Annals of Medicine

ISSN: 0785-3890 (Print) 1365-2060 (Online) Journal homepage: www.tandfonline.com/journals/iann20

Brain imaging and the effects of caffeine and

nicotine

Stephen R. Dager & Seth D. Friedman

To cite this article: Stephen R. Dager & Seth D. Friedman (2000) Brain imaging and the effects
of caffeine and nicotine, Annals of Medicine, 32:9, 592-599, DOI: 10.3109/07853890009002029

To link to this article: https://doi.org/10.3109/07853890009002029

Published online: 27 Sep 2010.

Submit your article to this journal

Article views: 3508

View related articles

Citing articles: 4 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=iann20
5 92

+ REVIEW ARTICLE *

Brain imaging and the effects

. of caffeine * e

and nicotine
Stephen R Dager',2j3and Seth D Friedman2

Caffeine and nicotine are the most common psychostimulant drugs used worldwide.
Structural neuroimaging findings associated with caffeine and nicotine consumption are
limited and primarily reflect the putative relationship between smoking and white matter
hyperintensities (WMH), a finding that warrants further appraisal of its clinical
implications. The application of newer brain imaging modalities that measure subtle
haemodynamic changes or tissue-based chemistry in order to better elucidate brain
functional processes, including mechanisms underlying addiction to nicotine and caffeine
and the brain functional consequences, provide intriguing findings. Potential influences
of caffeine and nicotine on the functional contrast, or metabolic response, to neural
activation also necessitates the careful appraisal of the effects that these commonly used
drugs may have on the results of functional imaging.
Key words: brain imaging; caffeine; functional imaging; magnetic resonance imaging; nicotine;
magnetic resonance spectroscopy.
Ann Med 2000; 32: 592-599.

Introduction resonance imaging and spectroscopy (MRI/MRS)

Tremendous advances made in brain imaging tech-
nology during the past two decades have provided
both improved 2-D and 3-D structural definition and
quantitative assessment of discrete brain substructures
(1, 2), as well as the ability to systematically appraise
brain functionality. The latter application has garnered
substantial interest in the scientific community, initially
with positron emission tomography (PET) and single-
photon emission tomography (SPECT) and more
recently through the use of magnetic resonance (MR)
techniques ( 3 ) . Although it is outside the scope of
this article to provide an in-depth review of magnetic
From the Departments of 'Psychiatry and Behavioral Sciences,
'Radiology and 'Bioengineering, University of Washington, Seattle,
WA, USA.
Correspondence: Stephen R Dagcr, MD, University of
Washington School of Medicine, Departments of Psychiatry and
Behavioral Sciences, Radiology and Bioengineering, 4225 Roosevelt
Way NE, Suite 306-C, Seattle, WA 98105-6099 USA. E-mail:
srd~u.washinaton.edu.Fax: +1 206 5437565.
(see (3-5)), these noninvasive MR techniques go
beyond structural assessment of the brain. Expanded
MR capabilities allow clinical appraisal of tissue-
based chemical composition, metabolic regulation
and functional activation, the latter resulting from
regional decoupling between energy demand and
blood flow in response to neural activation (6, 7).
These expanded brain imaging capabilities are
being used increasingly to elucidate the patho-
physiological underpinnings of addiction and the
brain consequences of drug abuse (8). However,
relatively little research has been undertaken in
humans to evaluate the structural and functional
consequences in brain of caffeine and nicotine con-
sumption, the most commonly used psychoactive
substances worldwide. The focus of this article is to
survey neuroimaging findings related to caffeine and
nicotine and to discuss the potential effects of these
agents on physiological mechanisms responsible for
the functional contrast underlying MR functional
imaging techniques, most typically the regional
mapping of blood oxygen level determination (BOLD
fMRI) or lactate elevations (fMRS) ( 3 ) .

