NEUROPSYCHOPHARMACOLOGY 1993-VOL. 9, NO.3 233

Effects of Tobacco Smoking on the Hoffmann Reflex Chitoshi Kadoya, M.D., Shigeaki Matsuoka, M.D., and Edward F. Domino, M.D.

Ten normal adult tobacco smoking volunteers 21 to 32 the H reflex recovery cycle. Individual differences were years of age were the subjects of this study. They were marked. Nevertheless, the data obtained are consistent asked not to smoke for 12 hours prior to testing. The with evidence in animals that nicotine and tobacco smoke Hoffmann (H) reflex and its recovery cycle were measured stimulate Renshaw inhibitory neurons in the spinal cord, before and just after smoking (on different days) one either directly or indirectly. This technique provides nonfiltered, zero, low (0.27 mg), or high (2.16 mg) another objective measure of the effects of tobacco smoking nicotine tobacco cigarette. After smoking the nicotine­ in human volunteers. [Neuropsychopharmacology containing cigarettes, the subjects showed a reduction of 9:233-238, 1993J

KEY WORDS: Tobacco smoking; Hoffmann reflex; heart rate were also measured as a control index of pe­ Recovery cycle ripheral vascular effects.

It is well known that tobacco smoking and nicotine re­ duce the patellar reflex of animals and man (Schweit­ SUBJECTS AND METHODS zer and Wright 1938; Clark and Rand 1964; Domino and von Baumgarten 1969). Another monosynaptic stretch Ten healthy normal tobacco cigarette smoking volun­ reflex is the ankle jerk, with an electrically evoked teers who inhaled and smoked at least one pack of cig­ monosynaptic response that is equivalent to the Hoff­ arettes per day were recruited for this study. They mann (H) reflex. Gassel (1973) reported in a nonpeer­ ranged in age from 21 to 32 years with a mean of 22.8 reviewed years. Two were women and eight were men. They 0.8 mg to 1 mg of nicotine in either 24-hour deprived were asked to stop smoking the night before each ex­ smokers or nonsmokers caused a concomitant decrease periment so that they would be approximately 12 hours in the ankle jerk as well as the conditioning and test without cigarette smoking at the time of each test. H reflex recovery cycle at 200- and 300-millisecond in­ Approximately 12 hours after the last tobacco ciga­ tervals. Recovery occurred within 30 minutes. The pres­ rette smoked, each subject, in a random design, was ent study was designed to determine the effects of asked to smoke three different nonfllteredcigarettes on smoking cigarettes of a wider range of nicotine content different days in the laboratory. The cigarettes con­ on the H reflex recovery cycle over a much longer 60- tained no (0 mg), low (0.27 mg), or high nicotine (2.16 to 340-millisecond interval. Arterial blood pressure and mg). The nicotine-free cigarettes were purchased over the counter and contained a blend of wheat, cocoa and From the Department of Pharmacology (CH, EFD), University of orange leaves. The low- and high-nicotine cigarettes Michigan, Ann Arbor, Michigan and Department of Neurosurgery (CH, SM), University of Occupational and Environmental Health, were research cigarettes obtained from the University Kitakyushu, Japan. of Kentucky (2R1; Tobacco and Health Research Insti­ Address all correspondence to: E.F. Domino, M.D., Department tute, Lexington, KY). Each subject smoked the ciga­ of Pharmacology, University of Michigan, Ann Arbor, MI48109-0626. Received February 23, 1993; revised June 29, 1993; accepted June rettes at their normal rate, depth of inhalation, and du­ 30,1993. ration of 5 to 8 minutes. The mean of two pre smoking

© 1993 American College of Neuropsychopharmacology Published by Elsevier Science Publishing Co., Inc. 655Avenue of the Americas, New York, NY 10010 0893-133X/93/$6.00 234 C. Kadoya et a1. NEUROPSYCHOPHARMACOLOGY 1993-VOL. 9, NO.3

