;4:1087–95. [14] Sanfilippo FM, Hobbs MS, Knuiman MW, Ridout SC, Bradshaw PJ, Finn JC, et al. Can we monitor heart attack in the troponin era? Evidence from a population-based cohort study. BMC Cardiovasc Disord 2011;11:35. [15] Teng TH, Finn J, Hung J, Geelhoed E, Hobbs M. A validation study: how effective is the Hospital Morbidity Data as a surveillance tool for heart failure in Western Australia? Aust N Z J Public Health 2008;32:405–7. [16] Briffa T, Hung J, Knuiman M, McQuillan B, Chew DP, Eikelboom J, et al. Trends in incidence and prevalence of hospitalization for atrial fibrillation and associated mortality in Western Australia, 1995-2010. Int J Cardiol 2016;208:19–25. [17] Jamrozik K, Dobson A, Hobbs M, McElduff P, Ring I, D’Este K, et al. Monitoring the Incidence of Cardiovascular Disease in Australia. Canberra: AIHW;[cited 23 Mar 2018]. Cardiovascular Disease series No.17. Cat No. CVD 16. Available from: https://www.aihw.gov.au/getmedia/ 7cfa7e5c-c37a-42f4-8e6d-e9bff0796029/micda.pdf.aspx?inline=true 2001. [18] Virmani R, Burke AP, Farb A. Sudden cardiac death. Cardiovasc Pathol 2001;10:211–8. [19] Feng JL, Hickling S, Nedkoff L, Knuiman M, Semsarian C, Ingles J, et al. Sudden cardiac death rates in an Australian population: a data linkage study. Aust Health Rev 2015;39:561–7. [20] Solomon SD, Zelenkofske S, McMurray JJ, Finn PV, Velazquez E, Ertl G, et al. Sudden death in patients with myocardial infarction and left ventricular dysfunction, heart failure, or both. N Engl J Med 2005;352:2581–8. [21] Hong MK. Recent Advances in the Treatment of ST-Segment Elevation Myocardial Infarction. Scientifica (Cairo) 2012;2012. 683683.
[22] Zipes DP, Wellens HJ. Sudden cardiac death. Circulation 1998;98:2334– 51. [23] Gorgels AP, Gijsbers C, de Vreede-Swagemakers J, Lousberg A, Wellens HJ. Out-of-hospital cardiac arrest–the relevance of heart failure. The Maastricht Circulatory Arrest Registry. Eur Heart J 2003;24:1204–9. [24] Meta-analysis Global Group in Chronic Heart Failure (MAGGIC). The survival of patients with heart failure with preserved or reduced left ventricular ejection fraction: an individual patient data meta-analysis. Eur Heart J 2012;33:1750–7. [25] Teng TH, Hung J, Finn J. The effect of evidence-based medication use on long-term survival in patients hospitalised for heart failure in Western Australia. Med J Aust 2010;192:306–10. [26] Hayashi M, Shimizu W, Albert CM. The spectrum of epidemiology underlying sudden cardiac death. Circ Res 2015;116:1887–906. [27] Turakhia M, Tseng ZH. Sudden cardiac death: epidemiology, mechanisms, and therapy. Curr Probl Cardiol 2007;32:501–46. [28] Somberg JC, Torres V, Keren G, Butler B, Tepper D, Kleinbaum H, et al. Enhancement of myocardial vulnerability by atrial fibrillation. Am J Ther 2004;11:33–43. [29] Denker S, Lehmann M, Mahmud R, Gilbert C, Akhtar M. Facilitation of ventricular tachycardia induction with abrupt changes in ventricular cycle length. Am J Cardiol 1984;53:508–15. [30] Roy D, Brugada P, Wellens HJ. Atrial tachycardia facilitating initiation of ventricular tachycardia. Pacing Clin Electrophysiol 1983;6:47–52. [31] Brevetti G, Schiano V, Verdoliva S, Silvestro A, Sirico G, et al. Peripheral arterial disease and cardiovascular risk in Italy. Results of the Peripheral Arteriopathy and Cardiovascular Events (PACE) study. J Cardiovasc Med (Hagerstown) 2006;7:608–13. [32] Brevetti G, Giugliano G, Brevetti L, Hiatt WR. Inflammation in peripheral artery disease. Circulation 2010;122:1862–75. [33] Henry RM, Kostense PJ, Bos G, Dekker JM, Nijpels G, Heine RJ, et al. Mild renal insufficiency is associated with increased cardiovascular mortality: The Hoorn Study. Kidney Int 2002;62:1402–7. [34] Tsagalis G, Akrivos T, Alevizaki M, Manios E, Stamatellopoulos K, Laggouranis A, et al. Renal dysfunction in acute stroke: an independent predictor of long-term all combined vascular events and overall mortality. Nephrol Dial Transplant 2009;24:194–200. [35] Lahousse L, Niemeijer MN, van den Berg ME, Rijnbeek PR, Joos GF, Hofman A, et al. Chronic obstructive pulmonary disease and sudden cardiac death: the Rotterdam study. Eur Heart J 2015;36:1754–61. [36] van den Berg ME, Stricker BH, Brusselle GG, Lahousse L. Chronic obstructive pulmonary disease and sudden cardiac death: A systematic review. Trends Cardiovasc Med 2016;26:606–13. [37] Makikallio TH, Barthel P, Schneider R, Bauer A, Tapanainen JM, Tulppo MP, et al. Frequency of sudden cardiac death among acute myocardial infarction survivors with optimized medical and revascularization therapy. Am J Cardiol 2006;97:480–4. [38] Holmes Jr DR, Rihal CS. Revascularization options: one size does not fit all. Cmaj 2012;184:E104–105. [39] Ghimire G, Gupta A, Hage FG. Guidelines in review: 2013 ACCF/AHA Guideline for the Management of ST-Elevation Myocardial Infarction. JNucl Cardiol 2014;21:190–1. [40] Yancy CW, Jessup M, Bozkurt B, Butler J, Casey Jr DE, Drazner MH, et al. ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2013;2013 (62):e147–239. [41] January CT, Wann LS, Alpert JS, Calkins H, Cigarroa JE, Cleveland JC, et al. AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the Heart Rhythm Society. Circulation 2014;2014(130):2071–104. [42] Kernan WN, Ovbiagele B, Black HR, Bravata DM, Chimowitz MI, Ezekowitz MD, et al. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014;45:2160–236. [43] Sales AE, Tipton EF, Levine DA, Houston TK, Kim Y, Allison J, et al. Are co-morbidities associated with guideline adherence? The MI-Plus study of Medicare patients. J Gen Intern Med 2009;24:1205–10.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
