- Email: [email protected]
Arterial stiffening and systemic endothelial activation induced by smoking
International Journal of Cardiology 189 (2015) 293–298
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Arterial stiffening and systemic endothelial activation induced by smoking☆ The role of COX-1 and COX-2 Charalambos Vlachopoulos ⁎, Konstantinos Aznaouridis, Athanassios Bratsas, Nikolaos Ioakeimidis, Ioanna Dima, Panagiotis Xaplanteris, Christodoulos Stefanadis, Dimitris Tousoulis Peripheral Vessels and Hypertension Units, 1st Department of Cardiology, Athens Medical School, Hippokration Hospital, Athens, Greece
a r t i c l e
i n f o
Article history: Received 2 March 2015 Accepted 3 April 2015 Available online 6 April 2015 Keywords: Smoking Arterial elasticity Arterial stiffness Cyclooxygenase
a b s t r a c t Background: Arterial stiffness is an established predictor of cardiovascular risk. We explored the effects of acute smoking on arterial stiffness, systemic inflammation and endothelial activation in chronic smokers and the contribution of cyclooxygenases-1 and 2 (COX-1 and COX-2). Methods and results: In a randomized, double-blind, cross-over study, we investigated in 28 young smokers the vascular and systemic effects of smoking one cigarette, 3 h after receiving 1000 mg of aspirin (a non-selective COX-1 and COX-2 inhibitor) or placebo (aspirin substudy), or 200 mg of celecoxib (a selective COX-2 inhibitor) or placebo (celecoxib substudy). Smoking increased carotid–femoral pulse wave velocity (PWV, a marker of aortic stiffness), indicating an adverse effect on arterial elastic properties. Similarly, circulating levels of asymmetric dimethylarginine (ADMA) were increased after smoking. Aspirin fully prevented the smoking-induced increase of PWV after smoking. In contrast, celecoxib only partially prevented the smoking-induced increase of PWV. Both aspirin and celecoxib prevented to a similar extent the increase of ADMA levels after smoking. Conclusions: Smoking one cigarette is associated with a deterioration of arterial stiffness and with systemic endothelial activation in chronic smokers. Both COX-1 and COX-2, but primarily COX-1, mediate these unfavorable effects of smoking. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Smoking affects adversely several cardiovascular homeostatic pathways and comprises the most important modifiable risk factor for the development and progression of atherosclerotic disease [1,2]. The detrimental impact of cigarette smoking on cardiovascular physiology may be mediated, at least partly, by unfavorable effects on arterial stiffness and endothelial homeostasis, which are important markers of cardiovascular disease and independent predictors of the corresponding risk [3–9]. We and other investigators have previously shown that smoking deteriorates arterial stiffness and endothelial function in several settings [10–15]. Activation of platelets, increased oxidative stress and reduced nitric oxide (NO) availability are considered common final mechanisms that mediate the unfavorable cardiovascular effects of smoking [1,2]. However, the exact antecedent biochemical pathways that are modified ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Profiti Elia 24, Athens 14575, Greece. E-mail address: [email protected] (C. Vlachopoulos).
http://dx.doi.org/10.1016/j.ijcard.2015.04.029 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.
adversely by smoking have not been fully elucidated. Interestingly, recent evidence suggests that cyclooxygenases (COXs) are upregulated in smokers and may contribute, through an excess production of COX1 and COX-2-derived prostanoids, to the smoking-related cardiovascular injury [16–18]. It is established that arterial performance is impaired in chronic smokers [1,2,15]. However, although it has been previously shown that active and passive smoking may impair certain indices of arterial function both in chronic smokers and in smoking-naïve subjects [10,11,13,14,19], so far the effects of smoking a single cigarette on arterial function and the potential mediators for those effects have not been thoroughly investigated in chronic smokers. Accordingly, in the present study we investigated, through a cause-and-effect testing, whether acute smoking alters arterial stiffness and systemic inflammatory and endothelial activation in subjects who smoke on a chronic basis. Furthermore, in order to explore the potential role of COX-1 and COX-2, we evaluated the effect of high-dose aspirin (a nonselective COX-1 and COX-2 inhibitor) and celecoxib (a selective COX-2 inhibitor) on the smoking-induced changes of arterial stiffness and circulating levels of markers of inflammation and endothelial activation.
294
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
2. Methods 2.1. Study population We studied 28 apparently healthy current smokers (mean age 32.9 years, range 26 to 45, 15 men) who were free of hypertension, diabetes, dyslipidemia, or family history of premature vascular disease. All subjects were clinically well and were not taking any regular cardiovascular medications or antioxidant vitamin supplementation. They had not had any bacterial or other inflammatory disorder and they did not report use of anti-inflammatory or steroid substances during the past 2 months. No female participant was on oral contraceptives or estrogen replacement therapy. The study protocol was approved by our Institutional Research Ethics Committee and all subjects gave written informed consent.
