Psychosomatic Medicine 66:63-71 (2004)
© 2004 American Psychosomatic Society
Critical Review of Dietary Caffeine and Blood Pressure: A Relationship That Should Be Taken More Seriously
Jack E. James, PhD
National University of Ireland, Galway, Ireland.
Address correspondence and reprint requests to Prof. Jack E. James, Department of Psychology, National University of Ireland, Galway, Ireland. E-mail: j.james{at}nuigalway.ie
 |
ABSTRACT
|
|---|
OBJECTIVE: This review aims to determine whether, and to what extent, dietary caffeine may be a risk to cardiovascular health.
METHODS: A critical review of the relevant experimental and epidemiologic literature was conducted, with particular reference to studies of caffeine and blood pressure (BP).
RESULTS: There is extensive evidence that caffeine at dietary doses increases BP. However, concern that the drug may contribute to cardiovascular disease appears to have been dampened by (1) the belief that habitual use leads to the development of tolerance, and (2) confusion regarding relevant epidemiologic findings. When considered comprehensively, findings from experimental and epidemiologic studies converge to show that BP remains reactive to the pressor effects of caffeine in the diet. Overall, the impact of dietary caffeine on population BP levels is likely to be modest, probably in the region of 4/2 mm Hg. At these levels, however, population studies of BP indicate that caffeine use could account for premature deaths in the region of 14% for coronary heart disease and 20% for stroke.
CONCLUSIONS: Current evidence supports the conclusion that the BP-elevating effects of dietary caffeine may be contributing appreciably to population levels of cardiovascular mortality and morbidity. Accordingly, strategies for encouraging reduced dietary levels of caffeine deserve serious consideration.
Key Words: caffeine, • blood pressure, • cardiovascular disease, • adenosine, • hypertension, • epidemiology.
Abbreviations: BP = blood pressure.
|
INTRODUCTION
|
|---|
This review examines experimental and epidemiologic studies of dietary caffeine and blood pressure (BP). Although findings from the many relevant studies have sometimes been said to be inconsistent, close examination suggests greater consistency than might generally have been assumed. Accordingly, by integrating experimental and epidemiologic findings concerning caffeine and BP, this review aims to determine whether, and to what extent, dietary caffeine may be a risk to cardiovascular health.
|
WHY CAFFEINE?
|
|---|
Several key facts regarding caffeine and its use give credence to the possibility that dietary use of the drug could contribute to population levels of cardiovascular disease. First, population exposure to caffeine is high. More than 80% of the world’s population ingests caffeine daily, making it the most widely consumed drug in history (1). Although caffeine is frequently ingested in food (eg, chocolate) and not infrequently in medication (eg, some compound analgesics), global consumption is overwhelmingly attributable to three beverages: coffee, tea, and caffeine soft drinks (including "energy" drinks). In addition, for the majority of consumers, exposure to caffeine is effectively life-long. The first exposure for most people occurs before birth. Caffeine crosses the placenta (2,3), and because most women consume caffeine while pregnant, the majority of newborns show pharmacologically active levels of plasma caffeine (4). Exposure typically continues during childhood, with patterns of use tending to consolidate during adolescence and early adulthood (5). Thereafter, usage tends to stabilize, generally undergoing little change for the remainder of life (1). The unparalleled prevalence of caffeine use introduces multipliers in relation to the possible impact of the drug on population health. That is, effects experienced at the level of the individual, even if small, could have a substantial cumulative impact when assessed across entire populations.
The suggestion that dietary caffeine could have a role in disease processes is biologically plausible. After oral ingestion, caffeine is rapidly absorbed from the gastrointestinal tract into the bloodstream (6). Approximately 90% of the caffeine contained in a cup of coffee is cleared from the stomach within 20 minutes (7), and peak plasma concentration is typically reached within approximately 40 to 60 minutes (8). Once absorbed, caffeine exerts a variety of pharmacological actions at diverse sites, both centrally and peripherally. The main mechanism of action is antagonism of the potent endogenous neuromodulator, adenosine (9). Because caffeine and adenosine have similar molecular structures, caffeine has the potential to occupy adenosine receptor sites (especially A1 and A2a), thereby blocking the regulatory effects of adenosine. By antagonizing adenosine, which has generalized inhibitory functions, the effect of caffeine is broadly stimulatory (10,11). Thus, there are multiple prima facie reasons for considering caffeine a possible health risk factor. Specifically, there are concerns regarding the demographics of caffeine use, which is widespread and essentially life-long. In addition, the extent of possible adverse caffeine effects is considerable because of caffeine’s antagonism of endogenous adenosine (biological plausibility).
|
WHY BLOOD PRESSURE?
|
|---|
When considering the demographics of caffeine use in the context of population health, it is well to note that additional multiplying factors come into play when the outcome variables under consideration are themselves widespread. Cardiovascular disease is the major cause of death and disease in the developed countries of the world (12–14) and is likely to assume a similar status in much of the developing world (especially Asia) in the coming years (15–17). Elevated BP is of acknowledged importance as a cause of cardiovascular disease (12–19), and BP is comparatively sensitive to caffeine because of the important regulatory actions of adenosine in the cardiovascular and central nervous systems (20,21). Indeed, the principal means by which caffeine influences BP is known to be vasoconstriction resulting from adenosine antagonism (23–26). This is not to ignore the fact that various other cardiovascular indices have been considered in relation to the possible implications of dietary caffeine and cardiovascular function.