0The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
 BRAINIMAGING A N D THE EFFECTS
Caffeine
Caffeine and its methylxanthine derivatives are gener-
ally consumed via dietary intake from beverages and
foods that include coffee, tea, soft drinks, chocolate
and a variety of over-the-counter medications. The
caffeine (equivalent) content of beverages and food is
quite variable, ranging from 40 to 180 mg per/100
mL for coffee, 2 to 8 mg/15O mL for decaffeinated
coffee, 16 to 50 mg/150 mL for tea, 15 to 40 mg/lSO
mL for soft drinks, and 1 to 118 mg per 28 g for
chocolates (reviewed in (9)). Per capita caffeine
consumption averages about 75 mg/person/day world-
wide but there are wide cultural differences with some
ethnic groups, such as Scandinavians, reporting rates
of caffeine consumption exceeding 400 mg/person/day
(10).
Because of its strong hydrophobic properties, there
is no blood-brain barrier to caffeine; absorption and
brain uptake of ingested caffeine typically occurs
within about 45 min (11-14). Caffeine has a plasma
half-life of approximately 3-8 h, and brain levels
remain stable for at least 1 h following peak absorp-
tion. The half-life of caffeine is markedly increased
during the neonatal period as a result of an inactive
cytochrome P-450 hepatic enzyme system (1.9, but
otherwise there are no clear age-related changes in
caffeine pharmacokinetics (12-14). The half-life of
caffeine can be reduced by up to 30-50% among
smokers (16) and approximately doubled by oral
contraceptives or medications that alter liver metab-
olism, including psychotropic medications such as
fluoxetine or fluvoxamine (12-14).
The widespread use of caffeine as the psycho-
stimulant of choice primarily reflects its effects on
subjective enhancement of energy, efficiency, self-
confidence, alertness, and sociability ( 13), although
double-blind tests have not demonstrated convincingly
objective evidence of these effects in regular caffeine
users (reviewed in (9)). Although caffeine is not
generally considered a substance of abuse (17),
dependency does occur among individuals who
habitually use caffeine. When caffeine consumption is
abruptly stopped, mild-to-moderate withdrawal symp-
toms can occur within 14-48 h, for some individuals
persisting up to a week, (9, 13, 18).
The primary physiological effect of caffeine for
the concentration range normally consumed is the
competitive blockage of endogenous adenosine (19,
20). Adenosine, a normal cellular constituent, is
widely distributed and strongly inhibits neurotransmit-
ter release. Adenosine may have a role in homeostatic
regulation by matching the rate of energy consumption
with the amount of substrate supply, such as in
response to hypoxia, as adenosine production is
increased in situations of energy deprivation (19). In
contrast to other stimulants, such as amphetamine,
0The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
OF CAFFEINE A N D NICOTINE 5 93
that increase both energy metabolism and cerebral
blood flow (CBF), caffeine increases energy metabol-
ism while at the same time markedly reducing CBF by
30% or more (20). As a consequence of its decoupling
effects that produce concurrent acceleration of energy
metabolism and reduction in CBF, caffeine ingestion
in humans produces regional increases in brain lactate
that can be measured by MRS (21).
Unlike reports of a possible association between
nicotine use and discrete white matter lesions (dis-
cussed elsewhere in this paper), caffeine is not known
to cause brain structural changes detectable by
imaging studies. However, the long-term brain con-
sequences of caffeine consumption have not been
addressed directly in the available literature on
imaging. Although caffeine consumption has been
identified as a potential cardiovascular risk factor
(22), more recent evidence fails to support this
hypothesis and, in fact, suggests an inverse relation-
ship between the risk of fatal or nonfatal stroke and
caffeine consumption (23). Moreover, not only does
chronic caffeine consumption appear to not increase
the risk of cerebral vascular accidents, it may in fact
decrease the extent of associated hypoxic damage
(24). This latter observation presumably reflects the
neuroprotective effects of adenosine receptors up-
regulated by caffeine (19, 24). Recent epidemiological
findings from a large cohort of Japanese-American
males also raises the intriguing possibility that caffeine
consumption may provide prophylaxis against the
occurrence of Parkinson’s disease ( 2 9 , possibly
through enhancement of dopaminergic transmission,
but confirmatory imaging studies and studies on
receptor binding in humans are needed.
A number of PET studies have documented diffuse
reductions in CBF, up to 30% or more, that occur
rapidly in response to orally or intravenously (iv)
administrated caffeine (26-28). More recent imaging
work has capitalized on caffeine’s dual mechanisms
of increasing energy expenditure while reducing
CBF, allowing measurement of regional brain lactate
changes in response to caffeine ingestion (21). A
rapid MRS imaging technique, proton echoplanar
spectroscopic imaging (PEPSI) (29), was used to
demonstrate differential brain lactate increases among
caffeine-sensitive individuals who avoided caffeinated
products in comparison to regular caffeine consumers
in response to administration of oral caffeine
(10 mg/kg caffeine citrate) (Figs l a and 1b). To
further evaluate the effects of tolerance to the meta-
bolic actions of caffeine, a subgroup of regular
caffeine consumers, who were able to discontinue all
caffeinated food or beverages for one month, were
rechallenged with the same oral caffeine dose; a
marked amplification of brain lactate response oc-
curred in the subjects during this period of caffeine
holiday as shown in Figure lc. In contrast to the
594 DAGER FRIEDMAN

