Dietary Fat Level as Affecting Running Performance and Other Performance-Related Parameters of Rats Restricted or Non-Restricted in Food Intake DIRK TOLLENAAR Nutrition Group, Food Sciences Laboratory, U.S. Army Natick Development Center, Natick, Massachusetts 01760 * ABSTRACT The effect of food energy density on certain physical per formance characteristics of rats was studied during ad libitum and weightrestricted feeding. After a 16-week treadmill training period, 48 young adult male Long-Evans rats were divided into six groups, receiving 20%, 40% or 70% of energy as dietary fat and fed ad libitum or 40% by weight of average ad libitum intake. They were run to exhaustion once a week. Increased dietary fat level was not correlated with superior running per formance at either food intake level, although during restricted feeding the high fat group received 40% more energy than the low fat group. Lower body weight was correlated with increased running performance at both food intake levels. The rate of performance increase with decreasing body weight during food restriction was significantly higher for the low fat than for the high fat group. Water consumption was considerably lower during restricted than during ad libitum feeding. Plasma glucose was lower when the 70% fat diet was fed than when compared to the other fat diets at both levels of food intake, lower from day 8 on of restricted than during ad libitum food intake, and lower on day 15 than on day 8 of the restricted intake. J. Nutr. 106: 1539-1546, 1976. exercise INDEXING KEY WORDS fat food restriction blood glucose •water consumption body weight During ad libitum consumption, highfat, low carbohydrate food has repeatedly been shown in the past to reduce exercise endurance, efficiency and physical wellbeing ( 1-9 ), as well as voluntary activity of rats (10), in comparison with a highcarbohydrate or "normal" diet. In turn, a high-carbohydrate diet, consumed ad lib itum, has been found to increase exercise endurance, compared to a "normal" diet (2, 6,7, 11-13). These effects of diet on exercise endurance have been observed in humans (1-3, 5-9, 12, 13), dogs (4), rats and mice (H). Hollmann et al. (13) have reported an average increase of 61% in ergometer cycling duration during con
sumption of an all-carbohydrate as com pared to a "normal" diet. Deuel, Scheer et al. (14, 15), on the other hand, reported that rats swam longer when fed a 20% fat diet as compared to a 5% fat diet. However, the weights attached to the rats during the course of the swimming period had the effect of favoring the heavier rats from predominantly the higher dietary fat group. Further, swimming times averaged only 13 to 20 minutes. With a restricted intake, 500 to 580 Received for publication August 8, 1975. 1This paper reports research undertaken at the U.S. Army Natick Development Center and has been assigned no. TP 1705-FSL in the series of papers ap proved for publication.
1539 This
One
14EX-QLN-61Q7
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1540
DIRK TOLLENAAR
kcal/day mostly from carbohydrate, no significant decrease in exercise endurance of human subjects occurred during a 10 to 12-day period (16, 17). However, during an average intake by human subjects of 1,500 or 1,800 kcal/day from low-carbohy drate diets, in one case (3) coupled with fairly heavy, long-term exercise, lack of energy, rapid incapacitation, and excessive tiredness within days have been observed (3, 18, 19). When consuming about onethird of the ad libitum energy intake, no significant improvement in swimming en durance of rats was found with increasing the percentage of dietary fat (15). When consuming about two-thirds of the ad libitum energy intake, a higher average swimming endurance was observed while feeding a 40% fat diet compared to feed ing less fat in the diet. However, the use of weights attached to the rats during the swimming period tended to favor those from the high fat group, as mentioned above. Further, swimming times were only about 5 to 20 minutes. Especially after consuming a high fat diet, a decrease in blood glucose levels close to the point of exhaustion, after several hours of moderate or fairly heavy exercise, has been reported in human sub jects (2, 6-8, 20). Under similar condi tions, Costili (21) did not observe this, al though no information on diet is given. Such a drop in blood glucose can be a factor limiting exercise endurance (7, 8). During fasting and semistarvation, an initial decrease in blood glucose level dur ing the first few days followed by a gradual return to normal has frequently been found in humans and most animals, when the preceding diet was of average composition (22-24). Blood glucose levels in men fed 1,000 kcal/day fell most for those consuming a very low carbohydrate diet and least for the group consuming the most sugar (23). There exists a need to know how physi cal performance, especially exercise en durance, is affected by dietary fat content and level of food intake. The present study reports the effects of changes in fat con tent of the food on certain physical per formance parameters, especially running endurance, in rats, restricted in food intake and fed ad libitum.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
MATERIALS AND METHODS
Young adult male Long-Evans rats2 were housed individually in stainless steel cages at 21 ±1° (mean ±so) and 50 ±3% relative humidity. When received, they were 60 days old and weighed 245 ± 14 g. For the first 4 months after arrival, they were fed a stock diet.3 For the next 3 weeks they were fed a semipurified diet containing 40% of the energy from fat, as described below. Following this, the ac tual experimental phase was started. The rats were divided into 6 groups and fed either ad libitum or restricted, the diet containing 20%, 40% or 70% of the energy from fat. The restricted food intake was a uniform weight for all groups, namely 40% by weight of .the average ad libitum intake of all groups before the start of the different diet regimen. The composition, in percent, of the diet con taining 40% of the energy from fat was: vitamin-free casein,4 20.0; vegetable short ening,-'"' 10.45; lard,9 10.45; corn starch, 30.3; dextrin, 10.1; dextrose, 10.1; cellu lose, 4.0; mineral mixture,7 4.2; vitamin mixture,8 0.4. The diets with 20% and 70% energy from fat contained, respectively, 4.55% and 23.3% of, each, vegetable shortening and lard, and were prepared by interchange with the amount of corn starch, keeping the weight percentage of the other ingredients constant. The diets were stored refrigerated. The treadmill, used to run the rats, was a slight modification of a design used by Kimeldorf (25) and employed a motor' Obtained from Simonsen Labs., Inc., Gllroy, Cali fornia 95020. 3 Purina Laboratory Chow, Ralston Pnrina Co., St. Louis, Missouri. «All diet ingredients, unless stated otherwise, were obtained from ICN, Life Sciences Group, Cleveland, Ohio 44128. 5 Commercial, general purpose, high stability (100 AOM) shortening. « Armour Co., Chicago, Illinois. 7 The mineral mixture was composed as follows (%) : CaCO3, 15.8; CaHPO,, 35.8; KC1, 16.5; Na-HPO,, 17.2; MgSO4-7H2O> 12.4; MnSO,H2O, 0.44; ZnCOs, 0.11; CuS04-5H2O, 0.19; ferric citrate, 1.46; KIO3, 0.005; Cr( III »-acetate, 0.05. All inorganic salts were "certified" reagent grade, unless stated otherwise, and obtained from Fisher Scientific Co., Pittsburgh. Pennsylvania 15219. Ferric citrate was U.S.P. VIII grade. 8 The vitamin mixture, per kg of diet, contained (in mg, unless stated otherwise) : retinyl palmita te, G.OOOI.U. ; ergocalciferol. 3,000 I.D. ; or-tocopherol, 80 ; menadione, 0.30 ; thiamin-HCl, 6.0 ; riboflavin, 5.0; pyridoxine-HCl, 3.0; niacin, 30; Ca-pantothenate, 1C ; choline dihydrogen citrate, 3.5 g ; vita min Bjn, 0.056 ; ascorbic acid, 60 ; inositol, 17 ; p-aminobenzoic acid, 17; folie acid, 2.0; biotÃ-n, 0.1.
