Breakfast skipping in prepubertal obese children: hormonal ... - Nature

European Journal of Clinical Nutrition (2012) 66, 314–321

& 2012 Macmillan Publishers Limited All rights reserved 0954-3007/12 www.nature.com/ejcn

ORIGINAL ARTICLE

Breakfast skipping in prepubertal obese children: hormonal, metabolic and cognitive consequences C Maffeis1, E Fornari1, MG Surano1, E Comencini2, M Corradi1, M Tommasi1, I Fasan1 and S Cortese2,3,4 1

Unit of Pediatric Diabetes, Clinical Nutrition and Obesity, Department of Life and Reproduction Sciences, University of Verona, Verona, Italy; 2Child Neuropsychiatry Unit, Department of Life and Reproduction Sciences, University of Verona, Verona, Italy; 3 Institute for Pediatric Neuroscience, New York University Child Study Center, New York City, NY, USA and 4UMR_S INSERMU 930, CNRS ERL 3106, Franc¸ois-Rabelais University, Child Psychiatry Centre, University Hospital, Tours, France

Background and Aims: Skipping breakfast influences cognitive performance. The aim of our study was to investigate the relationship between the variation of hormonal and metabolic postprandial parameters induced by breakfast consumption or fasting and cognitive performance in obese children. Methods: Cross-sectional study for repeated measures. Memory and attention assessment tests, hormones and nutrient oxidation were measured before and after consuming breakfast vs fasting in 10 prepubertal obese children. Results: Fasting induced a significant (Po0.05) increase of the Overall Index of the Continuous Performance Test II (a global index of inattention) and the Test of Memory and Learning Word Selective Reminding (a test of verbal memory), whereas no changes were found after breakfast. Fasting was associated with a reduction of insulin and an increase in glucagon, with no changes in glucose. The increase in inattention was associated with a reduction of carbohydrate oxidation (r ¼ 0.66, Po0.05). We found no difference in the area under the curve of peptide YY and glucagon-like peptide-1 after breakfast or fasting, whereas Ghrelin was significantly lower. No association between postprandial hormone variation and cognitive performance was found. Conclusions: Attention and visual memory performance in the morning were reduced when the children skipped breakfast. No association was found with hormones or metabolic changes, but we did find an association with a reduction of carbohydrate oxidation. Nevertheless, these preliminary findings need confirmation in larger sample size.

European Journal of Clinical Nutrition (2012) 66, 314–321; doi:10.1038/ejcn.2011.206; published online 11 January 2012 Keywords: children; obesity; hormones; nutrient balance; attention

Introduction

Correspondence: Dr C Maffeis, Regional Center for Pediatric Diabetes, Unit of Clinical Nutrition and Obesity, Department of Life and Reproduction Sciences, University of Verona, Via Bengasi, 4, 37134 Verona, Italy. E-mail: [email protected] Contributors: CM designed and coordinated the study, submitted the study to the Ethical Committee, collected the data, performed the statistical analysis and wrote the manuscript. EF collected the data, provided data input, collaborated on the statistical analysis and the writing of the manuscript. MGS recruited the children, collected the data, provided data input and collaborated on the statistical analysis. EC participated in the design of the study, collected the data and provided data input. MC participated in the design of the study, carried out sample analysis and provided data input. MT participated in the design of the study and collected the data. IF participated in the design of the study and collected the data. SC designed the study, performed the statistical analysis and wrote the manuscript. Received 8 June 2011; revised 5 October 2011; accepted 11 November 2011; published online 11 January 2012

Skipping breakfast influences cognitive performance and mood in children and adults, leading to a decline in attention and memory during the morning (Wesnes et al., 2003; Rampersaud et al., 2005; Hoyland et al., 2009). In particular, in a group of high school students, visuospatial memory and self-reported alertness improved significantly after consuming breakfast compared with fasting conditions ¨ ller et al., 2008). This effect may be poten(Widenhorn-Mu tially greater in obese children than in non-obese children owing to the higher energy requirement in the former (Bandini et al., 1990; Maffeis et al., 1993). Previous data have shown that increased body weight was associated with decreased visuospatial organization and general mental ability among children, independent of parental and familial socio-economic status or other potential confounders, such