this study, observations that grey matter demonstrated

a
NAA greater lactate response to caffeine ingestion than
white matter probably reflects differential brain energy
expenditure between compartments (21).
Following abrupt cessation of caffeine, CBF globally
increases with a similar time course as the clinical
withdrawal state (18). The magnitude and duration of
Baseline caffeine withdrawal and related CBF increases appear
to be closely associated with the prior amount of daily
l r l caffeine consumption and presumably reflect enhanced
adenosine activity. Excess regional brain lactate ele-
After caffeine
ingestion vations observed in response to caffeine rechallenge
J I I
2.5 2.0- 1.5 1.0 0.5 among regular caffeine users following a 1-month
Chemical shift (ppm) caffeine holiday are also postulated to reflect en-
hanced effects of caffeine on adenosine receptors that
have been down-regulated (or not up-regulated) in the

t
b absence of caffeine (21). As will be further discussed
U-El Caffeine-intolerant subjects ( n = 9)
W Regular caffeine users ( n = 9) below, state-dependent metabolic effects of caffeine
O'I2 have the potential to substantially impact the inter-

t
0
.-
c pretation of functional imaging findings.
e
a
a O.I0
? Caffeine
al
c ingestion Nicotine
m
c
0
m
-J
0.06 Similar to caffeine, nicotine is the drug-of-choice
among large segments of the general population (30).
In studies evaluating smoking prevalence within
-fo L7 Ib $0 do do sb $0
Time (min)
patient and control groups, approximately 30% of
subjects are smokers, a rate markedly elevated among
certain psychiatric populations, most notably schizo-
C phrenics (reviewed in (31)). The most common
0-0 Regular caffeine users ( n = 5 ) sources of nicotine are cigarettes, chewable forms of
Caffeine holiday ( n = 5)
tobacco and, more recently, nicotine patches or gum
0
.- 0.10 utilized to promote smoking cessation. Regardless of
c
!! the route of administration, nicotine is rapidly ab-
a sorbed into the blood stream reaching peak plasma
Q
za,
c
levels within a few minutes (32). The lipophyllic
m
c
structure of nicotine allows it to cross the blood-brain
0
m
1
barrier rapidly, with brain uptake and elimination
only slightly lagging behind changes in blood levels
(33). The plasma and elimination half-lives of nicotine
are approximately 45 min and 2 h, respectively,
0.04
-20 -10 0 10 20 30 40 50 60 suggesting a corresponding time course in brain
Time (min) pharmacokinetics, although, to our knowledge, this
Figure 1. Characteristic spectra acquired from a subject at has not been measured for the human brain (32-34).
baseline and 1 h after caffeine ingestion are shown in (a). The Similar to caffeine, nicotine metabolism is 3-4 times
differential brain lactate response (expressed as mean lactate/ longer in newborns, also reflecting the previously
N,acetyl aspartate (NAA) ratios) to caffeine ingestion for caffeine-
noted liver enzyme deficits in infancy (35).
intolerant individuals and regular caffeine users are demonstrated
in (b). The strikingly different metabolic response to caffeine As with caffeine, the widespread use of nicotine
among regular caffeine users following a 1-month caffeine holiday stems from its reinforcing effects, as has been borne
is shown in (c). (Adapted from (21) with permission). out by a number of investigations demonstrating
dopaminergic changes following nicotine adminis-
tration (13, 36, 37), and from subjective behavioural
effects of caffeine on globally reducing CBF, brain effects such as improvements in attention, arousal and
lactate rises in response to caffeine demonstrated reaction time (38). In humans, nicotine administration
time-dependent regional changes that correspond to causes neural activation and produces a variety of
brain regions associated with arousal or anxiety. In physiological effects involving the cardiovascular