artery) − (distance from carotid artery to the suprasternal notch). Three readings were performed, and the mean value of PWV was calculated. 2.5. Measurement of markers of endothelial activation and inflammation Immediately after acquisition of venous blood, plasma and serum were separated by centrifugation (3000 ×g at 4 °C for 15 min), then placed in aliquots and stored at − 70 °C. High-sensitivity C-reactive protein (hsCRP) was measured by immunophelometry (Dade Behring, Marburg, Germany). High sensitivity interleukin-6 (IL-6) was measured using ELISA (R&D Systems, Minneapolis, MN, USA). Asymmetric dimethylarginine (ADMA) was also measured using ELISA (Immundiagnostik AG, Bensheim, Germany). 2.6. Statistical analysis
2.2. Study design The study consisted of two separate substudies (aspirin and celecoxib substudies) and was carried out using a randomized, double-blind, placebo-controlled, cross-over design. Each substudy consisted of two treatment sessions, one with an anti-inflammatory agent (aspirin or celecoxib) and one with placebo. Subjects had fasted for at least 6 h and had abstained from caffeine, ethanol and flavonoid-containing beverage intake for at least 12 h before each session. They had been instructed not to smoke for at least 4 h before the vascular studies. All vascular studies were performed in the afternoon between 3 and 7 p.m. on two separate days, in a quiet, temperature-controlled room at 23 °C. Baseline studies for evaluation of aortic stiffness were performed after a 20-min rest period in the supine position. Then, in the aspirin substudy, the subjects were randomized to taking aspirin 1000 mg (Aspirin, Bayer, 2 tablets of 500 mg) or placebo. In the celecoxib substudy, the subjects were randomized to taking celecoxib 200 mg (Celebrex, Pfizer, 2 capsules of 100 mg) or placebo. Inert tablets were given as placebo. After repeating the vascular studies at 3 h after the administration of anti-inflammatory agent/placebo (pre-smoking time point), the subjects smoked one standard cigarette (1 mg nicotine, 14 mg tar) in a sitting position over 5 min. The vascular measurements were repeated immediately after the completion of smoking (time Sm0) and 20 min thereafter (Sm20). Venous blood for assessment of markers of inflammation and endothelial activation was drawn into vacutainer tubes after baseline studies and after the studies at Sm20. After 2–10 days, all participants of both substudies returned and all studies were done after taking the opposite treatment (anti-inflammatory agent/placebo), according to the same schedule. Both subjects and investigators were blinded to treatment allocation throughout the study. 2.3. Measurement of blood pressure Supine blood pressure (BP) was measured with mercury sphygmomanometry after a 5-min rest, just before evaluation of arterial stiffness. Three readings 1 min apart were performed, and the mean value of systolic and diastolic BP was calculated. 2.4. Evaluation of aortic stiffness The pulse travels at a higher velocity in a stiff aorta and vice versa. Carotid–femoral pulse wave velocity (PWV), an established index of aortic stiffness [3,5,8,17,18], was calculated from measurements of pulse transit time and the distance traveled between two recording sites (pulse wave velocity = distance [m] / transit time [s]) using a validated noninvasive device (Complior®, Artech Medical, Pantin, France), as previously described [5,17,18]. Two different pulse waves were obtained simultaneously at two sites (at the base of the neck for the common carotid and over the right femoral artery) with two transducers. The distance was defined as: (distance from the suprasternal notch to femoral
Sample size calculation was based on the hypothesis that smoking would be associated with changes in arterial and biochemical indices of at least 0.9 of 1 SD. Therefore, we estimated that 12 subjects would provide 80% power at the 5% level of significance to detect such a difference between anti-inflammatory agents (aspirin or celecoxib) and placebo sessions in a cross-over design study. Continuous variables are expressed as mean value ± SD, whereas categorical variables as absolute frequencies. All continuous variables were tested for normal distribution by using the Kolmogorov–Smirnov criterion. For skewed variables, data are expressed as median value (25th–75th percentile) and logarithmic transformation was performed prior to analysis. The within-sessions and between-sessions (anti-inflammatory agent/placebo) effects during the 3-h pre-smoking period were evaluated by paired t-test. The effects of smoking within the sessions of anti-inflammatory agents (aspirin or celecoxib) and placebo were evaluated with analysis of variance (ANOVA) for repeated measures (3 timepoints: 3 h/pre-smoking, Sm0 and Sm20). When a significant effect occurred, the location of pairwise differences was determined by using paired t-test followed by the Bonferroni correction. The betweensubstudies effect of active treatment (celecoxib or aspirin) on the levels of endothelial/inflammatory markers was evaluated by unpaired t-test. Exact P values b0.05 were considered statistically significant. Data analysis was performed with SPSS software, version 13.0 (Chicago, IL). 3. Results Clinical characteristics are shown in Table 1. There were no significant differences in baseline hemodynamic, vascular and biochemical characteristics between the study substudies/sessions. The smoking effect on hemodynamic and arterial function indices at Sm0 and Sm20 in each session (anti-inflammatory agent or placebo) of the substudies (aspirin or celecoxib) is reported as the respective post-smoking mean value (Sm0 or Sm20) minus pre-smoking mean value. The net effect of each anti-inflammatory agent (aspirin or celecoxib) in the 3-h pretreatment period of each substudy was calculated as anti-inflammatory agent (aspirin or celecoxib) effect (3 h/pre-smoking minus baseline mean value in the anti-inflammatory agent session) minus placebo effect (3 h/pre-smoking minus baseline mean value in the placebo session). Changes of biochemical variables were calculated as mean value at the end of study minus mean value at baseline. In case of skewed variables, geometric means (anti-logs of mean log-values) were used for calculating responses. 3.1. Systemic hemodynamic effects 3.1.1. Aspirin substudy 3.1.1.1. Pre-smoking treatment period. Compared to placebo, aspirin decreased significantly systolic BP (by 3.5 mm Hg, P = 0.039) and pulse
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
295
Table 1 Baseline characteristics of the study population according to substudies/sessions.
Age, year Males/females Body-mass index, kg/m2 Waist/hip Glucose, mmol/L Total cholesterol, mmol/L Triglycerides, mmol/L HDL-cholesterol, mmol/L
Systolic BP, mm Hg Diastolic BP, mm Hg Pulse pressure, mm Hg Heart rate, bpm Pulse wave velocity, m/s hsCRP, mg/L IL-6, pg/mL ADMA, μmol/L
Aspirin substudy (N = 12)
Celecoxib substudy (N = 16)
P-value
36.3 ± 4.6 7/5 22.9 ± 3.5 0.86 ± 0.09 4.72 ± 0.46 4.77 ± 0.73 0.88 ± 0.53 1.71 ± 1.24
33.2 ± 4.1 8/8 22.2 ± 3.2 0.85 ± 0.08 4.64 ± 0.63 4.80 ± 0.89 0.74 ± 0.38 1.48 ± 0.34
0.07 0.66 0.60 0.84 0.71 0.93 0.40 0.49
Aspirin
Placebo
P value
Celecoxib
Placebo
P value
116 ± 11 71 ± 10 44 ± 3 66 ± 9 6.69 ± 0.84 0.53 (0.18–1.64) 0.90 (0.48–1.46) 0.53 (0.41–1.18)
118 ± 9 74 ± 7 45 ± 7 67 ± 7 6.67 ± 0.97 0.56 (0.18–1.13) 0.98 (0.51–1.44) 0.56 (0.15–0.90)
0.50 0.47 0.91 0.73 0.96 0.94 0.96 0.21
108 ± 12 68 ± 8 41 ± 7 67 ± 9 5.11 ± 0.77 0.58 (0.36–0.88) 1.12 (0.85–1.42) 0.58 (0.47–0.65)
109 ± 13 67 ± 10 42 ± 6 67 ± 7 5.07 ± 0.81 0.61 (0.34–1.32) 0.96 (0.76–1.35) 0.57 (0.51–0.70)
0.87 0.77 0.46 0.99 0.89 0.34 0.37 0.32
Categorical variables are presented as absolute frequencies, while continuous variables as mean ± SD or median (interquartile range). Probability values derived from Student's t-test for unpaired (between substudies) or paired measures (within substudies).
pressure (by 4.6 mm Hg, P = 0.013) after 3 h. We observed no significant changes of diastolic pressure or heart rate (all P = NS).
Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 11.8 bpm, P b 0.001) and Sm20 (by 6.4 bpm, P b 0.001). 3.2. Effects on arterial stiffness
3.1.1.2. Smoking. Within the placebo session, smoking increased systolic BP (P = 0.013 by ANOVA) and diastolic BP (P = 0.002 by ANOVA). Significant effects were observed only at Sm0 (systolic BP: increases of 7.3 mm Hg at Sm0, P = 0.01 and 4.7 mm Hg at Sm20, P = NS; diastolic BP increases of 7.5 mm Hg at Sm0, P = 0.005 and 3.4 mm Hg at Sm20, P = NS). Pulse pressure did not change with smoking (P = NS by ANOVA, decrease of 1.8 mm Hg at Sm0 and increase of 1.3 mm Hg at Sm20). Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 11.8 bpm, P = 0.001) and Sm20 (by 5.5 bpm, P = 0.014). Aspirin did not influence the smoking-induced changes of any of the above variables. Within the aspirin session, smoking increased systolic BP (P = 0.010 by ANOVA, increases of 6.7 mm Hg at Sm0, P = 0.01 and 1.4 mm Hg at Sm20, P = NS) and diastolic BP (P = 0.019 by ANOVA, increases of 4.0 mm Hg at Sm0, P = 0.018 and 0.9 mm Hg at Sm20, P = NS). Pulse pressure did not change (P = NS by ANOVA). Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 12.4 bpm, P b 0.001) and Sm20 (by 4.5 bpm, P = 0.011).
3.2.1. Aspirin substudy 3.2.1.1. Pre-smoking treatment period. PWV did not change significantly after taking aspirin (response of −0.07 m/s, P = 0.73, Fig. 1). 3.2.1.2. Smoking. In the placebo session, we observed a significant rise of PWV with smoking (P b 0.001 by ANOVA, increases of 0.65 m/s at Sm0, P b 0.001 and by 0.47 m/s at Sm20, P = 0.06, Fig. 1). Aspirin fully prevented the smoking-induced increase of PWV (P = 0.24 by ANOVA within the aspirin session, increases of 2.4 m/s at Sm0 and 1.4 m/s at Sm20, both P = NS, Fig. 1). 3.2.2. Celecoxib substudy 3.2.2.1. Pre-smoking treatment period. PWV was not significantly decreased 3 h after taking celecoxib (response of − 0.14 m/s, P = 0.27, Fig. 1).
3.1.2. Celecoxib substudy 3.1.2.1. Pre-smoking treatment period. Compared to placebo, celecoxib decreased significantly systolic BP (by 5.1 mm Hg, P = 0.014) and diastolic BP (by 3.2 mm Hg, P = 0.041) after 3 h. We observed no significant changes of pulse pressure or heart rate (all P = NS). 3.1.2.2. Smoking. Within the placebo session, smoking increased systolic BP (P = 0.001 by ANOVA) and diastolic BP (P = 0.002 by ANOVA). Significant effects were observed only at Sm0 (systolic BP: increases of 9.4 mm Hg at Sm0, P b 0.001 and 1.9 mm Hg at Sm20, P = NS; diastolic BP increases of 6.8 mm Hg at Sm0, P = 0.01 and 2.4 mm Hg at Sm20, P = NS). Pulse pressure had a marginal increase after smoking (P = 0.06 by ANOVA, increase of 2.6 mm Hg at Sm0, P = 0.09). Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 11.0 bpm, P b 0.001) and Sm20 (by 7.3 bpm, P = 0.003). Celecoxib did not influence the smoking-induced changes of any of the above variables. Within the celecoxib session, smoking increased systolic BP (P b 0.001 by ANOVA, increases of 6.6 mm Hg at Sm0, P = 0.002 and 3.6 mm Hg at Sm20, P = 0.04) and diastolic BP (P = 0.001 by ANOVA, increases of 4.8 mm Hg at Sm0, P = 0.001 and 2.5 mm Hg at Sm20, P = NS). Pulse pressure did not change (P = NS by ANOVA).
3.2.2.2. Smoking. In the placebo session, we observed a significant rise of PWV with smoking (P b 0.001 by ANOVA, increases of 0.63 m/s at Sm0, P b 0.001 and 0.44 m/s at Sm20, P = 0.008, Fig. 1). In the celecoxib session, PWV increased significantly with smoking (P = 0.002 by ANOVA). However, this increase was observed only at Sm0 (by 0.38 m/s at Sm0, P = 0.004 and by 0.16 m/s at Sm20, P = NS, Fig. 1). Accordingly, celecoxib did not fully prevent the smokinginduced increase of PWV but accelerated the restoration of PWV at Sm20. 3.3. Effects on systemic inflammation and endothelial function Smoking or anti-inflammatory drug intake did not influence the level of hsCRP and IL-6 in the placebo sessions or after aspirin or celecoxib (Fig. 2). Circulating levels of ADMA increased after smoking in the placebo sessions of both the aspirin and celecoxib substudies (P b 0.05 in both substudies). Aspirin prevented the smoking-induced increase of ADMA (change of − 0.17 vs 0.47 μmol/L with aspirin and placebo respectively, P = 0.035) (Fig. 2). Similarly, celecoxib abrogated the smoking-induced increase of circulating levels of ADMA (change of
296
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
Fig. 1. Changes in carotid–femoral pulse wave velocity in the celecoxib and placebo sessions (celecoxib substudy, bottom panel) and in the aspirin and placebo sessions (aspirin substudy, top panel) before and 3 h after treatment (pretreatment substudy, open area) and immediately after and 20 min after smoking (Sm0 and Sm20 in the smoking substudy, shaded area). P value in the pretreatment study (open area) indicates comparison between anti-inflammatory drug (aspirin/celecoxib) and placebo. P values in the smoking period indicate repeated measures comparison by ANOVA within each session (anti-inflammatory drug or placebo). *P b 0.05 vs. pre-smoking (3 h) within the session (paired t test followed by Bonferroni correction). Triangles indicate mean changes from baseline; bars, their 95% CIs.