For example, although elevated heart rate is a risk factor for cardiovascular disease (27), extensive research shows that dietary levels of caffeine have little effect on heart rate (28,29). Similarly, caffeine has been found to have little effect on a variety of electrocardiographic variables (30). Although reports of the arrhythmogenic properties of caffeine in people who have cardiac disease (31,32) or in those free of disease (33) have not always been consistent, the consensus is that the drug poses little significant arrhythmogenic risk. Moreover, the role of cardiac arrhythmia in the development of cardiovascular disease remains uncertain (33,34). Finally, although various coffee preparations have been found to elevate the cardiovascular risk factors of serum cholesterol and plasma homocysteine significantly (35–39), neither caffeine specifically nor caffeine beverages other than coffee have these effects.
Thus, if caffeine does have implications for cardiovascular disease, its affect on BP would appear to be a key consideration. Regarding specific disease outcomes, no special distinction is made in this review between categories of cardiovascular disease. However, particular mention is made of coronary heart disease and cerebrovascular accident (stroke), because these are the cardiovascular diseases of highest incidence (19).
|
EXPERIMENTAL STUDIES OF THE EFFECTS OF ACUTE CAFFEINE EXPOSURE ON BLOOD PRESSURE
|
|---|
Acute BP elevations in the range of 5 to 15 mm Hg systolic and 5 to 10 mm Hg diastolic are typical after experimental administration of caffeine in amounts comparable to those consumed in everyday life (28,29,40–61). In general, effects may last several hours and are evident in both men and women of all ages (52,56–58). Moreover, caffeine-induced increases in BP have frequently been found to be additive to the effects of other pressor agents, including cigarette smoking (44,61) and psychological stress (28,29,44–52,59). In addition to interest in the general effects of caffeine on BP, there is also interest in the factors, if any, that encourage individual susceptibility to the BP-elevating effects of the drug. One such line of investigation suggests that increased sensitivity may be experienced by people with existing high BP, including people with borderline hypertension (48–50) and hypertension (26,42,43,48,55,58). Similarly, there is evidence that caffeine may sometimes act synergistically on BP during times of psychological stress, resulting in exaggerated increases in BP in people experiencing stress (47,62,63).
Although such findings indicate that caffeine could contribute to cardiovascular disease, concern has been dampened by doubts regarding the generality of experimental studies of acute effects vis-à-vis chronic reactions arising from dietary use. With few exceptions, until recently, studies have used a double-blind drug challenge protocol involving a single [acute] dose of caffeine. In contrast, dietary caffeine is characterized by ingestion of three to five caffeine beverages consumed at intervals during the day. Hence, it is doubtful whether the classic drug challenge paradigm is suitable for elucidating the chronic effects of repeated exposure characteristic of dietary use. Such doubt has been strengthened by 1) the belief that habitual use of caffeine leads to the development of hemodynamic tolerance, and 2) confusion surrounding the epidemiology of caffeine and BP.
|
CAFFEINE TOLERANCE AND BLOOD PRESSURE
|
|---|
There is only meager direct empirical support for the belief that habitual consumers are hemodynamically unresponsive (ie, develop tolerance) to caffeine. Moreover, the belief that dietary use leads to complete tolerance is inconsistent with a large body of direct and indirect evidence. Regarding the latter, it is noteworthy that participants in studies of acute hemodynamic reactivity to caffeine are typically drawn from the population of habitual consumers. As such, participants generally have a pattern of consumption characterized by intermittent ingestion of caffeine during the day, with fewer portions consumed later in the day, followed by overnight abstinence. In experimental studies of acute hemodynamic reactivity to caffeine, it has been usual to require participants to abstain from caffeine overnight before laboratory testing, and (as mentioned above) BP frequently increases in response to caffeine given under such circumstances. Thus, the following question arises: given that ordinary consumers after overnight abstinence show increased BP in response to caffeine ingested in the experimental laboratory, why would consumers engaged in everyday caffeine use show no such response? The answer to this question is to be found in empirical evidence summarized below, which shows, contrary to widespread belief, that moderate amounts of caffeine in the diet do not lead to immunity to the pressor effects of the drug.
Claims of hemodynamic tolerance to dietary caffeine stem largely from a single source, a study in 1981 by Robertson et al. (64). Although the authors themselves concluded that they had observed "partial" development of tolerance, the study has been widely misquoted as having demonstrated "complete" tolerance. Contrary to claims for the development of complete tolerance, participants in studies demonstrating caffeine-induced increases in BP represent the entire spectrum of consumers, ranging from essential nonusers to high consumers (28,29,40–63). With regard to this diversity, the tolerance hypothesis predicts greater reactivity in low compared with high habitual caffeine consumers. However, no such systematic difference in reactivity is evident in the findings (1,26). Moreover, no difference in reactivity to caffeine was found between high–caffeine-consuming and low–caffeine-consuming participants in studies that directly compared consumers with different dietary caffeine habits (28,46). Indeed, whereas the tolerance hypothesis predicts hemodynamic reactivity to be a function of caffeine-consumer status (consumers of larger amounts should be less reactive than consumers of smaller amounts), the pressor effect of caffeine has been shown to be proportional to systemic caffeine levels at the time the drug is ingested, irrespective of current history of use (57,65,66).
Notably, on a given day, the magnitude of BP elevations induced by each successive portion of caffeine beverage is approximately inversely proportional to the number of portions already consumed that day (57,65). That is, for the individual consumer, second and later cups of a caffeine beverage produce less hemodynamic reactivity than the initial cup of the day (57,65,66). However, near-complete sensitivity to the initial cup is restored the following day after overnight abstinence (67–74). Most importantly, the proportional diminution in pressor effect with progressive increases in caffeine plasma levels is consistent with current understanding regarding the main mechanism of action. During the process of competitive blockade, it appears that available adenosine receptors generally become saturated by caffeine at dietary levels of use (9). This same process has implications for clinical practice. Specifically, reflecting unease about the possible pressor effects of caffeine, it has become common clinical practice to recommend "moderation" and avoidance of "excessive" use of caffeine. In reality, however, the evidence shows that almost all of the pressor action of caffeine is achieved at ordinary (ie, moderate) dietary levels, equivalent to three or four cups of coffee, with little or no additional pressor effect occurring when larger daily amounts are ingested. The main consequence of higher levels of consumption is that pressor effects are likely to be prolonged rather than substantially more pronounced.