0 The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
 BRAINI M A G I N G A N D THE
system, including increases in CBF (39). Nicotine
consumption also stimulates the down-regulation of
neuronal nicotinic receptors (NAChRs) and produces
secondary effects involving multiple neurotransmitter
systems (ie dopamine and glutamate), mechanisms
that have been extensively reviewed elsewhere (40).
The high rate of nicotine use among schizophrenic
patients may reflect the ability of nicotine to regulate
dopaminergic innervation t o limbic structures, regions
putatively involved in schizophrenia (31). Moreover,
several investigators have suggested a decreased risk
for Parkinson’s disease and other motoric disorders
with nicotine use, possibly, the result of similar
dopaminergic modulatory effects (41).
It is well recognized that cigarette smoke also
contains an abundance of other substances that affect
the rate of nicotine absorption in addition to creating
an additive burden on respiratory and cardiovascular
health (eg carbon monoxide and formaldehyde)
(42). The occurrence of impairments in endothelial-
dependent vasodilatation (EDV) among smokers is
indicative of potential systemic vascular compromise
over the long term (43, 44).
White matter hyperintensities (WMH), a common
brain structural finding visible by MRI, have been
suggested to reflect small vessel disease (45). An
example of typical MRI findings of WMH is shown in
Figure 2. Several investigators have noted a marked
increase in WMH among smokers without other
known risk factors and an increased prevalence
among those smokers diagnosed with dementia (46)
or bipolar disorder (47), suggesting that nicotine may
be a specific risk factor for such vascular lesions. A
Figure 2. Standard T,-weighted magnetic resonance images (MRl)s from two patients demonstrating the presence of white
matter hyperintensities(WMH), as indicated by the arrows. (Adapted from (45) with permission).
EFFECTS OF CAFFEINE AND NICOTINE 5 95
causal relationship between lifetime nicotine dose and
the severity of WMH findings has been proposed as
MRI-defined WMH changes may parallel worsening
EDV over time (47). However, it must be emphasized
that WMH clearly have myriad aetiology; similarly,
not all patients who smoke demonstrate white matter
pathology (47, 48). The clinical significance of WMH
also remains uncertain, as there may not be associated
clinical antecedents or deleterious clinical outcome
(48). Thus, further work integrating perfusion meas-
ures and quantitative WMH measurements is needed
to help clarify the relationship between vascular
health and lesion load. Additionally, routine character-
ization of smoking behaviour within neuroimaging
studies will aid in evaluating the general relationship
between WMH and smoking.
In contrast to the effect of caffeine on reducing
global CBF by 30% or more, cigarette smoking or iv
nicotine administration increases CBF, as demon-
strated by a xenon 133 SPECT study that showed an
approximately 16% average CBF increase (39). In an
elegant blood flow study with transcranial Doppler,
rapid (-10 s) blood flow increases within four
cerebral vessels were demonstrated during smoking, a
response that was directly related to nicotine dose and
decreased at a similar time course following smoking
cessation (49). Moreover, in the latter study, simul-
taneous measurements of radial artery blood flow
demonstrated decreased flow during smoking, illus-
trating the differential effects of nicotine on the
peripheral vasculature. Regional accentuation of
nicotine-induced CBF increases have been demon-
strated within the frontal lobes and cerebellum with
0The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
5 96 DAGEK FRIEDMAN

PET (50), although not all human CBF studies have
demonstrated consistently unidirectional regional CBF
increases in response to nicotine. For example, a PET
study that investigated the effects of nicotine on
attention following a low-dose nicotine infusion over
80 min also demonstrated regional CBF decreases in
cingulate and cerebellum (51).
Recent investigations utilizing BOLD fMRI during
nicotine infusion and statistical models derived from
increasing plasma nicotine levels (52, 5 3 ) have shown
regional brain increases in BOLD response (neuronal
activation) that parallel brain areas implicated in
nicotine craving or abuse, as well as regions of
increased NAChR receptor density (52). The number
of activated regions increased in conjunction with
increasing nicotine dose, although a dose-related
magnitude of response was not demonstrated. The
absence of a clear dose response between nicotine and
CBF may result from the cumulative dosing paradigm
employed, or from the inherent difficulty in quantify-
ing conventional BOLD fMRI signal response ( 3 ) .As
the investigators expressly noted, BOLD fMRI re-
sponses in the negative direction were not evaluated,
which might have helped to address regional bi-

b Signal intensity
(MR-units)

Time (seconds)
Figure 3. In (a) four adjacent axial echoplanar images through the visual cortex are shown for a healthy control during visual
stimulation under conditions of normocapnia (end-tidal pC0, = 40 mm Hg) and hypocapnia (end-tidal pC0, = 19 mm Hg),
demonstrating loss of functional magnetic resonance imaging (fMRI) functional contrast. In (b) the time course of signal intensity
response to hyperventilation (indicated by dark horizontal line) during repeated visual stimulation (indicated by grey vertical lines is
demonstrated). (Reproducedfrom (64) with permission.)

