DIABETES/METABOLISM RESEARCH AND REVIEWS REVIEW Diabetes Metab Res Rev 2014; 30(Suppl. 1): 34–40 Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/dmrr.2516
ARTICLE
Mediterranean diet and type 2 diabetes
Katherine Esposito1* Dario Giugliano2 1
Department of Clinical and Experimental Medicine, Second University of Naples, Naples, Italy
2
Department of Medical, Surgical, Neurological, Metabolic Sciences, and Geriatrics, Second University of Naples, Naples, Italy *Correspondence to: Katherine Esposito, Department of Clinical and Experimental Medicine, Second University of Naples, Naples, Italy. E-mail: [email protected]
Abstract Consumption of selected dietary components is favourably associated with prevention of type 2 diabetes, but discordant results for some foods or single nutrients continue to appear. The study of complete dietary patterns represents the most adequate approach to assess the role of diet on the risk of diabetes. The term ‘Mediterranean diet’ essentially refers to a primarily plant-based dietary pattern whose greater consumption has been associated with higher survival for lower all-cause mortality. At least five large prospective studies report a substantially lower risk of type 2 diabetes in healthy people or at risk patients with the highest adherence to a Mediterranean diet. Five randomized controlled trials have evaluated the effects of a Mediterranean diet, as compared with other commonly used diets, on glycaemic control in subjects with type 2 diabetes. Improvement of HbA1c levels was greater with a Mediterranean diet and ranged from 0.1% to 0.6% for HbA1c. No trial reported worsening of glycaemic control with a Mediterranean diet. Although no controlled trial specifically assessed the role of a Mediterranean diet in reducing cardiovascular events in type 2 diabetes, there is evidence that post-infarct or high-risk patients, including diabetic patients, may have cardiovascular benefits from a Mediterranean diet. The evidence so far accumulated suggests that adopting a Mediterranean diet may help prevent type 2 diabetes; moreover, a lower carbohydrate, Mediterraneanstyle diet seems good for HbA1c reduction in persons with established diabetes. Copyright © 2013 John Wiley & Sons, Ltd. Keywords Mediterranean diet; type 2 diabetes; dietary patterns; diabetes prevention; glycaemic control; HbA1c
Introduction
Received: 23 October 2013 Accepted: 19 December 2013
Copyright © 2013 John Wiley & Sons, Ltd.
The pandemic of type 2 diabetes is an enormous public health problem, with 552 million cases projected by 2030 worldwide [1]. There is an increasing awareness for new therapeutic approaches that delay progression to type 2 diabetes in people at increased risk for the disease, including those with impaired glucose tolerance or impaired fasting glucose. Lifestyle intervention studies have demonstrated a reduction of new diabetes ranging from 30% to 67%, which remains after the individual lifestyle counselling was stopped [2–4]. However, little is known whether lifestyle intervention can also reduce cardiovascular disease morbidity or mortality. The 20-year follow-up results from the Chinese Da Qing Study and results of more than 10 years of followup from the Finnish Diabetes Prevention Study showed no statistically significant differences in cardiovascular outcomes between the intervention and
35
Mediterranean Diet, Type 2 Diabetes
control groups [4,5]. Although lifestyle interventions can prevent type 2 diabetes, current nutritional recommendations for the primary prevention of type 2 diabetes are limited, with little that is truly evidence based [6]. Little emphasis has also been put on lifestyle change, including diet, in current diabetic algorithms for management of hyperglycaemia in type 2 diabetes [7].
Diet and prevention of type 2 diabetes Consumption of selected dietary components, such as foods or single nutrients, is favourably associated with prevention of type 2 diabetes. Whole grains, coffee or tea, low-fat milk and dairy products, moderate alcohol consumption, fruits and vegetables, pulses, and nuts (in women only) are associated with a decreased risk of type 2 diabetes [8]. As far as single nutrients are concerned, moderate evidence from several prospective observational studies or meta-analysis indicate a protective role for cereal fibre and unsaturated fatty acids [8]. Although selected micronutrients may affect glucose and insulin metabolism, to date there, is no convincing evidence that documents their role in the development of diabetes [9]. Moreover, discordant results continue to appear for whole grains [10], fruit and vegetables [11] and fish [12]. In the case of fruit and vegetable, for example, this may, in part, be due to the effect of measurement error in the assessment of their consumption on the association with diabetes, as food frequency questionnaires are known to overestimate fruit and vegetable consumption [13]. A recent systematic review and meta-analysis [14] demonstrated a small, protective effect of fruit and vegetable consumption on incidence of type 2 diabetes; however, the evidence was limited to only leafy vegetables, mostly in women, and in four trials only, with 7422 cases out of 169 807 non-cases. By contrast, fruit was better than vegetable in reducing diabetes risk in the European Prospective Investigation of Cancer study during 12 years of follow-up [15].
