- Email: [email protected]
Cardiorespiratory response to exercise before and after acute beta-adrenoreceptor blockade in nonsmokers and chronic smokers
InternutronalJournal Elsevier
of Cardtologv, 11 (1986) 293-304
293
IJC 00400
Cardiorespiratory response to exercise before and after acute beta-adrenoreceptor blockade in nonsmokers and chronic smokers W.J. Penny
and M.A.
Mir
Departments of Medicrne and Cardiology, Unioersify of Wales College of Medicine, Card$ (Received
30 September
1985; revision accepted
17 January
U.K.
1986)
Penny WJ, Mir MA. Cardiorespiratory response to exercise before and after acute beta-adrenoreceptor blockade in nonsmokers and chronic smokers. Int J Cardiol 1986;11:293-304. To evaluate the effects of chronic smoking on exercise performance we studied 5 smokers and 7 nonsmokers of comparable age and physical characteristics. The resting heart rate in smokers (75 f 3 beats/min; mean f SD) was significantly (P < 0.01) higher than in nonsmokers (64 f 5). During exercise on a bicycle ergometer the heart rate remained significantly (P < 0.01) higher in smokers than in nonsmokers. After exercise, the heart rate in nonsmokers settled to 78 f 9 beats / min at 10 minutes compared with 105 f 11 (P < 0.01) in smokers. Oxygen consumption was similar in both groups throughout. Beta-adrenergic blockade reduced the exercise tachycardia in both groups but the heart rate for the same workload remained significantly (P < 0.01) higher in smokers. Beta-blockade significantly reduced (P < 0.05) oxygen consumption in nonsmokers but not in smokers who also incurred a significantly (P < 0.05) greater oxygen debt and had higher serum lactate levels. These differences were attributed mainly to carboxyhaemoglobinaemia and partly to the effect of prolonged smoking on the heart and on intermediary metabolism. (Key words:
carboxyhaemoglobin;
lactate;
oxygen consumption;
oxylog)
Introduction Numerous studies have shown a strong incriminating relationship between cigarette smoking and acute myocardial infarction and sudden death in patients
Correspondence to: M.A. Mir. Department Heath Park. Cardiff CF4 4XN, U.K.
0167-5273/86/$03.50
of Medicine,
G 1986 Elsevier Science Publishers
University
B.V. (Biomedical
of Wales College
Division)
of Medicine,
294
with ischaemic heart disease [l-4]. Surprisingly most of the studies investigating the haemodynamic effects of smoking have concentrated on the acute events immediately after smoking a cigarette [5-121. Notable among the few studies comparing the cardiorespiratory responses to exercise between smokers and nonsmokers have been the ones by Blackburn et al. [13] and by Chevalier et al. [14]. However, both these studies were carried out before the advent of beta-adrenoreceptor blocking agents, which have made an important contribution to our understanding of the haemodynamic responses to exercise, and are now an essential part of the medical management of patients with ischaemic heart disease. The purpose of this study was to investigate the chronic effects of cigarette smoking on heart rate and oxygen consumption during exercise, and on the effects of beta-blockade before and after exercise in a group of young healthy smokers and nonsmokers. Methods Subjects
Twelve sedentary male volunteers (medical students and technical staff) were studied. There were 5 smokers and 7 nonsmokers of similar age, exercise habits and physical characteristics (Table 1). The smokers had smoked at least 20 cigarettes daily for over 5 years and none of the subjects in either group had any previous medical history. All subjects gave informed consent. Clinical examination, full blood count, serum urea and electrolytes, serum thyroxine and a resting electrocardiogram were all normal. Exercise Tests
All subjects were fasting and smokers had not smoked for at least 6 hours before the study. A graded 8-min exercise test was performed on a bicycle ergometer (Monark, Sweden). Each subject pedalled at 50 revolutions/mm in time with a metronome and the resistance was increased by 300 kilopondmetres (kpm) every 2 min. Each subject had 5 exercise tests over a 7-day period. During the first three he was familiarized with the equipment and urged to reach his maximal exercise
TABLE
1
Physical
characteristics
of 12 male subjects Nonsmokers
Age (yr) Height (cm) Weight (kg) Body surface area (m’)
28.9 +5 177.3 +7.7 72.5 k9.2 1.89kO.16
in the two groups. (n = 7)
Results
are expressed
Smokers 28.5 174.8 72.5 1.87
(n = 5)
rt6 +4.4 +7.1 + 0.10
as mean
+ 1 SD.
295
performance. Exercise to fatigue at this maximal workload (3-4 min) was taken as the endpoint during the fourth and fifth tests. All subjects were randomised to receive either propranolol (0.3 mg/kg in 200 ml of 5% dextrose intravenously) or placebo (200 ml of 5% dextrose) before the fourth and fifth exercise tests. On each occasion the infusion was given over 20 min, and the subjects did not know whether they were receiving propranolol or placebo (single-blind). Basal measurements of oxygen consumption, minute ventilation, pulse rate, blood pressure, 12-lead electrocardiogram, forced expiratory volume in one second and forced vital capacity (Vitalograph spirometer) and peak expiratory flow rate (Wright’s peak flow meter) were recorded in that order in a sitting position, after the subject had rested for 1 hour in the laboratory. This routine ensured that there was little disturbance during the recording of oxygen consumption, pulse and blood pressure so that the resting values could be obtained. An infusion of propranolol or placebo was given and after 1 hour the above measurements were again recorded. A venous blood sample was taken for the estimation of serum lactate concentration [15] and then the exercise was started. During exercise the oxygen consumption was measured every minute (see below) and the electrocardiogram was monitored continuously (Hewlett Packard stress system 1505A). The electrocardioagram was recorded for 15 set at the end of each minute and the heart rate was obtained from the printed record. The systolic blood pressure was recorded with a sphygmomanometer during the last minute of the exercise. After exercise the oxygen consumption and ventilation were measured every minute for 10 min. A further blood sample was obtained for lactate estimation during the last minute of exercise. Oxygen consumption and ventilation were measured with a portable instrument Oxylog (PK Morgan Ltd., Chatham, Kent). The Oxylog assembly consisted of a mask fitted with a flowmeter and an analyser which provided a digital display of cumulative ventilation and oxygen consumption and the consumption during the preceding minute. The analyser was connected to the mask by means of a flexible air hose which conducted the expired air. The volume of inspired air was measured by the flow meter and the expired air was sampled as it passed through the hose past the analyser. The samples of inspired air and expired air were passed through drying tubes containing calcium phosphate and then passed through two stable polarographic oxygen sensors (Beckman No. 148373). Oxygen consumption was calculated from the partial pressures of inspired and expired air with a first order correction for changes in barometric pressure, temperature and an assumption of 50% relative humidity in the inspired air. The Oxylog is suitable for ventilation volumes of 6-80 l/min [16] and its accuracy has been shown to compare well (r = 0.9) with oxygen consumption as measured using a mass spectrometer [17]. It was not possible to record the pulmonary function tests simultaneously with the oxygen consumption. The exercise was repeated after a rest period of 1 hour and the peak exercise pulmonary function tests were measured. All subjects voided 2 hours before the exercise test and resting urinary catecholamine concentrations were measured as 4-hydroxy-3-methoxymandelic acid by the method of Van de Calseyde et al. [18] in the subsequent 2-hour collection before each test.