− 0.03 vs 0.03 μmol/L with celecoxib and placebo respectively, P = 0.033) (Fig. 2). Comparison between the 2 substudies showed that aspirin and celecoxib had a similar effect in abrogating the smoking-induced increase in ADMA (P = NS). 4. Discussion To our knowledge, this is the first study to show that, in chronic smokers, smoking a single cigarette transiently increases arterial stiffness. A novel finding of our study is that the deterioration of arterial elastic properties that follows acute smoking is dependent primarily on COX-1 activity and secondarily on COX-2 activity. This is because aspirin, a non-selective COX-1 and COX-2 inhibitor, fully abrogated the smoking-induced increase of PWV, whereas celecoxib, a selective COX-2 inhibitor, only partially prevented the acute smoking-related adverse effect on aortic PWV. 4.1. Clinical implications It is established that chronic cigarette smoking is associated with increased arterial stiffness and wave reflections, as well as with endothelial dysfunction [2,8,9,12] which are predictors of cardiovascular risk [3, 4,7,8]. However, so far few studies only have investigated whether chronic smoking “exhausts” the capacity of arterial function to further deteriorate when a harmful intervention (such as smoking one cigarette) is added [10,11,20,21]. Our findings support that individuals who smoke on a chronic basis do not develop tolerance, so that every single cigarette transiently stiffens the aorta even further, partly in a COX (mainly COX-1)-dependent way. These changes may be clinically
meaningful, as there is evidence that even acute stimuli that impair vascular function [22] carry an increased cardiovascular risk [23]. Accordingly, our data indicate that beyond the chronic arterial damage that accompanies chronic smoking, exacerbations of vascular dysfunction after smoking a single cigarette may contribute to the increased cardiovascular risk of chronic smokers. Furthermore, our study etiologically establishes that, in chronic smokers, COX activation is an important mediator of additional acute and transient arterial damage induced by acute smoking. Our findings imply that COX-1 is perhaps more important than COX-2 in mediating these acute and transient effects. This is because high-dose aspirin, which is a potent COX-1 and COX-2 inhibitor, abrogated the detrimental effect of acute smoking on aortic PWV, whereas a selective COX-2 inhibitor such as celecoxib only partially prevented this acute smokingrelated adverse effect. Given that the low-dose aspirin used in clinical practice for primary and secondary cardiovascular protection is an efficient inhibitor of COX-1, we assume that even a lower aspirin dose (than the 1000 mg we used) likely prevents the acute additional increase of PWV induced by acute smoking in chronic smokers. However, this cannot be substantiated without a proper dose–response analysis. This vasculoprotective effect of aspirin is possibly clinically relevant, given that preservation of vascular function is associated with normal coronary vasoreactivity [24] and a lower incidence of cardiovascular outcomes [25]. As far as pure COX-2 inhibition is concerned, there is evidence that questions the cardiovascular safety of COX-2 inhibitors [26]. The net effect of a drug on the cardiovascular system is dependent on the balance between favorable and unfavorable actions. The reversal of smoking toxicity with celecoxib, which was only partial, supports the evidence that this class of drugs has a worse safety profile compared to other agents such as aspirin. This effect may not be specific to smokers though. On the other hand, even this limited beneficial action is in support of the data showing that celecoxib has a less unfavorable cardiovascular risk profile compared to other COX-2 inhibitors and may not be related to increased cardiovascular outcomes [27–29]. Nevertheless, no clear vasculoprotective effect of celecoxib can be documented in the settings similar to that of our study. 4.2. Mechanisms Our findings are in line with previous studies showing that certain aspects of arterial function are impaired during a 30-min period after cigarette smoking [11,13,21]. ADMA levels increased after smoking, indicating a lower endothelial NO bioavailability. This finding confirms previous experimental studies showing that smoke products increase acutely the endothelial production of ADMA, an endogenous inhibitor of endothelial NO synthase, in chronic smokers [30]. As NO is also a determinant of arterial elastic properties [31], impaired NO availability may have contributed to the deterioration of PWV after smoking. However, this association seems to be of partial importance in our study, as both aspirin and celecoxib abrogated the rise of ADMA after smoking, yet only aspirin fully prevented the smoking-related deterioration of PWV. Chronic smoking is a risk factor [1,2] that is associated with an increase of COX activity [16]. Both COXs are upregulated in smokers and may contribute, through an excess production of COX-1 and COX-2derived vasoconstrictor and platelet pro-aggregatory prostanoids, to the smoking-related cardiovascular injury [16–18,32]. Our study provides evidence for a further rapid increase of both COX-1 and COX-2 activities after smoking a single cigarette, given that celecoxib partly and aspirin fully protected against the deleterious effects of smoking on aortic stiffness. Existing evidence suggests that smoking increases the levels of proinflammatory cytokines leading to impaired arterial function [33] in acute conditions, through a COX-2-dependent mechanism [34] and an altered balance of prostanoids [19,32,35–37]. Interestingly, smoking
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
297
Fig. 2. Box-and-whisker plots of levels of hsCRP, IL-6 and ADMA before and after smoking, according to treatment. The centreline of the box denotes the median value; the extremes of the box, the interquartile range; and the bars, the upper and lower limits of 95% of the data. The circles represent outlying data. Probability values by paired t-test of the changes (after minus before) between the inflammatory drug (aspirin/celecoxib) and placebo sessions.
can increase the arterial COX-2 expression even after a few minutes, and this effect is facilitated by the presence of pro-inflammatory cytokines [38]. It seems that CRP and IL6 do not play a major role in this setting, as we observed no change in their levels. Our findings suggest a role of acute smoking-induced induction of COX-2 for arterial stiffening, as celecoxib partially prevented the increase of PWV after smoking. On the other hand, cigarette smoking activates COX-1 in young smokers, while it is well known that COX-1 mediates the synthesis of the vasoconstrictor thromboxane A2 (TXA2) [18,39]. Selective COX-2 inhibition leaves the COX-1 induced formation of thromboxane unopposed, which may have adverse vascular effects. Furthermore, selective COX-2 inhibition is known to inhibit the formation of the vasodilator prostacyclin [40]. On the other hand, the formation of thromboxane is almost abolished with high-dose aspirin [41]. These data may explain our finding of a more favorable effect of aspirin on arterial stiffness compared with celecoxib. In conclusion, this is the first study to demonstrate that in chronic smokers, smoking a single cigarette impairs arterial stiffness and produces a systemic endothelial activation. In addition, the deterioration of aortic stiffness that follows acute smoking depends primarily on COX-1 activity and partly on COX-2 activity given that aspirin, a nonselective COX-1 and COX-2 inhibitor, fully abrogated the acute
smoking-related adverse effect on aortic PWV, whereas celecoxib, a selective COX-2 inhibitor, only partially prevented this effect. These findings may have clinical implications given the prognostic importance of arterial stiffness for cardiovascular performance and risk. Conflict of interest None. Source of funding None. Disclosures None. References [1] D.G. Yanbaeva, M.A. Dentener, E.C. Creutzberg, G. Wesseling, E.F. Wouters, Systemic effects of smoking, Chest 131 (2007) 1557–1566. [2] J.A. Ambrose, R.S. Barua, The pathophysiology of cigarette smoking and cardiovascular disease: an update, J. Am. Coll. Cardiol. 43 (2004) 1731–1737.
