ISOCALORIC DIET CHANGES AND ELECTROENCEPHALOGRAPHIC SLEEP

Saturday 18 October 1975

ISOCALORIC DIET CHANGES AND ELECTROENCEPHALOGRAPHIC SLEEP F. PHILLIPS A. H. CRISP B. MCGUINNESS E. C. KALUCY

C. N. CHEN

J. KOVAL R. S. KALUCY J. H. LACEY

female anorectic patients gaining weight in the way described above initially showed more s.w.s. but, on reaching a stable "normal" weight, this diminished again whilst the amount ofR.E.M. sleep remained greater than before.7 Correspondingly, in normal subjects, episodes of acute starvation have been found89 to be associated with an increase in s.w.s. and a reduction of R.E.M.

Academic Department of Psychiatry, St George’s Hospital Medical School, London SW17

Electroencephalographic (E.E.G.) sleep changes were studied in eight young male healthy subjects who were given a normal balanced diet or a high-carbohydrate/low-fat or low-carbohydrate/high-fat isocaloric diet, according to an experimental design. Significantly less slow-wave sleep (S.W.S.) was found after consuming a high-carbohydrate/low-fat diet than after consuming a normal balanced diet or a low-carbohydrate/high-fat diet. The latter two diets did not differ in terms of the amounts of s.w.s. Both highcarbohydrate/low-fat and low-carbohydrate/high-fat isocaloric diets, especially the former, were associated with significantly more rapid-eye-movement (R.E.M.) sleep than was the normal balanced diet. These findings emphasise the importance of daily diet for the following Summary

night’s sleep. Introduction RELATIONSHIPS between nutritional status and sleep have long been subject to comment in everyday life, yet they have rarely been systematically investigated in man, In 1967 one of us reportedl that patients with primary anorexia nervosa showed major sleep disturbance in the second half of the night, related to nutritional status and unrelated to mood. This observation wasi later confirmed using standardised sleep se1f-reports2! and, in the sleep laboratory, measures of motility3 andl

sleep.

Milk and corn oil introduced into the cat duodenum have a sedative effects When a milk cereal preparation was given before bedtime, human sleep became less restless, especially in the second half of the night." 12 Such studies invite further investigation of the impact of the degree of starvation, weight change, and diet on sleep. We report here a study within which we set out to investigate the effect of isocaloric diets (i.e., diets containing identical numbers of calories but different proportions of basic constituents in terms of carbohydrate and fat) ingested during the preceding day, upon E.E.G. sleep in healthy young males. Our main interest was to examine the effect of carbohydrate on sleep since deprivation of this foodstuff plays such an important role in the metabolic status of many patients with primary anorexia nervosa.1 Methods and Materials The experiment was carried out on eight healthy young adult males of normal and stable weight and stable dietary habits. Subjects took part in the experiment in pairs. Each pair had two courses of dietary manipulation, each lasting 4 days separated by a 2-week interval. During the 4 days the diet was consumed at fixed times, and the men reported to the sleep laboratory at 9.30 P.M. All-night sleep E.E.G.s were recorded and scored blind by an independent person, according to the standard technique." On the first 2 days of each 4-day period, the men ate a normal balanced diet (X) containing 350 g of carbohydrate, 140 g of fat, and 75 g of protein. During the next 2 days one of the pair consumed a high-carbohydrate/lowfat isocaloric diet (Y) containing 600 g of carbohydrate, 33 g of fat, and 75 g of protein. The other man consumed a low-car-

all-night electroencephalographic (E.E.G.) recordings,2 4 in studies of anorectic patients whose weight was being restored to normal levels while eating "normal" diets. Obese patients losing weight, on low-calorie diets, have also been studied,56 and have provided further confirmation of the link. In general the sleep changes in such states of chronic starvation were characterised by reduced duration of s:eep, especially in the second half of the night, more broken sleep, less slow-wave sleep (s.w.s.) and rapid-eye"iovement

tances 7938

(R.E.M.) sleep and earlier waking. Such disimproved with weight-gain. In particular,

TABLE I-EXPERIMENTAL DESIGN .

