Dietary fat composition and cardiac events in patients with type 2 diabetes

Atherosclerosis 236 (2014) 31e38

Contents lists available at ScienceDirect

Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis

Dietary fat composition and cardiac events in patients with type 2 diabetes Ana Luiza Teixeira dos Santos 1, Tanara Weiss 1, Camila Kümmel Duarte 1, Jorge Luiz Gross 1, Mirela Jobim de Azevedo 1, Themis Zelmanovitz* Endocrine Division, Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Rua Ramiro Barcelos, 2350/ Pr
edio 12, 4
andar, CEP 90035-003 Porto Alegre, RS, Brazil

a r t i c l e i n f o

a b s t r a c t

Article history: Received 5 January 2013 Received in revised form 7 May 2014 Accepted 18 June 2014 Available online 25 June 2014

Objective: To evaluate associations of dietary fat composition with the development of cardiac events in patients with type 2 diabetes, without ischemic heart disease who were followed for at least 12 months. Methods: In this prospective cohort study the usual diet of patients was retrospectively assessed by a 3-day weighed diet record (WDR). Compliance with the WDR technique was assessed by comparing protein intake estimated from 3-day WDR and 24-h urinary nitrogen output. The following were considered cardiac events: myocardial infarction, myocardial revascularization procedures, congestive heart failure, new-onset angina pectoris, and sudden death. Results: A total of 227 patients with type 2 diabetes (aged 59 ± 10 years; 46.0% male), were followed during 4.6 years. In a multivariate Cox regression analysis, the intake of polyunsaturated fatty acids had a protective effect for cardiac events (HR ¼ 0.31, 95% CI: 0.11e0.89; P ¼ 0.03) adjusted for age, gender, duration of diabetes, smoking, compliance with WDR, using hypolipidemic agents, and the presence of hypertension and diabetic nephropathy. When the fat intake was divided into quartiles, the highest intake of a-linolenic acid (>1.25% of energy) was negatively associated with cardiac events (HR ¼ 0.58, 95% CI: 0.39e0.85; P ¼ 0.006), adjusted for the same covariates.. Conclusion: In patients with type 2 diabetes without ischemic heart disease, a high intake of polyunsaturated fatty acids, especially alpha linolenic acid, was protective for the development of cardiac events.. © 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: Dietary fatty acids Type 2 diabetes Cardiac events

1. Introduction Cardiovascular disease (CVD) is the main cause of morbidity and mortality in patients with diabetes mellitus. Coronary artery disease is more frequent and more severe in diabetic than nondiabetic patients [1]. Diabetes itself, particularly due to sustained hyperglycemia, greatly contributes to the severity of atherosclerotic disease [2]. Additionally, the known high frequency of associated cardiovascular risk factors in patients with diabetes, such as obesity, dyslipidemia, and hypertension, can also explain the * Corresponding author. Serviço de Endocrinologia, do Hospital de Clínicas de
dio 12, 4
andar, CEP 90035-003 Porto Porto Alegre, Rua Ramiro Barcelos, 2350/Pre Alegre, RS, Brazil. Tel./fax: þ55 51 3359 8127, þ55 51 3331 3312. E-mail addresses: [email protected] (A.L.T. dos Santos), [email protected]. br (T. Weiss), [email protected] (C.K. Duarte), [email protected] (J.L. Gross), [email protected] (M.J. de Azevedo), themis.voy@terra. com.br (T. Zelmanovitz). 1 Tel./fax: þ55 51 3359 8127, þ55 51 3331 3312. http://dx.doi.org/10.1016/j.atherosclerosis.2014.06.014 0021-9150/© 2014 Elsevier Ireland Ltd. All rights reserved.

severity of CVD in diabetes. A cross-sectional, multicenter study including 927 Brazilian outpatients with type 2 diabetes showed a 36% prevalence of coronary artery disease, a 33% of peripheral artery disease, and a 73% of hypertension in these patients [3]. Though hypolipidemic agents have shown the same effectiveness for type 2 diabetic patients and nondiabetic patients, the absolute rates of atherosclerotic cardiovascular disease events remained elevated in patients with diabetes [4]. This reinforces the importance of cardiovascular prevention measures based especially on lifestyle interventions, such as dietary changes; these measures should be implemented to mitigate cardiovascular risk factors. Dietary intervention involves the management of the intake of nutrients associated with cardiovascular protection and risk factors [5]. Current recommendations for the prevention and treatment of CVD in diabetes suggest a diet rich in fruits, vegetables, whole grains, walnuts, and low-fat dairy products, and salt restriction both for normotensive and hypertensive patients [6,7]. The specific recommendations are: reduction in saturated fatty

