© Springer Science+Business Media New York 2014
Robert F. Kushner and Daniel H. Bessesen (eds.)Treatment of the Obese Patient10.1007/978-1-4939-1203-2_1414. Dietary Modification as a Weight Management Strategy
(1)
Public Health Nutrition, Department of Nutrition, The University of Tennessee-Knoxville, 1215 W. Cumberland Avenue, 229 Jessie Harris Building, Knoxville, TN 37996-1920, USA
(2)
Division of Behavioral Medicine and Clinical Psychology, Cincinnati Children’s Hospital Medical Center, MLC 3015, 3333 Burnet Avenue, Cincinnati, OH 45229, USA
Introduction
Dietary prescriptions for weight management alter the energy intake side of the energy balance equation. During weight loss, the goal of all dietary prescriptions is to reduce energy intake so that a meaningful energy deficit occurs, allowing a rate of weight loss that is clinically significant. In addition to assisting with reducing energy intake, the ideal dietary prescription for obesity treatment should also improve the cardiometabolic profile, enrich diet quality, and enhance appetite regulation. While weight loss itself greatly improves glucose and lipid parameters, a dietary prescription can also enhance these physiological outcomes, and ideally a dietary prescription for obesity treatment can augment cardiometabolic outcomes beyond what is achieved with weight loss alone [1]. For enriching diet quality, the prescription should assist with meeting recommendations that are provided in the Dietary Guidelines for Americans, 2010 [2]. Finally, as long-term adherence to a dietary prescription continues to be a challenge, especially during weight loss maintenance, a prescription that can enhance satiation and/or satiety and minimize hunger, either through physiological or cognitive factors, could increase the ease of continuing the prescription over an extended period of time [3].
The purpose of this chapter is to provide an overview of dietary prescriptions that have been investigated for obesity treatment in adults. Only studies examining dietary prescriptions within randomized controlled trials (RCTs) for weight loss of at least 3 months in length, with participants randomized to receive the same diet throughout the length of the trial, are included. This chapter reviews research on different dietary prescriptions by organizing the prescriptions into three main categories: energy-focused prescriptions, macronutrient-focused prescriptions, and dietary pattern-focused prescriptions. Additionally, factors that have been investigated in relation to energy-focused prescriptions, such as meal replacements and consumption patterning (i.e., frequency and timing of consumption) are examined. Outcomes for weight loss, as well as cardiometabolic parameters, diet quality, and appetite regulation, if available, are reported.
Energy-Focused Prescriptions
Energy-focused dietary prescriptions for obesity treatment are based upon the concept of thermodynamics, such that when a negative energy deficit of 3,500 kilocalories (kcal) occurs, approximately 1 pound (lb.) of weight is lost [4]. Energy-focused prescriptions generally function under the assumption that the magnitude of the effect of the diet on achieving a negative energy balance is independent of the composition of the diet; however, recent research does suggest that composition of the diet may influence the degree of negative energy balance achieved via differences in energy expenditure [5, 6]. Due to its focus on energy, dietary goals in these types of prescriptions always include a daily energy limit, but may also include other dietary goals that are believed to be helpful in reducing energy intake.
Low-Calorie Diet
A low-calorie diet (LCD) is usually greater than 800 kcal per day, and typically ranges from 1,200 to 1,600 kcal per day [7]. The LCD is designed to induce an energy deficit of 500 to 1,000 kcal per day, producing a weight loss of 1–2 lb. per week [7]. An LCD is most commonly prescribed using a traditional food regime [7, 8]. Fat restriction may be combined with an LCD to assist with reducing energy intake [8]. In 1998, the National Heart, Lung, and Blood Institute (NHLBI) published evidence-based clinical guidelines for the treatment of adult obesity [9]. These guidelines recommend an LCD, with energy prescriptions of 1,000–1,200 kcal per day for women and 1,200–1,500 kcal per day for men. The NHLBI guidelines also recommend restricting fat intake to no more than 30 % of daily energy intake, with the restriction designed to assist with reducing energy intake, thereby aiding with meeting the reduced-energy goal.
