Effect of Macronutrient Composition on Short-Term Food Intake and Weight Loss^,

Abstract

The purpose of this review is to describe the role of macronutrient composition on the suppression of short-term food intake (FI) and weight loss. The effects of macronutrient composition on short-term FI will be reviewed first, followed by a brief examination of longer-term clinical trials that vary in effects of dietary macronutrient composition on weight loss. The objectives were: 1) to examine the effect of macronutrient composition on the suppression of short-term FI, 2) to determine whether some macronutrient sources suppress FI beyond their provision of energy, 3) to assess the combined effects of macronutrients on FI and glycemic response, and 4) to determine whether knowledge of the effect of macronutrients on short-term FI has led to greater success in spontaneous weight loss, adherence to energy-restricted diets, and better weight maintenance after weight loss. Although knowledge of macronutrient composition on short-term FI regulation has advanced our understanding of the role of diet composition on energy balance, it has yet to lead to greater success in long-term weight loss and weight maintenance. It is clear from this review that many approaches based on manipulating dietary macronutrient composition can help people lose weight as long as they follow the diets. However, only by evaluating the interaction between the physiologic systems that govern FI and body weight may the benefits of dietary macronutrient composition be fully realized.

Introduction

Despite advances over the past few decades in obesity prevention, management, and treatment, 1 in 3 North American adults is obese. This highly visible phenomenon has become a major characteristic of many westernized populations, affects all major demographic and socioeconomic groups, and continues to increase over a relatively short time frame (1). It is associated with an elevated risk of cardiovascular diseases, diabetes, liver disease, and some types of cancer, and it predisposes an entire generation to increased lifelong morbidity and mortality. Obesity occurs when energy intake habitually exceeds energy expenditure. Its etiology, however, is multifactorial, involving an interaction between physiologic, behavioral, and environmental factors that collectively regulate energy homeostasis.

An understanding of macronutrient composition—the relative amounts of protein, fat, and carbohydrate—on short-term (between 30 min and 4–5 h) food intake (FI)³ may assist in the design of longer trials aimed at achieving healthier body weights. What remains in debate, however, is the most effective dietary approach for treating obesity. There is considerable confusion, for example, about the dietary mechanism that supports long-term weight reduction and the maintenance of reduced body weight. For this reason, a better understanding of the short-term effects of macronutrient composition may offer approaches and insights into the design of strategies and meal plans for successful long-term weight reduction.

This review will explore the effect of macronutrient composition on short-term FI and then examine the effect of longer-term clinical trials of varying dietary macronutrient composition on weight loss. To evaluate the merits of macronutrient composition on FI regulation and weight loss, we consider the following 4 objectives: 1) to examine the effect of macronutrient composition on the suppression of short-term FI, 2) to determine whether some macronutrient sources suppress FI beyond their provision of energy, 3) to assess the combined effects of macronutrients on FI and glycemic response, and 4) to determine whether knowledge of the effect of macronutrients on short-term FI has led to greater success in spontaneous weight loss, adherence to energy-restricted diets, and better weight maintenance after weight loss.

Effect of Macronutrient Composition on Short-Term FI

Food consumption triggers a multitude of neural and hormonal signals, originating from the periphery and interacting with the central nervous system, that regulate FI according to energy requirements. In response to the macronutrient composition in one’s diet, the body releases hormones—gastrointestinal, pancreatic, and adipose-derived—that ultimately signal the hypothalamus to contribute to the cessation of eating. This powerful feedback system is sensitive to the macronutrient composition of the diet and thus can easily be exploited to alter FI.

Satiety and satiation are 2 important principles in the study of FI control. Satiety is defined as the state of eating cessation, and it delays the initiation of subsequent meals. Satiation, on the other hand, is the process of feeling full during the course of eating, a form of intrameal satiety that is assessed by measuring FI (2). Satiety can be evaluated by perceived sensations, gastrointestinal hormonal responses, and eating initiation. Perceived satiety sensations, which are assessed by visual analog scales consisting of 4 questions on hunger, fullness, desire to eat, and prospective food consumption, are associated with the hormonal responses and predict both eating initiation and subsequent meal size.