FAT LEVEL, CALORIC INTAKE AND PERFORMANCE
1541
based on the performance of the respec tive groups in the preliminary period as shown in table 1. The adjustment factor for running time for each group was de rived by dividing the overall mean time (41.8 minutes) by the preexperimental running time of the respective group. Sim ilarly, an adjustment factor for treadmill running performance was derived for each exhaustion. These runs took place at a group from the overall mean ( 102.8 kgm ) speed of 22.3 meters/minute (mpm) on and the preliminary period group mean. an incline of 12.5°.Treadmill running per Table 1 presents the adjusted mean run formance values in kilogram meters (kgm) ning time and treadmill running per were obtained by multiplying individual formance. body weights by total elevation during Under the conditions used, fat level had each run. Food was removed 1 hour be no significant effect on running times or fore running. Of 72 rats received, 48 were kgm performance for either the rats fed selected for the experimental phase. These ad libitum or restricted. On day 8 and 15 rats were randomly divided equally be of the test period, the restricted fed rats tween the 6 groups. ran to Stu dent'ssignificantly unpaired tlonger test ofaccording the means, and Heparinized blood samples were col lected during exsanguination, within 2 had a significantly higher kgm perfor mance according to the Mann-Whitney minutes after exhaustion. Deproteinization of the plasma was effected with 6% nonparametric U test than the rats fed ad ZnSO4-7H2O and approximately 0.3 N libitum. On day 8 and 15, average running NaOH (titrated against the ZnSO4 solu time increases were for the 20% fat group tion), as a modification of the method of 32% and 63%, for the 40%, fat group 33% Somogyi (26). Plasma glucose levels were and 63%, and for the 70% fat group 12% determined employing an automated mi and 17%, respectively. Corresponding crotechnique 9 involving a modification of kgm performance increases were for the the ferri-ferrocyanide oxidation-reduction 20% fat group 22% and 43%, for the 40% method according to Hoffman (27). Since fat group 18% and 37%, and for the 70% heparin has been found to raise recorded fat group 2% and 4%, respectively. glucose values (28), standards were like If the performance values of each rat wise heparinized. are averaged for the four running events Statistical analyses of the data for the to exhaustion prior to the experimental tables were performed using the means for phase, the following linear relationship to body weight is obtained: Y = 0.161X + each group and period, as presented, 126.2, in which X = body weight, g (range: rather than individual data. These means were analyzed first according to Student's 428-620), and Y = running time, minutes (range: 27-67), according to a regression t test (29). If this test yielded no signifi cant difference, a Mann-Whitney nonparaanalysis (F ratio = 51.8, P < 0.01, and metric U test (a ranking test) (30) was higher than for a 2nd order equation) also performed. In the cases, indicated, (29). If performance is expressed in kgm, where regression analyses and analyses of likewise, a linear relationship to body covariance were performed, the method weight is found under these conditions: outlined by Steel and Torrie (29) was Y = -0.184X + 195.6, in which Y = kgm performance, range: 74-134 (F ratio = followed. 13.2, P < 0.01). Average kgm performance RESULTS was plotted against average body weight separately for the ad libitum and re Running performance. To facilitate com stricted-fed groups for day 8 and 15 of the parisons of running performance of the different dietary groups throughout the test period and the 20% and 70% fat experimental periods, the experimental •Technlcon Auto Analyzer Methodology N-9, Techobservations were adjusted by factors nicon Instrument Corp., Chauncey, New York. driven, endless belt with a number of ventilated cages, open at the bottom just above the belt. After an initial training period prior to this, the rats were run to exhaustion once a week during the 7 weeks preceding the actual experimental phase. One day after the commencement of the experimental phase the and rat's withwere one run week intervals thereafter, to
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1542
DIRK TOLLENAAR TABLE l Effect of dietary fat level and food intake on running performance of male, young adult rats fed1Dietary fedmin(kgm)'min(kgm)min(kgm)min(kgm)min(kgm)minminmin(kgm)20%41. libitum
Days of experimental periodPreexperimen
energy from fat 20%42.7(104.0)0.979(0.988)38.7(95.0)38.7(94.8)38.7(95.4)38.738.738.7(98.5)48.7(111.4)0.858(0.9 40% 70%
84(lOO.O)41.000(1.028)41.1*(99.5 tal2Multiplicationfactors61815223643Ad
)439.4(95.9)36.3(88.3)40.143.040.3(99.5)Restricted
140% by weight of average ad libitum intake of all groups before start of different diet regimen. 1Average of four runs preceding the experimental period; diet containing 40% calories from fat. 3 Kilogrammeters. 4 Mean values. N = 7 or 8. 6To equalize pre-experimental 'base' values ; the values shown throughout the experimental period have been corrected by these factors for each group. On each of days 8 and 15, the restricted fed rats, as a group, ran longer (P < 0.02) than the rats fed ad libitum. The so of the running times on day 8 for the ad libitum fed group = 4.37, for the restricted fed group = 0.38 ; for day 15 these values are 2.45 and 7.85, respectively.
levels. To better compare the slopes of the lines for the 20% and 70% fat levels, average lines for days 8 and 15 for each fat level were calculated by use of the ap propriate multiplication factors. Thus, re gression equations of Y = -0.439X + 309.9 for 20% dietary fat energy, and Y = -0.051X + 126.6 for 70% dietary fat energy were found, using the individual values. ( For both cases, range of X > 426-534 g, and fits were better for a linear than for an exponential relationship.) The
slope for the 20% fat values was different (P< 0.005) from that for the 70% fat values, according to an analysis of covariance, using randomly paired individual values. The slopes for the 40% fat values were between those of the others. Body weights. Mean (unadjusted) body weights and the percent changes from the values just preceding the experimental period for selected dates of the experimen tal period are given in table 2. Water consumption. Mean water con-
TABLE 2 Effect of dietary fat level and food intake on mean body weights and percent change1
fed20%491» libitum Days of experimental period8
%*43 IAd
fed2energy 70%468
from fat 20%481
(-11.0) (-8.1) (-2.0) (-0.7) (+0.6) 457 467 470 455 491 505 (-11.0) (-1.5) (-13.2)40%493(-15.9)70%465(-9.4) (-0.8) (+0-7) 472 498 522 (+0.6)40%503(+4.0)Dietary(-1.2)Restricted
1Percent change from the average weight of groups during the 2 week period preceding the experimental period. 240% of the average ad libitum intake by weight of all groups. 3N = 7 or 8. Mean so = 9.3% (range: 5.0-13.3%) for the individual values within each ad libitum fed diet group.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
FAT LEVEL,
CALORIC
INTAKE AND PERFORMANCE
1543
TABLE 3 Effect of dietary fat level and food intake on mean water consumption1 •*•* fedml/dayml/dayml/dayml/dayml/dayml/day20%33.4' libitum fedDietary fat,70% energy from 40%29.3(1.070)28.527.430.829.729.429.9 "' 20%
experimental periodPre-experimental (Multiplicationfactor1)49111623Ad
(1.016)21.919.519.7 (0.951)19.319.819.8470%30.8 (1.046)19.217.818.232.9 (0.939)30.9531.131.328.432.840%31.5 (0.993)32.932.932.130.232.2Restricted
1 Values adjusted for differences between groups appearing before start of experimental phase. 2Over all days measured and all fat levels, water consumption was significantly higher for the ad libitum fed than for the restricted fed rats (P < 0.01). *Over all days measured, water consumption for the ad libitum fed rats was significantly lower in the 70% dietary energy from fat group than in the 20% or 40% groups (P < 0.05). *40% by weight of average ad libitum intake of all groups. * Mean values. N for all groups = 7 or 8. Mean so = 4.3 (range: 1.5-6.5) % of the mean values for each diet group.
