Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Remote ischemic conditioning wikipedia , lookup
Cardiac contractility modulation wikipedia , lookup
Saturated fat and cardiovascular disease wikipedia , lookup
Cardiac surgery wikipedia , lookup
Jatene procedure wikipedia , lookup
Cardiovascular disease wikipedia , lookup
Antihypertensive drug wikipedia , lookup
Coronary artery disease wikipedia , lookup
Original Article | doi: 10.1111/j.1365-2796.2009.02088.x Chocolate consumption and mortality following a first acute myocardial infarction: the Stockholm Heart Epidemiology Program I. Janszky1, K. J. Mukamal2, R. Ljung1,3, S. Ahnve1, A. Ahlbom4 & J. Hallqvist1,5 From the 1Department of Public Health Sciences, Karolinska Institute, Stockholm, Sweden; 2Division of General Medicine and Primary Care, Beth Israel Deaconess Medical Center, Boston, MA, USA; 3Centre for Epidemiology, The National Board of Health and Welfare, Stockholm; 4Department of Epidemiology, Institute of Environmental Medicine, Karolinska Institute, Stockholm; and 5Department of Public Health and Caring Sciences, Uppsala University, Uppsala, Sweden Abstract. Janszky I, Mukamal KJ, Ljung R, Ahnve S, Ahlbom A, Hallqvist J (Karolinska Institute, Stockholm, Sweden; Beth Israel Deaconess Medical Center, Boston, MA, USA; The National Board of Health and Welfare, Stockholm; Karolinska Institute, Stockholm; and Uppsala University, Uppsala, Sweden). Chocolate consumption and mortality following a first acute myocardial infarction: the Stockholm Heart Epidemiology Program. J Intern Med 2009; 266: 248–257. Objectives. To assess the long-term effects of chocolate consumption amongst patients with established coronary heart disease. Design. In a population-based inception cohort study, we followed 1169 non-diabetic patients hospitalized with a confirmed first acute myocardial infarction (AMI) between 1992 and 1994 in Stockholm County, Sweden, as part of the Stockholm Heart Epidemiology Program. Participants self-reported usual chocolate consumption over the preceding 12 months with a standardized questionnaire distributed during hospitalization and underwent a health examination 3 months after discharge. Participants were followed for hospitalizations and mortality with national registries for 8 years. Introduction Health effects of chocolate have been of great interest in recent years. As chocolate is a significant source of 248 ª 2009 Blackwell Publishing Ltd Results. Chocolate consumption had a strong inverse association with cardiac mortality. When compared with those never eating chocolate, the multivariableadjusted hazard ratios were 0.73 (95% confidence interval, 0.41–1.31), 0.56 (0.32–0.99) and 0.34 (0.17– 0.70) for those consuming chocolate less than once per month, up to once per week and twice or more per week respectively. Chocolate consumption generally had an inverse but weak association with total mortality and nonfatal outcomes. In contrast, intake of other sweets was not associated with cardiac or total mortality. Conclusions. Chocolate consumption was associated with lower cardiac mortality in a dose dependent manner in patients free of diabetes surviving their first AMI. Although our findings support increasing evidence that chocolate is a rich source of beneficial bioactive compounds, confirmation of this strong inverse relationship from other observational studies or largescale, long-term, controlled randomized trials is needed. Keywords: acute myocardial infarction, chocolate, prognosis. flavonoid antioxidants [1], several short-term clinical trials have shown beneficial cardiovascular effects of chocolate or cocoa consumption, including improvement of endothelial function [2–5], inhibition of I. Janszky et al. | Chocolate consumption and post-AMI prognosis platelet activation [2, 6, 7] and LDL oxidation [8, 9] and decrease of blood pressure [10, 11]. Limited evidence also supports the hypothesis that chocolate might have long-term protective effects on cardiovascular events. Population-based studies have found that chocolate or cocoa consumption is associated with lower cardiovascular mortality both amongst elderly men free of known coronary heart disease [12] and postmenopausal women [13]. The long-term effects of chocolate consumption amongst patients with established coronary heart disease is largely unknown and to the best of our knowledge, the prospective association between chocolate consumption and prognosis in survivors of acute myocardial infarction (AMI) has not been explored. Therefore, we investigated the long-term prospective relationships between chocolate consumption and total mortality, cardiac mortality and nonfatal end-points including recurrent AMI, hospitalization for heart failure and stroke amongst patients surviving their first AMI in a population-based study of AMI. Methods Subjects and