32

A.L.T. dos Santos et al. / Atherosclerosis 236 (2014) 31e38

acids (SFAs) and cholesterol intake; restriction of trans fatty acids (FA) intake; increase in total fiber intake and increase in consumption of fish (as a source of polyunsaturated FA (PUFA) [6,7]. However, the scientific evidence for these recommendations is generally based on studies conducted with non-diabetic subjects. Furthermore, the effects of dietary fat intake on CVD prevention could be different in patients with and without diabetes, since diabetes has been associated with lipid abnormalities and a prothrombotic profile [8]. The aim of the present study was to evaluate the association between dietary fat composition and the development of cardiac events in patients with type 2 diabetes and without ischemic heart disease. In addition, the role of adherence to current cardiovascular dietary recommendations in the prevention of cardiac events was evaluated. 2. Patients and methods 2.1. Patients We carried out a retrospective data analysis of patients with type 2 diabetes belonging to a prospective cohort study. The patients included in the study were from the Diabetes research outpatient clinic at Hospital de Clínicas de Porto Alegre (Rio Grande do Sul, Brazil) and were followed for at least 12 months. At baseline, the patient exclusion criteria were: presence of ischemic heart disease, body mass index (BMI) > 40 kg/m2, serum creatinine >1.5 mg/dl, heart failure, symptomatic autonomic neuropathy (gastroparesis or diabetic diarrhea), dietary counseling by a registered dietitian during the previous 12 months, and inability to perform the weighed diet records (WDR). The study was approved by the Hospital Ethics Committee; all patients signed a written informed consent form. All patients underwent a standardized clinical, nutritional and laboratory examination at baseline. A three-day WDR was carried out, 24-h urine was collected, and on the 3rd day (after patients fasted overnight), blood samples were also collected. During the reevaluation period, patients were submitted to the same clinical and laboratory examinations. The follow-up was determined as the period between the first assessment and the date of a first cardiac outcome, death, or the last medical evaluation. Patients were divided into two groups according to the presence of cardiac outcomes at the end of the study. 2.2. Methods 2.2.1. Dietary assessment Diet was assessed using a 3-day WDR technique (two nonconsecutive weekdays and one day of the weekend). Patients were issued commercial scales (1e125 g) and measuring cups (25e250 mL; Pirex) and given a detailed explanation and demonstration of the procedures by a trained, registered dietitian. Next, patients performed a training session for a one-day WDR. The 3-day WDR was performed over two to three weeks and before nutritional recommendations for all patients (Fig. 1). Compliance with the WDR technique was assessed by comparison of protein intake estimated from the 3-day WDR (PI-WDR) and from the 24-h urinary nitrogen output (PI-N) performed on the third day of the WDR period [9]. According to Vaz et al. [9], patient compliance with the WDR protocol is established if the PI-WDR/PIN ratio remain between 0.79 and 1.26. The analysis of dietary nutrients from the 3-day WDRs was performed using the Nutribase Clinical Nutritional Manager software (version 7.14, 2007). The mean values of each nutrient consumed during the 3-day WDR were calculated. Nutritional data

for frequently consumed foods were updated if necessary, and/or complemented with data obtained from local manufacturers of industrialized foods. Patients were classified as adherent or non-adherent according to reaching or not each dietary fat goal based on American Diabetes Association (ADA) recommendations [6]. The items of ADA fat intake recommendations are: SFA below 7% of energy/day; cholesterol below 200 mg/day; increasing the intake PUFA, and reducing the intake of trans FAs. Besides using established cutoff points values of ADA dietary recommendations, we also categorized patients as adherent or non-adherent according to the fat intake of all studied sample: above or below the mean (for PUFA) or median (for trans FAs), and according to the quartiles of nutrient intake (for PUFA). 2.2.2. Anthropometric measurements The body weight and height of patients (without shoes or coats) were collected with an anthropometric scale; measurements were recorded to the nearest 100 g for weight, and to the nearest 0.1 cm for height. Waist circumference was measured midway between the lowest rib and the iliac crest, near the umbilicus. 2.2.3. Clinical evaluation at baseline and end-of-study The clinical assessment emphasized cardiovascular evaluation (blood pressure, WHO cardiovascular questionnaire, resting electrocardiogram, and evaluation of peripheral arterial pulses) and renal function. Patients were classified as “nonsmoker” or “smoker” (smoker in the past or currently a smoker). Physical activity was classified into four levels based on a standardized questionnaire, which was adapted to local habits [10]. Mean blood pressure was calculated based on two separate measures using a digital sphygmomanometer (OMRON® Automatic Blood Pressure Monitor, Model HEM-705CP, Vernon Hills, Illinois 60061). Hypertension was defined as blood pressure 140/ 90 mmHg or use of antihypertensive drugs on at least two separate occasions. Presence of baseline ischemic heart disease, an exclusion criterion, was defined as: presence of angina and/or possible infarction, using a World Health Organization (WHO) cardiovascular questionnaire as previously standardized in patients with type 2 diabetes by our group [11], or abnormalities on resting electrocardiogram (Minnesota Codes: Q and QS patterns [1-1 to 1-3]; S-T junction and segment depression [4-1 to 4-4]; T-wave items [5-1 to 5-3], or complete left bundle branch block [7-1]). In those cases, the ischemic heart disease was always confirmed by exercise electrocardiogram or radionuclide myocardial perfusion imaging with stress (exercise or pharmacological) compatible with the presence of myocardial ischemia [12]. Peripheral vascular disease was determined based on the presence of intermittent claudication and/or absence of posterior tibial pulse at clinical examination. The presence of cerebrovascular disease was established if there was a history of cerebrovascular accident (ischemic stroke) and/or compatible findings (sequelae). Renal function was evaluated by serum creatinine and urinary albumin excretion [12]. The diagnosis of increased urinary albumin, microalbuminuria (24-h urinary albumin excretion between 30 and 300 mg) or macroalbuminuria (24-h urinary albumin excretion 300 mg), was confirmed by a second measurement [12,13]. 2.2.4. Cardiac outcomes Cardiac endpoints were defined when a new cardiac event occurred (myocardial infarction [Minnesota codes 1-1 to 1-3 or 7-1

A.L.T. dos Santos et al. / Atherosclerosis 236 (2014) 31e38

33

Fig. 1. Diagram of the study design.