0The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
 BRAINIMAGING A N D THE EFFECTS OE CAFFEINE A N D NICOTINE
directional CBF responses to nicotine administration
(53).
Some consideration is necessary regarding the
interactive nature of caffeine and nicotine use. The
majority of smokers also consume substantial quan-
tities of caffeine (54), and a decrease in smoking
frequency can also be associated with increased
caffeine intake (and vice versa) (55). Recent work
evaluating cardiovascular response to combined caf-
feine and nicotine administration indicates that the
resultant blood pressure and heart rate increases from
the resting state reflect additive effects of these two
drugs (56). Conversely, blood pressure response to
physical exertion, such as standing, is reduced through
the combined actions of caffeine or nicotine adminis-
tration that are presumably mediated by the effects of
adenosine (56). This later observation emphasizes
the importance of state-dependent influences on the
actions of both nicotine and caffeine.
Influences of caffeine and nicotine on
functional imaging results
The ability to model or separate out the relative
contributions to signal response from neuronal acti-
vation and blood flow mechanisms observed in
human neuroimaging studies remains a significant
challenge. For example, recent work, in which a rat
model was used to investigate the relationship between
oxygen delivery and blood flow with a 'H MRS
technique edited for carbon ( I3C), showed significant
increases in CBF and global brain metabolic rates
following nicotine administration (approximately 41 YO
and 30% increases from baseline anaesthetized values,
respectively), similar to the effects of sensorimotor
stimulation (34% and 26% increases, respectively)
(57). Future technical developments that allow mul-
tiple perfusion or activation measures to partial out
the nonspecific effects of generalized blood flow or
metabolic changes from discrete measures of acti-
vation will greatly enhance our ability to characterize
neural response patterns. However, in the context of
not yet having fully achieved this goal, the con-
founding effects of caffeine consumption and smoking
(both acute and chronic) on CBF and metabolism, as
well as associated changes in receptor density, have
implications for interpretation of fMRI and fMRS
results.
Current MR functioning imaging methods depend
upon subtle changes in CBF (oxyhaemoglobin levels)
or energy metabolism (lactate) to provide indices of
task-specific neural activation (3, 7). Although earlier
functional imaging studies with PET provided quanti-
tative assessment of changes in regional CBF or
glucose metabolism from baseline, if assumptions of a
two-compartment model could be met, sensitivity
0 The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
597
Brain imaging technology can characterize the
brain metabolic and chemical consequences o f
caffeine and nicotine consumption in the brain:
similarly, careful appraisal o f the influences that
these commonly used drugs have on interpretation
o f functional imaging findings is critical.
constraints did nor allow for the appraisal of indi-
vidual response patterns (7). On the other hand,
BOLD fMRI techniques, which allow substantially
improved temporal resolution and more precise
anatomical coregistration, also provide sufficient
sensitivity to evaluate single-subject dynamic response
patterns of brain activation that can be monitored in
near-real time (3). However, a general limitation of
all current functional imaging techniques has been
the assumption that links between CBF or energy
metabolism and neuronal activity are invariant, such
that disease states or drug effects do not alter these
relationships (7).
In addition to the previously noted effects of
caffeine habituation on functional MRS brain lactate
response (21), respiratory status can also influence
brain lactate levels substantially, presumably through
alterations in CBF regulation, with different patterns
of functional MRS brain lactate response to controlled
hyperventilation found for specific clinical populations
(58).In the case of BOLD fMRI, additional problems
arise from basal differences in CBF as a result of
inherent difficulties in quantitating signal response or
