- Email: [email protected]
Energy Intake Compensation After 3 Weeks of Restricted Energy Intake in Young and Elderly Men
Energy Intake Compensation After 3 Weeks of Restricted Energy Intake in Young and Elderly Men Renate M. Winkels, PhD, Angelique Jolink-Stoppelenburg, PhD-fellow, Kees de Graaf, Prof, Els Siebelink, Research Dietician, Monica Mars, PhD, and Lisette de Groot, Prof
Objectives: Decreased energy intake in older persons poses these people at risk of progressive weight loss. It may result from a failure to regulate energy intake and expenditure after periods of underfeeding. The objective of this study was to investigate if a period of underfeeding differentially influences energy intake of older compared with young men and, additionally, to study potential underlying mechanisms, namely changes in gastric emptying rate and cholecystokinin (CCK) levels in blood. Design/setting: Dietary intervention of 3 phases. After a phase of energy balance, we fed participants in phase 2 by a mean of 70% of their needs for 21 days. During phase 3, we assessed ad libitum energy intake of the participants during 9 days. At the end of phases 1 and 2, we assessed appetite, gastric emptying, and CCK levels in blood in response to a test meal.
Results: During energy balance, mean energy intake of young men (14.3 2.3 MJ/day) was significantly higher than that of older men (11.3 1.8 MJ/day, P \ .001). After the period of underfeeding, energy intake in phase 3 amounted to 16.3 2.6 MJ/day in young men and to 14.4 3.2 MJ/day in older men. Ad lib energy intake after underfeeding did not differ between young and older men (analysis of covariance, with energy intake during phase 1 as covariate, P 5 .99). There were no differential changes in body weight, body composition, resting energy expenditure, gastric emptying rate, CCK-8 levels, and appetite between young and older men during the study. Conclusion: Our results do not indicate that older men have an impaired ability to control energy intake after a period of underfeeding compared with younger men. Trial registration: NCT00561145. (J Am Med Dir Assoc 2011; 12: 277–286)
Participants: Fifteen young (age 24 years [range 20–34], body mass index 23.0 kg/m2 2.3) and 17 older (age 68 years [64–85], body mass index 24.5 kg/m2 1.9) men participated in this study.
Keywords: Energy intake; gastric emptying; aging; energy restriction
Food intake decreases with ageing. The decrease in food intake tends to be higher in elderly people than the decrease in energy expenditure, resulting in unintentional loss of predominantly lean body mass.1,2 Both physiological and nonphysiological factors contribute to the decline in energy intake. Weight loss presents itself in healthy elderly as well as in elderly with compromised health.3,4 Loss of lean body mass, which is associated with functional impairments
and disability, predisposes the elderly to malnutrition, morbidity, and mortality.1,2,5–7 The unintentional weight loss in older people might be triggered by an impaired ability to regulate energy balance; however, data are equivocal. Some studies suggest that older adults do not compensate their energy intake after a period of underfeeding, and do not regain the weight lost during underfeeding, whereas young adults do increase their energy intake and regain the weight lost during energy restriction.8–10 It has been hypothesized that this is related to a slower gastric emptying rate in elderly,11 although the response in gastric emptying rate to energy restriction has not been studied. However, other investigators could not replicate age-related differences in energy intake compensation after energy restriction, neither could they show age-related differences in related metabolic responses.12 We studied the effect of underfeeding on compensation in energy intake and resting energy expenditure, body weight,
Department of Human Nutrition, Wageningen University, Wageningen, the Netherlands (R.M.W., A.J.-S., K.d.G., E.S., M.M., L.d.G.). This study was financially supported by the NZO (Dutch Dairy Association), Zoetermeer, the Netherlands. Address correspondence to Renate M. Winkels, PhD, Wageningen University, Human Nutrition, Bomenweg 4, 6703 HD Wageningen, the Netherlands. E-mail: [email protected]
Copyright Ó2011 American Medical Directors Association DOI:10.1016/j.jamda.2010.08.011 ORIGINAL STUDIES
Winkels et al 277
and body composition in young and older men, to contribute to the evidence base on the causes of decreased ability to regulate energy balance in older people and to clarify current controversy. In addition, we studied whether gastricemptying rates, appetite ratings, and cholecystokinin (CCK) levels in blood mediated any differential responses to energy restriction between young and older men. Gastric-emptying rate may be an important regulator of food intake.13 CCK causes gallbladder contraction and stimulates the release of several digestive enzymes in the duodenum13 and slows gastric emptying.14 CCK decreases food intake;15 it has been suggested that increased CCK activity in elderly may contribute to anorexia of aging.15 SUBJECTS AND METHODS Subjects We recruited 55 men, by publishing advertisements in local newspapers and by sending out general e-mails to an e-mail list of persons who had indicated their interest in participating in studies of our university. We included only men, as hormonal fluctuations in young women during the menstrual cycle may influence body weight and food intake. During a screening visit, men donated a fasting blood sample and gave written informed consent. In the fasting blood sample, we measured glucose, hemoglobin, and hematocrit. Men were excluded based on the following exclusion criteria: body mass index (BMI in kg/m2) less than 20 or higher than 30, adherence to a weight-reduction or medically prescribed diet, dementia (Mini-Mental State Examination score \21), diabetes, anemia, gastrointestinal disorders, use of drugs known to interfere with energy balance, or a history of medical or surgical events known to affect the study
outcome. Based on these criteria, 32 men (15 young and 17 older men) were included in the study. The Medical Ethical Committee of our university approved the study. Design The study consisted of 3 subsequent phases: phase 1 was a 14-day period of energy balance (days 1–14), phase 2 was a 21-day 30% energy restriction period (days 15–35), and phase 3 was a 9-day ad libitum food intake period (days 36–44). Before phase 1, dieticians from our study center estimated the habitual energy intake of the subjects with a validated food-frequency questionnaire.16 During all phases of the study, all foods and beverages were provided by the study center. At the end of phase 1 (morning of day 15) and at the end of phase 2 (morning of day 36), all participants took part in a gastric-emptying test procedure. Phase 1: Energy Balance During days 1 through 14, phase 1 of the study, we provided the participants with a diet containing approximately 90 energy percent of their estimated total daily energy requirement (Table 1). Participants chose the remaining 10% of energy from a list of choice items. This list contained fruits and vegetables, sweet sandwich fillings, crackers, candies, and nonalcoholic and alcoholic drinks. Participants could freely choose from this list, but daily had to consume the choice items. Their choice of items was recorded by the dieticians. Participants kept a diary to record any deviations from the diet, illness, and use of drugs. We weighed the participants daily and adjusted their diet to a higher or lower energy content if participants
Table 1. Foods and Drinks Provided during the 3 Phases of the Study, for a Participant with an Energy Intake of 11 MJ/day during Energy Balance (Phase 1) Product
Energy balance (phase 1))
Energy restriction (phase 2) †
Ad libitum (phase 3) ‡
Bread Margarine Sandwich fillings
231 g (7 slices) 40 g 48 g 16 g (1 portion) 54 g (3 portions) 3 200 mL 250 g (2 pieces) 30 g (3 cookies) 150 g 150 g 70 g 76 g 50 g 18 g 210 g 1.1 MJ
162 g (5 slices) 27 g 32 g 16 g (1 portion) 18 g (1 portion) 1 167 mL 125 g (1 piece) 20 g (2 cookies) 130 g 130 g 45 g 66 g 36 g 13 g 140 g 0.7 MJ
462 g (21 buns) 80 g 150 g 32 g (2 portions) 108 g (6 portions) 6 500 mL 500 g (4 pieces) 60 g (6 cookies) 300 g 300 g 140 g 152 g 100 g 36 g 420 g 1.1 MJ
Eggs (per week) Milk Fruit Cookies Hot meal
Salad Salad dressing Dessert Choice items 1
Sweet fillings Cheese Meat
Potatoes, rice, or pasta Boiled vegetables Sauce Meat
* During phase 1, participants had to be weight stable. Food and drinks covered 90 energy percent of the daily needs. Participants chose the remaining 10 energy percent from a list of choice items. During all phases, the amount of energy that participants had to ingest with the choice items was prescribed by the dieticians; in phase 3 the amount of energy in these items was the same as in phase 1. † During phase 2, participants consumed 70% of the energy intake in phase 1. ‡ During phase 3, all products were provided as 200% of the amount in phase 1. Participants were requested to eat as much or little as to feel normally satiated. 278 Winkels et al
JAMDA – May 2011
progressively lost or gained weight (.0.2 kg/d). Participants were weight stable during days 12, 1, and 14 of phase 1, the mean energy intake during these 3 days was considered as their energy intake for energy balance. We collected duplicates of the diets that the participants consumed and stored these duplicates at –20 C until analysis. During phase 1, we collected a duplicate diet for a fictive participant with an energy intake of 11 MJ/d based on food tables. The diet mimicked a typical Dutch diet: 33 energy percent of fat, 52 energy percent of carbohydrates, 13 energy percent of protein, and 2.6 g of fiber/MJ based on the Dutch food table.17 During all 3 phases of the study during weekdays, participants consumed a hot lunch at the study center. After lunch, participants received a takehome package with snacks and foods and beverages for their evening meal and for breakfast. On Fridays, subjects received a package with food items and beverages for the weekend, plus instructions for the preparation of the meals. We weighed out all foods and beverages for each participant. There was a 14-day menu cycle that was repeated every 14 days. Phase 2: Energy Restriction During days 15 through 35, phase 2 of the study, we provided the participants with a diet that contained 70 energy percent of the energy that each participant consumed during the last 3 days of phase 1 (Table 1). To compose this energyrestricted diet, the dieticians multiplied the amount of a product in phase 1 by 0.7 and rounded this off to obtain a distributable amount of this product. For example, 231 g of bread*0.7 5 162 g of bread is equal to 4.9 slices: this was rounded of to 5 slices. The macronutrient distribution of the diet in phase 2 was similar to phase 1. As in phase 1, approximately 90 energy percent of the energy-restricted diet was prescribed by the dieticians; participants chose the remaining 10% of energy from a list of choice items. To analyze the amount of energy and micronutrients in this diet, we collected a duplicate diet for a fictive participant with a daily energy intake of 7.7 MJ during phase 2. Phase 3: Ad Libitum In phase 3, participants consumed ad libitum from a diet provided by the study center, (Table 1). Participants were asked to consume as much or little of the foods and beverages as required to feel normally satiated. Most products were provided as 200% of the amount in phase 1. We provided foods in uncommon portion sizes to prevent participants from consuming amounts similar to those consumed habitually. In addition to the products provided by the study center, participants had to consume the choice items, as in phases 1 and 2. The items were also provided by the study center and the total amount of energy in these items was the same as during phase 1. For example, if a participant daily had to select 1.1 MJ from choice-items in phase 1, this participant also had to select 1.1 MJ/day from choice-items in phase 3. We collected and weighed leftovers and empty packages to record food energy intake.
