- Email: [email protected]
The association of Mediterranean diet with lower risk of acute coronary syndromes in hypertensive subjects
International Journal of Cardiology 82 (2002) 141–147 www.elsevier.com / locate / ijcard
The association of Mediterranean diet with lower risk of acute coronary syndromes in hypertensive subjects Demosthenes B. Panagiotakos*, Christine Chrysohoou, Christos Pitsavos, Kostas Tzioumis, Ioanna Papaioannou, Christodoulos Stefanadis, Pavlos Toutouzas Cardiology Department, School of Medicine, University of Athens, Athens, Greece Accepted 18 November 2001
Abstract Background: The elevation of blood pressure levels has been recognised as a determinant of the risk for several common cardiovascular diseases. In this work we examined the effect of Mediterranean type of diet on coronary risk in subjects with hypertension. Methods: CARDIO2000 consisted of 848 randomly selected hospitalised patients (695 males, 58610 years old; 153 females, 6569 years old) for first event of coronary heart disease (CHD) and 1078 paired, by sex–age, hospitalised controls without CHD. The adoption of the Mediterranean diet was assessed through a validated questionnaire developed by the National School of Public Health. Results: 418 (49%) of the patients and 303 (28%) of the controls were hypertensive. Of them 21 (5%) patients and 36 (12%) controls were unaware of their condition, 94 (22%) and 34 (11%) were untreated, 148 (35%) and 111 (36%) were uncontrolled and 155 (38%) and 122 (41%) were controlled (P,0.01). One hundred and sixty-two (19%) of the patients and 265 (25%) of the controls (P,0.01) adopted the Mediterranean type of diet. Our results suggest that the adoption of Mediterranean diet reduces the risk of developing acute coronary syndromes by 17% (odds ratio50.83, 95% CI 0.73–0.88, P,0.01) in controlled hypertensive subjects, by 8% (odds ratio50.92, 95% CI 0.87–0.95, P,0.05) in unaware, by 7% (odds ratio50.93, 95% CI 0.88–0.95, P,0.05) in acknowledged but uncontrolled and by 20% (odds ratio50.80, 95% CI 0.71–0.89, P,0.01) in normotensive subjects. Conclusion: According to our findings the adoption of the Mediterranean diet is associated with the reduction of coronary risk in hypertensive subjects. 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Hypertension; Risk; Coronary; Mediterranean; Diet
1. Introduction The level of blood pressure has been recognised as a determinant of the risk for several common cardiovascular diseases, including coronary heart disease, cerebrovascular disease and heart failure [1,2]. Evidence based on the literature supports the hypothesis that dietary factors as well as other lifestyle *Corresponding author. Present address: 48–50 Chiou Str. Glyfada, Attica, 165 61 Hellas. Tel.: 130-1-960-3116; fax: 130-1-960-0719. E-mail address: [email protected] (D.B. Panagiotakos).
habits may influence blood pressure levels. Many investigators, based on the results from the Seven Countries study [3,4], have recognized the beneficial role of the Mediterranean type of diet in cardiovascular diseases, but the pathway remains unclear. In trials of vegetarian diets replacing animal with vegetable products, a reduction of blood pressure levels was revealed in normotensive as well as in hypertensive subjects [5–7]. Aspects of vegetarian diets, which are believed to reduce blood pressure, include high levels of fiber and minerals (such as potassium and magnesium) and reduced fat content. In observa-
0167-5273 / 02 / $ – see front matter 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 01 )00611-8
142
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
tional studies significant inverse associations have been reported between blood pressure levels and intake of magnesium, potassium, calcium, fiber and protein. However, in trials that tested these nutrients often as dietary supplements, the reduction in blood pressure was found small and inconsistent [6]. In this work we aimed to evaluate the effect of Mediterranean-type of diet on the risk of developing acute coronary syndromes in hypertensive subjects. A stratified, random sampling from all Greek regions ensured representation of various cultural, socioeconomical and behavioural characteristics that aggregate in the investigated population and may influence the results.
2. Methods The CARDIO2000 is a multicentre retrospective case-control study that investigates the association of several demographic, nutritional, lifestyle and medical risk factors with the risk of developing acute coronary syndromes. From January 2000 to August 2001, 848 randomly selected cardiac patients and 1078 cardiovascular disease-free subjects entered into the study. The number of the subjects was decided through power analysis, in order to evaluate differences in the coronary risk greater than 7% (statistical power .0.80, significant level ,0.05). According to the population distribution provided by the National Statistical Services (Ministry of Economics, census 2000), we stratified our sampling into all the Greek regions, in order to include various socio-economical levels and cultural particularities of the investigated population. Firstly, we randomly selected the coronary patients from the admission listing of the cardiological clinics. The inclusion criteria for cardiac patients are as follows. 1. First event of acute myocardial infarction diagnosed by two or more of the following features: typical electrocardiographic changes, compatible clinical symptoms, specific diagnostic enzyme elevations [8], or 2. First diagnosed unstable angina corresponding to class III of the Braunwald classification [8]. For each selected cardiac patient we randomly
selected a subject without any clinical symptoms or suspicion of cardiovascular disease in his medical history (control), matched to the patient by age (63 years), sex, and region. Controls were, mainly, subjects that visited the out-patients department of the same hospital and at the same period with the coronary patients, for routine examinations or minor surgical operations. In a few cases (in country hospitals), in which the available number of outpatient controls was not sufficient for the matching procedure, we enrolled into the study friends or colleagues of the coronary patients. In order to eliminate recall bias we tried to retrieve precise information through medical records and a specific confidential feedback form including structured questions concerning living habits and sociodemographic background factors. The medical information was retrieved from subjects’ medical history through hospital or insurance records. The interview took place during the 2nd to 4th day of hospitalisation. At least 16 countries that border the Mediterranean Sea can be defined as Mediterranean countries. Between these populations many cultural, ethnic, religious, economic and agricultural differences may exist that result in different dietary habits and that precludes a single definition of a Mediterranean-type of diet. However, this traditional type of diet is not systematically defined and according to the majority of the investigators is characterized by a pattern that is high in fruits, vegetables, bread, other cereals, potatoes, poultry, beans, nuts, fish, little red meat, dairy products, moderate alcohol consumption and olive oil as an important fat source. The previous dietary pattern is based on food patterns typical of many regions in Greece and southern Italy [9,10]. In our study, long-term nutritional habits were assessed through a special, validated questionnaire that is regularly used in nutritional surveys, designed by the Department of Nutrition of the National School of Public Health [11]. For each of the investigated food items, the frequency of consumption was quantified approximately in terms of the number of times the food was consumed, during a month. Thus, weekly consumption multiplied by 4 and a value of 0 was assigned to food items rarely or never consumed. In order to perform an objective assessment of the nutritional characteristics, food frequencies were