|
EXPERIMENTAL STUDIES OF THE EFFECTS OF CHRONIC CAFFEINE EXPOSURE ON BLOOD PRESSURE
|
|---|
Before the last decade, there had been little direct examination of the chronic hemodynamic effects of dietary caffeine. Among the first experimental studies (controlled clinical trials) to undertake such an examination, modest sustained decreases in BP were reported when caffeine beverages were either removed from the diet or replaced by decaffeinated alternatives (68,74–76). Similar results were reported in a number of subsequent studies in which ambulatory monitoring was used to measure BP level for extended periods (67–74). Confirmation of modest persistent elevations in systolic and diastolic pressures has also been provided by a recent meta-analysis of experimental studies of the effects of coffee drinking on BP (77).
Again, these findings are consistent with the consensus that antagonism of endogenous adenosine is the main mechanism of action responsible for caffeine-induced elevations in BP (9,22–25). Plasma caffeine concentration is typically highest in the late afternoon and lowest on awakening in the morning (78,79), reflecting the fact that overnight abstinence (typically of 10–12 hours), a characteristic of dietary intake for the majority of consumers, leads to near-complete elimination of systemic caffeine by early morning (78–81). Given that the pressor effect of caffeine is proportional to systemic caffeine levels at the time the drug is ingested (57,65,66), when systemic levels are depleted, as is typically the case before the first caffeine beverage of the day, reactivity to caffeine is substantially reinstated.
More specifically, acute upregulation of adenosine (82,83) is believed to explain the progressive diminution in hemodynamic reactivity that accompanies second and later cups of caffeine beverage consumed on a given day (transient partial tolerance). In turn, recovery of adenosine receptor numbers, increased sensitivity during overnight abstinence, or both, explain the substantial restoration of hemodynamic caffeine sensitivity (reversal of tolerance) evident when caffeine is next ingested, typically shortly after awakening (78,81). The cumulative effect of typical dietary levels of caffeine is for BP to be elevated modestly throughout most of the waking hours (67–74). Accordingly, it follows that dietary caffeine may pose a threat to cardiovascular health, and if so, the impact should be evident from epidemiologic analyses. The relevant epidemiology includes population studies of caffeine and cardiovascular disease in general, and studies of caffeine and BP in particular.
|
EPIDEMIOLOGY OF CAFFEINE AND CARDIOVASCULAR DISEASE
|
|---|
More than 100 large epidemiologic studies in more than a dozen countries have examined the relationships that dietary caffeine has with cardiovascular function, morbidity, and mortality (1,84). One feature of this large literature is the many inconsistencies in the reported findings, with a proportion of studies reporting null effects (ie, nonsignificant associations). This fact has inclined some commentators toward dismissing concerns about caffeine. However, the overall level of inconsistency is not as great as is sometimes claimed. When all relevant epidemiologic studies are considered, both quantitative meta-analyses (85,86) and critical reviews (74,84) implicate dietary caffeine as a cardiovascular risk factor. The fact that greater consistency has not emerged in relation to the epidemiology of dietary caffeine is likely a reflection of the many methodological shortcomings that have characterized studies published to date.
A major shortcoming of many studies is poor measurement of the key exposure variable, namely, caffeine consumption. Although this shortcoming has long been the subject of criticism (87), epidemiologic studies conducted over the past 3 decades have shown little improvement or innovation in trying to overcome the problem. Although dietary caffeine levels can be measured reliably using suitably structured and detailed self-report protocols (87,88), most studies report no reliability data and little other detail (eg, "coffee intake was recorded"). Because measurement error arising from the use of insensitive survey methods is likely to have been largely undifferentiated (ie, random), the probable effect in many instances will have been to underestimate the true association between caffeine and cardiovascular disease. Moreover, although at least half of the relevant epidemiologic studies conducted to date collected blood samples (mostly for the purpose of measuring serum lipid levels), none took the obvious next step of measuring systemic levels of caffeine. As such, advantage has not been taken of the fact that good estimates of dietary caffeine levels can be obtained by conducting high performance liquid chromatographic (HPLC) analysis for caffeine, paraxanthine (the major metabolite in humans), or both, of a single blood or saliva sample taken in the late afternoon (79,88). Thus, whereas overall epidemiologic findings suggest that dietary caffeine has a modest detrimental effect on cardiovascular health, the actual association may be stronger considering the likely presence of many spurious null results.
Confounding in the Epidemiologic Research
A frequent erroneous observation about the epidemiology of caffeine and cardiovascular disease is that much of the research has ignored the influence of confounders, especially cigarette smoking. In reality, for the past 3 decades, epidemiologic studies of caffeine have routinely used standard methodological and statistical procedures to control for the effects of cigarette smoking on relevant outcome variables. At the same time, it is important to acknowledge the ever-present risk of possible confounding in relation to a host of other known, or as yet, unidentified lifestyle variables. However, such risk may be comparatively low in relation to the epidemiology of caffeine, because the range of potential confounders identified and controlled for has been very extensive, including age, gender, cigarette smoking, alcohol consumption, body mass index, various dietary factors, serum cholesterol, BP, medical history, use of oral contraceptives, family history of heart disease, physical activity, personality, region of residence, education level, and religion (1,84). Indeed, there has probably been a tendency toward overadjustment for confounders in epidemiologic studies of caffeine (89,90). In particular, findings have frequently been adjusted for BP and cholesterol level, which may be caffeine-related and coffee-related causal pathways in their own right. As with measurement error, the impact of overadjustment for confounder effects would be to underestimate the actual strength of the association between caffeine consumption and cardiovascular disease, thereby creating a further tendency toward spurious null epidemiologic findings.