Consequences of skipping breakfast in obese children C Maffeis et al

315 as participation in sports, amount of physical activity, hours of TV viewing, psychosocial development, blood pressure and serum lipid profile (Li et al., 2008). The explanations for these findings are not available. One potential factor in decreased morning cognitive performance is the influence on glucose metabolism. Glucose is the main fuel oxidized in the brain, which is also the greatest glucose consumer in the body at rest (Scholey et al., 2009). Moreover, children’s brains oxidize more glucose than those of adults (Chugani, 1998). Therefore, postprandial glucose disposal may potentially affect brain function and neuropsychological function in children (Micha et al., 2010). A recent study conducted on 40 healthy adults showed that the cognitive functions were enhanced by avoiding a sharp decline in blood glucose concentration and by maintaining a higher glycemia in the late postprandial period (Nilsson et al., 2009). Moreover, recent data obtained with positron emission topography showed that obese adults have brain insulin resistance associated with peripheral insulin resistance, suggesting the central nervous insulin resistance as a common denominator of metabolic disorders and possibly of cognitive dysfunction in these subjects (Hallschmid and Schultes, 2009). In particular, it has been demonstrated that selective blockade of endogenous intrahippocampal insulin signaling impairs memory performance (McNay et al., 2010). Also, it has been claimed that gastrointestinal hormones involved in regulating appetite (Ghrelin and glucagon-like peptide-1 (GLP-1)) play a role in memory function as well as learning and ˜eda et al., 2010; Irwin et al., 2010). cognitive decline (Castan To the best of our knowledge, no studies have explored the potential relationships between the changes in cognitive functioning occurring in the morning after breakfast or fasting and postprandial metabolic and hormonal changes. Therefore, the aims of this study were to explore: (i) the changes of various indexes of cognitive functions induced by breakfast or fasting; and (ii) their relationship with postprandial hormonal and metabolic parameters in a sample of obese prepubertal children.

Subjects and methods Subjects In all, 10 prepubertal obese children (males/females: 4/6) participated in the study. Selecting prepubertal-age children reduces the potential-confounding effect on hormonal and metabolic changes induced by puberty. None of the children had any overt condition other than obesity. Puberty development was clinically assessed on the basis of Tanner stages (Tanner and Whitehouse, 1976). Children were defined as obese on the basis of their body mass index, which was above the body mass index cutoffs for obesity proposed by the International Obesity Task Force (Cole et al., 2000). None of them were dieting at the time of the study or were on medication. Written informed consent was obtained from the children and their parents before taking part in the

study. The protocol was approved by the Ethics Committee of the University Hospital of Verona.

Study design The experiment was designed as a randomized, crosssectional study for repeated measures. The aim was to compare neuropsychological measures as well as the changes in several metabolic parameters, thermogenic response, nutrient oxidation and gastrointestinal hormone levels in the 3 h after breakfast or fasting in the same subjects. Each test lasted 5 consecutive hours, during which the children were under close medical supervision. During the days preceding the test, none of the subjects was on a low calorie diet. The children had consumed their last meal at 2000 hours the day before. Each child arrived at the Department of Pediatrics at 0800 hours on the day of the test and performed the neuropsychological tests (Continuous Performance Test and Test of Memory and Learning) (Reynolds and Bigler, 1994; Conners, 2002). A teflon catheter was inserted in the antecubital vein of one arm for blood sampling. After 30 min of rest in a comfortable temperature (B24 1C) and humidity-controlled environment, continuous respiratory exchange measurements were taken by indirect calorimetry. After baseline (preprandial) blood sample collection, the children consumed the test meal (p15 min). The meal was fully consumed by all the children. Blood samples were collected at 15, 30, 60, 120 and 180 min after meal ingestion. Continuous respiratory exchange measurements were performed every 30 min. The neuropsychological tests were repeated after 180 min (Figure 1). At 1 week after the first test, the children repeated the test with a different meal. The two meals were assigned randomly.

Dietary intake On the day of the study, after baseline respiratory exchange measurements (indirect calorimetry), the children were given a test meal. Two different meals were served on the 2 days of the test. The energy content of the breakfast meal— whole milk, toasted bread and marmalade (as shown in Table 1)—was calculated as 20% of the estimated daily energy requirements at rest for 10-year-old children (Istituto Nazionale della Nutrizione, 1977). The composition of the meal was based on cultural consistency and broad acceptability by most of the children. Fasting consisted of only one glass of water. The energy and nutrient contents of the breakfast meal were calculated using the National Institute of Nutrition tables for food composition (Istituto Nazionale della Nutrizione, 1977). The meal was consumed under medical supervision.

Measurements of energy expenditure After 30 min of absolute rest, considered to be an adaptation period during which the procedure was explained to the European Journal of Clinical Nutrition


316 BREAKFAST

ab

-90’

-60’

b b

-30’ -15’ 0’ 15’ 30’

ab

60’

ab

90’

120’

ab

150’

180’

210’

time (minutes)

3.0 h Neuropsychological tests Indirect calorimetry a = visual scale of appetite b = blood sample

Figure 1

Study design and timing of measurements of the experiment.

