![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
* Department of Medical Sciences, Nutrition and
Department of Medical Sciences, Clinical Chemistry, Uppsala University Hospital, SE-751 85 Uppsala, Sweden;
IPM, Karolinska Institute, SE-171 77 Stockholm, Sweden;
Department of Biosciences, Unit for Preventive Nutrition, Karolinska Institute, SE-141 57 Huddinge, Sweden.
** Swedish Dairy Association, SE-105 46 Stockholm, Sweden; and
Uppsala University Children’s Hospital, SE-751 85 Uppsala, Sweden
2To whom correspondence should be addressed. E-mail: Ulf.Holmback{at}medsci.uu.se.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
20% of the work force in the industrialized world have irregular working hours, it is pertinent to study the consequences of eating at irregular, especially nighttime hours. We studied the postprandial responses during nocturnal fasting vs. eating throughout a 24-h wake period. Seven healthy males were studied twice in a crossover design. After a 6-d diet adjustment period [high fat diet, 45 energy percent (en%) fat, 40 en% carbohydrates)] with sleep from 2300 to 0700 h, the men were kept awake for 24 h at the metabolic ward and given either 6 isoenergetic meals, i.e., every 4 h (N-eat) or 4 isoenergetic meals from 0800 to 2000 h followed by a nocturnal fast (N-fast), with the same 24-h energy intake. Energy expenditure, substrate utilization, activity, heat release, body temperature and blood variables were measured over 24 h. Energy expenditure and blood glucose, triacylglycerol, insulin and glucagon concentrations were lower and nonesterified fatty acids concentrations were higher during the nocturnal fast than during nocturnal eating (P < 0.05); however, no 24-h differences between the protocols were apparent. Nocturnal fasting slightly altered the secretory patterns of the thyroid hormones and cortisol (P < 0.05). We found no clear indication that it would be more favorable to ingest few larger daytime meals than smaller meals throughout the 24-h period. The body seems to be able to buffer small differences in meal size and timing provided energy balance is maintained.
KEY WORDS: • substrate utilization • energy expenditure • postprandial • endocrine variables • circadian
Although studies on what to eat are numerous, when to eat is an issue that has received less attention. Because 20% of the work force in Sweden (1), as well as in most other countries in the industrialized world (2), have irregular working hours, it is pertinent to study the consequences of eating at irregular, especially nighttime hours. The circadian distribution of macronutrient intake has been shown to be affected by two- and three-shift work (3), with as yet not completely identified effects on macronutrient utilization.
We showed recently that macronutrient utilization displays a circadian rhythm that depends on dietary macronutrients (4). Furthermore, responses of insulin, pancreatic polypeptide (PP),2 thyroid stimulating hormone (TSH), free thyroxin (fT4), cortisol and leptin to meal intake were found to differ with respect to time of day (5). We found a decreased evening/nocturnal responsiveness of cortisol and PP to meal intake. This lack of responsiveness might indicate a lack of metabolic adjustment to nocturnal eating and might have health implications for night work (5).
Because the human body is set for activity during the day and recuperation during the night, an eating pattern that corresponds to this activity schedule could be hypothesized to be more physiologic. However, because shift workers remain active throughout the normal sleep period, the body might be able to alter its "metabolic programs" and buffer the altered distribution of energy intake. Previous studies have shown that the triacylglycerol (TAG) concentrations increase from morning to evening/night (4,6,7). Moreover, it has been shown that redistributing most of the energy intake to the night shift increases total and LDL cholesterol (8). The aim of this study was thus to examine whether there might be any short-term benefits of ingesting the total 24-h energy requirement only during the day, compared with distributing the energy intake evenly throughout the 24-h period.