X = normal diet Y Z

=

=

High-carbohydrate/low-fat diet, Low-carbohydrate/high-fat diet

JI

724

bohydrate/high-fat

isocaloric diet (Z) containing 100 g of carbohydrate, 255 g of fat, and 75 g of protein. The experimental diets (Y and Z) were exchanged in the last 2 days of the 4-day period in the second course. The experimental design is displayed in table i. The design is unusual in that three treatments are introduced with only the two experimental diets balanced for order. Comparisons between the effects of these two experimental diets were made by a three-factor analysis of variance for a cross-over design in which effects of treatment, periods, and days were extracted from the within-subject variation; and order effect, or treatment x period interaction, from the between-subject variation. Since in all instances, except where otherwise indicated, all temporal effects (i.e., periods, days, and their interactions), were non-significant, it has also been possible to compare the effects of the normal diet with the other two in a single-factor repeated-measure analysis of variance. 14 However, since carry-over and adaptation effects in these latter comparisons cannot be eliminated, the interpretation of these results involving the normal diet is necessarily more

conservative. I

7B

Results

S.W.S. and R.E.M.

sleep

The means of the E.E.G. sleep measures, examined in relation to diet, are summarised in table n and the

figure.

percentage of total to three diets.

as a

&.

sleep-time in relation

whilst they were consuming the balanced isocaloric diet. A lesser but significant increase was also found at the time of consumption of the low-carbohydrate/highfat diet (p<0.05). Moreover, a significant difference in was observed in the amount of R.E.M. sleep (p
Slow-wave Sleep (Stages 3 and 4)

(both in terms of total time in minutes and as total percent sleep-time) was significantly less (p<0-03 in subjects consuming a high-carbohydrate/low-fat die) than when they were consuming a low-carbohydrate, high-fat diet or the balanced isocaloric diet. Comparisor of s.w.s. after a low-carbohydrate/high-fat diet and th( balanced isocaloric diet, however, showed no significant difference. s.w.s.

R.E.M.

Sleep R.E.M. (both in terms of total time in minutes and as per it cent total sleep) was significantly increased (P<0-01) subjects when they were consuming the high-carbohydrate/low-fat diet when compared with the amount pre TABLE II-E.E.G. SLEEP CHARACTERISTICS IN RELATION TO THE THREE DIETS

Other Sleep Stages

There were no significant differences in stage-2 sleep between any of the dietary regimens. However, there was significant reduction of stage-1 with both experimental diets (p< 0-03).

Sleep Latency and Total Sleep-time The mean time taken to fall asleep was shortened, and the total sleep-time was lengthened, with both experimental diets. These differences, however, were not sta-

tistically significant. Discussion

Certain sleep E.E.G. characteristics seem to be associated with and, perhaps, dependent upon the previous day’s diet. These associations have emerged despite design problems with which any such study is fraught and to which ours is no exception. The introduction of a balanced isocaloric diet presented some degree of dietary upheaval for our volunteers although this was less of a problem than it might have been had we used females. It is also difficult to take account of adaptation. Our subjects were allowed to adapt in the usual manner and seemed to do so well. However, the reduction of light sleep found in association with both the high-carbohydrate/low-fat and low-carbohydrate/high-fat diets, and possibly also the increased amount of R.E.M. sleep which occurred on these occasions, may have been partly due to adaptation effects within the laboratory arising separateh within each half of the experiment. Nevertheless some important findings appear to have arisen which cannot be accounted for in such ways. For instance there is a specific association between carbohydrate and fat contents in the diet and s.w.s. : high-carbohydrate/loBB-!’a:

725 POSITIVE END-EXPIRATORY PRESSURE IN WEANING PATIENTS FROM CONTROLLED VENTILATION A Prospective Randomised Trial

jletis associated with decreased s.w.s. The link between and probable dependence of growth-hormone secretion on s,w.s. is well documented. 11-18 Acute starvation has been shown to enhance

s.w.s.,9 growth hormone,19

or

both,’ and insulin-induced hypoglycaemia stimulates growth-hormone secretion. 20 Others21have reported that