|
EPIDEMIOLOGY OF CAFFEINE AND BLOOD PRESSURE
|
|---|
Although the general methodological shortcomings evident in studies of the epidemiology of caffeine and cardiovascular disease are also a feature of the epidemiology of caffeine and BP, the latter field presents particular challenges. Of 18 population studies identified to date and summarized in Table 1, five reported no association between dietary caffeine and BP (91–95), six reported a significant positive association for systolic or diastolic pressure or both (96–101), and seven reported an inverse association for either systolic or diastolic pressure (102–108). Although it is possible that this pattern of findings represents the actual diversity of responses to caffeine among the populations studied, such a possibility appears unlikely. Most of the populations were broadly similar, both culturally and economically, yet the findings could hardly be more disparate. Studies of populations in the United States, for example, have produced all three possible outcomes; that is, the association between dietary caffeine and BP has been variously observed to be nonexistent, positive, and inverse. Moreover, these disparate findings contradict the largely consistent pattern of pressor effects arising from experimental studies of the effects of chronic caffeine exposure on BP (63,67–77).
Indeed, closer examination gives rise to the suspicion that the apparently contradictory epidemiologic findings for caffeine and BP may be the result of methodological artifact. During the course of a given day, the influence of dietary caffeine on BP may vary from zero, when plasma caffeine levels are depleted, to several millimeters of mercury for much of the day when plasma caffeine levels have been replenished. In addition to this typical diurnal pattern, ambulatory monitoring has shown that an extension of overnight abstinence into the next day leads to modest BP decreases for much of the abstinent day (63,74,109). That is, a transient hypotensive response forms part of the well-characterized array of withdrawal effects that accompany caffeine abstinence in habitual consumers (110–112). This correlation between plasma caffeine concentration and BP level helps to explain the apparently contradictory diversity of epidemiologic findings. In short, the cross-sectional association between caffeine and BP depends on the population distribution of caffeine consumers who happen to be caffeine-sated vs. caffeine-withdrawn at the time of examination.
In epidemiologic studies of dietary caffeine and BP, it is typical for participants to be asked to report the amount of caffeine they generally consume, after which a cross-sectional correlation is computed for dietary caffeine and BP measured without attention to when caffeine was last consumed. However, as explained, the relationship between caffeine and BP for the individual consumer depends crucially on the time lapse between recent ingestion of the drug and the moment when BP is measured. That is, recent caffeine consumption will generally have a pressor effect (positive association), whereas brief caffeine abstinence (10–12 hours) may have no effect, and longer periods of abstinence (12–24 hours) may decrease BP modestly (inverse association). It is only after abstinence of more than a day that BP returns to levels unaffected by previous caffeine use. As Table 1 shows, all three patterns of effects (null, positive, and inverse) have been reported with approximate equal frequency in the published epidemiologic studies. A common feature of studies reporting an inverse association between dietary caffeine and BP is that participants were asked to fast before examination. As such, there would have been a tendency for consumers to show BP readings lower than normal for themselves and potentially lower also than their nonconsuming counterparts, thereby resulting in a spurious inverse association between reported dietary caffeine and BP.
Even ignoring studies that required participants to fast longer than overnight, there is substantial risk of confounding because of failure to take account of the time lapse between most recent caffeine ingestion and BP measurement. For example, in a recent report of the prospective Johns Hopkins Precursors Study (101), coffee intake was assessed at 33 years of follow-up in a cohort of 1017 former medical students. This project is distinctive in that participants, being trained medical practitioners, were asked to self-report BP, which they had been found to be capable of doing at an acceptable level of reliability (113). However, critically for present considerations, no advice was given to participants regarding the timing of BP measurement relative to caffeine ingestion. With that aspect of the project left to chance, it is likely that participants varied in relation to level of caffeine satiation vs. deprivation at the time BP measurements were taken. Consequently, the modest association observed between coffee drinking and increased BP in the Precursors Study (101) probably represents an underestimate of the true association. The same general line of reasoning can be used to argue for a likely underestimation of the association between dietary caffeine and BP in all but one of the relevant epidemiologic studies.
The one exception concerns the Sydney Hospital Screening Study, which involved 5147 male and female volunteers age 20 to 70 years (98). After adjustment for age, body mass, cigarette smoking, alcohol consumption, serum cholesterol, and family history of hypertension, no association was initially found between caffeine consumption and BP. In a second analysis, however, participants who had consumed caffeine during the 3 hours before examination were found to have significantly elevated BPs (4/2 mm Hg) compared with participants consuming no caffeine for the same period. Moreover, because the increases in BP associated with recent ingestion of caffeine were independent of average daily intake (a measure of habitual use), the results also confirm experimental findings that habitual caffeine consumption does not lead to complete tolerance to the pressor action of the drug (63,67–77).
|
DIETARY CAFFEINE AND POPULATION BLOOD PRESSURE LEVELS
|
|---|
The balance of evidence examined here indicates that dietary caffeine contributes to statistically significant and sustained increases in BP. At the same time, it must be emphasized that the effects are modest. The average population effect during the waking hours may be estimated conservatively to be approximately 2 to 4 mm Hg (68,74,77), and the question arises whether effects of this magnitude are likely to have any appreciable independent effect on population cardiovascular mortality and morbidity. The clearest insight into this question is provided by population statistics of the relationship between BP level and rates of cardiovascular disease (12–19). In the present context, the key point is that the relationship between BP and cardiovascular disease is continuous and essentially linear. As such, any increase in population BP, including that attributable to dietary caffeine, is likely to contribute to an increase in cardiovascular disease (12–19).
