The effect of coffee on the cardiovascular system still remains controversial although it is one of the most popular beverages consumed worldwide in large amounts.1, 2, 3 However, the available data regarding the effect of caffeine intake on atherosclerotic inflammatory and thrombotic mechanisms are rather scarce.3 Moreover, the combination of coffee intake with smoking is deleterious exhibiting an association with increased blood pressure (BP), pronounced vascular dysfunction and unfavorable prognosis either in healthy or hypertensive populations.4, 5, 6 On the basis of the above, we examined the impact of heavy coffee intake on plasma inflammatory and fibrinolytic markers such as inteurleukin-6 (IL-6) and plasminogen-activator inhibitor type 1 (PAI-1) in essential hypertensive smokers.

In our pool of middle-aged patients with newly diagnosed (within the last 2 years) untreated stage I–II essential hypertension,7 who referred to the outpatient hypertension unit during a period of 6 months, we determined in smokers (defined as those who smoke ⩾1 cigarette/day) the usual coffee consumption according to a previously reported food frequency questionnaire by Tuomilehto et al.8 More specifically, patients were asked, ‘How many cups of coffee do you drink per day or per week?’ It should be noted that all types of coffee were adjusted for 1 cup of coffee (average caffeine concentration of 90 mg/cup).2, 3 Exclusion criteria included heart failure, atherosclerotic cardiovascular disease, diabetes mellitus or glucose intolerance, familial hypercholesterolemia, augmented serum creatinine concentration or overt proteinuria, any clinical and laboratory evidence of inflammation or medical treatment over a period of 1 month before entry in the study and any other clinically significant concurrent medical conditions. On the basis of the above exclusion criteria and the obtained answers to the food frequency questionnaire, and given that we would like to focus on the two ‘edges’ of coffee consumption, our final study population consisted of 30 consecutive heavy coffee drinkers (>4 cups/day) and 30 consecutive light coffee drinkers (<1 cup/day or <7 cups/week) matched for age, sex, and pack-years. The study protocol was approved by our institutional ethics committee and all participants gave written informed consent.

Office BP measurements were obtained according to recent guidelines.7 In all subjects venous blood samples were obtained between 0800 and 0900 after overnight fast and 15 min of supine rest for the estimation of lipid levels, fasting glucose, hemoglobin A(1c), plasma IL-6 and PAI-1 concentrations according to established methods.9, 10, 11, 12

Significant differences between the study subgroups were determined using the Student independent-samples t-test or the χ2 test where appropriate. Pearson correlation was performed in order to determine correlations between any of the parameters. Analysis of covariance was performed in order to detect significant differences of IL-6 and PAI-1 levels between the heavy and light coffee drinkers, after the adjustment for a number of covariates.

Heavy (n=30) and light (n=30) coffee drinkers did not differ regarding age, sex, body mass index (BMI), waist to hip ratio, diastolic BP, heart rate, pack-years and metabolic profile (P=NS, for all cases) (Table 1). In contrast, heavy coffee drinkers compared to light coffee drinkers had significantly increased systolic BP (by 11 mm Hg, P=0.001) and pulse pressure (by 7 mm Hg, P=0.015). Interestingly, heavy coffee drinkers compared to light coffee drinkers exhibited on average 120% higher IL-6, and 84% higher PAI-1 plasma concentrations (P=0.005 and 0.0001, respectively).

Table 1 Demographic, clinical and laboratory data for the heavy coffee drinking and light coffee drinking group of essential hypertensive smokers
Full size table

In the entire population, IL-6 was associated with BMI (r=0.221, P<0.05), pack-years index (r=0.347, P<0.001), office systolic BP (r=0.372, P<0.05), and low-density lipoprotein cholesterol (r=0.206, P<0.05). There was also a positive association between PAI-1 and waist to hip ratio (r=0.204, P<0.05), pack-years index (r=0.228, P<0.05), office systolic BP (r=0.284, P<0.005), and low-density lipoprotein cholesterol (r=0.327, P<0.001). By applying analysis of covariance, it was shown that both IL-6 and PAI-1 levels were significantly different between the heavy and light coffee drinking group, after adjusting for age, sex, BMI, pack-years, office systolic BP, and low-density lipoprotein cholesterol levels (P=0.013 and P=0.04, respectively).