Table 1 Energy and macronutrient composition of the test meal Breakfast

Proteins Lipids Carbohydrates kcal (g) (13%) (g) (27%) (g) (60%)

Whole milk (200 ml)

6.36

7.28

9.44

124

Total energy (kcal) (%) Toast-like bread (31.6 g) Marmalade (25 g)

3.2 0.08

1.6 —

24 11.5

125 46

9.66 38

8.88 79

44.94 178

295

Nutrient (g) Total energy (kcal) %, percentage of total energy.

children and their parents, respiratory exchanges were measured continuously for 30 min at four different times during the study period: one before meal intake and three after meal intake. During measurement the child rested quietly. The post-absorptive resting energy expenditure (REE) measurement was taken at 0830 hours (preprandial baseline). Postprandial calorimetric measurements were taken at 9.45, 10.45 and 11.45 and lasted 30 min each. Respiratory exchange measurements were recorded using an open-circuit computerized indirect calorimeter (Deltatrac; Datex Inc., Helsinki, Finland), using a transparent ventilated hood system, as described previously (Maffeis et al., 1995). REE was calculated from oxygen production (VO2) and carbon dioxide production (VCO2) using Lusk’s formula (Lusk, 1928).

Meal-induced thermogenesis and macronutrient oxidation rate The thermic effect of the meal was calculated by subtracting the pre-meal REE value (kcal/min) from the respective European Journal of Clinical Nutrition

average postprandial energy expenditure (kcal/min), and this was multiplied by the duration of the postprandial period (180 min). Thermic effect of the meal was expressed as a percentage of the metabolizable energy value of the test meal. The macronutrient oxidation rate was calculated from VO2 and VCO2 using the following formulas (Maffeis et al., 2001): Fox (g/min) ¼ 1.67 VO2 (l/min)1.67 VCO2 (l/min) 0.307 Pox and Gox (g/min) ¼ 4.55 VCO2 (l/min)3.21 VO2 (l/min)0.459 Pox, where Fox is fat oxidation, Gox is glucose oxidation and Pox is protein oxidation. Protein oxidation was estimated as follows: Pox (g/min) ¼ (REE (kcal/ min) 0.12)/4 kcal. We assumed that protein oxidation covered 12% of REE, based on evidence that the difference in energy expenditure was due to protein oxidation (expressed as a percent of energy expenditure), and between the pre- and postprandial phases was minimal (Acheson et al., 1987). Postprandial macronutrient oxidation was quantified calculating the area under the respective 180 min plots.

Biochemical analysis Blood samples were collected in EDTA (plus aprotinin to 0.6 TIU/ml in the tubes for hormones), centrifuged shortly after collection at 1600 g for 10 min at 4 1C and the plasma stored at 80 1C. Plasma glucose was analyzed by using the glucose oxidase method. Plasma insulin level was measured using the specific chemiluminescence method with an Insulin Bridge kit (Adaltis Inc., Montreal, QC, Canada). Glucagon, Ghrelin, peptide YY (PYY) and total GLP-1 were analyzed by means of EIA kits (Phoenix Pharmaceuticals, Belmont, CA, USA) (intra-assay coefficient of variation 5–10% and inter-assay coefficient of variation o15%) according to the protocol provided by the supplier. Total


317 Ghrelin (acylated plus des-acyl) and not just its active form was measured.

Neuropsychological assessment Two neuropsychological functions were assessed, namely attention and memory. Attention performances were assessed by means of the Conners’ Continuous Performance Test (CPT) II Version 5 (Conners, 2002). The CPT II V.5 is a commonly used taskoriented computerized assessment of attention. It includes the following subscales: Response Times; Change in Reaction Time Speed and Consistency; Signal Detection Theory Statistics; Omission Errors; Commission Errors as well as Overall Statistics (Confidence Index and Overall Index) of attention functions. The Overall Index, which provides a weighted sum of the subscales, and therefore, a comprehensive measure of attention impairment, was considered the main outcome for the CPT, although the individual subscales were also analyzed. A higher score on the Overall Index indicates a higher level of inattention. Memory function was assessed using the Test of Memory and Learning (Reynolds and Bigler, 1994). The core battery consists of five verbal and five non-verbal subtests. Besides the core battery, four supplementary subtests (three verbal and one non-verbal) were included. In this study, the two following subtests were used: ‘Word Selective Reminding’ and ‘Visual Sequential Memory’ in order to assess specifically short-term memory function. The Italian version of the test was used (Reynolds, 1994). We chose only two (Visual Sequential Memory and the Word Selective Reminding) out of the eight core subtests of the Test of Memory and Learning-2 battery to reduce the administration time, and thus avoid a possible fatigue effect, which would have hampered the interpretation of the results. We chose Visual Sequential Memory and the Word Selective Reminding to obtain an index of verbal memory as well as an index of spatial memory.

