Reduced

Metabolic

Rate During

Stephen

(5Adrenergic

Blockade

in Humans

Welle, Ronald G. Schwartz, and Marcia Statt

We previously reported that 1 week of propranolol treatment (160 to 240 mgld, orally) reduced resting metabolic rate (RMR) an average of 9% in healthy men. To determine whether this response was caused by the 25% reduction in serum triiodothyronine (T,), rather than (I-adrsnergic blockade, we examined the effect of nadolol on RMR in five healthy men. Nadolol is a nonselective 6-adrenergic antagonist that does not affect T, production. After 6 to 10 days of nadolol treatment (240 mg/d), mean postabsorptive RMR declined 7% (P < .Ol), with no significant change in serum T3 or thyroxine (T,) concentrations. This effect is significantly different from that of a hospitalized control group that received no drug and had no change in mean RMR, and was not different from the response to propranolol (previously published data). Nadolol slightly reduced the mean thermic response to a meal (12%). but this effect was not statistically significant. Mean postprandial RMR was 6% lower after nadolol treatment (P < .Ol), mainly because of the reduced postabsorptive RMR, rather than a change in the response to the meal. These data suggest that (Sadrenergic activity makes a small but significant contribution to resting energy expenditure in man. Copyright 0 1991 by W.B. Saunders Company

I

T IS WELL KNOWN that the sympathetic nervous system mediates the increase in resting metabolic rate (RMR) induced by cold and overfeeding in rodents via activation of brown adipose tissue,‘.’ and that catecholamine infusions or conditions that stimulate sympathetic nervous system activity increase RMR in humans.3-’ The thermogenic action of catecholamines in humans is prevented by propranolol, a nonselective P-adrenergic antagonist.3.5The contribution of the sympathetic nervous system to basal (unstimulated) RMR is unclear. While intravenous propranolol does not reduce basal RMR within a few hours of adn~inistration.3~64 we and others have found that several days of oral propranolol treatment reduces basal RMR.‘“,” These data suggest that P-adrenergic activity has a role in regulating basal RMR, but this conclusion is hindered by the nonspecific effects of propranolol. Of particular interest is that propranolol reduces concentrations of triiodothyronine (T,), a known thermogenic hormone.“.” Moreover. propranolol has a high lipophilicity” and therefore readily crosses the blood-brain barrier, so that we cannot rule out central nervous system effects, rather than blockade of peripheral f3-receptors. as the cause of metabolic effects. Thus, in the present study, we examined the effect of another nonselective p-antagonist, nadolol, on RMR. Nad0101was chosen because it has a f3-adrenergic potency and relative bioavailability similar to propranolol,” does not affect T, production at concentrations needed for effective P-adrenergic blockade,13 and has a low lipophilicity” so that much less crosses the blood-brain barrier. Although it is known that meals induce a mild stimulation of the sympathetic nervous system,‘4~‘hthere is no general agreement about how much this response contributes to the postprandial increase in thermogenesis.3.6.X.“.‘5~‘x Thus, we also examined the effect of nadolol on the thermic response to a meal.

consent. The protocol was approved by the University of Rochester Research Subjects Review Board. The subjects stayed in the University of Rochester General Clinical Research Center (CRC) throughout the study. They were not allowed to leave the CRC unless accompanied by a staff member. Throughout the study, each subject received a weightmaintenance diet with an energy content equal to 150% of his resting energy expenditure (measured before admission), with 15% of energy from protein, 40% from fat, and 45% from carbohydrate. These procedures were identical to those used in the weightmaintenance phase of a previous study” of six control subjects (no drug) and eight propranolol-treated subjects, so that comparisons between the present group and the previous groups could be made. These previously studied men were similar to the nadolol-treated subjects in age (18 to 40 years); the untreated subjects had a mean body weight (75.8 2 5.0 kg) similar to that of the nadolol-treated subjects, whereas the propranolol-treated subjects tended to be slightly heavier (80.3 + 3.3 kg). No drug was administered on the first 3 days of the study. A single 80-mg tablet of nadolol was administered on days 4 and 5, 80-mg tablets were given every 12 hours on days 6 and 7, and 80-mg tablets were given every 8 hours on days 8 to 13. On day 14. an 80-mg tablet was administered before the final RMR determination. RMR was measured with a ventilated mask, open-circuit indirect calorimeter, as previously described.” RMR was determined for a l-hour period after an overnight fast on days 2,3 (pre-drug), 6, 8, 10, 13. and 14 (after 2 to 10 full days of nadolol treatment). The smoker abstained from cigarettes for at least 8 hours before each RMR measurement. Before every RMR measurement, a serum sample was taken for measurement of serum T, and thyroxine (T,) levels by radioimmunoassay. On one of the pre-drug days and one of the final 2 days of nadolol administration, the RMR measurement was continued for another 3 hours after the subjects drank a liquid meal (Ensure [Ross Laboratories, Colum-

