Diabetes mellitus,high BMI and low education level predict sudden cardiac death within 24 hours of incident myocardial infarction

Abstract

Background

More than half of cardiovascular mortality occurs outside the hospital, mainly due to consistently low survival rates from out-of-hospital cardiac arrest.

Methods

This is a prospective, nested, case–control study derived from the Västerbotten Intervention Programme and the World Health Organization’s Multinational Monitoring of Trends and Determinants in Cardiovascular Disease study in northern Sweden (1986–2006). To determine predictors for sudden cardiac death risk factors for cardiovascular disease were compared between incident myocardial infarction with sudden cardiac death (n = 363) and survivors of incident myocardial infarction (n = 1998) using multivariate logistic regression analysis.

Results

Diabetes had the strongest association with sudden cardiac death out of all evaluated risk factors (odds ratio (OR) 1.83, 95% confidence interval (CI) 1.30–2.59), followed by low education (OR 1.55, 95% CI 1.19–2.01), high body mass index (OR 1.05, 95% CI 1.02–1.08) and male sex (OR 1.42, 95% CI 1.001–2.01).

Conclusions

The pattern of risk factors for incident myocardial infarction is different among survivors and those who die within 24 hours. The risk factors that contribute the most to death within 24 hours are diabetes mellitus, high body mass index and low education level, and can be addressed at both the public health level and by general practitioners.

Keywords

Sudden cardiac death epidemiology myocardial infarction risk factors prevention prognosis

Introduction

Cardiovascular mortality is the leading cause of death, but reducing this mortality is challenging. Sudden cardiac death (SCD) is at the core of this challenge, as at least two-thirds of these cases occur out of hospital^1–4 and overall survival from a cardiac arrest is less than 10%,^5–7 constituting approximately 20% of the total mortality.⁸ Thus, those suffering from incident myocardial infarction (MI) with SCD do not benefit from the last few decades of medical advancements in secondary prevention.⁹ Consequently, the main potential in reducing mortality from SCD lies outside hospital settings, and primary preventive measures are the most promising strategy for improving survival rates.¹⁰ Many studies present methods for stratifying SCD risk among other risk groups, such as post-MI, structural heart disease, heart failure and diabetes.^11–14 Post-mortem studies on SCD have demonstrated a wide range of underlying conditions,^15,16 which indicates that SCD cannot simply be viewed as a severe onset of the typical MI. The traditional risk factors for incident MI may differentiate between those who survive more than 24 hours (i.e. MI survivors) and those who present with SCD. Cohort data are available but there are no population-based data on the risk markers for SCD or studies comparing the risk factors in MI survivors and those who succumb from post-MI SCD. Identifying the risk factors for SCD and their relative importance among previously healthy individuals is the first logical step to initiating preventive measures that can improve survival. Therefore, our aim was to study the risk factors for SCD based on the hypothesis that the pattern of risk factors for MI is different in survivors and those who die within 24 hours.

Materials and methods

Study population

The study population was derived from the Västerbotten Intervention Programme (VIP)¹⁷ and the World Health Organization’s (WHO) Multinational Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) study in northern Sweden.¹⁸ Both VIP and MONICA are health examination programmes for cardiovascular disease and diabetes with participation rates of 59% and 77%, respectively. The data collection methods in the VIP and MONICA study regarding cardiovascular risk markers were similar. The designs of the VIP and MONICA study have previously been described in detail.^17,18 All MI events in northern Sweden are registered in the population-based MONICA registry according to standardised WHO and MONICA criteria and based on reports from hospitals or general practitioners, as well as hospital discharge records and death certificates.¹⁸ All prospective cases of first MI in VIP and MONICA participants between 1986 and 2006 were identified in the MONICA registry. Post-MI deaths were classified as SCD if they occurred within 24 hours of the onset of symptoms.

A prospective, nested case–control study design was used. Subjects with a history of MI before the health examination were excluded, as experiencing a MI may influence the studied determinants. Up to four controls matched by sex, age (±2 years) and year and type of health examination were selected for each case. A total of 2361 cases of incident MI and 8502 controls were identified. Of the incident MI cases, 363 experienced SCD.