design We followed individuals enrolled as nonfatal AMI cases in the Stockholm Heart Epidemiology Program (SHEEP), a population-based case–control study of incident AMI [14]. The study base comprised all Swedish citizens living in the Stockholm County, 45– 70 years of age, free of previous clinically diagnosed AMI. Male cases were identified during a 2-year period (1992–93) and female cases during 3 years (1992–94). Cases were identified through a special organization at the 10 emergency hospitals in the region. Criteria for AMI included (i) symptoms consistent with AMI, (ii) positive blood levels of the enzymes CK and LD and (iii) specified ECG-changes. The diagnosis ‘acute myocardial infarction’ required two of the criteria (i–iii) to be met. Later comparison with a population-based register-based incidence register indicated close to complete ascertainment of all first AMIs [15]. A total of 2246 cases of MI were identified, of which 1603 were nonfatal defined as surviving the AMI for at least 28 days. Eighty-six percent or 1381 of the nonfatal cases participated in the questionnaires handed out a few days after the AMI and 1361 provided complete data on chocolate consumption. To minimize reverse confounding by patients who may have been counselled to avoid chocolate or other sweets, we excluded patients with diabetes mellitus (n = 192), yielding a total of 1169 patients in subsequent analyses. A health examination measuring blood pressure, height and weight with a blood sampling was undertaken 3 months after the AMI onset. A total of 1051 non diabetic patients with complete chocolate data participated in the health examination and hence had information on measured blood pressure and biomarkers. Chocolate consumption The questionnaire distributed a few days after AMI onset queried the number of usual (50 g) portions of chocolate that participants usually consumed per day, per week or per month during the last 12 months. The original consumption categories included: never, less than once per month, 1–3 times per month, once per week, twice per week, 3–4 times per week, 5–6 times per week, once per day, twice per day and 3 times or more per day. Based upon the distribution of these original categories within the SHEEP population, we categorized usual chocolate consumption as never, less than once per month, up to once per week and twice or more per week. Covariates Lipids, coagulation, inflammation. As previously described elsewhere [14], lipids, coagulation factors [16] and inflammatory markers [17] were measured from blood samples drawn by venous puncture after overnight fasting at the health examination. Glucose, insulin, insulin-like growth factor binding protein-1 (IGFBP-1), insulin resistance. Glucose, insulin and IGFBP-1 levels were determined from fasting blood samples. An estimate of insulin ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 249 I. Janszky et al. | resistance was calculated using the homeostasis model assessment (HOMA-IR) as follows: insulin resistance = fasting glucose · fasting insulin ⁄ 22.5 [18]. Hypertension. Hypertension was defined as (i) being on antihypertensive drug therapy, for the reason of hypertension, when included in the study; (ii) a history of regular antihypertensive drug therapy during the last 5 years (or a part of that time); (iii) a systolic blood pressure ‡170 mmHg or a diastolic blood pressure ‡95 mmHg. Blood pressure values were the mean of two measurements in supine position after 5 min rest at the health examination, i.e. 3 months after AMI. Obesity. Patients with a measured BMI-value over 30 kg m)2 were classified as being obese. Physical inactivity. Patients who reported inactive leisure time, which included occasional walks, during the last 5–10 years were categorized as physically inactive. Smoking. Subjects who had never smoked regularly (i.e. for at least 1 year) were considered as neversmokers. Subjects, who smoked when included into the study or had stopped smoking within the last 2 years, were classified as smokers. Subjects, who had stopped smoking for more than 2 years before inclusion, were classified as ex-smokers. Chocolate consumption and post-AMI prognosis Pharmacological treatment was collected through the questionnaires and also at the health examination. Patients were asked to report on current use of pharmacological drugs, all brand names and these were later assigned generic codes. Both cardiac and non cardiac medications were recorded. The first one included calcium channel blockers, b-blockers, ACE inhibitors, diuretics, digitalis, nitrates and aspirin treatment. Follow-up The health care system in Sweden provides virtually complete follow-up information for all patients by matching their unique 10 digit person identification numbers to health care registers. The average followup, from the AMI, was 3158 