Dose-Related Actions of Nicotine on Behavior and Physiology: Review and Implications for Replacement Therapy for Nicotine Dependence

Journal of Substance Abuse, 1, 301-317 (1989)

Dose-Related Actions of Nicotine on Behavior and Physiology: Review and Implications for Replacement Therapy for Nicotine Dependence Jack E. Henningfield Phillip P. Woodson Addiction Research Center National Institute on Drug Abuse

Studies of the behavioral and physiologic actions of nicotine have been conducted over nearly a century, during which time the relationship between dose administered and response produced has been extensively studied Oohnston, 1942; Langley, 1905; U.S. Department of Health and Human Services [US DHHS, 1988]). These studies have revealed that the biobehavioral mechanisms by which nicotine controls behavior are similar to those of other dependence-producing drugs. For both animals and humans, central nervous system effects of nicotine can be discriminated, can either reinforce or punish behavior, and can elicit behavioral and physiologic responses. Moreover, following chronic exposure to nicotine, physical dependence develops such that acute abstinence is accompanied by a cascade of neurochemical effects that also can control behavior (i.e., nicotine withdrawal syndrome). Together, these effects of nicotine administration, and of deprivation following chronic exposure, readily lend the drug to controlling the behavior of those who administer it, including the behavior of self-administration itself. The implications of these observations for both the understanding and treatment of tobacco dependence have been extensively reviewed in the Report of the Surgeon General on the Health Consequences of Smoking: Nicotine Addiction (US DHHS, 1988). Implicit to all of the foregoing observations, however, is the concept of drug dosage, whereby the behavioral and physiologic actions of nicotine that lead to controlled behavior and to physical dependence are determined by the dose of nicotine. In the present discussion, the term dose refers to the Correspondence and requests for reprints should be sent to Jack E. Henningfield, Addiction Research Center, National Institute on Drug Abuse, 4940 Eastern Ave., Baltimore, MD 21224. 301

J. E. Henningfield and P. P. Woodson

302

amount of nicotine that was systemically absorbed; not included is the nicotine that was present in the tobacco or polacrilex (gum) vehicle but not extracted, or that which was extracted but then exhaled or swallowed. The presence or absence of a given nicotine-associated response, the magnitude of a given response, and even the qualitative nature of the response may all be determined by the dose of nicotine (Table 1). These relationships between the dose of nicotine administered and the resulting response of the person or animal are fundamental to understanding and treating tobacco dependence. Research from which such observations have been derived has been exhaustivel?, reviewed in the compendia by Larson and his colleagues (Larson, Haag, & Silvette, 1961; Larson & Silvette, 1968, 1971, 1975) and in a recent Report of the Surgeon General (US DHHS, 1988). The present paper reviews studies that illustrate each of the doserelated effects of nicotine summarized in Table 1 and then discusses the implications of these observations for pharmacologic replacement approaches currently or potentially in use to treat tobacco dependence. DOSE IS A DETERMINANT OF RESPONSE MAGNITUDE Most >;esponses to nicotine that have been systematically studied are directly, although sometimes complexly, related to the magnitude of the nicotine dose level. That is, they are not "all-or-none" or "threshold" responses (see discussion of dose response relationships in Gilman, Goodman, Rail, & Murad, 1985). Most biological responses do, of course, show a ceiling effect, and, as discussed further on, some responses change in qualitative ways as the dose continues to increase. Early studies by Langley, Dixon, and others near the turn of the twentieth century showed that a wide range of physiological responses in a variety of species are directly related to the magnitude of the nicotine dose level (Langley, 1914; Langley & Dixon, 1889). In 1942,Johnston showed that the magnitude of subjective responses to nicotine also was related to the dose level of nicotine which was administered (Johnston, 1942). A more recent laboratory study provides a clear illustration of nicotine dose-response relationships involving two means of nicotine administration: cigarette smoke inhalation and intravenous injection (Henningfield, Miyasato, & Jasinski, 1985). Cigarettes with a range of nicotine yield ratings and TABLE 1.

Summary of Dose-Related Actions of Nicotine

• Dose is a determinant of response magnitude. • Responses are differentially affected by changes in nicotine dose. • The vehicle of nicotine delivery can affect the nature and magnitude of the nicotine-associated response. • The magnitude of a response to a given dose of nicotine is inversely related to the degree of tolerance resulting from prior exposure. • Dose is a determinant of the nature of the response.

Dose-related Effects of Nicotine

303

intravenous injections at varying levels of nicotine were given to human volunteers. A variety of measures for assessing the dependence potential (abuse liability) of drugs was used to characterize nicotine. Figure 1 (Henningfield et al., 1985) shows that nicotine given by either route produced dose-related changes in several self-reported and observer-reported responses. In addition, several physiologic measures also changed as a function of the nicotine dose. A dose-response relationship of considerable generality was most thoroughly documented by Gritz (1980), who concluded that, in general, when nicotine dose is increased, cigarette smoking tends to decrease. Although this conclusion has been subsequently drawn in various reviews (e.g., Henningfield, 1984; US DHHS, 1988), this dose-response relationship has not always been observed, apparently because the relationships between specific independent and GROUP MEANS: 3 MINUTES POST DRUG 1.0

3.0

1.0

SELF-REPORT, Fool d,,'lI

20

1.0

0.0

1.0 (f)

~

00

~

OBSERVED: No ch:Inve due to drug

10

OBSERVED: OrUQ effect

30

17l

OBSERVED: Signs of drug

2.0

~

.5

(f)

'" '" ~ 0

~

::l! '"

oj

1.0

00

1.0

1.0

SELF-REPORT, Relaxod fHllllQ

~ .L

' - - _ - ' - _ - l . J_ _--"

:

I

II

II

.':c,H,------I.-,---~--:2~,9IH 1.41H

PIocIbo

.7'5IY

I.SIV

3.01Y

J..