Statistical analysis All results are shown as means and standard errors of the mean (s.e.m.). The global response of hormones and metabolites to meal ingestion was measured as the area under the curve (AUC). Data were not normally distributed, so non-parametric statistical methods were used. The comparison between means of variables and AUC measured before or after breakfast or fasting was carried out by a Wilcoxon paired-sample test. The potential association between metabolic parameters, hormones, appetite and neuropsychological tests scores after the two meals were consumed was calculated by Spearman correlation analysis. Power analysis (two sided a ¼ 0.05) showed that the sample size (n ¼ 10 subjects) was adequate (1–b40.80) to detect a difference in attention performance (CPT II Overall Index) between baseline and 180 min of 7% (estimated s.d. ¼ 7) in

Table 2

Physical characteristics

Age (years) Height (cm) Weight (kg) BMI (kg/m2) BMI z-score Fat mass (%) Fat mass (kg) Fat free mass (kg) Waist circumference (cm)

Median

IQR

9.6 142.0 51.5 26.2 2.2 37.3 19.1 32.7 80.5

9.3–10.1 137.7–145.5 46.0–58.5 24.6–27.6 2.0–2.5 34.9–40.0 15.9–23.3 30.0–35.2 77.5–84.2

Abbreviations: BMI, body mass index; IQR, interquartile range. Data are presented as median and IQR.

children who did not have breakfast. Po0.05 was used to indicate statistical significance. Statistical analyses were performed using the SPSS 16.0 software for Windows (SPSS Inc., Chicago, IL, USA).

Results The physical characteristics of the subjects are shown in Table 2. Fasting induced a significant (Po0.05) increase in the CPT II Overall Index (pre- vs postprandial: 27.3 (4.5) vs 33.7 (6.7) error %) and World Selective Reminding of the Test of Memory and Learning (pre- vs postprandial: 34.2 (2.6) vs 38.6 (2.3) percentile). No significant changes were found for Visual Sequential Memory (pre- vs postprandial: 88.3 (3.1) vs 87.1 (7.4) percentile). No significant difference for any other neuropsychological measures, including the subscales of the CPT, was found before and after breakfast. Breakfast caused a significant (Po0.05) postprandial increase in energy expenditure above baseline, whereas fasting was followed by no change (P ¼ NS) of energy expenditure in the 3 h of measurement (Table 3). Postprandial respiratory quotient was significantly (Po0.05) higher after breakfast, whereas it did not change from baseline after fasting. Carbohydrate and protein oxidation measured in the 180 min after meal ingestion was significantly (Po0.05) higher after breakfast than after fasting. On the contrary, the total amount of lipid oxidized after breakfast was lower than after fasting. Nutrient balance measured in the 180 min after breakfast consumption was positive for proteins and carbohydrates and negative for lipids. Meal intake induced an increase in glucose and triglyceride concentrations, whereas fasting was followed by a substantial stability of blood glucose and triglyceride concentrations (Figure 2). The AUC of glucose and triglycerides were significantly (Po0.01) higher after breakfast than after fasting. Insulin increased above baseline after meal intake and decreased with fasting (breakfast (pre- vs postprandial): 9.3 (0.9) vs 40.7 (3.2), P ¼ o0.001; fasting (pre- vs postprandial) 12.7 (2.0) vs 7.3 (0.8), Po0.05). On the contrary, glucagon decreased after breakfast and increased after fasting European Journal of Clinical Nutrition


318 Table 3 Energy intake, energy expenditure and nutrient oxidation, measured before and after breakfast and fasting, respectively

Preprandial Energy expenditure (kcal/min) RQ Nutrient oxidation Protein (mg/min) Carbohydrate (mg/min) Lipid (mg/min) Test meal (kcal) CPT II Overall Index (error %) World Selective Reminding (percentile) Visual Sequential Memory (percentile) Postprandial Energy expenditure (kcal/min) Energy expenditure (kcal/180 min) Meal-induced thermogenesis % of metabolizable energy of the test meal RQ Nutrient oxidation Protein (mg/min) Carbohydrate (mg/min) Lipid (mg/min) Energy balance (kcal/180 min) Nutrient balance (g/180 min) Protein Carbohydrate Lipid CPT II Overall Index (error %) World Selective Reminding (percentile) Visual Sequential Memory (percentile)

