- Email: [email protected]
The Quantity of Meal Fat Influences the Profile of Postprandial Hormones as Well as Hunger Sensation in Healthy Elderly People
The Quantity of Meal Fat Influences the Profile of Postprandial Hormones as Well as Hunger Sensation in Healthy Elderly People Vincenzo Di Francesco, MD, Rocco Barazzoni, MD, Luisa Bissoli, MD, Francesco Fantin, MD, Paolo Rizzotti, MD, Luigi Residori, MD, Angela Antonioli, MD, Maria Stella Graziani, PhD, Michela Zanetti, MD, Ottavio Bosello, MD, Gianfranco Guarnieri, MD, and Mauro Zamboni, MD
Background: In healthy elderly people, reduced appetite and the consequent decrease in food intake has been defined as the ‘‘anorexia of aging’’; this condition may lead to malnutrition. The aim of this study was to investigate how different compositions of macronutrients affect satiety and hunger signals as well as subjective sensations after meals in healthy elderly subjects. Methods: Experimental controlled study. Ambulatory healthy community-dwelling subjects evaluation in a single center on 12 elderly subjects, (75.22 years old) and 12 younger controls (28.22 years old). Using a visual analogical scale, hunger was evaluated under fasting conditions and at 30-minute intervals for up to 4 hours after two 800-kcal meals, where 20% and 40% of the calories were derived from fat. Serum samples were collected at –30, 60, 120, and 240 minutes to determine the concentrations of GLP-1, Protein-energy malnutrition is a frequent condition in the elderly and is associated with a reduction of the adaptive response to physiological and pathological conditions in old age.1 The so-called ‘‘anorexia of aging,’’ ie, an age-related decrease of calorie intake, may be one of the main risk factors for malnutrition in old age. Impaired satiety/hunger control may cause anorexia at more advanced ages.2–4 Peripheral signals influence satiety and hunger both after a meal and under fasting conditions. Interestingly, both the short-term postprandial satiety signals such as cholecystoki-
Geriatric Unit, Department of Biomedical and Surgical Sciences, University of Verona, Verona, Italy (V.D.F., L.B., F.F., L.R., A.A., O.B., M.Z.); Clinica Medica, Department of Clinical, Morphological and Technological Sciences, University of Trieste, Italy (R.B., M.Z., G.G.); Laboratorio di Analisi Chimico Cliniche ed Ematologiche, Azienda Ospedaliera di Verona, Verona, Italy (P.R., M.S.G.). This work was funded by grants from the University of Verona. Address correspondence to Vincenzo Di Francesco, MD, PhD, Dipartimento Scienze Biomediche e Chirurgiche, Sezione di Geriatria, Ospedale Maggiore, p.le Stefani 1, 37126 Verona, Italy. E-mail: [email protected]
Copyright Ó2010 American Medical Directors Association DOI:10.1016/j.jamda.2009.08.004 188 Di Francesco et al
acylated and desacylated ghrelin, triglycerides, glucose, and insulin. Results: Serum concentrations of GLP-1 were higher after the 40% fat meal than after the 20% fat meal (P \ .01) in the elderly but not in the younger subjects. Acylated to desacylated ratio was lower after the 40% fat meal (P \.05) in the elderly. Only in the older group were triglycerides higher (P \ .05), whereas hunger was significantly lower (P\.05) after the 40% fat meal. Conclusion: In healthy elderly people relatively large amounts of fat increase the satiety signal from GLP-1 and lower the acylated to desacylated ratio of ghrelin, consequently decreasing hunger. This condition may lead to a reduction in calorie intake. (J Am Med Dir Assoc 2010; 11: 188–193) Keywords: Ageing; malnutrition; GLP-1; hunger nin (CCK) and peptide YY (PYY) and the long-term adiposity signals from insulin and leptin have been demonstrated to be stronger in the elderly.5–8 The glucagon-like peptide 1 (GLP-1) is a potent regulator of food intake in humans,9 and the short-term postprandial concentrations of GLP-1 have been shown to be similar in both young and older healthy subjects after the isoenergetic duodenal infusion of glucose or lipid.6 However, the GLP-1 levels in the midterm response to a physiological meal in older subjects have not to our knowledge been determined. Ghrelin, a 28 amino acid gastric peptide, generates an orexigenic peripheral signal,10 which triggers meal initiation.11 In fact, ghrelin production has been shown to rise during fasting and to fall after meals.12 Ghrelin has been shown to be activated by acylation of the Serine3 residue of the octanoyl group.13 Acylated ghrelin (A-Ghr) is proposed to be the only biologically active form of the hormone that is able to exert an orexigenic effect.14 Nonacylated ghrelin is the most abundant circulating form, which although devoid of direct neuroendocrine action, is able to bind to receptor subtypes.15 Acylated ghrelin was found to be significantly lower in healthy elderly women.16 Although ghrelin levels JAMDA – March 2010
have a typically dynamic pulsatile distribution in younger subjects, elderly subjects have reduced ghrelin levels and show a flat postprandial curve after meals.17 Lower desacylated ghrelin and relative increments of acylated ghrelin have been suggested to negatively modulate insulin action and to contribute to insulin resistance in adults with metabolic syndrome.18 The effects of acylated to desacylated proportion in the elderly need to be investigated, as meal composition, especially the proportion of lipids, might influence ghrelin response and activation. The aims of this study were (1) to investigate the fasting and postprandial dynamics of GLP-1 in elderly and young controls; and (2) to evaluate the levels of GLP-1 and acylated and desacylated ghrelin in the elderly group and their relationships with regard to subjective hunger sensation in response to isocaloric hyperlipidic or hypolipidic diets.
Experimental Design
PATIENTS AND METHODS
Hunger Evaluation
Excluding criteria were malnutrition or recent weight loss; oral, swallowing, and chewing problems; any abdominal surgical procedure; cholelithiasis; diabetes; neurological diseases (including cognitive impairment); chronic gastrointestinal peptic or inflammatory diseases; malignancies; and any acute ongoing disease. Subjects with known cardiac, renal, or respiratory impairment or a body mass index lower than 18.5 or greater than 30 kg/m2 were also excluded. Individuals taking drugs that may have interfered with gastrointestinal motility and visceral sensitivity (Ca-antagonists, nitrates, prokynetics, proton pump inhibitors, H2-antagonists, sedatives) were also excluded.
