Abstract
Consumption of fruits reduces the risk of chronic diseases such as cardiovascular disease; however, very few studies have investigated the effect of fruit consumption in overweight and obese children. We examined whether consuming dried apple as a snack is a practical solution for weight loss and improves body composition and metabolic markers. Thirty-eight overweight or obese children aged 10 to 16 years were randomly assigned to one of two groups consuming twice daily 120 kcal serving per day of either dried apple or a control snack (muffin) for 8 weeks. Body weight, height, waist circumference, and body composition were determined during an initial visit and after 8 weeks of intervention. Blood samples were collected to measure serum concentrations of blood lipids, glucose, insulin, proinsulin, total adiponectin, and C-reactive protein, as well as total antioxidant capacity and activity of glutathione peroxidase. Body weight increased in the muffin group (P = .01). BodPod and dual-energy X-ray absorptiometry showed that fat-free mass increased (P < .05) only in the muffin group. High-density lipoprotein cholesterol concentration increased (P = .04) after the 8-week treatment within the apple group. Overall, minor differences were detected in growing children who consumed snacks of either dried apples or muffins with similar macronutrient profiles for 8 weeks. Future research should evaluate the effects of consuming fresh apples that include the peel.
Introduction
Childhood obesity is a focus of public health efforts in the United States, at least, in part, because obesity has become twice as prevalent in children and four times as prevalent in adolescents in the past 30 years and because the condition becomes more severe during adulthood and increases risk for metabolic diseases. 1,2 The prevalence of obesity in school-aged children (6–19 years) between 2011 and 2014 was 17%. 3 Overall, the percentage of obesity in children has tripled since the 1980s, but it has remained stable for the past 10 years. 2,4 Despite reaching a plateau, obesity remains a serious problem that requires continuous surveillance and development of potential solutions.
Obesity can result in multiple metabolic complications for children. Studies have shown that overweight children have high-density lipoprotein cholesterol (HDL-C) abnormalities and also have a lower capacity to promote cholesterol efflux. 5 –7 In addition, other adverse lipid profiles, greater insulin resistance, and systemic inflammation are observed in overweight children. 8 Results of a cross-sectional study investigating the association of triglyceride-to-HDL-C (TG:HDL-C) ratio with other cardiovascular markers and symptoms of vascular damage in obese prepubertal children indicated that obese children had significantly higher values for the TG:HDL-C ratio than normal children. 9
Lipid profile abnormalities, higher than normal body mass index (BMI), and insulin resistance in children often result in metabolic syndrome. 10 Therefore, it is essential to develop strategies for weight loss and to improve markers of metabolic syndrome.
Obesity does not result from a single underlying element, but is attributed to a number of predisposing factors. Genetics, prenatal and perinatal influences, postnatal weight gain, feeding behaviors of parents such as prompting, pressuring or restricting the child to eat, decreased physical activity, sedentary lifestyle, and unhealthy changes in dietary patterns have been identified as some of the potential reasons for obesity. 11
Consumption of fruits has shown promising results in weight management and chronic disease prevention. 12,13 Studies in the adult population indicate that fruit intake is inversely related to risk of obesity and other chronic diseases since fruits have relatively low energy densities and are high in fiber, and contribute to an overall lower energy intake 12,14 and weight loss. 12 Dried fruit consumption in general has also been associated with improved nutrient intakes and reduced obesity. 15 However, the average diet among individuals in the United States falls short of meeting recommended guidelines for fruit intake. During 2007–2010, about half of the total U.S. population consumed less than 1 cup of fruit per day and 76% did not meet daily fruit intake recommendations of at least two cups per day. 16 Including fruits as a part of a regular diet may help reduce the risk of chronic diseases. 17 –19
In 2014, apple was second only to banana as the most consumed fruit in the United States, according to The Produce For Better Health Foundation. 20 Apple is rich in flavonoids and other phenolics. In fact, in the United States, apples are the largest source of phenolics; 22% of the phenolics consumed from fruits are from apples. 21 In addition, the fiber content of a medium-sized apple contributes 17% of the daily value of fiber. 20 In one study, daily consumption of apple for a period of 4 weeks increased antioxidant enzymes in an elderly population, likely because of the presence of flavonoids. 22
Apple consumption has also been shown to improve body composition. An intervention study conducted by de Oliveira et al. 12 designed to evaluate the effects of adding fruits to the diet on energy consumption and body weight among overweight women aged 30–50 years demonstrated a significant decrease in energy density as well as weight reduction in women who consumed apples. Similar results were observed in a study of postmenopausal women who consumed 75 g dried apples daily for 1 year. Total cholesterol (TC) decreased significantly in the apple group, which exhibited an average of 1.5 kg of body weight loss. 13
Several studies have examined the potential effects of apple on various metabolic markers in adults. 23 –25 However, no clinical study has focused on the effect of apple on overweight and obese children. Therefore, we conducted a randomized controlled clinical trial to examine whether consumption of 120 kcal of fiber-rich snack (dried apple) twice daily for 8 weeks in comparison with a carbohydrate-rich snack (muffin) would improve body composition, inflammatory markers, and blood lipid profiles of overweight and obese children.