on a resting ECG [11], fixed non-perfused area on myocardial scintigraphy (necrosis), or an episode of chest pain with a compatible ECG], congestive heart failure [class III and IV of the New York Heart Association classification], sudden death [death within 24 h after the last time the patient was seen alive in his/her usual state [14]] or death of cardiac origin), myocardial revascularization procedures, and new onset of angina pectoris (WHO cardiovascular questionnaire). Abnormal resting ECG and new onset of angina pectoris were considered as cardiac outcomes only after stress myocardial scintigraphy (exercise or dipyridamole) confirmation. 2.2.5. Laboratory analyses Glycemic control was evaluated by serum glucose (glucoseperoxidase colorimetric enzymatic method) and glycated hemoglobin (high precision chromatography, Merck-Hitachi 9100 apparatus). The lipid profile consisted of the measurement of total cholesterol and triglycerides using colorimetric assay. HDL cholesterol was measured using the direct enzymatic method. LDL cholesterol was calculated using the Friedewald formula. Serum

creatinine was measured by the Jaffe method and urea by kinetic UV assay. Urinary albumin excretion was evaluated by the immunoturbidimetry technique (Kit MICROALB, AMES) [15]. 3. Statistical analyses Student t test, ManneWhitney U test, and the Exact Fisher or ChieSquare tests were applied. Multivariate Cox's proportional hazard model was used to estimate the risk factors for the presence of cardiac outcomes (dependent variables). Different models were generated to evaluate the association of each nutrient (independent variables) with cardiac events. Independent variables were selected based on their significance (P < 0.10) at univariate analysis or according to their biological relevance. The independent variables were selected based on at least ten cardiac outcomes for each putative predictor. Data were expressed as mean ± standard deviation, or as median (P25 e P75) unless otherwise stated. The level of significance adopted was 5%. The software SPSS 18.0 (SPSS®, Chicago, IL) was used for the analysis.

34

A.L.T. dos Santos et al. / Atherosclerosis 236 (2014) 31e38

4. Results

Table 2 Daily dietary intake of patients with type 2 diabetes categorized according to the development of cardiac events.

4.1. Characteristics of patients according to cardiac events A total of 236 patients underwent clinical, laboratory, and nutritional evaluation at baseline. Nine patients were unable to attend the medical visits or declined participation in the study and thus did not perform the re-evaluation. Therefore, the study included 227 patients (96.2% of the initial population; 106 male) who were followed for 4.6 years (from one to 8.6 years). The mean age of patients was 59 ± 10 years, duration of diabetes 12 ± 8 years, and BMI 28.6 ± 4.0 kg/m2. Thirteen deaths occurred during follow up: one patient died due to cardiac reasons and 12 patients from other causes (complications due to systemic lupus erythematosus, sepsis, postoperative complications, neoplasms, and respiratory insufficiency). During the follow-up, the following cardiac events were identified in 36 patients: myocardial infarction (n ¼ 3), myocardial revascularization procedures (n ¼ 4), congestive heart failure (n ¼ 2), and new-onset angina pectoris (n ¼ 27). Clinical and laboratory characteristics of patients at baseline according to cardiac outcome are shown in Table 1. The patients who presented cardiac events had diabetes for a longer period of time. The other characteristics were not different between the two groups. It should be noted that the low percentages of hypolipidemic agents occurred because this information was recorded at baseline when patients did not have a diagnosis of ischemic heart disease. At the end-ofstudy, the proportion of patients using hypolipidemic drugs was 90% and 57% in the group with and without cardiac events, respectively (p ¼ 0.01). 4.2. Diet characteristics and cardiac events Compliance with the WDR technique was observed in 88% in the group with cardiac events and 83% in the group with no cardiac event (P ¼ 0.41). WDR were obtained in all seasons (27.9% in the winter, 18.6% in the summer, 26.1% in the fall and 27.4% in the spring; P ¼ 0.79). Table 1 Baseline clinical and laboratory characteristics of patients with type 2 diabetes categorized according to the development of cardiac events.

Age (years) Male Ethnicity: White Duration of diabetes (years) BMI (kg/m2) Hypertension Diabetic nephropathy Smoking (self-reported) Alcohol (self-reported) Sedentary lifestyle Diabetes treatment (D/OA/I or I þ OA) ACE inhibitors Beta-blockers Hypolipidemic agents Fasting plasma glucose (mg/dL) Glycated hemoglobin (%) Total cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL)

Without cardiac events (n ¼ 191)

With cardiac events (n ¼ 36)

P

59 ± 10 44% 88% 12 ± 7 28.6 ± 4.1 74% 32% 49% 44% 51% 3%/61%/36%

62 ± 9 58% 87% 14 ± 9 28.9 ± 4.6 85% 50% 58% 42% 66% 3%/65%/32%

0.09 0.15 1.0 0.04 0.72 0.27 0.06 0.36 1.0 0.14 1.0

55% 31% 22% 151 ± 56 7.5 ± 1.5 204 ± 43 53 ± 30 125 ± 38 133 (95e201)

54% 25% 9% 150 ± 53 7.2 ± 1.4 198 ± 56 51 ± 13 132 ± 37 136 (121e166)

1.0 0.68 0.15 0.90 0.32 0.42 0.79 0.33 0.57

Data are expressed as mean ± SD, patients with the characteristic (%), or median (P25eP75). BMI: body mass index; Diabetes treatment: D: diet; OA: oral agents; I: insulins. ACE: angiotensin converting enzyme.

Energy (Kcal) Carbohydrates (% of energy) Proteins (% of energy) Lipids (% of energy) Saturated FA (% of energy) Monounsaturated FA (% of energy) Polyunsaturated FA (% of energy) ALA (% of energy) Linoleic FA (% of energy) EPA (% of energy) DHA (% of energy) Arachidonic FA (% of energy) Trans FA (% of energy) P/S Ratio Cholesterol (mg) Total fiber (g/1000 Kcal)

Without cardiac events (n ¼ 191)

With cardiac events (n ¼ 36)

P

1794 ± 469 47 ± 7 19 ± 3 34 ± 7 9±3 10 ± 3 10 ± 3 0.97 ± 0.49 8.51 ± 3.23 0.00 (0.00e0.01) 0.01 (0.00e0.02) 0.05 (0.03e0.08) 1.2 ± 0.7 1.1 ± 0.4 210 ± 99 11 ± 4

1656 ± 522 46 ± 10 19 ± 3 32 ± 7 10 ± 2 10 ± 2 9±3 0.81 ± 0.43 7.70 ± 2.90 0.01 (0.00e0.03) 0.01 (0.00e0.04) 0.05 (0.04e0.08) 1.3 ± 0.8 0.9 ± 0.4 193 ± 94 11 ± 3

0.11 0.79 0.75 0.31 0.73 0.81 0.10 0.07 0.16 0.69 0.79 0.95 0.53 0.08 0.36 0.65

Data are expressed as mean ± SD or median (p25-p75). FA: fatty acid; ALA:

a-linolenic acid; EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; P/S ratio: polyunsaturated/saturated fat ratio.