The Diabetes Prevention Program (DPP) provides an example of weight, cardiometabolic, and dietary outcomes commonly found when an LCD with fat restriction is prescribed [10]. Within the DPP, overweight and obese individuals with elevated fasting glucose levels who were randomly assigned to the lifestyle intervention condition received an LCD (ranging from 1,200 to 1,800 kcal per day depending on initial body weight), with a fat restriction of 25 % energy from fat. This intervention also included a moderate- to vigorous-intensity physical activity goal of 150 min per week, and provided a behavioral intervention to assist with changing dietary and physical activity behaviors to aid with achieving a weight loss goal of 7 % of initial body weight. The other two conditions that participants were randomized to were a medication (metformin) or placebo intervention. In DPP, participants were followed for an average of 2.8 years, and in the lifestyle intervention, the mean weight loss was −5.6 kilograms (kg), which was significantly greater than the two other conditions, medication, or placebo, that had not been prescribed the LCD. The pattern of weight loss and maintenance of weight loss seen in the lifestyle intervention showed the greatest degree of weight loss occurring at 6 months (approximately −7.0 kg), which was maintained for another 6 months, with weight regain occurring over longer follow-up. However, at 4-year follow-up, the lifestyle intervention still maintained a weight loss of almost −4.0 kg.
For cardiometabolic outcomes, fasting glucose and glycosylated hemoglobin (HbA1) significantly decreased more so in the lifestyle intervention as compared to the medication and placebo intervention [10]. Over the mean 2.8-year follow-up, the lifestyle intervention showed a significantly reduced prevalence of hypertension and dyslipidemia as compared to the medication and placebo interventions. Additionally, significantly fewer lifestyle participants required medication for treatment of elevated triglycerides or high levels of low-density lipoprotein (LDL)-cholesterol than the other conditions. However, as weight loss alone enhances cardiovascular outcomes and the lifestyle intervention had the greatest weight loss, it is not clear if the enhanced cardiovascular outcomes found in the lifestyle intervention are a consequence of the LCD or other factors, such as greater weight loss, found within the lifestyle intervention.
Analyses of 1-year changes in dietary intake showed that the lifestyle intervention had significantly greater reductions of energy (−452 kcal per day) and percent energy from fat (−6.6 %) intake than the medication and placebo interventions [11]. Additionally, the lifestyle intervention showed significantly greater increases in intake of dietary fiber and weekly intake of servings from fruits and significantly greater reductions in weekly intake of servings from red meat and sweets than the other two conditions [11].
Meal Replacements
Meal replacements are portion-controlled products, usually liquid shakes and bars, containing a known amount of energy and macronutrient content, that are considered to be a useful strategy to reduce problematic food choices and decrease challenges with meal planning when engaging in an LCD [8]. Additionally, as adherence to any dietary prescription requires consuming foods of an appropriate portion size, meal replacements may enhance dietary adherence since they reduce the burden of weighing and measuring all foods consumed due to their portion-controlled quality [7, 12–15]. The most commonly investigated dietary prescription that has been examined with meal replacements is a partial meal replacement (PMR) plan, which prescribes two portioned-controlled, vitamin/mineral fortified meal replacements per day, with a reduced calorie meal and snack comprised of conventional foods [7, 16].
For weight loss outcomes, a meta-analysis of the effect of meal replacements in comparison to an isocaloric LCD composed of conventional foods found that at the 3- and 12-month follow-up, the meal replacement conditions produced more than 2 kg of greater weight loss, which was significant, than the comparison conditions [7]. The Look AHEAD (Action for Health in Diabetes) trial, which was designed to investigate the impact of a lifestyle intervention that produces a minimum weight loss of 7 % on cardiovascular disease morbidity and mortality in individuals with type 2 diabetes, also prescribed a PMR plan in the lifestyle intervention [16]. The lifestyle intervention lost significantly more percent body weight than a support and education control condition at 1-year (−8.6 ± 6.9 % vs. 0.7 ± 4.8 %, p < 0.001) [17]. One-year outcomes also found that the number of meal replacements consumed for the year was associated with weight loss at 52 weeks (r = 0.30, p < 0.001), and participants in the highest quartile for meal replacement use had four times greater odds of reaching the 7 % weight loss goal [18].
For cardiometabolic outcomes, the PMR plan shows improvements in glucose, cholesterol, and triglycerides, but improvements are not greater than those seen with isocaloric diets composed of conventional foods [7]. Little research has been conducted examining the impact a PMR plan has on diet quality. One RCT did compare a PMR plan to an isocaloric plan using conventional foods and found similar dietary outcomes between the two conditions. The only difference between the conditions was in regard to micronutrient intake, with the PMR plan having a significantly greater intake [19].