Dietary patterns Dietary-patterning analysis has been increasingly used as an alternative method to traditional single nutrient analysis because it can assess cumulative effects of the overall diet. The use of dietary patterns avoids focusing on single foods or nutrients that can be correlated with, or interact with, each other and assesses combinations of food which alone may have effects too small to be identified. Thus, the assessment of whole dietary patterns has become instrumental in nutritional epidemiology, even because Copyright © 2013 John Wiley & Sons, Ltd.
synergistic or antagonistic effects may exist between the different components of a food pattern [16]. Three main approaches have been used to define dietary patterns: factor analysis, cluster analysis and dietary indices [17]. The predominant a posteriori methods are factor and cluster analyses which identify the major dietary patterns independently of their relevance to disease. The a priori approach is used to describe the ideal diet for disease prevention based on available evidence of the disease. Therefore, a diet score is obtained by summing up foods considered to be important for a specific disease. Scores take into account both the quantity and quality of food, whereas assessment of nutrient intake cancels most dietary confounding by incorporating possible nutritional confounders in the score, and capture possible effect modification among nutritional variables through the same process. More important than isolated foods or nutrients, the study of complete dietary patterns represents the most adequate approach to assess the role of diet on the risk of diabetes.
Mediterranean diet The term ‘Mediterranean diet’ essentially refers to a primarily plant-based diet whose greater consumption has been associated with higher survival for lower allcause mortality [18,19]. The Mediterranean diet was first described in the 1960s by Ancel Keys, based on his observation of food habits of some populations in the Mediterranean region [20]: the traditional dietary patterns typical of Crete, much of the rest of Greece, and southern Italy in the early 1960s were considered to be largely responsible for the good health observed in these regions. There is no single Mediterranean diet, as more than 20 countries have a coastline in the Mediterranean basin; moreover, social, political, religious and economic differences introduce variations to the Mediterranean diet both between and within these countries. Despite this, a high consumption of foods of vegetable origin, such as cereals and whole grains, fruits, vegetables, legumes, nuts; olive oil as the principal source of fat; fish and poultry consumed in low-to-moderate amounts; relatively low consumption of red meat; and moderate consumption of wine, normally with meals, could be considered important characteristics of this dietary pattern [21].
Mediterranean diet and prevention of type 2 diabetes A recent systematic review and meta-analysis of prospective studies assessed the effect of dietary patterns, Diabetes Metab Res Rev 2014; 30(Suppl 1): 34–40. DOI: 10.1002/dmrr
36
including Mediterranean-style dietary patterns, on prevention of type 2 diabetes [22]. Ten large prospective studies were identified that used either the a posteriori or a priori approaches to explore the link between dietary factors and the risk of type 2 diabetes. The analysis comprised 190 301 subjects free of type 2 diabetes at baseline, followed for a time ranging from 2 to 23 years, and 8932 cases of incident diabetes. Overall, adherence to healthy dietary pattern was associated with reduced risk of developing type 2 diabetes: combined mean difference = 0.39, 95% confidence interval (CI) 0.54 to 0.24. A common characteristic of these ‘preventive’ dietary patterns is their abundant plant food content. Two studies specifically evaluated the role of Mediterranean diet in diabetes prevention. In 13 380 Spanish university graduates without diabetes at baseline followed up for a median of 4.4 years, participants who adhered closely to a Mediterranean diet had a 83% lower risk of diabetes compared with those with the lowest adherence score [23]. The Mediterranean score included legumes, grains, fruits and nuts, vegetables, fish, meat and meat products, alcohol, milk and dairy products, and ratio of monounsaturated to saturated fat. In the large post-infarct GISSI-Prevenzione trial including 8291 Italian patients with a recent myocardial infarction followed up for 3.5 years [24], a Mediterranean diet protected against new diabetes (35% lower risk). The Mediterranean score was based on consumption of cooked and raw vegetables, fruit, fish and olive oil. Not unexpectedly, the characteristics of dietary patterns associated with prevention of type 2 diabetes are reminiscent of the ‘prudent diet’ containing high amounts of vegetables, legumes, whole grains, fruit, fish, poultry and low-fat dairy products [25]. At the opposite, ‘the Western diet’ characterized by a high consumption of red meat, processed meat, eggs, refined grains, sweets and dessert, French fries and high-fat dairy products was associated with higher risk of developing type 2 diabetes [22]. Interestingly, the potential protection from diabetes by a Mediterranean-style diet reported in some cohorts [23,24] may be considered as anti-inflammatory and anti-oxidative, principally because of the increased intake of protective nutrients such as dietary fibre, vitamins and minerals, as well as antioxidants and polyphenols, whereas they are also characterized by lower intake of pro-inflammatory aliments and nutrients, such as saturated and trans fatty acids, refined sugars and starches and generally, foods that account for excess energy intake [26]. Supporting this line of thought, a recent metaanalysis of prospective studies identified two inflammatory markers, that is, interleukin-6 and C-reactive protein, significantly associated with diabetes, with an increased risk of 26% for elevated C-reactive protein levels and of 31% for elevated interleukin-6 levels [27]. Copyright © 2013 John Wiley & Sons, Ltd.