From the Departments of Medicine (Endocrine and Cardiology Physiology, and Radiology, Division of Nuclear Medicine, and the Department of Nursing Practice, University of Rochester School of Medicine and Dent&y, Rochester, NY. Supported by National Institutes of Health Grants No. DK-39063, RR-00044, and KO8HLO1823. Address reprint requests to Stephen Welle, PhD, Endocrine Unit. Monroe Community Hospital, 435 E Henrietta Rd. Rochester, NY 14620. Copyright 0 1991 by W.B. Saunders Company 0026-0495/9114006-0010$03.00l0 Units),

METHOD The subjects were five healthy men, 23 to 35 years old (mean, 29). Body weight ranged from 68 to 84 kg (mean, 74) and fat free mass, measured by ‘“K counting,” ranged from 51 to 66 kg (mean, 58). One subject smoked 10 cigarettes per day throughout the study; the others were nonsmokers. Before the study, all procedures and risks were explained verbally and in writing. All subjects gave verbal and written Metabolism,

Vol 40, No 6 (June), 1991:

pp 619-622

619

620

WELLE, SCHWARTZ, AND STATT

Table 2. Effect of Nadolol on Postabsorptive

bus, OH], 10 kcal/kg, 14% protein, 31% fat, 55% carbohydrate, as percent of energy). We previously found that this meal increases plasma norepinephrine levels an average of 40% in healthy men, indicating sympathetic nervous system stimulation.‘* RMR, hourly postprandial increases in RMR, and serum T, and T, values were analyzed by repeated-measures ANOVA. Post hoc comparisons were made with Dunnett’s test. which keeps the overall significance level of the whole set of comparisons at the stated level. A one-sided test was used because we specifically were looking for a reduction in RMR. Other comparisons were made with r tests. RESULTS

Body weight was constant

RMR

and Serum T, and T, RMR Day

[kcalhnin)

T3 lng/dLl

T, (&JU

2

1.16 + 0.03

112 + 15

7.6 -t 0.8

3

1.14 + 0.03

109 2 6

7.4 -f 0.5

6

1.08 + 0.05’

106?

5

7.8 5 0.5

8

1.08 + 0.03*

104?

6

7.5 + 0.6

10

1.07 + 0.03*

100 + 7

8.1 + 0.6

13

1.07 + 0.03x

106?

12

7.2 + 0.6

14

1.08 ? 0.03*

103 2 IO

7.7 + 0.5

NOTE. Values are mean & SEM. Nadolol treatment started on day 4.

throughout

the study (Table

1). Nadolol treatment was associated with the expected reductions in blood pressure and pulse rate (Table 1). No adverse clinical effects of nadolol administration were observed. ANOVA showed a significant change in RMR (P < .Ol). This effect was explained by the lower mean RMR each day during drug treatment, compared with the initial RMR (Table 2). On day 6 (the third day of nadolol treatment), RMR was unchanged in two subjects and reduced 7% to 12% in three subjects. On day 8 (the fifth day of nadolol treatment), RMR was increased 4% in one subject and reduced 3% to 12% in the other four subjects. On days 10, 13, and 14, after the subjects had received the full dose of nadolol for 2 days or more, all RMRs were lower than the baseline value. The mean RMR on days 10 to 14 was 0.08 ? 0.01 kcal/min (mean, 7%; range, 4% to 9%) lower than the mean baseline RMR (P < .Ol by paired t test) (Fig 1). This change was substantially greater than the day-to-dayvariability in RMR (mean coefficient of variation was 3.4% before drug treatment and 1.1% on days 10 to 14). The subject who smoked cigarettes had a response similar to that of the other subjects (9% decrease in RMR on days 10 to 14 compared with baseline). We also compared the effect of nadolol on RMR on day 10 (after 6 days of treatment) to changes in RMR previously observed in healthy men hospitalized for the same period of time with no drug treatment or with propranolol treatment.” As shown in Table 3, the group that did not receive any p-blocker had no change in mean RMR during the study, whereas propranolol-treated subjects had a mean decrease of 9% (P < .Ol) and nadolol-treated subjects had a mean decrease of 7% (P < .Ol). The ANOVA showed a significant group (no drug, propranolol, or nadolol) x time (pre-drug, post-drug) interaction (P < .025). This interaction also was significant (P < .Ol) when the propranolol-