on the average 9% less, according to Student's t test, than for the other two groups. From day 4, water consumption of rats fed at all fat levels was lower during the restricted feeding than feeding, according to during Student'sad libitum i test. These reductions between day 4 and 11 (not measured beyond this time) fpr the 20% and 40% fat groups were on the av erage 40%. For the group fed 10% of its
sumption of the rats during selected dates of the experimental period is given in table 3. Values obtained for each group during the experimental period were ad justed for differences between groups ap pearing prior to initiation of the experi mental period. During ad libitum feeding, no significant difference existed between the 2Qr/( and 40% fat groups. From day 2 to 23 of ad libitum feeding, the water con sumption of rats fed the high fat diet was
TABLE 4 Effect of dietary fat level and food intake on plasma glucose values1-1-'-4-6 fed20%101.0' libitum fedeDietary ofexperimental Days energy from fat 20%105.2(0.978)111.5101.697.099.1103.8 40% 70%
periodPre-experimental mlmg/100 (Multiplicationfactor2)181522mg/100
(1.031)75.172.051.9 (0.999)113.189.677.570%99.8 )63.574.280.968.1104.7 (0.991 (0.983)119.393.484.040%103.0 (1.019)109.47102.8101.095.7Restricted
mlmg/100 mlmg/100 mlmg/100 mlAd 1Blood was taken within 2 minutes after rats were run to exhaustion. *Values adjusted for differences between groups appearing before start of experimental phase. 3Over all days measured, plasma glucose at both food intake levels were lower in the 70% dietary energy from fat group than in the 20% or 40% groups (P < 0.05), and for the restricted fed rats was also lower in the 40% dietary energy from fat group than in the 20% group (P < 0.05). 4On day 8 and 15 plasma glucose was lower during restricted than during ad libitum feeding (P < 0.02). 6 For the restricted fed rats, plasma glucose was lower on day 15 than on day 8 (P < 0.05). ' 40% of the average ad libitum intake by weight. ' Mean values. N = 7 or 8. Depending on the kind of comparison made above, so of the means ranged from 2.0% to 23.9% of the values.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1.-.44
DIRK TOLLENAAR
energy from fat, this figure was 2,9% and significantly different (P < 0.05) from the reduction for the other groups. No signifi cant differences in water consumption oc curred within the restricted feeding regi men between groups. Plasma glucose levels. Glucose levels in plasma obtained within 2 minutes after the rats were run to exhaustion are given in table 4, These values were adjusted to the same average value (102.9 mg/100 ml), obtained just prior to the initiation of the experimental phase. During ad libi tum feeding, the plasma according to Student's pairedglucose / test levels, of the means, were significantly lower for the 70% fat group than for the 2JO%and 40% fat groups over the whole period mea sured. Further, on day 8 and 15 of the experimental period, the means for each fat level were lower during restricted than during ad libitum feeding. During the re stricted intake feeding period, plasma glu cose levels were lower for the 70% fat group than for the 40% fat group; and for the latter, in turn, lower than for the 20% fat group. Finally, plasma glucose levels on day 15 of the restricted feeding period were lower than on day 8. DISCUSSION On the average, no significant weight loss occurred among the ad libitum fed rats during the experimental period. Thus, the exercise conditions are unlikely to have played any significant role in the weight loss of the restricted fed group. The restricted intake group fed 20% of its dietary energy from fat ran on day 15 of the experimental period on the average 63% longer, with a 43% higher perfor mance in kgm, for a 12.4% weight loss, as compared to the ad libitum fed 20% fat group. These figures for the 40% and 70% fat groups were, respectively: 63, 37, 16.6 and 17, 4, 9.6. Thus, a 9.6% weight loss, assuming a linear relationship, for the restricted fed 20%, 40%, and 70% fat groups corresponded with, respectively, a 49%, 36%, and 17% increase in running time, and a 33%, 21%, and 4% increase, respectively, in kgm performance. This difference in rate of increase in kgm per formance between the 20% and 70% fat groups was significant (P < 0.005), ac
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
cording to an analysis of covariance, using individual values. During the period when 40% dietary energy from fat, fed ad lib itum, was provided, rats with a 9.6% lower weight in the same range, on the average, had a 19% longer running time and a 9% higher kgm performance. A correlation coefficient, r = —0.94existed in this latter case between body weight and running time, averaged for the future groups and four preliminary events, mentioned. Based on the equation obtained for the relationship between body weight and kgm performance for the period before the start of the experimental period, the equalization of the average preliminary running times and kgm performance intro duced an average error into the experi mental results given of not more than about 1.4% of the values, the largest error being about 3.3% (for the 20% fat, ad libitum fed group). This does not affect the relative significance of the results. Lower body fat may have been an im portant factor in the higher running per formance with lower body weight during ad libitum feeding and in, at least, the early stages of restricted feeding. An ad ditional factor promoting increased run ning performance with decreased body weight during restricted feeding might be an increased adrenalin secretion (31) as a result of the stress imposed by the re stricted food intake. Samuels et al. (32) have reported that running endurance of adult male rats was higher from the 9th day of fasting on, as compared with ihe 2nd to 4th day. No information on the magnitude of a training effect is provided by these authors, however, exhaustion tests having been initiated only aftsr the start of fasting. Tollenaar (33) has re ported a significant increase in running endurance of rats on the 7th, compared to the 3rd day of a restricted food intake, as well as compared to running times while fed ad libitum. The higher amount of food energy offered with the 70% fat diet than with the 20% fat diet during the restricted in take provided no advantage in terms of running time or kgm performance. During the ad libitum feeding, a tendency seemed to exist for the rats fed the diet with 70% energy from fat to initially show a better
7 FAT LEVEL, CALORIC INTAKE AND PERFORMANCE