Heart, Lung and Circulation (2019) xx, 1–8 1443-9506/04/$36.00 https://doi.org/10.1016/j.hlc.2019.03.020
ORIGINAL ARTICLE
Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event Jia-Li Feng, BMed, MMed, PhD School of Population and Global Health, Faculty of Medicine, Dentistry and Health Sciences, The University of Western Australia, Perth, WA, Australia Received 4 January 2019; received in revised form 2 February 2019; accepted 27 March 2019; online published-ahead-of-print xxx
Background
Sudden cardiac death (SCD) still accounts for the majority of deaths from the four major cardiovascular events (myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF), and stroke) despite substantial progress on prevention.
Methods
Four separate cohorts (one for each of the four major cardiovascular conditions) were captured through person-linked hospital morbidity and mortality data collections between 2000 and 2009 and followed-up for 11.5 years. The incidence rate for each cohort was total SCD cases divided by sum of follow-up time for each individual alive. Kaplan–Meier survival curve was used to calculate unadjusted risk of SCD. Predictors of SCD were identified by fitting multivariable adjusted Cox regression models in each of the cohorts.
Results
There were 1,174 cases of SCD from 53,614 total CVD events across the cohorts (35.6% for MI, 15.6% for HF, 22.4% for AF, 26.4% for stroke). The incidence rate and unadjusted risk of SCD were both highest after incident hospitalisation for HF, followed by MI, stroke and AF. The elevated risk of SCD was independently associated with MI, HF, arrhythmias, peripheral artery disease, diabetes, chronic kidney disease, and prior coronary heart disease (hazard ratios ranging from 1.1 to 2.8). Early revascularisation is protective in 28-day survivors after an incident MI event.
Conclusions
An appreciable incidence of SCD following an incident event of MI, HF, AF and stroke deserves greater prevention efforts. Major medical conditions such as MI, HF, peripheral artery disease, and arrhythmias are risk markers of SCD and coronary revascularisation is protective.
Keywords
Incidence Predictor Sudden cardiac death Major cardiovascular event
Introduction Four major cardiovascular events, myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF), and stroke annually affect around 1.9% of the total general population in Australia and 2.6% in the United States (US) [1–4]. Sudden cardiac death (SCD) still accounts for the majority of deaths from the four major cardiovascular events in spite of substantial progress on prevention [5]. Few observational
studies are available to report the incidence rates of SCD after AF or stroke and limited data show long-term incidence rates of SCD after MI or HF. The observational communitybased Olmsted County study showed a 30-day incidence of SCD (1.2%) and a 5-year unadjusted risk of 6.9% in 3,000 residents after MI from 1979 to 2005 [6]. A single-centre study in China reported the 5-year SCD unadjusted risk of 5.0% in 40 days survivors of MI with an ejection fraction 35% [7]. Another single-centre study in the US of HF patients with
Email: [email protected]
© 2019 Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia and New Zealand (CSANZ). Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
2
J.-L. Feng
preserved ejection fraction who underwent a cardiac procedure reported the unadjusted risk of SCD were 1.0% at 1 year and 3.0% at 5 years [8]. The impact of long-term conditions on the prognosis of these four major cardiovascular events is poorly understood. Thus identifying predictors of SCD as an indication of the prognosis becomes a critical first step for health care planning and future reduction of SCD risk. A Charlson Comorbidity Index (CCI) score of three or more and ejection fraction 25% were risk markers of SCD while coronary revascularisation reduced the risk of SCD in post-MI patients [6]. The single-centre study from the US mentioned above identified predictors of SCD such as diabetes and MI following an admission of HF with preserved ejection fraction [8]. However, few large-scale observational data are available investigating predictors of SCD in patients with AF, although randomised controlled trials suggest diabetes, hypertension, coronary artery disease were predictive of SCD in a population of patients with AF [9–11]. No study has specifically investigated the predictors of SCD following stroke [12]. Although the four major cardiovascular diseases (CVDs) appear to share some common cardiovascular risk factors such as diabetes and hypertension, there is a lack of investigation of such common medical conditions as predictors of SCD following the four major cardiovascular events altogether. The result of this investigation ideally would, first, provide a whole picture of the incidence burden of SCD in the survivors of the four major cardiovascular events. Second, it would broaden our understanding of SCD predictors and facilitate optimisation of treatment after the four major cardiovascular events for a further reduction in SCD incidence. Western Australia’s (WA) data linkage provides a unique source of examining the incidence and predictors of SCD in a whole-population setting following hospitalisation for incident MI, HF, AF or stroke patients over a longterm follow-up.