Baseline variables

Smoking habits were classified into ‘daily smoking’ or ‘non-smoking’ (including previous smokers and occasional smokers). Body mass index (BMI) was calculated as weight in kilograms divided by height in metres squared. Hypertension was defined as a systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg, or reported use of anti-hypertensive medication during the past 14 days. Blood samples were taken after a minimum 4-hour fast and analysed for total cholesterol and plasma glucose using a bench-top analyser (Reflotron, Boehringer Mannheim GmbH Diagnostica, Germany). Since 2005, a HemoCue bench-top analyser (Quest Diagnostics) has been used for glucose values. An oral glucose tolerance test was performed with a 75 g oral glucose load according to WHO standards. Diabetes was defined as self-reported disease on a questionnaire or a fasting glucose ≥7 mmol/L and/or a 2-hour post-load plasma glucose ≥11.0 mmol/L (≥12.2 mmol/L in the VIP, as capillary plasma was drawn). Education level was dichotomised into 9 years of compulsory education or a higher level of education.

Statistical analysis

Using logistic regression we analysed the associations between risk factors and MI, comparing SCD cases and MI survivors. Multivariate models included the traditional cardiovascular risk factors age, smoking, diabetes, education level, BMI, cholesterol level and systolic blood pressure. Univariate conditional logistic regression including a single potential risk variable was used to analyse crude odds ratios (ORs) and 95% confidence intervals (CIs) for each potential risk factor based on the matched case–control sets. The multivariate model included smoking, diabetes, education, BMI, cholesterol level and systolic blood pressure. The case–control sets were matched for age and sex. The statistical computations were carried out in SPSS for Windows (versions 15 and 17; SPSS Inc., Chicago, IL, USA). The study was approved by the research ethics committee of Umeå University. All participants provided informed consent.

Results

A total of 363 MI cases with SCD and 1318 controls matched for screening time, age and sex were included with 1998 MI survivors and 7184 matched controls. Of the 2361 MI cases, 16% of the men and 14% of the women presented with SCD; 288 (79%) of the 363 SCD cases occurred out of hospital. No significant difference was found in the time from screening to event between SCD cases and MI survivors (P = 0.67).

Diabetes or the use of diabetes medications was reported by 3.1% of all patients. When including those with blood glucose levels >6.9 mmol/L in fasting samples or 12.1 mmol/L in the oral glucose tolerance test, the prevalence more than doubled to 6.3%. At baseline, SCD cases generally had a higher risk factor profile than MI survivors, regardless of sex. Notably, the frequency of diabetes was almost doubled among SCD cases compared to MI survivors for both sexes (Table 1). For the baseline data of controls see Appendix 1. Diabetes was the single factor with the strongest association with SCD compared to MI survivors, followed by low education, BMI and systolic blood pressure. For men the results were similar as for both sexes combined, but among women the education level and systolic blood pressure did not differ significantly between SCD cases and MI survivors (Table 2). In the multivariate model compared to crude analysis, male sex, but not systolic blood pressure, was associated with SCD compared to MI survivors. Among men, diabetes and low education were associated with SCD. Among women, only BMI remained associated with SCD (Table 2).

Table 1.

Baseline characteristics of 1507 men and 491 women with myocardial infarction without sudden death and 284 men and 79 women with myocardial infarction with sudden death.

	Men		Women
	MI (n = 1507)	SCD (n = 284)	MI (n = 491)	SCD (n = 79)
Age at screening (years)	50/57/60	50/59/60	50/60/60	50/60/60
Age at event (years)	56/62/67	57/63/67	58/64/70	58/65/70
Smoking (%)	39	42	43	51
Diabetes (%)	9.4	16.2	12.6	21.5
Low level of education (%)	42	49	49	53
BMI (kg/m²)	26.8 ± 3.5	27.4 ± 3.8	26.7 ± 4.6	29.0 ± 6.1
Cholesterol (mmol/L)	6.4 ± 1.2	6.4 ± 1.2	6.6 ± 1.4	6.7 ± 1.5
SBP (mmHg)	138 ± 16.8	142 ± 18.3	141 ± 19.0	142 ± 19.6

Data are presented as median and interquartile range, proportion, and mean ± standard deviation respectively. Low level of education refers to 9 years of compulsory education.