Pattern of Consumption
Within an LCD, differing patterns of when energy is consumed have been examined in two RCTs. One pattern of consumption that has been examined is eating frequency. The amount of energy consumed can be spread out into a few meals per day, or into multiple smaller meals and snacks per day. Only one, small, 6-month RCT has examined the influence of eating frequency on weight loss within an LCD, in which participants were randomized into a condition in which participants consumed three meals per day or into a grazing condition in which participants consumed three meals and approximately three snacks per day [20]. While significant reductions in body mass index (BMI) occurred, there were no differences between the conditions. However, self-reported hunger significantly decreased in the condition with more frequent eating bouts, but hunger did not change in the condition consuming only three meals.
The effects of meal timing on weight loss have also been examined in one, 12-week RCT [21]. In this investigation, a prescription of 1,400 kcal per day was spread into three meals, with participants randomized into conditions in which either 1,200 or 700 kcal were consumed by the completion of lunch. The remaining 200 or 700 kcal were consumed at dinner. Results found the condition that consumed more energy earlier in the day lost significantly more weight (−8.7 ± 1.4 kg vs. −3.6 ± 1.5 kg, p < 0.0001). Significantly, greater reductions in triglycerides, cholesterol, and glucose, and increases in high-density lipoprotein (HDL)-cholesterol occurred in the condition that consumed more energy earlier in the day. However, it is not clear if the difference between the conditions on cardiometabolic outcomes is due to the timing of energy intake or the difference in achieved weight loss. Self-reported hunger was also significantly lower in the condition that consumed the majority of their energy intake earlier, rather than later, in the day.
Very-Low-Calorie Diet
A very-low-calorie diet (VLCD) provides ≤800 kcal per day [8, 22, 23]. VLCDs were developed to enhance weight loss by creating a greater energy deficit than what occurs with the LCD. The VLCD is designed to preserve lean body mass, thus large amounts of dietary protein, usually 70–100 grams (g) per day or 0.8–1.5 g protein per kg of ideal body weight, are prescribed [8, 22]. One recommended source of protein for VLCDs is from a milk-, soy-, or egg-based powder, which is mixed with water and consumed as a beverage [22]. These powders also include 100 % of the recommended daily allowance for essential vitamins and minerals [8, 22]. Another recommended source of protein is from lean meat, fish, and fowl, and this form of VLCD is called a protein-sparing modified fast, which must be supplemented with a multivitamin and 2–3 g per day of potassium [22]. VLCDs require consumption of 2 liters (L) per day of noncaloric fluids [22]. VLCDs are considered to be appropriate only for those with a BMI ≥ 30 kg/m2, and are increasingly used with individuals prior to having bariatric surgery to reduce overall surgical risks in those with severe obesity [8, 22].
One meta-analysis compared VLCDs to LCDs and examined short-term (varied from 8 to 50 weeks) and long-term (varied from 18 to 66 months) weight loss outcomes [22]. Results indicated that in the short-term, weight loss favored the VLCD (−16.1 ± 1.6 % vs. −9.7 ± 2.4 % of initial weight; p < 0.001), but long-term outcomes were similar between the two diets (−6.3 ± 3.2 % vs. −5.0 ± 4.0 % of initial weight; p > 0.2). The lack of difference between the two diets during the longer-term follow-up was due to greater weight regain occurring in the VLCD conditions. Thus, this meta-analysis suggests that in the long term, the VLCD does not enhance weight loss outcomes as compared to diets of higher energy prescriptions (LCD).
Improvements in glycemic control and blood lipids are also found with VLCDs; however, the degree of improvement appears to be a function of the amount of weight loss rather than any particular factor of the diet [22, 23]. Research on the effect of VLCDs on appetite regulation is very limited, with most research focused on the effect of VLCDs on binge eating, particularly in those with Binge Eating Disorder, and reported outcomes are mixed [23].
Summary of Energy-Focused Prescriptions
Energy-focused dietary prescriptions are successful at producing weight loss and improving cardiometabolic outcomes. Within these types of weight management dietary prescriptions, the improvements in cardiometabolic outcomes appear to mostly be a function of the degree of weight loss achieved, rather than any specific component of the dietary prescriptions. Those prescriptions that are better able to lower energy intake, such as the PMR plan and VLCDs, produce greater weight loss; however, the maintenance of the weight loss becomes more challenging if the degree of energy restriction is such that it cannot be maintained or is not designed to be maintained (VLCDs). Intervention research in the area of the pattern of consumption of energy is beginning to suggest that the frequency of eating bouts and when the greatest amount of energy is consumed during the day may be important in improving weight loss outcomes, but more research is needed in the area.