296
Statistical Methods
Standard statistical methods were used. The data from the two groups were compared using Student’s unpaired t-test. Student’s paired z-test was used to determine the significance of differences within the same group with and without beta-blockade. P < 0.05was taken to denote statistical significance. All data are presented as means f SD. Results The physical characteristics of the subjects in both groups were similar (Table 1). Table 2 summarises the results of the pulmonary function tests in the two groups at rest and before and after beta-blockade. At rest the mean values were within the normal range and there was no significant difference between the two groups. Beta-blockade produced a slight and statistically insignificant reduction in the forced expiratory volume in 1 set, forced vital capacity and peak expiratory flow rate in smokers. The resting ventilation rate was lower (P < 0.05) in nonsmokers after beta-blockade than in smokers (Table 2). Urinary catecholamines as 4-hydroxy-3-methoxymandelic acid were measured before exercise in 4 smokers and 4 nonsmokers. The mean level of 3.8 f 0.93 pmol/l in nonsmokers was not significantly different from 4.6 f 1.1 pmol 4-hydroxy-3-methoxymandelic acid/l in smokers. Exercise Data
Subjects in both groups achieved the same total maximal workload (5400 kilopondmetres) on the bicycle ergometer after a training period of 3 tests; non-
TABLE
2
Pulmonary function data of smokers and nonsmokers intravenously). Data expressed as means f 1 SD. Nonsmokers
after placebo
(n = 7)
and after propranolol
Smokers
(0.3 mg/kg
(n = 5)
Propranolol
Placebo
Vital capacity
5200
+320
5100
+300
5000
+700
4800
f 700
(ml) Forced
4400
+ 600
4400
*400
4000
+ 600
3800
&500
637
+133
Placebo
expiration
Propranolol
vol. in 1 set (ml) Peak expiratory flow rate (l/min) Ventilation
646
f141
7.3&
0.9
(l/fin) * P i 0.025 as compared
with nonsmokers.
6.6k
583
0.5
k
8.9+
68
2.1
558
f
8.16k
56
1.5*
291 T
TT
p--p4ol T _l_.X_
I
I
,
poo.001
~3 Smokers after A Non smokers = Non vrokers
beta blockade after beta blockade
4 1200
2400
3600
6000
4000
Work (hpm)
Fig. 1. Relation of heart rate to cumulative work load in kilopondmetres (kpm) before and after beta-blockade in healthy smokers and nonsmokers. Beta-blockade was achieved by the administration of propranolol 0.3 mg/kg body weight intravenously 1 hour before the exercise test. At comparable workloads smokers showed a significantly higher heart rate than nonsmokers before and after beta-blockade. All data mean +l SD.
smokers achieved this workload in the first test and their heart rate became linear with the increments in workload by the second test. However, the smokers found it difficult to match the exercise performance of nonsmokers until the third test; their
6000
smokers
2700
+eoo
2400 2 E E .s ‘i
5 0
s
2100
‘Ei
‘800
3600
2 ‘p H
2400
$ I
900 600
1200
300
6246824
6
8
IO
Minutes During
exercise
Post-exercise
Fig. 2. Effects of exercise on oxygen consumption in smokers and nonsmokers. Work load is cumulative increase in kilopondmetres (kpm) each minute. Resting oxygen consumption was smokers (420 ml/mm) than in nonsmokers (290 ml/mm) and it remained slightly higher increment in the workload. During the post-exercise recovery period the oxygen consumption higher (but not significantly) in smokers. Data expressed as mean f 1 SD.
shown as higher in for each remained
3
+9
rt5
+8
&6
+6
+7
4.4 +0.51
1.37kO.81 3.85 f 0.98
0.46
3.93*
0.93* 0.15 7.0 + 0.35
0.28 + 0.02 2.40 k 0.20
118
114 157
63
57 125
Propranolol
0.08 0.20
*6
f 5 *17
*9
+5 +7
(n = 7)
0.29+ 2.78&
149
114 186
78
64 157
Placebo
Nonsmokers
1.13+ 6.671
3.79+
0.25 0.99
0.68
0.20 0.50
* 15
f 5 *21*
+11**
C 0.05
n.s. < 0.01
(n = 5)
f 3** +11**
0.421 2.96$
163
118 220
105
75 176
Placebo
ns. < 0.05
< 0.001
n.s. ‘z 0.001
-=z0.01
n.s. < 0.01
Significance P (paired-t)
Smokers
k28
*7 k36
+
0.54*
1.15f 0.85 5.8 + 0.28*
5.1 f
0.02 0.40**
9**
f6’ f 5**
0.25f 2.85 i
146
110 183
86
69 136
Propranolol
systolic blood pressure (mm Hg), oxygen consumption (I/min) and serum lactate concentration (mmol/l) in smokers and nonsmokers before and after beta-blockade (propranolol 0.3 mg/kg intravenously).
* P < 0.05; **p < 0.01.
(5400 kpm)
(Vmin) Resting Peak exercise (5400 kpm) Oxygen debt Serum lactate (mmol/l) Resting At exercise
(5400 kpm) Post-exercise (1 min) Oxygen consumption
(5400 kpm) Post exercise (10 min) Systolic blood pressure (mm Hg) Resting Peak exercise
Heart rate (beats/min) Resting Peak exercise
Heart rate (beats/min), (5400 kilopondmetres)
TABLE
n.s. ‘C 0.05
< 0.001
< 0.05 n.s.
ns.
n.s. ns.
< 0.01
< 0.05 < 0.001
Significance P (paired-t)
at rest and at peak exercise
%
299
oxygen consumption became linear with the increments in workload by the fourth test, but their heart rate was significantly greater than nonsmokers at rest and at all increments in workload during the exercise test (Fig. 1; Table 3). Propranolol reduced the heart rate significantly at all levels of exercise in both groups but smokers continued to have a relative tachycardia even after beta-blockade when compared with nonsmokers (P < 0.01). Similarly, the mean systolic blood pressure at peak exercise during placebo was significantly (P < 0.05) higher in smokers (220 + 21 mm Hg) than in nonsmokers (186 + 17 mm Hg). Furthermore, beta-blockade significantly reduced the peak- and post-exercise systolic blood pressure in nonsmokers, but had only a slight and statistically insignificant effect in smokers (Table 3). The resting oxygen consumption was higher in smokers than in nonsmokers but there was a large scatter among smokers and the difference was not statistically significant. With the stepwise increase in workload, the oxygen consumption increased linearly in both groups; smokers had a slightly higher oxygen consumption than nonsmokers but the difference was not statistically significant (Fig. 2). During the recovery period oxygen consumption remained slightly but not significantly higher in smokers. Effects of Beta-Blockade
in Oxygen Consumption
There was no significant difference in the oxygen debt incurred after placebo by the two groups. With beta-blockade, there was a reduction in the oxygen consumption at each workload in nonsmokers; at the maximum workload oxygen consumption fell from 2.78 f 0.20 l/min to 2.40 f 0.20 l/min (P < 0.05). This was associated with a small increase in oxygen debt, from 3.93 f 0.46 to 4.40 + 0.5 1 (P = 0.05, Table 3). However, in smokers there was no change in oxygen consumption with beta-blockade; at the maximal load oxygen consumption was 2.96 f 0.50 I/min before and 2.85 _t 0.40 l/min after beta-blockade. There was a significant increase in oxygen debt, from 3.79 + 0.68 to 5.10 + 0.54 1 (P < 0.001). After propranolol the oxygen debt was significantly greater in smokers than in nonsmokers (5.10 + 0.54 vs. 4.40 + 0.501; p < 0.05). These differences in oxygen debt were also reflected in the serum lactate levels at peak exercise which were significantly (P < 0.05) higher in smokers than in nonsmokers (Fig. 3). In both groups the serum lactate level at peak exercise was significantly lower after beta-blockade (Table 3). Post-Exercise
Data
Throughout the recovery period, the heart rate remained significantly higher in smokers. At the tenth minute after exercise, the heart rate was 105 ~fr11 in smokers compared with 78 f 9 beats/mm (P -c0.001) in nonsmokers. As would be expected beta-blockade significantly reduced the heart rate in the recovery period in both groups, but in smokers the rate remained significantly (P < 0.01) higher 10 min after exercise than in nonsmokers (Table 3).