RICHARD SAUMAREZ T. GORDON MCNABB JOHN J. SKILLMAN

THOMAS W. FEELEY JAMES M. KLICK

arginine-induced growth-hormone secretion is significantly suppressed by a high carbohydrate content or a high carbohydrate/protein ratio in the diet; such

Departments of Anesthesia and Surgery of the Harvard Medical School and the Beth Israel Hospital, Boston,

changes, however, are unrelated to differences in caloric intake, plasma free-fatty-acids (F.F.A.) or glucose.22 We found

no

increase in

s.w.s.

after

a

low-carbohydrate/

high-fat diet, and this may seem inconsistent with some of the work previously mentioned. Nevertheless it may also indicate that a longer time latency is required for certain mechanisms to operate in the case of a low carbohydrate intake. In the study by Karacan et al.8, even with total starvation, changes of s.w.s. did not occur until 60-67 h after fasting. It would seem therefore that carbohydrate diet is closely related to both s.w.s. and more general growthhormone secretion. Thus acute starvation (3 or 4 days) enhances s.w.s., probably with an associated increase in growth-hormone secretion. Conversely a high carbohydrate diet suppresses growth-hormone secretion and, as shown in our experiment, appears also to reduce s.w,s.

We have found R.E.M. sleep to be especially significantly increased by the high-carbohydrate/low-fat diet. The inference that carbohydrate content in the diet has a specific effect on changes in R.E.M. activity can there fore be made with confidence, despite the possible adaptation effects referred to earlier. The lesser increase of R,E.M. sleep associated with the low-carbohydrate/highfat diet presents some problems in interpretation. Adaptation effects apart, it may be that a high-fat content in the diet also enhances R.E.M. sleep. This notion is supported by reports that plasma-F.F.A. may increase free plasma-tryptophan by competition at their binding site,23 and that free plasma-tryptophan correlates positively with the amount of R.E.M. sleep among normal in-

dividuals.24 Oswald has suggested that s.w.s. may be associated with widespread bodily restorative functions whilst R,E.M. sleep may be more associated with synthetic processes of brain reorganisation and repair.25 Patients with primary anorexia nervosa show reduced R.E.M. sleep in association with their generally impaired sleep in the second half of the night. After refeeding, the amount of R,E.M. sleep is greatly increased while an initial increase in s.w.s. is subsequently reversed.7 This divergence between s.w.s. and R.E.M. changes occurred after the subjects had reached within 15% of their matched population mean weight and may have arisen in relation to the pubertal endocrine changes which arise 26 at around this time and which presumably involve major cerebral activity. As suggested earlier, further studies of the relationship of degrees of starvation, weight change, and specific dietary constituents are required to clarify further the relationship between these factors and sleep and associated restorative processes. Meanwhile we suggest to those engaged in sleep research that they take into account the dietary habits and nutritional status of their subjects. RpnnMt
fnr ffnrints should he addressed

tn

C.N.C*.

Massachusetts 02215, U.S.A.

Summary 5

Twenty-five patients in acute respiratory failure were randomised to receive either

of

positive end-expiratory pressure (P.E.E.P.) or while weaning from controlled ventilation. The use of P.E.E.P. resulted in a significant reduction in the increase in alveolar-arterial oxygen tension gradient (AaDO21) which occurred in the group of patients who cm

no-P.E.E.P.