A visual analogue scale (VAS) was used to measure subjective hunger. Hunger was defined according to de Graaf et al19 as the subjective driving force for the search, choice, and ingestion of food. Subjects were asked to make a single vertical mark on a horizontal 10-cm bar indicating an assessment of their current feelings from ‘‘not hungry at all’’ to ‘‘really hungry.’’ Baseline evaluations were collected 30 minutes before the meal, at the end of the meal, and 30, 60, 120, and 240 minutes thereafter. Blood samples were collected 30 minutes before and 60, 120, and 240 minutes after the meal. After centrifugation, serum samples were immediately stored at –80 C for final evaluation. Serum levels of glucose and triglycerides were measured using standard methods. Acylated ghrelin was measured using Radio Immuno Assay (RIA) (Linco, St. Charles, MO) (intra-assay coefficient of variation 4.9%; interassay coefficient of variation 8.9%). Serum desacyl-ghrelin was measured using enzyme-linked immunosorbent assay (ELISA) (DRG, New York, NY) following the manufacturer’s recommendations (intra-assay coefficient of variation 6%; interassay coefficient of variation 9.6%). ELISA assay (Phoenix Pharmaceuticals, Belmont, CA) was used for GLP-1 7 to 36 (minimum detectable concentration 0.08 ng/mL). This test has an intra-assay variation of 5% and interassay variation of 14%. For technical reasons, ghrelin evaluation was possible only for the elderly group.
Study Sample A total of 20 elderly subjects were screened from the database of an ongoing study on the effect of physical activity on body composition in community-dwelling healthy autonomous elderly. Two women refused to participate. Two women and 1 man were excluded because they were taking antisecretive drugs. Two men were excluded as well because they were taking Ca-antagonists. One man referred previous cholecystectomy. All the subjects were examined at baseline, before starting the intervention (daily walking program). Fifteen healthy young controls were recruited from medical students and personnel working at our unit. One young woman was excluded because she was taking oral contraceptives and 1 man because he had a measured body mass index (BMI) greater than 30 kg/m2. One woman refused the meal because she referred tomato juice intolerance. The final study sample consisted of 12 elderly subjects, 5 men and 7 women, mean age 75.26 years (range 68–87), BMI 21.1 to 28.3 kg/m2. They all had normal cognitive function (Mini Mental State Examination .24/30) and no functional impairment (evaluated as activities of daily living [ADLs] and instrumental ADLs). As a control group, we evaluated 12 subjects (6 men, 6 women) with mean age 28.22.0 (range 25–33) and BMI 18.9 to 26.5 kg/m2 (P . .05). The study was approved by our institution’s ethical committee and each subject gave their informed consent. ORIGINAL STUDIES
After an overnight fast, both groups were evaluated before eating; they were then asked to consume the standard meal in about 20 minutes. The end of the meal was considered time 0, after which evaluations were repeated every 30 minutes for a further 4 hours. The 20% fat meal (20%FM) meal consisted of 100 g of pasta with 50 g of tomato sauce and 10 g of olive oil, 80 g of lean turkey ham slices, one 60-g roll, and 250 mL of tap water. The total amount of energy was 800 kcal: 15% from proteins, 20% from fat, 65% from carbohydrates. The 40% fat meal (40%FM) consisted of 60 g of pasta with 70 g of bolognese and 10 g of olive oil, 50 g of soft fatty cheese, one 60-g roll, and 250 mL of tap water. The total amount of energy was 800 kcal: 15% from proteins, 40% from fat, 45% from carbohydrates.
Statistical Analysis Results are shown as meansSE. Data were analyzed using a 3-factor repeated measures analysis of variance (ANOVA) for GLP-1, glucose, insulin, triglycerides, and VAS evaluations and a 2-factor repeated measures ANOVA for ghrelin evaluation. A Pearson’s correlation coefficient was calculated to evaluate point-by-point relationships between VAS and laboratory variables in the older group. A significance level of .05 was adopted. The study was designed to verify the hypothesis that mean postprandial GLP-1 value may vary by about 25% after a hyperlipidic meal. The power of the test with 12 subjects in each group was about 90% for Di Francesco et al 189
Fig. 1. Fasting and postprandial concentrations of GLP-1 in the elderly and in younger controls, after 20%FM (full lines) and 40%FM (broken lines).
repeated measurements and 80% for comparison between age groups. Analyses were performed using SPSS for Windows, release 16.0 (Macintosh version, SPSS Inc, Chicago, IL).