The daily dietary intake evaluated by 3-day WDR of patients with type 2 diabetes, categorized according to the development of cardiac events, is shown in Table 2. There were no differences between energy and macronutrient intakes of patients with and without cardiac events. The fractions of daily dietary PUFA were also evaluated. In relation to n-3 PUFA intake, a-linolenic acid (ALA) was the most frequently consumed, with a trend to a lower intake in patients with cardiac events than in patients without cardiac events (0.81 ± 0.43% vs. 0.97 ± 0.49%, respectively; P ¼ 0.07). Multivariate analysis showed that the intake of ALA was negatively associated with the development of cardiac events (HR 0.28, 95% CI 0.09e0.81; P ¼ 0.019), adjusted for age, gender, duration of diabetes, smoking, compliance with WDR, using hypolipidemic agents, and presence of hypertension and diabetic nephropathy. When we repeated multivariate analysis including the current use of hypolipidemic agents in the model, the inverse association of ALA intake with cardiac events remained significant. Regarding n-6 PUFA intake, linoleic acid, the principal component, did not differ between the two groups (7.69 ± 2.81 vs. 8.50 ± 3.23%, respectively; P ¼ 0.16). Moreover, no statistical difference was observed for other evaluated PUFA: eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA), and arachidonic acid. Soybean oil was the most widely consumed vegetable oil (70%); other oils consumed included corn, rice or the combination of more than one type of oil. The proportion of soybean oil intake of patients with cardiac events was significantly lower (55%) than the proportion for patients without end-of-study cardiac events (73%; P ¼ 0.042). In relation to other foods sources of n-3 PUFA, only 14% of patients consumed marine foods, with no difference between patients with (24%) and without (13%) cardiac events. The negative association between the intake of ALA and the development of cardiac events remained significant in multivariate analysis, adjusted for age, gender, duration of diabetes, smoking, compliance with WDR, using hypolipidemic agents, presence of hypertension and diabetic nephropathy, and also soybean oil intake. 4.3. Dietary adherence according to fat diet recommendations and cutoff points of fat intake and cardiac events Adherence to current cardiovascular dietary recommendations was compared in patients with and without cardiac outcomes at

A.L.T. dos Santos et al. / Atherosclerosis 236 (2014) 31e38

35

Fig. 2. Multivariate Cox regression analysis to evaluate the association between the incidence of cardiac events (dependent variable) and alpha linolenic acid intake of type 2 diabetes patients; patients were divided according to the quartiles of intake of this nutrient: 0.6% of energy ¼ blue line; 0.61e0.86% of energy ¼ green line; 0.87e1.24% of energy ¼ yellow line; and >1.25% of energy ¼ purple line. The analysis was adjusted for age, gender, duration of diabetes, smoking, compliance with WDR, using hypolipidemic agents, and presence of hypertension and diabetic nephropathy. P ¼ 0.006. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

the end-of-study. To establish adherence to PUFA intakes, the cutoff point >9.5% of energy was defined (based on the mean intake of the 227 patients studied). The proportion of patients with cardiac events and who adhered to PUFA recommendations was significantly lower than that of the patients without cardiac events (28 vs. 48%; P ¼ 0.03). The multivariate analysis showed that the intake of PUFA >9.5% of energy was negatively associated with the development of cardiac events (HR 0.31 95% CI 0.11e0.89; P ¼ 0.03), adjusted for age, gender, duration of diabetes, smoking, compliance with WDR, using hypolipidemic agents, and presence of hypertension and diabetic nephropathy. There were no significant differences in the proportion of patients with and without cardiac outcomes that adhered to SFA (14 vs. 17%, respectively; P ¼ 0.81) and cholesterol intake recommendations (61 vs. 50%, respectively; P ¼ 0.28). Similarly, the proportion of patients with cardiac events who consumed 1.05% of energy of trans FA (adherents) was not different from the patients without cardiac events (52 vs. 50%; P ¼ 1.00). 4.4. Analyses according to quartiles of PUFA intake and cardiac events Fractions of PUFA were analyzed according to the quartiles of intake. The quartiles of ALA intake were linearly associated with a decrease in cardiac events, as follows: 1st quartile (0.60% of energy): 38.9% of cardiac events; 2nd quartile (0.61e0.86% of energy): 27.8% of cardiac events; 3rd quartile (0.87e1.24% of energy): 19.4% of cardiac events; and 4th quartile (>1.25% of energy): 13.9% of

cardiac events (P for trend ¼ 0.01). This inverse association was confirmed by a multivariate analysis using the 1st quartile of ALA intake as the reference: HR 0.58, 95% CI 0.39e0.85; P ¼ 0.006 (Fig. 2). This analysis was adjusted for age, gender, duration of diabetes, smoking, compliance with WDR, using hypolipidemic agents, and presence of hypertension and diabetic nephropathy. 5. Discussion The current study demonstrated a negative association of PUFA intake with the incidence of cardiac events in a cohort of patients with type 2 diabetes. The association was also shown when the use of hypolipidemic agents and cardiovascular risk factors were taken into consideration. The protective role of PUFA intake was especially significant for ALA consumption. PUFA intake equal to or higher than 9.5% of total energy was associated with up to 70.0% risk reduction for cardiac events. The highest intake of ALA (largest quartile: >1.25% of energy) was associated with a 42% reduction in risk. The benefit of the intake of n-3 PUFA on primary and secondary prevention of CVD in individuals without diabetes has been demonstrated in observational and clinical trials [16], both as a result of dietary supplementation [17] and increased fish consumption [18]. Furthermore, a cardioprotective effect of ALA has also been suggested by a number of observational studies; most studies [19,20], but not all [21], have shown an inverse association between ALA intake and ischemic heart disease. Recently, a metaanalysis of prospective and retrospective studies reinforced the

