Nutrition and Cardiovascular and Metabolic Diseases

Additional content is available online at Elsevier eBooks for Practicing Clinicians

Together with physical activity, smoking cessation, sleep, and stress reduction, a healthy diet forms the foundation for prevention and treatment of cardiometabolic diseases, including coronary heart disease (CHD), stroke, type 2 diabetes mellitus (DM), obesity, and related conditions. Dietary factors represent 8 of the top 25 causes of global deaths, largely owing to effects on cardiometabolic diseases (see also Chapter 1 ). Insufficient intakes of healthy foods, such as fruits, nuts, whole grains, beans, vegetables, seafood, and yogurt, cause substantial burdens; as do excess intakes of salt, sugary beverages, and processed meats. In the United States, suboptimal diet is the leading cause of poor health, estimated to cause approximately 1 in 4 overall deaths.

Obesity, DM, and related conditions have increased in recent decades, owing to rapid social, cultural, and environmental transitions transmitted primarily through changes in diet and other lifestyle habits. Familiarity with the evidence for effects of different dietary factors, including controversies and uncertainties, is essential to prioritize interventions to improve eating habits and reduce diet-related diseases.

For much of the 20th century, research and policy focused on nutrient deficiency diseases (e.g., scurvy, pellagra) and increased agricultural production of inexpensive, shelf stable, starchy crops (e.g., rice, wheat, corn) to feed a rapidly growing world population. These efforts were successful at achieving their goals, leading to a modern global food system that emphasizes commodity crops and shelf-stable, inexpensive packaged and processed foods rich in starch and sugar, preserved by salt, and fortified with vitamins. This legacy food system was built to address caloric hunger and vitamin deficiencies, not diet-related chronic diseases.

It was not until the 1980s that modern nutrition science turned to focus more on chronic conditions like cardiovascular disease (CVD), DM, and obesity. Over the last 40 years, the emerging science has rapidly advanced from less reliable ecologic (cross-national) comparisons, short-term experiments, and animal studies to more robust evidence from prospective cohort studies of disease endpoints, well-conducted metabolic trials of diverse risk markers and pathways, and randomized clinical trials of disease endpoints. Several new conclusions have emerged. First, dietary habits influence not only blood cholesterol (a major focus of the 1980s) and obesity (a major current focus), but also multiple other established and emerging risk factors ( Fig. 29.1 ). Consequently, health effects of any dietary factor cannot be inferred from effects on any single intermediate endpoint. Second, specific foods and overall dietary patterns, rather than isolated single nutrients, are most important for cardiometabolic health. Third, insufficient intakes of protective foods—e.g., minimally processed, phytonutrient-rich foods like fruits, seeds, nuts, beans, vegetables, and whole grains—produce similar or greater disease burdens than excess intakes of unhealthy factors.

This chapter reviews the evidence for effects of diet on cardiometabolic diseases, and highlights key knowledge gaps. Because translation of knowledge into action is essential, this chapter also briefly reviews effective behavior change strategies.

Energy Balance (see also Chapter 30 )

The global obesity epidemic is strikingly recent, commencing in the 1980s in the United States and many nations after decades of relative stability. Abdominal adiposity, which produces largest metabolic harms, has also increased more than overall weight in many nations. The breadth, depth, and pace of this epidemic, including in young children, suggest strong environmental drivers, rather than purely behavioral or genetic explanations.

For years, the main scientific and policy responses to obesity have emphasized energy balance and calorie counting. Beyond the empiric failure of this approach to stem the obesity epidemic worldwide, new evidence indicates that the quality of food is the major driver of energy imbalance and weight gain. Foods represent complex matrices of nutrients, ingredients, and processing characteristics, each with pleotropic effects on a range of pathways and tissues. More highly processed foods drive greater ad libitum energy intake (+508 kcal/day) and weight gain, while carbohydrate quality and quantity also influence energy expenditure . Diet quality is thus a major determinant of long-term energy imbalance, which can be considered a downstream mediator, not an upstream determinant, of the obesity epidemic. While nearly any popular diet can work for short-term weight loss, healthful food-based patterns appear most important for long-term weight maintenance. Consequently, long-term obesogenic effects of foods should not be judged on the basis of calorie content alone, but also the molecular and physiologic effects that drive subsequent long-term energy intake and expenditure.