298
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
[3] C. Vlachopoulos, K. Aznaouridis, C. Stefanadis, Clinical appraisal of arterial stiffness: the Argonauts in front of the Golden Fleece, Heart 92 (2006) 1544–1550. [4] A.A. Quyyumi, Prognostic value of endothelial function, Am. J. Cardiol. 91 (2003) 19H–24H. [5] S. Laurent, J. Cockcroft, L. Van Bortel, European Network for Non-invasive Investigation of Large Arteries, et al., Expert consensus document on arterial stiffness: methodological issues and clinical applications, Eur. Heart J. 27 (2006) 2588–2605. [6] E. Agabiti-Rosei, G. Mancia, M.F. O'Rourke, et al., Central blood pressure measurements and antihypertensive therapy: a consensus document, Hypertension 50 (2007) 154–160. [7] C. Vlachopoulos, K. Aznaouridis, C. Stefanadis, Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and metaanalysis, J. Am. Coll. Cardiol. 55 (2010) 1318–1327. [8] C. Vlachopoulos, K. Aznaouridis, M.F. O'Rourke, M.E. Safar, K. Baou, C. Stefanadis, Prediction of cardiovascular events and all-cause mortality with central haemodynamics: a systematic review and meta-analysis, Eur. Heart J. 31 (2010) 1865–1871. [9] K. Aznaouridis, C. Vlachopoulos, A. Protogerou, C. Stefanadis, Ambulatory systolic– diastolic pressure regression index as a predictor of clinical events. A metaanalysis of longitudinal studies, Stroke 43 (2012) 733–739. [10] C. Stefanadis, E. Tsiamis, C. Vlachopoulos, et al., Unfavorable effect of smoking on the elastic properties of the human aorta, Circulation 95 (1997) 31–38. [11] C. Vlachopoulos, F. Kosmopoulou, D. Panagiotakos, et al., Smoking and caffeine have a synergistic detrimental effect on aortic stiffness and wave reflections, J. Am. Coll. Cardiol. 44 (2004) 1911–1917. [12] D.S. Celermajer, K.E. Sorensen, D. Georgakopoulos, et al., Cigarette smoking is associated with dose-related and potentially reversible impairment of endothelium dependent dilatation in healthy young adults, Circulation 88 (1993) 2149–2155. [13] J. Lekakis, C. Papamichael, C. Vemmos, K. Stamatelopoulos, A. Voutsas, S. Stamatelopoulos, Effect of acute cigarette smoking on endothelium-dependent arterial dilatation in normal subjects, Am. J. Cardiol. 81 (1998) 1225–1228. [14] O.T. Raitakari, M.R. Adams, R.J. McCredie, K.A. Griffiths, D.S. Celermajer, Arterial endothelial dysfunction related to passive smoking is potentially reversible in healthy young adults, Ann. Intern. Med. 130 (1999) 578–581. [15] N.A. Jatoi, P. Jerrard-Dunne, J. Feely, A. Mahmud, Impact of smoking and smoking cessation on arterial stiffness and aortic wave reflection in hypertension, Hypertension 49 (2007) 981–985. [16] B.F. McAdam, D. Byrne, J.D. Morrow, J.A. Oates, Contribution of cyclooxygenase-2 to elevated biosynthesis of thromboxane A2 and prostacyclin in cigarette smokers, Circulation 112 (2005) 1024–1029. [17] M. Hermann, H. Krum, F. Ruschitzka, To the heart of the matter: coxibs, smoking, and cardiovascular risk, Circulation 112 (2005) 941–945. [18] W.M. Loke, K.L.M. Sing, C.Y.J. Lee, et al., Cyclooxygenase-1 mediated platelet reactivity in young male smokers, Clin. Appl. Thromb. Hemost. 20 (2014) 371–377. [19] C. Stefanadis, C. Vlachopoulos, E. Tsiamis, et al., Unfavorable effects of passive smoking on aortic function in men, Ann. Intern. Med. 128 (1998) 426–434. [20] C. Papamichael, K. Karatzi, E. Karatzis, et al., Combined acute effects of red wine consumption and cigarette smoking on haemodynamics of young smokers, J. Hypertens. 24 (2006) 1287–1292. [21] K. Karatzi, C. Papamichael, E. Karatzis, et al., Acute smoke-induced endothelial dysfunction is more prolonged in smokers than in non-smokers, Int. J. Cardiol. 120 (2007) 404–406. [22] C. Vlachopoulos, I. Dima, K. Aznaouridis, et al., Acute systemic inflammation increases arterial stiffness and decreases wave reflections in healthy individuals, Circulation 112 (2005) 2193–2200. [23] L. Smeeth, S.L. Thomas, A.J. Hall, R. Hubbard, P. Farrington, P. Vallance, Risk of myocardial infarction and stroke after acute infection or vaccination, N. Engl. J. Med. 351 (2004) 2611–2618.
[24] T.J. Anderson, A. Uehata, M.D. Gerhard, et al., Close relation of endothelial function in the human coronary and peripheral circulations, J. Am. Coll. Cardiol. 26 (1995) 1235–1241. [25] B. Williams, P.S. Lacy, S.M. Thom, CAFE Investigators, Anglo-Scandinavian Cardiac Outcomes Trial Investigators, CAFE Steering Committee and Writing Committee, et al., Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study, Circulation 113 (2006) 1213–1225. [26] J.S. Bennett, A. Daugherty, D. Herrington, P. Greenland, H. Roberts, K.A. Taubert, The use of nonsteroidal anti-inflammatory drugs (NSAIDs). A science advisory from the American Heart Association, Circulation 111 (2005) 1713–1716. [27] M. Hermann, F. Ruschitzka, Cardiovascular risk of cyclooxygenase-2 inhibitors and traditional non-steroidal anti-inflammatory drugs, Ann. Med. 39 (2007) 18–27. [28] S.E. Kimmel, J.A. Berlin, M. Reilly, et al., Patients exposed to rofecoxib and celecoxib have different odds of nonfatal myocardial infarction, Ann. Intern. Med. 142 (2005) 157–164. [29] P. McGettigan, D. Henry, Cardiovascular risk and inhibition of cyclooxygenase. A systematic review of the observational studies of selective and nonselective inhibitors of cyclooxygenase 2, JAMA 296 (2006) 1633–1644. [30] W.Z. Zhang, K. Venardos, J. Chin-Dusting, D.M. Kaye, Adverse effects of cigarette smoke on NO bioavailability. Role of arginine metabolism and oxidative stress, Hypertension 48 (2006) 278–285. [31] I.B. Wilkinson, S.S. Franklin, J.R. Cockcroft, Nitric oxide and the regulation of large artery stiffness: from physiology to pharmacology, Hypertension 44 (2004) 112–116. [32] I. Ikonomidis, J. Lekakis, G. Vamvakou, F. Andreotti, P. Nihoyannopoulos, Cigarette smoking is associated with increased circulating proinflammatory and procoagulant markers in patients with chronic coronary artery disease: effects of aspirin treatment, Am. Heart J. 149 (2005) 832–839. [33] K. Bhagat, P. Vallance, Inflammatory cytokines impair endothelium-dependent dilatation in human veins in vivo, Circulation 96 (1997) 3042–3047. [34] R.K. Kharbanda, B. Walton, M. Allen, et al., Prevention of inflammation-induced endothelial dysfunction: a novel vasculo-protective action of aspirin, Circulation 105 (2002) 2600–2604. [35] W. Qi, J.X. Wei, I. Dorairaj, R.P. Mahajan, V.G. Wilson, Evidence that a prostanoid produced by cyclo-oxygenase-2 enhances contractile responses of the porcine isolated coronary artery following exposure to lipopolysaccharide, Br. J. Anaesth. 