days (SD = 257 days, median = 3164 days, interquartile range = 376 days). All-cause and cardiac mortality were used as primary end-points as provided by the National Causeof-death Register. The ICD 9 and 10 codes for cardiac mortality included 410–414, 420–429, I20–I25, I38–I51 respectively. Patients were also followed for nonfatal AMI using the Swedish Myocardial Infarction Register [22]. Information on hospitalization for heart failure (ICD-9 and 10 codes were 428, I50 respectively) and stroke (431, 434, I64, I63, I61) was derived from the Swedish Hospital Discharge Register [23–25]. Follow-up was closed on 1 December 2001. Educational attainment. We classified educational attainment as mandatory school only versus high school, college or university. Consumption of sweets and desserts. Patients were queried about their usual intake of six types of sweets – biscuits, cookies, cakes, pastries, confectioneries and ice cream, using the same response options as for chocolate. Each variable was categorized into quartiles and a value of 1–4 was assigned to each. The score values were added up to form a single sweet score variable ranging from 6–24. Consumption of alcohol and filtered coffee was assessed with a semiquantitative food frequency questionnaire, as previously described [19–21]. 250 Statistical analyses We used Cox proportional hazard models to examine the association between chocolate consumption and cardiac and all-cause mortality, new nonfatal AMI and hospitalization for stroke and heart failure. The group of never consumers was the reference category in these models. We tested the proportionality of hazards using log-log curves and formal tests of interaction with time or log-time. Primary confounding factors for evaluation of chocolate consumption we included were age (in 5 categories), gender, smoking (in 3 categories), obesity, physical inactivity, alcohol consumption (in g day)1), filtered coffee consumption (in cups per day), educational ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 I. Janszky et al. | Chocolate consumption and post-AMI prognosis attainment and sweet score. As blood pressure is a plausible mediator of the effects of chocolate on prognosis, we included it only in secondary analyses. In further analyses, we also added additional clinical characteristics of the index AMI (Killip class, Q-wave versus non-Q-wave MI and history of ventricular tachycardia), which were simultaneously available only in 920 participants. Rothman’s synergy index with 95% confidence intervals was used to evaluate interaction between chocolate consumption and clinical characteristics [26]. To examine potential mediators, we calculated adjusted mean levels of biomarkers according to chocolate consumption and then compared the change in regression coefficients associated with chocolate modelled as a continuous variable from multivariable proportional hazards models with and without inclusion of specific biomarkers. For tests of linear trend, we treated the median value of chocolate intake in grams within categories as a continuous variable. Statistical analyses were performed using sas 9 for Windows (SAS Inc., Cary, NC, USA). Results Patient characteristics Table 1 shows characteristics of SHEEP participants according to chocolate intake. In general, prevalence of hypertension, sedentary lifestyle and lower education decreased with increasing chocolate consumption. Chocolate consumption was strongly correlated with sweet score (Spearman’s rho = 0.41, P < 0.001). Chocolate consumption and mortality Table 2 shows the relationships of chocolate consumption with total and cardiac mortality. There was an inverse, dose–response relationship between chocolate consumption and cardiac mortality both in ageand gender-adjusted models and after controlling for other potential confounders. In contrast, sweet score had no statistically significant relationship to cardiac or total mortality. The multivariable-adjusted hazard ratios (HRs) for a single-unit increase in the sweet Table 1 Characteristics of the patients according to usual chocolate consumption Never n 166 Less than once Up to once Twice or more per month per week per week 250 453 300 Mean (SD) Mean (SD) Mean (SD) Mean (SD) Age (years) 60.7 (7.5) 59.1 (7.0) 58.6 (7.4) 59.5 (6.7) n (%) n (%) n (%) n (%) Male gender 110 (66.3) 188 (75.2) 328 (72.4) 205 (68.3) Hypertension 71 (43.0) 88 (35.3) 161 (35.7) 101 (34.2) Sedentary lifestyle 87 (52.7) 120 (48.2) 194 (43.5) 125 (42.0) Overweight (BMI > 30) 29 (17.5) 39 (15.7) 79 (17.5) 49 (16.3) Nonsmokers 40 (24.1) 63 (25.2) 106 (23.4) 63 (21.0) Previous smokers 34 (20.5) 64 (25.6) 120 (26.5) 64 (21.3) Current smokers 92 (55.4) 123 (49.2) 227 (50.1) 173 (57.7) 48 (29.5) 68 (27.5) 152 (33.6) 104 (34.9) 99 (62.7) 125 (54.6) 241 (57.5) 168 (61.1) 1 95 (67.4) 151 (72.3) 290 (75.1) 162 (66.7) 2+ 46 (32.6) 58 (27.8) 96 (24.9) 81 (33.3) Cigarette smoking High School ⁄ College or University Index hospitalization Q wave infarction Killip classification ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 251 I. Janszky et al. | Chocolate consumption and post-AMI prognosis Table 2 Risk of fatal outcomes following acute myocardial infarction amongst SHEEP participants according to usual chocolate consumption Never Total mortality, n (%) Age-gender adjusted HR Age-gender adjusted HR per week 41 (16.4) 82 (18.1) 52 (17.3) 0.81 0.92 0.85 (0.52–1.28) (0.62–1.37) (0.55–1.31) 1.0 0.89 0.96 0.94 (0.56–1.42) (0.63–1.48) (0.58–1.53) 24 (14.5) 24 (9.6) 41 (9.1) 18 (6.0) 1.0 0.69 0.67 0.43 (0.39–1.23) (0.40–1.11) (0.23–0.80) 0.73 0.56 0.34 (0.41–1.31) (0.32–0.99) (0.17–0.70) 95% CI Multivariable HR Twice or more per week 35 (21.1) 95% CI Cardiac mortality, n (%) Up to once per month 1.0 95% CI Multivariable HR Less than once 1.0 95% CI P trend 0.69 0.96 0.02 0.01 Adjusted HR indicated hazard ratio adjusted for age (in 5 categories), gender, smoking (in 3 categories), obesity, physical inactivity, alcohol consumption (in g day)1), filtered coffee consumption (in cups per day), educational attainment and sweet score, 95% CI indicates 95% confidence interval. score were 0.98 (0.95–1.02) and 0.98 (0.93–1.04) for total and cardiac mortality respectively. Additional adjustment for hypertension and systolic blood pressure did not attenuate the risk associated with chocolate intake with cardiac mortality. In further analyses, we additionally controlled for markers of infarct severity (Killip class, maximum CK levels, ventricular tachycardia and Q-wave MI) with no material change in the observed HRs. Our results concerning the strong inverse association between chocolate consumption and cardiac mortality were also robust to alternative categorizations or using chocolate consumption as a continuous variable. Chocolate consumers had slightly lower total mortality in all categories when compared to chocolate abstainers, but the relationship was weaker than that of with cardiac mortality. Chocolate consumption and nonfatal outcomes Table 3 shows the multivariable-adjusted associations of chocolate consumption with risks of hospitalization for nonfatal cardiovascular events alone and in combination. In general, chocolate consumers tended to have slightly lower risk for these events, but there was no clear evidence of association. 252 We observed a modest variation in the association between chocolate consumption and stroke over time (i.e. non proportionality of hazards) restricted to those consuming chocolate less than once per month compared with chocolate abstainers, but not for any other categories of intake. Those consuming chocolate less than once per month had a more substantially lower risk for stroke after the first 4 years of follow-up (HR 0.48, 95% CI, 0.20–1.18) than during the first half of follow-up (HR 0.91, 95% CI 0.39–2.08). The proportional hazards assumption was met for all other outcomes. Potential mediators Table 4 presents least squares means of the potential mediators of the effect of chocolate according to chocolate intake when adjusted for age, gender, obesity, physical inactivity, smoking, education, alcohol and filtered coffee consumption and sweet score (i.e. the same factors we included to the multivariable proportional hazard models). Insulin level showed a modest inverse relationship with chocolate intake, but no other significant graded associations were observed. We also compared the association of chocolate intake with cardiac mortality before and after adjustment for potentially mediating ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 I. Janszky et al. | Chocolate consumption and post-AMI prognosis Table 3 Risk of nonfatal outcomes following acute myocardial infarction amongst SHEEP participants according to usual chocolate consumption Recurrent AMI, n (%) Adjusted HR Less than once Up to once Twice or more Never per month per week per week 36 (21.7) 51 (20.4) 104 (23.0) 59 (19.7) 1.0 0.95 1.02 0.86 (0.61–1.49) (0.68–1.55) (0.54–1.37) 63 (25.2) 91 (20.1) 69 (23.0) 95% CI Congestive heart failure, n (%) Adjusted HR 56 (33.7) 1.0 0.82 0.68 0.78 (0.56–1.19) (0.47–0.97) (0.52–1.16) 22 (13.3) 24 (9.6) 35 (7.7) 30 (10.0) 1.0 0.67 0.54 0.62 (0.33–1.16) 95% CI Stroke, n (%) Adjusted HR 95% CI Any nonfatal event, n (%) Adjusted HR 95% CI (0.36–1.23) (0.30–0.96) 78 (47.0) 103 (41.2) 177 (39.1) 113 (37.7) 1.0 0.93 0.89 0.82 (0.68–1.26) (0.67–1.19) (0.59–1.14) P trend 0.38 0.78 0.65 0.30 Adjusted HR indicated hazard ratio adjusted for age (in 5 categories), gender, smoking (in 3 categories), obesity, physical inactivity, alcohol consumption (in g day)1), filtered coffee consumption (in cups per day), educational attainment and sweet score, 95% CI indicates 95% confidence