The evidence for effects of foods and diet patterns on obesity and weight control are described further in sections below. Mechanisms are being elucidated and involve, beyond conscious cues like hunger and satiety, unconscious drivers like brain craving and reward, glucose-insulin responses, hepatic fat synthesis, adipocyte function, visceral adiposity, metabolic expenditure, and the gut microbiome. ^, ^, Other interacting factors include physical activity, industry marketing, TV watching, sleep duration, circadian alignment, and maternal-fetal (e.g., trans-generational) influences. For example, lower sleep duration and altered circadian rhythms alter hunger, food preferences toward “comfort foods” and leptin, ghrelin, insulin, and gut-peptide concentrations. TV watching increases obesity and weight gain, but mediated by changes in diet (e.g., eating in front of the TV, less healthy choices due to TV marketing) rather than physical inactivity. Liquid calories from sugary drinks and alcohol, larger portion sizes, and more meals away from home also associate with adiposity. Changes in social norms and networks, industry marketing, and local food availability also appear important. ^, Because habitual excess energy intakes as small as approximately 50 to 100 kcal/d may explain much of the obesity epidemic, subtle effects on these pathways are sufficient, when sustained, to account for population shifts in weight (see Fig. 25.4).

Notably, while global obesity has appropriately highlighted the central role of nutrition in health, a focus on obesity alone misses the many other health consequences of dietary habits (see Fig. 29.1 ). Changes in diet quality substantially improve cardiometabolic risk factors within 6 to 8 weeks, even without weight loss. Thus, obesity should be considered only one mediating pathway for health effects of diet, and the main nutritional targets and metrics of success for individuals and populations should be for overall health, not weight.

Foods

The successes of nutritional science and dietary guidelines in the 20th century to address vitamin deficiency diseases ensconced a reductionist approach to food that emphasized isolated single nutrients. As chronic diseases emerged as a major public health problem in the 1980s, this scientific emphasis on single nutrients lingered. Thus, dietary fat was considered the major cause of obesity; and saturated fat and cholesterol, the major causes of heart disease. Advances in nutrition science demonstrate that, except for certain major additives like sodium, sugar, or trans fat, single nutrients in isolation are less relevant than food types and overall diet patterns —which comprise complex matrices of processing, carbohydrate types, fatty acids, proteins, micronutrients, and phytonutrients. ^,

Fruits and Vegetables

Higher fruit and vegetable intake associates with less long-term weight gain and lower incidence of CHD and stroke ( Fig. 29.2 ). ^, Total fruit or vegetable intake does not associate with DM, perhaps due to greater importance of certain subtypes. 100% fruit juice similarly appears fairly neutral for glycemia and DM, ^, and associates with modestly lower risk of CVD. In controlled trials lasting up to two years, diets including an emphasis on fruits and vegetables improve BP, lipid levels, insulin resistance, inflammation, adiposity, and endothelial function. Such benefits likely derive from the combinations of micronutrients, phytonutrients, and fiber in fruits and vegetables, as well as their replacing less healthful foods. Together these studies provide consistent evidence that fruits and vegetables improve cardiometabolic health. Phytonutrient-rich fruits, such as berries, may have particular benefit. ^, ^,

FIGURE 29.2, Meta-analyses of foods and associations with cardiometabolic outcomes. CHD , Coronary heart disease; CVD , cardiovascular disease; NR, not reported.

Nuts and Beans

Nuts are rich in unsaturated fats, vegetable protein, fiber, folate, minerals, tocopherols, and phenolic compounds. Consumption of nuts lowers total cholesterol, LDL-cholesterol, ApoB, triglycerides, and insulin resistance in trials; ^, associates with lower incidence of CHD in prospective studies (see Fig. 29.2 ); ^, and was a key component of a large Mediterranean diet trial that reduced abdominal obesity and risk of hard CVD endpoints by 30%. While their energy density has raised theoretical concerns for weight gain, nuts are rich in healthy fats, fiber, and phenolics, and both long-term observational studies and controlled trials demonstrate that nuts and seeds do not promote, and actually may reduce, weight gain and visceral adiposity. ^,

Cardiovascular effects of beans (used herein to include pulses [edible seeds] and legumes) are less well established. Like nuts, beans contain bioactive compounds including phenolics, minerals, and fiber; although also more starch, compared with unsaturated fat-rich nuts. In observational cohorts, bean intake inversely associates with total CVD, CHD, and incident hypertension, but not significantly with stroke or DM (see Fig. 29.2 ). ^, Meta-analyses of small trials of soy foods suggest modest improvements in blood cholesterol levels and arterial stiffness; and small to no effects on other risk factors such as glycemic control, blood pressure, inflammation, and body weight. Based on available evidence, eating nuts is a priority for cardiometabolic health; legumes may also be beneficial.