RESULTS Figure 1 shows the fasting and postprandial concentrations of GLP-1 in the elderly and in younger controls, after 20%FM (full lines) and 40%FM (broken lines). Serum GLP-1 was higher in the older group but only after the 40%FM (time group meal interaction, P \ .05). In the elderly, but not in the younger group, GLP-1 showed different curves after the 2 meals (meal time interaction, P \ .01) with higher postprandial concentrations after the 40%FM. GLP-1 at 240 minutes after 20%FM: mean 40.8 pg/mL (95% confidence interval [CI]: 27–54); after 40%FM: mean 55.6 pg/mL (95% CI: 45–66). Figure 2 shows the fasting and postprandial concentrations of glucose, insulin, and triglycerides after the 20%FM (full lines) and the 40%FM (broken lines). Serum glucose and insulin showed the expected postprandial rise without differences between the meals (time treatment interaction, P . .05) but with lower concentrations in the younger group for glucose (time group treatment interaction, P \ .01). Triglycerides were higher after the 40%FM in the older but not in the younger group (time treatment interaction, P \ .05). Triglycerides concentrations were higher in the older compared with younger controls (treatment group interaction, P \ .01). Figure 3 shows fasting and postprandial concentrations of acylated ghrelin, desacylated ghrelin, and the plot of the acylated to desacylated ghrelin ratio. Acylated ghrelin showed a flat curve both after the 40%FM and the 20%FM (mean time and time treatment interaction, P . .05). Desacylated ghrelin concentrations fell after the meal both after the 40%FM and the 20%FM (time interaction P \ .01) without meal differences (time treatment interaction, P . .05). The acylated to desacylated ratio was significantly affected 190 Di Francesco et al
by the quantity of meal fat (time treatment interaction, P \ .05) with lower values after the 40%FM. Figure 4 shows the postprandial hunger sensation scores after the 20%FM (grey area) or the 40%FM (white area) in the older and the younger groups. Hunger sensation was significantly more suppressed after the 40%FM compared with the 20%FM but only in the older (time treatment interaction, P \ .05) . Correlations between hunger sensation and laboratory parameters was evaluated in the older group: The hunger sensation was inversely correlated to serum glucose after 20%FM and 40%FM (R5–0.602, P \ .001, and R5 –0.417, P \ .03 respectively) and to insulin but only after 20%FM (R 5 –0.356, P \ .005). Serum triglycerides inversely correlated with acylated ghrelin after 20%FM or 40%FM (R5–0.631, P \ .001, and R5 –0.478, P \ .01 respectively) and directly with serum insulin after 20%FM or 40%FM (R50.575, P \ .001, R5 0.483, P \ .01 respectively). DISCUSSION This study demonstrates that in healthy elderly men and women the postprandial serum concentrations of GLP-1 were higher and the ratio of acyl to desacyl ghrelin was lower after the consumption of a meal where 40% of the calories were derived from fat, when compared with an isocaloric meal where only 20% of calories came from fat. Consequently, the hyperlipidic meal had a stronger inhibitory effect on hunger. In the younger control group, neither the hunger sensation nor the postprandial GLP-1 curves were influenced by the quantity of alimentary fat. The levels of GLP-1 were higher in the older group but only after the 40% fat meal. Blundell et al20 coined the term ‘‘fat-paradox’’ to describe how food rich in fat, notwithstanding the generation of higher satiety signals, causes higher intake of calories in adults, thus leading to obesity. Hunger suppression in the elderly seems to be much stronger after a fatty meal thereby counteracting the small volume effect and the high caloric density of lipids. JAMDA – March 2010
Fig. 2. Fasting and postprandial concentrations of glucose, insulin and triglycerides after 20%FM (full lines) and 40%FM (broken lines) in older and younger groups.
It has been suggested that lipids cause a more potent suppression of hunger and food intake than glucose in younger but not older subjects in a study that used small intestinal nutrient infusion.21 This does not seem to be the case in our study, where we used a physiological meal and the stomach passage may account for the apparent differences. Our data seem to confirm the study hypothesis, that is, that a relatively low amount of fat causes lower hunger inhibition, as alimentary fatty acids are among the major determinants of GLP-1 secretion.22 GLP-1 levels rise in blood also after glucose ingestion, through insulin stimulation,23 but our study demonstrated a significantly higher postprandial GLP-1 curve ORIGINAL STUDIES
in the elderly only after the hyperlipidic meal. Macintosh et al did not find postprandial differences in GLP-1 dynamics in healthy elderly compared with young controls.6 Nevertheless, the Australian researchers used liquid nutrients directly infused into the duodenum. Furthermore, their observations were limited to the 2 hours after infusion. We used mixed meals and the observation period was prolonged to 4 hours, which may have helped to detect the differences between the 2 age groups. Insulin increases GLP-1 secretion,22,24 although contradictory results have been published.23 In this study, meal macronutrient composition did not affect the insulin curves. So the Di Francesco et al 191
Fig. 3. Fasting and postprandial concentrations of acylated ghrelin, desacylated ghrelin, and the plot of the acylated to desacylated ghrelin ratio in the older group.
difference of GLP-1 response after the 40%FM cannot be directly attributed to insulin because this hormone was unaffected by the composition of the meal. GLP-1 secretion is considered to be a response to hyperinsulinemia, as it has been shown to improve peripheral insulin sensitivity.22 Higher concentrations of GLP-1 may raise hypothalamic sensitivity to insulin, thus indirectly improving hunger suppression. Leptin also acts on energetic control by stimulating the secretion of GLP-1.22–25 Hyperleptinemia has been described in the elderly,8 but leptin concentrations were not affected in the short term by the meal. It is therefore unclear how leptin may influence GLP-1 secretion after different alimentary loads. Furthermore, the hypothesis that a higher fat load may potentiate leptin stimulus on GLP-1 is contradicted by the experimental observation that a hyperlipidic meal causes greater resistance to the leptin signal.25 Different postprandial curves of GLP-1 and hunger sensations after 20% or 40% fat may be influenced by different 192 Di Francesco et al
gastric emptying dynamics, because both alimentary fat and GLP-1 delay gastric emptying.23 Therefore, the hyperlipidic meal may inhibit gastric emptying and consequently hunger, either directly or by elevating GLP-1 levels. We could not demonstrate a correlation between the hunger sensation scores and point-by-point GLP-1 concentrations, but with this model it is not possible to evaluate the net effect of GLP-1 on hunger, avoiding all the other neurohormonal signals of the network controlling energy balance. In other experimental models GLP-1 injection was shown to suppress hunger but only at very high doses.9 Using a similar experimental setting, we have previously demonstrated that elderly subjects have lower plasmatic concentrations of acylated ghrelin than their younger counterparts. Furthermore, acylated ghrelin showed a flat postprandial curve in the elderly and pulsating dynamic curves in the younger controls.17 In the present study, we evaluated the effects of lipids on the dynamics of the different forms of ghrelin in the postprandial period. The ratio of acylated to desacylated ghrelin was significantly affected by the quantity of fat in the meal with lower values detected after the 40%FM. The relative prevalence of the active acylated form over the desacylated ghrelin 4 hours after the hypolipidic meal may explain the higher hunger sensation. It is well known that acylated ghrelin stimulates hunger and food consumption.14 Relative increments of acylated ghrelin have been suggested to negatively modulate insulin action and to contribute to insulin resistance in adults with metabolic syndrome.18 It may be hypothesized that a reduced acyl to desacyl ratio, as observed after the hyperlipidic meal, could increase hypothalamic insulin sensitivity in the elderly, thus leading to an increased inhibition of hunger. Some limitations of our study deserve a comment. We studied a sample of healthy elderly people who were not suffering from chronic diseases and were not taking medications. Nevertheless, it is reasonable to assume that chronic illness, medications, and malnutrition would only worsen the abnormalities seen in our subjects.2 Because of the small size of our sample, we were unable to evaluate the effect of gender on satiety/hunger and peptide dynamics. In this observational study, results may be confounded by unmeasured factors such as unmeasured biological parameters that were not extensively evaluated and deserve further study. In conclusion, our study shows that in healthy elderly subjects, the consumption of an 800-kcal meal with 40% calories derived from fat, compared with an isocaloric meal with 20% calories from fat, caused higher postprandial serum concentrations of GLP-1 and lower acyl to desacyl ghrelin ratios. As a result, the hyperlipidic meal was more effective in suppressing hunger and delaying the search for a second meal. This condition may lead to calorie restriction and eventually to malnutrition. In other words, in the elderly, fatty foods may actually cause undernutrition so that the ‘‘fat paradox’’ is lost. Under a clinically practical point of view, these conclusions should discourage the frequent use of fatty food (including oil, butter, and so forth) as a choice for improving calorie intake in frail older persons living at home or in nursing JAMDA – March 2010
Fig. 4. Fasting and postprandial hunger sensation scores after 20%FM (grey area) or 40%FM (white area) in the 2 age groups.