Methods Data Source and Study Cohort Person-linked mortality and morbidity data for this study were obtained from the Mortality Register and Hospital Morbidity Data Collection and linked by WA Data Linkage System using probabilistic matching [13]. The linked dataset included death records for CVD and all public and private hospitalisations for CVD in WA from 1 January 1985 to 30 June 2011. Variables examined in the linked dataset included date of death, underlying and associated causes of death, indication of post-mortem, demographic fields (including age, sex, and Indigenous status), hospital admission and discharge dates, hospital discharge diagnosis fields (principal and 20 additional), and 11 fields of in-hospital procedures. Ethical approval for this study was obtained from the Human Research Ethics Committees of the WA Department of Health (#RA/4/1/1491) and The University of Western Australia (#2014/55).
The study included four cohorts of patients with a principal discharge diagnosis of incident MI, HF, AF or stroke from 1 January 2000 to 31 December 2009. The four cohorts were identified separately from each other, and were not mutually exclusive. The ICD codes used to identify the cohorts are listed in the Supplementary Table 1. The cohort of incident MI included patients with MI recorded in the principal discharge diagnosis field and no hospitalisation history of MI recorded in any discharge diagnosis fields in the preceding 15 years. Following the same practice, incident HF, AF, and stroke patients were identified separately. Validation studies on the coding of MI, HF, AF, and stroke in the Hospital Morbidity Data showed 92.2% of positive predictive values for MI, 99.5% for HF, 84.6% for AF, 85.0% for stroke [14–17]. Lower sensitivity for MI was noted from the Hospital Morbidity Data in the very old patients [14], thus patients 85years and -older were excluded from all four cohorts. In addition, patients aged <35 years were excluded as they are likely to have a lower rate of SCD and different predictors compared to those 35 years old [18]. Of note is that patients surviving <28 days following the admission date of incident MI were excluded from this study (<5% of the MI cohort) because a fatal first MI is less predictable thus providing few prevention efforts [19]. There were a total of 53,614 incident events of MI, HF, AF and stroke, representing 50,877 individuals, with 5.1% of individuals appearing in more than one incident cohort. As SCD incidence rate and predictors were identified separately in each of the four cohorts, individuals appearing in more than one cohort were retained in analyses.
Potential Predictors for SCD Potential predictors for SCD included common medical conditions, cardiac procedures, sex, age, and Indigenous status. The medical conditions included: hypertension, MI, HF, AF, arrhythmias (ventricular tachycardia, ventricular fibrillation, and cardiac arrest), diabetes, chronic kidney disease, peripheral arterial disease, stroke, chronic obstructive pulmonary disease, and coronary heart disease (CHD). Each of the four cohorts was not considered as a potential predictor in its respective cohort. For example, MI was not considered as a potential predictor in the cohort of incident MI cases. Potential predictors of SCD were identified from the 20 additional discharge diagnosis fields during the incident admission and any discharge diagnosis fields in a fixed 15-year lookback period from the discharge date of the incident admission. CHD was only considered in the incident MI cohort because in the other three cohorts, MI as the potential predictor was examined and this was identified from the 15-year prior hospitalisation history only. Coronary procedures were investigated to determine whether they are predictors of SCD in the incident MI cohort only, as coronary revascularisation is most commonly used in patients with CHD. Coronary procedures included coronary revascularisation (percutaneous coronary intervention (PCI) and coronary artery bypass grafting surgery (CABG)). Implantable cardioverter-defibrillator is an important treatment approach for SCD prevention and was, therefore, considered
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
3
Incidence and Predictors of Sudden Cardiac Death
Table 1 Characteristics of individuals with incident hospitalisation for myocardial infarction, heart failure, atrial fibrillation, and stroke, aged 35–84 years in Western Australia.
Mean Age, years (SD)
Myocardial infarction
Heart failure
Atrial fibrillation
Stroke
(n = 19,145)
(n = 8,294)
(n = 12,015)
(n = 14,160)
64.4 (12.3)
71.7 (10.9)
66.5 (11.8)
69.2 (12.0)
Age groups 35–54 years
25.2
9.7
17.9
15.1
55–69 years
37.6
25.8
37.7
28.8
70–84 years
37.2
64.5
44.4
56.1
70.8 4.7
57.4 5.7
59.1 1.7
55.6 4.1
Male Indigenous status Medical conditions§ myocardial infarction
–
17.9
6.2
9.0
heart failure
16.6
–
13.1
11.4
diabetes mellitus
24.9
34.6
13.5
25.8
chronic kidney disease
12.6
25.2
7.5
12.7
hypertension atrial fibrillation
55.4 14.9
62.8 38.1
37.4 –
66.0 22.7
arrhythmias*
5.5
4.7
2.0
2.2
strokez
4.1
7.6
2.4
–
peripheral artery disease
9.1
15.1
5.1
10.5
chronic obstructive pulmonary disease
10.9
24.8
9.8
13.2
21.8
–
–
–
5.2
7.8
4.7
3.9
0.2
0.8
0.3
0.2
coronary heart disease (excluding MI)
y
Prior procedures coronary revascularisation£ implantable cardioverter-defibrillator m
Procedures within 28 days of incident admission percutaneous coronary intervention
45.4
1.3
0.5
0.1
coronary artery bypass grafting
6.5
0.8
0.1
0.01
implantable cardioverter-defibrillator
0.5
1.5
0.2
0.03
Data are percentages (%) unless otherwise indicated. Abbreviations: SD, standard deviation; MI, myocardial infarction. § *
includes conditions at incident admission and prior medical history unless otherwise indicated; cases could have multiple medical conditions.
includes ventricular tachycardia, ventricular fibrillation, and cardiac arrest.
z
includes subarachnoid haemorrhage, intracerebral haemorrhage, occlusion and stenosis of precerebral arteries with cerebral infarction, and occlusion of cerebral arteries with cerebral infarction.
y
includes stable and unstable angina pectoris and chronic coronary heart disease, identified from prior medical history only.