Whole Grains, Refined Grains, Starches, Sweets

Carbohydrate-rich foods dominate most diets: bread, rice, white potatoes, breakfast cereals, crackers, pastas, chips and fries, salty snacks, muffins, sweet bakery products, sugar-sweetened beverages (SSBs), and candy. Health effects of such foods appear determined by several characteristics that only partly overlap: whole grain versus refined carbohydrate (starch+sugar) content, dietary fiber content, glycemic load, and food processing ( Fig. 29.3 ).

FIGURE 29.3, Meta-analyses of beverages and associations with cardiometabolic outcomes. BMI , Body mass index; CHD , coronary heart disease; CVD , cardiovascular disease; NR , not reported; SSB , sugar-sweetened beverage.

A whole grain is a seed including its bran (exterior skin; providing fiber, B-vitamins, minerals, flavonoids, and tocopherols), endosperm (starchy interior; nearly all glucose), and germ (plant embryo; providing fatty acids, antioxidants, and phytonutrients). Refined carbohydrates include refined grains (e.g., white flour, white rice), stripped of their bran and germ to leave only starchy endosperm, and added sugars. All refined carbohydrates (sugars or starch) produce rapid, dose-dependent glycemic responses, with similar overall health harms. ^, ^, Thus, refined carbohydrate (i.e., starch) in foods should be considered a “hidden sugar.”

Foods rich in refined grains, starches, and added sugars associate with risk of long-term weight gain. While the quantity of intake of refined grains does not significantly associate with CHD or DM (see Fig . 29.2 ), more discriminatory measures such as glycemic index and especially glycemic load, that account for both quantity and rapidity of digestion, strongly associate with CHD, stroke, and DM (see Fig . 29.3 ). Metabolic feeding studies confirm harms of refined carbohydrates, while clinical trials demonstrate substantial weight loss and improved glycemia on diets that reduce refined carbohydrates and glycemic load. Added sugars in beverages appear most deleterious, perhaps owing to a combination of large portion sizes, rapid intake patterns, and limited effects on satiety.

In contrast, foods containing whole grains or dietary fiber associate with lower risk of CVD, DM, and long-term weight gain (see Figs. 29.2 and 29.3 ). ^, ^, In trials, replacing refined grains with whole grains improves blood cholesterol levels, glycemia, and possibly systemic inflammation. Similarly, fruits, bean, legumes, whole grains, and yogurt also contain some sugar or starch, yet are linked to metabolic and cardiovascular benefits as well as long-term weight maintenance. Such benefits appear related to a combination of factors, rather than any one characteristic. ^, Glycemic responses of carbohydrate-rich foods can be further mitigated by food order or mixed meals, such as by adding fats or proteins preceding or accompanying the meal, or even by a healthier long-term background diet. ^,

Several uncertainties exist. First, it remains unclear whether benefits of whole grains relate to displacing refined carbohydrates in the diet; or to additional health benefits of the germ (providing minerals, fatty acids, phytonutrients) and bran (providing fiber, minerals, phytonutrients). Dietary fiber, for instance, appears essential for gut microbial health and their bioactive metabolites (e.g., short-chain fatty acids). Second, the independent relevance of food processing is unclear. Most commercial “whole grain” breads, cereals, and crackers are made by finely milling, separating, and reconstituting the endosperm, germ, and bran. Fiber and nutrient contents are retained, but loss of intact food structure exposes the endosperm to rapid digestion by salivary and pancreatic enzymes, increasing its glycemic index compared with less finely milled whole grains (e.g., steel-cut oats, stone-ground bread). Third, the best simple metric for selecting healthier grain products is uncertain (see Fig . 29.3 ). One pragmatic rule-of-thumb is to choose foods containing at least 1 g of fiber for every 10 g of carbohydrate per serving (carbohydrate:fiber ratio <10:1), which implicitly balances the relative proportions of starch, sugar, whole grain, bran, and added fiber.