and two post smoking H reflex recovery cycles was de­ the ratio of the amplitude of the H2 to the HI response termined for each subject. In addition, heart rate and at different millisecondintervals from 60 to 340 to de­ blood pressure were measured before and after smok­ termine the recovery cycle. Figure 1 illustrates the spi­ ing by feeling the pulse and by auscultation, respec­ nal reflex pathway studied. Figure 2 illustrates typical tively. Data were analyzed using both correlated and potentials of the M and H responses as recorded. noncorrelated Student t-tests. The general methods used for obtaining the H reflex are well described by Hugon (1973). A Nicolet Path­ RESULTS nnder II was used in these studies. The ampli&er band­ Effects on the Recovery Cycle of the H Reflex pass was 75 Hz to 1500 Hz. Stimulus Sl was applied to the left popliteal space for stimulating the posterior Smoking a nicotine-free cigarette had no signifIcant tibial nerve with a bipolar electrode. The cathode was effecton the recovery cycle of the H reflex. The recov­ placed centrally and the anode distally. A square-wave ery cycle for all 10 subjects is illustrated in Figure 3 and pulse of 0.5 milliseconds every 2 seconds was applied. is given to show normal intersubject variability. In con­ The subject'S motor response of the calf (soleus) mus­ trast, in most subjects smoking a low- or high-nicotine cle was recorded as the direct M response. At the site cigarette produced some depression of the recoverycy­ of maximal response, a bipolar EMG electrode was cle of the H reflex. Smoking a low-nicotine cigarette placed over the soleus muscle with a ground between depressed the H reflex only at certain intervals, partic­ the stimulating and the recording electrodes. The ularly in subjects SE, IK, OK, MA, OM, and KN. In stimulating and recording electrodes were placed so contrast, smoking a high-nicotine cigarette depressed that a maximal H response was obtained with a mini­ the H reflex at more intervals in subjects Y A, SE, IK, mal M response from the normally relaxed soleus mus­ MR, OK, MA, OM, and as. The mean ratio of the cle. The Nicolet was programmed to provide 16 chan­ H2/H1 amplitude was obtained for all 10 subjects smok­ nels of data. Channel l always displayed only the fIrst ing cigarettes of differing nicotine concentrations. A Hresponse (HI). The second H response (H2) was dis­ dose-effect relationship to nicotine existed in which played along with the fIrst on the 15 other channels. smoking the nicotine-free cigarette had no signifIcant The HI and H2 responses were separated at increasing effect on the mean H reflex recovery cycle, but it was 20-millisecondintervals. Thus, 52 would be elicited af­ progressively reduced after smoking the low- and es­ ter 51 at intervals of 60, 80, 100, 120, 140, 160, and so pecially the high-nicotine-containing cigarette. The on milliseconds. The data obtained were expressed as mean data are illustrated in Figure 4. The differences were statistically signincant at many different stimu­ lus intervals of the recovery cycle (p < .05).

Cardiovascular Effects of Tobacco Smoking

Smoking cigarettes of differing nicotinecontent had ei­ ther no effect or caused a slight increase in heart rate and arterial blood pressure. As can be seen in Figure 5, smoking nicotine-free cigarettes had no effect on

Group Ia fiber

M H M H M H

Recording Bectrodes

M H M H • , Figure 2. M and H electromyographs of the soleus muscle Muscle Splodle on - illustrating the effects of increasing stimulus strength the -: relative size of the M and H electro myographic response. In Figure 1. Schematic diagram of the neural pathways in­ the present study, the intensity of the electrical stimulus was volved in the H reflex. reduced for a minimum M and maximum H response. NEUROPSYCHOPHARMACOLOGY 1993-VOL. 9, NO.3 Tobacco Smoking and the Hoffmann Reflex 235

Subject Nicotine Free Low Nicotine High Nicotine

'"'0 LOO • •• Y. A. 0.0 0.40 0.20' .. � .... ._-. L�� , -- --

'.20 1.00 "0 S. E. � -- ••• -'�VU_... 0,40 •"'·l •• O,DO ------

u. ," 10K. • •• � 0,10 I r-.A-_I ... .. 0 •. 20 ', .. �- .,.. �-- �- -.�- ��-�- �- -

t.U 1 00

• o.

MR 0.0 c.. 0 .•• Figure 3. Effectsof smok­ 0',. ..20 � ing tobacco of differing ni­ 0.00 cotine content on the re­ covery cycle of the Hoff­ 1,'0 1,00 mann reflex in 10 individual 0.'. O. K. 0.'. smokers. 0.4. 0.10 The stimulus interval '.00".OL between 51 and 52 is plot­ ted on the x-axis. The mean

1.10 ratio of the amplitude of the 1.00 •.1. H2/H1 reflex is plotted on .l- M A. •••• 0.4. the y-axis. Each subject was • •• studied on three different •....• � 00 _ _ � " days on which a free- (0 mg), O low- (0.27 mg), or high­ ". (2.16 mg) nicotine content 1.00 0.'0 cigarette was smoked. Each O.M. 0.'0 'n . 0• •• horizontal row of the three •. 'JO� rf· •• OD graphs illustrates the data for one subject identifIedby his or her initials. The left ". 10. vertical column of graphs K. II. • •• 0,1. I represents the recovery cy­ ... 0',. 20.. � cle plotted for each of 10 0.00 ...... ' - - - .- subjects smoking a nicotine­ free cigarette. The middle

