| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Section of Endocrinology, Department of Medicine, University of Chicago (P.P., T.M., E.V.C.), Chicago, Illinois 60637; and Centre d’Etude des Rythmes Biologiques, Laboratoire de Physiologie, Université Libre de Bruxelles (K.S.), B-1070 Brussels, Belgium
Address all correspondence and requests for reprints to: Dr. Plamen Penev, Section of Endocrinology, University of Chicago, MC 1027, 5841 South Maryland Avenue, Chicago, Illinois 60637. E-mail: ppenev{at}medicine.bsd.uchicago.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of this study was to examine the changes in the impact of carbohydrate-rich meals on circulating catecholamines with aging.
Design: After iv glucose tolerance testing and 1 d of habituation, blood samples were collected every 10–30 min for 24 h. Daytime hours were spent at bed rest. Sleep was scheduled between 2300 and 0700 h with polygraphic monitoring.
Setting: The study was performed at a general clinical research center.
Participants: Nine young (age, 20–28 yr) and eight older (age, 50–69 yr) healthy men participated in this study.
Intervention: Identical mixed meals (62% carbohydrate) were given at 0900, 1400, and 1900 h.
Main Outcome Measures: The main outcome measures were 24-h plasma E and NE measurements.
Results: The profiles of E and NE were characterized by clear day-night differences, which were preserved in the older group. Young subjects showed a clear dissociation between postprandial adrenomedullary and sympathetic activities characterized by a rapid decline in plasma E and increased NE levels. There was an overall increase in NE levels and markedly dampened postprandial variation in plasma E in the older men.
Conclusions: In young adults, postprandial E levels follow a biphasic pattern that is inversely related to that of glucose and insulin. Aging is associated with a dysregulation of this response.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Under routine sleep-wake and rest-activity schedules, plasma catecholamines exhibit 24-h rhythmicity, with higher levels during the daytime period of activity and lower levels during recumbency and sleep (5, 6, 7, 8). The sympathoadrenomedullary system is also involved in regulation of the visceral, metabolic, and hemodynamic changes related to carbohydrate intake in humans (2). Most experimental studies indicate that ingestion of carbohydrate-rich or mixed meals in healthy young subjects is associated with a reduction in splanchnic and skeletal muscle vascular resistance associated with an increase in SNS outflow, heart rate, cardiac output, and plasma NE levels (9, 10, 11, 12). Fewer studies have examined the corresponding changes in adrenomedullary activity (12) and plasma E levels after ingestion of carbohydrates (13, 14). It has been generally accepted that plasma E levels remain relatively unchanged during the immediate postprandial period and rise some 4–5 h after carbohydrate ingestion (13). Other studies, however, have reported a decline in circulating E levels after ingestion of glucose (14, 15) or mixed meals (16, 17). In all studies to date, the need for a sensitive assay and/or arterial sampling has limited the detailed description of the impact of mixed meals on circulating E levels.
Although aging is accompanied by region-specific increases in SNS activity and higher plasma NE levels, adrenomedullary secretion has been found to decline (18). Circulating E levels, however, remain relatively unchanged due to a concomitant decrease in plasma E clearance (18). To date, the effects of aging on circulating E during a 24-h sleep-wake routine with three carbohydrate-rich meals have not been well documented. The present study compared the impact of mixed carbohydrate-rich meals on the postprandial and 24-h profiles of plasma catecholamines, and in particular E, in healthy young and older men. Given that 1) alterations in SNS activity and circulating E levels have been related to changes in glucose homeostasis (19, 20), and 2) there is age-related deterioration in glucose tolerance with increased insulin resistance and a relative decline in insulin secretion (21, 22), we also performed iv glucose tolerance testing and measured the postprandial and 24-h profiles of blood glucose and insulin in our subjects. These measures were then analyzed in relation to the differences in plasma catecholamines between the two age groups.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Nine young men [age, 20–28 yr; body mass index (BMI), 21–25 kg/m2] and eight older men (age, 50–69 yr; BMI, 22–28 kg/m2) completed the study (Table 1
). All of them were in good health based on review of their medical histories, physical exams, and laboratory tests (complete blood count, fasting lipid, thyroid and comprehensive metabolic panels, electrocardiogram, and urinalysis). All subjects were fully self-sufficient normotensive nonsmokers who did not take prescription medications (except for one older man receiving statin therapy for cholesterol reduction). Research volunteers with irregular life habits, shift or night work schedules, recent (<4 wk) travel across time zones, sleep complaints, excessive alcohol (>10 drinks/wk) or caffeine use (>300 mg/d), and personal history of endocrine, metabolic, neurological, or psychiatric disorders were excluded. All subjects gave written informed consent and were paid for their participation in the study.