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
related into food quantities in grams per day on the basis of standard portion size estimations. Total diet was described by the use of composite scores [12,13]. In particular, the Mediterranean diet was scored in terms of its component characteristics, as mentioned above. Subjects who adopt this special type of diet were categorized using as cut-off points the median values of the monthly food consumption score, as done by several investigators in the past [11–13]. Although the CARDIO2000 data have been selected from an epidemiologic retrospective study, special attention has been given to the clinical implications of the blood-pressure levels observed in persons with poorly controlled hypertension or unaware of their condition. Thus, we analysed the collected information to compare the actual blood-pressure levels in persons who were unaware that they had hypertension, those who were aware of their condition but were not under any treatment, those who were under treatment but remained uncontrolled hypertensives, and those in whom hypertension was controlled by treatment. In keeping with the long-standing classification criteria used in several population-based studies [14,15], subjects who reported never having been told of having hypertension and were not currently taking any antihypertensive medication and whose blood pressure was less than 140 / 90 mmHg were assigned to the normotensive group. Subjects whose blood pressure were greater than 140 / 90 mmHg or were under antihypertensive medication were assigned to the group of hypertension. Subjects with the same criteria for blood pressure as before but answered ‘no’ to the question ‘Have you ever been told you have hypertension?’ were acknowledged as unaware of hypertension, while subjects who had answered ‘yes ’ to the previous question but stated that they were not currently taking antihypertensive medication, were defined as untreated hypertension. Subjects who answered ‘yes’ to the question regarding awareness and treatment but their blood pressure levels were greater than 140 / 90 mmHg were defined as uncontrolled hypertension, and, finally, the rest of them were assigned into the group of controlled hypertension. Current smokers were defined as those who smoked at least 1 cigarette per day. Quantification of smoking status was based on the calculation of packyears adjusted for nicotine containment equal to 0.8-
143
mg per cigarette. Former smokers were defined as the subjects who stopped smoking for over 1 year. Physically active were those who reported engaging in non-occupational, physical activity during the past year more than one time per week. All the others were considered as physically inactive. The potential confounding effect of occupational physical activity was taken into account in the analysis as a dummy variable. Hypercholesterolemia was defined as cholesterol levels greater than 220 mg / dl or greater than 200 mg / dl when two other risk factors for coronary heart disease were present or use of special treatment. Obesity was defined as body mass index greater than 29.9 kg / m 2 and diabetic subjects were those with fast blood glucose .125 mg / dl. According to their medical records, the majority of the controls (86%) and the patients (83%) had laboratory measurements (at least one) during the past 12 months. We also took total cholesterol and blood glucose measurements during the first 12 h of hospitalisation. The mean values of the previous laboratory measurements were assisted in order to classify our subjects into hypertensive, hypercholesterolemic or diabetics. Alcohol consumption was measured by daily ethanol intake, in wine glasses (100 cc–12% ethanol). Further details of the aims and the methodology of CARDIO2000 project have been described in the literature [10,16].
2.1. Statistical analysis Continuous variables are presented as mean6one standard deviation as well as 95% confidence intervals (CI) while qualitative variables are presented as absolute and relative frequencies. In order to fit multivariate risk models an exploratory analysis was initially applied. Pearson’s correlation coefficient was used in order to measure associations between the continuous variables, while contingency tables with calculation of chi-squared test, as well as application of Student’s t-test evaluated associations between the categorical and continuous variables. Estimations of the relative risks of developing acute coronary syndromes, under several hypotheses, were performed by the calculation of odds ratio (OR) and the corresponding confidence intervals through multiple logistic regression analysis [17]. All reported P-values are based on two-sided tests and compared to a
144
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
significant level of 5%. STATA 6 software was used for the calculations (STATA Corp. College Station, TX, USA).
3. Results Six hundred and fifty-eight (77%) out of the 848 patients were males (59.0610 years old) and 190 (23%) were females (65.369 years old). Also, 830 out of the 1078 controls were males (58.0610 years old) and 248 were females (64.869 years old), of similar age with the patients. Five hundred and twenty-six (62%) coronary patients and 765 (71%) controls adopted the Mediterranean-type of diet. Mean age of those who defined ‘closer’ to the Mediterranean-type of diet was significantly higher than the others (62.569 vs. 56.1610 years old, P50.023). The prevalence of current smoking, hypertension, hypercholesterolemia, diabetes mellitus and obesity in the subgroups of subjects that adopted the Mediterranean diet was significantly lower compared to the others (Table 1). No associations were found between the adoption of Mediterranean-type of diet, the income and the educational level of the subjects ( x 2 520.12, P5 0.455 and x 2 521.33, respectively, P50.559). Also, no associations were found between leisure time physical activity, occupational class (unskilled, semiskilled and skilled) and the adoption of Mediterranean diet ( x 2 512.32, P50.514 and x 2 511.41, respectively, P50.423). Additionally, no interaction was assessed between occupational type physical activity and adoption of Mediterranean type of diet. All the patients and the controls were classified into the aforementioned hypertension groups. As we can see in Table 2, 418 (49%) of the