running performance than when fed the other diets. This tendency diminished or reversed itself with time, however. The contribution from carbohydrate varied between 17% and 61% of the energy intake (25%-62% by weight) in the present experiment. The lower amount of solutes presented during restricted as compared to ad lib itum feeding no doubt was the main, if not the sole, cause of the considerably lower water consumption in the former case. During the combustion of fat more water is formed per gram than during the com bustion of carbohydrate. A 9% lower water consumption during ad libitum feeding was observed for the 70% fat than for the other two groups. These dif ferences are largely in agreement with those observed earlier (34). The lower plasma glucose levels with the high fat than with the lower dietary fat levels, both during ad libitum and re stricted feeding, as observed in the rat experiment described above, have, like wise, been reported in humans (7, 23). The plasma glucose levels, observed in the rats right after exhaustion, were lower during the restricted than during the ad libitum feeding regimen, and were lower on the 15th than on the 8th day of re stricted feeding. Only on this 15th day, for the high fat diet group, were glucose levels observed indicative of moderately serious hypoglycemia. Lower blood glu cose levels during restricted feeding or fasting of rats have been reported pre viously (32, 34). Under the conditions of this experiment, no significant difference occurred in run ning performance between the groups with different dietary fat levels, even though during the restricted intake the amount of energy provided was 40% higher for the high than for the low fat group. However, for a given percent de crease in body weight within the same weight range, there was a significantly higher percent increase in performance for the low than for the high fat group. If the results reported above can be extrapolated to humans, the findings would be applicable to conditions of strenuous marches, such as where weight of food that can be carried is restricted.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1545
ACKNOWLEDGMENTS ^ The author is grateful to Dr. William K. Calhoun for his suggestions and to Julian W. Ratteree for his technical as sistance. LITERATURE CITED 1. Marsh, M. C. & Murlin, J. K. (1928) Mus cular efficiency on high carbohydrate and high fat diets. J. Nutr. 1, 105-137. 2. Christensen, E. H. & Hansen, O. (1939) Arbeitsfähigkeit und Ernährung. Skand. Arch Physiol. 81, 161-171. 3. Kark, R. M., Johnson, R. E. & Lewis, J. S. (1945) Defects of pemmican as an emer gency ration for infantry troops. War Med. 7, 345-352. 4. Takagi, K. (1946) Endurance and nutri tion. Sci. Foods (Japan) 1, 69-72; cf. Chem. Abstr. 46, 11,357 fgh. 5. Berghoff, A. & Clatzel, H. (1965) Aus wirkungen von Kostformen verschiedenen Fettgehaltes auf den menschlichen Organis mus. II. Energieumsatz und Leistungsfähig keit am Ergometer. Med. Pharmacol. exp. 12, 157-166. 6. Astrand, P.-O. (1967) Diet and athletic performance. Federation Proc. 26, 1772-1777. 7. Bergström, J., Hermansen, L., Hultman, E. & Saltin, B. (1967) Diet, muscle glycogen and physical performance. Acta Physiol. Scand. 71, 140-150. 8. Pruett, E. D. R. (1970) Glucose and in sulin during prolonged work stress in men living on different diets. J. Appi. Physiol. 28, 199-208. 9. Pernow, B. & Saltin, B. ( 1971 ) Availability of substrates and capacity for prolonged heavy exercise in man. J. Appi. Physiol. 31, 416-422. 10. Reed, L. L., Anderson, W. E. & Mendel, L. B. (1930) Factors influencing the dis tribution and character of adipose tissue in the rat. II. The influence of diet, undernutrition, fasting, and activity upon the distribu tion and character of fat. J. Biol. Chem. 87, 156-174. 11. Ershoff, B. H. (1954) Beneficial effect of low-fat diets on the swimming performance of rats and mice in cold water. J. Nutr. 53, 439-449. 12. Ekelund, L.-G. (1969) Exercise. Ann. Rev. Physiol. 37, 85-116. 13. Hollmann, W., Karcher, H. & Stolte, D. (1973) Effect of a carbohydrate diet on cardiopulmonary and endurance parameters of ability. Sportarzt Sportmed. 24, 55-60. 14. Deuel, H. J., Jr., Meserve, E. R., Sträub, E., Hendrick, C. & Scheer, B. T. (1947) The effect of fat level of the diet on general nutrition. I. Growth, reproduction and physi cal capacity of rats receiving diets contain ing various levels of cottonseed oil or mar garine fat ad libitum. J. Nutr. 33, 569-582. 15. Seheer, B. T., Dorst, S., Code, J. F. & Soule, D. F. (1947) Physical capacity of rats in
1546
16.
17.
18.
19.
20.
21. 22.
2.3.
24.
DIRK TOLLENAAR
relation to energy and fat content of the diet. Am. J. Physiol. 149, 194-203. Taylor, H. L., Buskirk, E. R., Brozek, J., Anderson, T. T. & Grande, F. (1957) Per formance capacity and effects of caloric re striction with hard physical work on young men. J. Appi. Physiol. 10, 421-429. Daws, T. A., Consolazio, C. F., Hilty, S. L., Johnson, H. L., Krzywicki, H. J., Nelson, R. A. & Witt, N. F. (1972) Evaluation of cardiopulmonary function and work per formance in man during caloric restriction. J. Appi. Physiol. 33, 211-217. Young, C. M., Scanlan, S. S., Im, H. S. & Lutwak, L. ( 1971 ) Effect on body compo sition and other parameters in obese young men of carbohydrate level of reduction diet. Am. J. Clin. Niitr. 24, 290-296. Bloom, W. L. & Azar, G. J. (1963) Simi larities of carbohydrate deficiency and fasting. I. Weight loss, electrolyte excretion and fatigue. Arch. Int. MecÃ-.112, 333-342. Ahlborg, B., Bergström, J., Ekelund, L.-G. & Hultman, E. (1967) Muscle glycogen and muscle electrolytes during prolonged physi cal exercise. Acta Physiol. Scand. 70, 129142. Costili, D. L. (1970) Metabolic responses during distance running. J. Appi. Physiol. 28, 251-255. Keys, A., Brozek, J., Henschel, A., Mickelsen, O. & Taylor, H. L. (1950) The Biology of Human Starvation. Vol. I. Univ. Minnesota Press, 763 pp. Vaughan, D. A., Drury, H. F., Hannon, J. P., Vaughan, L. N. & Larson, A. M. (1959) Some biochemical effects of restricted diets during successive field trials in winter. J. Nutr. 67, 99-108. Carlson, L. A. & Frosberg, S. O. (1967) Blood lipid and glucose levels during a 10day period of low-calorie intake and exer cise in man. Metab., Clin. Exp. 16, 624-634.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
25. Kimeldorf, D. J. (1961) The measurement of performance in small laboratory animals. In: Performance Capacity. A Symposium. Pp. 99-112. USQM Food & Cont. Inst. Armed Forces, Chicago. 26. Somogyi, M. (1930) A method for the preparation of blood filtrates for the deter mination of sugar. J. Biol. Chem. 86, 655— 663. 27. Hoffman, W. S. (1937) A rapid photo electric method for the determination of glu cose in blood and urine. J. Biol. Chem. 120, 51-55. 28. Brown, G. M., Zachwieja, A. & Stancer, H. C. ( 1966 ) An improved technique for con tinuous in vivo analysis of glucose. Clin. Chini. Acta 14, 386-390. 29. Steel, R. G. D. & Torrie, J. H. (I960) Principles and Procedures of Statistics with Special Reference to the Biological Sciences. McGraw-Hill Book Co., New York. 30. Siegel, S. (1956) Nonparametric Statistics for the Behavioral Sciences. Pp. 116-127. McGraw-Hill Book Co., New York. 31. Campos, F. A. de M., Cannon, W. B., Lundin, H. & Walker, T. T. (1929) Some condi tions affecting the capacity for prolonged muscular work. Am. J. Physiol. 87, 680-701. 32. Samuels, L. T., Gilmore, R. C. & Reinecke, R. M. (1948) The effect of previous diet on the ability of animals to do work during subsequent fasting. J. Nutr. 36, 639-651. 33. Tollenaar, D. (1970) The effects of vary ing ratios of minerals and vitamins to calories consumed during food intake restriction on physical condition of rats and man. U.S. Army Natick Labs. Tech. Rept. 71-21-FL, pp. 21. 34. Tollenaar, D. (1966) Effects of mineral and vitamin supplementation on swimming times and other parameters related to per formance of rats on a low calorie regimen. J. Nutr. 90, 441-448.