36

A.L.T. dos Santos et al. / Atherosclerosis 236 (2014) 31e38

evidence regarding the relation of ALA and CVD risk [22]. In these observational studies, higher ALA exposure was associated with a moderately lower risk of CVD. In the pooled dietary analyses, each 1-g/day increment of ALA intake was associated with a 10% lower risk of coronary heart disease death. However, the beneficial effect of ALA intake probably did not occur for others cardiac outcomes such as the incidence of congestive heart failure [23]. The beneficial effects of ALA intake for CVD may be associated with slowing down the formation and calcification of atherosclerotic plaque [24]. Other related mechanisms are plaque stabilization [25] and antiarrhythmic effects [26], either directly or through desaturation and elongation of ALA into EPA. The varied beneficial effects of ALA intake on cardiovascular outcomes in different populations might be influenced by the dietary consumption of EPA and DHA [27]. It is possible that ALA may reduce the risk of ischemic heart disease among subjects with low consumption of these long-chain n-3 PUFA. Among American men with little (<100 mg/day) or no n-3 long-chain PUFA intake, each 1 g of daily ALA intake was associated with a 47% reduction in the ischemic heart disease risk [28]. In the present study, the mean EPA and DHA intake was ~70 mg/day; this level of intake is considered low. However, other authors considered that the association between ALA intake and CVD risk is independent of EPA/DHA intake [29,30]. In a Chinese population-based cohort, individuals in the highest quartile of EPA/DHA (0.46 ± 0.18 g/day) and ALA (0.80 ± 0.36 g/day) intake, had a 33% lower risk of cardiovascular mortality (HR 0.67, 95% CI 0.57e0.78) as compared with subjects in the lowest quartiles of both types (EPA/DHA ¼ 0.19 ± 0.10 g/day; ALA ¼ 0.40 ± 0.11 g/day) [29]. The inverse association between ALA intake and cardiovascular mortality persisted along the quartiles of EPA/DHA intake and even decreased in the lowest one. Reinforcing the possible independent cardiovascular effect of ALA consumption from the amount of EPA and DHA intake, the Nurses' Health Study cohort also demonstrated an inverse association between the highest intake of ALA and sudden cardiac death [30]. Further studies are needed to determine whether the protection for cardiovascular events of dietary ALA intake is influenced or not by EPA/ DHA consumption. The ALA intake, instead of soybean oil itself, is probably responsible for the beneficial effects demonstrated in the heart. Firstly, as we demonstrated, the negative association between ALA intake and cardiac events remained significant even after being adjusted for soybean oil intake. Second, ALA represents a lower proportion of FA (~7%) in the composition of soybean oil as compared with linoleic acid (51%), oleic acid (22.8%), and palmitic acid (10.3%). Furthermore, we did not find an association of these FA with the development of cardiac events. In patients with diabetes, the beneficial effects of PUFA in the progression of CVD [31], as well as in relation to the cardiovascular risk profile [32,33] was demonstrated in cross-sectional [32] and observational [34] studies, and in clinical trials [35,36]. A large female cohort study of type 2 diabetic patients without coronary heart disease showed an inverse association of n-3 PUFA intake (from fish consumption) with the incidence of coronary heart disease and total mortality [34]. In that study, diet was evaluated by food frequency questionnaires answered by mail; the questionnaire is not a comparable tool in relation to the WDR technique used in the present study [37]. Additionally, few randomized controlled trials evaluated the effects of the intake of n-3 PUFA on diabetic CVD and only patients with previous macrovascular disease [35,36,38] or with high cardiovascular risk [39] were included. Moreover, these studies evaluated n-3 effects from supplements and not from dietary sources and a beneficial effect on the progression of the CVD was demonstrated in some of them [35,36]. On the other hand, in the ORIGIN trial, a recent trial conducted in pre-