interval. biomarkers. As expected from the weak associations of chocolate with these biomarkers, including them in multivariable analyses attenuated the observed association of chocolate with cardiac mortality by no more than 3%. Stratified analyses In stratified analyses, we found largely similar associations between chocolate consumption and outcome across several subgroups. Chocolate consumption was associated with lower cardiac mortality both amongst men and women, in patients below or equal to 60 years and in those older than that, amongst physically active and inactive subjects, amongst never-, current- and former smokers, amongst hypertensive patients and patients without hypertension and amongst patients having high school ⁄ college degree and lower education. Chocolate consumption was also associated with lower cardiac mortality for patients with a BMI value below 30 kg m)2 (HR for the trend variable 0.92; 95% CI 0.82–0.98), but not for those above this BMI (HR 1.01; 95% CI 0.91–1.11), but the confidence intervals for Rothman’s synergy index for this interaction included one. Discussion In this long-term follow-up study of early survivors of a first AMI without diabetes, usual chocolate consumption, assessed at the time of hospitalization, had a strong inverse association with subsequent cardiac mortality. The corresponding associations with total mortality and nonfatal cardiovascular events were considerably weaker. The observed association of chocolate with lower cardiac mortality was generally stable across different subgroups and with adjustment for a variety of confounders. In contrast, intake of other sweets was not associated with cardiac or total mortality. We know of no studies assessing the possible effects of chocolate consumption on post-AMI prognosis. Some previous studies have suggested that chocolate may have protective effects against cardiovascular events and mortality in the general population. In the Zutphen Elderly Study, men free of coronary heart disease in the highest tertile of cocoa intake had half of the cardiovascular mortality of those in the lowest tertile after adjustment for the potential confounding factors [12]. In the Harvard Alumni Study, men who consumed candy had lower all-cause mortality than ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 253 I. Janszky et al. | Chocolate consumption and post-AMI prognosis Table 4 Potential mediatory factors according to usual chocolate consumption Never 239 n Less than once Up to once Twice or more P for per month per week per week homogeneity 307 497 318 LSM* (SE) LSM (SE) LSM (SE) LSM (SE) 133 (1.7) 131 (1.3) 132 (1.0) 130 (1.3) 0.55 Total cholesterol (mmol L)1) 6.22 (0.10) 6.26 (0.08) 6.21 (0.06) 6.15 (0.07) 0.82 LDL 4.21 (0.09) 4.33 (0.07) 4.26 (0.05) 4.19 (0.07) 0.46 HDL 1.09 (0.03) 1.08 (0.02) 1.10 (0.01) 1.08 (0.02) 0.82 ApoB 1.63 (0.03) 1.67 (0.03) 1.63 (0.02 1.64 (0.02) 0.71 ApoA 1.39 (0.02) 1.40 (0.02) 1.40 (0.01) 1.39 (0.02) 0.77 Lp(a) 0.28 (0.03) 0.31 (0.02) 0.30 (0.02) 0.29 (0.02) 0.87 Triglycerides (mmol L)1) 2.24 (0.12) 1.93 (0.09) 1.93 (0.07) 2.02 (0.09) 0.12 0.76 Systolic blood pressure (mmHg) Serum levels of Glucose 5.34 (0.10) 5.25 (0.08) 5.35 (0.06) 5.33 (0.07) Insulin 13.96 (0.95) 13.13 (0.73) 11.65 (0.54) 11.03 (0.69) 0.05 HOMA 3.20 (0.28) 2.99 (0.22) 2.84 (0.16) 2.74 (0.20) 0.59 IGFBP1 29.64 (2.00) 24.22 (1.54) 29.84 (1.15) 27.75 (1.46) 0.02 CRP 4.63 (0.72) 4.07 (0.56) 3.52 (0.42) 4.28 (0.54) 0.50 IL-6 6.70 (3.36) 9.25 (2.61) 8.01 (1.86) 8.90 (2.38) 0.93 TNFalfa 1.83 (0.19) 2.16 (0.15) 2.05 (0.12) 2.13 (0.14) 0.53 PAI-1 20.39 (1.54) 19.01 (1.20) 18.70 (0.89) 21.90 (1.14) 0.13 Fibrinogen 3.83 (0.08) 3.74 (0.06) 3.72 (0.05) 3.66 (0.06) 0.42 tPA-PAI 6.21 (0.38) 5.46 (0.31) 6.09 (0.23) 5.69 (0.29) 0.26 Von Willebrand 1.41 (0.07) 1.55 (0.06) 1.39 (0.04) 1.53 (0.05) 0.07 12.37 (1.09) 11.78 (0.85) 12.99 (0.65) 12.28 (0.80) 0.72 Homocystein *LSM, least squares means, adjusted for age, gender, obesity, physical inactivity, smoking, education, alcohol, filtered coffee and sweets score; ApoA, apolipoprotein A; ApoB, apolipoprotein B; Lp (a), lipoprotein (a); HOMA, homeostasis model assessment; IGFBP-1, insulin-like growth factor binding protein-1; CRP, high sensitivity C-reactive protein; IL-6, interleukin-6; PAI-1, plasminogen activator inhibitor 1; tPA, tissue plasminogen activator; TNF, tumour necrosis factor. nonconsumers [27], although chocolate and non chocolate candies were not differentiated. Mink et al. [13] also found an inverse association between chocolate intake and cardiovascular disease mortality amongst postmenopausal women. In contrast, in the Nurses’ Health Study, there was no significant association between moderate chocolate intake and the incidence of coronary heart disease in multivariate analyses [28]. Two components of chocolate might be particularly relevant to our