Data from both animal and human experiments suggest that macronutrients exhibit a hierarchical effect on short-term FI suppression and satiety. In young adults, for example, a 240-kcal high-protein (77%), high-carbohydrate (84%), or high-fat (58% of energy from fat) snack enhanced satiety and delayed dinner requests by 60, 34, and 25 min, respectively (3). Snack composition had no impact on FI at dinner (3), but the effect of macronutrients, which was consistent with previous findings, showed that protein promotes satiety to a greater extent than carbohydrate or fat (4, 5). In another similar study in adults, although the consumption of each of 3 macronutrients 1 h before lunch reduced FI, there were greater reductions in FI after carbohydrate than after fat intake (6). Total daily energy intake (kcal meal + kcal treatment) was higher after the fat preload with no substantial differences between carbohydrate and protein compared with the control (6). The macronutrient dose is another factor that affects FI. In restrained adults, for example, a high-fat preload consumed before a meal (357 kcal, 20% carbohydrate, 65% fat) did not suppress FI as well as the low-fat treatment (357 kcal, 81% carbohydrate, 5% fat) (7).

These comparative short-term FI trials are based on the selection of representative macronutrients that are supposed to be indicative of all sources within their class. The satiety index of common foods developed by Holt et al. (8) showed that satiety scores varied both within and in comparison to other food groupings. Many sources within the fruit, carbohydrate- and protein-rich, breakfast cereals, snacks, confectionary, and bakery product groups were satiating. Indeed, among those who regularly study short-term FI regulation, the macronutrient source is understood to be an important determinant of FI. However, it is often overlooked in comparative FI trials, an omission that contributes to confusion about the effect of macronutrient composition on FI regulation.

Fats (Lipids)

Lipids are usually referred to as fats and oils. The lipid family includes TGs, phospholipids, and sterols. Lipids contain carbon, hydrogen, and oxygen and are not water-soluble. Among the lipids, TGs are the most common type found in foods and in the body. The functional units of TGs, or FAs, are characterized according to their chain length, degree of saturation or unsaturation, and shape. In addition, whereas fats provide the body with a major source of energy, their influence on gut hormones, including cholecystokinin (CCK), glucagon-like peptide 1 (GLP-1), and gastric inhibitory polypeptide (GIP) (9), slows gastric emptying and meal absorption (10) and reduces short-term FI (6). As a critical part of the alimentary canal, the upper small intestine is responsible for nutrient digestion and absorption. The presence of FAs in the upper small intestine triggers a gut-brain-liver neural axis that plays a role in the control of glucose homeostasis, delays the return of hunger, and inhibits FI (11).

Long-chain TGs (LCTs) greatly increase CCK and peptide tyrosine tyrosine (PYY). Yet, medium-chain TGs (MCTs), which contain saturated FAs with carbon chain lengths of 6–10 atoms, are more satiating. MCTs are absorbed into the portal system and are rapidly oxidized by the liver. Long-chain TG–induced chylomicrons, on the other hand, bypass the liver and are more likely to promote lipid accumulation.

Enhanced oxidation of FAs contributes to lower FI (12). Thus, MCTs may contribute to spontaneous energy-intake reduction and greater weight loss than LCTs (13). For example, the addition of 38 g of fat as monounsaturated LCTs, saturated LCTs, or MCTs to high-carbohydrate breakfasts delayed the request for subsequent FI compared with the low-fat breakfast in healthy adults (14). Furthermore, both high-LCT breakfasts led to the highest energy intake at lunch and dinner, but it was the MCT breakfast that resulted in the lowest FI, suggesting that, despite the high energy density of high-fat meals, the high-MCT diet enhances satiation and the suppression of FI.