The relevance of personalization in carbohydrate responses is also of growing interest. Health effects of carbohydrate-rich foods can vary depending on insulin sensitivity, background diet, physical activity, or microbiome composition (see Microbiome and Personalization, below). Given the high rates of adiposity, prediabetes and DM, poor diet, and physical inactivity in most populations, reducing refined grains, starches, and added sugars should be a top priority for most individuals.

Fish

Regular fish consumption (1 to 2 servings/week) associates with modestly lower risk of CHD and stroke (see Fig . 29.2 ). Mechanistic, observational, and trial data suggest that fish may have stronger benefits for fatal CHD rather than nonfatal myocardial infarction (MI) or stroke, suggesting potential specificity for pathways of acute ischemia-induced ventricular arrhythmia. Because fish are a rich source of omega-3 fats, fish oil supplements have also been evaluated in trials (see n-3 PUFA, below). In observational studies, fish consumption associates with less ischemic stroke, but fish oil supplements have not influenced stroke in post-hoc analyses of CHD trials. Observational studies of atrial fibrillation and heart failure have produced mixed findings. Meta-analyses suggest no significant associations with incident DM.

Types of fish consumed and preparation methods may influence CVD effects. Greatest benefits may accrue from nonfried oily (dark meat) fish, that contain up to 10-fold more omega-3 fats than other types. Fish also contain other unsaturated fats, selenium, and vitamin D, which could provide benefit. Methylmercury in fish has no detectable influence on CVD events or incident hypertension. ^, Presence of persistent organic pollutants (e.g., dioxins, polychlorinated biphenyls) may partly reduce but does not appear to fully offset cardiometabolic benefits of fish intake. ^, In sum, the evidence supports recommendations to consume 1 to 2 servings of fish, especially oily fish, per week.

Red Meats

Prevalent guidelines recommend lean meats to lower dietary cholesterol and saturated fat. However, effects on cardiometabolic risk may be more complex, with other factors such as preservatives, heme iron, and cooking methods being more relevant. Available evidence suggests that processed meats (i.e., preserved with sodium or other additives, such as deli meats, sausage, hot dogs, bacon, etc.) increase risk of CVD, stroke, and DM; with unprocessed red meats having generally smaller associations, gram for gram, with these endpoints (see Fig . 29.2 ). ^, ^, Because unprocessed versus processed meats contain similar average amounts of total fat, saturated fat, and cholesterol, the stronger associations for processed meats suggest that sodium content—about 400% higher in processed meats—or other preservatives such as nitrites (hidden as “celery juice” in “nitrate-free” processed meats) may contribute. Heme iron, a risk factor for DM in animal experiments, gestational DM, and inborn errors of iron metabolism such as hemochromatosis, may explain the higher risk of DM seen with both processed and unprocessed red meats. Based on available evidence, processed meats should be avoided, and unprocessed meats eaten up to 1 to 2 servings/wk or less.

Poultry, Eggs

In prospective observational studies, poultry intake appears generally neutral for CVD and DM risk (see Fig . 29.2 ). ^, When combined with its relatively low content of bioactive nutrients, these findings suggest that poultry consumption has minimal cardiometabolic effects. Eggs appear similarly neutral for CVD risk in general populations (see Fig . 29.2 ). Overall evidence suggests minimal cardiometabolic effects of poultry intake or occasional egg intake (e.g., up to 2 to 3 per week); consistent with recent similar conclusions on dietary cholesterol (see Dietary Cholesterol, below).

Dairy

Cardiometabolic effects of dairy foods have generally been considered in relation to a limited set of nutrients: saturated fat, calcium, vitamin D. Yet, more diverse dairy nutrients and processing methods appear relevant. These include fermentation of yogurt and cheese (creating menaquinones), branch-chain, odd-chain, and medium-chain fatty acid contents, probiotics in yogurt, milk fat globule membrane content, and more. ^, In long-term cohorts, milk intake associates with lower risk of stroke, cheese with lower risk of CHD, and yogurt, cheese, and possibly butter with lower risk of DM (see Fig . 29.2 ). In randomized trials, milk or total dairy consumption increases lean mass and reduces body fat and waist circumference. In long-term observational studies, yogurt associates with relative weight loss, potentially related to probiotic-microbiome interactions.