establishing baseline conditions, although newer MR
functional imaging techniques, such as quantitative
T,-weighted mapping of functional water changes,
may reduce this problem (59). For the currently
available fMRI techniques a variety of factors, such as
gender (60), age (61), respiratory status (62) or drug
use (63), can signifdandy influence the pattern of
BOLD fMRI signal response and confound the inter-
pretation of results. For example, investigators have
directly evaluated the effects of respiratory-induced
alterations in CBF on BOLD fMRI signal response,
demonstrating abolishment of functional contrast
within the visual cortex to light activation under
conditions of hyperventilation (64). An example of
this work is shown in Figures 3a and 3b. Further
investigation has demonstrated that the fMRI signal
response is directly proportional to end-tidal CO,
between 20 and 60 mmHg, consistent with the linear
relationship for CBF and PaCO, between those ranges
(65). As caffeine and nicotine substantially affect CBF,
brain metabolism and neural activation, their use
would also be expected to influence the BOLD fMRI
signal contrast to a considerable degree, producing
598 DACER FREDMAN
differential effects in the acute state, with chronic
usage and under withdrawal conditions. These variable
systematically.
References
1. Magnotta VA, Heckel D, Andreasen NC, Cizadlo T, Corson
PW, Ehrhardt JC, et al. Measurement of brain structures
with artificial neural networks: two- and three-dimensional
application. Radiology 1999; 21 1: 781-90.
2. Csernansky JC, Joshi S, Wang L, Haller JW, Gad0 M, Miller
JP, et al. Hippocampal morphometry in schizophrenia by
high dimensional brain mapping. Proc Nut1 Acud Sci U S A
1998; 95: 11406-11.
3. Posse S, Miiller-Gartner HW, Dager SR. Functional magnetic
resonance studies of brain activation. Sem Clin Neuro-
psychiatry 1996; 1: 76-88.
4. Gadian DG. N M R and its applications to living systems.
2nd edn. New York, NY Oxford University Press; 1995: 1-
283.
5. Dager SR, Steen RG. Applications of magnetic resonance
spectroscopy to the investigation of neuropsychiatric dis-
orders. Neuropsychopharmacology 1992; 6: 249-66.
6. Fox PT, Raichle ME. Focal physiological uncoupling of
cerebral blood flow and oxidative metabolism during
somatosensory stimulation in human subjects. Proc Natl
Acad Sci U S A 1986; 83: 1 1 4 0 4 .
7. Dager SR, S w a m AC. Advances in brain metabolism
research: towards a moving picture of neural activity. Biol
Psychiatry 1996; 39: 231-3.
8. Kaufman MJ, Levin JM. Magnetic resonance findings in
substance abuse. In: Kaufman MJ, ed. Brain imaging in
substance abuse: research, clinical, and forensic applications.
Totowa, NJ: Humane Press; 2000: 155-98.
9. Fredholm BB, Battig K, Holmen J, Nehlig A, Zvartau EE.
Actions of caffeine in the brain with special reference to
factors that contribute to its widespread use. Pharmaco(
Rev 1999; 51: 83-133.
10. Gilbert RM. Caffeine consumption. In: Spiller GA, ed. The
methylzanthine beverages and foods: chemistry, consumption,
and health effects. New York, NY: Alan R Liss; 1984: 185-
213.
11. Axelrod J, Reisenthal J. The fate of caffeine in man and a
method for its estimation in biological material. f Pharmacol
Exp Ther 1953; 107: 519-23.
12. Arnaud MG. The pharmacology of caffeine. Prog Drug Res
2987; 31: 273-319.
13. O'Brien CP. Drug addiction and drug abuse. In: Hadman
JC, Limbird LE, eds. Goodman and Gilman's The pharma-
cological basis o f therapeutics. 9th edn. New York, NY:
McCraw-Hill; 1996: 557-77.
14. Rowland M, Tozer TN. Clinical pharmacokinetics. Phila-
delphia, PA: Lea and Fibiger: 1980.
15. Aranda JV, Collinge JM, Zinman R, Watters G. Maturation
of caffeine elimination in infancy. Arch Dis Child 1979; 54:
946-49.
16. Hart P, Farrell GC, Cookeley WG, Powell LW. Enhanced
drug metabolism in cigarette smokers. Brain Med J 1976; 2:
147-9.
17. American Psychiatric Association. Diagnostic and statistical
manual of mental disorders; Fourth edition Text Revision
(DSM-IV-TRTM).Washington, DC: American Psychiatric
Association: 1994.
Q The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
We thank Marie Domsalla for her help with manuscript
preparation. This work was supported, in part, by a NARSAD
18. Mathews RJ, Wilson WH. Caffeine consumption, with-