£
includes percutaneous coronary intervention and coronary artery bypass grafting surgery.
m
identified within 28 days of the incident admission (inclusive).
as a potential protective factor. The procedures were included if they occurred in any of the 11 procedure fields within 28 days (inclusive) and prior procedures identified using a fixed 15year lookback period from the incident MI admission. Demographic factors were identified from the incident admission (for age and sex) and all admissions (for Indigenous status).
and results of post-mortem if carried-out. Follow-up was censored at the date of death or at 30 June 2011 (the end of the study period), whichever came first. Twenty-two (22) patients who died during follow-up and had a missing cause of death were excluded from all analyses (MI cohort (n = 7), HF (n = 6), AF (n = 2), stroke (n = 7)).
Follow-Up
Statistical Analysis
The primary endpoint was SCD. Four criteria used to identify SCD, were detailed in our previous paper [19], based on following elements using administrative data: place of death, proximity to hospital admission, prior and/or concurrent medical history, underlying and associated causes of death,
Baseline characteristics are presented as mean standard deviation or as frequency (%) for continuous and categorical variables respectively. The crude incidence rate of SCD for each study cohort was calculated using the total SCD cases occurring in the cohort as the numerator and the total
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
4
J.-L. Feng
number in each study cohort as the denominator. The incidence rate of SCD for each study cohort was calculated using the total SCD cases as the numerator and the sum of followup time for each patient in that cohort as the denominator. Kaplan–Meier survival curves were used to estimate the unadjusted risk of SCD in all four cohorts. Predictors of SCD were assessed for each study cohort by multivariable Cox proportional hazards regression models including age, sex, Indigenous status, medical conditions, and/or cardiac procedures (only for the MI cohort). The associations of the candidate predictors with SCD are presented as hazard ratios (HRs) (95% confidence interval (CI)) (Table 3). No evidence of violation of the proportional hazard assumption was found by including an interaction term with ln (1 + time) in the model for each potential predictor. Two-tailed values of p < 0.05 were considered statistically significant. All analyses were performed with SAS software version 9.4 (Cary, NC, USA).
Results The mean age in individuals with incident hospitalisation for HF was 7 years higher than 28-day survivors of incident MI, 5 years higher than incident hospitalisation for AF, and 3 years for hospitalisation for incident stroke. Over two-thirds of 28-day survivors with incident MI were men whereas the gender divide was less apparent for the other three cohorts. Individuals with incident AF had the lowest prevalence of all medical conditions examined compared to the other three cohorts (Table 1). Hypertension was the most common medical condition recorded but varied across the four cohorts, ranging from 37.4% in AF cohort to 66.0% in stroke cohort. Diabetes was the second most common medical condition recorded in the cohorts of incident MI, AF, and stroke whereas AF was the second most common in HF cohort. Approximately half of the individuals with incident MI underwent PCI therapy within 28 days of incident admission.
HF, AF, and stroke respectively. Taking into account how many years each individual contributed to the study, the incidence rate of SCD was higher in the cohort of HF, followed by MI, and then stroke and AF (Table 2). The median time to follow-up for SCD was 3.2 to 5.0 years across the cohorts. Figure 1 shows the unadjusted risk of SCD in the four cohorts. Significant difference was found in the risk of SCD incidence between the four cohorts (Log-Rank 523.8, p < 0.0001).
Predictors of SCD Table 3 presents multivariable-adjusted hazard ratios for the association of selected medical conditions with the hazard of SCD for the four cohorts after adjustment for age, sex, and Indigenous status. In the incident MI cohort, most heartrelated medical conditions including HF, hypertension, stroke, peripheral artery disease, diabetes, and chronic kidney disease were associated with 1.1 times to 2.1 times greater hazard of SCD. The hazard of SCD was 40% and 34% lower for individuals who underwent PCI (HR 0.60, 95% CI 0.48 to 0.75) and CABG (HR 0.66, 95% CI 0.48 to 0.90) within 28 days following incident hospitalisation for MI respectively. However, there was no significant association between the presence of implantable cardioverter-defibrillator and SCD (HR 1.19, 95% CI 0.43 to 3.28). In the incident HF cohort, arrhythmias and MI were associated with a two-fold greater hazard of SCD while peripheral artery disease was associated with a 1.4 times greater hazard of SCD. In the incident AF cohort, MI, HF, and arrhythmias were also strong risk markers of SCD. In the incident stroke cohort, MI, peripheral artery disease, and chronic kidney disease were associated with approximately a two-fold elevated hazard of SCD.