BMI: body mass index; SBP: systolic blood pressure.

Table 2.

Univariate and multivariate logistic regression analysis of incident myocardial infarction with sudden cardiac death compared to 24-hour survival.

	All		Men		Women
	Crude	Multivariate^a	Crude	Multivariate^a	Crude	Multivariate^a
Age	1.01 (0.997–1.03)	1.01 (0.98–1.02)	1.02 (0.998–1.03)	1.02 (0.98–1.02)	1.00 (0.97–1.03)	0.99 (0.95–1.02)
Sex^b	1.17 (0.90–1.53)	1.42 (1.001–2.01)	–	–	–	–
Smoking	1.28 (0.99–1.66)	1.21 (0.91–1.63)	1.19 (0.89–1.60)	1.16 (0.83–1.61)	1.64 (0.96–2.82)	1.37 (0.73–2.58)
Diabetes	1.85 (1.36–2.51)	1.83 (1.30–2.59)	1.86 (1.30–2.66)	1.90 (1.28–2.81)	1.90 (1.04–3.45)	1.66 (0.79–3.49)
Low education	1.53 (1.21–1.94)	1.55 (1.19–2.01)	1.52 (1.17–1.99)	1.55 (1.16–2.07)	1.67 (0.98–2.87)	1.69 (0.92–3.12)
BMI	1.06 (1.03–1.09)	1.05 (1.02–1.08)	1.05 (1.01–1.08)	1.03 (0.99–1.08)	1.09 (1.04–1.14)	1.09 (1.03–1.15)
Cholesterol	0.99 (0.90–1.08)	0.98 (0.89–1.08)	0.97 (0.87–1.08)	0.95 (0.85–1.06)	1.06 (0.89–1.27)	1.11 (0.91–1.35)
SBP	1.01 (1.002–1.01)	1.004 (0.996–1.01)	1.01 (1.003–1.02)	1.01 (0.996–1.01)	1.00 (0.99–1.02)	1.00 (0.98–1.02)

Data are presented as odds ratio (95% confidence interval). Low level of education refers to 9 years of compulsory education.

BMI: body mass index; SBP: systolic blood pressure.

Multivariate model including age, smoking, diabetes, education, BMI, cholesterol and SBP.

Men compared to women.

The strongest risk factor for SCD compared to age and sex-matched controls in the multivariate conditional model was diabetes (OR 3.51, 95% CI 2.34–5.51), followed by smoking (OR 2.97, 95% CI 2.11–4.17) and low education (OR 1.63, 95% CI 1.19–2.23) (Table 3).

Table 3.

Univariate and multivariate conditional logistic regression analysis of incident myocardial infarction with or without sudden cardiac death.

	SCD		MI survivors
	Crude	Multivariate ^a	Crude	Multivariate^a
Smoking	2.84 (2.13–3.77)	2.97 (2.11–4.17)	2.10 (1.87–2.36)	2.14 (1.88–2.43)
Diabetes	3.78 (2.61–5.5)	3.51 (2.34–5.51)	2.17 (1.81–2.60)	1.86 (1.52–2.28)
Low education	1.97 (1.50–2.60)	1.63 (1.19–2.23)	1.13 (1.01–1.26)	1.02 (0.90–1.14)
BMI	1.11 (1.08–1.15)	1.09 (1.05–1.14)	1.05 (1.04–1.07)	1.03 (1.01–1.04)
Cholesterol	1.35 (1.22–1.51)	1.28 (1.13–1.45)	1.29 (1.24–1.35)	1.24 (1.19–1.30)
SBP	1.02 (1.01–1.03)	1.01 (1.004–1.02)	1.02 (1.01–1.02)	1.01 (1.01–1.02)

Data are presented as odds ratio (95% confidence interval). Controls were matched for screening time, age, and sex. Low education refers to 9 years of compulsory education.

BMI: body mass index; SBP: systolic blood pressure.

Multivariate model including smoking, diabetes, education, BMI, cholesterol and SBP.