K. Esposito and D. Giugliano
Since the publication of our meta-analysis [22], other studies have addressed the topic of prevention of type 2 diabetes by dietary patterns. In the large prospective European Prospective Investigation into Cancer and Nutrition study [28], adherence to the Mediterranean dietary pattern was associated with a small reduction in the risk of developing type 2 diabetes: in particular, there was a 12% decreased risk among individuals with high adherence to the Mediterranean dietary pattern (score 11–18 points) as compared with individuals with low adherence (0–6 points). Participants from the Nurses’ Health Study II with prior gestational diabetes mellitus were followed up for 15 years: adherence to a Mediterranean diet pattern was associated with 40% lower risk of type 2 diabetes, partly attenuated by adjustment for body mass index [29]. Similarly, men from the Health Professional Follow-Up Study were followed up to 20 years for incidence of type 2 diabetes: greater adherence to a Mediterranean diet score was associated with 25% lower incidence of future diabetes (quintile 5, score 7–9, versus quintile 1, score 0–2) [30]. On the other hand, several diet-quality scores, including Dietary Approach to Stop Hypertension (DASH) and alternate Healthy Eating Index (aHEI) dietary patterns, were also associated with a lower risk of type 2 diabetes [29,30], indicating that healthful dietary patterns may reflect a common pattern characterized by high intakes of plant-based foods, such as whole grains, moderate alcohol, associated with low intakes of red and processed meat, sodium, sugar-sweetened beverages and trans fat. Equally important, these dietary patterns, including Mediterranean diet, may yield the greatest risk reduction in people with a high body mass index.
Mediterranean diet and glycaemic control An inverse association has been found between adherence to Mediterranean diet and indices of glucose homeostasis in the general population, including elderly people, and high-risk patients [31]. Only few studies have specifically evaluated the association between Mediterranean diet and glycaemic control in type 2 diabetes. According to a recent Italian study in 901 outpatients with type 2 diabetes [32], greater adherence to the traditional Mediterranean diet was associated with lower HbA1c levels and 2-h post-meal glucose levels independently of variations in age, adiposity, energy intake, physical activity and other potential confounders. The differences between diabetic patients with high adherence to a Mediterranean-type diet as compared with those with low adherence were 0.9% for HbA1c (p < 0.001) and 39 mg/dL (p < 0.001) for 2-h glucose. These data were Diabetes Metab Res Rev 2014; 30(Suppl 1): 34–40. DOI: 10.1002/dmrr
37
Mediterranean Diet, Type 2 Diabetes
in part confirmed by a smaller cross-sectional study of 262 individuals with type 2 diabetes participating in a Spanish trial at single centre [33]: after adjustment for age, sex, physical activity, smoking, time of evolution of the type 2 diabetes, body mass index and insulin treatment, the odds ratio for having HbA1c ≥7% was 0.69 (95% CI: 0.17–2.83) for those participants with the highest adherence to the Mediterranean diet. Although there was a potential inverse non-significant association between adherence to Mediterranean diet and HbA1c levels, it was not statistically significant. Five randomized controlled trials [34–38] evaluated the effects of Mediterranean diet, as compared with other commonly used diets, on glycaemic control in type 2 diabetes. In postmenopausal women with type 2 diabetes (n = 279) randomized to either usual care or comprehensive lifestyle self-management program including a Mediterranean low-saturated fat diet, HbA1c decreased by about 0.4% (p = 0.001) at 6 months, with no change in control women [34]. In 772 high-risk persons, including 421 (54.5%) diabetic patients, Estruch et al. [35] evaluated the short-term effects of two Mediterranean diets (supplemented with either 1 L/week of virgin olive oil or 30 g/day of tree nuts), compared with a low-fat diet, and found lower fasting glucose (7 mg/dL) in the Mediterranean diet groups. In a 2-year trial, Shai et al. [36] compared three weight-loss diets in 322 moderately obese subjects; among the 46 participants with diabetes, there was no difference in HbA1c decrease between the groups assigned to the Mediterranean diet as compared with the low-fat diet (mean difference: 0.1%, p = 0.45), although a significant decrease in fasting glucose concentration ( 32.8 mg/dL) was found in the Mediterranean diet group. In a 4-year randomized interventional trial, the longest study performed up to now, Esposito et al. [37] assigned a total of 215 patients with newly diagnosed type 2 diabetes to follow-up, with 108 in the Mediterranean diet group and 107 on the low-fat diet. The Mediterranean diet has a carbohydrate content of less than 50% of daily energy and 30% of calories from fat. Ten patients in each group were lost to follow-up. At year 1, HbA1c levels were lower in the Mediterranean diet group than the low-fat diet group (difference: 0.6%, 95% CI 0.9–0.3%). Lastly, in a 12-month trial, comparative study of low-carbohydrate Mediterranean diet versus the American Diabetes Association (ADA) diet in 259 overweight diabetic patients type 2 diabetic patients on stable anti-hyperglycaemic drug therapy, Elhayany et al. [38] found a significantly greater reduction in HbA1c level (0.4%, p = 0.02) in those allocated to the Mediterranean diet with a lower carbohydrate content. A recent systematic review evaluating the literature published since 2001 until 2010 [39] found no advantage in using the Mediterranean-style eating pattern compared Copyright © 2013 John Wiley & Sons, Ltd.