An understanding of the degree of saturation or unsaturation and the number of double bonds in FAs may help to identify fat sources that suppress FI (15). Increasing the number of double bonds within 18-carbon FAs associates with lower FI in humans (15), but the effect is not linear and it is unknown whether this relation persists with other FA chain lengths. Upper intestine infusion of intralipids, containing 54% linoleic acid (18:2) and 29% oleic acid (18:1), and linoleic acid increased CCK concentrations and reduced FI at 90 min compared with a saline infusion (15). However, neither oleic acid (18:1) nor stearic acid (18:0) suppressed FI.

Compared with TGs, diacylglycerols (DAGs) are found in low doses in the diets of humans. However, a novel DAG oil containing 80% DAG lowered subjective appetite to a greater extent than did TG oil of similar FA composition made from 45% rapeseed and 55% sunflower oil (16). In addition to lower appetite, it has been reported that DAG oil contributes to increased FA oxidation (16).

Carbohydrates

Carbohydrates are the major food sources in the human diet. Depending on their digestibility characteristics, carbohydrates elicit distinct physiologic responses. Dietary fiber, sugars, and starches (including resistant starch) are the 3 major categories of carbohydrate. Dietary fibers, including soluble and insoluble, benefit human health through their effects on satiety, glycemia, gut microflora composition, and lipid profile (17). Sugars include monosaccharides, such as glucose, fructose, and galactose, and disaccharides, such as sucrose, maltose, and lactose. Dietary polysaccharides consist of oligosaccharides, such as maltodextrins and polydextrose; starches, such as amylose and amylopectin; and the nonstarch category, such as cellulose and pectins.

Sugars and starches influence satiety and short-term FI primarily through their effect on blood glucose and insulin responses. The effect of dietary glucose on short-term FI suppression is consistent with the glucostatic theory of FI control in which higher blood glucose, resulting from meal ingestion, leads to satiety and subsequently to the termination of eating (18). Conversely, the effect of dietary fibers more likely occurs via gastric distension, secretion of incretins, and the production of hydrogen, methane, and SCFAs. Indigestible carbohydrates, such as resistant starches, suppress appetite and FI through fermentation in the colon (19). The production of hydrogen, methane, and SCFAs modulate the metabolism of glucose and lipids as well (20).

Reduced FI has been reported after consuming carbohydrates that produce both low (21) and high (22) glycemic responses. However, short-term FI is often reduced more at 1–2 h after intake of rapidly digestible, high-glycemic carbohydrates and at 2–6 h after slowly digestible, low-glycemic carbohydrates. Although the glycemic index (GI) was proposed and initially used as a classification for the blood glucose–increasing potential of carbohydrate foods, it is not useful for predicting the effect of carbohydrate on satiety or FI within mixed meals (23). For example, baked potatoes (GI of 117), containing 240 kcal and 50 g available carbohydrate, lowered the desire to eat to a greater extent than did pasta (GI of 108) and brown rice (GI of 132) (24).

The effect of sugars, including glucose, fructose, sucrose (25, 26), high-fructose corn syrup (22), and lactose (27), on satiety as consumed before a meal or as part of meals has been reported in short-term clinical studies. Like other macronutrients, the magnitude effect of sugars may be altered by the source, matrix, dose, and the intermeal interval. Several short-term studies (22, 25, 26) showed that sugars suppress FI and, like other macronutrient sources, activate the normal cascade of hormonal and metabolic signals in the suppression of FI. For example, in healthy young men, both sucrose and high-fructose corn syrup drinks consumed before a meal similarly reduced subjective appetite, FI, and the hunger hormone ghrelin at a test meal 80 min later (22). Lower FI was associated with a greater blood glucose response from drinks containing higher glucose to fructose ratios.