diabetes and diabetic patients, the n-3 supplementation did not reduce the incidence of cardiovascular events in the presence of CVD [38]. One possible concern associated with n-3 PUFA intake is the deterioration of glycemic control associated with high supplement doses in diabetic subjects [40]. Since hyperglycemia has been associated with an increase in the incidence of CVD [2], this undesirable effect could attenuate the protective cardiovascular effects of n-3 PUFA. However, this deleterious impact on glycemic homeostasis was not confirmed [41,42] and was observed only with high consumption of fish oil and not with moderate intake of n-3 PUFA from marine or plants sources. Regarding the lipid profile, n-3 PUFA intake in diabetic patients seems to have a similar effect as in non-diabetic patients. n-3 PUFA intake decreases serum triglycerides, a dose dependent effect that is influenced by baseline plasma triglyceride levels [42]. Moreover, n-3 consumption might cause a small and controversial increase in HDL and LDL cholesterol levels [41,42]. In the current study, although we demonstrated that ALA intake was a protective factor against the development of cardiac events, we could not demonstrate the expected favorable effect from other n-3 FAs such as EPA and DHA [36]. The absence of this protective effect may have occurred due to the main intake of n-3 PUFA being from vegetable oils, especially soybean oil; this oil is rich in ALA and represented 70% of the total oil consumed by our patients. Moreover, only 14% of patients consumed marine foods, such as fish, which are known to be rich sources of EPA and DHA. The availability of fish and other marine products is limited in many countries [43] and ALA from plant oils and nuts is the predominant source of n-3 PUFA in the typical Western diet. Perhaps, ALA may be a cardioprotective alternative and could be of great importance for public health. Regarding to the absence of association between EPA and DHA intake and incidence of cardiac events in our study, it is important to emphasize that this subject remains inconclusive. Previous randomized, double-blind, placebo-controlled trials reported the efficacy of n-3 fatty acid supplements in the secondary prevention of CVD [44]. However, a recently published meta-analysis showed that there is insufficient evidence for a secondary preventive effect of EPA and DHA supplements against overall cardiovascular events among patients with a history of CVD [45]. In the present study, we did not find an association between SFA intake and the incidence of cardiac events. Although the recommendation is to reduce the intake of SFA to <7% of energy, which is considered level A of evidence according to the current guidelines [6], the recommendation is based on studies that analyzed the effect of changes in the SFA content performed only in individuals without diabetes [46,47]. Our findings are consistent with results from the meta-analysis of prospective cohort studies that evaluated the association of saturated fat with CVD; the study showed no significant evidence for an increased risk of ischemic heart disease or CVD [48]. On the other hand, a recent meta-analysis of randomized controlled trials that evaluated the effect of dietary fat manipulation on cardiovascular mortality and morbidity, suggested that dietary saturated fat reduction (through reduction and/or modification of dietary fat) may be protective of cardiovascular events, reducing them by 14% [49]. However, it is important to emphasize that this cardioprotective effect was seen in studies of men (not of women) and of individuals with moderate or high cardiovascular risk at baseline. Moreover, this protective effect was more clearly observed when saturated fat was replaced with unsaturated fats and not with carbohydrates. So, the relationship between SFA and ischemic heart disease remains controversial and further studies are needed to clarify this question.

A.L.T. dos Santos et al. / Atherosclerosis 236 (2014) 31e38

One major limitation of our study was that the dietary assessment was performed at baseline only and modifications of dietary habits during the follow up could have been missed. However, our dietary data were accurate. We used a standardized 3-day weighed-diet record technique, which includes a 24-h urinary urea measurement to confirm dietary intake estimated from records [50]. In fact, this dietary tool has been largely used to confirm dietary compliance in studies with diabetic patients [32,51,52]. Also linked to dietary data limitation, the measurement of a biological marker of fat intake would have reinforced our results. In fact, serum fatty acids have been used to evaluate fat intake in patients with type 2 diabetes [53]. Another limitation is that the relatively small sample size of our cohort could have accounted for the small proportion of traditional hard cardiac outcomes. In our sample new onset of angina represented almost 65% of the observed cardiac endpoints, and it can be considered a soft cardiac endpoint. However, the identification of new onset angina (myocardial ischemia) was always confirmed by stress myocardial scintigraphy (exercise or dipyridamole). Finally, the retrospective data analyses of patients who belong to a prospective cohort study might be considered as a study weakness. However, all patients were treated in the same outpatient division by the same experienced team using a standardized clinical and laboratory protocol. This cohort has been prospectively followed by our research group since 2000 [9,32,51e53]. Exposure precedes the disease with acceptable confidence in this scenario. 6. Conclusion The present study suggests an association between PUFA intake, especially ALA, with the prevention of cardiac events in patients with type 2 diabetes. Additional studies, especially randomized controlled clinical trials, are needed to confirm these findings in diabetic patients both with and without ischemic heart disease. Conflicts of interests Nothing to declare. Acknowledgments This study was partially supported by Conselho Nacional de
gico (MCT/CNPq # 502050/ Desenvolvimento Científico e Tecnolo 2005-5) and FIPE-Hospital de Clínicas de Porto Alegre. ALTS was ~o Coordenaça ~o de Aperfeirecipient of scholarship from Fundaça çoamento de Pessoal de Nível Superior (CAPES). References [1] Bloomgarden ZT. Cardiovascular disease and diabetes. Diabetes Care 2003;26: 230e7. [2] Wei M, Gaskill SP, Haffner SM, Stern MP. Effects of diabetes and level of glycemia on all cause and cardiovascular mortality, The San Antonio Heart Study. Diabetes Care 1998;21(7):1167e72. [3] Scheffel RS, Bortolanza D, Weber CS, Costa LA, Canani LH, Santos KG, et al. Prevalence of micro- and macroangiopathic chronic complications and their risk factors in the care of outpatients with type 2 diabetes mellitus. Rev Assoc Med Bras 2004;50:263e7. [4] Shepherd J, Barter P, Carmena R, Deedwania P, Fruchart JC, Haffner S, et al. Effect of lowering LDL cholesterol substantially below currently recommended levels in patients with coronary heart disease and diabetes: the Treating to New Targets (TNT) study. Diabetes Care 2006;29:1220e6. [5] Cernea S, H^ ancu N, Raz I. Diet and coronary heart disease in diabetes. Acta Diabetol 2003;40:S389e400. [6] Nutrition Recommendations and Interventions for Diabetes - 2008: A position statement of the American Diabetes Association. Diabetes Care 2008; 31(Supplement 1):S61e78. [7] Evert AB, Boucher JL, Cypress M, Dunbar SA, Franz MJ, Mayer-Davis EJ, et al. Nutrition Therapy Recommendations for the Management of Adults With Diabetes. Diabetes Care 2014;37(Supplement 1):S120e43.