Breakfast N ¼ 10

Fasting N ¼ 10

P-value

1.04 (0.04) 0.79 (0.01)

1.04 (0.02) 0.78 (0.01)

NS NS

32.7 (1.2) 80.0 (9.7) 68.4 (5.3) 295

32.7 (0.9) 64.8 (9.8) 72.2 (5.4) 0 27.3 (4.5)* 34.2 (2.6)*

NS NS NS —

88.3 (3.1)

1.16 (0.02) 208.4 (4.6)

1.07 (0.03) 193.5 (5.5)

o0.05 o0.05

5.6 (0.8)

—

—

0.82 (0.01)

0.77 (0.01)

o0.05

32.1 (1.1) 64.6 (8.6) 73.8 (5.2) —

NS o0.05 o0.05 —

— —

— —

34.8 122.3 58.0 67.4

(1.4) (9.0) (4.8) (4.8)

3.3 (0.2) 18.9 (1.4) 6.1 (0.7)

(6.7) 38.6 (2.3) 87.1 (7.4)

Abbreviations: CPT, Continuous Performance Test; NS, not significant; RQ, respiratory quotient. Data are shown as mean and standard error of the mean (s.e.m.). *Po0.05, preprandial vs postprandial.

(breakfast (pre- vs postprandial): 3.7 (0.2) vs 3.1 (0.1), P ¼ o0.01; fasting (pre- vs postprandial) 2.7 (0.1) vs 3.2 (0.1), Po0.01). The AUC of postprandial insulin was significantly (Po0.001) higher after breakfast than after fasting. The AUC of postprandial glucagon was significantly lower (Po0.001) after breakfast than after fasting. Meal ingestion was followed by a decrease in Ghrelin concentrations: AUC of Ghrelin was significantly lower (Po0.02) after breakfast than after fasting. The AUC of PYY and GLP-1 were not significantly different after breakfast and fasting (Table 4).

Correlations Changes in attention or memory function did not show a significant correlation with AUC of glucose, insulin, glucagon European Journal of Clinical Nutrition

Figure 2 Changes in glucose and triglycerides concentrations over time with breakfast and fasting (*Po0.05).

Table 4 AUC of glucose, insulin, triglycerides, appetite, PYY, glucagon, GLP1 and Ghrelin Breakfast

Fasting

P-value

Glucose (mmol 180 min/l) 116.4 (19.7) 6.1 (11.4) 0.001 Insulin (pmol 180 min/l) 4856.2 (453.9) 927.0 (336.6) o0.001 Triglycerides 36.7 (11.5) 8.2 (5.5) 0.003 (mmol 180 min/l) PYY (pmol 180 min/l) 1.6 (5.2) 5.5 (7.5) NS Glucagon (pmol 180 min/l) 115.4 (25.9) 83.0 (21.3) 0.001 GLP1 (pmol 180 min/l) 16.3 (13.6) 4.9 (9.9) NS Ghrelin (pmol 180 min/l) 363 (103) 151 (81) o0.02 Abbreviations: AUC, area under the curve; GLP-1, glucagon-like peptide-1; PYY, peptide YY; NS, not significant. Data are shown as mean and standard error of the mean (s.e.m.).

or gastrointestinal hormones (PYY, -1 and Ghrelin) after either meal (data not shown). However, a significant correlation was found between changes in the CPT II Overall Index; for instance, increasing inattention and decreasing postprandial carbohydrate oxidation (r ¼ 0.66, Po0.05). No significant correlations were found between postprandial carbohydrate oxidation after breakfast or fasting and AUC of glucose (breakfast: r ¼ 0.28; fasting r ¼ 0.39), insulin (breakfast: r ¼ 0.23; fasting r ¼ 0.22), glucagon (breakfast: r ¼ 0.64; fasting r ¼ 0.23) and gastrointestinal hormones (PYY (breakfast: r ¼ 0.61; fasting r ¼ 0.22), GLP-1


319 (breakfast: r ¼ 0.54; fasting r ¼ 0.13) and Ghrelin (breakfast: r ¼ 0.01; fasting r ¼ 0.53)).