Conversely, population-wide reductions in dietary caffeine would be expected to lead to overall reductions in BP, which, although small, should be accompanied by reductions in the population burden of cardiovascular death and disease. It has been estimated that a downward shift of 2 to 3 mm Hg in the population distribution of BP would produce life-saving benefits equal to the cumulative benefits achieved by antihypertensive treatment (16,17). Similarly, extensive study has shown that a 2-mm Hg reduction in systolic BP would result in lower mortality of approximately 7% for coronary heart disease and 10% for stroke (19). Thus, assuming dietary caffeine is responsible for increasing population BP by at least 4/2 mm Hg, epidemiologic BP findings (13,16,19) suggest that population-wide cessation of caffeine use could lead to decreases in premature death in the region of 14% for coronary heart disease and 20% for stroke (74).
|
CAFFEINE PHYSICAL DEPENDENCE, REINFORCEMENT, AND REDUCED INTAKE
|
|---|
Although widespread cessation of caffeine use may be unlikely to occur in the immediate future, evidence reviewed here suggests potential benefits to consumers from reduced intake. To date, however, there has been relatively little systematic study of the efficacy of procedures for reducing caffeine intake among people and groups and virtually no such study of population intervention. The main immediate barrier to reduced intake is that prolonged use of caffeine fosters the development of physical dependence, characterized by a syndrome of behavioral, physiological, and subjective discomfort provoked by abrupt withdrawal of the drug. In humans, headache, sleepiness, and lethargy are the most frequent symptoms of caffeine withdrawal (114–116), and cessation of as little as 100 mg (one cup of coffee) per day may produce symptoms that begin within approximately 12 to 16 hours since the drug was last ingested. Symptoms peak at approximately 24 to 48 hours and abate within 1 week (117–119). In addition, studies using choice or self-administration procedures or both have confirmed the reinforcing properties of caffeine in consumers during the withdrawal period (120–124).
An early controlled study to reduce caffeine intake by Foxx and Rubinoff (125) used a program of behavioral intervention based on nicotine or cigarette fading methods that had been developed for use with smokers. Promising results were reported, and these were broadly confirmed in subsequent larger studies involving follow-up of 3 months and more (126,127). Overall, the available evidence suggests that people can achieve success when attempting to reduce their intake, and that adverse withdrawal effects can be avoided by the gradual removal of caffeine from the diet. As such, the findings are consistent with studies of caffeine physical dependence and reinforcement, which show that the reinforcing properties of caffeine are markedly potentiated by abrupt abstinence (115,117,121). Under such circumstances, caffeine functions as a rather strong negative reinforcer in that intermittent ingestion of the drug enables the consumer to avoid aversive withdrawal effects. However, when withdrawal effects are absent, studies show that caffeine functions as a weak and unreliable positive reinforcer, and one that is considerably less robust than many other drugs of habitual use, including opioids, alcohol, and nicotine (128,129).
Accordingly, there are theoretical and empirical grounds for believing that caffeine fading (ie, gradual removal of the drug from the diet) may provide a useful and successful means for reducing dietary caffeine. At the same time, caffeine is present in an extensive array of products, including prescription, nonprescription and herbal drugs, foods, and beverages (130). Consequently, consumer education and improved labeling of products are likely to be important elements of any concerted effort to encourage greater avoidance of dietary caffeine. Product pricing may also be a factor, because the consumer is likely to incur greater expense when choosing decaffeinated alternatives of such products as coffee and tea.
|
CONCLUSIONS
|
|---|
Despite extensive evidence of a direct causal relationship between dietary caffeine and elevated BP, caffeine consumption continues to receive relatively little systematic clinical and public health attention. There are probably many reasons for this. Most of the professionals engaged in assessing and advising on health risks are themselves caffeine consumers, which may discourage full acknowledgment of the implications of the empirical findings. Epidemiologists, in particular, have tended to underestimate the effects of caffeine on BP, and have thereby failed to appreciate the importance of dietary caffeine as a potential cardiovascular risk factor. In part, this failure reflects the mistaken belief that habitual caffeine use leads to the development of tolerance to the cardiovascular effects of the drug. Contrary to this belief, there is strong experimental evidence that BP remains reactive to caffeine in the diet. In addition, although marred by persistent methodological shortcomings, the overall epidemiologic evidence implicates dietary caffeine as a pressor agent. Finally, extensive study of BP in populations indicates that the pressor effects of dietary caffeine are of sufficient magnitude to contribute appreciably to population levels of cardiovascular mortality and morbidity.
|
ACKNOWLEDGMENTS
|
|---|
This work was supported by the European Commission Fifth Framework Program, Grant No. QLK1-CT-2000-00069.
Received for publication July 18, 2003.
Revision received September 30, 2003.
|
REFERENCES
|
|---|
- James JE. Understanding Caffeine: A Biobehavioral Analysis. Thousand Oaks, CA: Sage Publications; 1997.
- Van’t Hoff W. Caffeine in pregnancy. Lancet 1982; 2: 1020.
- Brazier JL, Salle B. Conversion of theophylline to caffeine by the human fetus. Semin Perinatol 1981; 5: 315–20.[Medline]
- Dumas M, Gouyon JB, Tenenbaum D, Michiels Y, Escousse A, Alison M. Systematic determination of caffeine plasma concentrations at birth in preterm and full-term infants. Dev Pharmacol Ther 1982; 4: 182–6.
- Barone JJ, Roberts H. Human consumption of caffeine. In: Dews PB, editor. Caffeine: Perspectives from Recent Research. Berlin: Springer-Verlag; 1984. p. 59–73.
- Arnaud MJ. The pharmacology of caffeine. Prog Drug Res 1987; 31: 273–313.[Medline]
- Chvasta TE, Cooke AR. Absorption and emptying of caffeine from the human stomach. Gastroenterology 1971; 61: 838–43.[Medline]
- Rall TW. Drugs used in the treatment of asthma: the methylxanthines, cromolyn sodium, and other agents. In: Gilman AG, Rall TW, Nies AS, Taylor P, editors. Goodman and Gilman’s the Pharmacological Basis of Therapeutics. New York: Pergamon Press; 1990. p. 618–37.