300 Placebo Propranolol
0.3mg/Kg
I .V
r P10.05
1
Mean+_ S
.D
1
Serum (mmole/L)
smokers
(n=5)
Fig. 3. Serum lactate and oxygen debt before and after beta-blockade in smokers and nonsmokers. Beta-blockade increased significantly (P -C 0.05) oxygen debt in smokers but not in nonsmokers and reduced significantly (P < 0.01) serum lactate in nonsmokers but not in smokers. Smokers had significantly (P < 0.05) higher oxygen debt and serum lactate after beta-blockade as compared with nonsmokers.
301
Discussion The smokers and nonsmokers in this study had similar physical characteristics and all followed a sedentary life style. Both groups had similar resting urinary catecholamines, and achieved the same maximal workload on the bicycle ergometer, with comparable oxygen consumption and oxygen debt. Despite these similarities smokers had a significantly faster heart rate at rest and at exercise than nonsmokers. Although smokers followed similar exercise pursuits as their nonsmoker colleagues, they required greater motivation and longer conditioning to achieve the same workload as nonsmokers. This greater motivation and effort made by smokers may in part explain their higher heart rate and systolic blood pressure, their partial insensitivity to beta-blockade and their greater oxygen debt. Smoking is known to activate hepatic enzymes, and orally taken propranolol is cleared faster in smokers than in nonsmokers [19]. Propranolol was given intravenously in large doses, and the study was completed within 90 min to minimise any possible pharmacokinetic variations in smokers. Propranolol reduced the heart rate in both groups but less so in smokers who still had a relative tachycardia. It reduced the oxygen consumption on exercise in nonsmokers but not in smokers. After beta-blockade smokers had a higher oxygen debt (P < 0.001) and significantly (P < 0.05) higher lactate levels at peak exercise than nonsmokers. In this study the lactate level was measured at the peak of exercise in both groups and it is possible that the peak level, which may occur at a variable time after exercise [20] was missed. These cardiorespiratory and metabolic differences cannot be attributed to a faster clearance of propranolol or to a higher adrenergic drive in smokers. Hepatic clearance of intravenously administered propranolol would mostly depend on the hepatic blood flow, which is not altered by smoking [21,22]. Even though the intrinsic hepatic clearance of propranolol may have been accelerated in our smokers, it is unlikely that the serum drug level would have been significantly lowered within 60 min of intravenous administration of propranolol to account for the relative tachycardia observed during and after exercise. Tachycardia after smoking [8,11,14,23] has been attributed to the effects of nicotine [ 11,241 and to hypoxaemia caused by carboxyhaemoglobinaemia [25]. Inhalation of carbon monoxide in nonsmoking volunteers to raise the carboxyhaemoglobin concentration to the levels seen in smokers produces similar haemodynamic effects as we have described [26]. However, since the half-life of carboxyhaemoglobin is about 4 hours [27] it cannot fully explain the relative tachycardia in our group of smokers who had not smoked for over 6 hours. There are conflicting reports of higher catecholamine levels in smokers [28] but this persistence of a relative tachycardia in smokers after beta-blockade suggests that it is not mediated wholly through excess adrenergic stimulation. Cryer et al. [12] found that the haemodynamic effects of smoking preceded and persisted long after the transient rise in catecholamines which follows smoking, whereas Strasser et al. [28] found no change in catecholamine levels after smoking. In this study there was no difference in the resting urinary catecholamine levels in the two groups. The effects of chronic cigarette smoking on the myocardium are unknown. In a study on the effect of
302
hypoxaemia on the electrocardiogram in patients with car pulmonale, Tirlapur and Mir [29] showed that tachycardia and a prolonged QT interval were reversed by oxygen in nonsmokers but not in smokers. Thus no single factor can explain the higher rate in smokers: hypoxaemia due to carboxyhaemoglobin, excessive adrenergic drive and a direct effect on the myocardium probably all contribute. Resting oxygen consumption was higher in smokers but there was a wide scatter in this group as has been observed by Krone et al. [30], and the differences did not reach statistical significance. Oxygen uptake on exercise and oxygen debt were not significantly different in the two groups. Chevalier et al. [7] observed a higher oxygen debt in smokers but these workers were unable to provide any supporting evidence. In this study the lack of any significant difference in the oxygen debt between smokers and nonsmokers was reflected in their similar serum lactate levels. Although myocardial oxygen usage forms only a minor part of the total oxygen consumption, a greater myocardial oxygen expenditure in smokers, caused by higher systolic blood pressure and heart rate, must have contributed to their greater oxygen consumption and debt. Beta-blockade revealed complex differences between the two groups. Propranolol reduced the resting oxygen consumption in both groups, though during exercise it reduced oxygen consumption only in nonsmokers. Previous studies have demonstrated this effect in nonsmokers [31-341 but a lack of this effect in smokers has not been previously described, and explains why studies on mixed groups have given varying results [35]. After propranolol smokers had a greater oxygen debt than nonsmokers. This difference is also reflected in the lactate levels at peak exercise. Beta-blockade reduced the lactate level in both groups, as has been previously described in nonsmokers [36], but this reduction was significantly (P -c0.05) less in smokers. This fall in oxygen consumption and a reduction in lactate levels in nonsmokers are suggestive of an increased efficiency after beta-blockade. In contrast smokers after beta-blockade had significantly higher heart rates during exercise, higher oxygen debt, higher ratio of oxygen debt to increased oxygen uptake and higher lactate levels than nonsmokers. While beta-blockers significantly reduced systolic blood pressure during and after exercise in nonsmokers, it failed to have any significant effect on smokers. These findings suggest that smokers were doing the same work after beta-blockade for a higher haemodynamic and metabolic cost than nonsmokers. While the underlying mechanisms responsible for these differences between smokers and nonsmokers before and after beta-blockade are not known at present, these results have two major implications. First, during exercise testing heart rate as an index of work done is an unreliable guide in smokers who have a higher heart rate even after full physical conditioning, than nonsmokers for the same work output. Exercise data should be interpreted with caution in a mixed population of smokers and nonsmokers. Secondly, since beta-blockade does not fully suppress exercise and resting tachycardia in smokers, its full benefit cannot be realised in smokers. The differences we have described help explain the difficulties experienced by some workers [37] in effectively treating angina pectoris in patients who continue to smoke.