'

,

SELF-RE~T:

1.0

0.0

0.0 1,41H

.7!lIV

e~ "

II

2.0

OllSERVED, _

.5

0.0

0.0

1.!5IV

2.9IH

3,OIV

DeWe10smoke

I~ .1.

"If

'

41M

PtocIbo

.7'5IV

1.41H 1.!51V

2.9IH 3.01Y

Figure 1. Data points are mean values (n = 8 subjects X 4 sessions) of the difference between the predrug values and those taken approximately 3 minutes following drug administration. Open circles = inhaled nicotine from research cigarettes (doses refer to nicotine yield ratings of the cigarettes); filled circles = intravenous nicotine. (Henningfield, Miyasato, & Jasinski, 1985)

304

J. E. Henningfield and P. P. Woodson

dependent variables are often complex or not even detected by the measures used. In addition, specific data have sometimes been interpreted differently by different reviewers. For example,Schachter (1977) concluded that decreases in cigarette consumption of approximately 25% when yield ratings of the cigarettes increased by 433% was evidence of titration to maintain relatively stable nicotine intake. Other theoreticians (e.g., Russell, 1979) suggested that rather than nicotine dose titration, what was being observed was diminished responding in response to high and aversive doses of nicotine. It is also possible that in some studies that involved an attempt to manipulate dose, the attempt itself was thwarted by unmeasured changes in smoking behavior by subjects (d. Henningfield, 1984). The boundary model of nicotine dose intake regulation proposed by Kozlowski and his colleagues provides another interpretation of the data (Benowitz, Jacob, Kozlowski, & Yu, 1986; Kozlowski & Herman, 1984). This model suggests that the extant data are overwhelmingly sufficient to indicate that people did adjust their nicotine intake in response to changes in delivery, but that the relations were not precise and were best described by broad upper and lower boundaries. The boundary model is a descriptive summary of data in which it is hypothesized that self-administration of high doses may be acutely aversive, leading to diminished intake, whereas low doses may lead to aversive symptoms of nicotine withdrawal, leading in turn to increased intake. Responses Are Differentially Affected by Changes in Nicotine Dose A second point illustrated in Figure 1 is that the magnitude of the changes as a function of dose differs across response measures. One possible explanation for this across-measure variation is that the biologic responses are differentially sensitive to nicotine and to nicotine dose manipulations. Figures 2 and 3 provide additional illustrations of this phenomenon. Volunteer cigarette smokers were permitted to smoke without restriction (except to use a cigarette holder that was used to collect data) during 90-minute sessions in which they had access to a television and reading materials. Thirty minutes before each session, the subjects were given two pieces of nicotine polacrilex to chew. The gum contained a total of 0, 2, 4, or 8 mg nicotine. Analysis of chewed gum and blood samples confirmed that the standardized chewing procedure (see Nemeth-Coslett, Henningfield, O'Keeffe, & Griffiths, 1987) had resulted in orderly increases in nicotine dose administered. Postsession plasma nicotine levels were also increased as a direct function of pretreatment dose. Figure 2 shows that cigarette smoking decreased as a function of increases in the nicotine dose; however, the sensitivity of the response varied across measures. For example, only latency to the first puff of the session was affected at the 2 mg dose; significant decreases in puffing, cigarettes smoked,

Dose-related Effects of Nicotine

305

and carbon monoxide (CO) intake required 4 mg; puffs per cigarette did not change under any condition. Figure 3 shows subjective report data obtained by questionnaire. In this study, the rating of "strength" of the dose was quite sensitive, being significantly increased by 2 mg and further increased by the 4 and 8 mg doses. Another subjective response, "liking," was similarly responsive but as an inverse function of dose level. In contrast, ratings of "desire to smoke" were not significantly changed by any of the doses, although there was a weak inverse relationship that approached significance at the 8 mg dose. Of the many studies of the effects of variations in nicotine dosing of cigarettes, or of nicotine pretreatments (see reviews by Gritz, 1980; Henningfield, 1984; and U S DHHS, 1988), none has utilized such a broad range of behavioral measures, and many have relied only upon measures that this study showed to be rather weak. For example, the most commonly used measures are cigarettes smoked, which showed less than a one-half cigarette per session decrease even at the 8-mg dose, and "craving"-related indices such as desire to smoke, which was not significantly decreased in this study.