98 (2007) 323–330. [36] J. Blanco-Rivero, V. Cachofeiro, V. Lahera, et al., Participation of prostacyclin in endothelial dysfunction induced by aldosterone in normotensive and hypertensive rats, Hypertension 46 (2005) 107–112. [37] J. Blanco-Rivero, I. Márquez-Rodas, F.E. Xavier, et al., Long-term fenofibrate treatment impairs endothelium-dependent dilation to acetylcholine by altering the cyclooxygenase pathway, Cardiovasc. Res. 75 (2007) 398–407. [38] S.S. Barbieri, B.B. Weksler, Tobacco smoke cooperates with interleukin-1β to alter βcatenin trafficking in vascular endothelium resulting in increased permeability and induction of cyclooxygenase-2 expression in vitro and in vivo, FASEB J. 21 (2007) 1831–1843. [39] J. Toivanen, O. Ylikorkala, L. Viinikka, Effects of smoking and nicotine on human prostacyclin and thromboxane production in vivo and in vitro, Toxicol. Appl. Pharmacol. 82 (1986) 301–306. [40] S. Roy, Effects of smoking on prostacyclin formation and platelet aggregation in users of oral contraceptives, Am. J. Obstet. Gynecol. 180 (1999) S364–S368. [41] P. Borgdorff, M.L. Handoko, Y.Y. Wong, G.T. Tangelder, COX-2 inhibition by use of rofecoxib or high dose aspirin enhances ADP-induced platelet aggregation in fresh blood, Open Cardiovasc. Med. J. 4 (2010) 198–205.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Arterial stiffening and systemic endothelial activation induced by smoking☆ The role of COX-1 and COX-2 Charalambos Vlachopoulos ⁎, Konstantinos Aznaouridis, Athanassios Bratsas, Nikolaos Ioakeimidis, Ioanna Dima, Panagiotis Xaplanteris, Christodoulos Stefanadis, Dimitris Tousoulis Peripheral Vessels and Hypertension Units, 1st Department of Cardiology, Athens Medical School, Hippokration Hospital, Athens, Greece
a r t i c l e
i n f o
Article history: Received 2 March 2015 Accepted 3 April 2015 Available online 6 April 2015 Keywords: Smoking Arterial elasticity Arterial stiffness Cyclooxygenase
a b s t r a c t Background: Arterial stiffness is an established predictor of cardiovascular risk. We explored the effects of acute smoking on arterial stiffness, systemic inflammation and endothelial activation in chronic smokers and the contribution of cyclooxygenases-1 and 2 (COX-1 and COX-2). Methods and results: In a randomized, double-blind, cross-over study, we investigated in 28 young smokers the vascular and systemic effects of smoking one cigarette, 3 h after receiving 1000 mg of aspirin (a non-selective COX-1 and COX-2 inhibitor) or placebo (aspirin substudy), or 200 mg of celecoxib (a selective COX-2 inhibitor) or placebo (celecoxib substudy). Smoking increased carotid–femoral pulse wave velocity (PWV, a marker of aortic stiffness), indicating an adverse effect on arterial elastic properties. Similarly, circulating levels of asymmetric dimethylarginine (ADMA) were increased after smoking. Aspirin fully prevented the smoking-induced increase of PWV after smoking. In contrast, celecoxib only partially prevented the smoking-induced increase of PWV. Both aspirin and celecoxib prevented to a similar extent the increase of ADMA levels after smoking. Conclusions: Smoking one cigarette is associated with a deterioration of arterial stiffness and with systemic endothelial activation in chronic smokers. Both COX-1 and COX-2, but primarily COX-1, mediate these unfavorable effects of smoking. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Smoking affects adversely several cardiovascular homeostatic pathways and comprises the most important modifiable risk factor for the development and progression of atherosclerotic disease [1,2]. The detrimental impact of cigarette smoking on cardiovascular physiology may be mediated, at least partly, by unfavorable effects on arterial stiffness and endothelial homeostasis, which are important markers of cardiovascular disease and independent predictors of the corresponding risk [3–9]. We and other investigators have previously shown that smoking deteriorates arterial stiffness and endothelial function in several settings [10–15]. Activation of platelets, increased oxidative stress and reduced nitric oxide (NO) availability are considered common final mechanisms that mediate the unfavorable cardiovascular effects of smoking [1,2]. However, the exact antecedent biochemical pathways that are modified ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Profiti Elia 24, Athens 14575, Greece. E-mail address: [email protected] (C. Vlachopoulos).
http://dx.doi.org/10.1016/j.ijcard.2015.04.029 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.
adversely by smoking have not been fully elucidated. Interestingly, recent evidence suggests that cyclooxygenases (COXs) are upregulated in smokers and may contribute, through an excess production of COX1 and COX-2-derived prostanoids, to the smoking-related cardiovascular injury [16–18]. It is established that arterial performance is impaired in chronic smokers [1,2,15]. However, although it has been previously shown that active and passive smoking may impair certain indices of arterial function both in chronic smokers and in smoking-naïve subjects [10,11,13,14,19], so far the effects of smoking a single cigarette on arterial function and the potential mediators for those effects have not been thoroughly investigated in chronic smokers. Accordingly, in the present study we investigated, through a cause-and-effect testing, whether acute smoking alters arterial stiffness and systemic inflammatory and endothelial activation in subjects who smoke on a chronic basis. Furthermore, in order to explore the potential role of COX-1 and COX-2, we evaluated the effect of high-dose aspirin (a nonselective COX-1 and COX-2 inhibitor) and celecoxib (a selective COX-2 inhibitor) on the smoking-induced changes of arterial stiffness and circulating levels of markers of inflammation and endothelial activation.
294
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
2. Methods 2.1. Study population We studied 28 apparently healthy current smokers (mean age 32.9 years, range 26 to 45, 15 men) who were free of hypertension, diabetes, dyslipidemia, or family history of premature vascular disease. All subjects were clinically well and were not taking any regular cardiovascular medications or antioxidant vitamin supplementation. They had not had any bacterial or other inflammatory disorder and they did not report use of anti-inflammatory or steroid substances during the past 2 months. No female participant was on oral contraceptives or estrogen replacement therapy. The study protocol was approved by our Institutional Research Ethics Committee and all subjects gave written informed consent.