ORIGINAL STUDIES
Energy in the Diet After the study, we thawed all the duplicates of the diets we collected during phase 1 and phase 2. We pooled the duplicates per phase and homogenized them. Total fat, protein, dry matter, ash, and fiber were analyzed in these duplicates, available carbohydrates were determined by difference. From this, we calculated the amount of fat, protein, carbohydrates, fiber, and energy in the duplicates. The mean energy intake of participants during phases 1 and 2 was calculated and subsequently adjusted for the difference between calculated and analyzed amount of energy in the diet. For phase 3, reported energy intakes were based on the food table only, because participants could eat ad libitum in that phase, which made the collection of duplicates not possible. Anthropometry, Body Composition, Respiratory Quotient, Resting Energy Expenditure, and Physical Activity We measured anthropometry, body composition, resting energy expenditure (REE), and respiratory quotient (RQ) on the last day of phase 1 (day 14), on the last day of phase 2 (day 35), and on the last day of phase 3 (day 44). On these days, we transported the participants by car from their residence to the study center. Participants had to be fasted and were instructed to avoid intensive physical activity on the morning of the measurements. The same observer performed all the anthropometric and body composition measurements. We weighed the participants to the nearest 0.1 kg using a digital weighing scale (Seca, Bascule, MT). Participants wore only their underwear during weighing and were asked to void beforehand. We did not inform participants about their weight and asked them not to weigh themselves during the study. Height was measured to the nearest 0.1 cm using a microtoise. Fat mass and bone mass were determined using a dual energy X-ray absorptiometer (DEXA, model DPX-L, Lunar Radiation Corp., Madison, WI, software version 1.31). Lean body mass was calculated from total body mass minus fat mass and bone mass. Appendicular lean soft tissue (ALST) was assessed with DEXA. Smooth muscle mass was estimated from ALST with the following formula:18 Total body SM5ð1:13 ALSTÞ ð0:02 ageÞ 1 1:58 REE was measured using indirect calorimetry (Deltatrac, Madison, WI). Participants lay on a bed for 35 minutes and were asked to stay awake and motionless. Resting VO2 and VCO2 were measured during the last 25 minutes. Weir’s equation was used to calculate REE.19 We monitored physical activity of the participants by assessing the number of steps taken each day with pedometers (Yamax Digi-walker, SW-200, Tokyo, Japan). Men were asked to maintain their usual lifestyle and activity pattern during the complete study period.
Winkels et al 279
Appetite Rating, Gastric Emptying Rate, and CCK Levels in Blood in Response to a Test Meal All participants took part in 2 gastric-emptying meal tests: at the end of energy balance (morning of day 15) and at the end of energy restriction (morning of day 36). Participants arrived at the study center in fasting state on the morning of these tests. A research nurse inserted a catheter in the antecubital vein of participants to facilitate the drawing of multiple blood samples during the morning. Saline was injected into the catheter after drawing of a blood sample, to prevent clogging. The first tube of each time point of blood sampling was discarded to remove the saline. Fifteen minutes before the start of the test procedure, a fasting blood sample was taken. In addition, participants rated their appetite by scaling their hunger, fullness, desire to eat, and prospective food consumption on a 10-cm visual analogue scale20,21 with statements expressing extremes (not at all versus very little/very much) and participants collected a breath sample with a straw into a 12-mL tube. The test procedure started with the consumption of a test meal at a steady pace during a period of approximately 10 minutes, followed by a glass of water. The test meal consisted of a pancake containing 100 mg of 13C-labeled octanoic acid (octanoic acid-1–13C, 99 atom%, OBT grade, Isotech, Miamisburg, OH). 13C-octanoic acid is retained in the solid phase of the meal in the stomach, but is rapidly absorbed and oxidized into 13CO2 on arrival in the duodenum. The rate of appearance of 13CO2 in breath is therefore a measure of gastricemptying rate.22 The pancake contained 2.1 MJ (506 kcal) and consisted of 18 energy percent protein, 52 energy percent carbohydrate, and 30 energy percent fat. During the day of the test meal, the amount of energy in the diet was corrected to adjust for the amount of energy in the pancake. Breath samples were collected every 15 minutes after the initiation of the meal for a period of 4 hours. Participants rated their appetite and donated a blood sample 15, 30, 45, 60, 90, 120, and 180 minutes after the initiation of the meal. The total areaunder-the-curve (AUC) was calculated for appetite scores. In breath samples, the 13C-enrichment of exhaled CO2 was analyzed with Isotope Ratio Mass Spectrometry. Total CO2 production was assumed to be 300 mmol/(m2*hour); body surface area in m2 was calculated with the formula of Haycock et al.23 Recovery curves of 13C in breath and cumulative curves over time were plotted and fitted according to formula 1 or 2 respectively:22,24 Formula 1: percentage of excretion per hour of the given 13 C-dose: y 5 atb ect , where t 5 time in hours; and a, b, and c are constants Formula 2: percentage of cumulative excretion per hour of the given 13C-dose: y 5 mð1 ekt Þb , where t 5 time in hours and m, k, and b are constants with m being the total cumulative recovery when time is infinite. We used an MS Excel macro sheet (Microsoft Corp, Redmond, WA) provided by the University of Leuven22 to derive 2 gastric-emptying parameters from these curves: the recovery time of 50% of the maximal recovered 13C (gastric half emptying time T1/2) and the time when the peak of the 13C excretion
280 Winkels et al
curve is reached (lag phase Tlag). We retained the value of the T1/2 and Tlag for the best-fitting curve. In blood, CCK-8 levels were analyzed with a radio-immuno assay (CCK-8 Euro-Diagnostics, Malmo¨, Sweden).25 This CCK assay has been optimized to reach a sensitivity of 0.05 pmol/L with no cross-reactivity toward gastrin-17, and sulphated gastrin. The intra- and interassay CV ranged from 9% at lower concentrations (0.85 pmol/L) to 5% at higher concentrations (2.04 pmol/L). We calculated the total AUC of CCK-levels in blood plotted against time in minutes. Statistical Analysis Our primary outcome was the difference in energy intake compensation after energy restriction between young and older men. To test this we used analysis of covariance (ANCOVA). We compared ad libitum energy intake during phase 3 in young men with that in elderly men and inserted energy intake during energy balance (phase 1) as covariate in the model. Additionally, we used analysis of variance (ANOVA) for repeated measures to assess if body weight, body composition, energy expenditure, and RQ differentially changed between young and older men during the 3 phases of the study (P for interaction between group and outcome). To test whether gastric-emptying rate, CCK levels, and appetite ratings differentially responded to energy restriction in young versus older men, we used mixed model ANOVA. In these models, we included group (young versus older men), time (phase 1: energy balance versus phase 2: energy restriction) and group*time effects; time was included as a repeated statement. Our main interest was whether the group*time interaction was significant. For the various appetite ratings, we used a mixed model ANOVA in which we added baseline appetite ratings as covariates to the model to adjust for baseline differences in these ratings. We used least square means to calculate the AUC adjusted for baseline differences. For gastric half-emptying time (T1/2), gastric emptying lag time (Tlag), AUC of the CCK and 13 C-excretion curves, we used a comparable mixed model ANOVA, but did not include covariates. Because CCK data were not normally distributed, these data were logtransformed before testing. Data were analyzed using PASW Statistics version 17.0.3 (SPSS Inc, Chicago, IL). Results were considered statistically significant at the .05 level. RESULTS All subjects completed the study. Young men were slightly taller than older men, but their BMI did not differ statistically significantly before the start of the intervention (Table 2). Fasting glucose levels were slightly higher in older men. The amount of energy and the macronutrient composition of the diet as chemically analyzed were comparable with what was expected based on the Dutch food table. During phase 1, the amount of energy in a prescribed daily diet of 9.9 MJ/d according to food tables contained
JAMDA – May 2011
Table 2. Baseline Characteristics of the Participants, Assessed before the Start of the Intervention
Age, y Height, m Weight, kg Body mass index, kg/m2 Glucose, mmol/L Hemoglobin, g/L Hematocrit
Young men (n 5 15)
Older men (n 5 17)
24 [20, 34] 1.81 0.08 75.8 11.3 23.0 2.3 4.5 [3.7, 5.1] 9.4 0.5 45% 2
68 [64, 85] 1.76 0.05* 75.8 7.6 24.5 1.9 5.0 [4.6, 5.5]* 9.4 0.4 45% 2
Data are means standard deviation, or median [range]. * Statistically significantly different between young and older men by unpaired t test for normally distributed variables or by Mann-Whitney U test (glucose).