|
The study protocol was approved by the institutional review board of University of Chicago. During the week preceding the study, the subjects were asked to adhere to a regular schedule, with bedtimes between 2300 and 0700 h; scheduled meals around 0900, 1400, and 1900 h (±30 min); and no daytime naps. Wrist activity was monitored continuously (Actiwatch, Mini-Mitter Co., Bend, OR) to verify compliance as previously described (23). The subjects were then admitted to the General Clinical Research Center (CRC) at University of Chicago, where they spent two 24-h periods with bedtimes between 2300 and 0700 h and polygraphic sleep monitoring. Except for scheduled bathroom time between 0700 and 0800 h as needed, the study participants spent all daytime hours at bed rest in a semirecumbent position. Caloric intake included a single 0.3 g/kg glucose bolus, which was part of an iv glucose tolerance test (IVGTT) starting at 0900 h on the first morning, followed by identical carbohydrate-rich meals (30 kcal/kg·d; 62% carbohydrate, 15% protein, and 23% fat) at 1400 and 1900 h on the first day and at 0900 h (morning meal), 1400 h (midday meal), and 1900 h (evening meal) on the second day. Participants were required to eat their entire meals within a 30-min period. There was no other source of calories, and only decaffeinated diet sodas were allowed in addition to ad libitum water intake.
Experimental procedures
All subjects underwent a tolbutamide-assisted IVGTT on the first morning after their CRC admission. Starting at 0900 h, blood samples were drawn every 5 min over a 25-min period, followed by a 0.3 g/kg iv glucose bolus. Additional blood samples were collected 2, 3, 4, 6, 8, 10, 12, 14, 16, 19, 22, 24, 25, 27, 30, 40, 50, 60, 70, 80, 90, 100, 120, 140, 160, and 180 min after the glucose challenge, and an iv tolbutamide injection (125 mg/m2 body surface area) was given 20 min after the glucose bolus.
During the second day at the CRC, blood samples were collected every 30 min for 24 h, except for the first hour after each carbohydrate-rich meal, when blood was drawn every 10 min, and the first 2 h of scheduled bedtime, when blood samples were collected every 15 min. During all bedtime hours, the sampling catheter was connected to plastic tubing extending to an adjacent room to allow blood collection without disturbing the subject’s sleep.
The first night of sleep recording was used for habituation, and sleep analysis was based on data obtained during the second night. Polysomnographic recordings consisted of central and occipital electroencephalogram, electrooculogram, and electromyogram tracings, which were scored according to standardized clinical criteria (24). The sleep recordings of two young subjects (no. 5 and 7) were incomplete due to equipment failure and were not included in the final analysis.
Assays
Plasma glucose concentrations were measured in duplicate at the bedside on a glucose analyzer (STAT Plus 2300, Yellow Springs Instrument Co., Yellow Springs, OH) with a coefficient of variation of less than 2%. All other blood samples were centrifuged at 4 C and stored at –20 C until assay for insulin, NE, and E. All samples collected from the same subject were run in the same assay. Serum insulin concentrations were measured in duplicate using a double-antibody RIA technique with a limit of sensitivity of 20 pmol/liter and an average within-assay coefficient of variation of 6%. Plasma catecholamines were measured using a HPLC system (Coulochem MD5001, ESA, Inc., Chelmsford, MA). With this method, catecholamines undergo initial extraction and purification by absorption to silica powder, followed by HPLC column separation and electrochemical detection. The procedure has a sensitivity of 10 pg/ml for NE and 5 pg/ml for E, and an average within assay coefficient of variation of 6–7.8% at physiological catecholamine levels. A value equal to 50% the sensitivity of the assay was assigned to all measured samples with values below that limit. Less than 9% of the daytime samples, but 42% of all nighttime samples, had values below the detection limit for E. The iv sampling line of one young subject (no. 8) had to be replaced in the middle of the night, which caused a transient surge in circulating catecholamine levels. Therefore, a 2-h block of biochemical measurements for this subject starting 30 min before the line failure was discarded from additional analysis. The sleep data of this subject were also excluded from the final analysis.