sumption of an all-carbohydrate as com pared to a "normal" diet. Deuel, Scheer et al. (14, 15), on the other hand, reported that rats swam longer when fed a 20% fat diet as compared to a 5% fat diet. However, the weights attached to the rats during the course of the swimming period had the effect of favoring the heavier rats from predominantly the higher dietary fat group. Further, swimming times averaged only 13 to 20 minutes. With a restricted intake, 500 to 580 Received for publication August 8, 1975. 1This paper reports research undertaken at the U.S. Army Natick Development Center and has been assigned no. TP 1705-FSL in the series of papers ap proved for publication.
1539 This
One
14EX-QLN-61Q7
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1540
DIRK TOLLENAAR
kcal/day mostly from carbohydrate, no significant decrease in exercise endurance of human subjects occurred during a 10 to 12-day period (16, 17). However, during an average intake by human subjects of 1,500 or 1,800 kcal/day from low-carbohy drate diets, in one case (3) coupled with fairly heavy, long-term exercise, lack of energy, rapid incapacitation, and excessive tiredness within days have been observed (3, 18, 19). When consuming about onethird of the ad libitum energy intake, no significant improvement in swimming en durance of rats was found with increasing the percentage of dietary fat (15). When consuming about two-thirds of the ad libitum energy intake, a higher average swimming endurance was observed while feeding a 40% fat diet compared to feed ing less fat in the diet. However, the use of weights attached to the rats during the swimming period tended to favor those from the high fat group, as mentioned above. Further, swimming times were only about 5 to 20 minutes. Especially after consuming a high fat diet, a decrease in blood glucose levels close to the point of exhaustion, after several hours of moderate or fairly heavy exercise, has been reported in human sub jects (2, 6-8, 20). Under similar condi tions, Costili (21) did not observe this, al though no information on diet is given. Such a drop in blood glucose can be a factor limiting exercise endurance (7, 8). During fasting and semistarvation, an initial decrease in blood glucose level dur ing the first few days followed by a gradual return to normal has frequently been found in humans and most animals, when the preceding diet was of average composition (22-24). Blood glucose levels in men fed 1,000 kcal/day fell most for those consuming a very low carbohydrate diet and least for the group consuming the most sugar (23). There exists a need to know how physi cal performance, especially exercise en durance, is affected by dietary fat content and level of food intake. The present study reports the effects of changes in fat con tent of the food on certain physical per formance parameters, especially running endurance, in rats, restricted in food intake and fed ad libitum.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
MATERIALS AND METHODS
Young adult male Long-Evans rats2 were housed individually in stainless steel cages at 21 ±1° (mean ±so) and 50 ±3% relative humidity. When received, they were 60 days old and weighed 245 ± 14 g. For the first 4 months after arrival, they were fed a stock diet.3 For the next 3 weeks they were fed a semipurified diet containing 40% of the energy from fat, as described below. Following this, the ac tual experimental phase was started. The rats were divided into 6 groups and fed either ad libitum or restricted, the diet containing 20%, 40% or 70% of the energy from fat. The restricted food intake was a uniform weight for all groups, namely 40% by weight of .the average ad libitum intake of all groups before the start of the different diet regimen. The composition, in percent, of the diet con taining 40% of the energy from fat was: vitamin-free casein,4 20.0; vegetable short ening,-'"' 10.45; lard,9 10.45; corn starch, 30.3; dextrin, 10.1; dextrose, 10.1; cellu lose, 4.0; mineral mixture,7 4.2; vitamin mixture,8 0.4. The diets with 20% and 70% energy from fat contained, respectively, 4.55% and 23.3% of, each, vegetable shortening and lard, and were prepared by interchange with the amount of corn starch, keeping the weight percentage of the other ingredients constant. The diets were stored refrigerated. The treadmill, used to run the rats, was a slight modification of a design used by Kimeldorf (25) and employed a motor' Obtained from Simonsen Labs., Inc., Gllroy, Cali fornia 95020. 3 Purina Laboratory Chow, Ralston Pnrina Co., St. Louis, Missouri. «All diet ingredients, unless stated otherwise, were obtained from ICN, Life Sciences Group, Cleveland, Ohio 44128. 5 Commercial, general purpose, high stability (100 AOM) shortening. « Armour Co., Chicago, Illinois. 7 The mineral mixture was composed as follows (%) : CaCO3, 15.8; CaHPO,, 35.8; KC1, 16.5; Na-HPO,, 17.2; MgSO4-7H2O> 12.4; MnSO,H2O, 0.44; ZnCOs, 0.11; CuS04-5H2O, 0.19; ferric citrate, 1.46; KIO3, 0.005; Cr( III »-acetate, 0.05. All inorganic salts were "certified" reagent grade, unless stated otherwise, and obtained from Fisher Scientific Co., Pittsburgh. Pennsylvania 15219. Ferric citrate was U.S.P. VIII grade. 8 The vitamin mixture, per kg of diet, contained (in mg, unless stated otherwise) : retinyl palmita te, G.OOOI.U. ; ergocalciferol. 3,000 I.D. ; or-tocopherol, 80 ; menadione, 0.30 ; thiamin-HCl, 6.0 ; riboflavin, 5.0; pyridoxine-HCl, 3.0; niacin, 30; Ca-pantothenate, 1C ; choline dihydrogen citrate, 3.5 g ; vita min Bjn, 0.056 ; ascorbic acid, 60 ; inositol, 17 ; p-aminobenzoic acid, 17; folie acid, 2.0; biotÃ-n, 0.1.