coronary patients and 303 (28%) of the controls were hypertensive. Table 2 also presents the distribution of several demographic and cardiovascular risk factors according to the group of hypertension. The analysis of the results presented in Table 2, raised a significant association between groups of hypertension and smoking status, in coronary patients ( x 2 538.11, P,0.001) but no association was observed in control groups ( x 2 51.10, P50.337). Additionally, the prevalence of hypercholesterolemia, obesity and diabetes mellitus, both in patients and controls, was significantly related with the hypertension group (P,0.05). Finally, significant relationships were observed between groups of hypertension and educational, financial status of the patients and controls, respectively ( x 2 521.11 and x 2 531.19, P, 0.05). After taking into account the effect of the previous associations and the effect of several potential confounders such as age and sex, by design, presence of hypercholesterolemia, diabetes mellitus, physical inactivity, smoking habit, premature history of coronary heart disease, dose and type of treatment, we performed conditional risk analysis. Our results regarding the effect of the adoption of Mediterranean type of diet on the risk of developing acute coronary syndromes in hypertensive as well in normotensive subjects are presented in Table 3. Further analysis showed that, male sex, low income and educational level, and not having seen a physician in the past year increased the risk of being classified as having hypertension but being unaware of the condition, by 51%, 59%, and 44%, respectively (P,0.01). Also, an age of at least 50 years for males and 55 years for females were found strong risk factors for the lack of awareness of hypertension. In the model that assessed the risk of having acknowl-
Table 1 Prevalence of the conventional cardiovascular risk factors in coronary patients and controls that adopted the Mediterranean diet Mediterranean diet
Current smoking Hypercholesterolemia Hypertension Obesity Diabetes mellitus Physical inactivity (leisure time)
Other type of diet
P-value
Patients n5162
Controls n5265
Patients n5686
Controls n5813
95 (59%) 89 (55%) 71 (44%) 62 (38%) 39 (24%) 129 (80%)
90 (34%) 72 (27%) 64 (24%) 64 (24%) 11 (4%) 164 (62%)
439 418 347 281 199 521
235 (28%) 285 (35%) 239 (29%) 227 (28%) 73 (9%) 479 (59%)
(64%) (61%) (50%) (41%) (29%) (76%)
,0.01 ,0.01 ,0.01 ,0.05 ,0.01 NS
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
145
Table 2 Distribution of cardiovascular risk factors and other socio-demographic variables, by hypertension group in patients (controls) group Normotensive
ACS patients 430 (51%) Controls 775 (72%) Prevalence of cardiovascular risk factors in patients (controls) Smoking 52% (48%) Hypercholesterolemia 58% (23%) Diabetes mellitus 19% (5%) Obesity (BMI.29.9 kg / m 2 ) 21% (16%) Physical inactivity Distribution of socio-demographic factors in patients (controls) Education None–primary school 28% (23%) Senior high or colleges 52% (54%) University 20% (23%) Annual income Low (,4750 US$) 7% (4%) Moderate (4750–8500 US$) 52% (48%) Good (8500–14,500 US$) 37% (39%) Very good (.14,500 US$) 4% (9%)
Uncontrolled hypertension Acknowledged untreated
Treated uncontrolled
21 (2%) 36 (3%)
94 (11%) 34 (3%)
148 (17%) 111 (10%)
155 (19%) 122 (12%)
48% (30%) 47% (28%) 29% (11%) 17% (18%)
69% 65% 32% 28%
45% 62% 33% 18%
(39%) (29%) (7%) (12%)
39% (37%) 43% (22%) 34% (7%) 18% (10%)
39% (33%) 49% (56%) 12% (11%)
37% (31%) 49% (55%) 14% (14%)
39% (30%) 44% (42%) 17% (28%)
32% (33%) 52% (42%) 16% (25%)
9% (3%) 52% (49%) 32% (33%) 7% (15%)
9% (4%) 60% (54%) 27% (33%) 4% (9%)
6% (3%) 60% (54%) 22% (29%) 12% (14%)
8% (5%) 52% (54%) 28% (30%) 12% (11%)
edged but uncontrolled hypertension, male sex, not having seen a physician in the preceding year, low educational and financial levels as well as an age of at least 45 years were significant risk factors for this condition.
4. Discussion In this work we present the findings from the CARDIO2000 study regarding the effect of the adoption of Mediterranean diet on the risk of developing acute coronary syndromes, in hypertensive subjects. The performed risk analysis showed that the adoption of Mediterranean diet is associated with the Table 3 Results from the multiple regression analysis for the effect of the adoption of Mediterranean diet on the risk of developing ACS, by hypertensive and normotensive subjects
Controlled hypertensives Unaware of their condition Acknowledged uncontrolled Normotensive
Treated, controlled
Unaware
Odds ratio
95% CI
P-value
0.83 0.92 0.93 0.80
0.78–0.88 0.87–0.95 0.88–0.95 0.71–0.89
0.003 0.038 0.031 0.001
The comparisons were made between subjects who adopted the Mediterranean diet versus the rest of them.
(41%) (40%) (8%) (21%)
reduction of the risk of developing acute coronary syndromes not only in normotensive subjects, but also in hypertensive as well as in subjects who are unaware of their condition or uncontrolled. There is extensive scientific evidence on the relation between diet and incidence of coronary heart disease, various types of cancer and other diseases [1,2,5,7,13]. These studies showed that dietary factors exert their influence largely through their effects on blood lipids and lipoproteins, as well as on the other established modifiable risk factors, with the exception of cigarette smoking. Our results support the hypothesis that the adoption of the Greek type of Mediterranean diet can significantly reduce the risk of developing acute coronary syndromes under the presence of hypertension, after controlling for several potential confounders. The investigated type of diet is low in saturated fat, high in monounsaturated fat, mainly from olive oil, high in complex carbohydrates, from legumes, and high in fibre, mostly from vegetables and fruits [9,12,18]. Based on the results from the Seven Countries Study [19], the protective role of Mediterranean type of diet against atherosclerosis was explained, mainly, due to the reduction of cholesterol and blood pressure levels. Recently, the findings from the Lyon Diet Heart study [9] illustrate the potential importance of the Mediterranean dietary
146
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
pattern, especially when compared to other recommended diets, like Step-I diet [20]. However it still remains a matter of debate if the protective influence is primarily caused by single nutrients, e.g. dietary fatty acids, potassium or dietary fibre or if it can be attributed to the Mediterranean diet as a whole [9]. Also, many investigators believe that the effect of the Mediterranean diet, on coronary risk, may be influenced through a number of confounders such as geographic and other non-measured cultural and social differences of the investigated populations [21]. In this study we aimed to explore the effect of Mediterranean diet on the risk of developing acute coronary syndromes. Based on a database of |2000 coronary patients and controls we found that the adoption of Mediterranean diet is associated with a substantial reduction of the coronary risk in hypertensive subjects, even if they were unaware of their condition or if it was uncontrolled. In addition, our risk analysis showed that the one fifth of the non-fatal acute coronary syndromes (attributable risk), due to hypertension, could be prevented in controlled hypertensive subjects through the adoption of this traditional diet. Additionally, 9% of the non-fatal cardiac events in unaware of their condition or uncontrolled subjects could be prevented through the adoption of the Greek type of Mediterranean diet.