The two protocols tested were six meals throughout the 24-h period (N-eat) or four larger meals from 0800 to 2000 h followed by a nocturnal fast (N-fast). We used a model mimicking the first night on a rotating 2- or 3-shift schedule (9,10). We also wanted to relate metabolic variables, such as energy expenditure and substrate utilization, with endocrine variables, such as insulin, cortisol and thyroid hormones, to better describe the metabolic situation during night work.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Seven men were recruited for the study. Their mean (range) age was 30 y (22–43); wt 79 kg (63–92); and body fat 22 g/100 g (17–31). Three of the subjects participated in an earlier study and some of their data were taken from that study (4). All were in good health as determined by medical history and physical examination; none of the subjects were smokers. They were screened for sleep disturbances, unusual sleep patterns and pathological blood lipid levels; one subject had slightly increased TAG concentrations [2.24 and 1.54 mmol/L, at d 1 and 0800 h on d 7, respectively (mean of the two sessions)]. All subjects gave their written informed consent, and the Ethical Committee of the Faculty of Medicine at Uppsala University approved the study.
Experimental design.
The subjects participated in two 7-d experimental sessions, consuming the same diet in two different eating protocols in a crossover design with a 1-mo washout period between the two sessions. During the sessions, they were followed on an outpatient basis during d 1–6. The procedures were previously described (4). Briefly, body composition was assessed before and during the preexperimental period (11). The subjects wore a wrist activity recorder (Actiwatch, measures movements/min; Cambridge Neurotechnology, Cambridge, UK) and kept a diary of their daily activities and sleep patterns. They were instructed to maintain a normal life but avoid strenuous physical activity. On d 7, the 24-h metabolic study was performed at the metabolic unit. In the morning, an intravenous catheter was inserted on the dorsal side of the left hand and the subjects were dressed in a direct calorimetry suit (12). At 0800 h, the 24-h study began. During the subsequent 24 h, the subjects remained awake. The 24-h study was divided into six 4-h periods. Two different eating protocols were used, i.e., nocturnal eating (N-eat) in which the subjects consumed six isoenergetic meals at 0800, 1200, 1600, 2000, 0000 and 0400 h, and nocturnal fasting (N-fast) in which the same energy content as N-eat was divided into four meals served at 0800, 1200, 1600 and 2000 h; water (200 mL) was given at 0000 h and 0400 h. Blood sampling occurred at 0.5, 1, 2, 3 and 4 h postprandially. At the end of each 4-h period, urine was collected. Activity, heat release and body temperature were measured continuously throughout the 24-h period. The subjects remained seated in a chair throughout the study and no physical activity was allowed.
Diet.
The diet contained 15 energy percent (en%) from protein, 40 en% from carbohydrates (CHO) and 45 en% from fat [see (4) for a more complete description of the diet and how it was prepared].
Procedures.
The rates of O2-consumption and CO2-production during the 24-h study were assessed using an ergospirometer (SensorMedics 2900Z, Yorba Linda, CA). Energy expenditure was calculated according to Schutz (13), and fat and CHO oxidation were calculated according to Jéquier et al. (14). Protein oxidation was calculated from urinary nitrogen excretion (analyzed with the Kjeldahl technique), corrected for changes in the blood urea pool according to Jéquier et al. (14). Heat release and body temperature were measured using the direct calorimetry suit (12). Blood samples were centrifuged (2500 x g for 10 min) and the supernatant was stored at –20°C until analysis. The following substances were analyzed: plasma glucose, glucagon and serum insulin, PP, nonesterified fatty acids (NEFA), TAG, TSH, fT4, total triiodothyronine (tT3) and cortisol [see (4) and (5) for description of blood sample analyses].
Statistics.