FAT LEVEL, CALORIC INTAKE AND PERFORMANCE
1541
based on the performance of the respec tive groups in the preliminary period as shown in table 1. The adjustment factor for running time for each group was de rived by dividing the overall mean time (41.8 minutes) by the preexperimental running time of the respective group. Sim ilarly, an adjustment factor for treadmill running performance was derived for each exhaustion. These runs took place at a group from the overall mean ( 102.8 kgm ) speed of 22.3 meters/minute (mpm) on and the preliminary period group mean. an incline of 12.5°.Treadmill running per Table 1 presents the adjusted mean run formance values in kilogram meters (kgm) ning time and treadmill running per were obtained by multiplying individual formance. body weights by total elevation during Under the conditions used, fat level had each run. Food was removed 1 hour be no significant effect on running times or fore running. Of 72 rats received, 48 were kgm performance for either the rats fed selected for the experimental phase. These ad libitum or restricted. On day 8 and 15 rats were randomly divided equally be of the test period, the restricted fed rats tween the 6 groups. ran to Stu dent'ssignificantly unpaired tlonger test ofaccording the means, and Heparinized blood samples were col lected during exsanguination, within 2 had a significantly higher kgm perfor mance according to the Mann-Whitney minutes after exhaustion. Deproteinization of the plasma was effected with 6% nonparametric U test than the rats fed ad ZnSO4-7H2O and approximately 0.3 N libitum. On day 8 and 15, average running NaOH (titrated against the ZnSO4 solu time increases were for the 20% fat group tion), as a modification of the method of 32% and 63%, for the 40%, fat group 33% Somogyi (26). Plasma glucose levels were and 63%, and for the 70% fat group 12% determined employing an automated mi and 17%, respectively. Corresponding crotechnique 9 involving a modification of kgm performance increases were for the the ferri-ferrocyanide oxidation-reduction 20% fat group 22% and 43%, for the 40% method according to Hoffman (27). Since fat group 18% and 37%, and for the 70% heparin has been found to raise recorded fat group 2% and 4%, respectively. glucose values (28), standards were like If the performance values of each rat wise heparinized. are averaged for the four running events Statistical analyses of the data for the to exhaustion prior to the experimental tables were performed using the means for phase, the following linear relationship to body weight is obtained: Y = 0.161X + each group and period, as presented, 126.2, in which X = body weight, g (range: rather than individual data. These means were analyzed first according to Student's 428-620), and Y = running time, minutes (range: 27-67), according to a regression t test (29). If this test yielded no signifi cant difference, a Mann-Whitney nonparaanalysis (F ratio = 51.8, P < 0.01, and metric U test (a ranking test) (30) was higher than for a 2nd order equation) also performed. In the cases, indicated, (29). If performance is expressed in kgm, where regression analyses and analyses of likewise, a linear relationship to body covariance were performed, the method weight is found under these conditions: outlined by Steel and Torrie (29) was Y = -0.184X + 195.6, in which Y = kgm performance, range: 74-134 (F ratio = followed. 13.2, P < 0.01). Average kgm performance RESULTS was plotted against average body weight separately for the ad libitum and re Running performance. To facilitate com stricted-fed groups for day 8 and 15 of the parisons of running performance of the different dietary groups throughout the test period and the 20% and 70% fat experimental periods, the experimental •Technlcon Auto Analyzer Methodology N-9, Techobservations were adjusted by factors nicon Instrument Corp., Chauncey, New York. driven, endless belt with a number of ventilated cages, open at the bottom just above the belt. After an initial training period prior to this, the rats were run to exhaustion once a week during the 7 weeks preceding the actual experimental phase. One day after the commencement of the experimental phase the and rat's withwere one run week intervals thereafter, to
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1542
DIRK TOLLENAAR TABLE l Effect of dietary fat level and food intake on running performance of male, young adult rats fed1Dietary fedmin(kgm)'min(kgm)min(kgm)min(kgm)min(kgm)minminmin(kgm)20%41. libitum
Days of experimental periodPreexperimen
energy from fat 20%42.7(104.0)0.979(0.988)38.7(95.0)38.7(94.8)38.7(95.4)38.738.738.7(98.5)48.7(111.4)0.858(0.9 40% 70%
84(lOO.O)41.000(1.028)41.1*(99.5 tal2Multiplicationfactors61815223643Ad
)439.4(95.9)36.3(88.3)40.143.040.3(99.5)Restricted
140% by weight of average ad libitum intake of all groups before start of different diet regimen. 1Average of four runs preceding the experimental period; diet containing 40% calories from fat. 3 Kilogrammeters. 4 Mean values. N = 7 or 8. 6To equalize pre-experimental 'base' values ; the values shown throughout the experimental period have been corrected by these factors for each group. On each of days 8 and 15, the restricted fed rats, as a group, ran longer (P < 0.02) than the rats fed ad libitum. The so of the running times on day 8 for the ad libitum fed group = 4.37, for the restricted fed group = 0.38 ; for day 15 these values are 2.45 and 7.85, respectively.
levels. To better compare the slopes of the lines for the 20% and 70% fat levels, average lines for days 8 and 15 for each fat level were calculated by use of the ap propriate multiplication factors. Thus, re gression equations of Y = -0.439X + 309.9 for 20% dietary fat energy, and Y = -0.051X + 126.6 for 70% dietary fat energy were found, using the individual values. ( For both cases, range of X > 426-534 g, and fits were better for a linear than for an exponential relationship.) The
slope for the 20% fat values was different (P< 0.005) from that for the 70% fat values, according to an analysis of covariance, using randomly paired individual values. The slopes for the 40% fat values were between those of the others. Body weights. Mean (unadjusted) body weights and the percent changes from the values just preceding the experimental period for selected dates of the experimen tal period are given in table 2. Water consumption. Mean water con-
TABLE 2 Effect of dietary fat level and food intake on mean body weights and percent change1
fed20%491» libitum Days of experimental period8
%*43 IAd
fed2energy 70%468
from fat 20%481
(-11.0) (-8.1) (-2.0) (-0.7) (+0.6) 457 467 470 455 491 505 (-11.0) (-1.5) (-13.2)40%493(-15.9)70%465(-9.4) (-0.8) (+0-7) 472 498 522 (+0.6)40%503(+4.0)Dietary(-1.2)Restricted
1Percent change from the average weight of groups during the 2 week period preceding the experimental period. 240% of the average ad libitum intake by weight of all groups. 3N = 7 or 8. Mean so = 9.3% (range: 5.0-13.3%) for the individual values within each ad libitum fed diet group.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
FAT LEVEL,
CALORIC
INTAKE AND PERFORMANCE
1543
TABLE 3 Effect of dietary fat level and food intake on mean water consumption1 •*•* fedml/dayml/dayml/dayml/dayml/dayml/day20%33.4' libitum fedDietary fat,70% energy from 40%29.3(1.070)28.527.430.829.729.429.9 "' 20%
experimental periodPre-experimental (Multiplicationfactor1)49111623Ad
(1.016)21.919.519.7 (0.951)19.319.819.8470%30.8 (1.046)19.217.818.232.9 (0.939)30.9531.131.328.432.840%31.5 (0.993)32.932.932.130.232.2Restricted
1 Values adjusted for differences between groups appearing before start of experimental phase. 2Over all days measured and all fat levels, water consumption was significantly higher for the ad libitum fed than for the restricted fed rats (P < 0.01). *Over all days measured, water consumption for the ad libitum fed rats was significantly lower in the 70% dietary energy from fat group than in the 20% or 40% groups (P < 0.05). *40% by weight of average ad libitum intake of all groups. * Mean values. N for all groups = 7 or 8. Mean so = 4.3 (range: 1.5-6.5) % of the mean values for each diet group.