converted from controlled ventilation to sponventilation without P.E.E.P. Patients who weaned without P.E.E.P. had a mean increase in AaDO21 of 102±35 mm Hg S.E. while those who weaned with P.E.E.P. had a mean increase of only 10±22 mm Hg (P<0·03). The use of P.E.E.P. was also associated with a significant improvement in the vital capacity and the maximum inspiratory force. Patients who weaned with P.E.E.P. had an increase in vital capacity of 258±108 ml (P<0·05) and an increase in inspiratory force of -15±5 cm H2O (P<0·01), while patients who weaned without P.E.E.P. did not have significant changes in these measurements. The use of P.E.E.P. during weaning may be helpful in patients who fail to wean because of the development of hypoxæmia due to rapid alveolar collapse, since P.E.E.P. appears to minimise the increase in intrapulmonary right-to-left shunt which normally occurs during weaning from controlled ventilation.

were

taneous

MISS PHILLIPS AND OTHERS: REFERENCES

1. Crisp; A. H. J. psychosom. Res. 1967, 11, 117. 2. Crisp, A. H., Stonehill, E., Fenton, G. W. Postgrad. med. J. 1971, 47, 207. 3. Crisp, A. H., Stonehill, E. J. psychosom. Res. 1971, 15, 501. 4. Lacey, J. H., Crisp, A. H., Kalucy, R. S., Hartmann, M., Chen, C. Postgrad. med. J. (in the press). 5. Crisp, A. H., Stonehill, E. J. psychosom. Res. 1970, 14, 353. 6. Crisp, A. H., Stonehill, E., Fenton, G. W., Fenwick, P. B. C. Psychother. Psychosom. 1973, 22, 159. 7. Lacey, J. H., Crisp, A. H., Kalucy, R. S., Hartmann, M. K., Chen, C. Br. med. J. (in the press). 8. Karacan, I., Rosenbloom, A. C., Londono, J. H., Salis, P. J., Thornby, J. I., Williams, R. L. Psychosomatics, 1973, 14, 33. 9. MacFadyen, U. M., Oswald, I., Lewis, S. A. J. appl. Physiol. 1973, 35, 391. 10. Fara, J. W., Rubenstein, E. H., Sonnenschein, R. R. Science, 1969, 166, 110. 11. Southwell, P. R., Evans, C. R., Hunt, J. N. Br. med. J. 1972, ii, 4. 12. Brezinova, V., Oswald, I. ibid. ii, p. 431. 13. Rechtschaffen, A., Kales, A. Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. U.S. Government Printing Office, 1968. 14. Winer, B. J. Statistical Principles in Experimental Design. New York, 1970. 15. Takahashi, Y., Kipnis, D. M., Daughaday, W. H. J. clin. Invest. 1968, 47, 2079. 16. Parker, D. G., Sassin, J. F., Mace, J. W., Gotlin, R. W., Rossman, L. G. J. clin. Endoc. Metab. 1969, 29, 871. 17. Sassin, J. F., Parker, D. C., Johnson, L. C., Rossman, L. G., Mace, J. W., Gotlin, R. W. Life Sci. 1969, 8, 1299. 18. Sassin, J. F., Parker, D. C., Mace, J. W., Gotlin, R. W., Johnson, L. C., Rossman, L. G. Science, 1969, 165, 513. 19. Merimee, T. J., Fineberg, S. E. J. clin. Endocr. Metab. 1974, 39, 385. 20. Roth, J., Glick, S. M., Yalow, R. S., Berson, S. A. Science, 1963, 140, 8987. 21. Merimee, T. J., Fineberg, S. E. Lancet, 1973, ii, 120. 22. Merimee, T. J., Rabin, D. Metabolism, 1973, 22, 1235. 23. Curzon, G., Knott, P. J. in Aromatic Amino Acids in Brain, (Ciba Fndn. Symp. new ser. 22); p. 217. Amsterdam, 1974. 24. Chen, C. N., Kalucy, R. S., Hartmann, M. K., Lacey, J. H., Crisp, A. H., Bailey, J. E., Eccleston, E. G., Coppen, A. Br. med. J. 1974, iv, 564. 25. Oswald, I. New Sci. 1970, 46, 170. 26. Palmer, R. L., Crisp, A. H., Mackinnon, P. C. B., Franklin, M., Bonnar, J., Wheeler, M. Br. med. J. 1975, i, 179.