5. Limitations of the study In this retrospective case-control study two, main, sources of systematic errors may exist, the selection and the recall bias. In order to eliminate selection bias we tried to set objective criteria, both for patients and controls. Concerning information bias we tried to avoided it through accurate and detailed data from patients’ medical records. However, recall bias may still exist, especially in the measurement of nutritional habits, smoking, and the onset of the investigated cardiovascular risk factors. Furthermore, regarding the potential effect of uncontrolled-unknown confounders, we tried to reduce it through multivariate analysis and using the same study base, both for patients and controls.
Acknowledgements The Hellenic Heart Foundation funds the CARDIO2000 study. This study was supported by research grants from the Hellenic Heart Foundation (11 / 1999–2001). The authors would like to thank the physicians and the specialists that coordinated this study: Dr. K. Tzioumis (Athens, Crete, Pelloponisos), Dr. ´I. Papaioannou (Athens, Thessalia), Dr. P. Stravopodis (Ionian Islands), Dr. L. Karra (Aegean Islands), Dr. D. Antoniades (Macedonia), Dr. G. Rembelos (Aegean Islands), Dr. D. Markou (Athens), A. Moraiti (Athens), D. Evagelou (Crete), Dr. S. Vellas (Attica, Hpeirous), B. Meidanis (Macedonia, Sterea Hellas, Thessalia), Dr. S. Loggos (Attica), Dr. I. Elefsiniotis (Athens), Dr. I. Vogiatzis (Central Macedonia), Dr. N. Marinakis (Aegean Islands), Dr. G. Koutsimbanis (Thrace), and Dr. T. Kyratzoglou (East Macedonia).
References [1] The Joint National Committee on Detection Evaluation and Treatment of High Blood Pressure. The fifth report of the Joint National Committee on Detection Evaluation and Treatment of High Blood Pressure (JNC V). Arch Intern Med 1993;53:54–83. [2] Working Group on Primary Prevention of Hypertension. National High Blood Pressure Education Program Working Group report on primary prevention of hypertension. Arch Intern Med 1993;53:86– 208. [3] Menotti A, Lanti M, Puddu PE, Kromhout D. Coronary heart disease incidence in northern and southern European populations: a reanalysis of the seven countries study for a European coronary risk chart. Heart 2000;84(3):238–44. [4] Kafatos A, Diacatou A, Voukiklaris G, Nikolakakis N, Vlachonikolis J, Kounali D, Mamalakis G, Dontas AS. Heart disease risk-factor status and dietary changes in the Cretan population over the past 30 years: the Seven Countries Study. Am J Clin Nutr 1997;65(6):1882– 6. [5] Burt VL, Whelton P, Roccella EJ et al. Prevalence of hypertension in the US adult population: results from the Third National Health and Nutrition Examination Survey 988-99. Hypertension 1995;25:305– 13. [6] Ascherio A, Rimm EB, Giovannucci EL et al. A prospective study of nutritional factors and hypertension among US men. Circulation 1992;86:475–84. [7] Espiro-Montoro A, Lopez-Miranda J, Castro P et al. Monounsaturated fatty acid enriched diets lower plasma insulin levels and blood pressure in healthy young men. Nutr Metab Cardiovasc Dis 1996;6:147–54. [8] Braunwald E. In: 5th ed, Heart disease, London: WB Saunders, 1997, pp. 1184–202.
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147 [9] de Lorgeril M, Salen P, Martin J-L, Monjaud I, Delaye J, Mamelle N. Mediterranean diet, traditional risk factors and the rate of cardiovascular complications after myocardial infarction. Final report of the Lyon Diet Heart Study. Circulation 1999;99:779–85. [10] Panagiotakos DB, Pitsavos C, Chrysohoou C, Stefanadis C, Toutouzas PK. The role of traditional Mediterranean-type of diet and lifestyle, in the development of acute coronary syndromes: preliminary results from cardio2000 study. C Eur J Pub Health 2002; in press. [11] Trichopoulou A. In: Composition of Greek foods and dishes, Athens, Greece: Athens School of Public Health Publications, 1992, p. 152, in Greek and English. [12] Trichopoulou A, Kouris-Blazos A, Wahlqvist M et al. Diet and overall survival in elderly people. Br Med J 1995;311:1457–60. [13] Assmann G, deBasker G, Bagnara S et al. International Consensus statement on olive oil and the Mediterranean diet: implications for health in Europe. Eur J Cancer Prev 1997;6:418–21. [14] Hyman DJ, Pavlik VN. Characteristics of patients with uncontrolled hypertension in the United States. N Engl J Med 2001;345:479–86. [15] Burt VL, Cutler A, Higgins M et al. Trends in the prevalence, awareness, treatment, and control of hypertension in the adult US population: data from the health examination surveys, 1960 to 1991. Hypertension 1995;26:60–9.
147
[16] Panagiotakos DB, Pitsavos C, Chrysohoou C, Stefanadis C, Toutouzas PK. Risk stratification of coronary heart disease through established and emerging lifestyle factors, in a Mediterranean population: CARDIO2000 epidemiological study. J Cardiovasc Risk 2001;8(6):329–39. [17] Armitage P, Berry U. In: Statistical methods in medical research, Oxford: Blackwell Scientific, 1994, p. 620. [18] Strazullo P, Ferro-Luzzi A, Saini A et al. Changing the Mediterranean diet: effects on blood pressure. J Hypertens 1986;4:407–12. [19] Keys A, Menotti A, Karvonen MJ et al. The diet and 15-year death rate in the Seven Countries Study. Am J Epidemiol 1986;124:903– 15. [20] Kris-Etherton P, Eckel R, Howard B, Jeor S, Bazzare T. Lyon Diet Heart Study. Benefits of a Mediterranean-style, National Education Program /AHA Step I dietary pattern on cardiovascular disease. Circulation 2001;103:1823–5. [21] Robertson RM, Smaha L. Can a Mediterranean style diet reduce heart disease? Circulation 2001;103:1821–2.
