Effects of cigarette smoking and oral nicotine on hand tremor. The effect of nicotine on hand tremor was assessed by measuring the current induced in a coil.
Effects of cigarette smoking and oral nicotine on hand tremor The effect of nicotine on hand tremor was assessed by measuring the current induced in a coil
assessed the amplitude of tremor induced by smoking two cigarettes in a group of 33 subjects drawn from two age groups. Smoking increased tremor significantly. Age, sex, and anxiety levels had no effect. Experiment 2 assessed the effect of smoking on tremor frequency. Analog recordings of tremor in 10 subjects were processed to yield the total power in 20 I-Hz bands from I to 20 Hz. The log amplitude showed a strong linear relationship to the log frequency, but this relationship was not affected by smoking. Smoking also had no effect on peak frequency in the frequency spectrum. Experiment 3 confirmed that the effects of smoking on tremor were attributable to nicotine. Twelve subjects were tested before and after smoking a cigarette and before and after chewing gum containing 4 mg nicotine. The two conditions induced comparable increases in tremor. A 2-mg nicotine preparation tested in eight subjects had no effect on tremor. Tremor may be a useful index of the central activity of nicotine and may help illuminate its mechanism of action. by a magnet attached to the finger. Experiment
1
Saul M. Shiffman, Ph.D., Ellen R. Gritz, Ph.D., Joan Maltese, Ph.D., Monroe A. Lee, Nina G. Schneider, Ph.D., and Murray E. Jarvik, M.D., Ph.D. Los Angeles, Calif., and Tampa, Fla. Department of Psychology, University of South Florida, Department of Psychiatry and
Biobehavioral Sciences, The Neuropsychiatric Institute, School of Medicine, University of California, Los Angeles, and the Veterans Administration Medical Center, Brentwood.
Nicotine is a powerful alkaloid with both central and peripheral nervous system effects. At nicotinic cholinergic ganglia, it exerts an initial stimulant effect, followed later by blockade. It is believed that such biphasic action can also occur in the central nervous system (CNS) and that the effect is dose-dependent. Although it is increasingly clear that nicotine plays a major role in the maintenance of a cigarette.
Supported by the Medical Research Service of the Veterans Administration. Received for publication Aug. 23, 1982. Accepted for publication Jan. 6, 1983. Reprint requests to: Saul M. Shiffman, Ph.D., Department of Psychology, University of South Florida, Tampa, Fla. 33620.
800
smoking habit,'° 13 CNS activities of nicotine are still inadequately understood. One way to study the central effects of nicotine is to examine a peripheral event affected by nicotine, such as hand tremor, from which central effects might be inferred. All voluntary muscular activity is accompanied by rhythmic oscillations known as normal or physiologic tremor.2' 12 Although some purely mechanical explanations of tremor have been proposed,2, 15, 19, 20, 22 tremor is probably controlled by pharmacologic mechanisms. Sweet et al.2' suggest that there may be a spectrum of tremor conditions with distinct biochemical defects but with a final common path-
Volume 33 Number 6
Effects of smoking and oral nicotine on tremor
way. Norepinephrine consistently increases the amplitude of tremor in a variety of pathologic tremor types, whereas other drugs, including nicotine, have inconsistent effects.5' 16 Cahen et al.' demonstrated blockade of nicotine tremor in rabbits by adrenergic blocking agents. Nicotine has sometimes been used to model the tremor associated with Parkinson's disease. Working with rabbits, Bovet and Longo3 found a parallel between the antagonism of some drugs against tremors produced by nicotine and the clinical results obtained with these drugs in the therapy of Parkinson's disease. They suggest that these results are produced by nicotine's cholinergic activity. The effect of smoking on normal tremor has been the subject of several studies. Frankenhaeuser et al.9 reported that the first cigarette induced the largest increase in tremor, with subsequent cigarettes inducing smaller changes. In a pilot study by Day ,6 smoking one cigarette was found to increase tremor amplitude in light smokers, and no change in tremor frequency was seen. This was consistent with findings in an earlier study by Day and co-workers.17 Unfortunately, Day's subjects were not typical smokers but light smokers and even nonsmokers. Edwards7 performed several early experiments to measure the effects of smoking on tremor. Finger tremor increased after smoking and inhaling half a cigarette but not in subjects who did not inhale. Tremor increased whether or not the subjects were previously abstinent