Very little research has examined changes in overall diet quality in energy-focused dietary prescriptions, but research that has been conducted shows that besides decreases in energy intake, reductions in fat (especially if this was part of the dietary prescription), red meat, and sweets intake; and increases in fruit and micronutrient (particularly with PMR plans) intake are found with these energy-focused dietary prescriptions. Research is lacking regarding the various energy-focused dietary prescriptions differing abilities to enhance appetite regulation.
Macronutrient-Focused Prescriptions
In addition to energy-focused dietary prescriptions, dietary prescriptions that alter macronutrient composition of the diet can promote weight loss. While an LCD often reduces fat to assist with reducing energy intake so that the energy goal can be more easily met, other macronutrient-focused dietary prescriptions that alter carbohydrate and protein intake were developed to induce ketosis and/or improve appetite control to assist with weight loss [24, 25]. These macronutrient-focused dietary prescriptions Emphasize reducing the amount of carbohydrate consumed without a specific energy restriction, or increasing the proportion of protein consumed within an energy-restricted diet. Furthermore, diets that focus on carbohydrate intake may alter the type of carbohydrate eaten by accentuating a reduction in glycemic index or load consumed.
Low-Carbohydrate Diet
A low-carbohydrate diet does not have a standard definition; however, most interventions define a low-carbohydrate diet as consuming no more than 20 g of carbohydrate per day [26–28]. In low-carbohydrate diets, energy is not restricted, yet research has found that energy intake does decrease when a low-carbohydrate diet is prescribed [29]. This reduction in energy intake, rather than ketosis as initially theorized [24], is the hypothesized mechanism by which a low-carbohydrate diet produces weight loss [30, 31]. A low-carbohydrate diet recommends consumption of conventional foods high in protein and fat, with a focus on consuming mono- and polyunsaturated fats [24]. Carbohydrates are to be consumed from non-starchy vegetables [26]. Once a desired weight is achieved, carbohydrate intake may gradually increase (5 g per day of carbohydrate per week), primarily in the form of vegetables, limited fruits, and eventually small amounts of whole grains and dairy products, to 50 g of carbohydrate per day [24].
A systematic review of RCTs examining the effect of low-carbohydrate diets on weight loss found that low-carbohydrate diets reduced body weight over a >3-month time period when compared to corresponding baseline values [32]. For more long-term outcomes, a meta-analysis examining the effect of low-carbohydrate and energy-restricted, low-fat diets on weight loss found the low-carbohydrate diet produced a significantly greater weight loss at 6 months (−4.3 kg; −5.6 to −3.0 kg, 95 % confidence interval), but not at 12 months (−1.0 kg; −3.5 to 1.5 kg, 95 % confidence interval) [33]. Finally, a 2-year trial also found that weight loss was not significantly different between a low-carbohydrate diet and an energy-restricted, low-fat diet (low-carbohydrate = −6.3 kg [−8.1 to −4.6 kg, 95 % confidence interval]; energy-restricted, low-fat = −7.4 kg [−9.1 to −5.6 kg, 95 % confidence interval]) [34]. Thus, weight loss appears to be greater in low-carbohydrate diets in the short term (<6 months), however over the long term (>12 months), weight loss appears to be comparable between a low-carbohydrate and an energy-restricted, low-fat diet.
For cardiometabolic outcomes, the results of one meta-analysis found that low-carbohydrate diets positively impact HDL-cholesterol and triglycerides, but negatively impact total cholesterol and LDL-cholesterol as compared to energy-restricted, low-fat diets at 6 months [33]. Little research has evaluated overall diet quality of a low-carbohydrate diet prescription. However, the carbohydrate goal in these diets does mean that intake of grains, fruit, starchy vegetables, and dairy is low. Research on the impact of a low-carbohydrate diet on appetite regulation is limited. A secondary data analysis of a 2-year RCT assessing appetite found that individuals who consumed a low-carbohydrate diet reported being less bothered or distracted by hunger than those on an energy-restricted, low-fat diet [27].