homes. They are already frequently suffering from secondary anorexia and malnutrition, thus fat overload at lunch would blunt dinner food consumption and net calorie intake could be even further reduced. REFERENCES 1. Mowe` M, Bohmer T, Kindt E. Reduced nutritional status in an elderly population (.70y) is probable before disease and possibly contributes to the development of disease. Am J Clin Nutr 1994;59:317–324. 2. Morley JE. Anorexia of aging: Physiologic and pathologic. Am J Clin Nutr 1997;66:760–773. 3. Roberts SB, Fuss P, Heyman MB, et al. Control of food intake in older men. JAMA 1994;272:1601–1606. 4. Di Francesco V, Fantin F, Omizzolo F, et al. The anorexia of aging. Dig Dis 2007;25:129–137. 5. Di Francesco V, Zamboni M, Dioli A, et al. Delayed postprandial gastric emptying and impaired gallbladder contraction together with elevated cholecystokinin and peptide YY serum levels sustain satiety and inhibit hunger in healthy elderly persons. J Gerontol A Biol Sci Med Sci 2005; 60:1581–1585. 6. MacIntosch GC, Andrews JM, Jones KL, et al. Effects of age on concentrations of plasma cholecystokinin, glucagon-like peptide 1 and peptide YY and their relation to appetite and pyloric motility. Am J Clin Nutr 1999;69:999–1006. 7. MacIntosh CG, Morley JE, Wishart J, et al. Effects 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. 8. Di Francesco V, Zamboni M, Zoico E, et al. Unbalanced serum leptin and ghrelin dynamics prolong postprandial satiety and inhibit hunger in healthy elderly: Another reason for the ‘‘anorexia of aging’’. Am J Clin Nutr 2006;83:1149–1152. 9. Gutzwiller JP, Go¨ke B, Drewe J, et al. Glucagon-like peptide-1: A potent regulator of food intake in humans. Gut 1999;44:81–86. 10. Asakawa A, Inui A, Kaga T, et al. Ghrelin is an appetite-stimulatory signal from stomach with structural resemblance to motilin. Gastroenterology 2001;120:337–345. 11. Cummings DE, Purnell JQ, Frayo RS, et al. A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 2001; 50:1714–1719.
ORIGINAL STUDIES
12. Tolle V, Bassant MH, Zizzari P, et al. Ultradian rhythmicity of ghrelin secretion in relation with GH, feeding behavior, and sleep-wake patterns in rats. Endocrinology 2002;143:1353–1361. 13. Hosoda H, Kojima M, Matsuo H, Kangawa K. Ghrelin and des-acyl ghrelin: Two major forms of rat ghrelin peptide in gastrointestinal tissue. Biochem Biophys Res Commun 2000;279:909–913. 14. Asakawa A, Inui A, Fujimiya M, et al. Stomach regulates energy balance via acylated ghrelin and desacyl ghrelin. Gut 2005;54:18–24. 15. van der Lely AJ, Tschop M, Heiman ML, Ghigo E. Biological, physiological, pathophysiological, and pharmacological aspects of ghrelin. Endocr Rev 2004;25:426–457. 16. Akamizu T, Murayama T, Teramukai S, et al. Plasma ghrelin levels in healthy elderly volunteers: The levels of acylated ghrelin in elderly females correlate positively with serum IGF-I levels and bowel movement frequency and negatively with systolic blood pressure. J Endocrinol 2003; 188:333–344. 17. Di Francesco V, Fantin F, Residori L, et al. Effect of age on the dynamics of acylated ghrelin in fasting conditions and in response to a meal. J Am Geriatr Soc 2008;56:1369–1370. 18. Barazzoni R, Zanetti M, Ferreira C, et al. Relationships between desacylated and acylated ghrelin and insulin sensitivity in the metabolic syndrome. J Clin Endocrinol Metab 2007;92:3935–3940. 19. de Graaf C, Blom WA, Smeets PA, et al. Biomarkers of satiation and satiety. Am J Clin Nutr 2004;79:946–961. 20. Blundell JE, Cotton JR, Delargy H, et al. The fat-paradox: Fat-induced satiety signals versus high fat overconsumption. Int J Obes 1995;19: 832–835. 21. Cook CG, Andrews JM, Jones KL, et al. Effects of small intestinal nutrient infusion on appetite and pyloric motility are modified by age. Am J Physiol 1997;273:R755–761. 22. Lim JE, Brubaker PL. Glucagon like peptide 1 secretion by the L-cell: The view from within. Diabetes 2006;55:S70–S77. 23. Baggio LL, Drucker DJ. Biology of incretins. GLP-1 and GIP. Gastroenterology 2007;132:2131–2157. 24. Niswender KD, Schwartz MW. Insulin and leptin revisited: Adiposity signals with overlapping physiological and intracellular signalling capabilities. Front Neuroendocrinol 2003;241:1–10. 25. Williams DL, Baskin DG, Schwartz MW. Leptin regulation of the anorexic response to gucagon-like peptide-1 receptor stimulation. Diabetes 2006;55:3387–3393.