with other eating patterns for glycaemic control. This summary statement, however, was based on imperfect reporting of published data and was also missing reporting of published studies [40]. The evidence so far accumulated from the long-term (from 12 to 48 months) randomized controlled trials suggests that adopting a Mediterranean diet may improve glycaemic control and cardiovascular risk in individuals with established diabetes [41]. Another systematic review with meta-analysis included 21 randomized controlled trials with interventions lasting 6 or more months and compared low carbohydrate, vegetarian, vegan, low glycaemic index, high fibre, Mediterranean and high protein diets with control diets in type 2 diabetes [42]. The low carbohydrate, low glycaemic index, Mediterranean and high protein diets all led to a greater improvement in glycaemic control compared with their respective control diets, with the largest effect size seen in Mediterranean diet (mean difference in HbA1c = 0.41% HbA1c, p < 0.001). The heterogeneity between the three studies [34,37,38] included in the meta-analysis was high (I2 = 82%, p = 0.004), probably owing to the choice of including the comparison between traditional Mediterranean diet and the ADA diet which yielded an advantage of 0.2% HbA1c for the ADA diet [38]. On the contrary, the reduction in HbA1c level was significantly greater (0.4% difference) in the low-carbohydrate Mediterranean diet than in the ADA diet. Interestingly enough, the best results in terms of HbA1c reduction in type 2 diabetic patients are obtained with a Mediterranean-style diet with a carbohydrate content of less that 50% of daily energy [37,38].
Mediterranean diet and vascular disease in type 2 diabetes Coronary heart disease risk is increased in the presence of type 2 diabetes [43]. Two controlled trials [36,37] evaluated the effects of Mediterranean diet on cardiovascular risk factors in type 2 diabetic patients and found more marked improvement of traditional cardiovascular risk factors, including systolic blood pressure, triglycerides, ratio of total to high-density lipoprotein (HDL) cholesterol and HDL cholesterol in those allocated to a Mediterranean diet, as compared with diabetic subjects receiving a control diet. There are no controlled clinical trials that specifically assessed the role of a Mediterranean diet in reducing cardiovascular events and mortality in type 2 diabetes. However, two large interventional studies included a representative diabetic population. In the GISSI-Prevenzione clinical trial [44], a total of 11 323 men and women, including 1700 diabetic subjects, with myocardial infarction from 172 centres in Italy, received advice to increase Diabetes Metab Res Rev 2014; 30(Suppl 1): 34–40. DOI: 10.1002/dmrr
38
K. Esposito and D. Giugliano
their consumption of Mediterranean foods, including fish, fruit, raw and cooked vegetables and olive oil. Compared with people in the worst dietary score quarter, the odds ratio of mortality for those in the best score quarter was 0.51 (95% CI 0.44–0.59). In the PREDIMED study [45], a multicentre trial in Spain, 7447participants (age range, 55–80 years) at highcardiovascular risk, but with no cardiovascular disease at enrolment, were randomly allocated to one of three diets: a Mediterranean diet supplemented with extra-virgin olive oil, a Mediterranean diet supplemented with mixed nuts or a control diet (advice to reduce dietary fat). The primary end point was the rate of major cardiovascular events (myocardial infarction, stroke or death from cardiovascular causes). The multivariable-adjusted hazard ratios were 0.70 (95% CI, 0.54–0.92) and 0.72 (95% CI, 0.54–0.96) for the group assigned to a Mediterranean diet with extra-virgin olive oil (96 events) and the group assigned to a Mediterranean diet with nuts (83 events), respectively, versus the control group (109 events). On the basis of these positive results, the trial was stopped after a median follow-up of 4.8 years. About one half of participants had diabetes at baseline: in this population, the risk of the composite primary endpoint was reduced by 29% (HR: 0.71, 0.53–0.96) and was not different from the reduction observed in the non-diabetic people. The Melbourne Collaborative Cohort Study [46] included 16 610 males and 23 860 females and aimed to determine whether a Mediterranean-style diet could reduce mortality in 2150 subjects who had previously been diagnosed with diabetes or had elevated blood glucose at baseline (1990–1994). There were 4082 deaths, including 874 cardiovascular deaths, over an average follow-up of 12.3 years. A personal Mediterranean Diet Score was calculated using data from a validated 121-item food frequency questionnaire. A trend for lower cardiovascular mortality in those adhering more closely to the Mediterranean diet was consistently seen across all glucose tolerance categories except impaired glucose tolerance in men but only in the women with normal glucose tolerance or newly identified diabetes. However, 95% CIs associated with these mortality rates were wide and overlapping. Although there is some evidence that adopting or adhering closely to a Mediterranean dietary patterns
may help protect the cardiovascular system from diabetic injuries, the lack of specific interventional trials represents a major limitation for a widespread adoption of this diet in every type 2 diabetic patient. On the other hand, there is a robust epidemiological evidence in the general population indicating that greater adherence to Mediterranean diet is significantly associated with a reduced risk of both overall and cardiovascular mortality [47].