9.6 MJ/d according to chemical analysis. During phase 2, a prescribed diet of 6.9 MJ/d based on food tables contained 6.8 MJ according to chemical analysis. Analyzed amounts of macronutrients were as follows: protein 12 to 14 energy%, carbohydrates 48 to 54 energy%, fat 33 to 37 energy%, fiber content 3.0 to 3.2 g/MJ (20–31 g/d). Mean energy intake during energy balance (phase 1) was significantly higher in young than in older men: 14.3 MJ/d for young men and 11.3 MJ/d for older men (Table 3, P \ .001). During energy restriction (phase 2), the energy intake was 30% lower than needed for energy balance: for young men the absolute deficit in energy was 4.3 MJ/d, for older men it was 3.4 MJ/d. Mean energy intake during the period of ad libitum food intake (phase 3) was 116% of energy intake during energy balance (phase 1) in young men and 127% in older men (Figure 1). Energy intake compensation after energy restriction did not differ between young and older men: energy intake in phase 3 did not differ between young and older men (tested with ANCOVA, Table 3, P 5 .99). Analysis of changescores yielded similar results: change in energy intake between phase 3 and phase 1 did not differ between young and older men (data not shown). For both young and elderly men, the ad libitum energy intake during the weekdays of phase 3 of the study tended to be higher (16.1 2.4 MJ for young and 14.2 3.2 MJ for elderly men during weekdays) than the energy intake during weekend days (15.1 2.7 MJ for young and 13.5 3.5 MJ for elderly men during weekend days). We did not find differences in product choice behavior between young and older men during phase 3 (data not shown). Weight loss during phase 2 was similar for both groups: 2.2 SD 0.9 kg for young men and 2.1 0.9 kg for older men, whereas weight gain was similar during phase 3: 1.4 1.3 kg for young men and 1.5 1.1 kg for older men. Older men had statistically significantly less smooth muscle mass and lower resting energy expenditure than young men had (Table 3, group effects), but there were no differences in body weight, fat mass, lean body mass, RQ, or physical activity. There were no differential changes between young and older men during the 3 phases of the study in any of these variables (Table 3, interaction tests). ORIGINAL STUDIES
During the test meal procedures, breath and blood samples of one young and breath samples of one older man could not be collected because of technical difficulties. The excretion curves of 13CO2 after consumption of a test meal labeled with 13C-octanoic acid in phases 1 and 2 were similar for young and elderly men (Figure 2). The change in gastric-emptying lag time (Tlag) from phase 1 to phase 2 differed between young and older men (P 5 .036 for group*time interaction), but not for gastric half emptying time (P 5 .098). There were no statistically significant group or time effects. CCK-8 concentrations in blood increased after consumption of the test meal in young and older men (Figure 3), both during energy balance and during energy restriction. There was no differential change between young and old men from phase 1 to phase 2 in total AUC (P 5 .20 for group*time interaction), neither were there statistically significant group or time effects in total AUC. Rated feelings of hunger and appetite are shown in Figure 4. The rating of these scores did not differentially change from energy balance to energy restriction in young men compared with older men: there were no statistically significant group*time-interaction effects (P values for group*time interaction: P 5 .34 for hunger, P 5 .37 for desire, P 5 .18 for fullness, and P 5 .50 for prospective consumption). For fullness, total AUC was significantly greater in phase 2 than in phase 1 (P for time-effect phase 1 versus phase 2 5 .013), but not for the other appetite ratings. There were no statistically significant group effects in total AUC for appetite ratings. DISCUSSION We did not find differences in the ability of older men to compensate their energy intake or expenditure after a period of undereating compared with young men. Young and older men both lost weight during a period of energy restriction; energy compensation behavior did not differ between young and older men after this period of energy restriction. There was only a slight differential response between young and older men in gastric-emptying lag time in response to a test meal. We did not find differential changes between young and elderly men in any of the other physiological responses. Thus, we were not able to confirm the findings from others8,9 that older men are less well able to compensate energy balance after a period of energy restriction than young men. We strictly controlled energy intake during phases 1 and 2 of our study, and carefully monitored intake during phase 3. In addition, we monitored energy expenditure by assessing resting energy expenditure and by assessing physical activity of the participants and we studied the effect of energy restriction on gastric-emptying rate and CCK excretion after a test meal. Within this strictly monitored regime, we did not find any evidence that regulation of energy balance in older men is less well maintained than in young men. Gastric-emptying rate may be an important regulator of food intake, although other factors like gastric distension and meal composition probably also contribute to satiation and meal termination.13 Habitual dietary intake may affect Winkels et al 281
282 Winkels et al
Table 3. Energy Intake, Body Weight, Body Composition, Resting Energy Expenditure, RQ and Physical Activity Of Young (N515) And Older Men (N517) During The Three Phases Of The Intervention. Data are means standard deviation
Young men Older men
Young men Older men Young men Older men Young men Older men Young men Older men Young men Older men Young men Older men Young men Older men
Energy intake in MJ/d
Body weight in kg Fat mass in kg Lean body mass in kg Smooth muscle mass in kg Resting energy expenditure in MJ/d RQ Physical activity in nr of kilo-steps/d
Phase 1
Phase 2
Phase 3
Energy Balance, 100% of Energy Needs
Energy Restriction, 70% of Energy Needs
Ad libitum, 200% of Energy Needs Provided
14.3 2.3† 11.3 1.8†
10.1 1.6 7.9 1.3
15.9 2.4 14.0 3.2
74.5 10.6 74.1 7.5 13.1 7.0 15.4 5.1 58.5 6.3 56.1 4.9 32.4 3.9
72.2 10.9 72.1 7.6 11.9 6.8 14.2 5.6 57.7 6.7 55.5 5.0 32.1 3.9
73.2 10.6 73.6 7.1 12.2 6.7 14.9 5.4 58.4 6.9 56.5 4.4 32.2 3.9
29.0 3.0 6.2 0.9
28.7 2.9 6.1 0.9
29.1 2.7 6.4 0.5
5.8 0.6 0.86 0.04 0.86 0.04 9.1 2.0
5.6 0.5 0.88 0.04 0.85 0.04 8.6 2.2
5.9 0.5 0.89 0.04 0.89 0.06 8.2 2.7
7.9 3.9
8.2 3.8
7.8 4.5
ANCOVA
P 5 .99* ANOVA for repeated measures Differences Differences Interaction within persons between (group* groups variable) P \.001 P 5 .94 P 5 .67 P \.001
P 5 .23
P 5 .94
P \.001
P 5 .32
P 5 .41
P 5 .025
P 5 .012
P 5 .56
P 5 .019
P 5 .050
P 5 .99
P 5 .049
P 5 .15
P 5 .37
P 5 .20
P 5 .55
P 5 .33
ANCOVA, analysis of covariance; RQ, respiratory quotient. * We tested the difference between young and older men in phase 3; in these analyses we inserted energy intake during energy balance (phase 1) as covariate in the model. † Differed statistically significantly between young and older men during phase 1, P \.05. JAMDA – May 2011
180 170
Energy intake during phase 3 as % of intake during phase 1
160 150 140 130 120 110 100 90 80 70 60 50 36
37
38
39
40
41
42
43
44
day Fig. 1. Mean energy intake ( SEM) during the 9-day period following energy restriction relative to the energy intake during weight maintenance in young (B) and older men (-). Energy intake during phase 1 was 100% (dotted line $$$$$), energy intake during phase 2 was 70% (dashed line - - - - - -).
the gastric-emptying response to a test meal.26,27 Because earlier studies suggested that gastric-emptying rate may be delayed in elderly men,28,29 we hypothesized that an altered gastric-emptying response to a lower habitual energy intake may be responsible for a dysregulation in energy balance in elderly. The response in gastric-emptying lag time to energy restriction in elderly men was statistically significantly different from the response in young men: in elderly men it changed from 132 minutes during energy balance to 145 minutes during energy restriction, whereas in young men it changed from 143 to 138 minutes (Figure 2). However, this could be a chance finding, as we did not correct for multiple testing. Moreover, it is debatable whether this minimal effect is physiologically relevant. Gastric half-emptying time did not differentially change in young versus elderly men in response to energy restriction, which is in line with our findings of CCK in blood: there was no differential response in CCK after energy restriction in young versus elderly men. Thus, our data do not provide substantial evidence for a different response to energy restriction in gastric-emptying rate in elderly men than in young men. Our findings are in line with the results from Yukawa et al12; older participants in that study also increased energy intake after a period of energy restriction and experienced an increase in body weight during that period. Although, Yukawa et al12 included men and women, whereas we included men only, both their and our study are not in agreement with the findings from the earlier studies from Roberts and coworkers.8–10 As expected, we found that energy intake during energy balance of elderly men was lower than that of young men. Also, elderly men had lower smooth muscle mass and lower ORIGINAL STUDIES
resting energy expenditure than young men. Nevertheless, we could not show an age-related difference in energy intake compensation. Possibly, dysregulation of energy balance occurs only at very old age, or in persons with underlying illnesses. Yet, age, weight, BMI, health status, and energy intake during energy balance in our study were comparable with these characteristics in the studies from Roberts and coworkers,8–10 who did find a diminished regulation of energy intake in elderly. Subtle differences in study design may have contributed to the opposing findings. In the studies from Roberts and coworkers,8–10 the absolute energy restriction was equal for old and young intervention groups, whereas in the study from Yukawa et al12 and our own study the relative energy restriction was equal between young and old groups. However, this small difference in study design will probably not fully explain the opposing findings. Absolute energy restriction was higher for young men than for older men in our study. Nevertheless, weight loss during phase 2 was similar for both groups (Table 3). For each individual, we compared the observed weight loss, with the expected weight loss, based on the assumption that an energy deficit of 3.2 MJ causes 100 g weight loss30 (data not shown). For 3 young men and 1 older man weight loss during phase 2 was much lower than expected. However, exclusion of these participants did not influence our outcomes. Regulation of food intake is a complex process, which involves neural, endocrine, and hedonic factors.31 Moreover, cognitive aspects, social factors, and atmosphere during mealtime are important determinants of energy intake.32,33 The older participants in our study were all raised just after the second world war. During that period, food was scarce in the Netherlands, and therefore many participants were taught during childhood to always finish their plate. Some of the older participants expressed that it was a waste of food to leave any leftovers on their plate during phase 3 of the study, although we instructed the participants to consume as much food to feel normally satiated. Although speculative, these childhood beliefs may have caused the older participants to overeat during phase 3. During weekdays, the participants of our study consumed their hot meal at the study center. Most older participants were retired and tended to stay long during meal times to enjoy their time with the other participants. Most young men had to return to their daily activities quickly after meal times. We therefore hypothesize that particularly the older men may have eaten more during phase 3 of the study than they might have eaten in their own residence. Indeed, our results show that participants of our study tended to eat more during weekdays, when they consumed their hot meal at the study center, than during weekend days, when they consumed all meals in their own residence. Nevertheless, both young and older men tended to have a higher energy intake during weekdays than on weekend days, and energy intake during weekend days was still higher than required for energy balance. Although consumption of meals in the earlier-mentioned studies8–10,12 also partly took place at the various research centers, we cannot compare their exact setting and ambiance, which may partly explain differences in findings among studies. Winkels et al 283
13
C recovery in breath (% of the administered dose/hour)
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0
30
60
90
120
150
180
210
240
time (minutes) 250 gastric half-emptying time T1/2 (min)
gastric emptying lag time Tlag (min)
200 150 100 50 0
200 150 100 50 0
young men elderly men
young men elderly men
Fig. 2. Mean recovery of 13C in breath (SEM) of young and elderly men after a pancake meal with 13C-octanic acid at t 5 0 minutes (hatched bar). Gastric emptying rate was determined at the end of the phase of energy balance (day 15) and at the end of the phase of energy restriction (day 36) in n 5 14 young men and n 5 16 elderly men. In the upper graph: C, young men energy balance; B, young men energy restriction; -, elderly men energy balance; ,, elderly men energy restriction. In bar graphs, black bars represent energy balance, white bars energy restriction.T1/2 did not and Tlag did differentially change from energy balance to energy restriction in young compared to elderly men (P 5 .098 for T1/2 and P 5 .036 for Tlag for group*time interaction).