Di Francesco et al 193
Background: In healthy elderly people, reduced appetite and the consequent decrease in food intake has been defined as the ‘‘anorexia of aging’’; this condition may lead to malnutrition. The aim of this study was to investigate how different compositions of macronutrients affect satiety and hunger signals as well as subjective sensations after meals in healthy elderly subjects. Methods: Experimental controlled study. Ambulatory healthy community-dwelling subjects evaluation in a single center on 12 elderly subjects, (75.22 years old) and 12 younger controls (28.22 years old). Using a visual analogical scale, hunger was evaluated under fasting conditions and at 30-minute intervals for up to 4 hours after two 800-kcal meals, where 20% and 40% of the calories were derived from fat. Serum samples were collected at –30, 60, 120, and 240 minutes to determine the concentrations of GLP-1, Protein-energy malnutrition is a frequent condition in the elderly and is associated with a reduction of the adaptive response to physiological and pathological conditions in old age.1 The so-called ‘‘anorexia of aging,’’ ie, an age-related decrease of calorie intake, may be one of the main risk factors for malnutrition in old age. Impaired satiety/hunger control may cause anorexia at more advanced ages.2–4 Peripheral signals influence satiety and hunger both after a meal and under fasting conditions. Interestingly, both the short-term postprandial satiety signals such as cholecystoki-
Geriatric Unit, Department of Biomedical and Surgical Sciences, University of Verona, Verona, Italy (V.D.F., L.B., F.F., L.R., A.A., O.B., M.Z.); Clinica Medica, Department of Clinical, Morphological and Technological Sciences, University of Trieste, Italy (R.B., M.Z., G.G.); Laboratorio di Analisi Chimico Cliniche ed Ematologiche, Azienda Ospedaliera di Verona, Verona, Italy (P.R., M.S.G.). This work was funded by grants from the University of Verona. Address correspondence to Vincenzo Di Francesco, MD, PhD, Dipartimento Scienze Biomediche e Chirurgiche, Sezione di Geriatria, Ospedale Maggiore, p.le Stefani 1, 37126 Verona, Italy. E-mail: [email protected]
Copyright Ó2010 American Medical Directors Association DOI:10.1016/j.jamda.2009.08.004 188 Di Francesco et al
acylated and desacylated ghrelin, triglycerides, glucose, and insulin. Results: Serum concentrations of GLP-1 were higher after the 40% fat meal than after the 20% fat meal (P \ .01) in the elderly but not in the younger subjects. Acylated to desacylated ratio was lower after the 40% fat meal (P \.05) in the elderly. Only in the older group were triglycerides higher (P \ .05), whereas hunger was significantly lower (P\.05) after the 40% fat meal. Conclusion: In healthy elderly people relatively large amounts of fat increase the satiety signal from GLP-1 and lower the acylated to desacylated ratio of ghrelin, consequently decreasing hunger. This condition may lead to a reduction in calorie intake. (J Am Med Dir Assoc 2010; 11: 188–193) Keywords: Ageing; malnutrition; GLP-1; hunger nin (CCK) and peptide YY (PYY) and the long-term adiposity signals from insulin and leptin have been demonstrated to be stronger in the elderly.5–8 The glucagon-like peptide 1 (GLP-1) is a potent regulator of food intake in humans,9 and the short-term postprandial concentrations of GLP-1 have been shown to be similar in both young and older healthy subjects after the isoenergetic duodenal infusion of glucose or lipid.6 However, the GLP-1 levels in the midterm response to a physiological meal in older subjects have not to our knowledge been determined. Ghrelin, a 28 amino acid gastric peptide, generates an orexigenic peripheral signal,10 which triggers meal initiation.11 In fact, ghrelin production has been shown to rise during fasting and to fall after meals.12 Ghrelin has been shown to be activated by acylation of the Serine3 residue of the octanoyl group.13 Acylated ghrelin (A-Ghr) is proposed to be the only biologically active form of the hormone that is able to exert an orexigenic effect.14 Nonacylated ghrelin is the most abundant circulating form, which although devoid of direct neuroendocrine action, is able to bind to receptor subtypes.15 Acylated ghrelin was found to be significantly lower in healthy elderly women.16 Although ghrelin levels JAMDA – March 2010
have a typically dynamic pulsatile distribution in younger subjects, elderly subjects have reduced ghrelin levels and show a flat postprandial curve after meals.17 Lower desacylated ghrelin and relative increments of acylated ghrelin have been suggested to negatively modulate insulin action and to contribute to insulin resistance in adults with metabolic syndrome.18 The effects of acylated to desacylated proportion in the elderly need to be investigated, as meal composition, especially the proportion of lipids, might influence ghrelin response and activation. The aims of this study were (1) to investigate the fasting and postprandial dynamics of GLP-1 in elderly and young controls; and (2) to evaluate the levels of GLP-1 and acylated and desacylated ghrelin in the elderly group and their relationships with regard to subjective hunger sensation in response to isocaloric hyperlipidic or hypolipidic diets.
Experimental Design
PATIENTS AND METHODS
Hunger Evaluation
Excluding criteria were malnutrition or recent weight loss; oral, swallowing, and chewing problems; any abdominal surgical procedure; cholelithiasis; diabetes; neurological diseases (including cognitive impairment); chronic gastrointestinal peptic or inflammatory diseases; malignancies; and any acute ongoing disease. Subjects with known cardiac, renal, or respiratory impairment or a body mass index lower than 18.5 or greater than 30 kg/m2 were also excluded. Individuals taking drugs that may have interfered with gastrointestinal motility and visceral sensitivity (Ca-antagonists, nitrates, prokynetics, proton pump inhibitors, H2-antagonists, sedatives) were also excluded.