Discussion This study showed that mental performance is reduced in children who skip breakfast and that the reduction in attentive performance (as indicated by the CPT II Overall Index) was associated with a reduction in carbohydrate oxidation. Under fasting conditions, despite no significant changes in carbohydrate oxidation over the 3 h compared with baseline, children showing the highest increase in the CPT II Overall Index during the 3 h of measurement were those who oxidized the lowest amount of carbohydrates (glucose). The body has powerful and efficient self-regulating mechanisms to maintain adequate brain function. Since brain activity uses glucose as its main fuel, glucose homeostasis plays a critical role (Chugani, 1998). Children’s brains oxidize much more glucose than those of adolescents and adults; between the ages of 4 and 10 years, children utilize, for a given amount of brain tissue, twice the glucose than adults do (Scholey et al., 2009). This high glucose requirement affects mechanisms that regulate food supply, such as appetite, meal frequency and composition. Insulin, glucagon, Ghrelin, GLP-1 and PYY are, among others, important factors in the process (Delzenne et al., 2010). In our study, we found that glycemia remained fairly constant in the 3 h of fasting. This was accompanied by a progressive increase in glucagon and a reduction of insulin, promoting glycogenolysis and gluconeogenesis. Postprandial changes in plasma PYY and GLP-1 were not significantly different after fasting or after breakfast, supporting their unlikely involvement in the regulation of cognitive function in short-term fasting. On the contrary, fasting was accompanied by a modest reduction of Ghrelin, significantly lower than that following breakfast. Apparently, reduction of Ghrelin during fasting was surprising. However, a recent study showed that the spontaneous 24-h Ghrelin secretion pattern is maintained in fasting subjects, so that a reduction of Ghrelin in the late morning occurs physiologically also in the fasting condition (Koutkia et al., 2005). In our study, postprandial changes in Ghrelin after both meals were not associated with changes in cognitive function. This finding, which requires confirmation with further studies, supports the claim that in shortterm postprandial conditions, Ghrelin does not have an important impact on the regulation of cognitive function. In fasting conditions as well as after food ingestion, insulin contributed to suppress Ghrelin secretion, as previously demonstrated by others (Flanagan et al., 2003). Brain insulin resistance, recently proposed as an important factor in body weight regulation, may contribute to explain this finding (Anthony et al., 2006). Insulin has a well-known appetite-suppressing effect in the brain (Suzuki et al., 2010). Circulating insulin crosses the blood–brain barrier in a

dose-dependent manner by a saturable receptor-mediated mechanism and acts at the arcuate nucleus where receptors are highly expressed, suppressing food intake (Porte and Woods, 1981). Brain insulin resistance favors the inefficiency of insulin to suppress appetite. Increased postprandial insulin levels after breakfast may force brain insulin resistance, contributing, together with other factors, to reducing appetite. Moreover, higher circulating glucose and insulin concentrations ensured more substrate availability to brain tissue than in fasting conditions. This is accompanied by a significant (Po0.01) increase in carbohydrate oxidation in the three postprandial hours compared with baseline and an increase in non-oxidative glucose disposal, as suggested by the positive carbohydrate balance. In resting conditions, the brain is the main glucose oxidizer of the body; therefore, the postprandial increase in carbohydrate oxidation is associated, by implication, with an increase in carbohydrate oxidation in the brain, an indirect index of higher metabolic activity. Therefore, glucose supply to the brain in (shortterm) fasting conditions allows the maintenance of ‘regular’ neuronal function, whereas breakfast intake, by postprandial increase in glucose and insulin, likely leads to ‘optimizing’ neuronal functions, contributing to maintain a higher mental performance level. Recent data in adults suggest that glycemic index of the evening meal may affect cognition in the following morning. In particular, high glycemic index dinner was associated with a better verbal recall 90 min after breakfast than a low glycemic index evening meal (Lamport et al., 2011). This effect was associated with increased blood glucose in the post-breakfast phase when a high glycemic index evening meal was consumed. In our study, the impact of previous evening meal upon cognition in the morning was not explored. However, inter-individual variability of blood glucose measured 180 min after breakfast was extremely small (median 88 mg/dl; interquartile range 87.5–91 mg/dl), suggesting that an evening meal effect is unlikely. Skipping breakfast had a negative impact on visual memory function, in accordance with other studies (Rampersaud et al., 2005). However, differently from findings on attention performance, no significant association was found with carbohydrate oxidation. We do not have an explanation for this finding, but it is possible that different metabolic pathways underlie these two different cognitive functions. Since no data are available on substrate oxidation and memory function in obese children, we believe that this is an important direction for future research in the field. Verbal memory was not affected by skipping breakfast. Given the mixed findings on this issue in the available literature (Rampersaud et al., 2005), no firm conclusions can be drawn. Further studies could clarify this inconsistency. Potential limitations of our study are sample size and the lack of comparison with normal weight children. Sample size was calculated to avoid a type 2 error (power: 1b40.8) in the analyses between psychological and metabolic variables. However, post-hoc analysis showed that sample size was not European Journal of Clinical Nutrition