Proteins

The effect of dietary proteins on satiety and FI depends on the source, digestion, and absorption characteristics and amino acid profile. Proteins that are rapidly digested and absorbed, such as whey and soy (28), suppress short-term FI (at 1 h) to a greater extent than do more slowly absorbed proteins, such as casein (29) and egg albumin (30). However, these effects of protein source on FI are time dependent. For example, ad libitum FI after whey, soy protein, and casein did not differ when FI was assessed 3 h later, despite difference in appetite regulatory hormone responses (31, 32). The hydrolyzed form of protein, which is more digestible, typically suppresses FI more than its intact protein form (33). Finally, the amino acid profile of the protein, such as those that contain BCAAs and bioactive peptides, contributes to the enhancement of FI suppression. This occurs because of the release of several gut peptides involved in the satiety cascade, including insulin, CCK, GLP-1, PYY, and GIP (30).

Do Some Macronutrient Sources Suppress FI Beyond Their Provision of Energy?

Several lines of evidence suggest that many macronutrient sources suppress FI to a greater extent than their calorie content. First, relatively low doses of protein suppress FI. Caloric compensation scores, a measure of the efficacy of a caloric treatment to suppress FI compared with a noncaloric treatment, of 5–40 g of whey protein drinks consumed 30 min before an ad libitum meal were up to 50% greater than could have been predicted from their energy content alone (34). Dietary protein also requires more energy for digestion and metabolism and facilitates weight loss from body fat stores while preserving resting energy expenditure (35).

Second, in healthy-weight boys, a 50-g glucose or whey-protein solution provided 30 or 60 min, respectively, before an ad libitum meal suppressed next-meal FI beyond its provision of energy (36). This may, however, be an artifact of a more highly sensitive regulatory system in children, an issue that merits further investigation.

Third, among the carbohydrate sources, the addition of fiber to a meal reduces caloric intake. Higher intake of dietary fiber is associated with lower abdominal obesity in older (37) and obese (38) adults as well as in individuals with type 2 diabetes (39). The mechanism of action of soluble fiber intake on lower body weight is related to its effect on FI by forming gels in the gastrointestinal tract, thereby delaying digestion and absorption of other macronutrients and through gastric distension. Insoluble fiber intake contributes to SCFA production and secretion of the incretins GLP-1 and GIP, which enhances satiety and reduces FI. Both soluble and insoluble fibers may contribute to weight loss. A 12-y prospective cohort study indicated that women with the greatest increase in dietary fiber intake from whole-grain products had lower body weight and maintained it for longer than those who consumed less fiber (40). High-carbohydrate diets can alter gut microbial composition and fermentation activities, which may link to improved body energy regulation (41). A 24-wk clinical study showed that an increase in fecal Bacteroides numbers after high-carbohydrate diets was associated with decreases in body weight, BMI, and waist circumference in 88 individuals with metabolic syndrome (41).

Fourth, the effect of several fat sources reduces FI more than the calories they provide. Because of the different rates of absorption and utilization, DAGs enhance satiety and decrease FI suppression more than TGs (16). For normal-weight, overweight, and obese individuals, the addition to yogurt of 2, 4, and 6 g of a fractionated palm (40%) and oat (2.5%) oil emulsion decreased FI 4 h later and subsequent energy intake for up to 36 h (42). Exposure of the ileum to lipids increases the secretion of GLP-1, and activation of the ileal brake (43), a powerful feedback mechanism that delays nutrient transit time in the small intestine, may provide the mechanism of action.

Finally, all macronutrients contribute to the release of a multitude of satiety hormones; there is substantial overlap, and thus there are no distinct macronutrient-specific biomarkers. For example, carbohydrate and protein both increase insulin and GLP-1. Fat is a potent CCK secretagogue, although protein and carbohydrate contribute to its release. Protein exerts a sustained effect on insulin, GLP-1, and PYY (44), and all macronutrient sources contribute to a decrease in postprandial ghrelin concentrations.