Discussion
Incidence of SCD Over a maximum follow-up of 11.5 years, the crude incidence rates of SCD were 2.3%, 4.2%, 1.5%, 1.4% in the cohorts MI,
This study showed that the incidence rate of SCD was highest following incident hospitalisation for HF, followed by MI, stroke and then AF. Among the four cohorts, MI and
Table 2 Incidence rates of sudden cardiac death (SCD) following incident hospitalised myocardial infarction, heart failure, atrial fibrillation, and stroke, in individuals aged 35–84 years in Western Australia. Study cohorts
Myocardial infarction (n = 19,145)
Heart failure (n = 8,294)
Atrial fibrillation (n = 12,015)
Stroke (n = 14,160)
Number of sudden cardiac death
439
350
183
202
Person-years
101,051
32,134
64,479
59,381
Crude incidence rate (%)
2.3 (439/19,145)
4.2 (350/8,294)
1.5 (183/12,014)
1.4 (202/14,160)
Incidence rate (per 1,000 person-years) Unadjusted risk at 1 year (%, 95% CI)
4.3 (439/10,1051) 0.7 (0.6–0.8)
10.9 (350/32,134) 1.6 (1.3–1.9)
2.8 (183/64,479) 0.4 (0.3–0.5)
3.4 (202/59,381) 0.4 (0.3–0.5)
Unadjusted risk at 5 year (%, 95% CI)
2.1 (1.9–2.4)
5.0 (4.4–5.6)
1.4 (1.2–1.6)
1.7 (1.4–2.0)
Abbreviation: CI, confidence interval.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
5
Incidence and Predictors of Sudden Cardiac Death
Figure 1 Kaplan–Meier cumulative mortality curves for sudden cardiac death following incident hospitalisation for myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF), and stroke (Str) in individuals aged 35–84 years in Western Australia.
peripheral artery disease are recognised as the two most common predictors of SCD, followed by arrhythmias and HF, and then chronic kidney disease and diabetes. In the incident MI cohort, there was a reduced hazard for SCD associated with receipt of coronary revascularisation procedures at the time of MI. These findings suggest that patients potentially benefit from optimised treatment of these medical conditions and preventing SCD. For patients following incident hospitalisation for HF, treating cardiac related conditions would be particularly important as they have the highest risk of succumbing to SCD. Several international studies have shown a range of incidence of SCD following MI, HF, or AF. The cohort of 28-day survivors of incident MI (4.3 per 1,000 person-years) is at the lower end of the range of incidence rate of SCD (4 to 18 per 1,000 person-years) compared to other studies on MI [6,7,20]. Several factors are likely to account for the difference. The cohort of incident MI includes 28-day survivors only, this differs from the above three studies where the full cohort or surviving at least 40 days of MI were included, probably accounting for the observed difference in incident rate. As advanced management of MI has improved the prognosis of these patients [21], different lengths of follow-up time (maximum 11.5 years versus 5 years) may explain the differing results between this
study and the study in the FuWei Hospital [7]. Also of note is that the cohort was the incident MI cohort who had no prior hospitalisation for MI in the preceding 15 years. This differs from two of studies mentioned above where MI patients and those with prior history of CHD were included [7,20]. Compared to the findings of Eisen and Marijon in a cohort of established AF (electrocardiographic evidence of AF within 12 months) or aged at least 65 years old respectively, a relatively lower risk of SCD was found in the incident AF cohort (1.2% versus 0.4% at 1 year; 3.8% versus 0.8% at 3 years respectively) [9,10]. Study cohorts with different types of HF (reduced left ventricular ejection fraction versus preserved) may lead to a difference in observed risk of SCD due to the inverse association between mortality and left ventricular ejection fraction [22,23]. A meta-analysis showed patients with reduced ejection fraction have a higher risk of mortality than those with preserved ejection fraction (75% of patients with reduced ejection fraction) [24]. As suggested by a previous study on the same WA population [25], approximately half of our HF cohort had CHD comprising patients with both reduced and preserved ejection fraction. All these may explain why the HF cohort had a slightly higher risk of SCD at both 1 and 5 years than the US study where only HF patients with preserved ejection fraction were included [8].
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
6
J.-L. Feng
Table 3 Multivariable-adjusted hazard ratios{ and 95% confidence intervals for sudden cardiac death following incident hospitalisation for myocardial infarction, heart failure, atrial fibrillation, and stroke in individuals aged 35–84 years in Western Australia. Myocardial infarction
Heart failure
Atrial fibrillation
Stroke
(n = 19,145)
(n = 8,294)
(n = 12,015)
(n = 14,160)
Age£
1.05 (1.04, 1.06)
1.02 (1.01, 1.04)
1.06 (1.04, 1.07)
1.05 (1.03, 1.06)
Sex (female as reference)
1.23 (1.00, 1.51)
1.22 (0.98. 1.52)
1.04 (0.77, 1.40)
1.34 (1.00, 1.79)
myocardial infarction heart failure
– 2.11 (1.69, 2.63)
1.82 (1.43, 2.32) –
1.94 (1.29, 2.92) 2.59 (1.86, 3.59)
1.86 (1.28, 2.71) 1.24 (0.83, 1.86)
diabetes
1.41 (1.15, 1.74)
1.01 (0.79, 1.28)
1.47 (1.02, 2.12)
1.06 (0.78, 1.45)
chronic kidney disease
1.66 (1.32, 2.09)
1.23 (0.96, 1.59)
1.37 (0.89, 2.11)
2.00 (1.41, 2.83)
hypertension
1.29 (1.02, 1.62)
0.87 (0.69, 1.10)
1.09 (0.79, 1.51)
1.29 (0.92, 1.81)
atrial fibrillation
1.13 (0.90, 1.43)
0.91 (0.73, 1.14)
–
1.37 (1.02, 1.88)
arrhythmias*
1.22 (0.84, 1.78)
2.78 (1.96, 3.92)
2.19 (1.17, 4.12)
1.49 (0.76, 2.90)
strokez
1.41 (1.03, 1.94)
0.90 (0.58, 1.40)
0.99 (0.46, 2.13)
–
peripheral artery disease chronic obstructive pulmonary disease
1.42 (1.11, 1.83) 1.10 (0.86, 1.42)
1.36 (1.03, 1.82) 1.25 (0.99, 1.59)
1.44 (0.90, 2.29) 1.27 (0.85, 1.88)
2.05 (1.45, 2.89) 1.42 (1.01, 2.04)
1.61 (1.30, 2.00)
–
–
–
§
Medical conditions
coronary heart disease
y
V
Procedures
{
percutaneous coronary intervention
0.60 (0.48, 0.75)
–
–
–
coronary artery bypass grafting
0.66 (0.48, 0.90)
–
–
–
implantable cardioverter-defibrillator
1.19 (0.43, 3.28)
–
–
–
adjusted for age, age2, sex, Indigenous status, and all medical conditions stated in this table.