Materials and Methods
Participants and study design
Eighty-seven overweight or obese boys and girls between the ages of 10–16 years residing in San Diego, California, with BMI equal to or greater than the 85th percentile on the Centers for Disease Control (CDC) growth chart were recruited for the study. Forty-eight qualified participants started the study in August 2014 and 38 participants completed the study (79.2% retention rate) by May 2016. After an initial screening by telephone, children who were eligible to participate were invited to the School of Exercise and Nutritional Sciences at San Diego State University for their first visit. Written consent and assent forms were obtained from the participants and the parents or legal guardians of the participants before the start of the study. All the study protocols were approved by the Institutional Review Board (IRB) of San Diego State University.
Participants were randomly assigned to one of two intervention groups using a pregenerated randomization list and were instructed to consume 120 kcal of either dried apple (experimental group) or muffin (control group) twice per day for 8 weeks. The dried apple and muffin were matched in energy, carbohydrates, and fat. Diced apples were provided by Tree Top, Inc. (Selah, WA). Muffins were prepared in the laboratory. Apples provided 240 kcal energy, 0 g fat, 59 g carbohydrate, 0.8 g protein, and 8 g fiber. Muffins contained 240 kcal energy, 0 g fat, 55 g carbohydrate, 4 g protein, and 0 g fiber. As reported previously, the dried apples included 732 μg gallic acid equivalents per gram of apple. 26
The snacks were provided to the participants in prepackaged bags at the beginning of the study and the participants were instructed to consume the snack between breakfast and lunch, and lunch and dinner along with 8 oz of water to control for potential hunger variations that could be caused by the consumption of other fluids. Subjects were provided with a daily calendar checklist to mark any day they missed consuming their snack and were instructed to record and return any portion of the leftover snack for the purpose of monitoring compliance. Participants were advised to maintain their usual physical activity and dietary patterns throughout the study period.
Participants visited the laboratory twice during the study at 0 weeks (baseline) and 8 weeks (final visit). A 24-h dietary recall was collected from the subjects at baseline and after 8 weeks. Height, weight, waist circumference, and body composition measurements as well as physical activity data and blood samples were collected at baseline and at 8 weeks.
24-h dietary recall and physical activity measures
A 24-h dietary recall was collected twice during the course of the study (0 and 8 weeks) to ensure that participants adhered to their usual diet and consumed the snack provided to them. During the recall, participants were asked to provide information about types of foods and beverages consumed along with meal timings and portion sizes. Studies have shown that 24-h recall measures are reliable for dietary assessment. 27 –29 All food records were analyzed using the Food Processor (ESHA, Salem, OR, USA) software.
The Fels physical activity questionnaire (PAQ) was administered to the participants to assess their physical activity level. Fels PAQ asks participants to report the type, location, domain, and context of the activity as well as provide estimates of time spent in activities of various levels of intensity. Fels PAQ is moderately reliable for children of all age groups. 30
Anthropometric measures
Subjects were asked to wear a bathing suit while being measured for waist circumference and body weight at each visit. Waist circumference was measured by placing a flexible tape measure laterally, midway between the lowest portion of the rib cage and iliac crest, and anteriorly midway between the xiphoid process of the sternum and the umbilicus. Waist circumference was recorded to the nearest centimeter while the participants were in the standing position at the end of gentle expiration. 31 Body weight was measured using a beam balance equipped with a stadiometer to the nearest 0.1 kg and height was measured standing barefoot to the nearest 0.5 cm. BMI was calculated using the standard formula weight (kg) divided by height (m2). The values were plotted on the CDC growth chart to determine participants' classification status (i.e., ≥85th percentile is overweight; ≥95th percentile is obese).