2.5
1.5 800
1.0
CCK: total AUC
CCK-8 (pmol/L)
2.0
0.5
0.0
600 400 200 0
0
30
60
90
120
150
180
young men
elderly men
time (minutes)
Fig. 3. Mean concentration of CCK-8 in blood (SEM) of young and elderly men after a pancake meal at t 5 0 (left graph) and total AUC under the CCK-graph (right graph) assessed during energy balance (day 15) and energy restriction (day 36) in n 5 14 young men and n 5 17 elderly men. In the left graph: C, young men energy balance; B, young men energy restriction; -, elderly men energy balance; ,, elderly men energy restriction. In the whisker graph medians 1 ranges are plotted, gray boxes are during energy balance, white boxes are during energy restriction. There were no statistically significant group, time, or group*time-interaction effects in total AUC. 284 Winkels et al
JAMDA – May 2011
10
CONCLUSION
Hunger (cm)
8
6
2000
Hunger: total AUC
4
2
1000 500 0
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
1500
young men
elderly men
time (minutes) 10
Desire (cm)
8
6
2000
Desire: total AUC
4
2
ACKNOWLEDGMENTS
1000 500 0
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
1500
young men
elderly men
time (minutes) 10
We thank all participants for their enthusiastic participation in the trial and the dieticians, students, and research nurses for their assistance during the trial. We thank Dr Kristen Verbeke of the Katholic University in Leuven for kindly providing us with the Excel program for modeling the gastric emptying data.
6
REFERENCES 4
Fullness: total AUC
Fullness (cm)
8
2
2000 1500 1000 500 0
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
young men
elderly men
time (minutes) 10
8
6
4
2
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
Prospective consumption: total AUC
Prospective consumption (cm)
In our study, older and young men did not differ in their ability to compensate energy balance and regulate body weight after a period of energy restriction. Energy intake is regulated by many factors, both physiological and social. Although social factors may have influenced older men in our study to eat more, physiological markers in our study did not suggest substantial differential response changes to energy restriction in young versus elderly men. Although we cannot completely rule out that other physiological factors that determine food intake were overruled by social and environmental factors in our study, our data do not give indications for the hypothesis that elderly have a declined ability to regulate energy balance. Future studies with older participants, preferably in the participants’ own residence, or in a setting that enables continuous measurement of metabolic responses, are needed to further clarify the etiology of decreased energy intake in older people.
2000 1500 1000 500 0
young men
elderly men
time (minutes)
Fig. 4. Mean score of several hunger and appetite ratings in young and elderly men after a pancake meal at t 5 0 (left graphs) and AUC of the different scores (right graphs). Fourteen young and 17 elderly men rated their hunger, fullness, desire to eat, and prospective food consumption on a visual analogue scale during energy balance (day 15) and energy restriction (day 36) after a test meal at t 5 0 minutes. In the left graphs: C, young men energy balance; B, young men energy restriction; -, elderly men energy balance; ,, elderly men energy restriction. In the right AUC graphs, black bars represent energy balance, white bars energy restriction. Total AUC (1 SEM) are adjusted for differences at baseline. There was no statistically significant group*time-interaction in any of the AUCs of appetite rating. ORIGINAL STUDIES
1. Chapman IM. Endocrinology of anorexia of ageing. Best Pract Res Clin Endocrinol Metab 2004;18:437–452. 2. Chapman IM, MacIntosh CG, Morley JE, Horowitz M. The anorexia of ageing. Biogerontology 2002;3:67–71. 3. Morley JE. Anorexia of aging: Physiologic and pathologic. Am J Clin Nutr 1997;66:760–773. 4. De Groot CPGM, Enzi G, Matthys C, et al. Ten-year changes in anthropometric characteristics of elderly Europeans. J Nutr Health Aging 2002;6:4–8. 5. De Groot CPGM, Van Staveren WA. Undernutrition in the European SENECA studies. Clin Geriatr Med 2002;18:699–708. 6. Persson MD, Brismar KE, Katzarski KS, et al. Nutritional status using Mini Nutritional Assessment and Subjective Global Assessment Predict Mortality in geriatric patients. J Am Geriatr Soc 2002;50:1996–2002. 7. Sullivan DH, Walls RC. Protein-energy undernutrition and the risk of mortality within six years of hospital discharge. J Am Coll Nutr 1998; 17:571–578. 8. Moriguti JC, Das SK, Saltzman E, et al. Effects of a 6-week hypocaloric diet on changes in body composition, hunger, and subsequent weight regain in healthy young and older adults. J Gerontol A Biol Sci Med Sci 2000;55(12):B580–587. 9. Roberts SB, Fuss P, Heyman MB, et al. Control of food intake in older men. JAMA 1994;272:1601–1606. 10. Das SK, Moriguti JC, McCrory MA, et al. An underfeeding study in healthy men and women provides further evidence of impaired regulation of energy expenditure in old age. J Nutr 2001;131:1833–1838. 11. Wilson MM, Morley JE. Invited review: Aging and energy balance. J Appl Physiol 2003;95:1728–1736. 12. Yukawa M, Cummings DE, Matthys CC, et al. Effect of aging on the response of ghrelin to acute weight loss. J Am Geriatr Soc 2006;54: 648–653. 13. Delzenne N, Blundell J, Brouns F, et al. Gastrointestinal targets of appetite regulation in humans. Obes Rev 2010;11:234–250. 14. Liddle RA, Morita ET, Conrad CK, Williams JA. Regulation of gastric emptying in humans by cholecystokinin. J Clin Invest 1986;77:992–996. Winkels et al 285
15. MacIntosh CG, Morley JE, Wishart J, et al. Effect of exogenous cholecystokinin (CCK)-8 on food intake and plasma CCK, leptin, and insulin concentrations in older and young adults: evidence for increased CCK activity as a cause of the anorexia of aging. J Clin Endocrinol Metab 2001;86:5830–5837. 16. Hiemstra G, de Roos N, de Vries J. Anamnesemethode: snel en doelmatig? [Method of anamnesis: quick and efficient?] Nederlands. tijdschrift voor dietisten 2005;60:88–96. Dutch. 17. Stichting NEVO. Nederlands voedingsstoffen bestand: NEVO table 2006. Dutch Nutrient Database. The Hague, the Netherlands: Voorlichtingsbureau voor de voiding; 2006. 18. Kim J, Wang Z, Heymsfield SB, et al. Total-body skeletal muscle mass: Estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr 2002;76:378–383. 19. Weir J. New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol 1949;109:1–9. 20. Parker BA, Sturm K, MacIntosh CG, et al. Relation between food intake and visual analogue scale ratings of appetite and other sensations in healthy older and young subjects. Eur J Clin Nutr 2004;58:212–218. 21. Blundell J, de Graaf C, Hulshof T, et al. Appetite control: Methodological aspects of the evaluation of foods. Obes Rev 2010;11:251–270. 22. Ghoos YF, Maes BD, Geypens BJ, et al. Measurement of gastric emptying rate of solids by means of a carbon-labeled octanoic acid breath test. Gastroenterology 1993;104:1640–1647. 23. Haycock GB, Schwartz GJ, Wisotsky DH. Geometric method for measuring body surface area: A height-weight formula validated in infants, children, and adults. J Pediatr 1978;93:62–66.
286 Winkels et al
24. Maes BD, Ghoos YF, Rutgeerts PJ, et al. [*C]octanoic acid breath test to measure gastric emptying rate of solids. Dig Dis Sci 1994;39:104S–106S. 25. Pasman WJ, Heimerikx J, Rubingh CM, et al. The effect of Korean pine nut oil on in vitro CCK release, on appetite sensations and on gut hormones in post-menopausal overweight women. Lipids Health Dis 2008; 7:10. 26. Cunningham KM, Daly J, Horowitz M, Read NW. Gastrointestinal adaptation to diets of differing fat composition in human volunteers. Gut 1991;32:483–486. 27. Cunningham KM, Horowitz M, Read NW. The effect of short-term dietary supplementation with glucose on gastric emptying in humans. Br J Nutr 1991;65:15–19. 28. Clarkston WK, Pantano MM, Morley JE, et al. Evidence for the anorexia of aging: Gastrointestinal transit and hunger in healthy elderly vs young adults. Am J Physiol 1997;272:R243–248. 29. Morley JE. Decreased food intake with aging. J Gerontol A Biol Sci Med Sci 2001;56:81–88. 30. Hall KD. What is the required energy deficit per unit weight loss? Int J Obes (Lond) 2008;32:573–576. 31. De Graaf C, Blom WAM, Smeets PAM, et al. Biomarkers of satiation and satiety. Am J Clin Nutr 2004;79:946–961. 32. Nijs KAND, De Graaf C, Siebelink E, et al. Effect of family-style meals on energy intake and risk of malnutrition in Dutch nursing home residents: A randomized controlled trial. J Gerontol A Biol Sci Med Sci 2006;61:935–942. 33. Stroebele N, De Castro JM. Effect of ambience on food intake and food choice. Nutrition 2004;20:821–838.