... vertical column represents I." • .•0 the H reflex recovery cycle K. N. • •• 0.4' obtained from the same in­ O.tO.... 0.00 dividuals smoking a low­ nicotine-containing ciga- - Pre rette. The vertical column of 1.60 graphs on the right repre- O.S. 0 1.00 I. o.o.t --0---- sents the H reflex effects in -0'0'0 0 Post "OO��b'.t. the same individuals smok­ '.00 ..... _ ...._. __ ...... -0 ___ ing a high-nicotine-contain­ =!!!!!l! =!!!!l!! =!!!!!.� ing cigarette. 236 C. Kadoya et aI. NEUROPSYCHOPHARMACOLOGY 1993-VOL. 9, NO.3

0.0 Nicotine Free 0.10

0.60 &J u;i. O.SO i :I: 0.40 "0 0 '':: CD G.30 a: o pol' C � : 0.20 �

0.10

0.00

60 80 100 120 140 160 Ito 200 no 240 260 210 aDo '20 JeD

0.110 * *

0.80 Low Nicotine

0.10 III

�... 0.60

:I:� O.so "0 0 � 0.40 a: C

: O.lO �

0.20

0.10

0.00

60 80 100 120 140 160 Ito 200 no 240 260 280 aGO '20 ).0

O.tO *

0.10 High Nicotine

0.10

Figure 4. The effects of smoking &J 060 tobacco of differingnicotine content eni' on the mean recovery cycle of the i O.so H reflex. :I: The mean data of the recovery "0 0 0.40 cycle of the H reflex are plotted for � a: 10 individuals smoking nicotine- O.lO ..c: free, low- (0.27 mg), and high- (2.16 CD � mg) nicotine-containing cigarettes. 0.20 The x- and y-axes of each of the

graphs are similar to those de- 0.10 scribed in Figure 3. The data are

statistically significant using a two- 000 tailed paired comparison t test. The 60 10 '00 no 140 ,�o 110 2DO 220 240 260 alO 3DO ll� 140 vertical lines represent ± SE. * P < .05, ** P < .01. InlolVill Belween 51 and 52 (msor.) NEUROPSYCHOPHARMACOLOGY 1993-VOL. 9, NO. 3 Tobacco Smoking and the Hoffmann Reflex 237

CIl Q) iii (ij a: cr t:: .Pre t:: .Pre tU III Q) OPost Q) o Post I: 150 I 150 '0 "0 c: c: tU III Q) ... � ::I :l VI IJ) VI IJ) ...Q) Q) a.. 0: '0 "0 0 0 0 0 iO co c: c CIl III Ql Q) :i: ::2

Nicotine Free Low Nicotine (0.27 mg)

Q) iii 200 a: • Systolic t:: .Pre CIl 8 Ql o Post o Diastolic * I: 6 o Heart Rate '0 Q) c: () tU c 4 ...Ql � ::I Q) Ii) :t= 2 Ii) II) is Q: c 0 III U Q) 0 -2 0 ::2 iO c: -4 CIl II) -6 � High Nicotine Low Nicotine Nicotine Free High Nicotine (2.16 mg)

Figure 5. Mean cardiovascular effects of smoking tobacco of differing nicotine content. The mean systolic and diastolic ar­ terial blood pressure and the heart rate are plotted. The vertical lines represent + SE. * P < .05. mean systolic or diastolic blood pressure and heart rate. (1973) who determined the effects of tobacco smoke in In contrast, smoking the low-nicotine cigarette pro­ 10 studies and the effects of intravenous nicotine (0.8 duced a slight increase in diastolic blood pressure mg to 1.0 mg) in four studies. It was not clear from his whereas smoking the high-nicotine-containing ciga­ report whether each study represented a separate sub­ rette slightly elevated systolic and diastolic blood pres­ ject. Both treatments caused a marked depression of sure and heart rate. the H reflex and its excitability. His subjects were non­ smokers or were tobacco smokers who stopped smok­ ing at least 24 hours before each test. The nicotine con­ DISCUSSION tent of the cigarette was not specifled.He measured the H reflex recovery cycle at only 200- and 300-millisecond The results of this study indicate that the recovery cy­ intervals. At both intervals, there was a dramatic sup­ cle of the H reflex is clearly depressed following the pression to approximately 50% of control. This effect smoking of tobacco-containing cigarettes. There is a was consistent with a similar depression of the ankle dose-effect relationship in that smoking a zero-nico­ jerk. The overall effect lasted approximately 20 to 30 tine-containing cigarette had no signifIcanteffect on the minutes following either tobacco smoking or nicotine recovery cycle of the H reflex as opposed to low- or high­ administration. nicotine-containing tobacco cigarettes. The data of the In the present study, there was considerable in­ present study are in complete agreement with Cassel dividual variation in the effects observed. Thisis evi- 238 C. Kadoya et al. NEUROPSYCHOPHARMACOLOGY 1993-VOL.9, NO.3