To analyse the combined effect of the risk factors on SCD, we constructed four groups based on the presence of pairs of the selected dichotomised risk factors in a conditional logistic regression analysis. The presence of diabetes mellitus and low education versus none of these had an OR of 6.33 for SCD compared to matched controls (95% CI 3.70–10.8). The risk for SCD among persons with diabetes mellitus and smoking had an OR of 8.45 (95% CI 4.25–16.8) compared to non-smokers without diabetes. Finally, smokers with low education had an OR of 4.45 (95% CI 3.00–6.58) compared to non-smokers with higher education.

The analysis also included the metabolic risk factors diabetes mellitus and BMI dichotomised at 30 kg/m². The combined variable exhibited a risk increase for SCD (OR 8.53, 95% CI 4.01–18.13) compared to individuals with neither of these risk factors (Table 4). The interaction variable with BMI and diabetes was not significant (P = 0.41). We conducted separate sensitivity analyses in which the two SCD cases with BMI <20 kg/m² were excluded, but it had no impact on the results (data not shown).

Table 4.

Univariate conditional logistic regression analysis of myocardial infarction with or without sudden cardiac death.

	SCD	Survivors
Non-diabetic with BMI <30 kg/m²	–	–
Non-diabetic with BMI >30 kg/m²	1.79 (1.27–2.51)	1.42 (1.23–1.65)
Diabetes with BMI <30 kg/m²	3.30 (2.16–5.05)	2.19 (1.75–2.74)
Diabetes with BMI >30 kg/m²	8.53 (4.01–18.13)^a	2.46 (1.83–3.31)

Data are presented as odds ratio (95% confidence interval). Controls were matched for screening time, age and sex.

BMI: body mass index; SCD: sudden cardiac death.

P-value for interaction 0.41.

Discussion

Our main finding in the present study is that the risk factors for dying within 24 hours of the onset of an incident MI compared to surviving is diabetes, low education level, high BMI and male gender. The combination of BMI >30 kg/m² and diabetes was associated with an 8-fold increased risk of SCD compared to matched controls without MI. Our findings are in line with previous research and provide further evidence that both diabetes and obesity increase the risk of SCD. A recent meta-analysis of 10 prospective and four case–control studies reported a 2-fold increased risk of SCD among people with diabetes.¹³ A Swedish cohort study from Göteborg found diabetes more common among fatal than non-fatal coronary events.¹⁹ Other cohort studies have similarly found associations between diabetes and SCD,³ and several mechanisms have been suggested. The increased risk of MI and SCD in patients with diabetes mellitus may be related to inflammation, diabetic cardiomyopathy and diabetic autonomic neuropathy with subsequent arrhythmia.^13,20 Previous cohort studies have shown up to 2 to 3-fold increased risk for SCD among obese patients,^21–23 supporting our results. The mechanism of increased SCD among obese individuals could be mediated through several mechanisms, such as haemodynamic disturbances, hypertension and left ventricular hypertrophy or cardiac electrical abnormalities.²¹

For patients with incident MI admitted alive to the hospital, the odds of survival are good.^9,24,25 The vast majority of patients who die from cardiovascular events do so out of hospital. Implantation of implantable cardioverter defibrillators in individuals at very high risk of SCD is one option, but is limited to a selected high-risk group. Early awareness of warning signs, early cardiopulmonary resuscitation, and public access to automated external defibrillators are important strategies, but they have limitations. Public access defibrillation requires early detection of out-of-hospital cardiac arrest and the presence of ventricular fibrillation or flutter.²⁶ Thus, the potential for improving MI survival rates is outside the hospital, mainly involving primary prevention.

Our study shows that lifestyle factors are important not just for the risk of developing incident MI, but also for early prognosis, and that the strongest contributors may be modified in the general population. We identified three possible ways to decrease the risk attributable to these risk factors in the general population and selected high-risk groups. These strategies aim to lower out-of-hospital cardiovascular mortality.

First, the factors most strongly associated with an early bad prognosis, diabetes and obesity should be addressed through preventive lifestyle and public health interventions to avoid or delay disease. In public health interventions, improving the education level of the population is important, especially to encourage and support further education beyond 9 years of compulsory school. Public health interventions reach the entire population and will also decrease overall MI incidence.