artery) − (distance from carotid artery to the suprasternal notch). Three readings were performed, and the mean value of PWV was calculated. 2.5. Measurement of markers of endothelial activation and inflammation Immediately after acquisition of venous blood, plasma and serum were separated by centrifugation (3000 ×g at 4 °C for 15 min), then placed in aliquots and stored at − 70 °C. High-sensitivity C-reactive protein (hsCRP) was measured by immunophelometry (Dade Behring, Marburg, Germany). High sensitivity interleukin-6 (IL-6) was measured using ELISA (R&D Systems, Minneapolis, MN, USA). Asymmetric dimethylarginine (ADMA) was also measured using ELISA (Immundiagnostik AG, Bensheim, Germany). 2.6. Statistical analysis
2.2. Study design The study consisted of two separate substudies (aspirin and celecoxib substudies) and was carried out using a randomized, double-blind, placebo-controlled, cross-over design. Each substudy consisted of two treatment sessions, one with an anti-inflammatory agent (aspirin or celecoxib) and one with placebo. Subjects had fasted for at least 6 h and had abstained from caffeine, ethanol and flavonoid-containing beverage intake for at least 12 h before each session. They had been instructed not to smoke for at least 4 h before the vascular studies. All vascular studies were performed in the afternoon between 3 and 7 p.m. on two separate days, in a quiet, temperature-controlled room at 23 °C. Baseline studies for evaluation of aortic stiffness were performed after a 20-min rest period in the supine position. Then, in the aspirin substudy, the subjects were randomized to taking aspirin 1000 mg (Aspirin, Bayer, 2 tablets of 500 mg) or placebo. In the celecoxib substudy, the subjects were randomized to taking celecoxib 200 mg (Celebrex, Pfizer, 2 capsules of 100 mg) or placebo. Inert tablets were given as placebo. After repeating the vascular studies at 3 h after the administration of anti-inflammatory agent/placebo (pre-smoking time point), the subjects smoked one standard cigarette (1 mg nicotine, 14 mg tar) in a sitting position over 5 min. The vascular measurements were repeated immediately after the completion of smoking (time Sm0) and 20 min thereafter (Sm20). Venous blood for assessment of markers of inflammation and endothelial activation was drawn into vacutainer tubes after baseline studies and after the studies at Sm20. After 2–10 days, all participants of both substudies returned and all studies were done after taking the opposite treatment (anti-inflammatory agent/placebo), according to the same schedule. Both subjects and investigators were blinded to treatment allocation throughout the study. 2.3. Measurement of blood pressure Supine blood pressure (BP) was measured with mercury sphygmomanometry after a 5-min rest, just before evaluation of arterial stiffness. Three readings 1 min apart were performed, and the mean value of systolic and diastolic BP was calculated. 2.4. Evaluation of aortic stiffness The pulse travels at a higher velocity in a stiff aorta and vice versa. Carotid–femoral pulse wave velocity (PWV), an established index of aortic stiffness [3,5,8,17,18], was calculated from measurements of pulse transit time and the distance traveled between two recording sites (pulse wave velocity = distance [m] / transit time [s]) using a validated noninvasive device (Complior®, Artech Medical, Pantin, France), as previously described [5,17,18]. Two different pulse waves were obtained simultaneously at two sites (at the base of the neck for the common carotid and over the right femoral artery) with two transducers. The distance was defined as: (distance from the suprasternal notch to femoral
Sample size calculation was based on the hypothesis that smoking would be associated with changes in arterial and biochemical indices of at least 0.9 of 1 SD. Therefore, we estimated that 12 subjects would provide 80% power at the 5% level of significance to detect such a difference between anti-inflammatory agents (aspirin or celecoxib) and placebo sessions in a cross-over design study. Continuous variables are expressed as mean value ± SD, whereas categorical variables as absolute frequencies. All continuous variables were tested for normal distribution by using the Kolmogorov–Smirnov criterion. For skewed variables, data are expressed as median value (25th–75th percentile) and logarithmic transformation was performed prior to analysis. The within-sessions and between-sessions (anti-inflammatory agent/placebo) effects during the 3-h pre-smoking period were evaluated by paired t-test. The effects of smoking within the sessions of anti-inflammatory agents (aspirin or celecoxib) and placebo were evaluated with analysis of variance (ANOVA) for repeated measures (3 timepoints: 3 h/pre-smoking, Sm0 and Sm20). When a significant effect occurred, the location of pairwise differences was determined by using paired t-test followed by the Bonferroni correction. The betweensubstudies effect of active treatment (celecoxib or aspirin) on the levels of endothelial/inflammatory markers was evaluated by unpaired t-test. Exact P values b0.05 were considered statistically significant. Data analysis was performed with SPSS software, version 13.0 (Chicago, IL). 3. Results Clinical characteristics are shown in Table 1. There were no significant differences in baseline hemodynamic, vascular and biochemical characteristics between the study substudies/sessions. The smoking effect on hemodynamic and arterial function indices at Sm0 and Sm20 in each session (anti-inflammatory agent or placebo) of the substudies (aspirin or celecoxib) is reported as the respective post-smoking mean value (Sm0 or Sm20) minus pre-smoking mean value. The net effect of each anti-inflammatory agent (aspirin or celecoxib) in the 3-h pretreatment period of each substudy was calculated as anti-inflammatory agent (aspirin or celecoxib) effect (3 h/pre-smoking minus baseline mean value in the anti-inflammatory agent session) minus placebo effect (3 h/pre-smoking minus baseline mean value in the placebo session). Changes of biochemical variables were calculated as mean value at the end of study minus mean value at baseline. In case of skewed variables, geometric means (anti-logs of mean log-values) were used for calculating responses. 3.1. Systemic hemodynamic effects 3.1.1. Aspirin substudy 3.1.1.1. Pre-smoking treatment period. Compared to placebo, aspirin decreased significantly systolic BP (by 3.5 mm Hg, P = 0.039) and pulse
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
295
Table 1 Baseline characteristics of the study population according to substudies/sessions.
Age, year Males/females Body-mass index, kg/m2 Waist/hip Glucose, mmol/L Total cholesterol, mmol/L Triglycerides, mmol/L HDL-cholesterol, mmol/L
Systolic BP, mm Hg Diastolic BP, mm Hg Pulse pressure, mm Hg Heart rate, bpm Pulse wave velocity, m/s hsCRP, mg/L IL-6, pg/mL ADMA, μmol/L
Aspirin substudy (N = 12)
Celecoxib substudy (N = 16)
P-value
36.3 ± 4.6 7/5 22.9 ± 3.5 0.86 ± 0.09 4.72 ± 0.46 4.77 ± 0.73 0.88 ± 0.53 1.71 ± 1.24
33.2 ± 4.1 8/8 22.2 ± 3.2 0.85 ± 0.08 4.64 ± 0.63 4.80 ± 0.89 0.74 ± 0.38 1.48 ± 0.34
0.07 0.66 0.60 0.84 0.71 0.93 0.40 0.49
Aspirin
Placebo
P value
Celecoxib
Placebo
P value
116 ± 11 71 ± 10 44 ± 3 66 ± 9 6.69 ± 0.84 0.53 (0.18–1.64) 0.90 (0.48–1.46) 0.53 (0.41–1.18)
118 ± 9 74 ± 7 45 ± 7 67 ± 7 6.67 ± 0.97 0.56 (0.18–1.13) 0.98 (0.51–1.44) 0.56 (0.15–0.90)
0.50 0.47 0.91 0.73 0.96 0.94 0.96 0.21
108 ± 12 68 ± 8 41 ± 7 67 ± 9 5.11 ± 0.77 0.58 (0.36–0.88) 1.12 (0.85–1.42) 0.58 (0.47–0.65)
109 ± 13 67 ± 10 42 ± 6 67 ± 7 5.07 ± 0.81 0.61 (0.34–1.32) 0.96 (0.76–1.35) 0.57 (0.51–0.70)
0.87 0.77 0.46 0.99 0.89 0.34 0.37 0.32
Categorical variables are presented as absolute frequencies, while continuous variables as mean ± SD or median (interquartile range). Probability values derived from Student's t-test for unpaired (between substudies) or paired measures (within substudies).