Data analysis
Minimal model analysis (MINMOD, version 5.01; Bergman & Stefanovski Assoc., Boston, MA) was used to derive measures of insulin sensitivity (SI index) and ß-cell function [acute insulin response to glucose (AIRg)] from the individual IVGTT datasets (25). The glucose disposition index (DI), derived by multiplying the SI and AIRg indices, was also calculated as an integrative measure of individual glucose handling.
Sleep onset was defined as the timing of the first epoch of stage II sleep. Total sleep time was calculated as the sum of all periods after sleep onset, scored as stage I, II, III, IV, or rapid eye movement sleep. Sleep efficiency was calculated as the percentage of the time in bed, which was scored as sleep. Sleep maintenance was calculated as the percentage of time between sleep onset and morning awakening, which was scored as sleep.
The 24-h profiles of glucose, insulin, E, and NE were analyzed during the following physiologically defined intervals: 1) nighttime sleep period between the individual times of sleep onset and morning awakening; 2) premeal period between 0800 and 0900 h for the morning meal, 1300–1400 h for the midday meal, and 1800–1900 h for the evening meal; and 3) postprandial period between 0900 and 1300 h for the morning meal, 1400–1800 h for the midday meal, and 1900–2300 h for the evening meal. Fasting levels were defined as the average of all blood measurements obtained between 0800 and 0900 h. The mean levels of the variables of interest were used in the comparisons of the nighttime sleep and the three daytime premeal periods. The change in the variables of interest relative to their premeal levels was used in the analysis of all postprandial periods. A comparison of the 24-h homeostatic model assessment of insulin resistance (fasting insulin x fasting glucose/22.5) profiles of the young subjects in the present study under varying sleep conditions was previously reported (26).
Statistics
The comparison of anthropometric and sleep-related parameters between the two age groups was performed using Student’s t test for independent samples. A two-way ANOVA, with age group (young vs. older) as an independent variable and time of day (sleep, morning, midday, and evening premeal periods) as a repeated measure variable, was used to examine the 24-h variation in all variables of interest (these 24-h fluctuations will be referred to as diurnal, because the endogenous or exogenous nature of their driving mechanisms cannot be defined with this experimental design). A second ANOVA, with age group as an independent variable and meal (morning, midday, and evening) and postprandial time (the four consecutive 1-h intervals after each meal) as repeated measures, was used to examine the biochemical changes related to the recurrent ingestion of identical carbohydrate-rich meals. Greenhouse-Geisser correction was used in these analyses whenever the assumption of sphericity was not met.
We also performed cross-correlation analysis to explore the temporal relationships among glucose, insulin, E, and NE during the daytime period of repeated meal ingestion as follows. The individual data series of these variables were interpolated at 10-min intervals between 0800 and 2300 h, and the cross-correlation coefficients between glucose-E, glucose-NE, insulin-E, and insulin-NE were calculated at time lags 0, ±10, ±20, and ±30 min. For each pair of variables, the lag time with the highest cross-correlation coefficient in the young group was identified and compared with the matching time lag value in the group of older subjects using Student’s t test for independent samples. The individual cross-correlation coefficients in both age groups were Fisher-transformed before this comparison.