JAMDA – May 2011
Objectives: Decreased energy intake in older persons poses these people at risk of progressive weight loss. It may result from a failure to regulate energy intake and expenditure after periods of underfeeding. The objective of this study was to investigate if a period of underfeeding differentially influences energy intake of older compared with young men and, additionally, to study potential underlying mechanisms, namely changes in gastric emptying rate and cholecystokinin (CCK) levels in blood. Design/setting: Dietary intervention of 3 phases. After a phase of energy balance, we fed participants in phase 2 by a mean of 70% of their needs for 21 days. During phase 3, we assessed ad libitum energy intake of the participants during 9 days. At the end of phases 1 and 2, we assessed appetite, gastric emptying, and CCK levels in blood in response to a test meal.
Results: During energy balance, mean energy intake of young men (14.3 2.3 MJ/day) was significantly higher than that of older men (11.3 1.8 MJ/day, P \ .001). After the period of underfeeding, energy intake in phase 3 amounted to 16.3 2.6 MJ/day in young men and to 14.4 3.2 MJ/day in older men. Ad lib energy intake after underfeeding did not differ between young and older men (analysis of covariance, with energy intake during phase 1 as covariate, P 5 .99). There were no differential changes in body weight, body composition, resting energy expenditure, gastric emptying rate, CCK-8 levels, and appetite between young and older men during the study. Conclusion: Our results do not indicate that older men have an impaired ability to control energy intake after a period of underfeeding compared with younger men. Trial registration: NCT00561145. (J Am Med Dir Assoc 2011; 12: 277–286)
Participants: Fifteen young (age 24 years [range 20–34], body mass index 23.0 kg/m2 2.3) and 17 older (age 68 years [64–85], body mass index 24.5 kg/m2 1.9) men participated in this study.
Keywords: Energy intake; gastric emptying; aging; energy restriction
Food intake decreases with ageing. The decrease in food intake tends to be higher in elderly people than the decrease in energy expenditure, resulting in unintentional loss of predominantly lean body mass.1,2 Both physiological and nonphysiological factors contribute to the decline in energy intake. Weight loss presents itself in healthy elderly as well as in elderly with compromised health.3,4 Loss of lean body mass, which is associated with functional impairments
and disability, predisposes the elderly to malnutrition, morbidity, and mortality.1,2,5–7 The unintentional weight loss in older people might be triggered by an impaired ability to regulate energy balance; however, data are equivocal. Some studies suggest that older adults do not compensate their energy intake after a period of underfeeding, and do not regain the weight lost during underfeeding, whereas young adults do increase their energy intake and regain the weight lost during energy restriction.8–10 It has been hypothesized that this is related to a slower gastric emptying rate in elderly,11 although the response in gastric emptying rate to energy restriction has not been studied. However, other investigators could not replicate age-related differences in energy intake compensation after energy restriction, neither could they show age-related differences in related metabolic responses.12 We studied the effect of underfeeding on compensation in energy intake and resting energy expenditure, body weight,
Department of Human Nutrition, Wageningen University, Wageningen, the Netherlands (R.M.W., A.J.-S., K.d.G., E.S., M.M., L.d.G.). This study was financially supported by the NZO (Dutch Dairy Association), Zoetermeer, the Netherlands. Address correspondence to Renate M. Winkels, PhD, Wageningen University, Human Nutrition, Bomenweg 4, 6703 HD Wageningen, the Netherlands. E-mail: [email protected]
Copyright Ó2011 American Medical Directors Association DOI:10.1016/j.jamda.2010.08.011 ORIGINAL STUDIES
Winkels et al 277
and body composition in young and older men, to contribute to the evidence base on the causes of decreased ability to regulate energy balance in older people and to clarify current controversy. In addition, we studied whether gastricemptying rates, appetite ratings, and cholecystokinin (CCK) levels in blood mediated any differential responses to energy restriction between young and older men. Gastric-emptying rate may be an important regulator of food intake.13 CCK causes gallbladder contraction and stimulates the release of several digestive enzymes in the duodenum13 and slows gastric emptying.14 CCK decreases food intake;15 it has been suggested that increased CCK activity in elderly may contribute to anorexia of aging.15 SUBJECTS AND METHODS Subjects We recruited 55 men, by publishing advertisements in local newspapers and by sending out general e-mails to an e-mail list of persons who had indicated their interest in participating in studies of our university. We included only men, as hormonal fluctuations in young women during the menstrual cycle may influence body weight and food intake. During a screening visit, men donated a fasting blood sample and gave written informed consent. In the fasting blood sample, we measured glucose, hemoglobin, and hematocrit. Men were excluded based on the following exclusion criteria: body mass index (BMI in kg/m2) less than 20 or higher than 30, adherence to a weight-reduction or medically prescribed diet, dementia (Mini-Mental State Examination score \21), diabetes, anemia, gastrointestinal disorders, use of drugs known to interfere with energy balance, or a history of medical or surgical events known to affect the study
outcome. Based on these criteria, 32 men (15 young and 17 older men) were included in the study. The Medical Ethical Committee of our university approved the study. Design The study consisted of 3 subsequent phases: phase 1 was a 14-day period of energy balance (days 1–14), phase 2 was a 21-day 30% energy restriction period (days 15–35), and phase 3 was a 9-day ad libitum food intake period (days 36–44). Before phase 1, dieticians from our study center estimated the habitual energy intake of the subjects with a validated food-frequency questionnaire.16 During all phases of the study, all foods and beverages were provided by the study center. At the end of phase 1 (morning of day 15) and at the end of phase 2 (morning of day 36), all participants took part in a gastric-emptying test procedure. Phase 1: Energy Balance During days 1 through 14, phase 1 of the study, we provided the participants with a diet containing approximately 90 energy percent of their estimated total daily energy requirement (Table 1). Participants chose the remaining 10% of energy from a list of choice items. This list contained fruits and vegetables, sweet sandwich fillings, crackers, candies, and nonalcoholic and alcoholic drinks. Participants could freely choose from this list, but daily had to consume the choice items. Their choice of items was recorded by the dieticians. Participants kept a diary to record any deviations from the diet, illness, and use of drugs. We weighed the participants daily and adjusted their diet to a higher or lower energy content if participants
Table 1. Foods and Drinks Provided during the 3 Phases of the Study, for a Participant with an Energy Intake of 11 MJ/day during Energy Balance (Phase 1) Product
Energy balance (phase 1))
Energy restriction (phase 2) †
Ad libitum (phase 3) ‡
Bread Margarine Sandwich fillings
231 g (7 slices) 40 g 48 g 16 g (1 portion) 54 g (3 portions) 3 200 mL 250 g (2 pieces) 30 g (3 cookies) 150 g 150 g 70 g 76 g 50 g 18 g 210 g 1.1 MJ
162 g (5 slices) 27 g 32 g 16 g (1 portion) 18 g (1 portion) 1 167 mL 125 g (1 piece) 20 g (2 cookies) 130 g 130 g 45 g 66 g 36 g 13 g 140 g 0.7 MJ
462 g (21 buns) 80 g 150 g 32 g (2 portions) 108 g (6 portions) 6 500 mL 500 g (4 pieces) 60 g (6 cookies) 300 g 300 g 140 g 152 g 100 g 36 g 420 g 1.1 MJ
Eggs (per week) Milk Fruit Cookies Hot meal
Salad Salad dressing Dessert Choice items 1
Sweet fillings Cheese Meat
Potatoes, rice, or pasta Boiled vegetables Sauce Meat
* During phase 1, participants had to be weight stable. Food and drinks covered 90 energy percent of the daily needs. Participants chose the remaining 10 energy percent from a list of choice items. During all phases, the amount of energy that participants had to ingest with the choice items was prescribed by the dieticians; in phase 3 the amount of energy in these items was the same as in phase 1. † During phase 2, participants consumed 70% of the energy intake in phase 1. ‡ During phase 3, all products were provided as 200% of the amount in phase 1. Participants were requested to eat as much or little as to feel normally satiated. 278 Winkels et al
JAMDA – May 2011
progressively lost or gained weight (.0.2 kg/d). Participants were weight stable during days 12, 1, and 14 of phase 1, the mean energy intake during these 3 days was considered as their energy intake for energy balance. We collected duplicates of the diets that the participants consumed and stored these duplicates at –20 C until analysis. During phase 1, we collected a duplicate diet for a fictive participant with an energy intake of 11 MJ/d based on food tables. The diet mimicked a typical Dutch diet: 33 energy percent of fat, 52 energy percent of carbohydrates, 13 energy percent of protein, and 2.6 g of fiber/MJ based on the Dutch food table.17 During all 3 phases of the study during weekdays, participants consumed a hot lunch at the study center. After lunch, participants received a takehome package with snacks and foods and beverages for their evening meal and for breakfast. On Fridays, subjects received a package with food items and beverages for the weekend, plus instructions for the preparation of the meals. We weighed out all foods and beverages for each participant. There was a 14-day menu cycle that was repeated every 14 days. Phase 2: Energy Restriction During days 15 through 35, phase 2 of the study, we provided the participants with a diet that contained 70 energy percent of the energy that each participant consumed during the last 3 days of phase 1 (Table 1). To compose this energyrestricted diet, the dieticians multiplied the amount of a product in phase 1 by 0.7 and rounded this off to obtain a distributable amount of this product. For example, 231 g of bread*0.7 5 162 g of bread is equal to 4.9 slices: this was rounded of to 5 slices. The macronutrient distribution of the diet in phase 2 was similar to phase 1. As in phase 1, approximately 90 energy percent of the energy-restricted diet was prescribed by the dieticians; participants chose the remaining 10% of energy from a list of choice items. To analyze the amount of energy and micronutrients in this diet, we collected a duplicate diet for a fictive participant with a daily energy intake of 7.7 MJ during phase 2. Phase 3: Ad Libitum In phase 3, participants consumed ad libitum from a diet provided by the study center, (Table 1). Participants were asked to consume as much or little of the foods and beverages as required to feel normally satiated. Most products were provided as 200% of the amount in phase 1. We provided foods in uncommon portion sizes to prevent participants from consuming amounts similar to those consumed habitually. In addition to the products provided by the study center, participants had to consume the choice items, as in phases 1 and 2. The items were also provided by the study center and the total amount of energy in these items was the same as during phase 1. For example, if a participant daily had to select 1.1 MJ from choice-items in phase 1, this participant also had to select 1.1 MJ/day from choice-items in phase 3. We collected and weighed leftovers and empty packages to record food energy intake.