Data were analyzed using values from the whole 24-h period with a three-factor repeated-measures ANOVA (RM-ANOVA). Unequal variances were corrected with the Huynh-Feldt correction factor (15). Three factors were used: "protocol" (24-h difference between N-eat and N-fast); "time-of-day" (difference between the six 4-h time periods throughout the 24-h experiment, using combined data from both protocols); and "meal" (difference between the five time points within each 4-h period, using combined data from both protocols). Student’s t test for dependent samples was used when comparing the day period (0800–0000 h) and the night period (0000–0800 h) between the two protocols. Statistical software (SuperANOVA, version 1.11; Abacus Concepts, Berkeley, CA and Statistica 6.0, StatSoft, Tulsa, OK) was used for the analyses. All results are reported as means ± SEM. Significance was accepted at P < 0.05 and P-values <0.07 are reported as tendencies. Data for TAG were analyzed without the subject with high TAG values. His results were consistent with the other subjects in all other variables and were therefore included. The sensitivity of the actimeters varied between the sessions; therefore, the data were changed to proportions of the 24-h mean before statistical analyses.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Energy expenditure did not differ between the protocols during the daytime period (0800–0000 h), but was lower within the nighttime period (0000–0800 h) during the N-fast protocol then during the N-eat protocol (Fig. 1, Tables 1 and 2).
|
|
|
Fat oxidation did not differ between the protocols during the daytime period or the nighttime period. A time-of-day x meal effect occurred during both protocols, probably from lower postprandial fat oxidation after the 0800-h meal compared with the 1600-h meal (Fig. 1, Table 2).
Protein oxidation did not differ between the protocols or show any time-of-day pattern (data not shown).
Activity pattern (proportions of 24-h mean) showed no time-of-day or meal pattern. Heat release did not differ between the protocols and a time-of-day pattern was observed during both protocols (Fig. 2, Table 2). During both protocols, postprandial heat release increased (Table 2). A tendency for a protocol x time-of-day interaction occurred during the day, probably due to a larger postprandial increase in heat release after the 1600- and 2000-h meals during the N-fast protocol than during the N-eat protocol (Fig. 2).
|
Glucose concentration did not differ between the protocols during the daytime period, but was lower during the N-fast than during the N-eat protocol in the nighttime period (Fig. 3, Table 2). During both protocols, lglucose concentrations were lower after the 0800-h meal compared with the 1600- and 2000-h meals (Fig. 3).
|
Nonesterified fatty acid concentration did not differ between the protocols during the daytime period, but was higher during the nighttime period during the N-fast protocol than N-eat (Fig. 3, Table 1). A time-of-day pattern was seen during both protocols, and during both protocols, meal intake decreased NEFA concentration; however, the response to meal intake seemed to be delayed after the 1600- and 2000-h meals with the N-fast protocol (Fig. 3). There was a protocol x time-of-day x meal interaction, most likely due to the increase in NEFA concentration during the first part of the nighttime period during the N-fast protocol, and a lower postprandial response 4 h after the 1200-, 1600- and 2000-h meals during the N-fast protocol, than during the N-eat protocol (Fig. 3, Table 2).
Insulin concentration was higher in the daytime period and lower in the nighttime period during the N-fast protocol than during the N-eat protocol (Fig. 4, Table 1). Meal intake increased insulin concentration, and this increase was affected by both protocol and time of day (Fig. 4, Table 2). A protocol x time-of-day x meal interaction occurred, most likely due to the plateau in the nighttime period during the N-fast protocol (Fig. 3, Table 2).
|
Glucagon concentration did not differ between the protocols during the daytime period but was lower during the nighttime period (Fig 4, Table 1). There was a protocol x time-of-day interaction, probably from higher postprandial concentrations after the 0800 and 1200; and lower concentration during the 0400- to 0800-h period (Fig. 4, Table 2).
Thyroid stimulating hormone concentration did not differ between the protocols during the daytime period or the nighttime period. Both protocols showed a time-of-day pattern with the highest concentrations during the nighttime period (Fig. 5, Table 2). There was a protocol x time-of-day x meal interaction, most likely due to the difference in concentration pattern between the protocols during the nighttime period (Fig. 5, Table 2).
|
Total T3 concentration did not differ between the protocols during the daytime or nighttime period. Both protocols showed a time-of-day pattern with the highest concentrations during the 0000- to 0400-h period (Fig. 5, Table 2).
Cortisol concentration did not differ between the protocols during the daytime period or the nighttime period. Both protocols showed a time-of-day pattern with the highest concentrations during the 0400- to 0800-h period (Fig. 5, Table 2).