Finally, partial correlation coefficients controlling for BMI and age group (young vs. older) were calculated to explore the associations of SI, AIRg, and DI with the mean levels of glucose, insulin, NE, and E during the period of nighttime sleep, a time interval that was free of the immediate effects of meal ingestion on these variables. All statistical analyses were performed using the SPSS 11.0 software package (SPSS, Inc., Chicago, IL). Group data in the text are presented as the mean ± SD and in the figures as the mean ± SE. Statistical significance is assumed at P < 0.05, and values between 0.05 and 0.08 are reported as trends.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Figure 1 shows the mean profiles of plasma E, NE, glucose, and insulin levels delineating their day-night and postprandial variations and age-related differences. Twenty-four-hour changes in E and NE levels were seen in both age groups (Fig. 1, upper four panels), with higher levels during the day and lowest levels during the sleep period (time of day effect, P < 0.001 for both E and NE). In contrast to the concomitant declines in E and NE at night, discordant changes in circulating catecholamines with an overall increase in NE and a decrease in E were seen during the postprandial periods.
|
, the fluctuations in E levels during the daytime period differed significantly between the two age groups. Postprandial decreases, which resembled the mirror image of the meal-related rise in glucose and insulin, were repeatedly seen in the young subjects, but appeared markedly blunted in the older subjects. Although the impact of meals on plasma E concentrations was clearly affected by age, the mean E levels of the older subjects during the wake period remained similar to those of the young subjects (P = 0.541). During sleep, a trend for higher E levels in the older group emerged, but failed to reach statistical significance.
Consistent with previous reports, the older subjects had significantly higher NE levels than the young (P = 0.003) throughout the 24-h sleep-wake cycle (Fig. 1, second row of panels from the top). The overall pattern of 24-h variation in plasma NE concentrations in the older group, however, was well preserved and remained comparable to that of the young subjects.
The 24-h profiles of glucose and insulin (Fig. 1, bottom four panels) were also similar in the two age groups, with the older subjects showing a trend (P = 0.056) toward higher 24-h mean glucose levels (105 ± 6 mg/dl) compared with the young subjects (97 ± 8 mg/dl).
Premeal and sleep levels of catecholamines, glucose, and insulin
To separate the 24-h variations in E, NE, glucose, and insulin from the impact of meals, Fig. 2 shows only the mean concentrations of these variables during nighttime sleep and the periods before the morning, midday, and evening meals. The premeal E concentrations in the young group increased across the daytime period, reaching their highest level before dinner, whereas those in the older subjects remained relatively constant throughout the day (Fig. 2, top panel; P = 0.035 for time of day x group effect). In contrast, the 24-h variation in NE was similar in both age groups, albeit occurring at a higher mean level in the older adults. The premeal NE levels did not change significantly from morning to evening in either age group (Fig. 2, second panel).
|
; P = 0.004 for time effect). This recurrent exposure to carbohydrate-rich meals tended to have a more pronounced impact on the preprandial insulin levels of the older subjects (P = 0.057 for age group effect; P = 0.074 for time x group interaction). In the older subjects, insulin levels failed to return to baseline after exposure to the first carbohydrate-rich meal, and the midday and evening preprandial insulin levels were, on the average, more than 2-fold higher than those in the young subjects. Postprandial changes in catecholamine, glucose, and insulin levels
Figure 3A illustrates the temporal pattern of postprandial change in catecholamines, glucose, and insulin based on the combined responses to all three meals, whereas Fig. 3B shows the impact of time of day (morning, midday, and evening) on the integrated meal effect during the corresponding 4-h postprandial period.
|
, top two panels). In the young subjects, food intake was accompanied by an initial rapid decline in E, followed by a gradual rise toward premeal levels. The postprandial E profile differed significantly in the older subjects (P < 0.001 for postprandial time x group effect), in whom the ingestion of carbohydrate-rich meals was associated with a relatively small initial decline in plasma E, stable levels during the next 3 h, and a late return to preprandial levels during the hour immediately before the next meal. Unlike E, the postprandial NE levels in both age groups rose rapidly to reach the highest levels during the first hour after the carbohydrate-rich meals and then decreased gradually toward premeal values during the subsequent 2 h. When adjusted for the difference in premeal NE levels between the two age groups, both the magnitude and the temporal profile of the postprandial NE rise of the older subjects were comparable to those in the young subjects (P = 0.296 for group effect).