ORIGINAL STUDIES
Energy in the Diet After the study, we thawed all the duplicates of the diets we collected during phase 1 and phase 2. We pooled the duplicates per phase and homogenized them. Total fat, protein, dry matter, ash, and fiber were analyzed in these duplicates, available carbohydrates were determined by difference. From this, we calculated the amount of fat, protein, carbohydrates, fiber, and energy in the duplicates. The mean energy intake of participants during phases 1 and 2 was calculated and subsequently adjusted for the difference between calculated and analyzed amount of energy in the diet. For phase 3, reported energy intakes were based on the food table only, because participants could eat ad libitum in that phase, which made the collection of duplicates not possible. Anthropometry, Body Composition, Respiratory Quotient, Resting Energy Expenditure, and Physical Activity We measured anthropometry, body composition, resting energy expenditure (REE), and respiratory quotient (RQ) on the last day of phase 1 (day 14), on the last day of phase 2 (day 35), and on the last day of phase 3 (day 44). On these days, we transported the participants by car from their residence to the study center. Participants had to be fasted and were instructed to avoid intensive physical activity on the morning of the measurements. The same observer performed all the anthropometric and body composition measurements. We weighed the participants to the nearest 0.1 kg using a digital weighing scale (Seca, Bascule, MT). Participants wore only their underwear during weighing and were asked to void beforehand. We did not inform participants about their weight and asked them not to weigh themselves during the study. Height was measured to the nearest 0.1 cm using a microtoise. Fat mass and bone mass were determined using a dual energy X-ray absorptiometer (DEXA, model DPX-L, Lunar Radiation Corp., Madison, WI, software version 1.31). Lean body mass was calculated from total body mass minus fat mass and bone mass. Appendicular lean soft tissue (ALST) was assessed with DEXA. Smooth muscle mass was estimated from ALST with the following formula:18 Total body SM5ð1:13 ALSTÞ ð0:02 ageÞ 1 1:58 REE was measured using indirect calorimetry (Deltatrac, Madison, WI). Participants lay on a bed for 35 minutes and were asked to stay awake and motionless. Resting VO2 and VCO2 were measured during the last 25 minutes. Weir’s equation was used to calculate REE.19 We monitored physical activity of the participants by assessing the number of steps taken each day with pedometers (Yamax Digi-walker, SW-200, Tokyo, Japan). Men were asked to maintain their usual lifestyle and activity pattern during the complete study period.
Winkels et al 279
Appetite Rating, Gastric Emptying Rate, and CCK Levels in Blood in Response to a Test Meal All participants took part in 2 gastric-emptying meal tests: at the end of energy balance (morning of day 15) and at the end of energy restriction (morning of day 36). Participants arrived at the study center in fasting state on the morning of these tests. A research nurse inserted a catheter in the antecubital vein of participants to facilitate the drawing of multiple blood samples during the morning. Saline was injected into the catheter after drawing of a blood sample, to prevent clogging. The first tube of each time point of blood sampling was discarded to remove the saline. Fifteen minutes before the start of the test procedure, a fasting blood sample was taken. In addition, participants rated their appetite by scaling their hunger, fullness, desire to eat, and prospective food consumption on a 10-cm visual analogue scale20,21 with statements expressing extremes (not at all versus very little/very much) and participants collected a breath sample with a straw into a 12-mL tube. The test procedure started with the consumption of a test meal at a steady pace during a period of approximately 10 minutes, followed by a glass of water. The test meal consisted of a pancake containing 100 mg of 13C-labeled octanoic acid (octanoic acid-1–13C, 99 atom%, OBT grade, Isotech, Miamisburg, OH). 13C-octanoic acid is retained in the solid phase of the meal in the stomach, but is rapidly absorbed and oxidized into 13CO2 on arrival in the duodenum. The rate of appearance of 13CO2 in breath is therefore a measure of gastricemptying rate.22 The pancake contained 2.1 MJ (506 kcal) and consisted of 18 energy percent protein, 52 energy percent carbohydrate, and 30 energy percent fat. During the day of the test meal, the amount of energy in the diet was corrected to adjust for the amount of energy in the pancake. Breath samples were collected every 15 minutes after the initiation of the meal for a period of 4 hours. Participants rated their appetite and donated a blood sample 15, 30, 45, 60, 90, 120, and 180 minutes after the initiation of the meal. The total areaunder-the-curve (AUC) was calculated for appetite scores. In breath samples, the 13C-enrichment of exhaled CO2 was analyzed with Isotope Ratio Mass Spectrometry. Total CO2 production was assumed to be 300 mmol/(m2*hour); body surface area in m2 was calculated with the formula of Haycock et al.23 Recovery curves of 13C in breath and cumulative curves over time were plotted and fitted according to formula 1 or 2 respectively:22,24 Formula 1: percentage of excretion per hour of the given 13 C-dose: y 5 atb ect , where t 5 time in hours; and a, b, and c are constants Formula 2: percentage of cumulative excretion per hour of the given 13C-dose: y 5 mð1 ekt Þb , where t 5 time in hours and m, k, and b are constants with m being the total cumulative recovery when time is infinite. We used an MS Excel macro sheet (Microsoft Corp, Redmond, WA) provided by the University of Leuven22 to derive 2 gastric-emptying parameters from these curves: the recovery time of 50% of the maximal recovered 13C (gastric half emptying time T1/2) and the time when the peak of the 13C excretion
280 Winkels et al
curve is reached (lag phase Tlag). We retained the value of the T1/2 and Tlag for the best-fitting curve. In blood, CCK-8 levels were analyzed with a radio-immuno assay (CCK-8 Euro-Diagnostics, Malmo¨, Sweden).25 This CCK assay has been optimized to reach a sensitivity of 0.05 pmol/L with no cross-reactivity toward gastrin-17, and sulphated gastrin. The intra- and interassay CV ranged from 9% at lower concentrations (0.85 pmol/L) to 5% at higher concentrations (2.04 pmol/L). We calculated the total AUC of CCK-levels in blood plotted against time in minutes. Statistical Analysis Our primary outcome was the difference in energy intake compensation after energy restriction between young and older men. To test this we used analysis of covariance (ANCOVA). We compared ad libitum energy intake during phase 3 in young men with that in elderly men and inserted energy intake during energy balance (phase 1) as covariate in the model. Additionally, we used analysis of variance (ANOVA) for repeated measures to assess if body weight, body composition, energy expenditure, and RQ differentially changed between young and older men during the 3 phases of the study (P for interaction between group and outcome). To test whether gastric-emptying rate, CCK levels, and appetite ratings differentially responded to energy restriction in young versus older men, we used mixed model ANOVA. In these models, we included group (young versus older men), time (phase 1: energy balance versus phase 2: energy restriction) and group*time effects; time was included as a repeated statement. Our main interest was whether the group*time interaction was significant. For the various appetite ratings, we used a mixed model ANOVA in which we added baseline appetite ratings as covariates to the model to adjust for baseline differences in these ratings. We used least square means to calculate the AUC adjusted for baseline differences. For gastric half-emptying time (T1/2), gastric emptying lag time (Tlag), AUC of the CCK and 13 C-excretion curves, we used a comparable mixed model ANOVA, but did not include covariates. Because CCK data were not normally distributed, these data were logtransformed before testing. Data were analyzed using PASW Statistics version 17.0.3 (SPSS Inc, Chicago, IL). Results were considered statistically significant at the .05 level. RESULTS All subjects completed the study. Young men were slightly taller than older men, but their BMI did not differ statistically significantly before the start of the intervention (Table 2). Fasting glucose levels were slightly higher in older men. The amount of energy and the macronutrient composition of the diet as chemically analyzed were comparable with what was expected based on the Dutch food table. During phase 1, the amount of energy in a prescribed daily diet of 9.9 MJ/d according to food tables contained
JAMDA – May 2011
Table 2. Baseline Characteristics of the Participants, Assessed before the Start of the Intervention
Age, y Height, m Weight, kg Body mass index, kg/m2 Glucose, mmol/L Hemoglobin, g/L Hematocrit
Young men (n 5 15)
Older men (n 5 17)
24 [20, 34] 1.81 0.08 75.8 11.3 23.0 2.3 4.5 [3.7, 5.1] 9.4 0.5 45% 2
68 [64, 85] 1.76 0.05* 75.8 7.6 24.5 1.9 5.0 [4.6, 5.5]* 9.4 0.4 45% 2
Data are means standard deviation, or median [range]. * Statistically significantly different between young and older men by unpaired t test for normally distributed variables or by Mann-Whitney U test (glucose).