THE VEHICLE OF NICOTINE DELIVERY CAN AFFECT THE NATURE AND MAGNITUDE OF THE NICOTINE-ASSOCIATED RESPONSE Inextricably linked to the drug itself is the route and vehicle of drug delivery. Route and vehicle are factors that influence the resulting response by (a) affecting the kinetics (pattern and rate of drug absorption and elimination), (b) determining the amount of administered drug that is systemically absorbed (via entry into the blood), and (c) providing ancillary stimulation due to the presence of other chemicals. In addition, response measures can be altered by previous history, or lack of history, of exposure to that route and by social/cognitive attributions of the route. These factors are not mutually exclusive and may often be co-operative. Nicotine preloading via cigarette smoke inhalation, nicotine polacrilex chewing, intravenous injection, and swallowing of nicotine-containing capsules has been shown capable of decreasing the subsequent behavior of cigarette smoking. Of these, the route of smoke inhalation (e.g., Chait, Russ, & Griffiths, 1985) appears most efficacious and potent, whereas the gastrointestinal route (e.g., Jarvik, Glick, & Nakamura, 1970) is certainly the least potent and possibly less efficacious. The relatively weak effects obtained with swallowed nicotine capsules may be explained by the effects of first-pass liver metabolism of nicotine when given by this route, as well as by the relative absence of sensory stimulation that may be important in the reduction of smoking and satiation of urges to smoke (US DHHS, 1988). With regard to measures of desire to smoke, Figure 1 shows that at the highest nicotine dose, cigarette smoke inhalation and intravenous delivery both produced reductions. Cigarette smoke, however, also reduced desire at low dose levels; in fact, the reduction in desire to smoke was as effective at

306

J. E. Henningfield

-

4

.2'

3

VI

~

ev ....

0

-... 0

0

ev .c

E

2

L

:::I

Z

I 4-4-

VI VI

~ 22 ....

ev a. en

:::I Q..

......ev ~

*

~ c: ev~

u

7»" -~-

-

O

-"-

:::I Q..

L....L..

*

600 u

-VI ev VI

.=-c: o

s .~

400

g(/)

...J

... VI

8

I

t-

!

7

L 9

0 0

CD

ev

*

"0

~

c:

a.

E 7 o a.

';C

.J..

I

*

0_

>.ev

ev... 0 o

L--I

VI

Q..~

U.

...L.

ev a.

1600

:::I

*+

9

0'

2000

C7'

C

L

0

o

...u ev

I

::: 20

2400

~

24

0

.J..

"-

-

c:

and P. P. Woodson

200

L

:IE

2

I .J..

L o '_ _- - '_ _- - '

P 2

4

8

(;

0

5

, -e......LP 2:::--~-----I , 4

I

8

Nicot i ne Dose (mg) Figure 2. Data points are mean values (n = 8 subjects X 4 sessions) of measures of cigarette smoking collected during 90 minute test session; bracketts indicate 1 SEM. The x-axis shows the nicotine gum dose; P = placebo gum. Asterisks indicate that the value was significantly different from that obtained at placebo (p < 0.05); plus signs indicate that value was significantly different from that obtained at 2 mg. (NemethCoslett, Henningfield, o 'Keeffe, & Griffiths, 1987)

the 0.4 mg dose delivered via tobacco smoke, as at the 1.5 mg dose delivered intravenously, and was nearly as effective as the 3.0 mg dose delivered intravenously. Presumably, these differences are related to the more efficient delivery of nicotine via the intrapulmonary route of smoke inhalation, the additional contribution of the array of sensory stimuli provided by cigarette smoke inhalation, and the presence of substances in tobacco smoke (e.g., carbon monoxide, alkaloids) other than nicotine.

Dose-related Effects of Nicotine

...

s:

5 QJ

""o

~4

QJ

;; 3

lJ)

QJ III

0

a

2

E

I

lJ)

{:.

...

QJ III

E

::J ~

QJ

a

.

0 3

CI

::CI

...

0' C

""

2

..J

E ::J

o

0

I ..L

C

ItJl

U o

~

CI

o

c:

*

~+ ,

,

I

CI

'0

.~ ~

l:7'

W

4 8 2 8 Nicotine Dose (mg) Figure 3. Data points are mean values (n = 8 subjects X 4 sessions) of subject ratings of variables shown on the y-axis following 90 minute test sessions; brackens indicate 1 SEM. The x-axis shows the nicotine gum dose; P = placebo gum. Asterisks indicate that the value was significantly different from that obtained at placebo (p < 0.05); plus signs indicate that value was significantly different from that obtained at 2 mg. (Nemeth-Coslett, Henningfield, O'Keeffe, & Griffiths, 1987) P

2

4

THE MAGNITUDE OF A RESPONSE TO A GIVEN DOSE OF NICOTINE IS INVERSELY RELATED TO THE DEGREE OF TOLERANCE RESULTING FROM PRIOR EXPOSURE The reduction in responsiveness occurring when nicotine doses are repeatedly given has been systematically studied for nearly a century. Pioneers in the study of tolerance to drugs in general, and to nicotine in particular, were Langley and Dixon. Langley reported diminished responsiveness to repeated doses of nicotine in a variety of species and response measures (Langley, 1905, 1907-1908). Figure 4 shows data from a study by Dixon and Lee (1912) on cardiovascular responses to nicotine administration in a study using rabbits as subjects. With respect to the electroencephalographic activating effects of nicotine, rapid tolerance occurs, with at least 30 minutes being needed between nicotine injections to repeat the initial magnitude of the effect (Domino, 1983). Multiple mechanisms account for the development of tolerance to nicotine and have been reviewed (see US DHHS, 1988). There have also been many studies of tolerance to the effects of nicotine in human subjects (see reviews,

308

J. E. Henningfield and P. P. Woodson

B

...

•

...

A

-wi"

wi

a

e ';

. . . . . . . .