The recurrent ingestion of carbohydrate-rich meals in both age groups was accompanied by the expected fluctuations (postprandial time effect, P < 0.001) in blood glucose and insulin. These variables rose rapidly, reaching the highest levels during the first 2 h after each meal and then declined toward their premeal baseline over the next 2 h (Fig. 3A, bottom two panels). When adjusted for the difference in premeal levels, the magnitude and temporal profile of the postprandial rise in glucose and insulin in the older subjects were comparable to those in the young subjects (group effect, P = 0.995 and 0.851, respectively).
Impact of time of day on meal-related changes in catecholamines, glucose, and insulin
Figure 3B shows that the impact of time of day on the meal-related changes in NE, glucose, and insulin was qualitatively similar in the two age groups. Age-related differences, however, were apparent in the E response to meals with varying time of day. Indeed, in the young group, the overall decline in E during the 4-h postprandial period became progressively more pronounced from morning to evening (time of day x postprandial time effect, P = 0.031). In contrast, time of day did not significantly affect the magnitude of the postprandial E decrease in the group of older subjects (Fig. 3B, top panel).
In both age groups, the overall NE increase during the postprandial period was most pronounced in the morning (Fig. 3B, second panel) and declined progressively with the midday and evening meals (P = 0.003 for time of day effect). Unlike NE, the 4-h postprandial increase in glucose in the young and older subjects was smallest in the morning (Fig. 3B, third panel) and became progressively larger after the midday and evening meals (time of day effect, P = 0.005). In contrast to the glucose response and similar to the NE response to meals, the postprandial insulin increase was highest after the morning meal and became smaller in the afternoon and evening (time of day effect, P < 0.001) in both young and older subjects (Fig. 3B, bottom panel).
Relationship between daytime catecholamine variations and those in glucose and insulin
Cross-correlation analysis allowed for quantification of the inverse relationship between the daytime variations in E levels and those in glucose and insulin in the young subjects. The highest cross-correlation coefficients (–0.42 ± 0.15 and –0.47 ± 0.11 for glucose and insulin, respectively; P < 0.01 for significance of mean cross-correlation) were detected when E values preceded those of glucose and insulin by 10 min. The inverse association of E with glucose and insulin was significantly attenuated in the older subjects (cross-correlation coefficient, –0.16 ± 0.15; P = 0.002 for comparison by age group; P > 0.10 for significance of mean cross-correlation; and cross-correlation coefficient, –0.21 ± 0.14; P < 0.001 for comparison by age group; P > 0.10 for significance of mean cross-correlation, respectively).
A positive association of the changes in NE with the patterns of glucose and insulin was detected in the young group; the cross-correlation coefficients were maximal when NE values preceded those of glucose and insulin by 20 min (0.43 ± 0.15 and 0.46 ± 0.15, respectively; P < 0.01 for significance of mean cross-correlation). The association of daytime NE levels with glucose and insulin in the older subjects was qualitatively similar, but somewhat more variable (cross-correlation coefficient, 0.22 ± 0.15; P = 0.081 for comparison by age group; P > 0.10 for significance of mean cross-correlation; and cross-correlation coefficient, 0.37 ± 0.20; P = 0.312 for comparison by age group; P < 0.01 for significance of mean cross-correlation, respectively).
IVGTT
Minimal model IVGTT analysis showed a significantly lower SI in the older group (P = 0.026). The reduction in SI with increasing age was confirmed by analysis of covariance (P = 0.045) when the higher BMI of the older subjects (P = 0.010) was controlled for. The older men also had higher fasting blood glucose levels (P = 0.046) and AIRg indices (P = 0.045) than the young subjects. The age-related increase in ß-cell responsiveness, as quantified by the AIRg, compensated for the decrease in insulin sensitivity, such that the mean DIs of the two age groups were essentially identical (Table 1).
Sleep
The sleep measures of the two age groups are summarized in Table 1. The older subjects had significantly less total sleep and more waking time during the scheduled bedtimes (P = 0.016), which resulted in lower sleep efficiency (P = 0.011) and sleep maintenance (P = 0.005) compared with the young group.