303
References 1 Doll R. Hill AB. Mortality
in relation
J 1964;l: 139991410. 2 Doyle JT, Dawber TR, Kannel 3 4 5 6 7 8 9 10
11
12
13 14 15 16 17 18 19 20 21 22 23
24 25
to smoking:
WB, Kinch
ten years’ observations
SH. Kahn
of British doctors.
HA. The relationship
of cigarette
Br Med
smoking
to
coronary heart disease. J Am Med Assoc 1964:190: 886-890. Frank CW, Weinblatt E, Shapiro S, Sager RV. Myocardial infarction in men: role of physical activity and smoking in incidence and mortality. J Am Med Assoc 1966;198: 1241-1245. Ball K, Turner R. Smoking and the heart: the basis for action. Lancet 1974;2: 822-826. Armitage AK, Hall GH. Mode of action of intravenous nicotine in causing a fall of blood pressure in the cat. Eur J Pharmacol 1969:7: 23-30. Cellina GU, Honour AJ, Littler WA. Direct arterial pressure, heart rate. and electrocardiogram during cigarette smoking in unrestricted patients. Am Heart J 1974:89: 18-25. Chevalier RB, Krumholz RA, Ross JC. Reaction of non-smokers to carbon monoxide inhalation: cardiopulmonary responses at rest and during exercise. J Am Med Assoc 1966;198 : 1061-1064. Aronow WS. Goldsmith JR. Kern JC, Johnson LL. Effect of smoking cigarettes on cardiovascular haemodynamics. Arch Environ Health 1974;28 : 330-332. Goldberg AN, Krone RJ, Resnekov L. Effects of cigarette smoking on haemodynamics at rest and during exercise. 1. Normal subjects. Chest 1971;60: 531-536. Ayres SM. Mueller HS. Gregory JJ. Giannelli S, Penny JL. Systemic and myocardial haemodynamic responses to relatively small concentrations of carboxyhemoglobin (CoHB). Arch Environ Health 1969:18:699-709. Aronow WS. Dendinger J, Rokaw SN. Heart rate and carbon monoxide level after smoking high-, low-, and non-nicotine cigarettes. A study in male patients with angina pectoris. Ann Intern Med 1971;74:697-702. Cryer PE, Haymond MW, Santiago JV, Shah SD. Norepinephrine and epinephrine release and adrenergic mediation of smoking-associated haemodynamic and metabolic events. N Engl J Med 1976:295 : 573-517. Blackburn HJ, Brozek J. Taylor HL. Common circulatory measurements in smokers and nonsmokers. Circulation 1960;22: 1112-1124. Chevalier RB. Bowers JA, Bondurant S, Ross JC. Circulatory and ventilatory effects of exercise in smokers and nonsmokers. J Appl Physiol 1963:lS: 357-360. Leese HJ. Bronk JR. Automated fluorometric analysis of micromolar quantities of ATP. glucose and lactic acid. Anal Biochem 1972;45 : 211-221. Humphrey, SJE, Wolff HS. The Oxylog. J Physiol (London) 1977;267 : 1-12. Belyarin AJ. Brown GA, Harrison MH. The Oxylog: an evaluation. Royal Air Force Institute of Aviation Medicine IAM Report No. 608, 1981. Van de Calseyde JF, Scholtis RJH. Schmidt NA. Leyton CJJA. Gas chromatography in the estimation of urinary metanephrines and VMA. Clin Chim Acta 1971:32 : 361-366. Wood AJJ. Vestal RE, Branch RA. Wilkinson GR, Shand DG. Effect of aging and cigarette smoking on antipyrine and indocyanine green elimination. Pharmacol Ther 1979:26 : 16-20. Astrand PO, Rodahl K. Textbook of work physiology. New York: McGraw-Hill. 1970. Vestal RE, Wood AJJ. Branch RA, Shand DG, Wilkinson GR. The effects of aging and cigarette smoking on propranolol’s disposition. Clin Pharmacol Ther 1979~26: 8-15. Nies AS. Wilkinson GR, Rush BD, Strother JT, McDevitt DG. Effects of alteration of hepatic microsomal enzyme activity on liver blood flow in the rat. Biochem Pharmacol 1976;25 : 1491-1994. Thomas CB, Bateman JL. Lindberg EF, Bronhold HJ. Observation on the individual effects of smoking on the blood pressure, heart rate, stroke volume, and cardiac output of healthy young adults. Ann Intern Med 1956:44: 874-892. Aronow WS, Kaplan MA. Jacob D. Tobacco: a precipitating factor in angina pectoris. Ann Intern Med 1968;69 : 529-536. Asmussen E. Chiodi H. The effects of hypoxaemia on ventilation and circulation in man. Am J Physiol 1941;132 : 4266436.
304 26 Chevalier RB, Krumholtz RA, Ross JC. Reaction of nonsmokers to carbon monoxide inhalation, cardiopulmonary responses at rest and during exercise. J Am Med Assoc 7966;198 : 1061-1064. 27 Shephard RJ. Smoking withdrawal and changes of cardiorespiratory fitness. Am Rev Respir Dis 1971:104:933-935. 28 Strasser R, Dietz R, Schomig A, Rascher W, Kubler W. Sympathetic activity during smoking before and after beta-blockade (abstract). Eur Heart J 1981;2: 203. 29 Tirlapur VG. Mir MA. Nocturnal hypoxaemia and associated electrocardiographic changes in patients with chronic obstructive airways disease. N Engl J Med 1982;306 : 125-130. 30 Krone RJ, Goldberg AN, Balkoura M. Schuessler R. Resnekow L. Effects of cigarette smoking at rest and during exercise. II. Role of venous return. J Appl Physiol 1972;32 : 745-748. 31 Schroder G, Werko L. Haemodynamic studies and clinical experience with nethalide. a beta adrenergic blocking agent. Am .I Cardiol 1965;15 : 58-65. 32 Astrom H. Haemodynamic effects of beta adrenergic blockade. Br Heart J 1968;30 : 44-49. 33 Parker JO, West RO, Di Giorgi S. Haemodynamic effects of propranolol in coronary heart disease. Am J Cardiol 1968;21: 11-19. 34 Sonnenblick EH, Braunwald E, Williams JF, Glick G. Effects of exercise on myocardial force velocity relations in intact unanaesthetised man. Relative roles of change in heart rate, sympathetic activity and ventricular dimensions. J Clin Invest 1965;44: 2051-2062. 35 Gibson DG. Pharmacodynamic properties of /I-adrenergic receptor blocking drugs in man. Drugs 1974:7:8-38. 36 Pillet AP, Bemand BV, Saunders RA. Effects of propranolol on blood sugar, insulin and free fatty acids. Diabetologia 1969;5 : 339. 37 Fox K, Johnathan A, Williams H, Selwyn A. Interaction between cigarettes and propranolol in treatment of angina pectoris. Br Med J 198O:l: 191-193.
of Cardtologv, 11 (1986) 293-304
293
IJC 00400
Cardiorespiratory response to exercise before and after acute beta-adrenoreceptor blockade in nonsmokers and chronic smokers W.J. Penny
and M.A.