37

[8] Seligman BG, Biolo A, Polanczyk CA, Gross JL, Clausell N. Increased plasma levels of endothelin 1 and von Willebrand factor in patients with type 2 diabetes and dyslipidemia. Diabetes Care 2000;23(9):1395e400. [9] Vaz JS, Bittencourt M, Almeida JC, Gross JL, De Azevedo MJ, Zelmanovitz T. Protein intake estimated by weighed diet records in patients with type 2 diabetes: misreporting and intra-individual variability using 24-hour nitrogen output as criterion standard. J Am Diet Assoc 2008;108(5):867e72. €m J, Eriksson JG, Valle TT, Ha €ma €l€ [10] Tuomilehto J, Lindstro ainem H, IlanneParikka P, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with Impaired glucose tolerance. N Engl J Med 2001;344(18): 1343e50. [11] Azevedo MJ, Neto AFR, Caramori MLA, Beck MO, Moreira JSR, Ludwig R, Gross JL. Value of diagnostic tools for myocardial ischemia used in routine clinical practice to predict cardiac events in patients with Type 2 Diabetes Mellitus: A Prospective Study. Arq Bras Endocrinol Metab 2006;50(1):46e52. [12] American Diabetes Association: Standards of Medical Care in Diabetes. Diabetes Care 2014;37(Suppl. 1):S14e80. [13] Gross JL, de Azevedo MJ, Silveiro SP, Canani LH, Caramori ML, Zelmanovitz T. Diabetic nephropathy: diagnosis, prevention, and treatment. Diabetes Care 2005;28:164e76. [14] Burke AP, Farb A, Malcom GT, Liang YH, Smialek J, Virmani R. Coronary risk factors and plaque morphology in men with coronary disease who died suddenly. N Engl J Med 1997;336:1276e82.
todo [15] Zelmanovitz T, Oliveira JR, Lulier F, Gross JL, Azevedo MJ. Avaliaç~ ao do me
trico para medida da excreça ~o urina
ria de albumina em imunoturdibime pacientes com diabete melito. Arq Bras Endocrinol Metab 1995;38:207e11. [16] De Caterina R. n-3 fatty acids in cardiovascular disease. N Engl J Med 2011;364(25):2439e50. [17] Yokoyama M, Origasa H, Matsuzaki M, Matsuzawa Y, Saito Y, Ishikawa Y, et al. Japan EPA lipid intervention study (JELIS) Investigators. Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis. Lancet 2007;369: 1090e8. [18] Wang C, Harris WS, Chung M, Lichtenstein AH, Balk EM, Kupelnick B, et al. n-3 Fatty acids from fish or fish-oil supplements, but not alpha-linolenic acid, benefit cardiovascular disease outcomes in primary- and secondary-prevention studies: a systematic review. Am J Clin Nutr 2006;84(1):5e17.
L, Pankow JS, Eckfeldt JH. Relation between dietary linolenic acid and [19] Djousse coronary artery disease in the National Heart, Lung, and Blood Institute Family Heart Study. Am J Clin Nutr 2001;74(5):612e9. [20] Campos H, Baylin A, Willett WC. Alpha-linolenic acid and risk of nonfatal acute myocardial infarction. Circulation 2008;118(4):339e45.
MC, Feskens EJ, Kok FJ, Kromhout D. alpha-Linolenic acid [21] Oomen CM, Ocke intake is not beneficially associated with 10-y risk of coronary artery disease incidence: the Zutphen Elderly Study. Am J Clin Nutr 2001;74(4):457e63. [22] Pan A, Chen M, Chowdhury R, Wu JH, Sun Q, Campos H, et al. a-Linolenic acid and risk of cardiovascular disease: a systematic review and meta-analysis. Am J Clin Nutr 2012;96(6):1262e73. [23] Lemaitre RN, Sitlani C, Song X, King IB, McKnight B, Spiegelman D, et al. Circulating and dietary a-linolenic acid and incidence of congestive heart failure in older adults: the Cardiovascular Health Study. Am J Clin Nutr 2012;96(2):269e74.
L, Arnett DK, Carr JJ, Eckfeldt JH, Hopkins PN, Province MA, et al. [24] Djousse Dietary linolenic acid is inversely associated with calcified atherosclerotic plaque in the coronary arteries: the National Heart, Lung, and Blood Institute Family Heart Study. Circulation 2005;111:2921e6. [25] Thies F, Garry JM, Yaqoob P, Rerkasem K, Williams J, Shearman CP, et al. Association of n-3 polyunsaturated fatty acids with stability of atherosclerotic plaques: a randomised controlled trial. Lancet 2003;361:477e85. [26] London B, Albert C, Anderson ME, Giles WR, Van Wagoner DR, Balk E, et al. Omega-3 fatty acids and cardiac arrhythmias: prior studies and recommendations for future research: a report from the National Heart, Lung, and Blood Institute and Office of Dietary Supplements Omega-3 Fatty Acids and their Role in Cardiac Arrhythmogenesis Workshop. Circulation 2007;116(10): e320e35. [27] Siscovick DS, Lemaitre RN, Mozaffarian D. The Fish Story: A Diet-Heart Hypothesis With Clinical Implications: n-3 Polyunsaturated Fatty Acids, Myocardial Vulnerability, and Sudden Death. Circulation 2003;107:2632e4. [28] Mozaffarian D, Ascherio A, Hu FB, Stampfer MJ, Willett WC, Siscovick DS, et al. Interplay between different polyunsaturated fatty acids and risk of coronary heart disease in men. Circulation 2005;111(2):157e64. [29] Koh AS, Pan A, Wang R, Odegaard AO, Pereira MA, Yuan JM, et al. The association between dietary omega-3 fatty acids and cardiovascular death: the Singapore Chinese Health Study. Eur J Prev Cardiol 2013 Dec 16. [30] Albert CM, Oh K, Whang W, Manson JE, Chae CU, Stampfer MJ, et al. Dietary alpha-linolenic acid intake and risk of sudden cardiac death and coronary heart disease. Circulation 2005;112(21):3232e8. € AT, Lichtenstein AH, Mozaffarian D, Herrington DM. Fish intake is [31] Erkkila associated with a reduced progression of coronary artery atherosclerosis in postmenopausal women with coronary artery disease. Am J Clin Nutr 2004;80:626e32. [32] Almeida JC, Zelmanovitz T, Vaz JS, Steemburgo T, Perassolo MS, Gross JL, et al. Sources of protein and polyunsaturated fatty acids of the diet and microalbuminuria in type 2 diabetes mellitus. J Am Coll Nutr 2008;27(5): 528e37.