The association of Mediterranean diet with lower risk of acute coronary syndromes in hypertensive subjects Demosthenes B. Panagiotakos*, Christine Chrysohoou, Christos Pitsavos, Kostas Tzioumis, Ioanna Papaioannou, Christodoulos Stefanadis, Pavlos Toutouzas Cardiology Department, School of Medicine, University of Athens, Athens, Greece Accepted 18 November 2001
Abstract Background: The elevation of blood pressure levels has been recognised as a determinant of the risk for several common cardiovascular diseases. In this work we examined the effect of Mediterranean type of diet on coronary risk in subjects with hypertension. Methods: CARDIO2000 consisted of 848 randomly selected hospitalised patients (695 males, 58610 years old; 153 females, 6569 years old) for first event of coronary heart disease (CHD) and 1078 paired, by sex–age, hospitalised controls without CHD. The adoption of the Mediterranean diet was assessed through a validated questionnaire developed by the National School of Public Health. Results: 418 (49%) of the patients and 303 (28%) of the controls were hypertensive. Of them 21 (5%) patients and 36 (12%) controls were unaware of their condition, 94 (22%) and 34 (11%) were untreated, 148 (35%) and 111 (36%) were uncontrolled and 155 (38%) and 122 (41%) were controlled (P,0.01). One hundred and sixty-two (19%) of the patients and 265 (25%) of the controls (P,0.01) adopted the Mediterranean type of diet. Our results suggest that the adoption of Mediterranean diet reduces the risk of developing acute coronary syndromes by 17% (odds ratio50.83, 95% CI 0.73–0.88, P,0.01) in controlled hypertensive subjects, by 8% (odds ratio50.92, 95% CI 0.87–0.95, P,0.05) in unaware, by 7% (odds ratio50.93, 95% CI 0.88–0.95, P,0.05) in acknowledged but uncontrolled and by 20% (odds ratio50.80, 95% CI 0.71–0.89, P,0.01) in normotensive subjects. Conclusion: According to our findings the adoption of the Mediterranean diet is associated with the reduction of coronary risk in hypertensive subjects. 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Hypertension; Risk; Coronary; Mediterranean; Diet
1. Introduction The level of blood pressure has been recognised as a determinant of the risk for several common cardiovascular diseases, including coronary heart disease, cerebrovascular disease and heart failure [1,2]. Evidence based on the literature supports the hypothesis that dietary factors as well as other lifestyle *Corresponding author. Present address: 48–50 Chiou Str. Glyfada, Attica, 165 61 Hellas. Tel.: 130-1-960-3116; fax: 130-1-960-0719. E-mail address: [email protected] (D.B. Panagiotakos).
habits may influence blood pressure levels. Many investigators, based on the results from the Seven Countries study [3,4], have recognized the beneficial role of the Mediterranean type of diet in cardiovascular diseases, but the pathway remains unclear. In trials of vegetarian diets replacing animal with vegetable products, a reduction of blood pressure levels was revealed in normotensive as well as in hypertensive subjects [5–7]. Aspects of vegetarian diets, which are believed to reduce blood pressure, include high levels of fiber and minerals (such as potassium and magnesium) and reduced fat content. In observa-
0167-5273 / 02 / $ – see front matter 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 01 )00611-8
142
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
tional studies significant inverse associations have been reported between blood pressure levels and intake of magnesium, potassium, calcium, fiber and protein. However, in trials that tested these nutrients often as dietary supplements, the reduction in blood pressure was found small and inconsistent [6]. In this work we aimed to evaluate the effect of Mediterranean-type of diet on the risk of developing acute coronary syndromes in hypertensive subjects. A stratified, random sampling from all Greek regions ensured representation of various cultural, socioeconomical and behavioural characteristics that aggregate in the investigated population and may influence the results.
2. Methods The CARDIO2000 is a multicentre retrospective case-control study that investigates the association of several demographic, nutritional, lifestyle and medical risk factors with the risk of developing acute coronary syndromes. From January 2000 to August 2001, 848 randomly selected cardiac patients and 1078 cardiovascular disease-free subjects entered into the study. The number of the subjects was decided through power analysis, in order to evaluate differences in the coronary risk greater than 7% (statistical power .0.80, significant level ,0.05). According to the population distribution provided by the National Statistical Services (Ministry of Economics, census 2000), we stratified our sampling into all the Greek regions, in order to include various socio-economical levels and cultural particularities of the investigated population. Firstly, we randomly selected the coronary patients from the admission listing of the cardiological clinics. The inclusion criteria for cardiac patients are as follows. 1. First event of acute myocardial infarction diagnosed by two or more of the following features: typical electrocardiographic changes, compatible clinical symptoms, specific diagnostic enzyme elevations [8], or 2. First diagnosed unstable angina corresponding to class III of the Braunwald classification [8]. For each selected cardiac patient we randomly
selected a subject without any clinical symptoms or suspicion of cardiovascular disease in his medical history (control), matched to the patient by age (63 years), sex, and region. Controls were, mainly, subjects that visited the out-patients department of the same hospital and at the same period with the coronary patients, for routine examinations or minor surgical operations. In a few cases (in country hospitals), in which the available number of outpatient controls was not sufficient for the matching procedure, we enrolled into the study friends or colleagues of the coronary patients. In order to eliminate recall bias we tried to retrieve precise information through medical records and a specific confidential feedback form including structured questions concerning living habits and sociodemographic background factors. The medical information was retrieved from subjects’ medical history through hospital or insurance records. The interview took place during the 2nd to 4th day of hospitalisation. At least 16 countries that border the Mediterranean Sea can be defined as Mediterranean countries. Between these populations many cultural, ethnic, religious, economic and agricultural differences may exist that result in different dietary habits and that precludes a single definition of a Mediterranean-type of diet. However, this traditional type of diet is not systematically defined and according to the majority of the investigators is characterized by a pattern that is high in fruits, vegetables, bread, other cereals, potatoes, poultry, beans, nuts, fish, little red meat, dairy products, moderate alcohol consumption and olive oil as an important fat source. The previous dietary pattern is based on food patterns typical of many regions in Greece and southern Italy [9,10]. In our study, long-term nutritional habits were assessed through a special, validated questionnaire that is regularly used in nutritional surveys, designed by the Department of Nutrition of the National School of Public Health [11]. For each of the investigated food items, the frequency of consumption was quantified approximately in terms of the number of times the food was consumed, during a month. Thus, weekly consumption multiplied by 4 and a value of 0 was assigned to food items rarely or never consumed. In order to perform an objective assessment of the nutritional characteristics, food frequencies were
