PRELIMINARY REPORT

The Effect of Ephedrine/Caffeine Mixture on Energy Expenditure Body Composition in Obese Women

and

Arne Astrup, Benjamin Buemann, Niels Juel Christensen, Smren Toubro, Grete Thorbek, Ole J. Victor, and Flemming Quaade Treatment with pz-agonists promotes fat loss and muscle growth in numerous species, but human studies are lacking. We studied the effect of a compound with pz-agonistic properties (ephedrine 20 mg/caffeine 200 mg [E+C]). Fourteen obese women were treated with a 4.2-MJ/d diet and either E+C or placebo (P) three times per day for 8 weeks in a double-blind study. Weight-loss was not different in the groups, but the E+C group lost 4.5 kg more body fat and 2.8 kg less fat-free mass (FFM). The decrease in 24-hour energy expenditure (EE) seen in the P group was 10% at day 1 and 13% at day 56, but was only 7% and 8% in thetreated group (P = .044). The higher EE in the E+C gtoup was entirely covered by fat oxidation. These findings provide evidence that promotion of fat loss and preservation of FFM during weight reduction may also be achieved pharmacologically in humans. Copyright o 1992 by W.B. Saunders Company

N

REPORTS to date have shown that substances with selective P2-adrenergic properties reduce body fat and increase lean, mainly skeletal muscle in several species such as mice, rats, sheep, cattle, poultry, and pigs.’ This principle may be an attractive pharmacological target for the treatment of human obesity, where promotion of fat loss with preservation of lean body mass may be UMEROUS

favorable.

However,

until now no study has reported

the

therapeutic effect of a compound with Pz-agonistic properties on body composition and energy expenditure (EE) in humans. In a previous controlled clinical trial we used ephedrine, a sympathomimetic drug with Pragonistic properties, in combination with caffeine, and found that the weight-loss induced by ephedrine was potentiated by caffeine.2 This subsequent study was performed in a double-blind placebo-controlled design, where a third party allocated 16 healthy obese women to form two matched groups. They received 8 weeks of treatment with a 4.2MJ/d diet and either ephedrine/caffeine (E+C) or placebo {P). All patients underwent measurement of body composition and 24-hour EE on three occasions: (1) during pretreatment on a weight-maintenance diet; (2) at the first day of treatment by diet and drug; and (3) at the last day of treatment (56th day).

From the Research Department of Human Nutrition, Royal Veterinary and Agricultural University, Copenhagen; and The Department of Internal Medicine and Endocrinology, Herlev Hospital, University of Copenhagen, Denmark. Supported in pan by Danish Medical Research Councii Grant No. 12-9084 and Danish Veterinary and Agricultural Research Council Grant No. 13-4268. Address reprint requests to Ante Astrup, MD, Dr Med Sci, Research Department of Human Nuti’tion, Royal Veterinary and Agricultural University Rolighedsvej 25, 1958 Fredetihberg C, Copenhagen, Denmark. Copyright 0 1992 by W.B. Saunders Company 0026-0495/92/4107-0002$03.00/O

686

SUBJECTS AND METHODS

Twenty-four-hour EE and substrate oxidation rates were measured in two open-circuit respiratory chambers.3,4 A fixed physical activity program was followed, including three bicycling sessions of 10 minutes each (75 W), and meals were served at 9:30 AM, 12:30 PM, and 6:00 PM. The subjects were kept under 24-hour surveillance by the staff. On the first stay in the respiratory chambers, the weight-maintenance diet consisted of conventional foods (energy: 55% as carbohydrate, 30% as fat, and 15% as protein). On both the second and the last stay in the respiratory chamber, the patients were fed a 4.2-MJ/d, high-carbohydrate, low-fat, conventional diet. They were given dietetic instruction and nutritional advice in groups. The pharmacological treatment consisted of either E+C (ephedrine 20 mg plus caffeine 200 mg) or P three times a day 1 hour before meals. The tablets were indistinguishable in weight, appearance, and taste. Before discharge from the respiratory chamber, body composition was estimated by the bioimpedance method.3 Group means of baseline data were compared with a two-sided Student’s unpaired t test, while changes in EE, substrate oxidations, and body composition were compared by an ANOVA, with the pretreatment value as a covariate. Results are presented as the means @EM). RESULTS