drawal and cerebral blood flow. Headache 1985; 25: 305-9.
19. Dunwiddie TV. Endogenously released adenosine in the
central nervous system. Int Rev Neurobiol 1985; 27: 63-
139.
20. Nehlig A, Daval JL, Debry G. Caffeine and the central
nervous system: mechanisms of action, biochemical, meta-
bolic and psychostimulant effects. Brain Res Rev 1992; 17:
139-70.
21. Dager SR, Layton ME, Strauss W, Richards TL, Heide A,
Friedman SD, et al. Human brain metabolic response to
caffeine and the effects of tolerance. A m J Psychiatry 1999;
156: 229-37.
22. LaCroix AZ, Mead LA, Liang K-Y, Thomas CB, Pearson
TA. Coffee consumption and the incidence of coronary
heart disease. N Engl f Med 1986; 315: 977-82.
23. Crobbee DE, Rimm EB, Giovannucci E, Colditz G, Stampfer
M, Willett W. Coffee, caffeine, and cardiovascular disease in
men. N EnglJ Med 1990; 323: 1026-32.
24. Sutherland GR, Peeling J, Lesiuk HJ, Brownstone RM,
Rydzy M, Saunders JK, et al. The effects of caffeine on
ischemic neuronal injury as determined by magnetic reson-
ance imaging and histopathology. Neuroscterrce 1991; 42:
171-82.
25. Ross GW, Abbott RD, Petrovitch H, Morens DM,
Grandinetti A, Tung KH, et al. Association of coffee and
caffeine intake with the risk of Parkinson's disease. J A M A
2000; 283: 2674-9.
26. Gibbs FA, Gibbs EL, Lennon WG. Cerebral blood flow in
man as influenced by adrenaline, caffeine, amyl nitrite and
histamine. A m Heart J 1935; 10: 916-24.
27. Mathew RJ, Wilson WH. Caffeine-induced changes in
cerebral circulation. Stroke 1985; 16: 814-17.
28. Cameron OG, Modell JG, Hariharan M. Caffeine and
human cerebral blood flow: a positron emission tomography
study. Life Sci 1990; 47: 1141-6.
29. Posse S, Dager SR, Richards TL, Yuan C, Artru AA, Ogg R,
et al. In vivo measurement of regional brain metabolic
response to hyperventilation using functional proton echo-
planar spectroscopic imaging (PEPSI). Magn Reson Med
1997; 37: 858-65.
30. Tate JC, Pomerleau CS, Pomerleau OF. Pharmacological and
nonpharmacological smoking motives: a profile. J Subst
Abuse 1993; 5 : 391-400.
31. Dalack GW, Healy DJ, Meador-Woodruff JH. Nicotine
dependence in schizophrenia: clinical phenomena and labora-
tory findings. A m J Psychiatry 1998; 155: 1490-501.
32. Rosenberg J, Benowitz NL, Jacob P, Wilson KM. Disposition
kinetics and effects of intravenous nicotine. Clin Pharmacol
Ther 1980; 28: 517-22.
33. Porchet HC, Benowitz NL, Scheiner LB, Copeland JR.
Apparent tolerance to the acute effect of nicotine results in
part from distribution kinetics. f Clin fnvest 1987; 80:
1466-71.
34. Benowitz NL, Jacob P 111, Jones RT, Rosenberg J. Inter-
individual variability in the metabolism and cardiovascular
effects of nicotine in man. f Pharmacol Exp Tber 1982; 221:
 BRAINIMAGING A N D THE EFFECTS
368-71.
35. Dempsey D, Jacob P 111, Benowitz NL. Nicotine metabolism
and elimination in newborns. Clin Pharmacol Ther 2000;
67: 458-65.
36. Dewey SL, Brodie JD, Gerasimov M, Horan B, Gardner EL,
Ashby CR Jr. A pharmacologic strategy for the treatment of
nicotine addiction. Synapse 1999; 31: 76-86.
37. Salokangas RKR, Vilkman H, llonen T, Taiminen T,
Bergman J, Haaparanta M, et al. High levels of dopamine
activity in the basal ganglia of cigarette smokers. A m J
Psychiatry 2000; 157: 632-34.
38. Jones GMM, Sahakiam BJ, Levy R, Warburton DM, Gray
JA. Effects of subcutaneous nicotine on attention, infor-
mation processing and short-term memory in Alzheimer's
disease. Psychopharmacology 1992; 108: 485-94.
39. Skinhoj E, Olesen J, Paulson OB. Influence of smoking and
nicotine on cerebral blood flow and metabolic rate of
oxygen in man. ] Appl Physiol 1973: 35, 820-2.
40. Gotti C, Fornasari D, Clementi F. Human neuronal nicotinic
receptors. Prog Neurobiol 1997; 53: 199-237.
41. Decina P, Carraci G, Sandik R, Berman W, Mukherjee S,
Scapicchio P. Cigarette smoking and neuroleptic-induced
Parkinsonism. Biol Psychiatry 1990; 28: 502-8.
42. Rylander R. Relative role of aerosol and volatile constituents
of cigarette smoker as agents toxic to the respiratory tract.
Natl Cancer Inst Monogr 1968; 28: 221-9.
43. Sarabi M, Lind L. Short-term effects of smoking and nicotine
chewing gum on endothelium-dependent vasodilation in
young healthy habitual smokers. J Cardiovasc Pharmacol
2000; 35: 451-6.
44. Neunteufl T, Priglinger U, Heher S, Zehetgruber M, Soregi