_

Figure 4. The data presented are continuous recordings of the blood pressure of decerebrate rabbits. The effects of a nicotine injection on an animal pretreated with nicotine every other day for one month prior to the test are shown in panel B; panel A shows data from an untreated (nontolerant) rabbit. (Dixon & Lee, 1912)

US DHHS, 1988). Jones, Farrell, and Heming (1978) reported that both nicotine-induced increases in heart rate and subjective effects decreased when doses were repeated at hourly intervals. Similarly, as shown in Figure 5, Henningfield (1984) showed that nicotine injections repeated at 10 min intervals resulted in pronounced tolerance to the "positive" ("desirable") effects of nicotine in a human subject. Studies of nicotine tolerance suggest that the basic phenomena are not unlike tolerance-related phenomena observed when other dependence-producing drugs are given in other species for a wide range of response measures (see reviews by Abood, 1984; Martin, 1977; US DHHS, 1988). A few generalities might be drawn from such observations: (a) the degree of tolerance development varies across response measure; (b) increasing dose levels of the drug is often partially but not always completely effective at restoring responsiveness; (c) tolerance development is often partial, such that some level of responsiveness generally remains; and (d) development of tolerance to a

309

Dose-related Effects of Nicotine

drug implies that diminished responsiveness will be observed when alternate formulations of the drug are given. Implications of these observations for replacement therapy of tobacco dependence are discussed further on in this paper.

~ 100 C) o

H-ALB

...J

w « a::z

o «

Ow (f)

Z

~..J ::l'

...J

50

I

Mec Pretreat

..J

« :J (f)

>

o

I 5

10

I I 15

NICOTINE INJECTIONS (3.0mg) Figure 5. Data points are scores on visual line analogue scales of positive ("liked") effects by a human volunteer following 3.0 mg intravenous injections of nicotine. These injections were given at 10 minute intervals. The function indicated as "Mec Pretreat" shows data collected on a different test day in which the test session occurred 60 minutes following an oral dose of 10 mg mecamylamine. (Henningfield, 1984)

Dose Is a Determinant of the Nature of the Response

Nicotine may produce stimulating actions when given at low dose levels, whereas depressant-like actions may be produced when the nicotine dose levels are high and/or when exposure is prolonged. Observations of dual stimulant/depressant actions are not unique to nicotine but rather, resemble observations on a number of other drugs, some categorized as sedatives and others as stimulants (e.g., Gilman et al., 1985; Thompson & Schuster, 1968). This complexity of nicotine's effects created, early on, differences of opinion as to whether nicotine should be categorized primarily as stimulant-like (Lewin, 1964) or sedative-like (Armstrong-jones, 1927). Some examples are provided below of how nicotine dose, in combination with the response measure itself as well as the time at which the measure is collected, may determine whether nicotinic effects are "stimulant-like" or "depressant-like" in nature. Responses mediated by the central nervous system may be either stimulantor depressant-like. For example, Ashton, Marsh, Millman, Rawlins, Telford, and Thompson (1980) showed that low intravenous doses of nicotine produced stimulant-like effects on the electroencephalographically-measured contingent

310

J. E. Henningfield and P. P. Woodson

negative variauon (CNV) response in human subjects; higher intravenous doses of nicotine produced depressant-like effects on the CNV response. Woodson, Buzzi, Nil, and Battig (1986) showed that smoking can produce diminished skin conductance amplitudes (depressant-like action), yet simultaneously induce heart-rate acceleration, peripheral vasoconstriction, and increased pulse velocity (stimulant-like activity); in turn, these effects of smoking were accompanied by a suppression of noise-induced tachycardia as well as a partial inhibition of noise-induced vasoconstriction. Such studies indicate that stimulation or depression depends not only on dose (as is the case with the CNV, for example), but also on the response system measured (cardiovascular vs, electrodermal), as well as on the conditions under which measurements are being made (noise vs. quiet). An example of how reliance upon a single response or use of a single dose level might be misleading in an effort to categorize nicotine as either predominantly a stimulant or predominantly a depressant is provided by the ganglionically-mediated actions of nicotine on skeletal muscle activity. Domino and Von Baumgarten (1969) found that the patellar reflex of human subjects was reduced by nicotine in a dose-dependent manner. These effects were presumably mediated, however, by nicotine's initial stimulation of Renshaw cells jn the spinal cord, which led to inhibition of responsiveness of certain skeletal muscles (US DHHS, 1988). To further complicate the interpretation, (a) tolerance occurs, such that repeated or prolonged exposure to nicotine. results in restoration of the responsiveness of the patellar reflex, and (b) different muscle groups are affected differentially (e.g., muscle tonus may increase in trapezius muscle; Fagerstrom & Gotestam, 1977). Behavioral effects of nicotine can also be either increased or decreased, depending upon the specific response measured and the dose given. One study compared the effects of nicotine and d-amphetamine on the behavior of squirrel monkeys working under different contingencies (reinforcement schedules; Spealman, Goldberg, & Gardner, 1981). In one experiment, the monkeys. pressed a lever that led to food delivery, whereas in another, they pressed a lever that prevented the delivery of impending electric shocks. Different doses of either d-amphetamine or nicotine were given prior to the test sessions. In general, the effects of nicotine and amphetamine were similar, and it mattered little whether the animals were working to avoid electric shock or to obtain food. Major determinants of the resulting behavior were (a) the dose of the drug, and (b) the specific schedule of reinforcement, for example, whether the response-shock or response-food relationship was determined by the number of responses emitted (fixed-ratio schedule) or the temporal pattern of responding (fixed-interval schedule). As shown in Figure 6, increasing doses of both nicotine and d-amphetamine first increased, then decreased response rates on the fixed-interval schedule, and both drugs produced dose-related decreases in responding on the fixed-ratio schedule. The nature of the response produced by a drug is often used to categorize the drug and to develop predictive theories of drug action. Inaccurate theoretical predictions may result, however, when not made with consideration

Dose-related Effects of Nicotine

311

FI XED INTERVAL

STIMULUS-SHOCK TERMINATION

220 200 180160 -

6

FIXED RATIO

140

120 a:: 100 .... 80 Z o 60

u

40

~

20

....oZ

0

C S .01.03.1 .3

W

I 3

n,...,

iii

C S'oI.03.1 .3

I 3

FOOD PRESENTATION

u a:: 280

w a. 260 240

w 220 ~

200

a:: 180

w 160

en oZ enWCl.

a::

.....