Conclusions At the question ‘which diet is best for diabetes?’, Sawyer and Gale [48] gave their perspective about the ideal diet that should be spare, sensible and taken with exercise, a diet for which pharmacotherapy is an adjunct, rather than a substitute, a diet that the patient believes in. An ideal diet should include the many components thought to be beneficial for cardio-metabolic risk. Although a description of such an all-inclusive diet is still lacking, it seems that a low-carbohydrate (
ARTICLE
Mediterranean diet and type 2 diabetes
Katherine Esposito1* Dario Giugliano2 1
Department of Clinical and Experimental Medicine, Second University of Naples, Naples, Italy
2
Department of Medical, Surgical, Neurological, Metabolic Sciences, and Geriatrics, Second University of Naples, Naples, Italy *Correspondence to: Katherine Esposito, Department of Clinical and Experimental Medicine, Second University of Naples, Naples, Italy. E-mail: [email protected]
Abstract Consumption of selected dietary components is favourably associated with prevention of type 2 diabetes, but discordant results for some foods or single nutrients continue to appear. The study of complete dietary patterns represents the most adequate approach to assess the role of diet on the risk of diabetes. The term ‘Mediterranean diet’ essentially refers to a primarily plant-based dietary pattern whose greater consumption has been associated with higher survival for lower all-cause mortality. At least five large prospective studies report a substantially lower risk of type 2 diabetes in healthy people or at risk patients with the highest adherence to a Mediterranean diet. Five randomized controlled trials have evaluated the effects of a Mediterranean diet, as compared with other commonly used diets, on glycaemic control in subjects with type 2 diabetes. Improvement of HbA1c levels was greater with a Mediterranean diet and ranged from 0.1% to 0.6% for HbA1c. No trial reported worsening of glycaemic control with a Mediterranean diet. Although no controlled trial specifically assessed the role of a Mediterranean diet in reducing cardiovascular events in type 2 diabetes, there is evidence that post-infarct or high-risk patients, including diabetic patients, may have cardiovascular benefits from a Mediterranean diet. The evidence so far accumulated suggests that adopting a Mediterranean diet may help prevent type 2 diabetes; moreover, a lower carbohydrate, Mediterraneanstyle diet seems good for HbA1c reduction in persons with established diabetes. Copyright © 2013 John Wiley & Sons, Ltd. Keywords Mediterranean diet; type 2 diabetes; dietary patterns; diabetes prevention; glycaemic control; HbA1c
Introduction
Received: 23 October 2013 Accepted: 19 December 2013
Copyright © 2013 John Wiley & Sons, Ltd.
The pandemic of type 2 diabetes is an enormous public health problem, with 552 million cases projected by 2030 worldwide [1]. There is an increasing awareness for new therapeutic approaches that delay progression to type 2 diabetes in people at increased risk for the disease, including those with impaired glucose tolerance or impaired fasting glucose. Lifestyle intervention studies have demonstrated a reduction of new diabetes ranging from 30% to 67%, which remains after the individual lifestyle counselling was stopped [2–4]. However, little is known whether lifestyle intervention can also reduce cardiovascular disease morbidity or mortality. The 20-year follow-up results from the Chinese Da Qing Study and results of more than 10 years of followup from the Finnish Diabetes Prevention Study showed no statistically significant differences in cardiovascular outcomes between the intervention and
35
Mediterranean Diet, Type 2 Diabetes
control groups [4,5]. Although lifestyle interventions can prevent type 2 diabetes, current nutritional recommendations for the primary prevention of type 2 diabetes are limited, with little that is truly evidence based [6]. Little emphasis has also been put on lifestyle change, including diet, in current diabetic algorithms for management of hyperglycaemia in type 2 diabetes [7].
Diet and prevention of type 2 diabetes Consumption of selected dietary components, such as foods or single nutrients, is favourably associated with prevention of type 2 diabetes. Whole grains, coffee or tea, low-fat milk and dairy products, moderate alcohol consumption, fruits and vegetables, pulses, and nuts (in women only) are associated with a decreased risk of type 2 diabetes [8]. As far as single nutrients are concerned, moderate evidence from several prospective observational studies or meta-analysis indicate a protective role for cereal fibre and unsaturated fatty acids [8]. Although selected micronutrients may affect glucose and insulin metabolism, to date there, is no convincing evidence that documents their role in the development of diabetes [9]. Moreover, discordant results continue to appear for whole grains [10], fruit and vegetables [11] and fish [12]. In the case of fruit and vegetable, for example, this may, in part, be due to the effect of measurement error in the assessment of their consumption on the association with diabetes, as food frequency questionnaires are known to overestimate fruit and vegetable consumption [13]. A recent systematic review and meta-analysis [14] demonstrated a small, protective effect of fruit and vegetable consumption on incidence of type 2 diabetes; however, the evidence was limited to only leafy vegetables, mostly in women, and in four trials only, with 7422 cases out of 169 807 non-cases. By contrast, fruit was better than vegetable in reducing diabetes risk in the European Prospective Investigation of Cancer study during 12 years of follow-up [15].