To further explore TAG concentration, simple linear regressions were performed with 4-h energy balance and 4-h fat balances against 4-h TAG area under the curve concentration (Fig. 6). Energy balance (r = 0.44, P < 0.001) and fat balance (r = 0.64, P < 0.001) were positively correlated with TAG.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Energy expenditure, activity and substrate utilization.
We find it somewhat surprising that the larger meals during the N-fast protocol did not result in a larger postprandial increase in energy expenditure during the day because the meals contained 50% more energy than the N-eat meals. The theoretical difference in energy expenditure between the N-eat and N-fast meals due to diet-induced thermogenesis (DIT) would be 0.4 MJ (DIT representing 10% of ingested energy); however, we observed a difference between the protocols of about 0.15 MJ. The positive energy balance (15% of ingested energy) observed with both protocols might have negated any smaller differences. The activity factor (PAL 1.4) we used has previously (4) been quite accurate in keeping subjects in energy balance, making the large positive energy balance unexpected.
In light of the increased activity after the 0400-h meal in the previous study (4), we performed a two-factor RM-ANOVA at 0500 h ± 2 h because this was the time at which body temperature was at its lowest. Sopowski et al. (6) showed that the nadir of body temperature coincides with maximum sleepiness. There was a weak tendency (P = 0.065) for activity to be lower during the N-fast protocol. Moreover, the subjects felt more tired during the N-fast protocol (16). These findings contrast with the hypothesis that fidgeting is a strategy to stay awake (4). Perhaps the food is required as a stimulus for activity because Levine et al. (17) observed that some subjects responded to overfeeding by increased fidgeting. Thus it is possible that a rise in glucose concentration, for example, acts as an activity stimulus. The unexpected variance in the sensitivity of the actimeter limits the conclusions that can be drawn from the activity data. However, the energy expenditure data support the activity data.
The N-fast protocol led to a higher carbohydrate oxidation during the feeding period and lower oxidation during the nighttime period, whereas fat oxidation was not affected by protocol. Because carbohydrate oxidation precedes fat oxidation in the oxidative hierarchy (18), the larger meals should lead to transiently higher carbohydrate oxidation. We expected fat oxidation to be higher during the fasting period, but no difference was observed; it is likely that the positive energy balance decreased fat oxidation.
Heat release and body temperature.
We found a larger postprandial increase in heat release after the 1600- and 2000-h meals during the N-fast protocol than during the N-eat protocol. These two peaks were not observed in energy expenditure or body temperature, i.e., the body reacts to larger meals by radiating more heat. The difference in body temperature between the protocols was probably due to a shift in position of the rectal temperature probe in two of the subjects, and not a physiologic phenomenon.
Glucose and pancreas hormones.
As expected, glucose concentration was lower during the fasting period. Perhaps more surprisingly, smaller oscillations were observed with the larger meals during the N-fast protocol, although carbohydrate oxidation was higher at the same time during the N-fast protocol. The more even secretory pattern observed with larger meals should not have been an effect of decreased gastric emptying rate. Energy delivery to the intestine is increased by increasing energy content and meal volume (19), and people adapted to a high fat (HF) diet have an increased gastric emptying rate (20). The insulin pattern, however, was the opposite of the glucose pattern during the N-fast protocol, with higher postprandial peaks during the daytime period during the N-fast protocol. This was most likely a result of the larger meals (21) in which insulin "buffered" the effect of the larger meals on glucose concentration.
Pancreatic polypeptide is a hormone-like peptide, released from the pancreatic islet in a biphasic manner in response to meals (22); it has been hypothesized to be a marker for vagal tone (23). As observed in our previous study (5), the postprandial response of PP decreased from morning to evening. Despite food intake during the N-eat, there was little difference in PP concentration during the night compared with during the N-fast protocol. Possibly, the reduced amplitude of PP at night could be related to the changes in metabolism and appetite observed in night wake (4,24). The highest glucagon concentration was observed during the 0800- to 1200-h period after which the glucagon concentration more or less leveled out, thus corroborating that glucagon is an important gluconeogenic hormone (25).