Mir
Departments of Medicrne and Cardiology, Unioersify of Wales College of Medicine, Card$ (Received
30 September
1985; revision accepted
17 January
U.K.
1986)
Penny WJ, Mir MA. Cardiorespiratory response to exercise before and after acute beta-adrenoreceptor blockade in nonsmokers and chronic smokers. Int J Cardiol 1986;11:293-304. To evaluate the effects of chronic smoking on exercise performance we studied 5 smokers and 7 nonsmokers of comparable age and physical characteristics. The resting heart rate in smokers (75 f 3 beats/min; mean f SD) was significantly (P < 0.01) higher than in nonsmokers (64 f 5). During exercise on a bicycle ergometer the heart rate remained significantly (P < 0.01) higher in smokers than in nonsmokers. After exercise, the heart rate in nonsmokers settled to 78 f 9 beats / min at 10 minutes compared with 105 f 11 (P < 0.01) in smokers. Oxygen consumption was similar in both groups throughout. Beta-adrenergic blockade reduced the exercise tachycardia in both groups but the heart rate for the same workload remained significantly (P < 0.01) higher in smokers. Beta-blockade significantly reduced (P < 0.05) oxygen consumption in nonsmokers but not in smokers who also incurred a significantly (P < 0.05) greater oxygen debt and had higher serum lactate levels. These differences were attributed mainly to carboxyhaemoglobinaemia and partly to the effect of prolonged smoking on the heart and on intermediary metabolism. (Key words:
carboxyhaemoglobin;
lactate;
oxygen consumption;
oxylog)
Introduction Numerous studies have shown a strong incriminating relationship between cigarette smoking and acute myocardial infarction and sudden death in patients
Correspondence to: M.A. Mir. Department Heath Park. Cardiff CF4 4XN, U.K.
0167-5273/86/$03.50
of Medicine,
G 1986 Elsevier Science Publishers
University
B.V. (Biomedical
of Wales College
Division)
of Medicine,
294
with ischaemic heart disease [l-4]. Surprisingly most of the studies investigating the haemodynamic effects of smoking have concentrated on the acute events immediately after smoking a cigarette [5-121. Notable among the few studies comparing the cardiorespiratory responses to exercise between smokers and nonsmokers have been the ones by Blackburn et al. [13] and by Chevalier et al. [14]. However, both these studies were carried out before the advent of beta-adrenoreceptor blocking agents, which have made an important contribution to our understanding of the haemodynamic responses to exercise, and are now an essential part of the medical management of patients with ischaemic heart disease. The purpose of this study was to investigate the chronic effects of cigarette smoking on heart rate and oxygen consumption during exercise, and on the effects of beta-blockade before and after exercise in a group of young healthy smokers and nonsmokers. Methods Subjects
Twelve sedentary male volunteers (medical students and technical staff) were studied. There were 5 smokers and 7 nonsmokers of similar age, exercise habits and physical characteristics (Table 1). The smokers had smoked at least 20 cigarettes daily for over 5 years and none of the subjects in either group had any previous medical history. All subjects gave informed consent. Clinical examination, full blood count, serum urea and electrolytes, serum thyroxine and a resting electrocardiogram were all normal. Exercise Tests
All subjects were fasting and smokers had not smoked for at least 6 hours before the study. A graded 8-min exercise test was performed on a bicycle ergometer (Monark, Sweden). Each subject pedalled at 50 revolutions/mm in time with a metronome and the resistance was increased by 300 kilopondmetres (kpm) every 2 min. Each subject had 5 exercise tests over a 7-day period. During the first three he was familiarized with the equipment and urged to reach his maximal exercise
TABLE
1
Physical
characteristics
of 12 male subjects Nonsmokers
Age (yr) Height (cm) Weight (kg) Body surface area (m’)
28.9 +5 177.3 +7.7 72.5 k9.2 1.89kO.16
in the two groups. (n = 7)
Results
are expressed
Smokers 28.5 174.8 72.5 1.87
(n = 5)
rt6 +4.4 +7.1 + 0.10
as mean
+ 1 SD.
295
performance. Exercise to fatigue at this maximal workload (3-4 min) was taken as the endpoint during the fourth and fifth tests. All subjects were randomised to receive either propranolol (0.3 mg/kg in 200 ml of 5% dextrose intravenously) or placebo (200 ml of 5% dextrose) before the fourth and fifth exercise tests. On each occasion the infusion was given over 20 min, and the subjects did not know whether they were receiving propranolol or placebo (single-blind). Basal measurements of oxygen consumption, minute ventilation, pulse rate, blood pressure, 12-lead electrocardiogram, forced expiratory volume in one second and forced vital capacity (Vitalograph spirometer) and peak expiratory flow rate (Wright’s peak flow meter) were recorded in that order in a sitting position, after the subject had rested for 1 hour in the laboratory. This routine ensured that there was little disturbance during the recording of oxygen consumption, pulse and blood pressure so that the resting values could be obtained. An infusion of propranolol or placebo was given and after 1 hour the above measurements were again recorded. A venous blood sample was taken for the estimation of serum lactate concentration [15] and then the exercise was started. During exercise the oxygen consumption was measured every minute (see below) and the electrocardiogram was monitored continuously (Hewlett Packard stress system 1505A). The electrocardioagram was recorded for 15 set at the end of each minute and the heart rate was obtained from the printed record. The systolic blood pressure was recorded with a sphygmomanometer during the last minute of the exercise. After exercise the oxygen consumption and ventilation were measured every minute for 10 min. A further blood sample was obtained for lactate estimation during the last minute of exercise. Oxygen consumption and ventilation were measured with a portable instrument Oxylog (PK Morgan Ltd., Chatham, Kent). The Oxylog assembly consisted of a mask fitted with a flowmeter and an analyser which provided a digital display of cumulative ventilation and oxygen consumption and the consumption during the preceding minute. The analyser was connected to the mask by means of a flexible air hose which conducted the expired air. The volume of inspired air was measured by the flow meter and the expired air was sampled as it passed through the hose past the analyser. The samples of inspired air and expired air were passed through drying tubes containing calcium phosphate and then passed through two stable polarographic oxygen sensors (Beckman No. 148373). Oxygen consumption was calculated from the partial pressures of inspired and expired air with a first order correction for changes in barometric pressure, temperature and an assumption of 50% relative humidity in the inspired air. The Oxylog is suitable for ventilation volumes of 6-80 l/min [16] and its accuracy has been shown to compare well (r = 0.9) with oxygen consumption as measured using a mass spectrometer [17]. It was not possible to record the pulmonary function tests simultaneously with the oxygen consumption. The exercise was repeated after a rest period of 1 hour and the peak exercise pulmonary function tests were measured. All subjects voided 2 hours before the exercise test and resting urinary catecholamine concentrations were measured as 4-hydroxy-3-methoxymandelic acid by the method of Van de Calseyde et al. [18] in the subsequent 2-hour collection before each test.