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
related into food quantities in grams per day on the basis of standard portion size estimations. Total diet was described by the use of composite scores [12,13]. In particular, the Mediterranean diet was scored in terms of its component characteristics, as mentioned above. Subjects who adopt this special type of diet were categorized using as cut-off points the median values of the monthly food consumption score, as done by several investigators in the past [11–13]. Although the CARDIO2000 data have been selected from an epidemiologic retrospective study, special attention has been given to the clinical implications of the blood-pressure levels observed in persons with poorly controlled hypertension or unaware of their condition. Thus, we analysed the collected information to compare the actual blood-pressure levels in persons who were unaware that they had hypertension, those who were aware of their condition but were not under any treatment, those who were under treatment but remained uncontrolled hypertensives, and those in whom hypertension was controlled by treatment. In keeping with the long-standing classification criteria used in several population-based studies [14,15], subjects who reported never having been told of having hypertension and were not currently taking any antihypertensive medication and whose blood pressure was less than 140 / 90 mmHg were assigned to the normotensive group. Subjects whose blood pressure were greater than 140 / 90 mmHg or were under antihypertensive medication were assigned to the group of hypertension. Subjects with the same criteria for blood pressure as before but answered ‘no’ to the question ‘Have you ever been told you have hypertension?’ were acknowledged as unaware of hypertension, while subjects who had answered ‘yes ’ to the previous question but stated that they were not currently taking antihypertensive medication, were defined as untreated hypertension. Subjects who answered ‘yes’ to the question regarding awareness and treatment but their blood pressure levels were greater than 140 / 90 mmHg were defined as uncontrolled hypertension, and, finally, the rest of them were assigned into the group of controlled hypertension. Current smokers were defined as those who smoked at least 1 cigarette per day. Quantification of smoking status was based on the calculation of packyears adjusted for nicotine containment equal to 0.8-
143
mg per cigarette. Former smokers were defined as the subjects who stopped smoking for over 1 year. Physically active were those who reported engaging in non-occupational, physical activity during the past year more than one time per week. All the others were considered as physically inactive. The potential confounding effect of occupational physical activity was taken into account in the analysis as a dummy variable. Hypercholesterolemia was defined as cholesterol levels greater than 220 mg / dl or greater than 200 mg / dl when two other risk factors for coronary heart disease were present or use of special treatment. Obesity was defined as body mass index greater than 29.9 kg / m 2 and diabetic subjects were those with fast blood glucose .125 mg / dl. According to their medical records, the majority of the controls (86%) and the patients (83%) had laboratory measurements (at least one) during the past 12 months. We also took total cholesterol and blood glucose measurements during the first 12 h of hospitalisation. The mean values of the previous laboratory measurements were assisted in order to classify our subjects into hypertensive, hypercholesterolemic or diabetics. Alcohol consumption was measured by daily ethanol intake, in wine glasses (100 cc–12% ethanol). Further details of the aims and the methodology of CARDIO2000 project have been described in the literature [10,16].
2.1. Statistical analysis Continuous variables are presented as mean6one standard deviation as well as 95% confidence intervals (CI) while qualitative variables are presented as absolute and relative frequencies. In order to fit multivariate risk models an exploratory analysis was initially applied. Pearson’s correlation coefficient was used in order to measure associations between the continuous variables, while contingency tables with calculation of chi-squared test, as well as application of Student’s t-test evaluated associations between the categorical and continuous variables. Estimations of the relative risks of developing acute coronary syndromes, under several hypotheses, were performed by the calculation of odds ratio (OR) and the corresponding confidence intervals through multiple logistic regression analysis [17]. All reported P-values are based on two-sided tests and compared to a
144
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
significant level of 5%. STATA 6 software was used for the calculations (STATA Corp. College Station, TX, USA).
3. Results Six hundred and fifty-eight (77%) out of the 848 patients were males (59.0610 years old) and 190 (23%) were females (65.369 years old). Also, 830 out of the 1078 controls were males (58.0610 years old) and 248 were females (64.869 years old), of similar age with the patients. Five hundred and twenty-six (62%) coronary patients and 765 (71%) controls adopted the Mediterranean-type of diet. Mean age of those who defined ‘closer’ to the Mediterranean-type of diet was significantly higher than the others (62.569 vs. 56.1610 years old, P50.023). The prevalence of current smoking, hypertension, hypercholesterolemia, diabetes mellitus and obesity in the subgroups of subjects that adopted the Mediterranean diet was significantly lower compared to the others (Table 1). No associations were found between the adoption of Mediterranean-type of diet, the income and the educational level of the subjects ( x 2 520.12, P5 0.455 and x 2 521.33, respectively, P50.559). Also, no associations were found between leisure time physical activity, occupational class (unskilled, semiskilled and skilled) and the adoption of Mediterranean diet ( x 2 512.32, P50.514 and x 2 511.41, respectively, P50.423). Additionally, no interaction was assessed between occupational type physical activity and adoption of Mediterranean type of diet. All the patients and the controls were classified into the aforementioned hypertension groups. As we can see in Table 2, 418 (49%) of the