Triacylglycerols and NEFA.
During the N-fast protocol, postprandial TAG levels were highest at 1700 h, which is similar to other studies (26,27), although Rivera-Coll et al. (27) also found a large postprandial peak at 0325 h. This peak agrees well with the highest postprandial TAG levels observed during the N-eat protocol and in our previous study (4). Moreover, Sopowski et al. (6) observed increased postprandial TAG levels at 2000 h compared with 0800 h. Furthermore, the postprandial concentration levels were higher during the N-fast protocol than during the N-eat protocol. The TAG response could be described as a simple dose-response; when observed over 24 h, it does not matter how the meals are taken. For the postprandial periods after the 0800-, 1200-, 1600- and 2000-h meals, the dose-response scheme seems to be correct with TAG levels 25% higher during the N-fast protocol. However, in this study as well as our previous study (4), the postprandial concentration after the 0400-h meal during the N-eat protocol was almost as high (1.57 mmol/L) as after the meals during the N-fast protocol (1.83 mmol/L), despite a much smaller meal. It has been postulated that 1.5 mmol TAG/L is the threshold for formation of large VLDL particles (28). These large VLDL particles then cause the formation of small dense artherogenic LDL particles (28). It therefore seems to be marginally better to eat smaller meals around the clock than larger meals during the day. However, it has been shown that those shift workers who redistributed most of their energy intake to the night shift had the highest levels of total and LDL cholesterol (8). Whether and how the increased TAG response after the 0400-h meal relates to the increased TAG concentrations observed in shift workers (29,30) remains to be explained.
Nonesterified fatty acid concentration varied less during the N-fast protocol, perhaps as an effect of the larger insulin variations (31) observed during the N-fast protocol. Nonesterified fatty acids have been proposed to be involved in the circadian rhythm of insulin sensitivity (32), presumably via their role in the formation of intramuscular TAG (31). In the study by Morgan et al. (32), fasting NEFA concentration was higher in the evening than in the morning. In this and a previous study (4), the NEFA concentration curve showed a bimodal shape with a maximum during the 1200- to 2000-h period and an additional peak after the 0400-h meal. A similar afternoon peak was observed in a study by Van Gent et al. (33) in which subjects consumed three HF meals between 0900 and 1700 h or eight HF meals between 0900 and 2300 h (33). However, when the subjects consumed eight HF meals throughout the 24-period, the NEFA concentration curve was flattened. Nevertheless, the diurnal variations observed in our studies and those of others, when normal meals have been provided, seem to be smaller (between 0.1 and 0.5 mmol NEFA/L) (32–34) than the 0.7 mmol NEFA/L above fasting levels required to show a decreased insulin sensitivity (31).
Thyroid hormones.
We found some differences between the protocols in TSH levels during the fasting period. However, as Goichot et al. (35) showed, sleep strongly affects TSH. Although the differences between protocols differed significantly, they were much smaller than the effect of sleep (35). The fT4 and tT3 concentrations were not affected by differences in meal intake, except that fT4 concentration tended to be slightly lower during the day during the N-fast protocol. In contrast to Goichot et al. (35), we observed a time-of-day pattern in fT4 and tT3 concentrations. Hirschfeld et al. (36) observed a time-of-day pattern in tT3 but not fT4 concentration. Because our protocols were similar, it is not apparent why the results differed.
Cortisol.
Cortisol concentration did not differ between the protocols; however, the graphs suggest that some small differences might still exist. The concentration pattern differed slightly between the protocols around the 0400-h meal, indicating that the meal intake affects cortisol concentration. As observed in a previous study (5), cortisol concentration was decreased after meals during the day and early morning, but not during the evening. This lack of nocturnal meal feedback might mean that the central drive to increase cortisol concentration during the evening/night is stronger than the potential meal effect. This warrants further study because cortisol affects substrate utilization and lipid storage (37).
"Health" effects.