NICOTINE

0-0

!!-AMPHETAMINE

140

120

t?e~-----

100

80 60

40 20

o

n n

,....-..--,.......,r---'~

C 5 .01 .03.1 ,3

I 3

nn~ C 5.01,03,1.3 I 3

DOSE (mg/kg) Figure 6. Data points are mean values from one or two test sessions from each of three or four squirrel monkeys. Each value is the overall rate of lever pressing that occurred during either the fixed-interval (left side panels) or fixed-ratio (right side panels) schedules of electric shock avoidance (top panels) or food reinforcement (bottom panels). The dashed lines indicate the mean values obtained when no injection was given ("C") or when saline ("S") injections were given. Injections of saline, nicotine, or d-amphetamine were given intramuscularly immediately before the sessions. (Spealman, Goldberg, & Gardner, 1981)

given to the fact that the dose of a given drug, and not just drug type, can qualitatively determine the nature of the response. For example, Schachter and others (see Nesbitt, 1973; Schachter, 1973; Woodson, 1985; for review see Gilbert, 1979) have developed various theories to reconcile nicotine's predominantly stress-like effects on the autonomic nervous system (i.e., sym-

312

J. E. Henningfield

and P. P. Woodson

pathomimetic effects resulting in end-organ arousal such as tachycardia, increased blood pressure, increased vasoconstriction, increased release of circulating catecholamines) and brain (i.e., electroencephalographic alpha blocking coupled with induction of hippocampal theta) with its predominantly tranquilizing effects on mood and behavior, especially when the organism is under stress. Briefly, one theory of Schachter states that nicotine drives the nervous system into an activated state. According to the "law of initial value," when stress is applied during this activated state, little or no further activation can occur. As the smoker normally identifies a stressor via the physiological activation which it induces, the smoker may conclude that no stress is present under the smoking state since little or no change from baseline was detected. This cognitive theory, based on the "law of initial value," may be contrasted with the psychophysiological theory of Woodson (1985), which states that nicotine, in addition to producing physiologically activating effects, also induces a second effect, distinct from the first, which may be characterized as physiologically stabilizing in nature. This stabilizing effect reduces the plasticity of nervous responsitivity to stress and thereby serves to protect the organism psychopharmacologically. Regardless of the mechanisms underlying the biphasic effect, attention need, to be given to the dose-related determinants of the nature of the response. Such consideration will facilitate the rational use of nicotine replacement to treat cigarette smoking (see review by Jarvik & Henningfield, 1988).

IMPLICATIONS OF DOSE-RELATED ACTIONS OF NICOTINE FOR REPLACEMENT THERAPY Nicotine polacrilex (Nicorette®) is available as a replacement source of nicotine that is used to assist smokers to maintain abstinence from tobacco (see reviews in American Hospital Formulary Services, 1987; Grabowski & Hall, 1985; Pomerleau & Pomerleau, 1988; US DHHS, 1988). Its safety and efficacy have been confirmed in clinical trials (American Hospital Formulary Services, 1987; US DHHS, 1988); however, since the percentage of outcomes in which abstinence is achieved ranges from about 10 to 50%, there is clearly room for improvement in this area. These high rates of relapse are not surprising and in fact are comparable to those obtained following treatment of other drug dependencies (US DHHS, 1988). Several lines of evidence suggest that consideration of some of the dose-related issues relevant to replacement chemotherapy for nicotine dependence could enhance clinical outcome. Specifically, by understanding the factors that may complicate attempts to maintain desired dosing regimens, it may be possible to devise practical strategies to improve treatment. This section reviews some characteristics of the nicotine polacrilex formulation and provides some specific suggestions to enable better control of the nicotine dose delivered via this route. The therapeutic intent of nicotine polacrilex use is to reduce the unpleasant symptomatology of the nicotine withdrawal syndrome, as well as to reduce