320 enough to exclude a type 2 error in analyzing correlations between biochemical and metabolic variables. Therefore, the nonsignificant correlation between gastrointestinal hormones and postprandial carbohydrate oxidation after breakfast found in this study should be cautiously considered. Moreover, as several variables have been included in the analysis and several tests have been performed, it is useful to be prudent also in considering differences at Po0.05. Finally, not-normal distribution of the data did not allow one to perform repeated measures analysis of variance to explore relationship between breakfast consumption and biochemical outcome. Another potential limit may be the lack of comparison with normal weight children due to Ethics Committee restrictions. However, the aim of the study was to compare different meals in obese children, who are more prone than normal weight children to have postprandial metabolic abnormalities, and who may get the greatest benefit from a healthy meal. Another potential limitation is the impact of possible habituation and fatigue effects on the neuropsychological performances. However, to our knowledge, no empirical study has specifically quantified these effects in children; therefore, we cannot establish to which extent these effects really impacted on our results. As stated previously, we chose only two subtests from the Test of Memory and Learning battery to limit the administration time and, consequently, the fatigue effect (the total administration time of the neuropsychological battery was B30 min). Most importantly, the randomized cross-over design of this study provides a way to overcome this issue. Finally, sedentary behavior could be considered as potential limitations in exporting the results in a normal daylife condition, in which children are more active. The strengths of this study include the following: (a) novelty: this is, to the best of our knowledge, the first study to explore the association between several metabolic and hormonal parameters and mental performance in two experimental conditions—fasting and after breakfast intake; (b) the specific age range: young children have a different brain metabolism than adults or adolescents, which deserves specific investigation; (c) absence of morbidity associated with obesity, such as hypertension, diabetes and dislipidemia, which also affect the brain function (Nash and Fillit, 2006). In conclusion, in a 3-h postprandial interval, metabolic and hormonal changes induced by breakfast were associated with better mental performance than fasting conditions. Reduction of carbohydrate oxidation rate was associated with decreasing mental performance in fasting children. These preliminary findings need to be confirmed by further studies conducted with larger sample size.

Conflict of interest Dr Cortese has received financial support to attend medical meetings from Eli Lilly and Company (2007–9) and Shire European Journal of Clinical Nutrition

Pharmaceuticals (2009–10), and has been a co-investigator in studies sponsored by GlaxoSmithKline (2006), Eli Lilly and Company (2007–8) and Genopharm (2008). He has served as a consultant for Shire Pharmaceuticals (2009–10). Dr Cortese is currently supported by a grant from the European Commission (‘Marie Curie’ grant for Career Development, Outgoing International Fellowship, POIF-253103). The other co-authors declare no conflict of interest.

Acknowledgements The study was sponsored by the Ministry of Health, Research Project of National Interest (PRIN) No. 2008CJ7CTW and supported by funds from the University of Verona and Galbusera Spa (I).

References Acheson KJ, Schutz Y, Bessard T, Flatt JP, Je´quier E (1987). Carbohydrate metabolism and de novo lipogenesis in human obesity. Am J Clin Nutr 45, 78–85. Anthony K, Reed LJ, Dunn JT, Bingham E, Hopkins D, Marsden PK et al. (2006). Attenuation of insulin-evoked responses in brain networks controlling appetite and reward in insulin resistance: the cerebral basis for impaired control of food intake in metabolic syndrome? Diabetes 55, 2986–2992. Bandini LG, Schoeller DA, Dietz WH (1990). Energy expenditure in obese and nonobese adolescents. Pediatr Res 27, 198–203. ˜eda TR, Tong J, Datta R, Culler M, Tscho ¨ p MH (2010). Castan Ghrelin in the regulation of body weight and metabolism. Front Neuroendocrinol 31, 44–60. Chugani HT (1998). A critical period of brain development: studies of cerebral glucose utilization with PET. Prev Med 27, 184–188. Cole TJ, Bellizzi MC, Flegal KM, Diez WH (2000). Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 320, 1240–1243. Conners CK (2002). Conners’ Continuous Performance Test. MultiHealth System: Toronto, Canada. Delzenne N, Blundell J, Brouns F, Cunningham K, De Graaf K, Erkner A et al. (2010). Gastrointestinal targets of appetite regulation in humans. Obes Rev 11, 234–250. Flanagan DE, Evans ML, Monsod TP, Rife F, Heptulla RA, Tamborlane WV et al. (2003). The influence of insulin on circulating ghrelin. Am J Physiol Endocrinol Metab 284, E313–E316. Hallschmid M, Schultes B (2009). Central nervous insulin resistance: a promising target in the treatment of metabolic and cognitive disorders? Diabetologia 52, 2264–2269. Hoyland A, Dye L, Lawton CL (2009). A systematic review of the effect of breakfast on the cognitive performance of children and adolescents. Nutr Res Rev 22, 220–243. Koutkia P, Schurgin S, Berry J, Breu J, Lee BS, Klibanski A et al. (2005). Reciprocal changes in endogenous ghrelin and growth hormone during fasting in healthy women. Am J Physiol Endocrinol Metab 289, E814–E822. Irwin N, Gault V, Flatt PR (2010). Therapeutic potential of the original incretin hormone glucose-dependent insulinotropic polypeptide: diabetes, obesity, osteoporosis and Alzheimer’s disease? Expert Opin Invest Drugs 19, 1039–1048. Istituto Nazionale della Nutrizione. (1977). Tabelle di Composizione Degli Alimenti. A cura di Carnovale E, Marletta L. Occhipinti Sisar & Gioja Press: Milan. Lamport DJ, Hoyle E, Lawton CL, Mansfield MW, Dye L (2011). Evidence for a second meal cognitive effect: glycaemic responses