£
hazard ratios at mean age.
§
includes concurrent conditions at incident admission and prior medical history unless otherwise indicated.
*
includes ventricular tachycardia, ventricular fibrillation, and cardiac arrest.
z
includes subarachnoid haemorrhage, intracerebral haemorrhage, occlusion and stenosis of precerebral arteries with cerebral infarction, and occlusion of cerebral arteries with cerebral infarction.
y
includes stable and unstable angina and chronic coronary heart disease, identified from prior medical history only.
V
identified within 28 days of the discharge date of the incident admission (inclusive) from any of the 11 procedure fields.
Myocardial infarction, peripheral artery disease, arrhythmias, and HF are identified as common predictors of SCD after the hospitalisation of four major cardiovascular events. Consistent with literature reporting strong evidence that acute ischaemia and infarction increase the risk of arrhythmic SCD, the present study observes both history of MI and arrhythmias as strongly predictive of SCD [26]. In addition, the findings of prior CHD (excluding MI) and HF which are associated with an increased risk of SCD are in line with prior studies [22,27]. This study also adds to the literature to report arrhythmias (including ventricular tachycardia, ventricular fibrillation, and cardiac arrest) is a strong risk indication in patients following hospitalisation for incident AF. This is supported by two possible mechanisms available to explain the pathways, namely where AF may reduce myocardial threshold for ventricular tachycardia induction and may elevate electrical instability [28,29]. One case study and one animal study have shown AF facilitates provocation of lethal arrhythmias although the pathophysiologic substrates are not completely clear among AF, fatal arrhythmias, and SCD [28,30]. Another finding of the present study is that
peripheral artery disease doubled the risk of SCD after stroke, indicating the presence of peripheral artery disease is a powerful prognosis marker of fatal events [31], particularly SCD. Polyvascular disease with inflammation and ischaemia of thrombotic origin may explain the relationship between peripheral artery disease and SCD [32]. This study extends the accumulating evidence showing chronic kidney disease as an independent marker for longterm cardiac mortality [33,34], including SCD, which is suggested by the mechanism of chronic kidney disease with a worsening kidney function increasing the risk of SCD, although the complete mechanism requires more investigation. The significant association between chronic obstructive pulmonary disease and SCD after stroke is observed, in agreement with studies from the Netherlands which have shown that chronic obstructive pulmonary disease increases the risk of SCD independent of cardiovascular risk factors [35,36]. The interplay between chronic obstructive pulmonary disease and SCD following stroke is of interest as chronic obstructive pulmonary disease is often encountered as a comorbidity of stroke.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
7
Incidence and Predictors of Sudden Cardiac Death
Consistent with the literature [6,7,37], coronary revascularisation is found to be associated with a lower risk of SCD, suggesting this procedure can interrupt the pathway to SCD in the post-MI population. The favourable risk of SCD driven by the coronary revascularisation may be overestimated due to selection bias where patients underwent the procedure. The reasons for non-revascularisation are multifactorial and may relate to the severity of MI, comorbidities, old age, and patient preferences [38], warranting further investigation of the indication for revascularisation. From a prevention perspective, this study provides a unique context of comparing the common medical conditions as predictors of SCD in patients surviving after the four major cardiovascular events. The risk of SCD is found to be varying widely, higher in the HF cohort and lower in the AF cohort. The results underscore the common medical conditions such as MI, HF, arrhythmias, peripheral artery disease, chronic kidney disease, and diabetes are predictive of SCD although some of them were not always predicting SCD through all the cohorts. The findings suggest two potential challenges of offering optimal care for patients. First, current management models for primary-disease-oriented care may not effectively address the health care needs for the four major CVD events for which, currently, clinical guidelines mainly focus treatment [39–42]. Second, polyvascular disease may mediate the relationship between chronic kidney disease, chronic obstructive pulmonary disease and SCD and, therefore, increasing the challenge of successful treatment as this adds to the overall complexity of care [43]. A coordinated comprehensive system of care delivery may be required to address the potentially multiple comorbid conditions for a patient regarding SCD prevention.
Strengths and Limitations The strengths of this study are the utilisation of high-quality administrative health data to capture the study cohort of all incident hospitalised CVD events occurring in a population setting, and the availability of providing long-term follow-up data for this cohort. This study design of using incident cases of major CVD events indicated the natural history of SCD and excluded the potential over-representation derived from prevalent patients with a long course of disease. Application of 15-year look-back period enabled determination of prior hospitalisations of MI, HF, AF and stroke, maximising the ability to identify the incident cohorts. A fixed 15-year lookback period for prior medical conditions of hospitalisation also minimises the under-ascertainment of the status of the medical conditions as potential predictors reported in this study. A number of possible limitations of this study must also be considered. Similar to other studies, identification of SCD is difficult and susceptible to error. Multiple sources were applied for SCD cases ascertainment, although some cases may have been missed and no validation of the cases has taken place. Other factors such as smoking, alcohol use, body mass index, and left ventricular function were not either well recorded or available in the administrative health
data. Thus, whether these factors confound the observed association remains unknown. The results from the cohort of MI may not be generalisable to the whole incident MI population as patients who died within 28 days following the admission date of incident hospitalisation for MI were not included.