9.6 MJ/d according to chemical analysis. During phase 2, a prescribed diet of 6.9 MJ/d based on food tables contained 6.8 MJ according to chemical analysis. Analyzed amounts of macronutrients were as follows: protein 12 to 14 energy%, carbohydrates 48 to 54 energy%, fat 33 to 37 energy%, fiber content 3.0 to 3.2 g/MJ (20–31 g/d). Mean energy intake during energy balance (phase 1) was significantly higher in young than in older men: 14.3 MJ/d for young men and 11.3 MJ/d for older men (Table 3, P \ .001). During energy restriction (phase 2), the energy intake was 30% lower than needed for energy balance: for young men the absolute deficit in energy was 4.3 MJ/d, for older men it was 3.4 MJ/d. Mean energy intake during the period of ad libitum food intake (phase 3) was 116% of energy intake during energy balance (phase 1) in young men and 127% in older men (Figure 1). Energy intake compensation after energy restriction did not differ between young and older men: energy intake in phase 3 did not differ between young and older men (tested with ANCOVA, Table 3, P 5 .99). Analysis of changescores yielded similar results: change in energy intake between phase 3 and phase 1 did not differ between young and older men (data not shown). For both young and elderly men, the ad libitum energy intake during the weekdays of phase 3 of the study tended to be higher (16.1 2.4 MJ for young and 14.2 3.2 MJ for elderly men during weekdays) than the energy intake during weekend days (15.1 2.7 MJ for young and 13.5 3.5 MJ for elderly men during weekend days). We did not find differences in product choice behavior between young and older men during phase 3 (data not shown). Weight loss during phase 2 was similar for both groups: 2.2 SD 0.9 kg for young men and 2.1 0.9 kg for older men, whereas weight gain was similar during phase 3: 1.4 1.3 kg for young men and 1.5 1.1 kg for older men. Older men had statistically significantly less smooth muscle mass and lower resting energy expenditure than young men had (Table 3, group effects), but there were no differences in body weight, fat mass, lean body mass, RQ, or physical activity. There were no differential changes between young and older men during the 3 phases of the study in any of these variables (Table 3, interaction tests). ORIGINAL STUDIES
During the test meal procedures, breath and blood samples of one young and breath samples of one older man could not be collected because of technical difficulties. The excretion curves of 13CO2 after consumption of a test meal labeled with 13C-octanoic acid in phases 1 and 2 were similar for young and elderly men (Figure 2). The change in gastric-emptying lag time (Tlag) from phase 1 to phase 2 differed between young and older men (P 5 .036 for group*time interaction), but not for gastric half emptying time (P 5 .098). There were no statistically significant group or time effects. CCK-8 concentrations in blood increased after consumption of the test meal in young and older men (Figure 3), both during energy balance and during energy restriction. There was no differential change between young and old men from phase 1 to phase 2 in total AUC (P 5 .20 for group*time interaction), neither were there statistically significant group or time effects in total AUC. Rated feelings of hunger and appetite are shown in Figure 4. The rating of these scores did not differentially change from energy balance to energy restriction in young men compared with older men: there were no statistically significant group*time-interaction effects (P values for group*time interaction: P 5 .34 for hunger, P 5 .37 for desire, P 5 .18 for fullness, and P 5 .50 for prospective consumption). For fullness, total AUC was significantly greater in phase 2 than in phase 1 (P for time-effect phase 1 versus phase 2 5 .013), but not for the other appetite ratings. There were no statistically significant group effects in total AUC for appetite ratings. DISCUSSION We did not find differences in the ability of older men to compensate their energy intake or expenditure after a period of undereating compared with young men. Young and older men both lost weight during a period of energy restriction; energy compensation behavior did not differ between young and older men after this period of energy restriction. There was only a slight differential response between young and older men in gastric-emptying lag time in response to a test meal. We did not find differential changes between young and elderly men in any of the other physiological responses. Thus, we were not able to confirm the findings from others8,9 that older men are less well able to compensate energy balance after a period of energy restriction than young men. We strictly controlled energy intake during phases 1 and 2 of our study, and carefully monitored intake during phase 3. In addition, we monitored energy expenditure by assessing resting energy expenditure and by assessing physical activity of the participants and we studied the effect of energy restriction on gastric-emptying rate and CCK excretion after a test meal. Within this strictly monitored regime, we did not find any evidence that regulation of energy balance in older men is less well maintained than in young men. Gastric-emptying rate may be an important regulator of food intake, although other factors like gastric distension and meal composition probably also contribute to satiation and meal termination.13 Habitual dietary intake may affect Winkels et al 281
282 Winkels et al
Table 3. Energy Intake, Body Weight, Body Composition, Resting Energy Expenditure, RQ and Physical Activity Of Young (N515) And Older Men (N517) During The Three Phases Of The Intervention. Data are means standard deviation
Young men Older men
Young men Older men Young men Older men Young men Older men Young men Older men Young men Older men Young men Older men Young men Older men
Energy intake in MJ/d
Body weight in kg Fat mass in kg Lean body mass in kg Smooth muscle mass in kg Resting energy expenditure in MJ/d RQ Physical activity in nr of kilo-steps/d
Phase 1
Phase 2
Phase 3
Energy Balance, 100% of Energy Needs
Energy Restriction, 70% of Energy Needs
Ad libitum, 200% of Energy Needs Provided
14.3 2.3† 11.3 1.8†
10.1 1.6 7.9 1.3
15.9 2.4 14.0 3.2
74.5 10.6 74.1 7.5 13.1 7.0 15.4 5.1 58.5 6.3 56.1 4.9 32.4 3.9
72.2 10.9 72.1 7.6 11.9 6.8 14.2 5.6 57.7 6.7 55.5 5.0 32.1 3.9
73.2 10.6 73.6 7.1 12.2 6.7 14.9 5.4 58.4 6.9 56.5 4.4 32.2 3.9
29.0 3.0 6.2 0.9
28.7 2.9 6.1 0.9
29.1 2.7 6.4 0.5
5.8 0.6 0.86 0.04 0.86 0.04 9.1 2.0
5.6 0.5 0.88 0.04 0.85 0.04 8.6 2.2
5.9 0.5 0.89 0.04 0.89 0.06 8.2 2.7
7.9 3.9
8.2 3.8
7.8 4.5
ANCOVA
P 5 .99* ANOVA for repeated measures Differences Differences Interaction within persons between (group* groups variable) P \.001 P 5 .94 P 5 .67 P \.001
P 5 .23
P 5 .94
P \.001
P 5 .32
P 5 .41
P 5 .025
P 5 .012
P 5 .56
P 5 .019
P 5 .050
P 5 .99
P 5 .049
P 5 .15
P 5 .37
P 5 .20
P 5 .55
P 5 .33
ANCOVA, analysis of covariance; RQ, respiratory quotient. * We tested the difference between young and older men in phase 3; in these analyses we inserted energy intake during energy balance (phase 1) as covariate in the model. † Differed statistically significantly between young and older men during phase 1, P \.05. JAMDA – May 2011
180 170
Energy intake during phase 3 as % of intake during phase 1
160 150 140 130 120 110 100 90 80 70 60 50 36
37
38
39
40
41
42
43
44
day Fig. 1. Mean energy intake ( SEM) during the 9-day period following energy restriction relative to the energy intake during weight maintenance in young (B) and older men (-). Energy intake during phase 1 was 100% (dotted line $$$$$), energy intake during phase 2 was 70% (dashed line - - - - - -).
the gastric-emptying response to a test meal.26,27 Because earlier studies suggested that gastric-emptying rate may be delayed in elderly men,28,29 we hypothesized that an altered gastric-emptying response to a lower habitual energy intake may be responsible for a dysregulation in energy balance in elderly. The response in gastric-emptying lag time to energy restriction in elderly men was statistically significantly different from the response in young men: in elderly men it changed from 132 minutes during energy balance to 145 minutes during energy restriction, whereas in young men it changed from 143 to 138 minutes (Figure 2). However, this could be a chance finding, as we did not correct for multiple testing. Moreover, it is debatable whether this minimal effect is physiologically relevant. Gastric half-emptying time did not differentially change in young versus elderly men in response to energy restriction, which is in line with our findings of CCK in blood: there was no differential response in CCK after energy restriction in young versus elderly men. Thus, our data do not provide substantial evidence for a different response to energy restriction in gastric-emptying rate in elderly men than in young men. Our findings are in line with the results from Yukawa et al12; older participants in that study also increased energy intake after a period of energy restriction and experienced an increase in body weight during that period. Although, Yukawa et al12 included men and women, whereas we included men only, both their and our study are not in agreement with the findings from the earlier studies from Roberts and coworkers.8–10 As expected, we found that energy intake during energy balance of elderly men was lower than that of young men. Also, elderly men had lower smooth muscle mass and lower ORIGINAL STUDIES
resting energy expenditure than young men. Nevertheless, we could not show an age-related difference in energy intake compensation. Possibly, dysregulation of energy balance occurs only at very old age, or in persons with underlying illnesses. Yet, age, weight, BMI, health status, and energy intake during energy balance in our study were comparable with these characteristics in the studies from Roberts and coworkers,8–10 who did find a diminished regulation of energy intake in elderly. Subtle differences in study design may have contributed to the opposing findings. In the studies from Roberts and coworkers,8–10 the absolute energy restriction was equal for old and young intervention groups, whereas in the study from Yukawa et al12 and our own study the relative energy restriction was equal between young and old groups. However, this small difference in study design will probably not fully explain the opposing findings. Absolute energy restriction was higher for young men than for older men in our study. Nevertheless, weight loss during phase 2 was similar for both groups (Table 3). For each individual, we compared the observed weight loss, with the expected weight loss, based on the assumption that an energy deficit of 3.2 MJ causes 100 g weight loss30 (data not shown). For 3 young men and 1 older man weight loss during phase 2 was much lower than expected. However, exclusion of these participants did not influence our outcomes. Regulation of food intake is a complex process, which involves neural, endocrine, and hedonic factors.31 Moreover, cognitive aspects, social factors, and atmosphere during mealtime are important determinants of energy intake.32,33 The older participants in our study were all raised just after the second world war. During that period, food was scarce in the Netherlands, and therefore many participants were taught during childhood to always finish their plate. Some of the older participants expressed that it was a waste of food to leave any leftovers on their plate during phase 3 of the study, although we instructed the participants to consume as much food to feel normally satiated. Although speculative, these childhood beliefs may have caused the older participants to overeat during phase 3. During weekdays, the participants of our study consumed their hot meal at the study center. Most older participants were retired and tended to stay long during meal times to enjoy their time with the other participants. Most young men had to return to their daily activities quickly after meal times. We therefore hypothesize that particularly the older men may have eaten more during phase 3 of the study than they might have eaten in their own residence. Indeed, our results show that participants of our study tended to eat more during weekdays, when they consumed their hot meal at the study center, than during weekend days, when they consumed all meals in their own residence. Nevertheless, both young and older men tended to have a higher energy intake during weekdays than on weekend days, and energy intake during weekend days was still higher than required for energy balance. Although consumption of meals in the earlier-mentioned studies8–10,12 also partly took place at the various research centers, we cannot compare their exact setting and ambiance, which may partly explain differences in findings among studies. Winkels et al 283
13
C recovery in breath (% of the administered dose/hour)
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 0
30
60
90
120
150
180
210
240
time (minutes) 250 gastric half-emptying time T1/2 (min)
gastric emptying lag time Tlag (min)
200 150 100 50 0
200 150 100 50 0
young men elderly men
young men elderly men
Fig. 2. Mean recovery of 13C in breath (SEM) of young and elderly men after a pancake meal with 13C-octanic acid at t 5 0 minutes (hatched bar). Gastric emptying rate was determined at the end of the phase of energy balance (day 15) and at the end of the phase of energy restriction (day 36) in n 5 14 young men and n 5 16 elderly men. In the upper graph: C, young men energy balance; B, young men energy restriction; -, elderly men energy balance; ,, elderly men energy restriction. In bar graphs, black bars represent energy balance, white bars energy restriction.T1/2 did not and Tlag did differentially change from energy balance to energy restriction in young compared to elderly men (P 5 .098 for T1/2 and P 5 .036 for Tlag for group*time interaction).