Dose-related Effects of Nicotine

313

directly the tendency to smoke, much as caloric preloading reduces the tendency to eat available food, or at least decreases the amount consumed (Henningfield & Jasinski, 1988). The degree to which the nicotine polacrilex suppresses the urge to smoke cigarettes is limited by the dose of nicotine absorbed as well as by characteristics of the delivery vehicle itself. For example, the sensory properties (taste, smell, appearance, touch) of nicotine polacrilex differ considerably from those of cigarettes and thus may provide less satisfaction than smoking, in that the gum produces an unpleasant taste and can fatigue the jaw muscles from the extensive chewing required to extract the nicotine. In addition, the hundreds of thousands of pairings of nicotine delivery (in each of the approximately 70,000 puffs taken each year in the 20 cigarette per day smoker) with its neuroregulatory consequences (Pomerleau & Pomerleau, 1984) may account for strong urges to smoke in situations in which reinforcement was obtained. Such urges may not be as reliably or robustly affected by administration of nicotine polacrilex. The nicotine withdrawal syndrome itself includes heightened urges to smoke; these urges seem to be reduced by nicotine replacement, but this effect is also not robust (US DHHS, 1988). The main limitation to the achievement of maximal efficacy of the polacrilex that is relevant to this paper concerns the issue of control over dosing. All of the effects of nicotine described in this paper are determined in part by the dose of nicotine and are not simple all-or-none consequences of any level of nicotine exposure. This conclusion is not unique to nicotine; the importance of controlling dose and determining the proper dosing regimen is a critical factor in other forms of chemotherapy for behavioral and physiologic disorders. Measurement and control of nicotine dose when administered either by cigarette smoke inhalation or by nicotine polacrilex chewing is unusually complicated. For example, the typical American cigarette contains about 10 mg of nicotine, of which approximately .5 to 1.5 mg is actually absorbed (US DHHS, 1988). Nicotine polacrilex contains either 2 or 4 mg of nicotine, of which between 40 and 60% appears to be extracted under standardized chewing conditions and somewhat less actually absorbed, resulting in delivery of an estimated .86 mg from the 2 mg formulation and 1.2 mg from the 4 mg formulation (Benowitz, Jacob, & Savanapridi, 1987). With both cigarettes and nicotine polacrilex gum, however, a variety of factors affects actual nicotine extraction as well as absorption. With both delivery systems, either practice or explicit instruction may be necessary to achieve desired effects rapidly. Control of nicotine administration in persons to whom nicotine polacrilex is given may be impaired by lack of experience with this route of drug administration and by insufficient instructions. Table 2 lists various conditions that may preclude reliable administration of therapeutic nicotine doses or otherwise reduce the validity of conclusions resulting from nicotine polacrilex administration in research or therapeutic applications. As indicated in the table, a variety of factors can prevent intended dosing, and failure to consider them may lead to false negative conclusions

314

J. E. Henningfield and P. P. Woodson

TABLE 2. Dose-Related Reasons for Lack of Response to Nicotine Polacrilex Administration •

Insufficient number of polacrilex units were administered (clinicians in countries where the 4 mg preparation is approved have the option of varying the unit dose level and/or the number of units administered).

•

Chewing patterns that were either too slow or insufficiently vigorous may have produced inadequate nicotine extraction from the polacrilex.

• The nicotine may have been swallowed rather than absorbed through the buccal mucosa. Nicotine extracted from the polacrilex into the saliva must be held in the mouth for at least 30 to 60 seconds. • Absorption may have been reduced by consumption of an acidic food (e.g., tomatoes) or beverage (e.g., coffee or soft drinks), within a few minutes prior to gum use, and even blocked altogether by simultaneous consumption of acidic substances and nicotine polacrilex. • The response measure itself (e.g., urges to smoke) may not have been a sensitive and reliable indicator of therapeutic doses of nicotine administration.

regarding the efficacy of the polacrilex, that is, that the treatment did not work, when in actuality the treatment was not provided adequately. Tbe data reviewed in this paper and elsewhere (e.g., Jarvik & Henningfield, 1988; US DHHS, 1988) also have implications for clinical treatment of tobacco dependence. Some of these, as related to nicotine dose, are summarized in Table 3. It would seem plausible that consideration of such implications might lead to improved treatment or at least to more definitive conclusions regarding efficacy of the nicotine polacrilex. A final conclusion follows from the observations concerning the importance of controlling nicotine dose and ensuring therapeutic dosing: Establishment of adequate dosing may be constrained where only one dose of the polarcilex gum formulation is available, as in the United States. This issue is especially problematic during the initial stages of replacement therapy, when inadequate nicotine dosing may be accompanied by severe withdrawal symptoms. For example, a person who has been smoking 40 cigarettes per day may be accustomed to an intake of approximately 40-60 mg of nicotine by this highly efficient route of nicotine delivery. Achieving even 50% of such nicotine dose levels via nicotine polacrilex could require the chewing of 20 to 35 2 mg pieces per day, entailing most of the waking day of the smoker. Such a dose regimen may be unacceptable or even impossible for some people; minimally it requires a considerably greater physical response than do most other widely-used medications. The availability of a stronger preparation that could reduce the burden of self-administration could provide a major clinical benefit. Higher doses of nicotine polacrilex may be less critical after initial stages of nicotine replacement, when nicotine dose dependence has been substantially reduced. At later points in therapy, it may be sufficient for patients to use the replacement on a symptomatic basis, at low overall doses (e.g., an occasional piece of the polacrilex when pressure to relapse is intense). Other routes of nicotine replacement also may be necessary to