“P

< .05 compared with day 2 by Dunnett’s test

treated treated

subjects were excluded, indicating that nadololsubjects had a significant decrease in RMR relative

to the changes in RMR in the untreated subjects. The effects of nadolol and propranolol were not significantly different, as indicated by the lack of a group x time interaction when the no-drug group was excluded from the ANOVA (P > .50). Nadolol had no significant effect on serum T, (P > .80 by ANOVA) or T, concentrations (P > .25) (Table 2). Although there was no significant change in T, levels during nadolol treatment, four of five subjects had small decreases in serum T, after 6 days of treatment, when comparisons with the effect of propranolol were made as indicated above. However, the mean (GEM) decrease in serum T, levels was three times greater (P < .05) in propranololtreated subjects (24% ? 5%) than in nadolol-treated subjects (8% ? 5%), and there was no significant correlation (r = -.38, P > .lO) between the percent change in serum T, and the percent change in RMR among all 13 subjects treated with a P-blocker. Hence, there is no evidence that

Table 1. Effect of Nadolol on Weight, Blood Pressure. and Pulse Rate Baseline Body weight (kg)

Nadolol

74.4 + 3.2

74.4 + 3.4

Systolic blood pressure (mm Hg)

116 _f 3

104 * 3”

Diastolic blood pressure (mm Hg)

74 * 2

63 f 2*

Pulse rate (beats/min)

77 + 3

61 + lx

NOTE. All values are mean + SEM, based on the mean of the last 2 days before drug treatment and the last 2 days of nadolol treatment. All measurements were made in the morning after an overnight fast. *P < .Ol compared with baseline, paired r test.

BASELINE

NAOOLOL

Fig 1. RMR of five healthy young men before drug treatment (baseline) and during the last 5 days of nadolol treatment. Baseline value is mean of RMRs measured on days 2 and 3, and nadolol value is mean of RMRs measured on days 10.13, and 14.

B-ADRENERGIC BLOCKADE

621

Table 3. Comparison of RMRs of Nadolol-Treated

Subjects to

Subjects Hospitalized Without Drug Treatment Propranolol-Treated

Subjects

Group

Treated

No drug (n = 6) Propranolol

1.13 k 0.08

(n = 8)

Nadolol (n = 5) NOTE.

Values are kcalimin,

and

6 to 7 Days

1.13 2 0.07

1.28 ? 0.03

1.17 + 0.03*

1.15 2 0.03

1.07 + 0.03*

mean + SEM.

lf < .05 by paired t test compared

with value before drug treatment.

the decrease in RMR associated with P-adrenergic blockade is mediated by reduced T, concentrations. ANOVA of the increments in RMR each hour after the standard meal (Table 4) showed no significant effect of nadolol (P > .20). Three subjects had small decreases in the integrated (0 to 3 hour) response to the meal (0.05 to 0.09 kcal/min) after nadolol, and two subjects had small increases (0.01 and 0.03 kcal/min). The mean effect was a 12% decrease in the thermic response to the meal, but the 95% confidence interval for this effect was wide (35% decrease to 11% increase). The mean postprandial RMR (baseline plus thermic effect of meal) was significantly lower (mean, 8%) after nadolol treatment (Table 4). DISCUSSION

P-Adrenergic blockade with nadolol was associated with a 7% reduction in RMR in healthy men. This effect is similar to the 9% reduction in RMR previously observed with propranolol treatment, even though nadolol did not reduce serum T, concentrations as propranolol did.” Hence, the effect of propranolol on RMR appears to be related primarily to l3-adrenergic inhibition, rather than inhibition of T, production. This conclusion is consistent with the finding that iopanoic acid, which inhibits T, production to a similar extent as propranolol, does not reduce RMR in humans.‘” Not all investigators have observed an effect of l3-adrenergic blockade on RMR in humans. Zwillich et al” found no change in RMR after giving propranoiol (80 mg every 6 hours) for 3 days, but the figures in their report suggested a small reduction in mean Oz consumption and CO1 production (exact values were not given). Acheson et al” found no significant effect of 2 weeks of propranolol treatment (160 mg/d as a single dose each morning) on 24-hour energy expenditure in healthy men (RMR values were not reported). The mean decrease in total energy expenditure during propranolol treatment in their study was 64 kcal/d. while the average reduction in RMR with nadolol treatTable 4. Effect of Nadolol on Postprandial RMR Pm-drug