In contrast, lower dairy fat content (reduced-fat or non-fat) does not consistently relate to lower cardiometabolic risk. Indeed, in a pooled analysis of 16 global cohorts, individuals with higher blood biomarkers of dairy fat consumption had significantly lower incidence of DM. Recent experimental evidence suggests that odd-chain saturated fats such as 15:0, fairly unique to dairy fat, reduce inflammation and dyslipidemia by binding to key metabolic regulators and repairing mitochondrial function. In sum, current evidence supports recommendations for modest dairy intake (2 to 3 servings/day), especially of yogurt and cheese, with more research needed to define the relevant active ingredients and health effects of dairy fat.

Plant Oils

While “industrial” or “refined” plant oils, especially those rich in n-6 PUFA, have been theorized to be harmful to health, little evidence supports this concern. The great majority of interventional and observational studies of total PUFA, n-6 PUFA, and MUFA have assessed industrialized plant oils like soybean, canola, and safflower oil. The cumulative evidence supports clear benefits of such plant oils ( see Macronutrients, below). Whether virgin versions of these oils would have even greater benefits, for example due to greater concentrations of phenolics and phytosterols, ^, is possible and requires investigation. Few studies have evaluated tropical oils (e.g., palm, coconut), beyond their known effects on blood lipids (raising both LDL-cholesterol and HDL-cholesterol).

Beverages

Sugar-Sweetened Beverages

Ecologic data, prospective cohorts, and trials provide convincing evidence that SSB intake increases adiposity. In the United States, calories consumed from beverages rapidly increased after the 1960s, doubling to 21.0% of all calories consumed by 2002—an increase of 222 daily kcal per person. This was largely due to increased SSBs (sodas, energy drinks, sweetened ice teas, fruit drinks), although that peak has been followed by gradual declines in SSB consumption thereafter. ^, Per serving, SSBs more strongly associate with long-term weight gain than nearly any other dietary factor. Randomized trials confirm that reducing SSBs decreases weight gain and fat accumulation. ^, Calories in liquid form, compared with solid foods, appear to be less satiating and increase total calories consumed. SSB intake also associates with significantly higher incidence of CVD and especially DM (see Fig . 29.3 , ^, likely related to weight gain and other additional harms of the high sugar and glycemic load. Given clear evidence for harms, and multiple alternatives (e.g., water, seltzer, unsweetened tea, diet soda, milk), SSBs should be largely eliminated from the diet.

Alternative sweeteners can be artificial (e.g., saccharin, aspartame) or naturally low-calorie (e.g., stevia). Based on observational studies and clinical trials, alternative sweeteners are better for cardiometabolic health than added sugars. ^, Yet, alternative sweeteners may not be completely benign: animal experiments and limited human data suggest influences on brain reward, taste perception, oral-gastrointestinal taste receptors, glucose-insulin and energy homeostasis, metabolic hormones, and the gut microbiome. For instance, if a child becomes accustomed to intense sweet tastes, will that reduce attractiveness of naturally sweet foods such as apples or carrots? In sum, alternative sweeteners can be a useful bridge to eliminate SSBs, but should not be considered innocuous; and subsequent shifts to non-sweetened drinks (e.g., seltzer, tea) should be encouraged.

Milk

See Dairy.

Coffee, Tea

While coffee and tea provide caffeine, these plant extracts—derived from beans and leaves—contain other bioactive compounds. Unrelated to caffeine content, frequent coffee intake (e.g., 3 to 4 cups/day) associates with less insulin resistance, DM, CVD, and in a few studies, heart failure (see Fig. 29.3 ). ^, Yet, clear physiologic benefits have not been established to support these observations. Acutely, caffeinated coffee worsens BP, insulin resistance, and glucose intolerance; ^, longer-term, habitual coffee intake does not affect BP, endothelial function, nor glucose metabolism, suggesting tachyphylaxis and/or other partly offsetting factors. ^, In mendelian randomization analyses, genetic variants linked to coffee intake do not associate with CHD, stroke, or DM. Based on lack of physiologic benefits in trials or lower risk in mendelian randomization studies, the cardiometabolic benefits of coffee intake should still be considered questionable.

Like coffee, frequent tea drinking (e.g., 3+ cups/day) associates with lower CVD and DM, with borderline statistical significance (see Fig. 29.3 ). ^, In randomized trials, certain types of tea modestly lower BP (green, black, sour), LDL-cholesterol (green), and fasting glucose (green, sour). ^, In contrast to coffee, these physiologic effects support potential causal cardiometabolic benefits of tea, but larger and more rigorous studies are still needed.

Alcohol

See Chapter 84.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here