315

Dose-related Effects of Nicotine

TABLE 3. Summary of Implications of Dose-Related Actions of Nicotine for Pharmacologic Replacement Therapy. • Several potential indices of a therapeutic response to nicotine polacrilex gum should be used in clinical practice (e.g., ability to concentrate, control mood, and handle other nicotine withdrawal symptoms). • Measures such as "craving for cigarettes" and "urges" to smoke are not reliably controlled by therapeutic doses of nicotine and should not serve as primary indices of treatment efficacy. The physician should be aware, however, that the continued presence of urges to smoke during gum administration may lead the patient to conclude that the gum is not effective. This could lead to poor compliance with the dosing regimen which could lead in turn to smoking relapse. The patient should be reassured that, with proper compliance, these urges will gradually abate, although perhaps never completely. • It should not be concluded that the polacrilex was not efficacious in a given treatment intervention if there was no means of confirming that known therapeutic doses of nicotine were actually extracted and absorbed (e.g., chemical verification or use of a method of using the gum previously shown to result in adequate dosing). • Confirmation that the patient has learned to extract and absorb nicotine can be accomplished by (a) chemical verification or (b) determination that use of the polacrilex when nicotinedeprived (e.g., after sleeping for several hours) results in a discriminable subjective effect. • Since the within-day development of tolerance to nicotine (from either polacrilex or tobacco use) may prevent discrimination of effects of nicotine administration, clear instructions should be given to ensure maintenance of therapeutic dose administration.

provide acceptable treatment for some tobacco-dependent persons (see Russell, 1988, and U.S. DHHS, 1988 for a review of alternative nicotine delivery systems). REFERENCES Abood, L.G. (1984). Mechanisms of tolerance and dependence: An overview. In C.W. Sharp (Ed.), Mechanisms of tolerance and dependence (NIDA Research Monograph 54, DHHS Publication No. ADM 84-1330, pp. 4-11). Washington, DC: U.S. Government Printing Office. American Hospital Formulary Services (1987). Drug information. Miscellaneous autonomic drugs: Nicotine polacrilex (pp. 657-662). Bethesda, MD: American Society of Hospital Pharmaceuticals. Armstrong-Jones, R. (1927). Tobacco, its use and abuse: From the nervous and mental aspect. Practitioner (London), 118, 6-19. Ashton, H., Marsh, V.R., Millman, J.E., Rawlins, M.D., Telford, R., & Thompson, J.W. (1980). Biphasic dose-related responses of the CNV (contingent negative variation) to i.v. nicotine in man. British Journal of Clinical Pharmacology, 10, 579-589. Benowitz, N.L., Jacob, P., III, Kozlowski, L.T., & Yu, L. (1986). Influence of smoking fewer cigarettes on exposure to tar, nicotine, and carbon monoxide. Nno England Journal of Medicine, 315, 1310-1313. Benowitz, N.L., Jacob, P., III, & Savanapridi, C. (1987). Determinants of nicotine intake while chewing nicotine polacrilex gum. Clinical Pharmacology and Therapeutics, 41, 467-473. Chait, L.D., Russ, N. W., & Griffiths, R.R. (1985). Effects of graded smoke inhalation on subsequent cigarette smoking behavior. Addictive Behaviors, 10, 273-280.

316

J. E. Henningfield and P. P. Woodson

Dixon, W.E., & Lee, W.E. (1912). Tolerance to nicotine. Quarterl),Journal of Experimental Phssiolog; (London), 5, 373-383. Domino, E.F. (1983). Neuropsychopharmacology of nicotine and tobacco smoking. Psvchopharmacolog: Bulletin, 19, 398-401. Domino, E.F., & Von Baumgarten, A.M. (1969). Tobacco cigarette smoking and patellar reflex depression. Clinical Phnrmacologx and Therapeutics, 10, 72-79. Fagerstrom, K.O., & Gotestam, K.G. (1977). Increase of muscle tonus after tobacco smoking. Addirtn» Behariors, 2, 203-206. Gilbert, D.G. (1979). Paradoxical tranquilizing and emotion-reducing effects of nicotine. Psychological Bulletin, 86, 643-661. Gilman, A.G., Goodman, L.S., Rail, T.W., & Murad, F. (Eds.). (1985). Goodman and Gilman's the pharmacological basis of therapeutics (7th ed.), New York: Macmillan. Grabowski, J., & Hall, S.M. (Eds.). (1985). Pharmacological adjuncts in smoking cessation (NIDA Research Monograph 53, DHHS Publication No. ADM 85-1333). Washington, DC: U.S. Government Printing Office. Gritz, E.R. (1980). Smoking behavior and tobacco abuse. In N.K. Mello (Ed.), Adrances in substance abuse: Behavioral and biological research (Vol. 1, pp. 91-158). Greenwich, CT: JAI Press. Henningfield,J.E. (1984). Behavioral pharmacology of cigarette smoking. In T. Thompson, P.B. Dews, &J.E. Barrett (Eds.), AdmnCfS in behavioral pharmacolog; (Vol. 4, pp. 131-210). New York: Academic. Henningfield, J.E., & Jasinski, D.R. (1988). Pharmacological basis for nicotine replacement. In O.F. Pomerleau & C.S. Pomerleau (Eds.), Nicotine replacement: A critical evaluation. (pp, ~5-61). New York: Alan R. Liss. Henningfield, J.E., Miyasato, K., & Jasinski, D.R. (1985). Abuse liability and pharmacodynamic characteristics of intravenous and inhaled nicotine. Journal of Pharmacology and Experimental Therapeutics, 234, 1-12. Jarvik, M.E., Glick, S.D., & Nakamura, R.K. (1970). Inhibition of cigarette smoking by orally administered nicotine. Clinical Pharmacology and Therapeutics, II, 574-576. Jarvik, M.E., & Henningfield, J.E. (1988). Pharmacological treatment of tobacco dependence. Pharmacology, Biochemistry & Behavior, 30, 279-294. Johnston, L.M. (1942). Tobacco smoking and nicotine. Lancet, 2, 742. Jones, R.T., Farrell, T.R., III, & Heming, R.I. (1978). Tobacco smoking and nicotine tolerance. In N.A. Krasnegor (Ed.), Self-administration oj abused substances: Methods Jor study (NIDA Research Monograph 20, DHEW Publication No. ADM 78-727, pp. 202-208). Washington, DC: U.S. Government Printing Office. Kozlowski, L.T., & Herman, c.P. (1984). The interaction of psychosocial and biological determinants of tobacco use: More on the boundary model. Journal oj Applied Social Psycholog)" 14, 244-256. Langley, J.N. (1905). On the reaction of cells and of nerve-endings to certain poisons, chiefly as regards the reaction of striated muscle to nicotine and to curari. Journal oj Physiology (London), 33, 374-413. Langley, J.N. (1907-1908). On the contraction of muscle, chiefly in relation to the presence of "receptive" substances. Part I. Journal oj Ph)'siology (London), 36, 347-384. Langley, J.N. (1914). The antagonism of curari and nicotine in skeletal muscle. Journal of Physiolog)' (London], 48, 73-108. Langley, J.H., & Dixon, W.L. (1889). On the local paralysis of the peripheral ganglia and on the connexion of different classes of nerve fibres with them. Proceedings oj the Royal Society oj London (Biology), 46, 423-431. Larson, P.S., Haag, H.B., & Silvette, H. (1961). Tobacco: Experimental and clinical studies. A comprehensire account of the world literature. Baltimore: Williams & Wilkins. Larson, P.S., & Silvette, H. (1968). Tobacco: Experimental and clinical studies. A comprrhensio» account oj the world literature (Supplement 1), Baltimore: Williams & Wilkins. Larson, P.S., & Silvette, H. (1971). Tobacco: Experimental and clinical studies. A comprrhrnsm« account oj the world literature (Supplement 2). Baltimore: Williams & Wilkins.