RMR before meal First-hour

increase

Second-hour

1.15 2 0.03

1.07 + 0.03’

0.23 2 0.03

0.20 + 0.03

0.29 2 0.05

0.25 -f 0.02

0.27 -r 0.04

0.23 + 0.03

Mean increase (3 h)

0.26 5 0.04

0.23 + 0.02

Postprandial

1.41 ? 0.04

1.30 + 0.03*

Third-hour

NOTE.

increase

Nadolol

increase RMR

Values are kcal/min, mean + SEM.

lP < .Ol compared

with pre-drug, paired t test.

ment in the present study was 100 to 115 kcal/d, and the average reduction in RMR during propranolol treatment in our previous study” was approximately 150 kcal/d. The role of the sympathetic nervous system in mediating the postprandial increase in RMR (the thermic effect of food) is unclear. Although P-adrenergic blockade reduces the thermic response to peripheral glucose plus insulin infusions,h.R,” these infusions do not mimic the natural metabolic events that follow a mixed meal. When pure carbohydrate meals are given orally (100 to 250 g carbohydrate), propranolol does not significantly inhibit the thermic response.3,” Vernet et al9 found no consistent effect of intravenous (IV) propranolol on the thermic response to an IV nutrient mixture in lean and obese women, although there was a slight trend toward a lower response. However, we previously reported that IV propranolol inhibited the thermic response to an oral mixed meal (the same meal used in the present study) an average of 22% over the first 2 hours,jh and Astrup et al” found that the thermic response to a carbohydrate-rich mixed meal was 26% lower during propranolol infusion. Schwartz et al” found that the thermic response during the first 2 hours after a similar meal was reduced an average of 33% by clonidine, which reduced the increase in norepinephrine appearance rate (an index of sympathetic nervous system activity) by 70%. Tappy et alz4 found that IV propranolol reduced the thermic response to a 75-g fructose meal an average of 40%. In the present study, P-adrenergic blockade with nadolol reduced the mean thermic response to a mixed meal an average of 12%, an effect that was not statistically significant. However, the 95% confidence interval for the effect of nadolol on the thermic effect of food did not exclude a reduction of up to 35%. Examination of the above studies shows no obvious relationship between meal composition or route of delivery and the effect of P-blockers on the thermic effect of nutrients. The main problem seems to be that the variability of the thermic effect of food is large, making it difficult to detect small changes. For example, in the present study, we would have had to include approximately 25 to 30 subjects to have an 80% chance of defining a 12% decrease in the thermic effect of food (the mean change we observed) as statistically significant at P < .05. Nevertheless, we can be fairly confident that P-adrenergic stimulation does not account for more than about one third of the total thermic response to food. It should be noted that the important value from the standpoint of the energy balance of these subjects is not the difference between basal and postparandial RMR, but rather the absolute postprandial RMR. This latter value was significantly reduced (8%) during nadolol treatment. The implication of these results for regulation of body weight in patients being treated with @blockers is uncertain. RMR accounts for 50% to 70% of the total energy expenditure of adult humans.‘5.‘h Thus, small differences in RMR maintained over prolonged periods could cause major changes in energy stores. Although a reduction in energy expenditure should theoretically lead to a compensatory reduction in food intake that would prevent weight gain, Ravussin et al” have demonstrated that subjects with

WELLE, SCHWARTZ, AND STATT

622

a low RMR have an increased risk of excessive weight gain. The difference in RMR between subjects with low and normal RMRs in their study” was generally smaller than the effect of P-blockade reported here. As pointed out by Acheson et al,” it is difficult to interpret the effects of p-blockers on body weight by surveying patients being treated with P-blockers, because improvement or deteriora-

tion of the disease being treated may alter body weight. A long-term study of the effects of P-blockers on energy balance is needed to establish the clinical significance of the present results. ACKNOWLEDGMENT

We thank the staff of the CRC for their technical assistance.