Conclusions This population-based study using administrative health data reported the incidence of SCD after an incident hospitalisation for MI, HF, AF or stroke event is appreciable, calling for prevention efforts. Medical conditions such as MI, HF, peripheral artery disease, and arrhythmias are predictors of SCD, broadening our understanding of SCD and requiring optimal care for the survivors of the four major cardiovascular events. Early revascularisation is protective in the MI cohort, emphasising timely intervention.
Authorship Statement The author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and my discussed interpretation.
Acknowledgements The author wishes to thank the staff at the Western Australian Data Linkage Brach and the Department of Health Inpatient Data Collection and Registry of Births, Deaths and Marriages for the provision of data, the Victorian Department of Justice and Regulation, and the National Coronial Information System. This work was supported by a project grant from the National Health and Medical Research Council of Australia, Perth, Western Australia, Australia (No.572558). JLF was supported by an International Postgraduate Research Scholarship (IPRS) and Australian Postgraduate Award when she was doing this work. The funding bodies played no part in the study design, analysis or interpretation of results.
Conflicts of Interest None declared.
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.hlc. 2019.03.020.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020
HLC 2888 No. of Pages 8
8
J.-L. Feng
References [1] Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, et al. Heart disease and stroke statistics—2015 update: a report from the American Heart Association. Circulation 2015;131:e29–322. [2] Australian Institute of Health and Welfare. Cardiovascular disease: Australian facts 2011. Canberra: AIHW;[cited 23 Mar 2018]. Cardiovascular disease series. Cat. no. CVD 53. Available from: 2011, http://www.aihw. gov.au/WorkArea/DownloadAsset.aspx?id=10737418530. [3] Ball J, Thompson DR, Ski CF, Carrington MJ, Gerber T, Stewart S. Estimating the current and future prevalence of atrial fibrillation in the Australian adult population. Med J Aust 2015;202:32–5. [4] Key statistics-Heart Research Australia [cited 23 Mar 2018]. Available from: https://www.heartfoundation.org.au/about-us/what-we-do/ heart-disease-in-australia. [5] Wong CX, Brown A, Lau DH, Chugh SS, Albert CM, Kalman JM, et al. Epidemiology of Sudden Cardiac Death: Global and Regional Perspectives. Heart Lung and Circulation 2019;28:6–14. [6] Adabag AS, Therneau TM, Gersh BJ, Weston SA, Roger VL. Sudden death after myocardial infarction. JAMA 2008;300:2022–9. [7] Fan X, Hua W, Xu Y, Ding L, Niu H, Chen K, et al. Incidence and predictors of sudden cardiac death in patients with reduced left ventricular ejection fraction after myocardial infarction in an era of revascularisation. Heart 2014;100:1242–9. [8] Al-Khatib SM, Shaw LK, O’Connor C, Kong M, Califf RM. Incidence and predictors of sudden cardiac death in patients with diastolic heart failure. J Cardiovasc Electrophysiol 2007;18:1231–5. [9] Eisen A, Ruff CT, Braunwald E, Nordio F, Corbalan R, Dalby A, et al. Sudden Cardiac Death in Patients With Atrial Fibrillation: Insights From the ENGAGE AF-TIMI 48 Trial. J Am Heart Assoc 2016;5. [10] Marijon E, Le Heuzey JY, Connolly S, Yang S, Pogue J, et al. Causes of death and influencing factors in patients with atrial fibrillation: a competing-risk analysis from the randomized evaluation of long-term anticoagulant therapy study. Circulation 2013;128:2192–201. [11] Hagens VE, Rienstra M, Van Veldhuisen DJ, Crijns HJ, et al. Determinants of sudden cardiac death in patients with persistent atrial fibrillation in the rate control versus electrical cardioversion (RACE) study. Am J Cardiol 2006;98:929–32. [12] Soros P, Hachinski V. Cardiovascular and neurological causes of sudden death after ischaemic stroke. Lancet Neurol 2012;11:179–88. [13] Waien SA. Linking large administrative databases: a method for conducting emergency medical services cohort studies using existing data. Acad Emerg Med 1997;4:1087–95. [14] Sanfilippo FM, Hobbs MS, Knuiman MW, Ridout SC, Bradshaw PJ, Finn JC, et al. Can we monitor heart attack in the troponin era? Evidence from a population-based cohort study. BMC Cardiovasc Disord 2011;11:35. [15] Teng TH, Finn J, Hung J, Geelhoed E, Hobbs M. A validation study: how effective is the Hospital Morbidity Data as a surveillance tool for heart failure in Western Australia? Aust N Z J Public Health 2008;32:405–7. [16] Briffa T, Hung J, Knuiman M, McQuillan B, Chew DP, Eikelboom J, et al. Trends in incidence and prevalence of hospitalization for atrial fibrillation and associated mortality in Western Australia, 1995-2010. Int J Cardiol 2016;208:19–25. [17] Jamrozik K, Dobson A, Hobbs M, McElduff P, Ring I, D’Este K, et al. Monitoring the Incidence of Cardiovascular Disease in Australia. Canberra: AIHW;[cited 23 Mar 2018]. Cardiovascular Disease series No.17. Cat No. CVD 16. Available from: https://www.aihw.gov.au/getmedia/ 7cfa7e5c-c37a-42f4-8e6d-e9bff0796029/micda.pdf.aspx?inline=true 2001. [18] Virmani R, Burke AP, Farb A. Sudden cardiac death. Cardiovasc Pathol 2001;10:211–8. [19] Feng JL, Hickling S, Nedkoff L, Knuiman M, Semsarian C, Ingles J, et al. Sudden cardiac death rates in an Australian population: a data linkage study. Aust Health Rev 2015;39:561–7. [20] Solomon SD, Zelenkofske S, McMurray JJ, Finn PV, Velazquez E, Ertl G, et al. Sudden death in patients with myocardial infarction and left ventricular dysfunction, heart failure, or both. N Engl J Med 2005;352:2581–8. [21] Hong MK. Recent Advances in the Treatment of ST-Segment Elevation Myocardial Infarction. Scientifica (Cairo) 2012;2012. 683683.