on the average 9% less, according to Student's t test, than for the other two groups. From day 4, water consumption of rats fed at all fat levels was lower during the restricted feeding than feeding, according to during Student'sad libitum i test. These reductions between day 4 and 11 (not measured beyond this time) fpr the 20% and 40% fat groups were on the av erage 40%. For the group fed 10% of its
sumption of the rats during selected dates of the experimental period is given in table 3. Values obtained for each group during the experimental period were ad justed for differences between groups ap pearing prior to initiation of the experi mental period. During ad libitum feeding, no significant difference existed between the 2Qr/( and 40% fat groups. From day 2 to 23 of ad libitum feeding, the water con sumption of rats fed the high fat diet was
TABLE 4 Effect of dietary fat level and food intake on plasma glucose values1-1-'-4-6 fed20%101.0' libitum fedeDietary ofexperimental Days energy from fat 20%105.2(0.978)111.5101.697.099.1103.8 40% 70%
periodPre-experimental mlmg/100 (Multiplicationfactor2)181522mg/100
(1.031)75.172.051.9 (0.999)113.189.677.570%99.8 )63.574.280.968.1104.7 (0.991 (0.983)119.393.484.040%103.0 (1.019)109.47102.8101.095.7Restricted
mlmg/100 mlmg/100 mlmg/100 mlAd 1Blood was taken within 2 minutes after rats were run to exhaustion. *Values adjusted for differences between groups appearing before start of experimental phase. 3Over all days measured, plasma glucose at both food intake levels were lower in the 70% dietary energy from fat group than in the 20% or 40% groups (P < 0.05), and for the restricted fed rats was also lower in the 40% dietary energy from fat group than in the 20% group (P < 0.05). 4On day 8 and 15 plasma glucose was lower during restricted than during ad libitum feeding (P < 0.02). 6 For the restricted fed rats, plasma glucose was lower on day 15 than on day 8 (P < 0.05). ' 40% of the average ad libitum intake by weight. ' Mean values. N = 7 or 8. Depending on the kind of comparison made above, so of the means ranged from 2.0% to 23.9% of the values.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1.-.44
DIRK TOLLENAAR
energy from fat, this figure was 2,9% and significantly different (P < 0.05) from the reduction for the other groups. No signifi cant differences in water consumption oc curred within the restricted feeding regi men between groups. Plasma glucose levels. Glucose levels in plasma obtained within 2 minutes after the rats were run to exhaustion are given in table 4, These values were adjusted to the same average value (102.9 mg/100 ml), obtained just prior to the initiation of the experimental phase. During ad libi tum feeding, the plasma according to Student's pairedglucose / test levels, of the means, were significantly lower for the 70% fat group than for the 2JO%and 40% fat groups over the whole period mea sured. Further, on day 8 and 15 of the experimental period, the means for each fat level were lower during restricted than during ad libitum feeding. During the re stricted intake feeding period, plasma glu cose levels were lower for the 70% fat group than for the 40% fat group; and for the latter, in turn, lower than for the 20% fat group. Finally, plasma glucose levels on day 15 of the restricted feeding period were lower than on day 8. DISCUSSION On the average, no significant weight loss occurred among the ad libitum fed rats during the experimental period. Thus, the exercise conditions are unlikely to have played any significant role in the weight loss of the restricted fed group. The restricted intake group fed 20% of its dietary energy from fat ran on day 15 of the experimental period on the average 63% longer, with a 43% higher perfor mance in kgm, for a 12.4% weight loss, as compared to the ad libitum fed 20% fat group. These figures for the 40% and 70% fat groups were, respectively: 63, 37, 16.6 and 17, 4, 9.6. Thus, a 9.6% weight loss, assuming a linear relationship, for the restricted fed 20%, 40%, and 70% fat groups corresponded with, respectively, a 49%, 36%, and 17% increase in running time, and a 33%, 21%, and 4% increase, respectively, in kgm performance. This difference in rate of increase in kgm per formance between the 20% and 70% fat groups was significant (P < 0.005), ac
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
cording to an analysis of covariance, using individual values. During the period when 40% dietary energy from fat, fed ad lib itum, was provided, rats with a 9.6% lower weight in the same range, on the average, had a 19% longer running time and a 9% higher kgm performance. A correlation coefficient, r = —0.94existed in this latter case between body weight and running time, averaged for the future groups and four preliminary events, mentioned. Based on the equation obtained for the relationship between body weight and kgm performance for the period before the start of the experimental period, the equalization of the average preliminary running times and kgm performance intro duced an average error into the experi mental results given of not more than about 1.4% of the values, the largest error being about 3.3% (for the 20% fat, ad libitum fed group). This does not affect the relative significance of the results. Lower body fat may have been an im portant factor in the higher running per formance with lower body weight during ad libitum feeding and in, at least, the early stages of restricted feeding. An ad ditional factor promoting increased run ning performance with decreased body weight during restricted feeding might be an increased adrenalin secretion (31) as a result of the stress imposed by the re stricted food intake. Samuels et al. (32) have reported that running endurance of adult male rats was higher from the 9th day of fasting on, as compared with ihe 2nd to 4th day. No information on the magnitude of a training effect is provided by these authors, however, exhaustion tests having been initiated only aftsr the start of fasting. Tollenaar (33) has re ported a significant increase in running endurance of rats on the 7th, compared to the 3rd day of a restricted food intake, as well as compared to running times while fed ad libitum. The higher amount of food energy offered with the 70% fat diet than with the 20% fat diet during the restricted in take provided no advantage in terms of running time or kgm performance. During the ad libitum feeding, a tendency seemed to exist for the rats fed the diet with 70% energy from fat to initially show a better
7 FAT LEVEL, CALORIC INTAKE AND PERFORMANCE