296
Statistical Methods
Standard statistical methods were used. The data from the two groups were compared using Student’s unpaired t-test. Student’s paired z-test was used to determine the significance of differences within the same group with and without beta-blockade. P < 0.05was taken to denote statistical significance. All data are presented as means f SD. Results The physical characteristics of the subjects in both groups were similar (Table 1). Table 2 summarises the results of the pulmonary function tests in the two groups at rest and before and after beta-blockade. At rest the mean values were within the normal range and there was no significant difference between the two groups. Beta-blockade produced a slight and statistically insignificant reduction in the forced expiratory volume in 1 set, forced vital capacity and peak expiratory flow rate in smokers. The resting ventilation rate was lower (P < 0.05) in nonsmokers after beta-blockade than in smokers (Table 2). Urinary catecholamines as 4-hydroxy-3-methoxymandelic acid were measured before exercise in 4 smokers and 4 nonsmokers. The mean level of 3.8 f 0.93 pmol/l in nonsmokers was not significantly different from 4.6 f 1.1 pmol 4-hydroxy-3-methoxymandelic acid/l in smokers. Exercise Data
Subjects in both groups achieved the same total maximal workload (5400 kilopondmetres) on the bicycle ergometer after a training period of 3 tests; non-
TABLE
2
Pulmonary function data of smokers and nonsmokers intravenously). Data expressed as means f 1 SD. Nonsmokers
after placebo
(n = 7)
and after propranolol
Smokers
(0.3 mg/kg
(n = 5)
Propranolol
Placebo
Vital capacity
5200
+320
5100
+300
5000
+700
4800
f 700
(ml) Forced
4400
+ 600
4400
*400
4000
+ 600
3800
&500
637
+133
Placebo
expiration
Propranolol
vol. in 1 set (ml) Peak expiratory flow rate (l/min) Ventilation
646
f141
7.3&
0.9
(l/fin) * P i 0.025 as compared
with nonsmokers.
6.6k
583
0.5
k
8.9+
68
2.1
558
f
8.16k
56
1.5*
291 T
TT
p--p4ol T _l_.X_
I
I
,
poo.001
~3 Smokers after A Non smokers = Non vrokers
beta blockade after beta blockade
4 1200
2400
3600
6000
4000
Work (hpm)
Fig. 1. Relation of heart rate to cumulative work load in kilopondmetres (kpm) before and after beta-blockade in healthy smokers and nonsmokers. Beta-blockade was achieved by the administration of propranolol 0.3 mg/kg body weight intravenously 1 hour before the exercise test. At comparable workloads smokers showed a significantly higher heart rate than nonsmokers before and after beta-blockade. All data mean +l SD.
smokers achieved this workload in the first test and their heart rate became linear with the increments in workload by the second test. However, the smokers found it difficult to match the exercise performance of nonsmokers until the third test; their
6000
smokers
2700
+eoo
2400 2 E E .s ‘i
5 0
s
2100
‘Ei
‘800
3600
2 ‘p H
2400
$ I
900 600
1200
300
6246824
6
8
IO
Minutes During
exercise
Post-exercise
Fig. 2. Effects of exercise on oxygen consumption in smokers and nonsmokers. Work load is cumulative increase in kilopondmetres (kpm) each minute. Resting oxygen consumption was smokers (420 ml/mm) than in nonsmokers (290 ml/mm) and it remained slightly higher increment in the workload. During the post-exercise recovery period the oxygen consumption higher (but not significantly) in smokers. Data expressed as mean f 1 SD.
shown as higher in for each remained
3
+9
rt5
+8
&6
+6
+7
4.4 +0.51
1.37kO.81 3.85 f 0.98
0.46
3.93*
0.93* 0.15 7.0 + 0.35
0.28 + 0.02 2.40 k 0.20
118
114 157
63
57 125
Propranolol
0.08 0.20
*6
f 5 *17
*9
+5 +7
(n = 7)
0.29+ 2.78&
149
114 186
78
64 157
Placebo
Nonsmokers
1.13+ 6.671
3.79+
0.25 0.99
0.68
0.20 0.50
* 15
f 5 *21*
+11**
C 0.05
n.s. < 0.01
(n = 5)
f 3** +11**
0.421 2.96$
163
118 220
105
75 176
Placebo
ns. < 0.05
< 0.001
n.s. ‘z 0.001
-=z0.01
n.s. < 0.01
Significance P (paired-t)
Smokers
k28
*7 k36
+
0.54*
1.15f 0.85 5.8 + 0.28*
5.1 f
0.02 0.40**
9**
f6’ f 5**
0.25f 2.85 i
146
110 183
86
69 136
Propranolol
systolic blood pressure (mm Hg), oxygen consumption (I/min) and serum lactate concentration (mmol/l) in smokers and nonsmokers before and after beta-blockade (propranolol 0.3 mg/kg intravenously).
* P < 0.05; **p < 0.01.
(5400 kpm)
(Vmin) Resting Peak exercise (5400 kpm) Oxygen debt Serum lactate (mmol/l) Resting At exercise
(5400 kpm) Post-exercise (1 min) Oxygen consumption
(5400 kpm) Post exercise (10 min) Systolic blood pressure (mm Hg) Resting Peak exercise
Heart rate (beats/min) Resting Peak exercise
Heart rate (beats/min), (5400 kilopondmetres)
TABLE
n.s. ‘C 0.05
< 0.001
< 0.05 n.s.
ns.
n.s. ns.
< 0.01
< 0.05 < 0.001
Significance P (paired-t)
at rest and at peak exercise
%
299
oxygen consumption became linear with the increments in workload by the fourth test, but their heart rate was significantly greater than nonsmokers at rest and at all increments in workload during the exercise test (Fig. 1; Table 3). Propranolol reduced the heart rate significantly at all levels of exercise in both groups but smokers continued to have a relative tachycardia even after beta-blockade when compared with nonsmokers (P < 0.01). Similarly, the mean systolic blood pressure at peak exercise during placebo was significantly (P < 0.05) higher in smokers (220 + 21 mm Hg) than in nonsmokers (186 + 17 mm Hg). Furthermore, beta-blockade significantly reduced the peak- and post-exercise systolic blood pressure in nonsmokers, but had only a slight and statistically insignificant effect in smokers (Table 3). The resting oxygen consumption was higher in smokers than in nonsmokers but there was a large scatter among smokers and the difference was not statistically significant. With the stepwise increase in workload, the oxygen consumption increased linearly in both groups; smokers had a slightly higher oxygen consumption than nonsmokers but the difference was not statistically significant (Fig. 2). During the recovery period oxygen consumption remained slightly but not significantly higher in smokers. Effects of Beta-Blockade
in Oxygen Consumption
There was no significant difference in the oxygen debt incurred after placebo by the two groups. With beta-blockade, there was a reduction in the oxygen consumption at each workload in nonsmokers; at the maximum workload oxygen consumption fell from 2.78 f 0.20 l/min to 2.40 f 0.20 l/min (P < 0.05). This was associated with a small increase in oxygen debt, from 3.93 f 0.46 to 4.40 + 0.5 1 (P = 0.05, Table 3). However, in smokers there was no change in oxygen consumption with beta-blockade; at the maximal load oxygen consumption was 2.96 f 0.50 I/min before and 2.85 _t 0.40 l/min after beta-blockade. There was a significant increase in oxygen debt, from 3.79 + 0.68 to 5.10 + 0.54 1 (P < 0.001). After propranolol the oxygen debt was significantly greater in smokers than in nonsmokers (5.10 + 0.54 vs. 4.40 + 0.501; p < 0.05). These differences in oxygen debt were also reflected in the serum lactate levels at peak exercise which were significantly (P < 0.05) higher in smokers than in nonsmokers (Fig. 3). In both groups the serum lactate level at peak exercise was significantly lower after beta-blockade (Table 3). Post-Exercise
Data
Throughout the recovery period, the heart rate remained significantly higher in smokers. At the tenth minute after exercise, the heart rate was 105 ~fr11 in smokers compared with 78 f 9 beats/mm (P -c0.001) in nonsmokers. As would be expected beta-blockade significantly reduced the heart rate in the recovery period in both groups, but in smokers the rate remained significantly (P < 0.01) higher 10 min after exercise than in nonsmokers (Table 3).