Body composition measures
Body composition was assessed using dual-energy X-ray absorptiometry (DXA; GE Healthcare Lunar, Madison, WI, USA) and Bod Pod (Life Measurement, Inc., Concord, CA, USA) measures. DXA followed by Bod Pod measures are the gold standards for measuring children's body fat percentage. 32
Serum analyses
Fasted blood samples (20 mL) were collected and centrifuged at 4°C to obtain serum. Samples were stored in cryogenic vials at −70°C until batch analysis. Serum glucose, TC, HDL-C, and TG were analyzed colorimetrically with kits from Stanbio Laboratory (Boerne, TX, USA). Low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald formula: TC − HDL − TG/5. 33 Insulin and proinsulin were determined using ELISA kits from Quidel (San Diego, CA, USA). Homeostatic model assessment of insulin resistance (HOMA-IR) was calculated using the equation described by Vogeser et al.: fasting insulin (μUI/mL) × fasting glucose (mM)/22.5. 34 C-reactive protein (CRP) and total adiponectin were assessed using ELISA kits from ALPCO (Salem, NH, USA). Total antioxidant capacity and glutathione peroxidase (GPx) activity were measured using assay kits from Cayman Chemical Company (Ann Arbor, MI, USA).
Statistical analyses
The data for this study were analyzed using IBM SPSS version 22 (Armonk, NY, USA). Two (group: control and apple) × 2 (time: baseline and 8 weeks) mixed design analyses of variance (ANOVA) were conducted to analyze effects on body composition and biochemical markers. P < .05 was considered statistically significant (95% CI). Significant main effects and interactions were followed by Bonferroni post hoc tests.
Results
Baseline characteristics, dietary intakes, and physical activity, anthropometric measurements, body composition assessments
Of the 87 subjects recruited for the study, 48 were eligible and 38 completed the study. The attrition rates were not similar between the groups (apple = 10.6%, muffin = 6.4%). The compliance to the dried apple and muffin regimens in those participants who completed the study was on average 87% for each group. Dietary intake and physical activity levels did not significantly differ between groups, suggesting that the intake patterns were similar between the two groups (Tables 1 and 2). Gallic acid intake for the apple group was estimated to be 54.9 mg/day (732 μg/kg body weight per day). For physical activity, sport index decreased within the apple group (P = .02); however, the total score was similar (P = .4) between the groups. Anthropometric measures of the participants such as weight, waist circumference, and BMI percentile did not significantly differ from baseline to 8 weeks within both groups. However, there was a significant increase in height within both the apple (P = .01) and muffin groups (P = .01; Table 3). Also, a significant increase in weight (P = .01) was detected within the muffin group over the 8 weeks. Results from BodPod showed that fat-free mass increased significantly within the muffin group over the 8-week period (P = .01), whereas there was no significant change in the fat-free mass in the apple group (P = .07; Table 4). In the muffin group, trunk fat mass increased significantly over the 8 weeks according to DXA (P = .05; Table 5).
Daily Nutrient Intake Calculated from Dietary Recall at Baseline and 8 Weeks Supplemented Without Dried Apple or Muffin
Values are expressed as mean ± SD. Data were analyzed using mixed design ANOVA. In case of a significant main effect (time, group) or significant interaction (time × treatment), follow-up was done using one-way repeated measure ANOVA. P < .05 was considered significant.
ANOVA, analysis of variance; CHO, Carbohydrate; SD, standard deviation.
Physical Activity Index at Baseline and 8 Weeks
Values are expressed as mean ± SD. Data were analyzed using mixed design ANOVA. In case of a significant main effect (time, group) or significant interaction (time × treatment), follow-up was done using one-way repeated measure ANOVA. P < .05 was considered significant.
Significantly different compared with baseline within group.
Anthropometric Indices at Baseline and 8 Weeks
Values are expressed as mean ± SD. Data were analyzed using mixed design ANOVA. In case of a significant main effect (time, group) or significant interaction (time × treatment), follow-up was done using one-way repeated measure ANOVA. P < .05 was considered significant.
Significantly different compared with baseline within group.
BMI, body mass index.
Effects of Dried Apple and Muffin Consumption on Body Composition at Baseline and 8 Weeks as Measured by BodPod
Values are expressed as mean ± SD. Data were analyzed using mixed design ANOVA. In case of a significant main effect (time, group) or significant interaction (time × treatment), follow-up was done using one-way repeated measure ANOVA. P < .05 was considered significant.