A visual analogue scale (VAS) was used to measure subjective hunger. Hunger was defined according to de Graaf et al19 as the subjective driving force for the search, choice, and ingestion of food. Subjects were asked to make a single vertical mark on a horizontal 10-cm bar indicating an assessment of their current feelings from ‘‘not hungry at all’’ to ‘‘really hungry.’’ Baseline evaluations were collected 30 minutes before the meal, at the end of the meal, and 30, 60, 120, and 240 minutes thereafter. Blood samples were collected 30 minutes before and 60, 120, and 240 minutes after the meal. After centrifugation, serum samples were immediately stored at –80 C for final evaluation. Serum levels of glucose and triglycerides were measured using standard methods. Acylated ghrelin was measured using Radio Immuno Assay (RIA) (Linco, St. Charles, MO) (intra-assay coefficient of variation 4.9%; interassay coefficient of variation 8.9%). Serum desacyl-ghrelin was measured using enzyme-linked immunosorbent assay (ELISA) (DRG, New York, NY) following the manufacturer’s recommendations (intra-assay coefficient of variation 6%; interassay coefficient of variation 9.6%). ELISA assay (Phoenix Pharmaceuticals, Belmont, CA) was used for GLP-1 7 to 36 (minimum detectable concentration 0.08 ng/mL). This test has an intra-assay variation of 5% and interassay variation of 14%. For technical reasons, ghrelin evaluation was possible only for the elderly group.
Study Sample A total of 20 elderly subjects were screened from the database of an ongoing study on the effect of physical activity on body composition in community-dwelling healthy autonomous elderly. Two women refused to participate. Two women and 1 man were excluded because they were taking antisecretive drugs. Two men were excluded as well because they were taking Ca-antagonists. One man referred previous cholecystectomy. All the subjects were examined at baseline, before starting the intervention (daily walking program). Fifteen healthy young controls were recruited from medical students and personnel working at our unit. One young woman was excluded because she was taking oral contraceptives and 1 man because he had a measured body mass index (BMI) greater than 30 kg/m2. One woman refused the meal because she referred tomato juice intolerance. The final study sample consisted of 12 elderly subjects, 5 men and 7 women, mean age 75.26 years (range 68–87), BMI 21.1 to 28.3 kg/m2. They all had normal cognitive function (Mini Mental State Examination .24/30) and no functional impairment (evaluated as activities of daily living [ADLs] and instrumental ADLs). As a control group, we evaluated 12 subjects (6 men, 6 women) with mean age 28.22.0 (range 25–33) and BMI 18.9 to 26.5 kg/m2 (P . .05). The study was approved by our institution’s ethical committee and each subject gave their informed consent. ORIGINAL STUDIES
After an overnight fast, both groups were evaluated before eating; they were then asked to consume the standard meal in about 20 minutes. The end of the meal was considered time 0, after which evaluations were repeated every 30 minutes for a further 4 hours. The 20% fat meal (20%FM) meal consisted of 100 g of pasta with 50 g of tomato sauce and 10 g of olive oil, 80 g of lean turkey ham slices, one 60-g roll, and 250 mL of tap water. The total amount of energy was 800 kcal: 15% from proteins, 20% from fat, 65% from carbohydrates. The 40% fat meal (40%FM) consisted of 60 g of pasta with 70 g of bolognese and 10 g of olive oil, 50 g of soft fatty cheese, one 60-g roll, and 250 mL of tap water. The total amount of energy was 800 kcal: 15% from proteins, 40% from fat, 45% from carbohydrates.
Statistical Analysis Results are shown as meansSE. Data were analyzed using a 3-factor repeated measures analysis of variance (ANOVA) for GLP-1, glucose, insulin, triglycerides, and VAS evaluations and a 2-factor repeated measures ANOVA for ghrelin evaluation. A Pearson’s correlation coefficient was calculated to evaluate point-by-point relationships between VAS and laboratory variables in the older group. A significance level of .05 was adopted. The study was designed to verify the hypothesis that mean postprandial GLP-1 value may vary by about 25% after a hyperlipidic meal. The power of the test with 12 subjects in each group was about 90% for Di Francesco et al 189
Fig. 1. Fasting and postprandial concentrations of GLP-1 in the elderly and in younger controls, after 20%FM (full lines) and 40%FM (broken lines).
repeated measurements and 80% for comparison between age groups. Analyses were performed using SPSS for Windows, release 16.0 (Macintosh version, SPSS Inc, Chicago, IL).