- Dunwiddie TV, Masino SA. The role and regulation of adenosine in the central nervous system. Ann Rev Neurosci 2001; 24: 31–55.[CrossRef][Medline]
- Biaggioni I, Paul S, Puckett A, Arzubiaga C. Caffeine and theophylline as adenosine receptor antagonists in humans. J Pharmacol Exp Ther 1991; 258: 588–93.[Abstract/Free Full Text]
- Carter AJ, O’Connor WT, Carter MJ, Ungerstedt U. Caffeine enhances acetylcholine release in the hippocampus in vivo by a selective interaction with adenosine A1 receptors. J Pharmacol Exp Ther 1995; 273: 637–42.[Abstract/Free Full Text]
- Wilson PW, Culleton BF. Epidemiology of cardiovascular disease in the United States. Am J Kidney Dis 1998; 32: S56–65.[Medline]
- Van den Hoogen PC, Seidell JC, Menotti A, Kromhout D. Blood pressure and long-term coronary heart disease mortality in the Seven Countries study: implications for clinical practice and public health. Eur Heart J 2000; 21: 1639–42.[Free Full Text]
- Wolf-Maier K, Cooper RS, Banegas JR, Giampaoli S, Hense H-W, Joffres M, Kastarinen M, Poulter N, Primatesta P, Rodriguez-Artalejo F, Stegmayr B, Thamm M, Tuomilehto J, Vanuzzo D, Vescio F. Hypertension prevalence and blood pressure levels in 6 European countries, Canada, and the United States. JAMA 2003; 289: 2363–9.[Abstract/Free Full Text]
- Pearson TA. Cardiovascular disease in developing countries: myths, realities, and opportunities. Cardiovasc Drugs Ther 1999; 13: 95–104.[CrossRef][Medline]
- Rodgers A, Lawes C, MacMahon S. Reducing the global burden of blood pressure-related cardiovascular disease. J Hypertens 2000; 18: S3–6.[CrossRef]
- Rodgers A, MacMahon S. Blood pressure and the global burden of cardiovascular disease. Clin Exp Hypertens 1999; 21: 543–52.
- MacMahon S. Blood pressure and the risk of cardiovascular disease. N Engl J Med 2000; 342: 50–2.
- Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002; 360: 1903–13.[CrossRef][Medline]
- Bush A, Busst CM, Clarke B, Barnes PJ. Effect of infused adenosine on cardiac output and systemic resistance in normal subjects. Br J Clin Pharmacol 1989; 27: 165–71.[Medline]
- Marangos PJ, Boulenger JP. Basic and clinical aspects of adenosinergic neuromodulation. Neurosci Biobehav Rev 1985; 9: 421–30.[CrossRef][Medline]
- Biaggioni I, Olafsson B, Robertson RM, Hollister AS, Robertson D. Cardiovascular and respiratory effects of adenosine in conscious man. Circ Res 1987; 61: 779–86.[Abstract/Free Full Text]
- Fuller RW, Maxwell DL, Conradson T-BG, Dixon CMS, Barnes PJ. Circulatory and respiratory effects of infused adenosine in conscious man. Br J Pharmacol 1987; 24: 309–17.
- Smits P, Boekema P, De Abreu R, Thien T, van’t Laar A. Evidence for an antagonism between caffeine and adenosine in the human cardiovascular system. J Cardiovasc Pharmacol 1987; 10: 136–43.[Medline]
- Coney AM, Marshall JM. Role of adenosine and its receptors in the vasodilation induced in the cerebral cortex of the rat by systemic hypoxia. J Physiol 1998; 509: 507–18.[Abstract/Free Full Text]
- Hartley TR, Lovallo WR, Whitsett TL, Sung BH, Wilson MF. Caffeine and stress: implications for risk, assessment, and management of hypertension. J Clin Hypertens 2001; 3: 3.