Significantly different compared with baseline within group.
Effect of Dried Apple and Muffin Consumption on Body Composition at Baseline and 8 Weeks as Measured by Dual-Energy X-Ray Absorptiometry
Values are expressed as mean ± SD. Data were analyzed using mixed design ANOVA. In case of a significant main effect (time, group) or significant interaction (time × treatment), follow-up was done using one-way repeated measure ANOVA. P < .05 was considered significant.
Significantly different compared with baseline within group.
Serum analysis
There were no significant differences in TC, TG, or LDL-C between the two groups. HDL-C levels in the apple group significantly increased after the 8-week period (P = .04) (Table 6). There were no significant differences in glucose, insulin, proinsulin, HOMA-IR, CRP, or adiponectin between groups from baseline to 8 weeks (Table 7). No significant differences were found between groups for total antioxidant capacity or GPx activity (Table 7).
Effect of Dried Apple and Muffin Consumption on Lipid Profile at Baseline and 8 Weeks
Values are expressed as mean ± SD. Data were analyzed using mixed design ANOVA. In case of a significant main effect (time, group) or significant interaction (time × treatment), follow-up was done using one-way repeated measure ANOVA. P < .05 was considered significant.
Significantly different compared with baseline within group.
HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.
Effect of Dried Apple and Muffin Consumption on Blood Serum Biomarkers at Baseline and 8 Weeks
Values are expressed as mean ± SD. Data were analyzed using mixed design ANOVA. In case of a significant main effect (time, group) or significant interaction (time × treatment), follow-up was done using one-way repeated measure ANOVA. P < 0.05 was considered significant.
CRP, C-reactive protein; GPx, glutathione peroxidase; HOMA-IR, homeostatic model assessment of insulin resistance.
Discussion
Childhood obesity increases the risk for many metabolic diseases in later life and has increasingly become a public health concern. Foods and diets that mitigate a positive energy balance and alter metabolic pathways associated with fat storage or weight gain complement physical activity approaches to weight management. 35,36 Cross-sectional studies have shown that fruit consumption improves the overall diet quality and nutrient intake in the adult population. 15 Fruit intake has been shown to lower the risk of obesity, improve anthropometric indices, and lower cardiovascular disease (CVD) and metabolic syndrome risk factors. 15,37 –40 Cross-sectional studies in children have also shown that the consumption of fruit reduces the risk of childhood obesity. 40,41
Dried fruits, in particular, have been shown to produce many health benefits. 41 Apples are associated with decreased risk of obesity in children according to the National Health and Nutrition Examination Survey (2003–2010). 42 Therefore, in this study, we examined whether consumption of dried apple (120 kcal) twice daily for an 8-week period in comparison with a carbohydrate-rich snack (muffin) would improve body composition, inflammatory markers, and blood lipid profile in overweight and obese children.
In this study, waist circumference did not change significantly between the apple and muffin groups, but weight, primarily due to an increase in lean mass, increased significantly in the muffin group, while the weight of children in the apple group remained relatively unchanged. Previous studies have provided discrepant results with respect to weight loss associated with fruit consumption. Findings of a study by Chai et al. 13 comparing dried plum and dried apple consumption found that postmenopausal women who consumed 75 g of dried apple daily, the same amount used in our study, had a 1.5 kg weight loss over 1 year. Similarly, Aprikian et al. 43 observed a significant weight loss of 1.32 kg after 10 weeks of daily intake of 300 g apple among overweight women with high cholesterol levels. However, a meta-analysis by Kaiser et al. 44 suggested that increased fruit/vegetable intake cannot cause discernable weight loss without reducing energy intake from other sources.
In this study, participants in the apple group did not lose weight but remained at a relatively stable weight throughout the study. This outcome suggests that fruits rich in fiber and low in energy density, like apple, may not promote weight loss in children, but also do not promote weight gain in already overweight and obese children. Research suggests that eating solid fruit, as opposed to pureed fruit or fruit juice, at the start of a meal can reduce energy intake. 45 Consistent with this, children in the muffin group significantly gained weight (1.25 kg) over an 8-week period, whereas no similar effect was detected for children eating dried apple. This could potentially be attributed to the consumption of muffin, which is a refined carbohydrate snack. However, it is important to note that much of the weight gain was due to increases in lean mass.