321 to high and low glycaemic index evening meals are associated with cognition the following morning. Nutr Neurosci 14, 66–71. Li Y, Dai Q, Jackson JC, Zhang J (2008). Overweight is associated with decreased cognitive functioning among school-age children and adolescents. Obesity (Silver Spring, MD) 16, 1809–1815. Lusk G (1928). The Elements of the Science of Nutrition, 4th edn. WV Saunders Co: Philadelphia. Maffeis C, Pinelli L, Schutz Y (1995). Increased fat oxidation in prepubertal obese children: a metabolic defense against further weight gain? J Pediatr 126, 15–20. Maffeis C, Schutz Y, Schena F, Zaffanello M, Pinelli L (1993). Energy expenditure during walking and running in obese and nonobese prepubertal children. J Pediatr 123, 193–199. Maffeis C, Schutz Y, Grezzani A, Provera S, Piacentini G, Tato` L (2001). Meal-induced thermogenesis and obesity: is a fat meal a risk factor for fat gain in children? J Clin Endocrinol Metab 86, 214–219. McNay EC, Ong CT, McCrimmon RJ, Cresswell J, Bogan JS, Sherwin RS (2010). Hippocampal memory processes are modulated by insulin and high-fat-induced insulin resistance. Neurobiol Learn Mem 93, 546–553. Micha R, Rogers PJ, Nelson M (2010). The glycaemic potency of breakfast and cognitive function in school children. Eur J Clin Nutr 64, 948–957. Nash DT, Fillit H (2006). Cardiovascular diseases risk factors and cognitive impairment. Am J Cardiol 97, 1262–1265. ¨ rck I (2009). Effects of differences in Nilsson A, Radeborg K, Bjo postprandial glycaemia on cognitive functions in healthy middleaged subjects. Eur J Clin Nutr 63, 113–120.

Porte Jr D, Woods SC (1981). Regulation of food intake and body weight in insulin. Diabetologia 20(Suppl), 274–280. Rampersaud GC, Pereira MA, Girard BL, Adams J, Metzl JD (2005). Breakfast habits, nutritional status, body weight, and academic performance in children and adolescents. J Am Diet Assoc 105, 743–760; quiz 761–762. Reynolds CR (1994). Bigler Test TEMA—Memoria e apprendimento. Ed Erickson: Trento. Reynolds CR, Bigler ED (1994). Test of Memory and Learning (TOMAL). Pro-Ed: Austin, TX. ¨ nram-Lea SI, Greer J, Elliott J, Kennedy DO (2009). Scholey AB, Su Glucose administration prior to a divided attention task improves tracking performance but not word recognition: evidence against differential memory enhancement? Psychopharmacology (Berl) 202, 549–558. Suzuki K, Simpson KA, Minnion JS, Shillito JC, Bloom SR (2010). The role of gut hormones and the hypothalamus in appetite regulation. Endocr J 57, 359–372. Tanner JM, Whitehouse RH (1976). Clinical longitudinal standards from birth to maturity for height, weight, velocity and stages of puberty. Arch Dis Child 51, 170–179. Wesnes KA, Pincock C, Richardson D, Helm G, Hails S (2003). Breakfast reduces declines in attention and memory over the morning in schoolchildren. Appetite 41, 329–331. ¨ller K, Hille K, Klenk J, Weiland U (2008). Influence of Widenhorn-Mu having breakfast on cognitive performance and mood in 13- to 20-year-old high school students: results of a crossover trial. Pediatrics 122, 279–284.

European Journal of Clinical Nutrition