RESULTS Figure 1 shows the fasting and postprandial concentrations of GLP-1 in the elderly and in younger controls, after 20%FM (full lines) and 40%FM (broken lines). Serum GLP-1 was higher in the older group but only after the 40%FM (time group meal interaction, P \ .05). In the elderly, but not in the younger group, GLP-1 showed different curves after the 2 meals (meal time interaction, P \ .01) with higher postprandial concentrations after the 40%FM. GLP-1 at 240 minutes after 20%FM: mean 40.8 pg/mL (95% confidence interval [CI]: 27–54); after 40%FM: mean 55.6 pg/mL (95% CI: 45–66). Figure 2 shows the fasting and postprandial concentrations of glucose, insulin, and triglycerides after the 20%FM (full lines) and the 40%FM (broken lines). Serum glucose and insulin showed the expected postprandial rise without differences between the meals (time treatment interaction, P . .05) but with lower concentrations in the younger group for glucose (time group treatment interaction, P \ .01). Triglycerides were higher after the 40%FM in the older but not in the younger group (time treatment interaction, P \ .05). Triglycerides concentrations were higher in the older compared with younger controls (treatment group interaction, P \ .01). Figure 3 shows fasting and postprandial concentrations of acylated ghrelin, desacylated ghrelin, and the plot of the acylated to desacylated ghrelin ratio. Acylated ghrelin showed a flat curve both after the 40%FM and the 20%FM (mean time and time treatment interaction, P . .05). Desacylated ghrelin concentrations fell after the meal both after the 40%FM and the 20%FM (time interaction P \ .01) without meal differences (time treatment interaction, P . .05). The acylated to desacylated ratio was significantly affected 190 Di Francesco et al
by the quantity of meal fat (time treatment interaction, P \ .05) with lower values after the 40%FM. Figure 4 shows the postprandial hunger sensation scores after the 20%FM (grey area) or the 40%FM (white area) in the older and the younger groups. Hunger sensation was significantly more suppressed after the 40%FM compared with the 20%FM but only in the older (time treatment interaction, P \ .05) . Correlations between hunger sensation and laboratory parameters was evaluated in the older group: The hunger sensation was inversely correlated to serum glucose after 20%FM and 40%FM (R5–0.602, P \ .001, and R5 –0.417, P \ .03 respectively) and to insulin but only after 20%FM (R 5 –0.356, P \ .005). Serum triglycerides inversely correlated with acylated ghrelin after 20%FM or 40%FM (R5–0.631, P \ .001, and R5 –0.478, P \ .01 respectively) and directly with serum insulin after 20%FM or 40%FM (R50.575, P \ .001, R5 0.483, P \ .01 respectively). DISCUSSION This study demonstrates that in healthy elderly men and women the postprandial serum concentrations of GLP-1 were higher and the ratio of acyl to desacyl ghrelin was lower after the consumption of a meal where 40% of the calories were derived from fat, when compared with an isocaloric meal where only 20% of calories came from fat. Consequently, the hyperlipidic meal had a stronger inhibitory effect on hunger. In the younger control group, neither the hunger sensation nor the postprandial GLP-1 curves were influenced by the quantity of alimentary fat. The levels of GLP-1 were higher in the older group but only after the 40% fat meal. Blundell et al20 coined the term ‘‘fat-paradox’’ to describe how food rich in fat, notwithstanding the generation of higher satiety signals, causes higher intake of calories in adults, thus leading to obesity. Hunger suppression in the elderly seems to be much stronger after a fatty meal thereby counteracting the small volume effect and the high caloric density of lipids. JAMDA – March 2010
Fig. 2. Fasting and postprandial concentrations of glucose, insulin and triglycerides after 20%FM (full lines) and 40%FM (broken lines) in older and younger groups.
It has been suggested that lipids cause a more potent suppression of hunger and food intake than glucose in younger but not older subjects in a study that used small intestinal nutrient infusion.21 This does not seem to be the case in our study, where we used a physiological meal and the stomach passage may account for the apparent differences. Our data seem to confirm the study hypothesis, that is, that a relatively low amount of fat causes lower hunger inhibition, as alimentary fatty acids are among the major determinants of GLP-1 secretion.22 GLP-1 levels rise in blood also after glucose ingestion, through insulin stimulation,23 but our study demonstrated a significantly higher postprandial GLP-1 curve ORIGINAL STUDIES
in the elderly only after the hyperlipidic meal. Macintosh et al did not find postprandial differences in GLP-1 dynamics in healthy elderly compared with young controls.6 Nevertheless, the Australian researchers used liquid nutrients directly infused into the duodenum. Furthermore, their observations were limited to the 2 hours after infusion. We used mixed meals and the observation period was prolonged to 4 hours, which may have helped to detect the differences between the 2 age groups. Insulin increases GLP-1 secretion,22,24 although contradictory results have been published.23 In this study, meal macronutrient composition did not affect the insulin curves. So the Di Francesco et al 191
Fig. 3. Fasting and postprandial concentrations of acylated ghrelin, desacylated ghrelin, and the plot of the acylated to desacylated ghrelin ratio in the older group.
difference of GLP-1 response after the 40%FM cannot be directly attributed to insulin because this hormone was unaffected by the composition of the meal. GLP-1 secretion is considered to be a response to hyperinsulinemia, as it has been shown to improve peripheral insulin sensitivity.22 Higher concentrations of GLP-1 may raise hypothalamic sensitivity to insulin, thus indirectly improving hunger suppression. Leptin also acts on energetic control by stimulating the secretion of GLP-1.22–25 Hyperleptinemia has been described in the elderly,8 but leptin concentrations were not affected in the short term by the meal. It is therefore unclear how leptin may influence GLP-1 secretion after different alimentary loads. Furthermore, the hypothesis that a higher fat load may potentiate leptin stimulus on GLP-1 is contradicted by the experimental observation that a hyperlipidic meal causes greater resistance to the leptin signal.25 Different postprandial curves of GLP-1 and hunger sensations after 20% or 40% fat may be influenced by different 192 Di Francesco et al
gastric emptying dynamics, because both alimentary fat and GLP-1 delay gastric emptying.23 Therefore, the hyperlipidic meal may inhibit gastric emptying and consequently hunger, either directly or by elevating GLP-1 levels. We could not demonstrate a correlation between the hunger sensation scores and point-by-point GLP-1 concentrations, but with this model it is not possible to evaluate the net effect of GLP-1 on hunger, avoiding all the other neurohormonal signals of the network controlling energy balance. In other experimental models GLP-1 injection was shown to suppress hunger but only at very high doses.9 Using a similar experimental setting, we have previously demonstrated that elderly subjects have lower plasmatic concentrations of acylated ghrelin than their younger counterparts. Furthermore, acylated ghrelin showed a flat postprandial curve in the elderly and pulsating dynamic curves in the younger controls.17 In the present study, we evaluated the effects of lipids on the dynamics of the different forms of ghrelin in the postprandial period. The ratio of acylated to desacylated ghrelin was significantly affected by the quantity of fat in the meal with lower values detected after the 40%FM. The relative prevalence of the active acylated form over the desacylated ghrelin 4 hours after the hypolipidic meal may explain the higher hunger sensation. It is well known that acylated ghrelin stimulates hunger and food consumption.14 Relative increments of acylated ghrelin have been suggested to negatively modulate insulin action and to contribute to insulin resistance in adults with metabolic syndrome.18 It may be hypothesized that a reduced acyl to desacyl ratio, as observed after the hyperlipidic meal, could increase hypothalamic insulin sensitivity in the elderly, thus leading to an increased inhibition of hunger. Some limitations of our study deserve a comment. We studied a sample of healthy elderly people who were not suffering from chronic diseases and were not taking medications. Nevertheless, it is reasonable to assume that chronic illness, medications, and malnutrition would only worsen the abnormalities seen in our subjects.2 Because of the small size of our sample, we were unable to evaluate the effect of gender on satiety/hunger and peptide dynamics. In this observational study, results may be confounded by unmeasured factors such as unmeasured biological parameters that were not extensively evaluated and deserve further study. In conclusion, our study shows that in healthy elderly subjects, the consumption of an 800-kcal meal with 40% calories derived from fat, compared with an isocaloric meal with 20% calories from fat, caused higher postprandial serum concentrations of GLP-1 and lower acyl to desacyl ghrelin ratios. As a result, the hyperlipidic meal was more effective in suppressing hunger and delaying the search for a second meal. This condition may lead to calorie restriction and eventually to malnutrition. In other words, in the elderly, fatty foods may actually cause undernutrition so that the ‘‘fat paradox’’ is lost. Under a clinically practical point of view, these conclusions should discourage the frequent use of fatty food (including oil, butter, and so forth) as a choice for improving calorie intake in frail older persons living at home or in nursing JAMDA – March 2010
Fig. 4. Fasting and postprandial hunger sensation scores after 20%FM (grey area) or 40%FM (white area) in the 2 age groups.