running performance than when fed the other diets. This tendency diminished or reversed itself with time, however. The contribution from carbohydrate varied between 17% and 61% of the energy intake (25%-62% by weight) in the present experiment. The lower amount of solutes presented during restricted as compared to ad lib itum feeding no doubt was the main, if not the sole, cause of the considerably lower water consumption in the former case. During the combustion of fat more water is formed per gram than during the com bustion of carbohydrate. A 9% lower water consumption during ad libitum feeding was observed for the 70% fat than for the other two groups. These dif ferences are largely in agreement with those observed earlier (34). The lower plasma glucose levels with the high fat than with the lower dietary fat levels, both during ad libitum and re stricted feeding, as observed in the rat experiment described above, have, like wise, been reported in humans (7, 23). The plasma glucose levels, observed in the rats right after exhaustion, were lower during the restricted than during the ad libitum feeding regimen, and were lower on the 15th than on the 8th day of re stricted feeding. Only on this 15th day, for the high fat diet group, were glucose levels observed indicative of moderately serious hypoglycemia. Lower blood glu cose levels during restricted feeding or fasting of rats have been reported pre viously (32, 34). Under the conditions of this experiment, no significant difference occurred in run ning performance between the groups with different dietary fat levels, even though during the restricted intake the amount of energy provided was 40% higher for the high than for the low fat group. However, for a given percent de crease in body weight within the same weight range, there was a significantly higher percent increase in performance for the low than for the high fat group. If the results reported above can be extrapolated to humans, the findings would be applicable to conditions of strenuous marches, such as where weight of food that can be carried is restricted.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
1545
ACKNOWLEDGMENTS ^ The author is grateful to Dr. William K. Calhoun for his suggestions and to Julian W. Ratteree for his technical as sistance. LITERATURE CITED 1. Marsh, M. C. & Murlin, J. K. (1928) Mus cular efficiency on high carbohydrate and high fat diets. J. Nutr. 1, 105-137. 2. Christensen, E. H. & Hansen, O. (1939) Arbeitsfähigkeit und Ernährung. Skand. Arch Physiol. 81, 161-171. 3. Kark, R. M., Johnson, R. E. & Lewis, J. S. (1945) Defects of pemmican as an emer gency ration for infantry troops. War Med. 7, 345-352. 4. Takagi, K. (1946) Endurance and nutri tion. Sci. Foods (Japan) 1, 69-72; cf. Chem. Abstr. 46, 11,357 fgh. 5. Berghoff, A. & Clatzel, H. (1965) Aus wirkungen von Kostformen verschiedenen Fettgehaltes auf den menschlichen Organis mus. II. Energieumsatz und Leistungsfähig keit am Ergometer. Med. Pharmacol. exp. 12, 157-166. 6. Astrand, P.-O. (1967) Diet and athletic performance. Federation Proc. 26, 1772-1777. 7. Bergström, J., Hermansen, L., Hultman, E. & Saltin, B. (1967) Diet, muscle glycogen and physical performance. Acta Physiol. Scand. 71, 140-150. 8. Pruett, E. D. R. (1970) Glucose and in sulin during prolonged work stress in men living on different diets. J. Appi. Physiol. 28, 199-208. 9. Pernow, B. & Saltin, B. ( 1971 ) Availability of substrates and capacity for prolonged heavy exercise in man. J. Appi. Physiol. 31, 416-422. 10. Reed, L. L., Anderson, W. E. & Mendel, L. B. (1930) Factors influencing the dis tribution and character of adipose tissue in the rat. II. The influence of diet, undernutrition, fasting, and activity upon the distribu tion and character of fat. J. Biol. Chem. 87, 156-174. 11. Ershoff, B. H. (1954) Beneficial effect of low-fat diets on the swimming performance of rats and mice in cold water. J. Nutr. 53, 439-449. 12. Ekelund, L.-G. (1969) Exercise. Ann. Rev. Physiol. 37, 85-116. 13. Hollmann, W., Karcher, H. & Stolte, D. (1973) Effect of a carbohydrate diet on cardiopulmonary and endurance parameters of ability. Sportarzt Sportmed. 24, 55-60. 14. Deuel, H. J., Jr., Meserve, E. R., Sträub, E., Hendrick, C. & Scheer, B. T. (1947) The effect of fat level of the diet on general nutrition. I. Growth, reproduction and physi cal capacity of rats receiving diets contain ing various levels of cottonseed oil or mar garine fat ad libitum. J. Nutr. 33, 569-582. 15. Seheer, B. T., Dorst, S., Code, J. F. & Soule, D. F. (1947) Physical capacity of rats in
1546
16.
17.
18.
19.
20.
21. 22.
2.3.
24.
DIRK TOLLENAAR
relation to energy and fat content of the diet. Am. J. Physiol. 149, 194-203. Taylor, H. L., Buskirk, E. R., Brozek, J., Anderson, T. T. & Grande, F. (1957) Per formance capacity and effects of caloric re striction with hard physical work on young men. J. Appi. Physiol. 10, 421-429. Daws, T. A., Consolazio, C. F., Hilty, S. L., Johnson, H. L., Krzywicki, H. J., Nelson, R. A. & Witt, N. F. (1972) Evaluation of cardiopulmonary function and work per formance in man during caloric restriction. J. Appi. Physiol. 33, 211-217. Young, C. M., Scanlan, S. S., Im, H. S. & Lutwak, L. ( 1971 ) Effect on body compo sition and other parameters in obese young men of carbohydrate level of reduction diet. Am. J. Clin. Niitr. 24, 290-296. Bloom, W. L. & Azar, G. J. (1963) Simi larities of carbohydrate deficiency and fasting. I. Weight loss, electrolyte excretion and fatigue. Arch. Int. MecÃ-.112, 333-342. Ahlborg, B., Bergström, J., Ekelund, L.-G. & Hultman, E. (1967) Muscle glycogen and muscle electrolytes during prolonged physi cal exercise. Acta Physiol. Scand. 70, 129142. Costili, D. L. (1970) Metabolic responses during distance running. J. Appi. Physiol. 28, 251-255. Keys, A., Brozek, J., Henschel, A., Mickelsen, O. & Taylor, H. L. (1950) The Biology of Human Starvation. Vol. I. Univ. Minnesota Press, 763 pp. Vaughan, D. A., Drury, H. F., Hannon, J. P., Vaughan, L. N. & Larson, A. M. (1959) Some biochemical effects of restricted diets during successive field trials in winter. J. Nutr. 67, 99-108. Carlson, L. A. & Frosberg, S. O. (1967) Blood lipid and glucose levels during a 10day period of low-calorie intake and exer cise in man. Metab., Clin. Exp. 16, 624-634.
Downloaded from https://academic.oup.com/jn/article-abstract/106/11/1539/4768752 by University of Glasgow user on 23 April 2018
25. Kimeldorf, D. J. (1961) The measurement of performance in small laboratory animals. In: Performance Capacity. A Symposium. Pp. 99-112. USQM Food & Cont. Inst. Armed Forces, Chicago. 26. Somogyi, M. (1930) A method for the preparation of blood filtrates for the deter mination of sugar. J. Biol. Chem. 86, 655— 663. 27. Hoffman, W. S. (1937) A rapid photo electric method for the determination of glu cose in blood and urine. J. Biol. Chem. 120, 51-55. 28. Brown, G. M., Zachwieja, A. & Stancer, H. C. ( 1966 ) An improved technique for con tinuous in vivo analysis of glucose. Clin. Chini. Acta 14, 386-390. 29. Steel, R. G. D. & Torrie, J. H. (I960) Principles and Procedures of Statistics with Special Reference to the Biological Sciences. McGraw-Hill Book Co., New York. 30. Siegel, S. (1956) Nonparametric Statistics for the Behavioral Sciences. Pp. 116-127. McGraw-Hill Book Co., New York. 31. Campos, F. A. de M., Cannon, W. B., Lundin, H. & Walker, T. T. (1929) Some condi tions affecting the capacity for prolonged muscular work. Am. J. Physiol. 87, 680-701. 32. Samuels, L. T., Gilmore, R. C. & Reinecke, R. M. (1948) The effect of previous diet on the ability of animals to do work during subsequent fasting. J. Nutr. 36, 639-651. 33. Tollenaar, D. (1970) The effects of vary ing ratios of minerals and vitamins to calories consumed during food intake restriction on physical condition of rats and man. U.S. Army Natick Labs. Tech. Rept. 71-21-FL, pp. 21. 34. Tollenaar, D. (1966) Effects of mineral and vitamin supplementation on swimming times and other parameters related to per formance of rats on a low calorie regimen. J. Nutr. 90, 441-448.