- Palatini P, Julius S. Relevance of heart rate as a risk factor in hypertension. Curr Hypertens Rep 1999; 1: 219–24.[Medline]
- James JE. The influence of user status and anxious disposition on the hypertensive effects of caffeine. Int J Psychophysiol 1990; 10: 171–9.[CrossRef][Medline]
- Lane JD, Williams RB. Caffeine affects cardiovascular responses to stress. Psychophysiology 1985; 22: 648–55.[Medline]
- Ammar R, Son JC, Kluger J, White CM. Evaluation of electrocardiographic and hemodynamic effects of caffeine with acute dosing in healthy volunteers. Pharmacotherapy 2001; 21: 437–42.[CrossRef][Medline]
- Chelsky LB, Cutler JE, Griffith K, Kron H, McClelland JH, McAnulty JH. Caffeine and ventricular arrhythmias: an electrophysiological approach. JAMA 1990; 264: 2236–40.[Abstract]
- Sutherland DJ, McPherson DD, Renton KW, Spencer CA, Montague TJ. The effect of caffeine on cardiac rate, rhythm, and ventricular repolarization. Chest 1985; 87: 319–24.[Abstract/Free Full Text]
- Stamler JS, Goldman ME, Gomes J, Matza D, Horowitz SF. The effect of stress and fatigue on cardiac rhythm in medical interns. J Electrocardiol 1992; 25: 333–8.[CrossRef][Medline]
- Graboys TB. The effect of caffeine on ventricular ectopic activity in patients with malignant ventricular arrhythmia. Arch Intern Med 1989; 149: 637–9.[Abstract]
- Ahola I, Jauhiainen M, Aro A. The hypocholesterolaemic factor in boiled coffee is retained by a paper filter. J Intern Med 1991; 230: 293–7.[Medline]
- Christensen B, Mosdol A, Rettstol L, Thelle DS. Abstention from filtered coffee reduces the concentrations of plasma homocysteine and serum cholesterol—a randomized controlled trial. Am J Clin Nutr 2001; 74: 302–7.[Abstract/Free Full Text]
- De Roos B, Van Tol A, Urgert R, Scheek LM, Van Gent T, Buytenhek R, Princen HM, Katan MB. Consumption of French-press coffee raises cholesteryl ester transfer protein activity levels before LDL cholesterol in normolipidaemic subjects. J Intern Med 2000; 248: 211–6.[CrossRef][Medline]
- Fried RE, Levine DM, Kwiterovich PO, Diamond EL, Wilder LB, Moy TF, Pearson TA. The effect of filtered-coffee consumption on plasma lipid levels: results of a randomized clinical trial. JAMA 1992; 267: 811–5.[Abstract]
- Urgert R, van Vliet T, Zock PL, Katan MB. Heavy coffee consumption and plasma homocysteine: a randomized controlled trial in healthy volunteers. Am J Clin Nutr 2000; 72: 1107–10.[Abstract/Free Full Text]
- Casiglia E, Paleari CD, Petucco S, Bongiovi S, Colangeli G, Baccilieri MS, Pavan L, Pernice M, Pessina AC. Haemodynamic effects of coffee and purified caffeine in normal volunteers: a placebo-controlled clinical study. J Hum Hypertens 1992; 6: 95–9.[Medline]
- France C, Ditto B. Cardiovascular responses to occupational stress and caffeine in telemarketing employees. Psychosom Med 1989; 51: 145–51.[Abstract/Free Full Text]
- Goldstein IB, Shapiro D. The effects of stress and caffeine on hypertensives. Psychosom Med 1987; 49: 226–35.[Abstract/Free Full Text]
- Hartley TR, Sung BH, Pincomb GA, Whitsett TL, Wilson MF, Lovallo WR. Hypertension risk status and effect of caffeine on blood pressure. Hypertension 2000; 36: 1337–41.
- James JE, Richardson M. Pressor effects of caffeine and cigarette smoking. Br J Clin Psychol 1991; 30: 276–8.
- Lane JD. Caffeine and cardiovascular responses to stress. Psychosom Med 1983; 45: 447–51.[Abstract/Free Full Text]
- Lane JD, Adcock RA, Williams RB, Kuhn CM. Caffeine effects on cardiovascular and neuroendocrine responses to acute psychosocial stress and their relationship to level of habitual caffeine consumption. Psychosom Med 1990; 52: 320–36.[Abstract/Free Full Text]
- Lane JD, Williams RB. Cardiovascular effects of caffeine and stress in regular coffee drinkers. Psychophysiology 1987; 24: 157–64.[Medline]
- Lovallo WR, Pincomb GA, Sung BH, Passey RB, Suasen KP, Wilson MF. Caffeine may potentiate adrenocortical stress responses in hypertension-prone men. Hypertension 1989; 14: 170–6.[Abstract/Free Full Text]
- Lovallo WR, Pincomb GA, Sung BH, Everson SA, Passey RB, Wilson MF. Hypertension risk and caffeine’s effect on cardiovascular activity during mental stress in young men. Health Psychol 1991; 10: 236–43.[CrossRef][Medline]
- Lovallo WR, al’Absi M, Pincomb GA, Everson SA. Caffeine and behavioral stress effects on blood pressure in borderline hypertensive Caucasian men. Health Psychol 1996; 15: 11–7.[CrossRef][Medline]
- Lovallo WR, al’Absi M, Pincomb GA, Green A, Passey RB, Sung BH, Wilson MF. Caffeine extended stress, and blood pressure in borderline hypertensive men. Int J Behav Med 2000; 7: 183–8.
- Pincomb GA, Lovallo WR, Passey RB, Wilson MF. Effect of behavior state on caffeine’s ability to alter blood pressure. Am J Cardiol 1988; 61: 798–802.[CrossRef][Medline]
- Pincomb GA, Wilson MF, Sung BH, Passey RB, Lovallo WR. Effects of caffeine on pressor regulation during rest and exercise in men at risk for hypertension. Am Heart J 1991; 122: 1107–15.[CrossRef][Medline]
- Pincomb GA, Sung BH, Lovallo WR, Wilson MF. Consistency of cardiovascular response pattern to caffeine across multiple studies using impedance and nuclear cardiography. Biol Psychol 1993; 36: 131–8.[CrossRef][Medline]
- Shepard JD, al’Absi M, Whitsett TL, Passey RB, Lovallo WR. Additive pressor effects of caffeine and stress in male medical students at risk for hypertension. Am J Hypertens 2000; 13: 475–81.[CrossRef][Medline]
- Smits P, Thien T, van ’t Laar A. The cardiovascular effects of regular and decaffeinated coffee. Br J Clin Pharmacol 1985; 19: 852–4.[Medline]
- Smits P, Thien T, van ’t Laar A. Circulatory effects of coffee in relation to the pharmacokinetics of caffeine. Am J Cardiol 1985; 56: 958–63.[CrossRef][Medline]
- Smits P, Pieters G, Thien T. The role of epinephrine in the circulatory effects of coffee. Clin Pharmacol Ther 1986; 40: 431–7.[Medline]
- Sung BH, Whitsett TL, Lovallo WR, al’Absi M, Pincomb GA, Wilson MF. Prolonged increase in blood pressure by a single oral dose of caffeine in mildly hypertensive men. Am J Hypertens 1994; 7: 755–8.[Medline]
- Vlachopoulos C, Hirata K, Stefanadis C, Toutouzas P, O’Rourke MF. Caffeine increases aortic stiffness in hypertensive patients. Am J Hypertens 2003; 16: 63–6.[CrossRef][Medline]
- Freestone S, Yeo WW, Ramsay LE. Effect of coffee and cigarette smoking on the blood pressure of patients with accelerated (malignant) hypertension. J Hum Hypertens 1995; 9: 89–91.[Medline]
- al’Absi M, Lovallo WR, McKey B, Sung BH, Whitsett TL, Wilson MF. Hypothalamic-pituitary-adrenocortical responses to psychological stress and caffeine in men at high and low risk for hypertension. Psychosom Med 1998; 60: 521–7.[Abstract/Free Full Text]
- Lane JD, Pieper CF, Phillips-Bute BG, Bryant JE, Kuhn CM. Caffeine affects cardiovascular and neuroendocrine activation at work and home. Psychosom Med 2002; 64: 323–31.