2.5
1.5 800
1.0
CCK: total AUC
CCK-8 (pmol/L)
2.0
0.5
0.0
600 400 200 0
0
30
60
90
120
150
180
young men
elderly men
time (minutes)
Fig. 3. Mean concentration of CCK-8 in blood (SEM) of young and elderly men after a pancake meal at t 5 0 (left graph) and total AUC under the CCK-graph (right graph) assessed during energy balance (day 15) and energy restriction (day 36) in n 5 14 young men and n 5 17 elderly men. In the left graph: C, young men energy balance; B, young men energy restriction; -, elderly men energy balance; ,, elderly men energy restriction. In the whisker graph medians 1 ranges are plotted, gray boxes are during energy balance, white boxes are during energy restriction. There were no statistically significant group, time, or group*time-interaction effects in total AUC. 284 Winkels et al
JAMDA – May 2011
10
CONCLUSION
Hunger (cm)
8
6
2000
Hunger: total AUC
4
2
1000 500 0
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
1500
young men
elderly men
time (minutes) 10
Desire (cm)
8
6
2000
Desire: total AUC
4
2
ACKNOWLEDGMENTS
1000 500 0
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
1500
young men
elderly men
time (minutes) 10
We thank all participants for their enthusiastic participation in the trial and the dieticians, students, and research nurses for their assistance during the trial. We thank Dr Kristen Verbeke of the Katholic University in Leuven for kindly providing us with the Excel program for modeling the gastric emptying data.
6
REFERENCES 4
Fullness: total AUC
Fullness (cm)
8
2
2000 1500 1000 500 0
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
young men
elderly men
time (minutes) 10
8
6
4
2
-15 0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225 240
0
Prospective consumption: total AUC
Prospective consumption (cm)
In our study, older and young men did not differ in their ability to compensate energy balance and regulate body weight after a period of energy restriction. Energy intake is regulated by many factors, both physiological and social. Although social factors may have influenced older men in our study to eat more, physiological markers in our study did not suggest substantial differential response changes to energy restriction in young versus elderly men. Although we cannot completely rule out that other physiological factors that determine food intake were overruled by social and environmental factors in our study, our data do not give indications for the hypothesis that elderly have a declined ability to regulate energy balance. Future studies with older participants, preferably in the participants’ own residence, or in a setting that enables continuous measurement of metabolic responses, are needed to further clarify the etiology of decreased energy intake in older people.
2000 1500 1000 500 0
young men
elderly men
time (minutes)
Fig. 4. Mean score of several hunger and appetite ratings in young and elderly men after a pancake meal at t 5 0 (left graphs) and AUC of the different scores (right graphs). Fourteen young and 17 elderly men rated their hunger, fullness, desire to eat, and prospective food consumption on a visual analogue scale during energy balance (day 15) and energy restriction (day 36) after a test meal at t 5 0 minutes. In the left graphs: C, young men energy balance; B, young men energy restriction; -, elderly men energy balance; ,, elderly men energy restriction. In the right AUC graphs, black bars represent energy balance, white bars energy restriction. Total AUC (1 SEM) are adjusted for differences at baseline. There was no statistically significant group*time-interaction in any of the AUCs of appetite rating. ORIGINAL STUDIES
1. Chapman IM. Endocrinology of anorexia of ageing. Best Pract Res Clin Endocrinol Metab 2004;18:437–452. 2. Chapman IM, MacIntosh CG, Morley JE, Horowitz M. The anorexia of ageing. Biogerontology 2002;3:67–71. 3. Morley JE. Anorexia of aging: Physiologic and pathologic. Am J Clin Nutr 1997;66:760–773. 4. De Groot CPGM, Enzi G, Matthys C, et al. Ten-year changes in anthropometric characteristics of elderly Europeans. J Nutr Health Aging 2002;6:4–8. 5. De Groot CPGM, Van Staveren WA. Undernutrition in the European SENECA studies. Clin Geriatr Med 2002;18:699–708. 6. Persson MD, Brismar KE, Katzarski KS, et al. Nutritional status using Mini Nutritional Assessment and Subjective Global Assessment Predict Mortality in geriatric patients. J Am Geriatr Soc 2002;50:1996–2002. 7. Sullivan DH, Walls RC. Protein-energy undernutrition and the risk of mortality within six years of hospital discharge. J Am Coll Nutr 1998; 17:571–578. 8. Moriguti JC, Das SK, Saltzman E, et al. Effects of a 6-week hypocaloric diet on changes in body composition, hunger, and subsequent weight regain in healthy young and older adults. J Gerontol A Biol Sci Med Sci 2000;55(12):B580–587. 9. Roberts SB, Fuss P, Heyman MB, et al. Control of food intake in older men. JAMA 1994;272:1601–1606. 10. Das SK, Moriguti JC, McCrory MA, et al. An underfeeding study in healthy men and women provides further evidence of impaired regulation of energy expenditure in old age. J Nutr 2001;131:1833–1838. 11. Wilson MM, Morley JE. Invited review: Aging and energy balance. J Appl Physiol 2003;95:1728–1736. 12. Yukawa M, Cummings DE, Matthys CC, et al. Effect of aging on the response of ghrelin to acute weight loss. J Am Geriatr Soc 2006;54: 648–653. 13. Delzenne N, Blundell J, Brouns F, et al. Gastrointestinal targets of appetite regulation in humans. Obes Rev 2010;11:234–250. 14. Liddle RA, Morita ET, Conrad CK, Williams JA. Regulation of gastric emptying in humans by cholecystokinin. J Clin Invest 1986;77:992–996. Winkels et al 285
15. MacIntosh CG, Morley JE, Wishart J, et al. Effect of exogenous cholecystokinin (CCK)-8 on food intake and plasma CCK, leptin, and insulin concentrations in older and young adults: evidence for increased CCK activity as a cause of the anorexia of aging. J Clin Endocrinol Metab 2001;86:5830–5837. 16. Hiemstra G, de Roos N, de Vries J. Anamnesemethode: snel en doelmatig? [Method of anamnesis: quick and efficient?] Nederlands. tijdschrift voor dietisten 2005;60:88–96. Dutch. 17. Stichting NEVO. Nederlands voedingsstoffen bestand: NEVO table 2006. Dutch Nutrient Database. The Hague, the Netherlands: Voorlichtingsbureau voor de voiding; 2006. 18. Kim J, Wang Z, Heymsfield SB, et al. Total-body skeletal muscle mass: Estimation by a new dual-energy X-ray absorptiometry method. Am J Clin Nutr 2002;76:378–383. 19. Weir J. New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol 1949;109:1–9. 20. Parker BA, Sturm K, MacIntosh CG, et al. Relation between food intake and visual analogue scale ratings of appetite and other sensations in healthy older and young subjects. Eur J Clin Nutr 2004;58:212–218. 21. Blundell J, de Graaf C, Hulshof T, et al. Appetite control: Methodological aspects of the evaluation of foods. Obes Rev 2010;11:251–270. 22. Ghoos YF, Maes BD, Geypens BJ, et al. Measurement of gastric emptying rate of solids by means of a carbon-labeled octanoic acid breath test. Gastroenterology 1993;104:1640–1647. 23. Haycock GB, Schwartz GJ, Wisotsky DH. Geometric method for measuring body surface area: A height-weight formula validated in infants, children, and adults. J Pediatr 1978;93:62–66.
286 Winkels et al
24. Maes BD, Ghoos YF, Rutgeerts PJ, et al. [*C]octanoic acid breath test to measure gastric emptying rate of solids. Dig Dis Sci 1994;39:104S–106S. 25. Pasman WJ, Heimerikx J, Rubingh CM, et al. The effect of Korean pine nut oil on in vitro CCK release, on appetite sensations and on gut hormones in post-menopausal overweight women. Lipids Health Dis 2008; 7:10. 26. Cunningham KM, Daly J, Horowitz M, Read NW. Gastrointestinal adaptation to diets of differing fat composition in human volunteers. Gut 1991;32:483–486. 27. Cunningham KM, Horowitz M, Read NW. The effect of short-term dietary supplementation with glucose on gastric emptying in humans. Br J Nutr 1991;65:15–19. 28. Clarkston WK, Pantano MM, Morley JE, et al. Evidence for the anorexia of aging: Gastrointestinal transit and hunger in healthy elderly vs young adults. Am J Physiol 1997;272:R243–248. 29. Morley JE. Decreased food intake with aging. J Gerontol A Biol Sci Med Sci 2001;56:81–88. 30. Hall KD. What is the required energy deficit per unit weight loss? Int J Obes (Lond) 2008;32:573–576. 31. De Graaf C, Blom WAM, Smeets PAM, et al. Biomarkers of satiation and satiety. Am J Clin Nutr 2004;79:946–961. 32. Nijs KAND, De Graaf C, Siebelink E, et al. Effect of family-style meals on energy intake and risk of malnutrition in Dutch nursing home residents: A randomized controlled trial. J Gerontol A Biol Sci Med Sci 2006;61:935–942. 33. Stroebele N, De Castro JM. Effect of ambience on food intake and food choice. Nutrition 2004;20:821–838.
JAMDA – May 2011