300 Placebo Propranolol
0.3mg/Kg
I .V
r P10.05
1
Mean+_ S
.D
1
Serum (mmole/L)
smokers
(n=5)
Fig. 3. Serum lactate and oxygen debt before and after beta-blockade in smokers and nonsmokers. Beta-blockade increased significantly (P -C 0.05) oxygen debt in smokers but not in nonsmokers and reduced significantly (P < 0.01) serum lactate in nonsmokers but not in smokers. Smokers had significantly (P < 0.05) higher oxygen debt and serum lactate after beta-blockade as compared with nonsmokers.
301
Discussion The smokers and nonsmokers in this study had similar physical characteristics and all followed a sedentary life style. Both groups had similar resting urinary catecholamines, and achieved the same maximal workload on the bicycle ergometer, with comparable oxygen consumption and oxygen debt. Despite these similarities smokers had a significantly faster heart rate at rest and at exercise than nonsmokers. Although smokers followed similar exercise pursuits as their nonsmoker colleagues, they required greater motivation and longer conditioning to achieve the same workload as nonsmokers. This greater motivation and effort made by smokers may in part explain their higher heart rate and systolic blood pressure, their partial insensitivity to beta-blockade and their greater oxygen debt. Smoking is known to activate hepatic enzymes, and orally taken propranolol is cleared faster in smokers than in nonsmokers [19]. Propranolol was given intravenously in large doses, and the study was completed within 90 min to minimise any possible pharmacokinetic variations in smokers. Propranolol reduced the heart rate in both groups but less so in smokers who still had a relative tachycardia. It reduced the oxygen consumption on exercise in nonsmokers but not in smokers. After beta-blockade smokers had a higher oxygen debt (P < 0.001) and significantly (P < 0.05) higher lactate levels at peak exercise than nonsmokers. In this study the lactate level was measured at the peak of exercise in both groups and it is possible that the peak level, which may occur at a variable time after exercise [20] was missed. These cardiorespiratory and metabolic differences cannot be attributed to a faster clearance of propranolol or to a higher adrenergic drive in smokers. Hepatic clearance of intravenously administered propranolol would mostly depend on the hepatic blood flow, which is not altered by smoking [21,22]. Even though the intrinsic hepatic clearance of propranolol may have been accelerated in our smokers, it is unlikely that the serum drug level would have been significantly lowered within 60 min of intravenous administration of propranolol to account for the relative tachycardia observed during and after exercise. Tachycardia after smoking [8,11,14,23] has been attributed to the effects of nicotine [ 11,241 and to hypoxaemia caused by carboxyhaemoglobinaemia [25]. Inhalation of carbon monoxide in nonsmoking volunteers to raise the carboxyhaemoglobin concentration to the levels seen in smokers produces similar haemodynamic effects as we have described [26]. However, since the half-life of carboxyhaemoglobin is about 4 hours [27] it cannot fully explain the relative tachycardia in our group of smokers who had not smoked for over 6 hours. There are conflicting reports of higher catecholamine levels in smokers [28] but this persistence of a relative tachycardia in smokers after beta-blockade suggests that it is not mediated wholly through excess adrenergic stimulation. Cryer et al. [12] found that the haemodynamic effects of smoking preceded and persisted long after the transient rise in catecholamines which follows smoking, whereas Strasser et al. [28] found no change in catecholamine levels after smoking. In this study there was no difference in the resting urinary catecholamine levels in the two groups. The effects of chronic cigarette smoking on the myocardium are unknown. In a study on the effect of
302
hypoxaemia on the electrocardiogram in patients with car pulmonale, Tirlapur and Mir [29] showed that tachycardia and a prolonged QT interval were reversed by oxygen in nonsmokers but not in smokers. Thus no single factor can explain the higher rate in smokers: hypoxaemia due to carboxyhaemoglobin, excessive adrenergic drive and a direct effect on the myocardium probably all contribute. Resting oxygen consumption was higher in smokers but there was a wide scatter in this group as has been observed by Krone et al. [30], and the differences did not reach statistical significance. Oxygen uptake on exercise and oxygen debt were not significantly different in the two groups. Chevalier et al. [7] observed a higher oxygen debt in smokers but these workers were unable to provide any supporting evidence. In this study the lack of any significant difference in the oxygen debt between smokers and nonsmokers was reflected in their similar serum lactate levels. Although myocardial oxygen usage forms only a minor part of the total oxygen consumption, a greater myocardial oxygen expenditure in smokers, caused by higher systolic blood pressure and heart rate, must have contributed to their greater oxygen consumption and debt. Beta-blockade revealed complex differences between the two groups. Propranolol reduced the resting oxygen consumption in both groups, though during exercise it reduced oxygen consumption only in nonsmokers. Previous studies have demonstrated this effect in nonsmokers [31-341 but a lack of this effect in smokers has not been previously described, and explains why studies on mixed groups have given varying results [35]. After propranolol smokers had a greater oxygen debt than nonsmokers. This difference is also reflected in the lactate levels at peak exercise. Beta-blockade reduced the lactate level in both groups, as has been previously described in nonsmokers [36], but this reduction was significantly (P -c0.05) less in smokers. This fall in oxygen consumption and a reduction in lactate levels in nonsmokers are suggestive of an increased efficiency after beta-blockade. In contrast smokers after beta-blockade had significantly higher heart rates during exercise, higher oxygen debt, higher ratio of oxygen debt to increased oxygen uptake and higher lactate levels than nonsmokers. While beta-blockers significantly reduced systolic blood pressure during and after exercise in nonsmokers, it failed to have any significant effect on smokers. These findings suggest that smokers were doing the same work after beta-blockade for a higher haemodynamic and metabolic cost than nonsmokers. While the underlying mechanisms responsible for these differences between smokers and nonsmokers before and after beta-blockade are not known at present, these results have two major implications. First, during exercise testing heart rate as an index of work done is an unreliable guide in smokers who have a higher heart rate even after full physical conditioning, than nonsmokers for the same work output. Exercise data should be interpreted with caution in a mixed population of smokers and nonsmokers. Secondly, since beta-blockade does not fully suppress exercise and resting tachycardia in smokers, its full benefit cannot be realised in smokers. The differences we have described help explain the difficulties experienced by some workers [37] in effectively treating angina pectoris in patients who continue to smoke.