homes. They are already frequently suffering from secondary anorexia and malnutrition, thus fat overload at lunch would blunt dinner food consumption and net calorie intake could be even further reduced. REFERENCES 1. Mowe` M, Bohmer T, Kindt E. Reduced nutritional status in an elderly population (.70y) is probable before disease and possibly contributes to the development of disease. Am J Clin Nutr 1994;59:317–324. 2. Morley JE. Anorexia of aging: Physiologic and pathologic. Am J Clin Nutr 1997;66:760–773. 3. Roberts SB, Fuss P, Heyman MB, et al. Control of food intake in older men. JAMA 1994;272:1601–1606. 4. Di Francesco V, Fantin F, Omizzolo F, et al. The anorexia of aging. Dig Dis 2007;25:129–137. 5. Di Francesco V, Zamboni M, Dioli A, et al. Delayed postprandial gastric emptying and impaired gallbladder contraction together with elevated cholecystokinin and peptide YY serum levels sustain satiety and inhibit hunger in healthy elderly persons. J Gerontol A Biol Sci Med Sci 2005; 60:1581–1585. 6. MacIntosch GC, Andrews JM, Jones KL, et al. Effects of age on concentrations of plasma cholecystokinin, glucagon-like peptide 1 and peptide YY and their relation to appetite and pyloric motility. Am J Clin Nutr 1999;69:999–1006. 7. MacIntosh CG, Morley JE, Wishart J, et al. Effects 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. 8. Di Francesco V, Zamboni M, Zoico E, et al. Unbalanced serum leptin and ghrelin dynamics prolong postprandial satiety and inhibit hunger in healthy elderly: Another reason for the ‘‘anorexia of aging’’. Am J Clin Nutr 2006;83:1149–1152. 9. Gutzwiller JP, Go¨ke B, Drewe J, et al. Glucagon-like peptide-1: A potent regulator of food intake in humans. Gut 1999;44:81–86. 10. Asakawa A, Inui A, Kaga T, et al. Ghrelin is an appetite-stimulatory signal from stomach with structural resemblance to motilin. Gastroenterology 2001;120:337–345. 11. Cummings DE, Purnell JQ, Frayo RS, et al. A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 2001; 50:1714–1719.
ORIGINAL STUDIES
12. Tolle V, Bassant MH, Zizzari P, et al. Ultradian rhythmicity of ghrelin secretion in relation with GH, feeding behavior, and sleep-wake patterns in rats. Endocrinology 2002;143:1353–1361. 13. Hosoda H, Kojima M, Matsuo H, Kangawa K. Ghrelin and des-acyl ghrelin: Two major forms of rat ghrelin peptide in gastrointestinal tissue. Biochem Biophys Res Commun 2000;279:909–913. 14. Asakawa A, Inui A, Fujimiya M, et al. Stomach regulates energy balance via acylated ghrelin and desacyl ghrelin. Gut 2005;54:18–24. 15. van der Lely AJ, Tschop M, Heiman ML, Ghigo E. Biological, physiological, pathophysiological, and pharmacological aspects of ghrelin. Endocr Rev 2004;25:426–457. 16. Akamizu T, Murayama T, Teramukai S, et al. Plasma ghrelin levels in healthy elderly volunteers: The levels of acylated ghrelin in elderly females correlate positively with serum IGF-I levels and bowel movement frequency and negatively with systolic blood pressure. J Endocrinol 2003; 188:333–344. 17. Di Francesco V, Fantin F, Residori L, et al. Effect of age on the dynamics of acylated ghrelin in fasting conditions and in response to a meal. J Am Geriatr Soc 2008;56:1369–1370. 18. Barazzoni R, Zanetti M, Ferreira C, et al. Relationships between desacylated and acylated ghrelin and insulin sensitivity in the metabolic syndrome. J Clin Endocrinol Metab 2007;92:3935–3940. 19. de Graaf C, Blom WA, Smeets PA, et al. Biomarkers of satiation and satiety. Am J Clin Nutr 2004;79:946–961. 20. Blundell JE, Cotton JR, Delargy H, et al. The fat-paradox: Fat-induced satiety signals versus high fat overconsumption. Int J Obes 1995;19: 832–835. 21. Cook CG, Andrews JM, Jones KL, et al. Effects of small intestinal nutrient infusion on appetite and pyloric motility are modified by age. Am J Physiol 1997;273:R755–761. 22. Lim JE, Brubaker PL. Glucagon like peptide 1 secretion by the L-cell: The view from within. Diabetes 2006;55:S70–S77. 23. Baggio LL, Drucker DJ. Biology of incretins. GLP-1 and GIP. Gastroenterology 2007;132:2131–2157. 24. Niswender KD, Schwartz MW. Insulin and leptin revisited: Adiposity signals with overlapping physiological and intracellular signalling capabilities. Front Neuroendocrinol 2003;241:1–10. 25. Williams DL, Baskin DG, Schwartz MW. Leptin regulation of the anorexic response to gucagon-like peptide-1 receptor stimulation. Diabetes 2006;55:3387–3393.
Di Francesco et al 193