Dose-related Effects of Nicotine

317

Larson, P.S., & Silvette, H. (1975). Tobacco: Experimental and clinical studies. A comprehensive account of the world literature (Supplement 3). Baltimore: Williams & Wilkins. Lewin, L. (1964). Phantastica: Narcotic and stimulating drugs, their USf and abuse (P.H.A. Wirth, Trans.). New York: E.P. Dutton. (Original work published in second German edition, 1927) Martin, W.R. (1977). Assessment of depressant abuse potentiality. In T. Thompson & K.R. Unna (Eds.), Predicting dependence liability of stimulant and depressant drug» (pp. 9-15). Baltimore: University Park Press. Nemeth-Coslett, R., Henningfield, J.E., O'Keeffe, M.K., & Griffiths, R.R. (1987). Nicotine gum: Dose-related effects on cigarette smoking and subjective ratings. Psvchophannacolog: (Balin), 92, 424-430. Nesbitt, P.O. (1973). Smoking, physiological arousal, and emotional response. [ournal of Personality and Social Psychology 25, 137-144. Pomerleau, O.F., & Pomerleau, C.S. (1984). Neuroregulators and the reinforcement of smoking: Towards a biobehavioral explanation. Neuroscience and Biobrhaoioral Rerieu.s, 8, 503-513. Pomerleau, O.F., & Pomerleau, C.S. (Eds.). (1988). Nicotine rrplncement. A critical roatuation. New York: Alan R. Liss. Russell, M.A.H. (1979). Tobacco dependence: Is nicotine rewarding or aversive? In N.A. Krasnegor (Ed.), Cigarfttf smoking as a dependence process (NIDA Research Monograph 23, DHEW Publication No. ADM 79-800, pp. 100-122). Washington, DC: U.S. Government Printing Office. Russell, M.A.H. (1988). Nicotine replacement. The role of blood nicotine levels, their rate of change, and nicotine tolerance. In O.F. Pomerleau & C.S. Pomerleau (Eds.), Nicotine replacement. A critical evaluation (pp, 63-94). New York: Alan R. Liss. Schachter, S. (1973), Nesbitt's paradox. In W.L. Dunn, Jr. (Ed.), Smoking behavior: Motn.es and incflltil'fS (pp. 147-155). Washington, DC: V.H. Winston. Schachter, S. (1977). Nicotine regulation in heavy and light smokers. Joumal uf Experimental PS,1'dlO!og)' (Cenrml), 106, 5-12. Spealman, R.D., Goldberg, S.R., & Gardner, M.L. (1981). Behavioral effects of nicotine: Schedulecontrolled responding by squirrel monkeys. The [ournal of P/umnaco!og.1' and Experimental Therapeutics, 216, 484-491. Thompson, T., & Schuster, C.R. (1968). Beharioral jJlzarlllacolog,r. Englewood Cliffs, NJ: PrenticeHall. U.S. Department of Health and Human Services (1988). The healtti conSN/umffS of smoking: Nicotine addiction, a report of the Surgfon General (DHHS Publication No. CDC 88-8406). Washington, DC: U.S. Government Printing Office. Woodson, P. P. (1985). The neuropharmacological double action of nicotine: Mediation of cigarette smoking's energizing yet stabilizing effects on psychophysiological function (Swiss Federal Institute of Technology Dissertation No. 7560). Dissertation Abstracts International, 45, 2640C. Woodson, P,P., Buzzi, R" Nil, R., & Biittig, K. (1986). Effects of smoking on vegetative reactivity to noise in women. Ps)'rllOph)'siolog)', 23, 272-282.