The mean E level during the sleep period was correlated negatively with SI (partial correlation coefficient, –0.48; P = 0.069) and positively with the corresponding mean glucose level (partial correlation coefficient, 0.51; P = 0.051) when age group and BMI were controlled for. Thus, higher plasma E during the sleep period tended to be associated with increased insulin resistance and higher blood glucose levels in the combined data from both age groups. There were no significant associations between mean NE levels during sleep and SI. Within each age group, there were no correlations between sleep amount, efficacy, or maintenance and mean E or NE levels during sleep.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Measurements of circulating E levels during the early postprandial period in the past have produced conflicting results. Early reports indicated that carbohydrate intake had little immediate impact on plasma E levels (13, 27). More recent studies using oral glucose tolerance testing and venous blood sampling (14, 15) or mixed meals and arterial blood measurements (16, 17) have detected lower E levels when glucose and insulin increased in the early postprandial period. In agreement with the later reports, our study demonstrates reproducible declines in venous blood E after ingestion of mixed carbohydrate-rich meals by the young subjects. In addition, our data indicate that the postprandial E profile follows a biphasic pattern, reflecting those of glucose and insulin in a mirror fashion. As a result, the initial rapid decline in plasma E was followed by a delayed rise some 3–4 h after the meal. The transition from exogenous glucose delivery to endogenous glucose production during the late postprandial period is known to involve the active release of glucagon and E to maintain plasma glucose homeostasis (27). The demonstration of reciprocal changes in plasma E and glucose/insulin levels at the very onset of the carbohydrate-rich meals in our study suggests an extended role for E during the entire postprandial period, including its earliest stages. In our cross-correlation analysis, the decline in E preceded the rise in glucose/insulin, suggesting that it may be part of the cephalic phase of the integrated response to meals. These observations lead us to speculate that the rapid E decline at the onset of carbohydrate-rich meals may facilitate the neurohumoral and metabolic transition from endogenous glucose production to exogenous glucose delivery.
The blunting of the biphasic postprandial profile of plasma E in older men is another interesting finding of our study. It may be the result of an age-related decline in the secretory function of adrenomedullary chromaffin cells (28) and/or decreased adrenomedullary responsiveness to neurohumoral stimuli. The blunted suppression of plasma E in the early postprandial period suggests the presence of decreased responsiveness of the adrenomedullary axis to rising glucose and/or insulin levels in older subjects. The association of higher mean blood glucose levels (in the setting of increased insulin resistance) with higher E levels during sleep in our study may be another manifestation of reduced adrenomedullary responsiveness to such humoral signals. Consistent with previous reports (27), during the late postprandial period, insulin levels in the young subjects remained transiently elevated relative to the rapid decline in plasma glucose (Fig. 4). In agreement with the role of E as a counterregulatory factor, this relative insulin excess coincided with the time of rising E levels (27). The absolute glucose levels in the older adults during the late postprandial period were significantly higher compared with those in the young subjects (fourth hour plasma glucose, 101 ± 10 mg/dl in the old vs. 88 ± 9 mg/dl in the young group; P < 0.01). As a result, the stimulus for E secretion during the late postprandial period in the older group may have been significantly weaker (Fig. 4), which, combined with the reported attenuation in stress-induced E secretion in older adults (18), could account for the age-related differences in E profiles seen 3–5 h after meals.
|
Because the meal-related changes in plasma catecholamines preceded the changes in glucose and insulin, our cross-correlation analysis supports previous observations that stimuli, such as food presentation, deglutition, gut distension, and circulatory changes related to food digestion and/or absorption, may trigger sympathetic neural reflexes and result in NE release (10, 27). The rise in circulating insulin levels after carbohydrate intake has also been shown to contribute to the alterations in postprandial SNS activity and vascular tone (31). Ingestion of carbohydrate-containing meals by healthy adults is accompanied by reduced peripheral vascular resistance, increased heart rate and cardiac output, and sympathetic outflow to the kidneys and skeletal muscle (9, 10, 11, 32). Although not dependent on the presence of a postprandial rise in insulin, these cardiovascular responses were more pronounced when insulin secretion was stimulated by carbohydrate-rich or mixed meals (10, 11). The parallel changes in the prandial release of insulin and NE as a function of the time of day in both age groups indicate that the hormonal modulation of meal-related SNS activity is well preserved in healthy older men.