and decreased 2 hr after withdrawal. Edwards' data were not analyzed statistically, leaving the conclusion ambiguous. These reports suggest that inhaled nicotine increases tremor. The smoking status of subjects, however, was not controlled. It is not clear how nicotine affects the typical habitual smoker who has developed tolerance to nicotine. Furthermore, earlier studies measured tremor with mechanical accelerometers, which tend to dampen some components of tremor and to distort its frequency spectrum. In our study we were interested in the effect of smoking on tremor in the habitual smoker and in ascertaining whether this effect is caused by nicotine per se. Smokers without pathologic tremor were studied by an electromagnetic transducer to re-
801
cord motion accurately. Experiment 1 concerned the effect of smoking on tremor amplitude, experiment 2 concerned the effects of smoking on tremor frequency, and experiment 3 compared the tremor effects of orally administered nicotine and a standard cigarette. Experiment
1
Our subjects were 23 smokers who usually smoked 20 cigarettes or more a day. All smoked brands delivering at least 0.7 mg nicotine (mean = 1.1 mg) per cigarette. Twelve subjects were women; 11 were men. All were righthanded. Subjects were drawn from two distinct age groups. Eleven were more than 50 yr old (mean age = 57.5), and 12 were under 30 yr old (mean age = 22.6). The older subjects smoked an average of 34.8 cigarettes a day and had smoked for a mean of 41.3 yr. The younger subjects smoked an average of 19.3 cigarettes a day and had smoked for a mean of 5.9 yr. Subjects were tested in the morning after overnight abstinence from tobacco and caffeine. First, subjects provided a breath sample for carbon monoxide (CO) analysis (as a check of abstinence) and resting heart rate (HR) was recorded. A smoking history was obtained and the State-Trait Anxiety Inventory (STAI) was administered.'s Each subject's tremor was then measured by using the "tremometer" described below. Subjects then smoked two cigarettes of their own brand (mean nicotine delivery = 1.1 mg) in a prescribed fashion, inhaling for 5 sec every 30 sec. After CO and HR measurements, tremor was retested. The following procedures were followed identically before and after the subjects smoked. While standing, subjects held their left (nondominant) hand inside the "tremometer" coil (see below) for a series of six 10-sec trials. Each trial produced a single tremor score. Analysis of the data by trials showed that the series of six trials had high internal consistency (Cronbach's coefficient: a = 0.87). Scores were averaged over trials for subsequent analysis. The "tremometer" was based on the principle that movement of a magnet in relation to a coil induces a current in the coil. The coil was wrapped around a Plexiglas rectangle (15.8 cm wide, 7.0 cm high, and 5.3 cm deep) open on
802
Shiffman et al.
Clin. Pharmacol. Ther. June 1983
POWER SPECTRUM TREMOR, PRE-SMOKING
0.6 LOG
0.8
PRIQUINCY
POWER SPECTRUM TREMOR, POST-SMOKING
'
0.0
0.2
U481
0.4
0.6
LOO
08
FILIQUINCT
1.0
1.2
1.4
111.1
Fig. 1. Power spectra of tremor before and after smoking.
two opposite sides. The subject's hand, to which a small samarium cobalt magnet had been taped, was held in the open volume of the Plexiglas coil. Movement of the subject's hand induced current to flow into the coil. This current was then amplified to induce a flat amplification function over the range of 2 to 20 Hz. (The signal was amplified x 10,000 at 1 Hz, and the amplification reduced 12 db per octave across the spectrum.) The resulting amplitude was accumulated over a 10-sec trial. Results. The CO and HR data were analyzed to check on the experimental manipulation. The relatively low mean CO reading (12.2 ppm, SD = 4.9) at the pretest confirmed that subjects followed the instruction to refrain from smoking on the morning of the experiment. (Our experience is that this is well within the range in nonsmokers in the Los Angeles basin.) Pretest HR averaged 67.9 bpm, (SD = 6.1). Both CO and HR were significantly elevated as a result of the smoking manipulation (after-test CO = 25.2 ppm, SD = 11.1; after-test HR = 81.0 bpm, SD = 10.6), demonstrating that subjects received an effective dose of nicotine. The tremor data were analyzed by a 2 x 2 (age by condition) repeated measures analysis of variance. The analysis showed that smoking induced a marked and significant increase in tremor (from 30.25 to 60.36 arbitrary units; P < 0.0001). No effects of the age or sex of
subjects were observed, nor were STAI scores related to tremor scores. Experiment 2