G , Lehr, et al. Effects of vitamin E on chronic and acute
endothelial dysfunction in smokers. ] A m Coll Cardiol
2000; 35: 277-83.
45. Friedman SD, Stidley CA, Brooks WM, Hart BL, Sibbitt
WLJr. Brain injury and neurometabolic abnormalities in
systemic lupus erythematosus. Radiology 1998; 209: 79-84.
46. Roman GC. Senile dementia of the Binswanger type: a
vascular form of dementia in the elderly. ] A M A 1987; 257:
1782-8.
47. Dupont RM. Structural brain changes in mood disorder:
clinical correlates and significance. Sem Clin Neuropsychia-
try 1996; 1: 20-31.
48. Brooks WM, Wesley MH, Koditawakku P, Carry PH,
Rosenberg GA. 'H-MRS differentiates white matter hyper-
intensities in subcortical arteriosclerotic encephalopathy
from those in normal elderly. Stroke 1997; 28: 1940-3.
49. Morioka C, Kondo H, Akashi K, Matsumura K, Ochi N,
Makinaga G , et al. The continuous and simultaneous blood
flow velocity measurement of four cerebral vessels and
peripheral vessel during cigarette smoking. Psychopharmaco-
logy 1997; 131: 220-9.
50. Nagata K, Shinohara T, Kanno I, Hatazawa J, Domino EF.
Effects of tobacco and cigarette smoking on cerebral blood
flow in normal adults. In: Domino EF, ed. Bruin imaging of
nicotine and tobacco smoking. MC: NPP Books; 1995: 95-
108.
51. Ghatan PH, Ingvar M, Eriksson L, Stone-Elander S,
0 The Finnish Medical Society Duodecim, Ann Med 2000; 32: 592-599
O F CAFFEINE AND N I C O T I N E 5 99
Serrander M, Ekberg K, et al. Cerebral effects of nicotine
during cognition in smokers and non-smokers. Psycho-
pharmacology 1998; 126: 179-89.
52. Stein EA, Pankiewicz J, Harsch HH, Cho H-K, Fuller SA,
Hoffmann RG, et al. Nicotine-induced limbic cortical
activation in the human brain: a functional MRI study. A m J
Psychiatry 1998; 155: 1009-15.
53. Bloom AS, Hoffmann RG, Fuller SA, Pankiewicz J, Harsch
HH, Stein EA. Determination of drug-induced changes in
functional MRI using a pharmokinetic model. Hum Bruin
Mapping 1999; 8 : 235-44.
54. Schubert DSP. Arousal seeking as a central factor in tobacco
smoking among college students. Znt J Soc Psychiatry 1965;
11: 221-5.
55. Kozlowski LT. Effects of caffeine consumption on nicotine
consumption. Psychopharmacology 1976; 47: 165-8.
56. Smits P, Temme L, Thien T. The cardiovascular interaction
between caffeine and nicotine in humans. Clin Pharmacol
Ther 1993; 54: 194-204.
57. Hyder F, Kennan RP, Kida I, Mason GF, Behar KL,
Rothman D. Dependence of oxygen delivery on blood flow
in rat brain: a 7 tesla nuclear magnetic resonance study. J
Cereb Blood Flow Metab 2000; 20: 485-98.
58. Dager SR, Strauss WL, Marro KI, Richards TL, Metzger
GD, Artru AA. Proton magnetic resonance spectroscopic
investigation of hyperventilation in subjects with panic
disorder and comparison subjects. A m ] Psychiatry 1995;
152: 666-72.
59. Posse S, Wiese S, Gembris D, Mathiak K, Kessler C, Grosse-
Ruyken ML, et al. Enhancement of BOLD-contrast sensitiv-
ity by single-shot multi-echo functional MR imaging. Magn
Reson Med 1999; 42: 87-97.
60. Levin JM, Ross MH, Mendelson JH, Mello NK, Cohen BM,
Renshaw PF. Sex differences in blood-oxygenation-level-
dependent functional MRI with primary visual stimulation.
A m J Psychiatry 1998; 155: 434-6.
61. Ross MH, Yurgelun-Todd DA, Renshaw PF, Maas LC,
Mendelson JH, Mello NK, et al. Age-related reduction in
functional MRI response to photic stimulation. Neurology
1997; 48: 173-6.
62. Posse S, Olthoff U, Jancke L, Weckesser M, Miiller-Gartner
HW, Dager SR. Regional dynamic signal changes during
initiation of controlled hyperventilation assessed by BOLD-
dependent fMRI. A m ] Neuroradiol 1997; 18: 1763-70.
63. Kaufman MJ, Levin JM, Maas LC, Rose SL, Lukas SE,
Mendelson JH, et al. Cocaine decreases relative cerebral
blood volume in humans: a dynamic susceptibility contrast
magnetic resonance imaging study. Psychopharmacology
1998; 138: 76-81.
64. Weckesser M, Posse S, Olthoff U, Kemna L, Dager SR,
Miiller-Gartner H-W. Functional imaging of the visual
cortex with BOLD-contrast MRI: hyperventilation decreases
signal response. Magn Reson Med 1999; 41: 213-6.
65. Posse S, Elghahwagi B, Kemna LJ. Dependence of the fMR1
time course of the hemodynamic response function in visual
cortex on global cerebral blood flow. Proc Intl Soc Magn
Reson Med 2000: 984.