Dietary patterns Dietary-patterning analysis has been increasingly used as an alternative method to traditional single nutrient analysis because it can assess cumulative effects of the overall diet. The use of dietary patterns avoids focusing on single foods or nutrients that can be correlated with, or interact with, each other and assesses combinations of food which alone may have effects too small to be identified. Thus, the assessment of whole dietary patterns has become instrumental in nutritional epidemiology, even because Copyright © 2013 John Wiley & Sons, Ltd.
synergistic or antagonistic effects may exist between the different components of a food pattern [16]. Three main approaches have been used to define dietary patterns: factor analysis, cluster analysis and dietary indices [17]. The predominant a posteriori methods are factor and cluster analyses which identify the major dietary patterns independently of their relevance to disease. The a priori approach is used to describe the ideal diet for disease prevention based on available evidence of the disease. Therefore, a diet score is obtained by summing up foods considered to be important for a specific disease. Scores take into account both the quantity and quality of food, whereas assessment of nutrient intake cancels most dietary confounding by incorporating possible nutritional confounders in the score, and capture possible effect modification among nutritional variables through the same process. More important than isolated foods or nutrients, the study of complete dietary patterns represents the most adequate approach to assess the role of diet on the risk of diabetes.
Mediterranean diet The term ‘Mediterranean diet’ essentially refers to a primarily plant-based diet whose greater consumption has been associated with higher survival for lower allcause mortality [18,19]. The Mediterranean diet was first described in the 1960s by Ancel Keys, based on his observation of food habits of some populations in the Mediterranean region [20]: the traditional dietary patterns typical of Crete, much of the rest of Greece, and southern Italy in the early 1960s were considered to be largely responsible for the good health observed in these regions. There is no single Mediterranean diet, as more than 20 countries have a coastline in the Mediterranean basin; moreover, social, political, religious and economic differences introduce variations to the Mediterranean diet both between and within these countries. Despite this, a high consumption of foods of vegetable origin, such as cereals and whole grains, fruits, vegetables, legumes, nuts; olive oil as the principal source of fat; fish and poultry consumed in low-to-moderate amounts; relatively low consumption of red meat; and moderate consumption of wine, normally with meals, could be considered important characteristics of this dietary pattern [21].
Mediterranean diet and prevention of type 2 diabetes A recent systematic review and meta-analysis of prospective studies assessed the effect of dietary patterns, Diabetes Metab Res Rev 2014; 30(Suppl 1): 34–40. DOI: 10.1002/dmrr
36
including Mediterranean-style dietary patterns, on prevention of type 2 diabetes [22]. Ten large prospective studies were identified that used either the a posteriori or a priori approaches to explore the link between dietary factors and the risk of type 2 diabetes. The analysis comprised 190 301 subjects free of type 2 diabetes at baseline, followed for a time ranging from 2 to 23 years, and 8932 cases of incident diabetes. Overall, adherence to healthy dietary pattern was associated with reduced risk of developing type 2 diabetes: combined mean difference = 0.39, 95% confidence interval (CI) 0.54 to 0.24. A common characteristic of these ‘preventive’ dietary patterns is their abundant plant food content. Two studies specifically evaluated the role of Mediterranean diet in diabetes prevention. In 13 380 Spanish university graduates without diabetes at baseline followed up for a median of 4.4 years, participants who adhered closely to a Mediterranean diet had a 83% lower risk of diabetes compared with those with the lowest adherence score [23]. The Mediterranean score included legumes, grains, fruits and nuts, vegetables, fish, meat and meat products, alcohol, milk and dairy products, and ratio of monounsaturated to saturated fat. In the large post-infarct GISSI-Prevenzione trial including 8291 Italian patients with a recent myocardial infarction followed up for 3.5 years [24], a Mediterranean diet protected against new diabetes (35% lower risk). The Mediterranean score was based on consumption of cooked and raw vegetables, fruit, fish and olive oil. Not unexpectedly, the characteristics of dietary patterns associated with prevention of type 2 diabetes are reminiscent of the ‘prudent diet’ containing high amounts of vegetables, legumes, whole grains, fruit, fish, poultry and low-fat dairy products [25]. At the opposite, ‘the Western diet’ characterized by a high consumption of red meat, processed meat, eggs, refined grains, sweets and dessert, French fries and high-fat dairy products was associated with higher risk of developing type 2 diabetes [22]. Interestingly, the potential protection from diabetes by a Mediterranean-style diet reported in some cohorts [23,24] may be considered as anti-inflammatory and anti-oxidative, principally because of the increased intake of protective nutrients such as dietary fibre, vitamins and minerals, as well as antioxidants and polyphenols, whereas they are also characterized by lower intake of pro-inflammatory aliments and nutrients, such as saturated and trans fatty acids, refined sugars and starches and generally, foods that account for excess energy intake [26]. Supporting this line of thought, a recent metaanalysis of prospective studies identified two inflammatory markers, that is, interleukin-6 and C-reactive protein, significantly associated with diabetes, with an increased risk of 26% for elevated C-reactive protein levels and of 31% for elevated interleukin-6 levels [27]. Copyright © 2013 John Wiley & Sons, Ltd.