Despite some conflicting reports (33, 34), a number of studies support the existence of diurnal variation in plasma E and NE levels (5, 6, 7, 8, 35, 36). The use of polygraphic sleep monitoring enabled us to analyze these changes in relation to the individual times of sleep onset and morning awakening independent from the scheduled transitions between daytime semirecumbency and supine sleeping posture at night (data not shown). The analysis indicated that the major part of the nocturnal decline in plasma E and NE was achieved before sleep onset, and conversely, that the major part of the morning rise in catecholamine concentrations occurred after sleep offset. Because a decrease in SNS activity is thought to precede sleep onset, circadian effects may underlie some of these changes (35, 37); however, this and other studies (6, 29) indicate that postural changes between daytime semirecumbency and supine sleep contribute significantly to the nighttime decline and morning rise in circulating catecholamines.
As illustrated by the progressive rise in premeal insulin levels during the daytime, the repeated ingestion of carbohydrate-rich meals was a more severe metabolic challenge for the older subjects. This is consistent with the presence of increased insulin resistance in the older group, requiring appropriate increments in insulin secretion to maintain blood glucose homeostasis in the face of recurrent carbohydrate loads during the day. The maximal insulin response to meals in the older subjects occurred in the morning and was associated with the smallest 4-h postprandial rise in blood glucose. Therefore, the morning hours appear to be a more favorable time for higher carbohydrate intake by older subjects. In contrast to the preserved insulin secretory response to meals, the increase in morning fasting blood glucose in the older subjects was not accompanied by a corresponding increase in circulating insulin levels (Fig. 4, age group differences in glucose and insulin between 0800 and 0900 h in the top two panels), suggesting the presence of reduced responsiveness of the aging ß-cells to increments in ambient glucose levels (18). The mechanisms that underlie this age-related shift in the set point for fasting blood glucose remain to be explored. These may include vascular and degenerative changes affecting the sensing and/or responsiveness of the central nervous system or pancreatic islets to changes in ambient glucose. Physiological increments in plasma E, such as those related to stressful events, can adversely affect insulin secretion and action (19, 20, 38); however, the impact of age-related changes in plasma E in the low end of the physiological range on glucose homeostasis remains unknown.
In summary, the results of this study demonstrate that ingestion of carbohydrate-rich meals is associated with a rapid decline in E, raising the possibility that physiological changes in plasma E play a role in the normal regulation of early postprandial metabolism. The blunting of the postprandial E profile in older men indicates the presence of age-related dysregulation in the adrenomedullary response to meals, the significance of which deserves future investigation.
|
Footnotes |
|---|
Abbreviations: AIRg, Acute insulin response to glucose; BMI, body mass index; DI, glucose disposition index; E, epinephrine; IVGTT, iv glucose tolerance test; NE, norepinephrine; SI, insulin sensitivity.
Received February 25, 2005.
Accepted August 2, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. V. Nedeltcheva, L. Kessler, J. Imperial, and P. D. Penev Exposure to Recurrent Sleep Restriction in the Setting of High Caloric Intake and Physical Inactivity Results in Increased Insulin Resistance and Reduced Glucose Tolerance J. Clin. Endocrinol. Metab., September 1, 2009; 94(9): 3242 - 3250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Melanson, W. T. Donahoo, G. K. Grunwald, and R. Schwartz Changes in 24-h substrate oxidation in older and younger men in response to exercise J Appl Physiol, November 1, 2007; 103(5): 1576 - 1582. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
). Lights were off between 2300 and 0700 h (
). The conversion factor from picograms per milliliter to picomoles per liter is 5.458 for E and 5.911 for NE.
) and older (•) men. The single nighttime and three daytime data points reflect the diurnal variation in these variables in both age groups. The conversion factor from picograms per milliliter to picomoles per liter is 5.458 for E and 5.911 for NE.