findings. Fat in chocolate is especially high in stearic acid (around 30% of all fatty acids) [29]. The effect of stearic acid on cardiovascular health remains controversial, but a recent systematic review concluded that stearic acid has beneficial or neutral 254 effects on blood pressure and clotting parameters and does not adversely affect traditional lipid risk factors [30]. Chocolate is also a very rich source of flavonoid antioxidants [1]. Analysing major American brands, Gu et al. found that milk chocolate contains 5– 12 mg 100 g)1 of catechin, 18–24 lg 100 g)1 of epicatechin and 0.22–0.31 mg 100 g)1 of procyanidins in total. For dark chocolate, the corresponding concentrations were 11–33 mg 100 g)1, 52–125 lg 100 g)1 and 0.85–1.99 mg 100 g)1 respectively. In randomized trials, chocolate lowers blood pressure [11] and may increase bioactive nitric oxide availability [10]. Other potentially bioactive substances in chocolate include zinc, selenium, magnesium and vitamin E. ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 I. Janszky et al. | Chocolate consumption and post-AMI prognosis According to the Swedish National Food Administration [31], the concentrations of these substances in Swedish chocolates are 1.37 mg 100 g)1 in milk chocolate and 2.89 mg 100 g)1 in dark chocolate for zinc, 3.6 lg 100 g)1 and 4.6 lg 100 g)1 for selenium, 63 mg 100 g)1 and 174 mg 100 g)1 for magnesium and 0.74 mg 100 g)1 and 0.66 mg 100 g)1 for vitamin E respectively. We explored several potentially mediating biomarkers for the inverse association between chocolate intake and lower cardiac mortality. Surprisingly, none of the markers we tested, including lipids, inflammatory markers, coagulation factors, indicators of glucose homeostasis or systolic blood pressure, appeared to be important in linking chocolate to lower cardiac mortality. Based on the results of short-term intervention trials, improvement in endothelial function [2–5], inhibition of platelet activation [2, 6, 7] and LDL oxidation [8, 9] are other proposed mechanisms that we could not test in this study. Chocolate intake in our study had a considerably stronger and a more directly inverse association with fatal cardiac events than nonfatal ones. Although the HRs for recurrent AMI, stroke and heart failure were all numerically <1.0 in the top category of chocolate intake, none of these showed a significant or clear pattern of a dose-dependent trend. This finding mirrors that seen for n-3 fatty acid supplementation in secondary prevention trials [32] and could imply an anti-arrhythmic effect of chocolate intake in this setting. Limited animal and human data support the hypothesis that antioxidants have anti-arrhythmic properties in the setting of myocardial ischemia and ⁄ or reperfusion [33–37]. On the other hand, it is also likely that nonfatal events, especially stroke and heart failure were more misclassified as they were derived from hospital discharge registers [25]. Limitations Observational studies inherently limit causal inference. Chocolate intake might be associated with healthier lifestyle characteristics that could themselves explain the observed inverse association with cardiac mortality. Although we adjusted for several potential confounders in our multivariable analyses and found no attenuation of our results in multivariable models compared with age- and gender-adjusted models, we cannot exclude the possibility of uncontrolled confounding. However, any remaining confounder potentially able to influence our results considerably would need to be strongly associated with both chocolate intake and cardiovascular mortality following an AMI and generally unrelated to the factors included in our models. It is also possible that some patients ceased chocolate consumption prior to hospitalization because of poor health. Diabetes would appear to be the most likely medical reason that patients would be counselled to remove chocolate from their diets and hence, we excluded patients with diabetes from our analyses. Moreover, we found a weak and statistically nonsignificant association of non chocolate sweet consumption with cardiac mortality, despite the fact that those patients who stop chocolate consumption due to health problems would likely also reduce intake of other sweets. The beneficial effect of chocolate on cardiac mortality also remained after adjusting for intake of other sweets. We queried our patients about chocolate consumption in general and did not differentiate between dark and milk chocolate. In the European Union, milk chocolate has to contain a minimum of 25% of cocoa solids, dark chocolate 35% [38]. The corresponding proportions in United States are 10% and 15% [39]. According to the main chocolate producer (Marabou owned by Kraft Foods Sverige AB, Upplands Väsby, Sweden) in the decade of the 1990s, 90% of the