coronary patients and 303 (28%) of the controls were hypertensive. Table 2 also presents the distribution of several demographic and cardiovascular risk factors according to the group of hypertension. The analysis of the results presented in Table 2, raised a significant association between groups of hypertension and smoking status, in coronary patients ( x 2 538.11, P,0.001) but no association was observed in control groups ( x 2 51.10, P50.337). Additionally, the prevalence of hypercholesterolemia, obesity and diabetes mellitus, both in patients and controls, was significantly related with the hypertension group (P,0.05). Finally, significant relationships were observed between groups of hypertension and educational, financial status of the patients and controls, respectively ( x 2 521.11 and x 2 531.19, P, 0.05). After taking into account the effect of the previous associations and the effect of several potential confounders such as age and sex, by design, presence of hypercholesterolemia, diabetes mellitus, physical inactivity, smoking habit, premature history of coronary heart disease, dose and type of treatment, we performed conditional risk analysis. Our results regarding the effect of the adoption of Mediterranean type of diet on the risk of developing acute coronary syndromes in hypertensive as well in normotensive subjects are presented in Table 3. Further analysis showed that, male sex, low income and educational level, and not having seen a physician in the past year increased the risk of being classified as having hypertension but being unaware of the condition, by 51%, 59%, and 44%, respectively (P,0.01). Also, an age of at least 50 years for males and 55 years for females were found strong risk factors for the lack of awareness of hypertension. In the model that assessed the risk of having acknowl-
Table 1 Prevalence of the conventional cardiovascular risk factors in coronary patients and controls that adopted the Mediterranean diet Mediterranean diet
Current smoking Hypercholesterolemia Hypertension Obesity Diabetes mellitus Physical inactivity (leisure time)
Other type of diet
P-value
Patients n5162
Controls n5265
Patients n5686
Controls n5813
95 (59%) 89 (55%) 71 (44%) 62 (38%) 39 (24%) 129 (80%)
90 (34%) 72 (27%) 64 (24%) 64 (24%) 11 (4%) 164 (62%)
439 418 347 281 199 521
235 (28%) 285 (35%) 239 (29%) 227 (28%) 73 (9%) 479 (59%)
(64%) (61%) (50%) (41%) (29%) (76%)
,0.01 ,0.01 ,0.01 ,0.05 ,0.01 NS
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
145
Table 2 Distribution of cardiovascular risk factors and other socio-demographic variables, by hypertension group in patients (controls) group Normotensive
ACS patients 430 (51%) Controls 775 (72%) Prevalence of cardiovascular risk factors in patients (controls) Smoking 52% (48%) Hypercholesterolemia 58% (23%) Diabetes mellitus 19% (5%) Obesity (BMI.29.9 kg / m 2 ) 21% (16%) Physical inactivity Distribution of socio-demographic factors in patients (controls) Education None–primary school 28% (23%) Senior high or colleges 52% (54%) University 20% (23%) Annual income Low (,4750 US$) 7% (4%) Moderate (4750–8500 US$) 52% (48%) Good (8500–14,500 US$) 37% (39%) Very good (.14,500 US$) 4% (9%)
Uncontrolled hypertension Acknowledged untreated
Treated uncontrolled
21 (2%) 36 (3%)
94 (11%) 34 (3%)
148 (17%) 111 (10%)
155 (19%) 122 (12%)
48% (30%) 47% (28%) 29% (11%) 17% (18%)
69% 65% 32% 28%
45% 62% 33% 18%
(39%) (29%) (7%) (12%)
39% (37%) 43% (22%) 34% (7%) 18% (10%)
39% (33%) 49% (56%) 12% (11%)
37% (31%) 49% (55%) 14% (14%)
39% (30%) 44% (42%) 17% (28%)
32% (33%) 52% (42%) 16% (25%)
9% (3%) 52% (49%) 32% (33%) 7% (15%)
9% (4%) 60% (54%) 27% (33%) 4% (9%)
6% (3%) 60% (54%) 22% (29%) 12% (14%)
8% (5%) 52% (54%) 28% (30%) 12% (11%)
edged but uncontrolled hypertension, male sex, not having seen a physician in the preceding year, low educational and financial levels as well as an age of at least 45 years were significant risk factors for this condition.
4. Discussion In this work we present the findings from the CARDIO2000 study regarding the effect of the adoption of Mediterranean diet on the risk of developing acute coronary syndromes, in hypertensive subjects. The performed risk analysis showed that the adoption of Mediterranean diet is associated with the Table 3 Results from the multiple regression analysis for the effect of the adoption of Mediterranean diet on the risk of developing ACS, by hypertensive and normotensive subjects
Controlled hypertensives Unaware of their condition Acknowledged uncontrolled Normotensive
Treated, controlled
Unaware
Odds ratio
95% CI
P-value
0.83 0.92 0.93 0.80
0.78–0.88 0.87–0.95 0.88–0.95 0.71–0.89
0.003 0.038 0.031 0.001
The comparisons were made between subjects who adopted the Mediterranean diet versus the rest of them.
(41%) (40%) (8%) (21%)
reduction of the risk of developing acute coronary syndromes not only in normotensive subjects, but also in hypertensive as well as in subjects who are unaware of their condition or uncontrolled. There is extensive scientific evidence on the relation between diet and incidence of coronary heart disease, various types of cancer and other diseases [1,2,5,7,13]. These studies showed that dietary factors exert their influence largely through their effects on blood lipids and lipoproteins, as well as on the other established modifiable risk factors, with the exception of cigarette smoking. Our results support the hypothesis that the adoption of the Greek type of Mediterranean diet can significantly reduce the risk of developing acute coronary syndromes under the presence of hypertension, after controlling for several potential confounders. The investigated type of diet is low in saturated fat, high in monounsaturated fat, mainly from olive oil, high in complex carbohydrates, from legumes, and high in fibre, mostly from vegetables and fruits [9,12,18]. Based on the results from the Seven Countries Study [19], the protective role of Mediterranean type of diet against atherosclerosis was explained, mainly, due to the reduction of cholesterol and blood pressure levels. Recently, the findings from the Lyon Diet Heart study [9] illustrate the potential importance of the Mediterranean dietary
146
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147
pattern, especially when compared to other recommended diets, like Step-I diet [20]. However it still remains a matter of debate if the protective influence is primarily caused by single nutrients, e.g. dietary fatty acids, potassium or dietary fibre or if it can be attributed to the Mediterranean diet as a whole [9]. Also, many investigators believe that the effect of the Mediterranean diet, on coronary risk, may be influenced through a number of confounders such as geographic and other non-measured cultural and social differences of the investigated populations [21]. In this study we aimed to explore the effect of Mediterranean diet on the risk of developing acute coronary syndromes. Based on a database of |2000 coronary patients and controls we found that the adoption of Mediterranean diet is associated with a substantial reduction of the coronary risk in hypertensive subjects, even if they were unaware of their condition or if it was uncontrolled. In addition, our risk analysis showed that the one fifth of the non-fatal acute coronary syndromes (attributable risk), due to hypertension, could be prevented in controlled hypertensive subjects through the adoption of this traditional diet. Additionally, 9% of the non-fatal cardiac events in unaware of their condition or uncontrolled subjects could be prevented through the adoption of the Greek type of Mediterranean diet.