K. Esposito and D. Giugliano
Since the publication of our meta-analysis [22], other studies have addressed the topic of prevention of type 2 diabetes by dietary patterns. In the large prospective European Prospective Investigation into Cancer and Nutrition study [28], adherence to the Mediterranean dietary pattern was associated with a small reduction in the risk of developing type 2 diabetes: in particular, there was a 12% decreased risk among individuals with high adherence to the Mediterranean dietary pattern (score 11–18 points) as compared with individuals with low adherence (0–6 points). Participants from the Nurses’ Health Study II with prior gestational diabetes mellitus were followed up for 15 years: adherence to a Mediterranean diet pattern was associated with 40% lower risk of type 2 diabetes, partly attenuated by adjustment for body mass index [29]. Similarly, men from the Health Professional Follow-Up Study were followed up to 20 years for incidence of type 2 diabetes: greater adherence to a Mediterranean diet score was associated with 25% lower incidence of future diabetes (quintile 5, score 7–9, versus quintile 1, score 0–2) [30]. On the other hand, several diet-quality scores, including Dietary Approach to Stop Hypertension (DASH) and alternate Healthy Eating Index (aHEI) dietary patterns, were also associated with a lower risk of type 2 diabetes [29,30], indicating that healthful dietary patterns may reflect a common pattern characterized by high intakes of plant-based foods, such as whole grains, moderate alcohol, associated with low intakes of red and processed meat, sodium, sugar-sweetened beverages and trans fat. Equally important, these dietary patterns, including Mediterranean diet, may yield the greatest risk reduction in people with a high body mass index.
Mediterranean diet and glycaemic control An inverse association has been found between adherence to Mediterranean diet and indices of glucose homeostasis in the general population, including elderly people, and high-risk patients [31]. Only few studies have specifically evaluated the association between Mediterranean diet and glycaemic control in type 2 diabetes. According to a recent Italian study in 901 outpatients with type 2 diabetes [32], greater adherence to the traditional Mediterranean diet was associated with lower HbA1c levels and 2-h post-meal glucose levels independently of variations in age, adiposity, energy intake, physical activity and other potential confounders. The differences between diabetic patients with high adherence to a Mediterranean-type diet as compared with those with low adherence were 0.9% for HbA1c (p < 0.001) and 39 mg/dL (p < 0.001) for 2-h glucose. These data were Diabetes Metab Res Rev 2014; 30(Suppl 1): 34–40. DOI: 10.1002/dmrr
37
Mediterranean Diet, Type 2 Diabetes
in part confirmed by a smaller cross-sectional study of 262 individuals with type 2 diabetes participating in a Spanish trial at single centre [33]: after adjustment for age, sex, physical activity, smoking, time of evolution of the type 2 diabetes, body mass index and insulin treatment, the odds ratio for having HbA1c ≥7% was 0.69 (95% CI: 0.17–2.83) for those participants with the highest adherence to the Mediterranean diet. Although there was a potential inverse non-significant association between adherence to Mediterranean diet and HbA1c levels, it was not statistically significant. Five randomized controlled trials [34–38] evaluated the effects of Mediterranean diet, as compared with other commonly used diets, on glycaemic control in type 2 diabetes. In postmenopausal women with type 2 diabetes (n = 279) randomized to either usual care or comprehensive lifestyle self-management program including a Mediterranean low-saturated fat diet, HbA1c decreased by about 0.4% (p = 0.001) at 6 months, with no change in control women [34]. In 772 high-risk persons, including 421 (54.5%) diabetic patients, Estruch et al. [35] evaluated the short-term effects of two Mediterranean diets (supplemented with either 1 L/week of virgin olive oil or 30 g/day of tree nuts), compared with a low-fat diet, and found lower fasting glucose (7 mg/dL) in the Mediterranean diet groups. In a 2-year trial, Shai et al. [36] compared three weight-loss diets in 322 moderately obese subjects; among the 46 participants with diabetes, there was no difference in HbA1c decrease between the groups assigned to the Mediterranean diet as compared with the low-fat diet (mean difference: 0.1%, p = 0.45), although a significant decrease in fasting glucose concentration ( 32.8 mg/dL) was found in the Mediterranean diet group. In a 4-year randomized interventional trial, the longest study performed up to now, Esposito et al. [37] assigned a total of 215 patients with newly diagnosed type 2 diabetes to follow-up, with 108 in the Mediterranean diet group and 107 on the low-fat diet. The Mediterranean diet has a carbohydrate content of less than 50% of daily energy and 30% of calories from fat. Ten patients in each group were lost to follow-up. At year 1, HbA1c levels were lower in the Mediterranean diet group than the low-fat diet group (difference: 0.6%, 95% CI 0.9–0.3%). Lastly, in a 12-month trial, comparative study of low-carbohydrate Mediterranean diet versus the American Diabetes Association (ADA) diet in 259 overweight diabetic patients type 2 diabetic patients on stable anti-hyperglycaemic drug therapy, Elhayany et al. [38] found a significantly greater reduction in HbA1c level (0.4%, p = 0.02) in those allocated to the Mediterranean diet with a lower carbohydrate content. A recent systematic review evaluating the literature published since 2001 until 2010 [39] found no advantage in using the Mediterranean-style eating pattern compared Copyright © 2013 John Wiley & Sons, Ltd.