The subjects in the older group of experiment returned to the laboratory for this experiment. One male subject could not be reached, resulting in n = 10. Following procedures similar to those in experiment 1, subjects participated in two 30-sec "tremometer" trials, one before and one after smoking two cigarettes. The analog output of the measuring device, which provides an exact analog representation of the subject's hand movement, was recorded on magnetic tape. This analog signal was then digitized at a sampling rate of 100 samples a sec and was subject to a power spectrum analysis through fast Fourier transformation. For purposes of analysis we used the frequency bands from 1 to 20 Hz, dividing this spectrum into 20 1-Hz bands. Results. Fig. 1 shows the mean power spectra before and after smoking. There is a marked peak in the 1-Hz frequency band, probably because of whole-body movement rather than specific hand tremor. The spectra suggest that smoking may act specifically to increase tremor in the range from 10 to 15 Hz. Two methods were applied in an attempt to identify any effects nicotine might have on the frequency of tremor. The first attempted to characterize each spectrum using a single math1
Volume 33
Number 6
Effects of smoking and oral nicotine on tremor
ematical function. Evaluation of the spectra showed that the log amplitude was a linear function of the log frequency. Linear regression of the log amplitude on the log frequency for each spectrum accounted for an average of 89.3% of the variance in each spectrum. (The data for 1 Hz were excluded because they were so deviant that they tended to markedly decrease the overall fit of the regression line.) Dependent t tests compared the mean slopes and intercepts of the regression lines for the spectra before and after smoking. The summed squared deviations from the regression, an inverse measure of the "fit" achieved, was similarly tested. These parameters were not significantly affected by smoking. There were also no smoking-related differences in the frequency at which maximum power was observed. Thus smoking does not appear to affect the frequency of hand tremor. Experiment 3
Twelve subjects (six men and six women) participated in this experiment. Subjects ranged in age from 18 to 58 yr (mean age = 27.5), and had smoked an average of 21.8 cigarettes a day for at least a year (mean = 11 yr). Subjects smoked brands of cigarettes delivering at least 0.72 mg nicotine and averaging 1.05 mg nicotine. All subjects were offered the opportunity to return for a third session to be tested on the 2-mg nicotine gum, and eight accepted. Subjects who participated in the third session did not differ demographically from those who declined. All testing sessions were held in the morning after overnight abstinence from cigarettes and caffeine. The CO concentration in expired air and HR were measured at the beginning and end of each session. In each session tremor was measured before and after the treatment by use of the procedures of experiment 1. In one session subjects smoked one of their own cigarettes. Subjects inhaled for 5 sec every 30 sec, and the cigarette was followed immediately by the "tremometer" test. In another session subjects chewed Nicorette gum, a preparation of nicotine bound to an ion exchange resin for the controlled release of nicotine.' The gum
O
803
pre
post
I
SEM N.12
20
2 mg Gum
4 mg Gum
1
mg Cigarettes
Fig. 2. Hand tremor measured before and after chewing 2-mg and 4-mg Nicorette gum or smoking a cigarette yielding 1 mg nicotine. The 4-mg Nicorette and the 1-mg cigarette were roughly equivalent in tremogenic effect.
contained 4 mg nicotine. Subjects were instructed to chew the gum slowly for 20 min and were then given the tremor test. The order of the cigarette and 4-mg nicotine gum conditions was counterbalanced. A third session, in which only eight subjects participated, was held last. Procedures were identical to those in the 4-mg nicotine gum condition with the exception that a Nicorette preparation containing 2 mg nicotine was used. Results. Analysis of CO at the beginning of each session confirmed subject abstinence. Expired CO concentration averaged 12.2 ppm. The CO measures also confirmed inhalation in the cigarette condition, rising to an average of 21.1 ppm (t = 7.08, df = 11, P < 0.001). There were significant HR increases in all three conditions, although they were attenuated in the 2-mg nicotine gum condition (a rise of 5.5 bpm and 11 bpm in the other conditions). Thus physiologic effects of nicotine were evident in all conditions. The tremogenic effects of the cigarette and the 4-mg Nicorette were analyzed by a 2 x 2 (condition by tremor before and after treatment) repeated measures analysis of variance. There was a significant increase in tremor (from 30.78 to 45.03 arbitrary units) in both conditions (F = 5.71; df = 1, 11; P