Combined Effects of Macronutrients on FI and Glycemic Response

Epidemiologic studies of dietary patterns suggest that macronutrients may act additively or synergistically to confer benefits. The combination of high protein and fiber from low-fat dairy products, whole grains, fruit, and vegetables was associated with lower body weights and improved metabolic regulation (45). An understanding into the combined effects of macronutrient sources on FI suppression and glycemic regulation may offer insight into the design of long-term weight-loss trials and may contribute to greater weight-loss success and improved metabolic health.

The coingestion of fiber with fat, for example, prolongs gastrointestinal transit time and decreases fat digestion, thereby leading to greater satiety and reduced energy intake than does fat alone (46). The addition of protein (47) or fat (48, 52) to carbohydrate contributes to reduced appetite and glycemic response by reducing gastric emptying (50) and by increasing incretins (51) and insulin (49). Although blood glucose was reduced by coingestion of fat and/or protein with carbohydrate, the addition of protein (25 g as tuna fish) to a high-GI carbohydrate (25 g as potato) acted synergistically to stimulate insulin secretion. When fat (25 g as margarine) was added to a combination of carbohydrate + protein, however, it did not increase insulin secretion (49). Although this finding is in line with the insulin-secreting effects of protein, the effects on FI are unknown.

The addition of fat or protein to a carbohydrate-rich meal lowers the blood glucose response compared with meals containing carbohydrate alone (34, 47, 48). Although the exact mechanisms of action are unclear, it has been suggested that a reduced gastric emptying rate (50) and the increased secretion or concentration of insulin (49) and of incretins, such as GLP-1 and GIP (51), may explain the hypoglycemic effect of combined macronutrients.

The effect on satiety and on glycemic and insulin responses of adding fat to carbohydrate depends on the fat source and dose and the food matrix. Supplementation of white bread with solid fat (5–40 g canola margarine) reduced blood glucose concentrations in a dose-dependent, but nonlinear manner (53). However, adding corn oil (5–30 g) to a glucose beverage reduced blood glucose in a linear manner (54). Compared with monounsaturated fats (80 g olive oil), saturated fats (100 g butter) added to carbohydrates (50 g as mashed potatoes) reduced blood glucose and insulin in individuals with type 2 diabetes (55).

Effect of Macronutrient Composition on Weight Loss

Many macronutrient sources contribute to the suppression of FI. Included in a habitual diet, macronutrients can lead to spontaneous reductions in energy intake, improved adherence to energy-restricted diets, and successful short-term weight loss. This suggests that several dietary approaches can help people lose weight.

Fats.

Nutritional guidelines advise individuals to focus on lower-fat, high-carbohydrate diets (56) to lose and to maintain weight. Compliance, however, is the major barrier to following a long-term low-fat, high-carbohydrate diet, and weight regain through a return to a habitual diet is common. Adherence to energy-restricted diets may be enhanced when fat intake is increased (57), but low adherence remains a problem for all energy-restricted diets (58).

The Acceptable Macronutrient Distribution Range (AMDR) for fat is between 20% and 35%. Although low-fat diets typically fall below this recommendation, some may contain up to 35% fat. A recent meta-analysis of studies examining various approaches to body weight management (59) found that fat diets near the higher end of the AMDR and moderate-fat diets (>35%) were associated with greater compliance and perhaps great weight-loss success. Other diets—low- carbohydrate (13–45% of energy from carbohydrate), high-protein, and Mediterranean (24–35% fat, high in unsaturated fats)—were more effective in achieving weight loss than low-fat, low-GI diets (60). Compared with a lower-fat diet (20% of energy from fat), an energy-restricted, moderate-fat diet (30% of energy from fat) resulted in similar weight loss at 7 mo; however, at the end of 14 mo only the group following the moderate-fat diet maintained their weight loss (57).

The 24-month Dietary Intervention Randomized Controlled Trial showed that weight maintenance after weight loss was better when dietary fat was >30%. The results were best after the Mediterranean and low-carbohydrate diets, which contained 33% and 39% of energy from fat, respectively (61). These benefits may be due to the fat content and to the composition of the diet. Diets rich in unsaturated fats, including MUFAs and PUFAs, compared with diets high in SFAs, are more metabolically beneficial for weight loss due to the higher thermogenesis and fat oxidation (62).