Second, the early detection and treatment of type 2 diabetes mellitus is important, especially among those with BMI >30 kg/m², who have an 8-fold increased risk of dying within 24 hours of their first MI. In the analysis of blood glucose levels sampled approximately 5 years before the SCD event, 50% of the individuals with diabetes were not treated or stated that they did not have diabetes. Delayed diagnosis and treatment may have contributed to the fatal outcome of incident MI. Previous studies have shown that impaired fasting plasma glucose increases the risk of SCD 1.51-fold.²⁷

Third, medical treatment of the remaining cardiovascular risk factors in the general population should be addressed. At this point, reducing cholesterol levels and hypertension are important ways to improve not only early prognosis, but also long-term prognosis, and reduce overall MI events. Smoking cessation should be encouraged at all levels of the health care system. A cardiovascular risk prevention strategy among risk groups with a high incidence of diabetes was associated with better prognosis and could be a suitable model for the general practitioner.²⁸

However, further studies are needed to confirm our findings and verify that our suggested measures are effective at reducing the out-of-hospital cardiovascular mortality. Studies with interventions in high-risk subjects with diabetes and obesity are especially important. To the best of our knowledge, this study is the first prospective, nested, case–control study investigating whether the established cardiovascular risk factors for incident MI are different between survivors and those who die within 24 hours. We also want to emphasise that we studied incident MIs, as these patients are not subject to secondary preventive medication or other interventions that may affect the studied risk factors.

Our study is unique for two reasons. Firstly, it includes all incident MI cases with SCD, even those who died before being admitted to hospital care and cases in which no cardiopulmonary resuscitation was attempted. This is especially important as a majority of cardiovascular deaths^4,11,14,27 occur outside hospital settings. Secondly, questionnaires, BMI, blood pressure, blood samples and other data were collected during a health examination prior to the SCD events.

Our study has some limitations. We only had 79 female SCD cases, limiting the gender analysis. The participants in this study were mainly middle-aged Caucasians from northern Sweden. Therefore, the results cannot be applied to other populations with certainty. We have adjusted our results for traditional risk factors but not biomarkers, and as such we cannot exclude the possibility that biomarkers, such as C-reactive protein, affected the results.

Conclusion

The pattern of risk factors for MI is different among survivors and those who die within 24 hours. Except for the male sex, all risk factors associated with death within 24 hours can be addressed at both the public health level and by general practitioners.

Footnotes

Author contribution

DL, SE and JH contributed to the conception or design of the work. JA, PW, DL, SE and JH contributed to the acquisition, analysis, or interpretation of data for the work. JA drafted the manuscript. JA, PW, DL, SE and JH critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Appendix 1

References

Chambless

Keil

Dobson

. Population versus clinical view of case fatality from acute coronary heart disease: results from the Who MONICA Project 1985–1990. Multinational monitoring of trends and determinants in cardiovascular disease. Circulation 1997; 96: 3849–3859.

Müller

Agrawal

Arntz

. How sudden is sudden cardiac death? Circulation 2006; 114: 1146–1150.

Norris

. Fatality outside hospital from acute coronary events in three British health districts, 1994–5. United Kingdom Heart Attack Study Collaborative Group. BMJ 1998; 316: 1065–1070.

Zheng

Croft

Giles

. Sudden cardiac death in the United States, 1989 to 1998. Circulation 2001; 104: 2158–2163.

Berdowski

Berg

Tijssen

. Global incidences of out-of-hospital cardiac arrest and survival rates: systematic review of 67 prospective studies. Resuscitation 2010; 81: 1479–1487.

Fredriksson

Herlitz

Nichol

. Variation in outcome in studies of out-of-hospital cardiac arrest: a review of studies conforming to the Utstein guidelines. Am J Emerg Med 2003; 21: 276–281.

Strömsöe

Svensson

Claesson

. Association between population density and reported incidence, characteristics, and outcome after out-of-hospital cardiac arrest in Sweden. Resuscitation 2011; 82: 1307–1313.

Wellens

Schwartz

Lindemans

. Risk stratification for sudden cardiac death: current status and challenges for the future. Eur Heart J 2014; 35: 1642–1651.

Tunstall-Pedoe

Vanuzzo

Hobbs

. Estimation of contribution of changes in coronary care to improving survival, event rates, and coronary heart disease mortality across the WHO MONICA Project populations. Lancet 2000; 355: 688–700.