consumption was milk chocolate in Sweden and Swedish milk chocolate normally contains 30% cocoa solids. Milk chocolate contains more fat and sugar than dark and is less rich in cocoa solids and hence flavonoid antioxidants [40]. The milk content may also decrease the absorption of cocoa-related antioxidants in the gastrointestinal tract [41], although this effect remains uncertain [42, 43] and Fraga et al. observed lower oxidative stress in a short-term trial for milk chocolate ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 255 I. Janszky et al. | [44]. Thus, our findings may or may not have been stronger had we queried our patients specifically about dark chocolate. Conclusion Chocolate consumption was associated with lower cardiac mortality in a dose dependent manner in patients free of diabetes surviving their first AMI. Although our findings support increasing evidence that chocolate is a rich source of beneficial bioactive compounds and suggest that individuals with coronary heart disease need not avoid chocolate if consumed in moderation, confirmation of this strong inverse relationship from other studies is needed. Conflict of interest statement No conflict of interest was declared. Acknowledgements The SHEEP study was supported by grants from the Swedish Council for Social Research, the Swedish Council for Work Life and the County Council of Stockholm. Dr Janszky was further supported by fellowship grant 2006-1146 of the Swedish Council of Working Life and Social Research and the Ansgarius Foundation. References 1 Waterhouse AL, Shirley JR, Donovan JL. Antioxidants in chocolate. Lancet 1996; 348: 834. 2 Flammer AJ, Hermann F, Sudano I et al. Dark chocolate improves coronary vasomotion and reduces platelet reactivity. Circulation 2007; 116: 2376–82. 3 Heiss C, Finis D, Kleinbongard P et al. Sustained increase in flow-mediated dilation after daily intake of high-flavanol cocoa drink over 1 week. J Cardiovasc Pharmacol 2007; 49: 74–80. 4 Hermann F, Spieker LE, Ruschitzka F et al. Dark chocolate improves endothelial and platelet function. Heart 2006; 92: 119–20. 5 Heiss C, Kleinbongard P, Dejam A et al. Acute consumption of flavanol-rich cocoa and the reversal of endothelial dysfunction in smokers. J Am Coll Cardiol 2005; 46: 1276–83. 6 Pearson DA, Paglieroni TG, Rein D et al. The effects of flavanol-rich cocoa and aspirin on ex vivo platelet function. Thromb Res 2002; 106: 191–7. 256 Chocolate consumption and post-AMI prognosis 7 Rein D, Paglieroni TG, Wun T et al. Cocoa inhibits platelet activation and function. Am J Clin Nutr 2000; 72: 30–5. 8 Osakabe N, Baba S, Yasuda A et al. Daily cocoa intake reduces the susceptibility of low-density lipoprotein to oxidation as demonstrated in healthy human volunteers. Free Radic Res 2001; 34: 93–9. 9 Baba S, Natsume M, Yasuda A et al. Plasma LDL and HDL cholesterol and oxidized LDL concentrations are altered in normo- and hypercholesterolemic humans after intake of different levels of cocoa powder. J Nutr 2007; 137: 1436–41. 10 Taubert D, Roesen R, Lehmann C, Jung N, Schomig E. Effects of low habitual cocoa intake on blood pressure and bioactive nitric oxide: a randomized controlled trial. JAMA 2007; 298: 49–60. 11 Taubert D, Roesen R, Schomig E. Effect of cocoa and tea intake on blood pressure: a meta-analysis. Arch Intern Med 2007; 167: 626–34. 12 Buijsse B, Feskens EJ, Kok FJ, Kromhout D. Cocoa intake, blood pressure, and cardiovascular mortality: the Zutphen Elderly Study. Arch Intern Med 2006; 166: 411–7. 13 Mink PJ, Scrafford CG, Barraj LM et al. Flavonoid intake and cardiovascular disease mortality: a prospective study in postmenopausal women. Am J Clin Nutr 2007; 85: 895–909. 14 Reuterwall C, Hallqvist J, Ahlbom A et al. Higher relative, but lower absolute risks of myocardial infarction in women than in men: analysis of some major risk factors in the SHEEP study. The SHEEP Study Group. J Intern Med 1999; 246: 161–74. 15 Linnersjo A, Hammar N, Gustavsson A, Reuterwall C. Recent time trends in acute myocardial infarction in Stockholm, Sweden. Int J Cardiol 2000; 76: 17–21. 16 Wiman B, Andersson T, Hallqvist J, Reuterwall C, Ahlbom A, deFaire U. Plasma levels of tissue plasminogen activator ⁄ plasminogen activator inhibitor-1 complex and von Willebrand factor are significant risk markers for recurrent myocardial infarction in the Stockholm Heart Epidemiology Program (SHEEP) study. Arterioscler Thromb Vasc Biol 2000; 20: 2019– 23. 17 Bennet AM, Prince JA, Fei GZ et al. Interleukin-6 serum levels and genotypes influence the risk for myocardial infarction. Atherosclerosis 2003; 171: 359–67. 18 Bennet AM, Brismar K, Hallqvist J, Reuterwall C, De Faire U. The risk of myocardial infarction is enhanced by a synergistic interaction between serum insulin and smoking. Eur J Endocrinol 2002; 147: 641–7. 