Shift work has been shown to be associated with a number of conditions such as high TAG concentrations (29) and obesity (38), leading to an increased risk of myocardial infarction (39). Could the type of macronutrient intake affect these metabolic disturbances? The differences between a high carbohydrate diet and a HF diet in a previous study, i.e., higher TAG concentration, lower energy expenditure, increased irritability and sleepiness after consumption of the high carbohydrate diet (4,24) (Lowden et al., unpublished data), might be of concern from a shift work perspective. Is the nocturnal energy intake also an issue? On the one hand, there was no clear advantage of any of the studied protocols regarding TAG concentrations; if anything, small meals around the clock would be preferred to keep the postprandial TAG concentrations low. Moreover, significant positive correlations were observed between 4-h energy and fat balances and 4-h TAG area-under-the curve concentrations. These correlations were due mainly to values from the N-fast protocol, i.e., energy and fat balance have to be substantially different from zero to affect TAG values.
Furthermore, the subjects seemed to be more active if they ate at night. On the other hand, the increased TAG and glucose concentrations toward the night and the decreased responsiveness of cortisol and PP to nighttime meal intake compared with daytime may have health implications. Preliminary "mood" data indicate that nocturnal fasting increases sleepiness ratings more than nocturnal eating (16). However, it has been shown that redistributing most of the energy intake to the night shift increases total and LDL cholesterol (8). One must be cautious when trying to link results from short-term studies to possible long-term effects. However, this study points out that the physiologically sensible strategy might differ from the psychologically sensible strategy.
We found no clear indication that it would be more favorable to ingest few larger meals during the daytime than more frequent smaller meals around the clock. The body seems to be able to buffer small differences in meal size and timing provided energy balance is maintained.
|
FOOTNOTES |
|---|
3 Abbreviations used: CHO, carbohydrate; DIT, diet-induced thermogenesis; en%, energy percent; fT4, free thyroxin; HF, high fat; N-eat, nocturnal eating; NEFA, nonesterified fatty acids; N-fast, nocturnal fasting; PAL, physical activity level; PP, pancreatic polypeptide; RM, repeated measures; TAG, triacylglycerol; TSH, thyroid stimulating hormone; tT3, total triiodothyronine.
Manuscript received 7 May 2003. Initial review completed 3 June 2003. Revision accepted 23 June 2003.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Press release from StatisticsSweden Nr 1998:203 [Swedish]. http://www.scb.se/press/press98/p203.htm (accessed July 16, 2002).
2. Lund, J., Arendt, J., Hampton, S. M., English, J. & Morgan, L. M. (2001) Postprandial hormone and metabolic responses amongst shift workers in Antarctica. J. Endocrinol. 171:557-564.[Abstract]
3. Lennernäs, M., Hambraeus, L. & Åkerstedt, T. (1995) Shift related dietary intake in day and shift workers. Appetite 25:253-265.[Medline]
4. Holmbäck, U., Forslund, A., Forslund, J., Hambraeus, L., Lennernäs, M., Lowden, A., Stridsberg, M. & Åkerstedt, T. (2002) Metabolic responses to nocturnal eating in men are affected by sources of dietary energy. J. Nutr. 132:1892-1899.
5. Holmbäck, U., Forslund, A., Lowden, A., Forslund, J., Åkerstedt, T., Lennernäs, M., Hambraeus, L. & Stridsberg, M. (2003) Endocrine responses to nocturnal eating—possible implications for night work. Eur. J. Nutr. 42:75-83.[Medline]
6. Sopowski, M. J., Hampton, S. M., Ribeiro, D. C., Morgan, L. & Arendt, J. (2001) Postprandial triacylglycerol responses in simulated night and day shift: gender differences. J. Biol. Rhythms 16:272-276.
7. Romon, M., Le Fur, C., Lebel, P., Edme, J. L., Fruchart, J. C. & Dallongeville, J. (1997) Circadian variation of postprandial lipemia. Am. J. Clin. Nutr. 65:934-940.