303
References 1 Doll R. Hill AB. Mortality
in relation
J 1964;l: 139991410. 2 Doyle JT, Dawber TR, Kannel 3 4 5 6 7 8 9 10
11
12
13 14 15 16 17 18 19 20 21 22 23
24 25
to smoking:
WB, Kinch
ten years’ observations
SH. Kahn
of British doctors.
HA. The relationship
of cigarette
Br Med
smoking
to
coronary heart disease. J Am Med Assoc 1964:190: 886-890. Frank CW, Weinblatt E, Shapiro S, Sager RV. Myocardial infarction in men: role of physical activity and smoking in incidence and mortality. J Am Med Assoc 1966;198: 1241-1245. Ball K, Turner R. Smoking and the heart: the basis for action. Lancet 1974;2: 822-826. Armitage AK, Hall GH. Mode of action of intravenous nicotine in causing a fall of blood pressure in the cat. Eur J Pharmacol 1969:7: 23-30. Cellina GU, Honour AJ, Littler WA. Direct arterial pressure, heart rate. and electrocardiogram during cigarette smoking in unrestricted patients. Am Heart J 1974:89: 18-25. Chevalier RB, Krumholz RA, Ross JC. Reaction of non-smokers to carbon monoxide inhalation: cardiopulmonary responses at rest and during exercise. J Am Med Assoc 1966;198 : 1061-1064. Aronow WS. Goldsmith JR. Kern JC, Johnson LL. Effect of smoking cigarettes on cardiovascular haemodynamics. Arch Environ Health 1974;28 : 330-332. Goldberg AN, Krone RJ, Resnekov L. Effects of cigarette smoking on haemodynamics at rest and during exercise. 1. Normal subjects. Chest 1971;60: 531-536. Ayres SM. Mueller HS. Gregory JJ. Giannelli S, Penny JL. Systemic and myocardial haemodynamic responses to relatively small concentrations of carboxyhemoglobin (CoHB). Arch Environ Health 1969:18:699-709. Aronow WS. Dendinger J, Rokaw SN. Heart rate and carbon monoxide level after smoking high-, low-, and non-nicotine cigarettes. A study in male patients with angina pectoris. Ann Intern Med 1971;74:697-702. Cryer PE, Haymond MW, Santiago JV, Shah SD. Norepinephrine and epinephrine release and adrenergic mediation of smoking-associated haemodynamic and metabolic events. N Engl J Med 1976:295 : 573-517. Blackburn HJ, Brozek J. Taylor HL. Common circulatory measurements in smokers and nonsmokers. Circulation 1960;22: 1112-1124. Chevalier RB. Bowers JA, Bondurant S, Ross JC. Circulatory and ventilatory effects of exercise in smokers and nonsmokers. J Appl Physiol 1963:lS: 357-360. Leese HJ. Bronk JR. Automated fluorometric analysis of micromolar quantities of ATP. glucose and lactic acid. Anal Biochem 1972;45 : 211-221. Humphrey, SJE, Wolff HS. The Oxylog. J Physiol (London) 1977;267 : 1-12. Belyarin AJ. Brown GA, Harrison MH. The Oxylog: an evaluation. Royal Air Force Institute of Aviation Medicine IAM Report No. 608, 1981. Van de Calseyde JF, Scholtis RJH. Schmidt NA. Leyton CJJA. Gas chromatography in the estimation of urinary metanephrines and VMA. Clin Chim Acta 1971:32 : 361-366. Wood AJJ. Vestal RE, Branch RA. Wilkinson GR, Shand DG. Effect of aging and cigarette smoking on antipyrine and indocyanine green elimination. Pharmacol Ther 1979:26 : 16-20. Astrand PO, Rodahl K. Textbook of work physiology. New York: McGraw-Hill. 1970. Vestal RE, Wood AJJ. Branch RA, Shand DG, Wilkinson GR. The effects of aging and cigarette smoking on propranolol’s disposition. Clin Pharmacol Ther 1979~26: 8-15. Nies AS. Wilkinson GR, Rush BD, Strother JT, McDevitt DG. Effects of alteration of hepatic microsomal enzyme activity on liver blood flow in the rat. Biochem Pharmacol 1976;25 : 1491-1994. Thomas CB, Bateman JL. Lindberg EF, Bronhold HJ. Observation on the individual effects of smoking on the blood pressure, heart rate, stroke volume, and cardiac output of healthy young adults. Ann Intern Med 1956:44: 874-892. Aronow WS, Kaplan MA. Jacob D. Tobacco: a precipitating factor in angina pectoris. Ann Intern Med 1968;69 : 529-536. Asmussen E. Chiodi H. The effects of hypoxaemia on ventilation and circulation in man. Am J Physiol 1941;132 : 4266436.
304 26 Chevalier RB, Krumholtz RA, Ross JC. Reaction of nonsmokers to carbon monoxide inhalation, cardiopulmonary responses at rest and during exercise. J Am Med Assoc 7966;198 : 1061-1064. 27 Shephard RJ. Smoking withdrawal and changes of cardiorespiratory fitness. Am Rev Respir Dis 1971:104:933-935. 28 Strasser R, Dietz R, Schomig A, Rascher W, Kubler W. Sympathetic activity during smoking before and after beta-blockade (abstract). Eur Heart J 1981;2: 203. 29 Tirlapur VG. Mir MA. Nocturnal hypoxaemia and associated electrocardiographic changes in patients with chronic obstructive airways disease. N Engl J Med 1982;306 : 125-130. 30 Krone RJ, Goldberg AN, Balkoura M. Schuessler R. Resnekow L. Effects of cigarette smoking at rest and during exercise. II. Role of venous return. J Appl Physiol 1972;32 : 745-748. 31 Schroder G, Werko L. Haemodynamic studies and clinical experience with nethalide. a beta adrenergic blocking agent. Am .I Cardiol 1965;15 : 58-65. 32 Astrom H. Haemodynamic effects of beta adrenergic blockade. Br Heart J 1968;30 : 44-49. 33 Parker JO, West RO, Di Giorgi S. Haemodynamic effects of propranolol in coronary heart disease. Am J Cardiol 1968;21: 11-19. 34 Sonnenblick EH, Braunwald E, Williams JF, Glick G. Effects of exercise on myocardial force velocity relations in intact unanaesthetised man. Relative roles of change in heart rate, sympathetic activity and ventricular dimensions. J Clin Invest 1965;44: 2051-2062. 35 Gibson DG. Pharmacodynamic properties of /I-adrenergic receptor blocking drugs in man. Drugs 1974:7:8-38. 36 Pillet AP, Bemand BV, Saunders RA. Effects of propranolol on blood sugar, insulin and free fatty acids. Diabetologia 1969;5 : 339. 37 Fox K, Johnathan A, Williams H, Selwyn A. Interaction between cigarettes and propranolol in treatment of angina pectoris. Br Med J 198O:l: 191-193.