The main finding of our study is that in essential hypertensive smokers, heavy coffee drinking compared to light coffee drinking is associated with 120% higher IL-6 and 84% higher PAI-1 plasma concentrations, independently of confounding factors. This study is the first to show that in this specific setting, heavy coffee consumption has additional detrimental effects on inflammatory and prothrombotic mechanisms.

In our study, office systolic BP levels were more augmented in the heavy compared to light coffee drinking group. This is in accordance with several studies that have shown an interplay between coffee drinking and BP status.1, 2, 3, 4 Although it is possible that increased systolic BP leads to enhanced inflammation and impaired fibrinolysis, the cross-sectional nature of our data opens the possibility that increased inflammation, triggered by excess coffee drinking, may lead to higher systolic BP levels.3, 9, 10, 11, 12 From another point of view, heavy coffee consumption may activate inflammatory pathways and affect fibrinolysis through additional pathways not associated with hemodynamic load. The above are supported by the fact that the differences in markers estimated persisted even after adjusting for BP status. However, whatever the pathophysiological underpinning for the abovementioned interrelationships, the main point is that heavy coffee intake is accompanied by increased systolic BP and heightened levels of both IL-6 and PAI-1, providing a possible link to augmented cardiovascular risk. Further properly designed studies are needed to elucidate the causal and temporal ordering of the association of coffee intake, BP and the markers estimated.

In these lines, the possible effect of adipose tissue-derived inflammatory and atherogenic processes could be attenuated by the fact that all participants were not clinically obese. Herein, the exhibited difference in IL-6 values between heavy and light coffee drinkers were not affected by BMI, which did not differ between groups. Additionally, the normoglycemic and rather non-high lipid profile may explain the lack of interrelationships of PAI-1, BMI, and IL-6.9, 10, 11, 12 Regarding the regulation of the sympathetic system, as assessed by heart rate, we showed that the latter was not further deteriorated by heavy coffee drinking decreasing the possibility of the implication of autonomic imbalance in inflammatory activation.

The complex cardiovascular response of coffee intake could be attributed not only to caffeine but also to the other biologically active substances. The contained caffeine causes adenosine receptor blocking, cathecholamines release, augmentation of sympathetic nervous activity, involvement of the renin-angiotensin-aldosterone axis, and hyperhomocysteinemia1, 2, 3, 4 leading to pro-inflammatory activation and impairment of fibrinolysis/thrombosis.9, 10, 11, 12 On the other hand, the contained polyphenols and kahweol may have an antioxidant and anti-inflammatory beneficial action.1, 2, 3 On the basis of our findings, one could suggest that coffee when consumed in large amounts (above 4 cups/day) alters unfavorably the equilibrium between the beneficial and harmful actions of the different substances contained, causing IL-6 and PAI-1 augmentation. Further prospective population studies are needed in order to define the amount of coffee that can be consumed without increasing the cardiovascular risk, in this setting.

There are some limitations that may affect the global application of our results. These are the cross-sectional nature of the study that could not definitively rule out chance as a potential explanation of the results, the self-reported questionnaires and the fact that groups consisted of light smokers and were matched for pack-years. Finally, a potential strength of the present study is that dietary interventions which influence BP and atherosclerotic processes seem to have a more profound effect in hypertensive patients.1, 2, 3

Concluding, heavy coffee drinking of more than 4 cups/day, is associated with augmented IL-6 and PAI-1 levels in hypertensive smokers. These findings suggest that heavy coffee intake through pronounced inflammatory activation and impaired fibrinolysis could unfavorably affect subclinical atherosclerosis progression and cardiovascular risk in this setting.