10.

Ilkhanoff

Goldberger

. Out-of-hospital cardiac arrest getting beyond the tip of the iceberg. Circulation 2012; 126: 793–796.

11.

Goldberger

Cain

Hohnloser

. American Heart Association/American College of Cardiology Foundation/Heart Rhythm Society scientific statement on noninvasive risk stratification techniques for identifying patients at risk for sudden cardiac death: a scientific statement from the American Heart Association Council on Clinical Cardiology Committee on Electrocardiography and Arrhythmias and Council on Epidemiology and Prevention. Circulation 2008; 118: 1497–1518.

12.

Piccini

Zhang

Pieper

. Predictors of sudden cardiac death change with time after myocardial infarction: results from the Valiant Trial. Eur Heart J 2010; 31: 211–221.

13.

Zaccardi

Khan

Laukkanen

. Diabetes mellitus and risk of sudden cardiac death: a systematic review and meta-analysis. Int J Cardiol 2014; 177: 535–537.

14.

Zaman

Kovoor

. Sudden cardiac death early after myocardial infarction: pathogenesis, risk stratification, and primary prevention. Circulation 2014; 129: 2426–2435.

15.

Hertz

Ferrero-Miliani

Frank-Hansen

. A comparison of genetic findings in sudden cardiac death victims and cardiac patients: the importance of phenotypic classification. Europace 2015; 17: 350–357.

16.

Pouleur

Barkoudah

Uno

. Pathogenesis of sudden unexpected death in a clinical trial of patients with myocardial infarction and left ventricular dysfunction, heart failure, or both. Circulation 2010; 122: 597–602.

17.

Norberg

Wall

Boman

. The Västerbotten Intervention Programme: background, design and implications. Glob Health Action 2010; 4643. DOI: 10.3402/gha.v3i0.4643.

18.

Stegmayr

Lundberg

Asplund

. The events registration and survey procedures in the Northern Sweden MONICA Project. Scand J Public Health Suppl 2003; 61: 9–17.

19.

Wilhelmsen

Köster

Harmsen

. Differences between coronary disease and stroke in incidence, case fatality, and risk factors, but few differences in risk factors for fatal and non-fatal events. Eur Heart J 2005; 26: 1916–1922.

20.

Vasiliadis

Kolovou

Mavrogeni

. Sudden cardiac death and diabetes mellitus. J Diabetes Complications 2014; 28: 573–579.

21.

Adabag

Huxley

Lopez

. Obesity related risk of sudden cardiac death in the Atherosclerosis Risk in Communities study. Heart 2015; 101: 215–221.

22.

Bertoia

Allison

Manson

. Risk factors for sudden cardiac death in post-menopausal women. J Am Coll Cardiol 2012; 60: 2674–2682.

23.

Noheria

Teodorescu

Uy-Evanado

. Distinctive profile of sudden cardiac arrest in middle-aged vs. older adults: a community-based study. Int J Cardiol 2013; 168: 3495–3499.

24.

Herlitz

Dellborg

Karlson

. Prognosis after acute myocardial infarction continues to improve in the reperfusion era in the community of Göteborg. Am Heart J 2002; 144: 89–94.

25.

Yang

Dzayee

Beiki

. Incidence and case fatality after day 28 of first time myocardial infarction in Sweden 1987–2008. Eur J Prev Cardiol 2012; 19: 1304–1315.

26.

Ringh

Herlitz

Hollenberg

. Out-of-hospital cardiac arrest outside home in Sweden, change in characteristics, outcome, and availability for public access defibrillation. Scand J Trauma Resusc Emerg Med 2009; 17:18. DOI: 10.1186/757-7241-17-18.

27.

Laukkanen

Mäkikallio

Ronkainen

. Impaired fasting plasma glucose and type 2 diabetes are related to the risk of out-of-hospital sudden cardiac death and all-cause mortality. Diabetes Care 2013; 36: 1166–1171.

28.

Avanzini F, Marzona I, Baviera M, et al. Improving cardiovascular prevention in general practice: results of a comprehensive personalized strategy in subjects at high risk. Eur J Prev Cardiol 2016; 23: 947–955.