One patient from each treatment group dropped out before the final assessment. The two groups that completed the study were comparable with respect to physical characteristics and body composition (body mass index [BMI]: E+C, 34.9 [1.6]; P, 33.3 [1.8]). After 8 weeks of E+C treatment, the mean weight-loss was 10.1 (0.4) kg, and on P treatment it was 8.4 (1.2) kg (NS). However, 4.5 kg more body fat was lost in the E+C group than in the P group. Only 1.1 (1.4) kg fat-free mass (FFM) was lost in the EiC group, while 3.9 (1.7) kg was lost in the P group (P < .05). In the P group, the slimming diet caused 24-hour EE to decrease by 19.4 (3.6) kJ/kg FFM at the first day of treatment (Table l), and by 26.5 (3.7) kJ/kg FFM after 8 weeks (P < .OOOl).The decreases in 24-hour EE in the E+C group were only 12.5 (2.0) kJ/kg FFM at day 1 and 15.6 (2.4) kJ/kg FFM after 8 weeks (P < .OOl). The decreases in 24-hour EE corresponded to 10% and 13% in the P group, and to only 7% and 8% in the E+C group

Metabolism,

Vol41,

No 7

(July), 1992: pp

686-688

6-AGONISTS

AND

ENERGY

687

EXPENDITURE

Table 1. Changes in EE and Substrate Oxidation During Treatment

With Either E + C or P as Adjuvant to a 4.2MJ/d

Diet

Pretreatment

Day 1

Day 56

Diet

Drug

Effect

Effect

-

24-hour

EE (kJ/d)

E+C P A24-hour

EE (kJ/kg

t 436

9,585

lr 369

8,836

2 39.2

9,839

-c 522

8,942

2 312

8,207

? 296

-

- 12.5 -c 2.0

-15.6

+ 2.4

P

-

- 19.4 f 3.6

~26.5

t 3.7

466 t 14

424 -c 11

421 + 18

P

461 r 10

422 + 7

404 ? 9

308 + 10

304 -c 8

292 r 9

P

313 + 6

304 2 5

291 2 7

lipid oxidation

(g/14

59 f 4

90 t 6

98 + 6

P

62 + 5

84 i 5

83 & 4

lipid oxidation

(g/8

31 * 2

41 t 2

41 * 2

P

28 r 1

37 2 1

36 +- 2

Lipid to carbohydrate

oxidation

0.10

0.06

0.82

0.0001

0.13

0.0001

0.01

0.0001

0.07

ratio (g/g)

E+C

0.40

+ 0.02

1.04 2 0.07

1.37 2 0.11

P

0.43

+ 0.05

0.98 2 0.14

1.04 + 0.09

NOTE.

0.0001

h)

E+C

EE, from

0.04

h)

E+C Sleeping

0.0002

EE (kJ/h)

E+C Daytime

0.05

EE (kJ/h)

E+C Sleeping

0.0007

FFM/d)

E+C Daytime

-

10,316

EE has been adjusted

for difference

in FFM,S except

where

expressed

per kg FFM. Daytime

EE, from

10:00 AM to 12:00

midnight,

Sleeping

1:OO AM to 9:00 AM.

(E+C L’P, P = .044), but the effect of E+C was unchanged with time. Although the statistical power of the study does not allow a firm conclusion, most of the effect of E+C seemed to be exerted during the daytime (Table 1). Pronounced decreases in carbohydrate oxidation and increases in lipid oxidation were seen in both groups on the first day of energy restriction (Fig 1). While the changes in carbohydrate oxidation were similar in the two groups, the higher 24-hour EE caused by E+C treatment was entirely covered by a higher lipid oxidation. Lipid oxidation was enhanced by 13 and 15 g/d, at the first and last day of

Carboydrate

treatment, respectively (Fig I), in the E+C group compared with the P group. Quantitatively, most of the effect of E+C was found during the daytime, but the effect only reached statistical significance during sleep. The ratio of fat oxidation to carbohydrate oxidation increased substantially during energy restriction, but the ratio was only insignificantly further increased by E+C (P = .07). Protein oxidation did not change at the first treatment day, but decreased by 35% at the last day of treatment (P < .003); no group difference was observed (P < .13). Three patients in the E+C group complained of insomnia. palpitations, and

Oxidation (g!kl_

Lipid Oxidation (g/d) r ~----~~~-I 150

275

260 140 225 200

130

I75 120

150 125

110

100 Fig 1. Effect on carbohydrate oxidation and lipid oxidation of oral administration of E+C or P as adjuvant to a 4.2-f&J/d diet. Effect of diet (carbohydrate oxidation), P c .OoOl; effect of drug, P = .69. Effect of diet (lipid oxidation), P < .OOOl; effect of drug, P = ,032. f6, P; w, E+C.