38

A.L.T. dos Santos et al. / Atherosclerosis 236 (2014) 31e38

[33] Hartweg J, Farmer AJ, Holman RR, Neil HA. Meta-analysis of the effects of n-3 polyunsaturated fatty acids on haematological and thrombogenic factors in type 2 diabetes. Diabetologia 2007;50(2):250e8. [34] Hu FB, Cho E, Rexrode KM, Albert CM, Manson JE. Fish and long-chain omega3 fatty acid intake and risk of coronary heart disease and total mortality in diabetic women. Circulation 2003;107(14):1852e7. [35] Mita T, Watada H, Ogihara T, Nomiyama T, Ogawa O, Kinoshita J, et al. Eicosapentaenoic acid reduces the progression of carotid intima-media thickness in patients with type 2 diabetes. Atherosclerosis 2007;191(1):162e7. [36] Kromhout D, Giltay EJ, Geleijnse JM. Alpha Omega Trial Group. n-3 Fatty Acids and Cardiovascular Events after Myocardial Infarction. N Engl J Med 2010;363(21):2015e26. [37] Thompson FE & Subar AF. Dietary Assessment Methodology. In “Nutrition in the Prevention and Treatment of Disease”. Second Edition (Edited by Ann M. Coulston and Carol J. Boushey), pp. 3e39. Elsevier Academic Press, California. [38] ORIGIN Trial Investigators, Bosch J, Dagenais GR, Díaz R, Dyal L, Jung H, et al. n-3 fatty acids and cardiovascular outcomes in patients with dysglycemia. N Engl J Med 2012;367(4):309e18. [39] Singh RB, Dubnov G, Niaz MA, Ghosh S, Singh R, Rastogi SS, et al. Effect of an Indo-Mediterranean diet on progression of coronary artery disease in high risk patients (Indo-Mediterranean Diet Heart Study): a randomised singleblind trial. Lancet 2002;360(9344):1455e61. [40] Woodman RJ, Mori TA, Burke V, Puddey IB, Watts GF, Beilin LJ. Effects of purified eicosapentaenoic and docosahexaenoic acids on glycemic control, blood pressure, and serum lipids in type 2 diabetic patients with treated hypertension. Am J Clin Nutr 2002;76:1007e15. [41] Friedberg CE, Janssen MJ, Heine RJ, Grobbee DE. Fish oil and glycemic control in diabetes. A meta-analysis. Diabetes Care 1998;21(4):494e500. [42] Balk EM, Lichtenstein AH, Chung M, Kupelnick B, Chew P, Lau J. Effects of omega-3 fatty acids on serum markers of cardiovascular disease risk: a systematic review. Atherosclerosis 2006;189(1):19e30. [43] Lang T. Food control or food democracy? Re-engaging nutrition with society and the environment. Public Health Nutr 2005;8:730e7. [44] von Schacky C, Angerer P, Kothny W, Theisen K, Mudra H. The effect of dietary omega-3 fatty acids on coronary atherosclerosis. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 1999;130(7):554e62.

[45] Kwak SM, Myung SK, Lee YJ, Seo HG. Efficacy of omega-3 fatty acid supplements (eicosapentaenoic acid and docosahexaenoic acid) in the secondary prevention of cardiovascular disease: a meta-analysis of randomized, double-blind, placebo-controlled trials. Arch Intern Med 2012;172(9): 686e94. [46] Third report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III): Final report. US Department of Health and Human Services; Public Health Service; National Institutes of Health; National Heart, Lung, and Blood Institute. Circulation 2002;106:3143. [47] Ginsberg HN, Kris-Etherton P, Dennis B, Elmer PJ, Ershow A, Lefevre M, et al. Effects of reducing dietary saturated fatty acids on plasma lipids and lipoproteins in healthy subjects: the Delta Study, Protocol 1. Arterioscler Thromb Vasc Biol 1998;18:441e9. [48] Siri-Tarino PW, Sun Q, Hu FB, Krauss RM. Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease. Am J Clin Nutr 2010;91(3):535e46. [49] Hooper L, Summerbell CD, Thompson R, Sills D, Roberts FG, Moore HJ, et al. Reduced or modified dietary fat for preventing cardiovascular disease. Cochrane Database Syst Rev. 2012:5. [50] Moulin CC, Tiskievicz F, Zelmanovitz T, de Oliveira J, Azevedo MJ, Gross JL. Use ofweighed diet records in thee valuation of diets with different protein contents in patients with type 2 diabetes. Am J Clin Nutr 1998;67:853e7. ~o C, et al. [51] Gross JL, Zelmanovitz T, Moulin CC, De Mello V, Perassolo M, Leita Effect of a chicken-based diet on renal function and lipid profile in patients with type 2 diabetes: a randomized crossover trial. Diabetes Care 2002;25: 645e65. [52] de Mello VD, Zelmanovitz T, Azevedo MJ, de Paula TP, Gross JL. Long-term effect of a chicken-based diet versus enalapril on albuminuria in type 2 diabetic patients with microalbuminuria. J Ren Nutr 2008;18(5):440e7. [53] Perassolo M, Almeida JC, Steemburgo T, Dall'Alba V, de Mello VD, Zelmanovitz T, et al. Endothelial dysfunction and serum fatty acid composition in patients with type 2 diabetes mellitus. Metabolism Clinical and Experimental 2008;57:1167e72.