The increase in weight gain is consistent with a study by Mozaffarian et al. who found that refined carbohydrate-rich foods were positively associated with weight gain in adults. 46 Those researcher did not report body composition changes of their participants. In this study, overall macronutrient composition of the diet was consistent from baseline to 8 weeks within each group. For instance, carbohydrate intake at baseline and after 8 weeks was, respectively, 50% and 53% among the apple group and 49% and 52% among the muffin group.
Some of the protective effects of apple against CVD have been attributed to apple's cholesterol lowering potential. 47 Apple may contribute to improved lipid profiles due to its high fiber and polyphenol content. 48,49 Aprikian et al. 47 observed a significant drop in plasma cholesterol and liver cholesterol and an increase in HDL-C among cholesterol-fed rats supplemented with lyophilized apples. Similarly, de Oliveira et al. 50 demonstrated that apple produced a greater cholesterol lowering effect in rats in comparison with pears and peaches. It has been suggested that phenolic compounds in apples contribute to this effect. 47 Similarly, a study conducted by Cho et al. 51 found that obese rats fed a high-fat diet containing 10% apple pomace exhibited significantly lower serum TC, LDL-C, and TG concentrations and significantly higher HDL-C after 5 weeks compared with the high-fat diet control group.
Although there were no significant changes in LDL-C or TC in this study, HDL-C levels significantly increased in subjects who consumed dried apple daily for 8 weeks. Similar results were obtained in a study by Codoner-Franch et al. 52 in which obese children who consumed 40 g of dried apples enriched with mandarin orange juice had significantly higher HDL-C levels after 4 weeks than baseline. Likewise, in a study by Chai et al., 13 HDL-C significantly increased in postmenopausal women after 12 months of daily consumption of 75 g dried apple.
Oxidative damage may contribute to CVD due to disruptions in lipids and other cellular components by reactive oxygen species. 53 Apple ranked second after cranberries in terms of highest level of total phenolic content when compared with many other common household fruits in the United States. 54 In vivo studies have shown that apple and/or apple juice provides antioxidant activity. 55,56 Furthermore, Codoner-Franch et al. 52 demonstrated increased antioxidant capacity in obese children who consumed 40 g of dried apple enriched with mandarin orange juice after 4 weeks compared with baseline as evidenced by a significant increase in ferric reducing antioxidant power and 2,2-azinobis (3 ethylbenzothiazoline-6-sulfonic acid. However, in this study, there were no significant differences between the two groups in total antioxidant capacity or GPx activity (markers of oxidative stress). 57
Fructose is associated with significantly lower serum glucose and decreased insulin responses as compared with sucrose-sweetened foods such as muffins. 58 Moderate fructose intake (7.5 g) improves the glycemic response without enhancing the TG response. 58 Nevertheless, very high fructose intake significantly increases fasting plasma triacylglycerol. 59 In this study, similar to the study conducted by de Oliveira et al., 60 daily fructose consumption did not affect insulin. However, in contrast to the same study, we did not observe any association between fruit intake and serum glucose.
Although this study did not reveal significant changes in glucose, insulin, or proinsulin levels in either group, high dietary fiber intake is typically associated with greater insulin sensitivity in adolescents. 61 In a study of 17 overweight and obese women, intake of 31.2 g/day of insoluble fiber from enriched bread for 3 days significantly improved whole-body glucose disposal, which equates to an 8% improvement of insulin sensitivity. 62 However, studies examining the effects of fiber on glucose and insulin in overweight and obese children are limited.
In conclusion, although only minor differences were detected, dried apple can be recommended as a snack for overweight and obese children due to its high-fiber and polyphenol content. Using dried apple as an alternative to a refined carbohydrate-rich snack may help stabilize body weight and increase HDL-C, both of which may aid in the prevention of CVD. Strengths of our study include high compliance and low attrition rate. Weaknesses include the relatively short intervention period that may have prevented our ability to detect additional significant results. Future studies should consider a similar design with a larger sample size for a longer duration and should include the effects of apple snacking on satiety in overweight and obese children. In addition, studies of whole fresh apples that include the peel may produce differing results in comparison with dried apple without the peel.
Footnotes
Acknowledgments
We thank the following SDSU students for help with data collection: Lama Almaiman, Amanda Cravinho, Arshya Ahouraei, Christine Fernandez, Nicole Herrmann, Diana Lee, Fanny Lee, Heather Tsunoda, Jennifer Beadles, Rose Miller, Sofie Blicher, and Theresa Carmichael.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
This study was funded by the U.S. Apple Association.