8. Lennernäs, M., Åkerstedt, T. & Hambraeus, L. (1994) Nocturnal eating and serum cholesterol of three-shift workers. Scand. J. Work Environ. Health 20:401-406.[Medline]
9. Morgan, L., Arendt, J., Owens, D., Folkard, S., Hampton, S., Deacon, S., English, J., Ribeiro, D. & Taylor, K. (1998) Effects of the endogenous clock and sleep time on melatonin, insulin, glucose and lipid metabolism. J. Endocrinol. 157:443-451.[Abstract]
10. Aring;kerstedt, T. & Torsvall, L. (1985) Napping in shift work. Sleep 8:105-109.[Medline]
11. Forslund, A. H., Johansson, A. G., Sjodin, A., Bryding, G., Ljunghall, S. & Hambraeus, L. (1996) Evaluation of modified multicompartment models to calculate body composition in healthy males. Am. J. Clin. Nutr. 63:856-862.
12. Hambraeus, L., Sjodin, A., Webb, P., Forslund, A., Hambraeus, K. & Hambraeus, T. (1994) A suit calorimeter for energy balance studies on humans during heavy exercise. Eur. J. Appl. Physiol. 68:68-73.
13. Schutz, Y. (1995) The basis of direct and indirect calorimetry and their potentials. Diabetes Metab. Rev 11:383-408.[Medline]
14. Jéquier, E., Acheson, K. & Schutz, Y. (1987) Assessment of energy expenditure and fuel utilization in man. Annu. Rev. Nutr. 7:187-208.[Medline]
15. Baguley, T. C. () http://www-staff.lboro.ac.uk/hutsb/Spheric.htm. 2002 (accessed May 28, 2003).
16. Lowden, A., Åkerstedt, T., Holmbäck, U., Forslund, A., Forslund, J. & Lennernäs, M. (2002) Partial food deprivation vs. high fat-meal - effects on performance and wakefulness during the night. J. Sleep Res. 11:140 (abs.).
17. Levine, J. A., Eberhardt, N. L. & Jensen, M. D. (1999) Role of nonexercise activity thermogenesis in resistance to fat gain in humans. Science (Washington, DC) 283:212-214 [see comments].
18. Jebb, S. A., Prentice, A. M., Goldberg, G. R., Murgatroyd, P. R., Black, A. E. & Coward, W. A. (1996) Changes in macronutrient balance during over- and underfeeding assessed by 12-d continuous whole-body calorimetry. Am. J. Clin. Nutr. 64:259-266.
19. Hunt, J. N., Smith, J. L. & Jiang, C. L. (1985) Effect of meal volume and energy density on the gastric emptying of carbohydrates. Gastroenterology 89:1326-1330.[Medline]
20. Castiglione, K. E., Read, N. W. & French, S. J. (2002) Adaptation to high-fat diet accelerates emptying of fat but not carbohydrate test meals in humans. Am. J. Physiol. 282:R366-R371.
21. Service, F. J. (1990) Meal-related insulin requirements. Horm. Metab. Res. Suppl. 24:77-83.[Medline]
22. Slezak, L. A. & Andersen, D. K. (2001) Pancreatic resection: effects on glucose metabolism. World J. Surg. 25:452-460.[Medline]
23. Rigaud, D., Mignon, M., Accary, J. P., Vatier, J., Cantowitz, F. & Bonfils, S. (1988) Pancreatic polypeptide response to insulin in duodenal ulcer.Different levels in accordance with ulcer activity and its response to treatment. Scand. J. Gastroenterol. 23:595-601.[Medline]
24. Lowden, A., Holmbäck, U., Åkerstedt, T., Forslund, A., Forslund, J. & Lennernäs, M. (2003) Time of day and type of food—relation to mood and hunger during 24 hours of constant conditions. J. Hum. Ergol. (in press).
25. Bertin, E., Arner, P., Bolinder, J. & Hagstrom-Toft, E. (2001) Action of glucagon and glucagon-like peptide-1-(7–36) amide on lipolysis in human subcutaneous adipose tissue and skeletal muscle in vivo. J. Clin. Endocrinol. Metab. 86:1229-1234.