19 Romelsjo A, Branting M, Hallqvist J et al. Abstention, alcohol use and risk of myocardial infarction in men and women taking account of social support and working conditions: the SHEEP case–control study. Addiction 2003; 98: 1453–62. 20 Janszky I, Ljung R, Ahnve S, Hallqvist J, Bennet AM, Mukamal KJ. Alcohol and long-term prognosis after a first acute myocardial infarction: the SHEEP study. Eur Heart J 2008; 29: 45–53. 21 Hammar N, Andersson T, Alfredsson L et al. Association of boiled and filtered coffee with incidence of first nonfatal myocardial infarction: the SHEEP and the VHEEP study. J Intern Med 2003; 253: 653–9. ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 I. Janszky et al. | Chocolate consumption and post-AMI prognosis 22 Hammar N, Nerbrand C, Ahlmark G et al. Identification of cases of myocardial infarction: hospital discharge data and mortality data compared to myocardial infarction community registers. Int J Epidemiol 1991; 20: 114–20. 23 Rosen M. National Health Data Registers: a Nordic heritage to public health. Scand J Public Health 2002; 30: 81–5. 24 Schaufelberger M, Swedberg K, Köster M, Rosen M, Rosengren A. Decreasing one-year mortality and hospitalization rates for heart failure in Sweden; Data from the Swedish Hospital Discharge Registry 1988 to 2000. Eur Heart J 2004; 25: 300– 7. 25 Ingelsson E, Arnlov J, Sundstrom J, Lind L. The validity of a diagnosis of heart failure in a hospital discharge register. Eur J Heart Fail 2005; 7: 787–91. 26 Lundberg M, Fredlund P, Hallqvist J, Diderichsen F. A SAS program calculating three measures of interaction with confidence intervals. Epidemiology 1996; 7: 655–6. 27 Lee IM, Paffenbarger RS Jr. Life is sweet: candy consumption and longevity. BMJ 1998; 317: 1683–4. 28 Hu FB, Stampfer MJ, Willett WC. Reply to PM Kris-Etherton et al. Am J Clin Nutr 2000; 72: 1059–60. 29 Connor WE. Harbingers of coronary heart disease: dietary saturated fatty acids and cholesterol. Is chocolate benign because of its stearic acid content? Am J Clin Nutr 1999; 70: 951–2. 30 Ding EL, Hutfless SM, Ding X, Girotra S. Chocolate and prevention of cardiovascular disease: a systematic review. Nutr Metab 2006; 3: 2. 31 Swedish National Food Administration. Available at: http://192. 121.81.11/livsmedelsok/sok.aspx?lang=2 (accessed on 15 January 2009). 32 GISSI-Prevenzione Investigators. Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione trial. Lancet 1999; 354: 447–55. 33 Hicks JJ, Montes-Cortes DH, Cruz-Dominguez MP, MedinaSantillan R, Olivares-Corichi IM. Antioxidants decrease reperfusion induced arrhythmias in myocardial infarction with ST-elevation. Front Biosci 2007; 12: 2029–37. 34 Assemand E, Lacroix M, Chahine R, Nadeau R, Mateescu MA. Gamma-irradiated ceruloplasmin affords antifibrillatory protec- 35 36 37 38 39 40 41 42 43 44 tion against ischemia ⁄ reperfusion damage in the isolated rat heart. Int J Radiat Biol 2007; 83: 309–17. Chen WP, Su MJ, Hung LM. In vitro electrophysiological mechanisms for antiarrhythmic efficacy of resveratrol, a red wine antioxidant. Eur J Pharmacol 2007; 554: 196–204. Maffei Facino R, Carini M, Aldini G et al. Procyanidines from Vitis vinifera seeds protect rabbit heart from ischemia ⁄ reperfusion injury: antioxidant intervention and ⁄ or iron and copper sequestering ability. Planta Med 1996; 62: 495–502. Atanasiu R, Dumoulin MJ, Chahine R, Mateescu MA, Nadeau R. Antiarrhythmic effects of ceruloplasmin during reperfusion in the ischemic isolated rat heart. Can J Physiol Pharmacol 1995; 73: 1253–61. McKevith B. New chocolate regulations. Nutr Bull 2003; 28: 197–8. Food and Drug Administration. Title 21 – Food and Drugs, Chapter I – Food and Drug Administration Department of Health and Human Services, Subchapter B – Food for Human Consumption, Part 163 – Cacao Products. Available at: http:// www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch. cfm?CFRPart=163 (accessed on 15 January 2009). Dreosti IE. Antioxidant polyphenols in tea, cocoa, and wine. Nutrition 2000; 16: 692–4. Serafini M, Bugianesi R, Maiani G, Valtuena S, De Santis S, Crozier A. Plasma antioxidants from chocolate. Nature 2003; 424: 1013. Cooper KA, Donovan JL, Waterhouse AL, Williamson G. Cocoa and health: a decade of research. Br J Nutr 2008; 99: 1– 11. Schroeter H, Holt RR, Orozco TJ, Schmitz HH, Keen CL. Nutrition: milk and absorption of dietary flavanols. Nature 2003; 426: 787–8. discussion 788. Fraga CG, Actis-Goretta L, Ottaviani JI et al. Regular consumption of a flavanol-rich chocolate can improve oxidant stress in young soccer players. Clin Dev Immunol 2005; 12: 11–7. Correspondence: Dr Imre Janszky, Department of Public Health Sciences, Karolinska Institute, Norrbacka, 6th floor, Karolinska University Hospital, SE-171 76 Stockholm, Sweden. (fax: + 46 8 73 73 888; e-mail: [email protected]). ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 266; 248–257 257