dent from the summary graphs of the actual data of all plasma levels of nicotine with the H reflex recovery cy­ 10 subjects under three different conditions of smok­ cle after tobacco smoking are indicated and are now be­ ing tobacco of varying nicotine content. This may be ing pursued. related, in part, to the fact that the subjects smoked at their own pace, as they would normally. Whether the effectsof the inhaled nicotine act directly on the spinal ACKNOWLEDGMENT cord or on inputs from peripheral sites is a subject of future study. There is no question that nicotine itself Funding for this investigation was provided in part by NIDA has a direct action to suppress the patellar reflex by stim­ Grant DA 07226. ulation of Renshaw inhibitory neurons in the spinal cord (Eccles et al. 1954; Curtis and Ryall 1964; Ueki et al. 1961). Ginzel (1967) suggested that nicotine also excites REFERENCES a variety of peripheral sites including a large variety of peripheral chemoreceptors and sensory receptors such Clark M5G, Rand MJ (1964): A pharmacological effect of as the muscle spindle, which provide additional inputs tobacco smoke. Nature 102:507-508 that stimulate Renshaw inhibitory neurons and thus Curtis RD, Ryall RW (1964): Nicotine and muscarinic recep­ indirectly act to suppress the recovery cycle of the H tors of Renshaw cells. Nature 203:652-653 reflex. Domino EF, von Baumgarten AM (1969): Tobacco cigarette Irrespective of whether the site of action of nico­ smoking and patellar reflex depression. Clin Pharmacol tine is peripheral or central, it is clear that the present Ther 10:72-79 data support the notion that smoking tobacco cigarettes Eccles JC, Fatt P, Koketsu K (1954): Cholinergic and inhibi­ of adequate nicotine content has important actions on tory synapses in a pathway from motor-axon collate rals the spinal cord H reflex recovery cycle. The implications to motoneurons. J Physiol (Lond) 126:524-562 of this research are that humans who smoke tobacco Gassel MM (1973): An objective technique for the analysis receive a skeletal muscle relaxant effect.Webster (1964) of the clinical effectiveness and physiology of action of reported that tobacco smoking reduced spasticity in a drugs in man. In Desmedt JE (ed), New Developments patient with progressive familial spastic paraplegia, as in Electromyography and Clinical Neurophysiology. Basel, Karger, pp 342-359 measured quantitatively using passive motion resis­ tance. It would be of interest to do additional studies Ginzel KH (1967): Introduction to the effects of nicotine on the central nervous system. Ann NYAcad Sci 142:101-120 with nicotine gum or patch to determine whether there is any signi&cant tachyphylaxis or tolerance with con­ Hugon M (1973): Methodology of the Hoffmannreflex inman. In Desmedt JE(ed), New Developments in Electromyog­ tinued nicotine administration. raphy and Clinical Neurophysiology, Vol 3. Basel, Kar­ The cardiovascular effectsof tobacco smoking were ger, pp 277-293 relatively negligible in this study. Only after smoking Schweitzer A, Wright S (1938): Action of nicotine on the spi­ cigarettes containing the highest concentration of nico­ nal cord. J Physiol 94:136-147 tine were statistically signifIcant but small changes ob­ Ueki 5, Koketsu K, Domino EF (1961): Effects of mecamyla­ served in systolic and diastolic blood pressure and heart mine on the Golgi recurrent collateral-Renshaw cell syn­ rate. The signi&cance of this &nding is that the plasma apse in the spinal cord. Exp Neurol 3:141-148 levels of nicotine after smoking must have been rela­ Webster DD (1964): The dynamic quantifIcationof spasticity tively low and yet de&nite effects were observed on the with automated integrals of passive motion resistance. H reflex recovery cycle. Further studies correlating Clin Pharmacol Ther 5:900-908