REFERENCES 1. Foster DO, Frydman ML: Nonshivering thermogenesis in the rat. Can J Physiol Pharmacol56:110-122,1978 2. Rothwell NJ, Stock MJ: A role for brown adipose tissue in diet-induced thermogenesis. Nature 281:31-35,1979 3. Seaton T, Welle S, Alex S, et al: The effect of adrenergic blockade on glucose-induced thermogenesis. Metabolism 33:415419,1984 4. Staten MA, Matthews DE, Cryer PE. et al: Physiological increments in epinephrine stimulate metabolic rate in humans. Am J Physiol253:E322-E330,1987 5. Welle SL, Thompson DA, Campbell RG: Beta-adrenergic blockade inhibits thermogenesis and lipolysis during glucoprovation in humans. Am J Physiol243:R379-R382, 1982 6. Acheson K, Jequier E, Wahren J: Influence of betaadrenergic blockade on glucose-induced thermogenesis in man. J Clin Invest 72:981-986, 1983 7. Christin L, Ravussin E, Bogardus C, et al: The effect of propranolol on free fatty acid mobilization and resting metabolic rate. Metabolism 38:439-444,1989 8. DeFronzo RA, Thorin D, Felber JP, et al: Effect of beta and alpha adrenergic blockade on glucose-induced thermogenesis in man. J Clin Invest 73:633-639, 1984 9. Vernet 0, Nacht C-A, Christin L, et al: Beta-adrenergic blockade and intravenous nutrient-induced thermogenesis in lean and obese women. Am J Physiol253:E65-E71, 1987 10. Jung RT, Shetty PS, James WPT: The effect of betaadrenergic blockade on metabolic rate and peripheral thyroid metabolism in obesity. Eur J Clin Invest 10:179-182,198O 11. Welle SL, Nair KS. Campbell RG: Failure of chronic beta-adrenergic blockade to inhibit overfeeding-induced thermogenesis in humans. Am J Physiol256:R653-R658, 1989 12. Frishman WH: Beta-adrenoceptor antagonists: New drugs and new indications. N Engl J Med 305:500-506,198l 13. Shulkin BL, Peele ME, Utiger RD: Beta-adrenergic antagonist inhibition of hepatic 3,5,3’-triiodothyronine production. Endocrinology 115:858-861, 1984 14. Berne C, Fagius J, Nikiasson F: Sympathetic response to oral carbohydrate administration. J Clin Invest 84:1403-1409, 1989 15. Schwartz RS, Jaeger LF, Veith RC: Effect of clonidine on

the thermic R94, 1988 16. Welle carbohydrate

effect of feeding S, Campbell overfeeding.

in humans.

Am J Physiol

254:R90-

RG: Stimulation of thermogenesis J Clin Invest 71:916-925, 1983

by

17. Astrup A, Simonsen L, Bulow J, et al: Epinephrine mediates facultative carbohydrate-induced thermogenesis in human skeletal muscle. Am J Physiol257:E340-E345, 1989 18. Schwartz RS, Jaeger LF, Silberstein S, et al: Sympathetic nervous system activity and the thermic effect of feeding in man. Int J Obesity 11:141-149, 1986 19. Forbes GB, Schultz F, Cafarelli C, et al: Effects of body size on potassium-40 measurement in the whole body counter. Health Phys 151435-442. 1968 20. Acheson KJ. Burger AG: A study of the relationship between thermogenesis and thyroid hormones. J Clin Endocrinol Metab 51:84-89. 1980 21. Zwillich C, Martin B. Hofeldt F. et al: Lack of effects of beta sympathetic blockade on the metabolic and respiratory responses to carbohydrate feeding. Metabolism 30:451-455. 1981 22. 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 23. Acheson KJ, Ravussin E, Wahren J, et al: Thermic glucose in man. J Clin Invest 74:1572-1580. 1984

effect of

24. Tappy L, Randin J-P. Felber J-P, et al: Comparison of thermogenic effect of fructose and glucose in normal humans. Am J Physiol250:E718-E724. 1986 25. Prentice AM, Coward WA. Davies HL. et al: Unexpectedly low levels of energy expenditure in healthy women. Lancet 1:14191422.1985 26. Schulz S, Westerterp KR, Bruck K: Comparison of energy expenditure by the doubly labeled water technique with energy intake, heart rate, and activity recording in man. Am J Clin Nutr 49:1146-1154.1989 27. Ravussin E, Lillioja S, Knowler WC, et al: Reduced rate of energy expenditure as a risk factor for body-weight gain. N Engl J Med 318:467-472, 1988