with other eating patterns for glycaemic control. This summary statement, however, was based on imperfect reporting of published data and was also missing reporting of published studies [40]. The evidence so far accumulated from the long-term (from 12 to 48 months) randomized controlled trials suggests that adopting a Mediterranean diet may improve glycaemic control and cardiovascular risk in individuals with established diabetes [41]. Another systematic review with meta-analysis included 21 randomized controlled trials with interventions lasting 6 or more months and compared low carbohydrate, vegetarian, vegan, low glycaemic index, high fibre, Mediterranean and high protein diets with control diets in type 2 diabetes [42]. The low carbohydrate, low glycaemic index, Mediterranean and high protein diets all led to a greater improvement in glycaemic control compared with their respective control diets, with the largest effect size seen in Mediterranean diet (mean difference in HbA1c = 0.41% HbA1c, p < 0.001). The heterogeneity between the three studies [34,37,38] included in the meta-analysis was high (I2 = 82%, p = 0.004), probably owing to the choice of including the comparison between traditional Mediterranean diet and the ADA diet which yielded an advantage of 0.2% HbA1c for the ADA diet [38]. On the contrary, the reduction in HbA1c level was significantly greater (0.4% difference) in the low-carbohydrate Mediterranean diet than in the ADA diet. Interestingly enough, the best results in terms of HbA1c reduction in type 2 diabetic patients are obtained with a Mediterranean-style diet with a carbohydrate content of less that 50% of daily energy [37,38].
Mediterranean diet and vascular disease in type 2 diabetes Coronary heart disease risk is increased in the presence of type 2 diabetes [43]. Two controlled trials [36,37] evaluated the effects of Mediterranean diet on cardiovascular risk factors in type 2 diabetic patients and found more marked improvement of traditional cardiovascular risk factors, including systolic blood pressure, triglycerides, ratio of total to high-density lipoprotein (HDL) cholesterol and HDL cholesterol in those allocated to a Mediterranean diet, as compared with diabetic subjects receiving a control diet. There are no controlled clinical trials that specifically assessed the role of a Mediterranean diet in reducing cardiovascular events and mortality in type 2 diabetes. However, two large interventional studies included a representative diabetic population. In the GISSI-Prevenzione clinical trial [44], a total of 11 323 men and women, including 1700 diabetic subjects, with myocardial infarction from 172 centres in Italy, received advice to increase Diabetes Metab Res Rev 2014; 30(Suppl 1): 34–40. DOI: 10.1002/dmrr
38
K. Esposito and D. Giugliano
their consumption of Mediterranean foods, including fish, fruit, raw and cooked vegetables and olive oil. Compared with people in the worst dietary score quarter, the odds ratio of mortality for those in the best score quarter was 0.51 (95% CI 0.44–0.59). In the PREDIMED study [45], a multicentre trial in Spain, 7447participants (age range, 55–80 years) at highcardiovascular risk, but with no cardiovascular disease at enrolment, were randomly allocated to one of three diets: a Mediterranean diet supplemented with extra-virgin olive oil, a Mediterranean diet supplemented with mixed nuts or a control diet (advice to reduce dietary fat). The primary end point was the rate of major cardiovascular events (myocardial infarction, stroke or death from cardiovascular causes). The multivariable-adjusted hazard ratios were 0.70 (95% CI, 0.54–0.92) and 0.72 (95% CI, 0.54–0.96) for the group assigned to a Mediterranean diet with extra-virgin olive oil (96 events) and the group assigned to a Mediterranean diet with nuts (83 events), respectively, versus the control group (109 events). On the basis of these positive results, the trial was stopped after a median follow-up of 4.8 years. About one half of participants had diabetes at baseline: in this population, the risk of the composite primary endpoint was reduced by 29% (HR: 0.71, 0.53–0.96) and was not different from the reduction observed in the non-diabetic people. The Melbourne Collaborative Cohort Study [46] included 16 610 males and 23 860 females and aimed to determine whether a Mediterranean-style diet could reduce mortality in 2150 subjects who had previously been diagnosed with diabetes or had elevated blood glucose at baseline (1990–1994). There were 4082 deaths, including 874 cardiovascular deaths, over an average follow-up of 12.3 years. A personal Mediterranean Diet Score was calculated using data from a validated 121-item food frequency questionnaire. A trend for lower cardiovascular mortality in those adhering more closely to the Mediterranean diet was consistently seen across all glucose tolerance categories except impaired glucose tolerance in men but only in the women with normal glucose tolerance or newly identified diabetes. However, 95% CIs associated with these mortality rates were wide and overlapping. Although there is some evidence that adopting or adhering closely to a Mediterranean dietary patterns
may help protect the cardiovascular system from diabetic injuries, the lack of specific interventional trials represents a major limitation for a widespread adoption of this diet in every type 2 diabetic patient. On the other hand, there is a robust epidemiological evidence in the general population indicating that greater adherence to Mediterranean diet is significantly associated with a reduced risk of both overall and cardiovascular mortality [47].
Conclusions At the question ‘which diet is best for diabetes?’, Sawyer and Gale [48] gave their perspective about the ideal diet that should be spare, sensible and taken with exercise, a diet for which pharmacotherapy is an adjunct, rather than a substitute, a diet that the patient believes in. An ideal diet should include the many components thought to be beneficial for cardio-metabolic risk. Although a description of such an all-inclusive diet is still lacking, it seems that a low-carbohydrate (