5. Limitations of the study In this retrospective case-control study two, main, sources of systematic errors may exist, the selection and the recall bias. In order to eliminate selection bias we tried to set objective criteria, both for patients and controls. Concerning information bias we tried to avoided it through accurate and detailed data from patients’ medical records. However, recall bias may still exist, especially in the measurement of nutritional habits, smoking, and the onset of the investigated cardiovascular risk factors. Furthermore, regarding the potential effect of uncontrolled-unknown confounders, we tried to reduce it through multivariate analysis and using the same study base, both for patients and controls.
Acknowledgements The Hellenic Heart Foundation funds the CARDIO2000 study. This study was supported by research grants from the Hellenic Heart Foundation (11 / 1999–2001). The authors would like to thank the physicians and the specialists that coordinated this study: Dr. K. Tzioumis (Athens, Crete, Pelloponisos), Dr. ´I. Papaioannou (Athens, Thessalia), Dr. P. Stravopodis (Ionian Islands), Dr. L. Karra (Aegean Islands), Dr. D. Antoniades (Macedonia), Dr. G. Rembelos (Aegean Islands), Dr. D. Markou (Athens), A. Moraiti (Athens), D. Evagelou (Crete), Dr. S. Vellas (Attica, Hpeirous), B. Meidanis (Macedonia, Sterea Hellas, Thessalia), Dr. S. Loggos (Attica), Dr. I. Elefsiniotis (Athens), Dr. I. Vogiatzis (Central Macedonia), Dr. N. Marinakis (Aegean Islands), Dr. G. Koutsimbanis (Thrace), and Dr. T. Kyratzoglou (East Macedonia).
References [1] The Joint National Committee on Detection Evaluation and Treatment of High Blood Pressure. The fifth report of the Joint National Committee on Detection Evaluation and Treatment of High Blood Pressure (JNC V). Arch Intern Med 1993;53:54–83. [2] Working Group on Primary Prevention of Hypertension. National High Blood Pressure Education Program Working Group report on primary prevention of hypertension. Arch Intern Med 1993;53:86– 208. [3] Menotti A, Lanti M, Puddu PE, Kromhout D. Coronary heart disease incidence in northern and southern European populations: a reanalysis of the seven countries study for a European coronary risk chart. Heart 2000;84(3):238–44. [4] Kafatos A, Diacatou A, Voukiklaris G, Nikolakakis N, Vlachonikolis J, Kounali D, Mamalakis G, Dontas AS. Heart disease risk-factor status and dietary changes in the Cretan population over the past 30 years: the Seven Countries Study. Am J Clin Nutr 1997;65(6):1882– 6. [5] Burt VL, Whelton P, Roccella EJ et al. Prevalence of hypertension in the US adult population: results from the Third National Health and Nutrition Examination Survey 988-99. Hypertension 1995;25:305– 13. [6] Ascherio A, Rimm EB, Giovannucci EL et al. A prospective study of nutritional factors and hypertension among US men. Circulation 1992;86:475–84. [7] Espiro-Montoro A, Lopez-Miranda J, Castro P et al. Monounsaturated fatty acid enriched diets lower plasma insulin levels and blood pressure in healthy young men. Nutr Metab Cardiovasc Dis 1996;6:147–54. [8] Braunwald E. In: 5th ed, Heart disease, London: WB Saunders, 1997, pp. 1184–202.
D.B. Panagiotakos et al. / International Journal of Cardiology 82 (2002) 141 – 147 [9] de Lorgeril M, Salen P, Martin J-L, Monjaud I, Delaye J, Mamelle N. Mediterranean diet, traditional risk factors and the rate of cardiovascular complications after myocardial infarction. Final report of the Lyon Diet Heart Study. Circulation 1999;99:779–85. [10] Panagiotakos DB, Pitsavos C, Chrysohoou C, Stefanadis C, Toutouzas PK. The role of traditional Mediterranean-type of diet and lifestyle, in the development of acute coronary syndromes: preliminary results from cardio2000 study. C Eur J Pub Health 2002; in press. [11] Trichopoulou A. In: Composition of Greek foods and dishes, Athens, Greece: Athens School of Public Health Publications, 1992, p. 152, in Greek and English. [12] Trichopoulou A, Kouris-Blazos A, Wahlqvist M et al. Diet and overall survival in elderly people. Br Med J 1995;311:1457–60. [13] Assmann G, deBasker G, Bagnara S et al. International Consensus statement on olive oil and the Mediterranean diet: implications for health in Europe. Eur J Cancer Prev 1997;6:418–21. [14] Hyman DJ, Pavlik VN. Characteristics of patients with uncontrolled hypertension in the United States. N Engl J Med 2001;345:479–86. [15] Burt VL, Cutler A, Higgins M et al. Trends in the prevalence, awareness, treatment, and control of hypertension in the adult US population: data from the health examination surveys, 1960 to 1991. Hypertension 1995;26:60–9.
147
[16] Panagiotakos DB, Pitsavos C, Chrysohoou C, Stefanadis C, Toutouzas PK. Risk stratification of coronary heart disease through established and emerging lifestyle factors, in a Mediterranean population: CARDIO2000 epidemiological study. J Cardiovasc Risk 2001;8(6):329–39. [17] Armitage P, Berry U. In: Statistical methods in medical research, Oxford: Blackwell Scientific, 1994, p. 620. [18] Strazullo P, Ferro-Luzzi A, Saini A et al. Changing the Mediterranean diet: effects on blood pressure. J Hypertens 1986;4:407–12. [19] Keys A, Menotti A, Karvonen MJ et al. The diet and 15-year death rate in the Seven Countries Study. Am J Epidemiol 1986;124:903– 15. [20] Kris-Etherton P, Eckel R, Howard B, Jeor S, Bazzare T. Lyon Diet Heart Study. Benefits of a Mediterranean-style, National Education Program /AHA Step I dietary pattern on cardiovascular disease. Circulation 2001;103:1823–5. [21] Robertson RM, Smaha L. Can a Mediterranean style diet reduce heart disease? Circulation 2001;103:1821–2.