- Robertson D, Wade D, Workman R, Woosley RL, Oates JA. Tolerance to the humoral and hemodynamic effects of caffeine in man. J Clin Invest 1981; 67: 1111–7.
- Goldstein IB, Shapiro D, Hui KK, Yu JL. Blood pressure response to the "second cup of coffee." Psychosom Med 1990; 52: 337–45.[Abstract/Free Full Text]
- Lane JD, Manus DC. Persistent cardiovascular effects with repeated caffeine administration. Psychosom Med 1989; 51: 373–80.[Abstract/Free Full Text]
- Green PJ, Suls J. The effects of caffeine on ambulatory blood pressure, heart rate, and mood in coffee drinkers. J Behav Med 1996; 19: 111–28.[CrossRef][Medline]
- James JE. Chronic effects of habitual caffeine consumption on laboratory and ambulatory blood pressure levels. J Cardiovasc Res 1994; 1: 159–64.
- Jeong D, Dimsdale JE. The effects of caffeine on blood pressure in the work environment. Am J Hypertens 1990; 3: 749–53.[Medline]
- Rakic V, Burke V, Beilin LJ. Effects of coffee on ambulatory blood pressure in older men and women: a randomized controlled trial. Hypertension 1999; 33: 869–73.[Abstract/Free Full Text]
- Superko HR, Myll J, DiRicco C, Williams PT, Bortz WM, Wood PD. Effects of cessation of caffeinated-coffee consumption on ambulatory and resting blood pressure in men. Am J Cardiol 1994; 73: 780–4.[CrossRef][Medline]
- Eggertsen R, Andreasson A, Hedner T, Karlberg BE, Hansson L. Effect of coffee on ambulatory blood pressure in patients with treated hypertension. J Intern Med 1993; 233: 351–5.[Medline]
- MacDonald TM, Sharpe K, Fowler G, Lyons D, Freestone S, Lovell HG, Webster J, Petrie JC. Caffeine restriction: Effect of mild hypertension. BMJ 1991; 303: 1235–8.
- James JE. Caffeine and blood pressure: habitual use is a preventable cardiovascular risk factor. Lancet 1997; 349: 279–81.[CrossRef][Medline]
- Bak AA, Grobbee DE. A randomized study on coffee and blood pressure. J Hum Hypertens 1990; 4: 259–64.[Medline]
- van Dusseldorp M, Smits P, Thien T, Katan MB. Effect of decaffeinated versus regular coffee on blood pressure: a 12-week, double-blind trial. Hypertension 1989; 14: 563–9.[Abstract/Free Full Text]
- Jee SH, He J, Whelton PK, Suh I, King J. The effect of chronic coffee drinking on blood pressure: a meta-analysis of controlled clinical trials. Hypertension 1999; 33: 647–52.[Abstract/Free Full Text]
- Denaro CP, Brown CR, Jacob PI, Benowitz NL. Effects of caffeine with repeated dosing. Eur J Clin Pharmacol 1991; 40: 273–8.[CrossRef][Medline]
- Lelo A, Miners JO, Robson R, Birkett DJ. Assessment of caffeine exposure: caffeine content of beverages, caffeine intake, and plasma concentrations of methylxanthines. Clin Pharmacol Ther 1986; 39: 54–9.[Medline]
- Pfeifer RW, Notari RE. Predicting caffeine plasma concentrations resulting from consumption of food or beverages: a simple method and its origin. Drug Intell Clin Pharm 1988; 22: 953–9.[Abstract]
- Shi J, Benowitz NL, Denaro CP, Sheiner LB. Pharmacokinetic-pharmacodynamic modeling of caffeine: tolerance to pressor effects. Clin Pharmacol Ther 1993; 53: 6–14.[Medline]
- Johansson B, Georgiev V, Lindström K, Fredholm BB. A1 and A2A adenosine receptors and A1 mRNA in mouse brain: effect of long-term caffeine treatment. Brain Res 1997; 762: 153–64.[CrossRef][Medline]
- Ramkuma V, Bumgarner JR, Jacobson KA, Stiles GL. Multiple components of A1 adenosine receptor-adenylate cyclase system are regulated in rat cerebral cortex by chronic caffeine ingestion. J Clin Invest 1988; 82: 242–7.
- James JE. Caffeine and Health. London: Academic Press; 1991.
- Greenland S. A meta-analysis of coffee, myocardial infarction, and coronary death. Epidemiology 1993; 4: 366–74.[Medline]
- Kawachi I, Colditz GA, Stone CB. Does coffee drinking increase the risk of coronary heart disease? results from a meta-analysis. Br Heart J 1994; 72: 269–75.[Abstract/Free Full Text]
- Schreiber GB, Robins M, Maffeo CE, Masters MN, Bond AP, Morganstein D. Confounders contributing to the reported associations of coffee or caffeine with disease. Prev Med 1988; 17: 295–309.[CrossRef]
TOP
ABSTRACT
INTRODUCTION