[22] Zipes DP, Wellens HJ. Sudden cardiac death. Circulation 1998;98:2334– 51. [23] Gorgels AP, Gijsbers C, de Vreede-Swagemakers J, Lousberg A, Wellens HJ. Out-of-hospital cardiac arrest–the relevance of heart failure. The Maastricht Circulatory Arrest Registry. Eur Heart J 2003;24:1204–9. [24] Meta-analysis Global Group in Chronic Heart Failure (MAGGIC). The survival of patients with heart failure with preserved or reduced left ventricular ejection fraction: an individual patient data meta-analysis. Eur Heart J 2012;33:1750–7. [25] Teng TH, Hung J, Finn J. The effect of evidence-based medication use on long-term survival in patients hospitalised for heart failure in Western Australia. Med J Aust 2010;192:306–10. [26] Hayashi M, Shimizu W, Albert CM. The spectrum of epidemiology underlying sudden cardiac death. Circ Res 2015;116:1887–906. [27] Turakhia M, Tseng ZH. Sudden cardiac death: epidemiology, mechanisms, and therapy. Curr Probl Cardiol 2007;32:501–46. [28] Somberg JC, Torres V, Keren G, Butler B, Tepper D, Kleinbaum H, et al. Enhancement of myocardial vulnerability by atrial fibrillation. Am J Ther 2004;11:33–43. [29] Denker S, Lehmann M, Mahmud R, Gilbert C, Akhtar M. Facilitation of ventricular tachycardia induction with abrupt changes in ventricular cycle length. Am J Cardiol 1984;53:508–15. [30] Roy D, Brugada P, Wellens HJ. Atrial tachycardia facilitating initiation of ventricular tachycardia. Pacing Clin Electrophysiol 1983;6:47–52. [31] Brevetti G, Schiano V, Verdoliva S, Silvestro A, Sirico G, et al. Peripheral arterial disease and cardiovascular risk in Italy. Results of the Peripheral Arteriopathy and Cardiovascular Events (PACE) study. J Cardiovasc Med (Hagerstown) 2006;7:608–13. [32] Brevetti G, Giugliano G, Brevetti L, Hiatt WR. Inflammation in peripheral artery disease. Circulation 2010;122:1862–75. [33] Henry RM, Kostense PJ, Bos G, Dekker JM, Nijpels G, Heine RJ, et al. Mild renal insufficiency is associated with increased cardiovascular mortality: The Hoorn Study. Kidney Int 2002;62:1402–7. [34] Tsagalis G, Akrivos T, Alevizaki M, Manios E, Stamatellopoulos K, Laggouranis A, et al. Renal dysfunction in acute stroke: an independent predictor of long-term all combined vascular events and overall mortality. Nephrol Dial Transplant 2009;24:194–200. [35] Lahousse L, Niemeijer MN, van den Berg ME, Rijnbeek PR, Joos GF, Hofman A, et al. Chronic obstructive pulmonary disease and sudden cardiac death: the Rotterdam study. Eur Heart J 2015;36:1754–61. [36] van den Berg ME, Stricker BH, Brusselle GG, Lahousse L. Chronic obstructive pulmonary disease and sudden cardiac death: A systematic review. Trends Cardiovasc Med 2016;26:606–13. [37] Makikallio TH, Barthel P, Schneider R, Bauer A, Tapanainen JM, Tulppo MP, et al. Frequency of sudden cardiac death among acute myocardial infarction survivors with optimized medical and revascularization therapy. Am J Cardiol 2006;97:480–4. [38] Holmes Jr DR, Rihal CS. Revascularization options: one size does not fit all. Cmaj 2012;184:E104–105. [39] Ghimire G, Gupta A, Hage FG. Guidelines in review: 2013 ACCF/AHA Guideline for the Management of ST-Elevation Myocardial Infarction. JNucl Cardiol 2014;21:190–1. [40] Yancy CW, Jessup M, Bozkurt B, Butler J, Casey Jr DE, Drazner MH, et al. ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2013;2013 (62):e147–239. [41] January CT, Wann LS, Alpert JS, Calkins H, Cigarroa JE, Cleveland JC, et al. AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the Heart Rhythm Society. Circulation 2014;2014(130):2071–104. [42] Kernan WN, Ovbiagele B, Black HR, Bravata DM, Chimowitz MI, Ezekowitz MD, et al. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014;45:2160–236. [43] Sales AE, Tipton EF, Levine DA, Houston TK, Kim Y, Allison J, et al. Are co-morbidities associated with guideline adherence? The MI-Plus study of Medicare patients. J Gen Intern Med 2009;24:1205–10.
Please cite this article in press as: Feng J-L. Incidence and Predictors of Sudden Cardiac Death After a Major Non-Fatal Cardiovascular Event. Heart, Lung and Circulation (2019), https://doi.org/10.1016/j.hlc.2019.03.020