Low Glycemic Index/Load Diet
Within carbohydrate-focused dietary prescriptions for weight loss, low glycemic index, or low glycemic load diets have been examined. Glycemic index is a postprandial measure used to gauge the impact of a certain amount of carbohydrate from a specific food source on blood glucose levels as compared to a reference food, such as white bread or glucose, with the same amount of carbohydrate from each food compared [35]. Thus, a carbohydrate-based food with a high glycemic index will raise blood glucose more quickly and to a higher level than a carbohydrate-based food of low glycemic index [36]. While the glycemic index does not take into account the amount of carbohydrate actually consumed, glycemic load does, and is calculated as the product of the glycemic index of the food multiplied by the grams of the available carbohydrate in the food divided by 100 [37]. A standard dietary prescription for a low glycemic index or load diet does not exist. To alter the glycemic index of the diet, foods are defined as being low or high in glycemic index based on specific cut-offs, and for a low glycemic index diet, low glycemic index foods are encouraged to be eaten instead of higher glycemic index foods. It has been theorized that consuming foods low in glycemic index or reducing the glycemic load of the diet may improve appetite regulation via enhancing glycemic control and this may augment weight loss [38–40].
The effectiveness of an ad libitum low glycemic index diet on weight loss is fairly poor [41]. Moreover, within an energy-restricted diet, several studies have failed to find significant differences in weight loss between energy-restricted, low glycemic index diets as compared to energy-restricted, high glycemic index diets [42, 43]. Furthermore, an energy-restricted, low glycemic load diet did not produce significantly different weight loss outcomes when compared to an energy-restricted, low-fat diet [44]. These findings suggest that consuming a diet low in glycemic index does not enhance weight loss as compared to other diets.
Cardiometabolic outcomes appear to be minimally improved by a low glycemic diet. Glycemic control at 40 weeks measured by changes in HbA1c, appeared to be significantly greater in an energy-restricted, low glycemic diet compared to an energy-restricted, low-fat diet (low glycemic: −0.8 ± 1.3 %; low-fat: −0.1 ± 1.2 %; p = 0.01) [44]. Significant differences in fasting glucose, insulin, HOMA-IR, total cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides, and high-sensitivity C-reactive protein were not found. Further, an energy-restricted, low glycemic index or load diet does not improve cardiometabolic outcomes when compared to a high glycemic index or load diet [42, 43]. While feeding studies have shown mixed results regarding the impact of a low glycemic index or load diets on appetite regulation [45, 46], RCTs have not adequately tested how reducing the glycemic index or load diet effects satiation, especially when energy intake is reduced.
High-Protein Diet
A high-protein diet is defined as consuming 20–30 % energy from protein [47]. For weight loss, high-protein diets also include an energy restriction. Consumption of a high-protein diet is believed to enhance weight loss due to two mechanisms: increasing dietary-induced thermogenesis [48, 49] and enhancing satiation [47, 50, 51]. Greater dietary-induced thermogenesis boosts overall energy expenditure, which could increase the degree of energy deficit incurred, assisting with weight loss. The enhanced satiation experienced with a diet high in protein may assist with appetite regulation, increasing adherence to a diet that is reduced in energy content [25]. A high-protein diet can be achieved through consumption of conventional foods, particularly meats, dairy, eggs, beans, and nuts; however, high-protein, portion-controlled liquid, and solid meal replacement products are available that may enhance dietary adherence to an energy-restricted, high-protein dietary prescription.
For weight loss outcomes, a meta-analysis compared energy-restricted, high-protein diets (mean percent energy from protein: 30.5 ± 2.4 %) to energy-restricted, standard protein diets (mean percent energy from protein: 17.5 ± 1.5 %) [52]. A subgroup analysis that only included studies with a duration ≥12 weeks found weight loss was not significantly greater in energy-restricted, high-protein diets compared to energy-restricted, standard protein diets (weighted mean difference = −0.49; −1.34 to 0.37, 95 % confidence interval).
A completers analysis from a 52-week RCT found that an energy-restricted, high-protein, low-fat diet had similar increases in HDL-cholesterol, and reductions in total cholesterol, LDL-cholesterol, triglycerides, glucose, insulin, blood pressure, and C-reactive protein as compared to an energy-restricted, high-carbohydrate, low-fat diet [53]. As weight loss was not significantly different between conditions and both groups lost significant weight over time, weight loss may be the contributor to improvements in these cardiometabolic measures.
While associations have been found between high-protein diets and satiation, RCTs investigating the impact of energy-restricted, high-protein diets on satiation are limited. One 12-week trial investigated the impact of an energy-restricted, high-protein diet (30 % energy from protein) vs. an energy-restricted, normal protein diet (18 % energy from protein) on weight loss and appetite [54]. Appetite sensations, specifically hunger, fullness, and desire to eat, were not significantly different between groups.
Diets of Varying Macronutrient Composition
To better understand the impact of differing macronutrient alternations on weight loss, Sacks and colleagues [55