26. Mirani-Oostdijk, C. P., Havekes, L., Terpstra, J., Frolich, M., Van Gent, C. M. & Jansen, H. (1983) Diurnal changes in serum triglycerides as related to changes in lipolytic enzymes, (apo) lipoproteins and hormones in normal subjects on a carbohydrate-rich diet. Eur. J. Clin. Investig. 13:301-309.[Medline]
27. Rivera-Coll, A., Fuentes-Arderiu, X. & Diez-Noguera, A. (1994) Circadian rhythmic variations in serum concentrations of clinically important lipids. Clin. Chem. 40:1549-1553.
28. Griffin, B. A. (1999) Lipoprotein atherogenicity: an overview of current mechanisms. Proc. Nutr. Soc. 58:163-169.[Medline]
29. Romon, M., Nuttens, M. C., Fievet, C., Pot, P., Bard, J. M., Furon, D. & Fruchart, J. C. (1992) Increased triglyceride levels in shift workers. Am. J. Med. 93:259-262.[Medline]
30. Knutsson, A. (1989) Relationships between serum triglycerides and gamma-glutamyltransferase among shift and day workers. J. Intern. Med. 226:337-339.[Medline]
31. Boden, G., Lebed, B., Schatz, M., Homko, C. & Lemieux, S. (2001) Effects of acute changes of plasma free fatty acids on intramyocellular fat content and insulin resistance in healthy subjects. Diabetes 50:1612-1617.
32. Morgan, L. M., Aspostolakou, F., Wright, J. & Gama, R. (1999) Diurnal variations in peripheral insulin resistance and plasma non-esterified fatty acid concentrations: a possible link?. Ann. Clin. Biochem. 36:447-450.
33. Van Gent, C. M., Pagano Mirani-Oostdijk, C., van Reine, P. H., Frolich, M., Hessel, L. W. & Terpstra, J. (1979) Influence of meal frequency on diurnal lipid, glucose and insulin levels in normal subjects on a high fat diet; comparison with data obtained on a high carbohydrate diet. Eur. J. Clin. Investig. 9:443-446.[Medline]
34. Terpstra, J., Hessel, L. W., Seepers, J. & Van Gent, C. M. (1978) The influence of meal frequency on diurnal lipid, glucose and cortisol levels in normal subjects. Eur. J. Clin. Investig. 8:61-66.[Medline]
35. Goichot, B., Weibel, L., Chapotot, F., Gronfier, C., Piquard, F. & Brandenberger, G. (1998) Effect of the shift of the sleep-wake cycle on three robust endocrine markers of the circadian clock. Am. J. Physiol. 275:E243-E248.
36. Hirschfeld, U., Moreno-Reyes, R., Akseki, E., L’Hermite-Baleriaux, M., Leproult, R., Copinschi, G. & Van Cauter, E. (1996) Progressive elevation of plasma thyrotropin during adaptation to simulated jet lag: effects of treatment with bright light or zolpidem. J. Clin. Endocrinol. Metab. 81:3270-3277.[Abstract]
37. Rosmond, R. & Björntorp, P. (1998) The interactions between hypothalamic-pituitary-adrenal axis activity, testosterone, insulin-like growth factor I and abdominal obesity with metabolism and blood pressure in men. Int. J. Obes. Relat. Metab. Disord. 22:1184-1196.[Medline]
38. van Amelsvoort, L. G., Schouten, E. G. & Kok, F. J. (1999) Duration of shiftwork related to body mass index and waist to hip ratio. Int. J. Obes. Relat. Metab. Disord. 23:973-978.[Medline]
39. Knutsson, A., Hallquist, J., Reuterwall, C., Theorell, T. & Akerstedt, T. (1999) Shiftwork and myocardial infarction: a case-control study. Occup. Environ. Med. 56:46-50.
This article has been cited by other articles:
|
|
 |
|
  Minerva BMJ, September 20, 2003; 327(7416): 690 - 690. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