Several functional fats may also contribute to successful weight loss when included as part of an energy-restricted meal plan. In obese middle-aged men, conjugated linoleic acid supplementation (4 g/d) for 4 wk reduced sagittal abdominal diameter (abdominal fat) and metabolic variables, including serum cholesterol, LDL cholesterol, TGs, FFAs, glucose, and insulin (63). Olibra (DSM), a fat emulsion of fractionated palm (40%) and oat (2.5%) oils, suppressed FI and contributed to weight loss when included as part of an energy-restricted meal plan and lifestyle intervention (42). Olibra (2.1 g) given twice daily to obese and overweight individuals in a longer-term study (>12 wk) resulted in sustained weight loss, although the reduction was similar to results seen with milk fat (64). The effect of MCTs is of interest because of their absorption characteristics and preferential oxidation in the liver. MCT oil (18–24 g/d) for 16 wk resulted in greater weight loss than olive oil when included as part of a weight-loss program (65).

Carbohydrates.

Both quantitative and qualitative aspects of carbohydrates may have an impact on the management of obesity and obesity-related conditions. The GI, which ranks carbohydrate-containing foods on the basis of their postprandial glycemic-increasing effects, has received considerable attention. Its proponents suggest that low-GI diets have favorable effects on weight loss and other biomarkers of metabolic health. Critics of the GI state that the concept is inaccurate, imprecise, and ignores other potentially important components in the food and thus does not confer additional benefits beyond current dietary guidance (66).

Notwithstanding, low-GI diets enhance weight control by contributing to satiety, sparing carbohydrate in favor of increasing fat oxidation, and attenuating postprandial insulin secretion (67). A systematic review and meta-analysis of 14 long-term randomized clinical trials found that low-GI diets were associated with improvements in the proinflammatory biomarker C-reactive protein but associated with significant losses in fat-free mass (68), an effect that may contribute to weight regain. In a large European study, overweight adults followed ad libitum high- or low-protein diets paired with either a high- or low-GI diet for 26 wk. The results showed that participants consuming the low-GI, higher-protein diet gained less weight than those following the high-GI diet (69). However, the absolute differences in weight regain between the treatment diets and the control were small and not clinically significant (∼1 kg).

Proteins.

Protein enhances dietary adherence by increasing satiation and satiety. Because of greater long-term adherence and follow-up, high-protein, energy-restricted diets promote weight loss, primarily from adipose tissue, while preserving lean body mass (70). In addition, high-protein diets preserve resting energy expenditure during weight loss, an essential physiologic adaptation that is required for weight maintenance (71). The long-term effectiveness of high-protein diets was examined by using both ad libitum and energy-restricted diets. In the ad libitum studies, a high-protein diet (25–30% of total energy), compared with high-carbohydrate (72, 73) or high-fat (74, 75) diets providing 10–20% protein, showed more favorable effects on body weight and fat loss and maintenance. However, because of poor compliance and high dropout rates (76–78), the effect of energy-restricted diets was less consistent. Both the medium-term (8–16 wk) (79–81) and long-term (12 mo) (76, 82) studies—comparing high-protein diets with high-carbohydrate, low-fat diets—reported nonsignificant differences in weight loss.

Conclusions

In summary, short-term trials have contributed to our understanding of the physiologic functionality of macronutrients on short-term FI. A satiety hierarchy is not supported by this review, because many macronutrient sources suppress FI. Thus, many dietary approaches can help people lose weight, if they are followed. In addition, because differences in weight loss are often small between diets of varying macronutrient composition, and less than predicted, the focus should be on finding a dietary regimen that an individual can follow. Future investigations must be predicated on an understanding of the physiologic systems that govern FI and energy balance. Only then may the merits of dietary macronutrient composition on weight loss be fully realized.