100

75 50

90 26 0

80 Pie-treatment

Day 1

Dey of meawremenf

Dsy 56

Pru-treetmentDay 1

Day 56

Day of measurement

688

ASTRUP ET AL

tremor, respectively. The side effects were transient and disappeared after 6 to 14 days of treatment. No side effects were reported in the P group. We have previously reported that this E+C combination administered with a low-energy diet resulted in a 3.6-kg better weight-loss than did diet and P after 24 weeks.2 Although not significant, due to the small sample size in the present study, the weight-loss of 1.7 kg attributed to 8 weeks of E+C treatment resembles that of the former study. That the additional weight-loss of 1.7 kg actually disguised a fat loss of 4.5 kg suggests that E+C possesses a repartitioning impact in humans. The additional 4.5kg body fat loss in the E+C group was not reflected in a similar body weight decrease due to the 2.8-kg smaller reduction in FFM. The larger decrease in protein oxidation during treatment in the E+C group did not reach statistical significance, and because FFM consists of both lean body mass and water, it is not possible to say if the impact of E+C on FFM is due to an anabolic effect on skeIeta1 muscle. Although tolerance is known to develop to most of the hemodynamic and renal effects of caffeine during long-term treatment, a persisting mild diuretic effect after 8 weeks would only tend to underestimate the magnitude of the preservation of FFM. We know of only one other human study dealing with repartitioning agents. This was reported by Acheson et al, who found a positive protein balance and a negative fat balance during 2 weeks of treatment with the l32-agonist, terbutaline, but no change in body composition was detected.6 Twenty-four-hour EE decreased in both groups when their diet was changed from weight maintenance to 4.2 MJ/d, but the decrease was partly prevented by E+C.

The stimulation of 24-hour EE was entirely based on increased lipid oxidation, and the effect was maintained during the 8 weeks of treatment. The effect of E+C on EE could not be explained by the smaller reduction in FFM, mainly because the effect appeared from the first day of treatment. We cannot rule out that the increase in EE caused by E+C was due to an increased spontaneous physical activity, which may accompany stimulation of the sympathetic nervous system. This possibility is supported by the finding that, although statistically insignificant, most of the effect of E+C seemed to take place during sleep (Table 1). Assuming that the thermogenic effect of E+C would be similar during free-living conditions, it was possible to estimate that approximately 20% of the weight-loss caused by E+C was due to stimulation of EE, and the remaining 80% must be due to decreased energy intake caused by the anorectic effect of the compound. In conclusion, this study provides evidence that stimulation of EE, promotion of fat loss, and preservation of FFM during dieting may be achieved pharmacologically, not only in animals, but also in humans. However, due to the small number of subjects in the study, we find that the results should be confirmed and extended in larger trials with more detailed measures of body composition. Such studies are in progress.

ACKNOWLEDGMENT

We thank John Lind, Lene Kromann-Larsen, mann, Bente Mathiasen, and Tina Cuthbertson tance.

Inge Timmerfor their assis-

REFERENCES

1. Stock MJ: New approaches to the control of obesity in animals and their clinical potential, in Somogyi JC, Hejda S (eds): Nutrition in Prevention of Disease. Base], Switzerland, Karger, 1989, ov 32-37 2. .&up A, Breum L, Toubro S, et al: The effect and safety of an ephedrine/caffeine compound compared to ephedrine, caffeine and placebo in obese subjects on an energy restricted diet. Int J Obesitv 16:269-277.1992 3. Astrup A, Thorbek G, Lind J, et al: Prediction of 24-h energy expenditure and its components from physical characteristics and body composition in normal-weight humans. Am J Clin Nutr 521777-783, 1990

4. Astrup A, Buemann B, Christensen NJ, et al: The contribution of body composition, substrates and hormones to the variability in energy expenditure and substrate utilization in premenopausal women. J Clin Endocrinol Metab 74:279-286, 1992

_

5. Ravussin E, Bogardus C: Relationship of genetics, age and physical fitness to daily energy expenditure and fuel utilization. Am J Clin Nutr 49:968-97.5, 1989 6. Acheson KJ, Ravussin E, Schoeller DA, et al: Two-week stimulation or blockade of the sympathetic nervous system in man: Influence on body weight, body composition, and 24-hour energy expenditure. Metabolism 37:91-98, 1988