pressure (by 4.6 mm Hg, P = 0.013) after 3 h. We observed no significant changes of diastolic pressure or heart rate (all P = NS).
Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 11.8 bpm, P b 0.001) and Sm20 (by 6.4 bpm, P b 0.001). 3.2. Effects on arterial stiffness
3.1.1.2. Smoking. Within the placebo session, smoking increased systolic BP (P = 0.013 by ANOVA) and diastolic BP (P = 0.002 by ANOVA). Significant effects were observed only at Sm0 (systolic BP: increases of 7.3 mm Hg at Sm0, P = 0.01 and 4.7 mm Hg at Sm20, P = NS; diastolic BP increases of 7.5 mm Hg at Sm0, P = 0.005 and 3.4 mm Hg at Sm20, P = NS). Pulse pressure did not change with smoking (P = NS by ANOVA, decrease of 1.8 mm Hg at Sm0 and increase of 1.3 mm Hg at Sm20). Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 11.8 bpm, P = 0.001) and Sm20 (by 5.5 bpm, P = 0.014). Aspirin did not influence the smoking-induced changes of any of the above variables. Within the aspirin session, smoking increased systolic BP (P = 0.010 by ANOVA, increases of 6.7 mm Hg at Sm0, P = 0.01 and 1.4 mm Hg at Sm20, P = NS) and diastolic BP (P = 0.019 by ANOVA, increases of 4.0 mm Hg at Sm0, P = 0.018 and 0.9 mm Hg at Sm20, P = NS). Pulse pressure did not change (P = NS by ANOVA). Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 12.4 bpm, P b 0.001) and Sm20 (by 4.5 bpm, P = 0.011).
3.2.1. Aspirin substudy 3.2.1.1. Pre-smoking treatment period. PWV did not change significantly after taking aspirin (response of −0.07 m/s, P = 0.73, Fig. 1). 3.2.1.2. Smoking. In the placebo session, we observed a significant rise of PWV with smoking (P b 0.001 by ANOVA, increases of 0.65 m/s at Sm0, P b 0.001 and by 0.47 m/s at Sm20, P = 0.06, Fig. 1). Aspirin fully prevented the smoking-induced increase of PWV (P = 0.24 by ANOVA within the aspirin session, increases of 2.4 m/s at Sm0 and 1.4 m/s at Sm20, both P = NS, Fig. 1). 3.2.2. Celecoxib substudy 3.2.2.1. Pre-smoking treatment period. PWV was not significantly decreased 3 h after taking celecoxib (response of − 0.14 m/s, P = 0.27, Fig. 1).
3.1.2. Celecoxib substudy 3.1.2.1. Pre-smoking treatment period. Compared to placebo, celecoxib decreased significantly systolic BP (by 5.1 mm Hg, P = 0.014) and diastolic BP (by 3.2 mm Hg, P = 0.041) after 3 h. We observed no significant changes of pulse pressure or heart rate (all P = NS). 3.1.2.2. Smoking. Within the placebo session, smoking increased systolic BP (P = 0.001 by ANOVA) and diastolic BP (P = 0.002 by ANOVA). Significant effects were observed only at Sm0 (systolic BP: increases of 9.4 mm Hg at Sm0, P b 0.001 and 1.9 mm Hg at Sm20, P = NS; diastolic BP increases of 6.8 mm Hg at Sm0, P = 0.01 and 2.4 mm Hg at Sm20, P = NS). Pulse pressure had a marginal increase after smoking (P = 0.06 by ANOVA, increase of 2.6 mm Hg at Sm0, P = 0.09). Heart rate increased (P b 0.001 by ANOVA) both at Sm0 (by 11.0 bpm, P b 0.001) and Sm20 (by 7.3 bpm, P = 0.003). Celecoxib did not influence the smoking-induced changes of any of the above variables. Within the celecoxib session, smoking increased systolic BP (P b 0.001 by ANOVA, increases of 6.6 mm Hg at Sm0, P = 0.002 and 3.6 mm Hg at Sm20, P = 0.04) and diastolic BP (P = 0.001 by ANOVA, increases of 4.8 mm Hg at Sm0, P = 0.001 and 2.5 mm Hg at Sm20, P = NS). Pulse pressure did not change (P = NS by ANOVA).
3.2.2.2. Smoking. In the placebo session, we observed a significant rise of PWV with smoking (P b 0.001 by ANOVA, increases of 0.63 m/s at Sm0, P b 0.001 and 0.44 m/s at Sm20, P = 0.008, Fig. 1). In the celecoxib session, PWV increased significantly with smoking (P = 0.002 by ANOVA). However, this increase was observed only at Sm0 (by 0.38 m/s at Sm0, P = 0.004 and by 0.16 m/s at Sm20, P = NS, Fig. 1). Accordingly, celecoxib did not fully prevent the smokinginduced increase of PWV but accelerated the restoration of PWV at Sm20. 3.3. Effects on systemic inflammation and endothelial function Smoking or anti-inflammatory drug intake did not influence the level of hsCRP and IL-6 in the placebo sessions or after aspirin or celecoxib (Fig. 2). Circulating levels of ADMA increased after smoking in the placebo sessions of both the aspirin and celecoxib substudies (P b 0.05 in both substudies). Aspirin prevented the smoking-induced increase of ADMA (change of − 0.17 vs 0.47 μmol/L with aspirin and placebo respectively, P = 0.035) (Fig. 2). Similarly, celecoxib abrogated the smoking-induced increase of circulating levels of ADMA (change of
296
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
Fig. 1. Changes in carotid–femoral pulse wave velocity in the celecoxib and placebo sessions (celecoxib substudy, bottom panel) and in the aspirin and placebo sessions (aspirin substudy, top panel) before and 3 h after treatment (pretreatment substudy, open area) and immediately after and 20 min after smoking (Sm0 and Sm20 in the smoking substudy, shaded area). P value in the pretreatment study (open area) indicates comparison between anti-inflammatory drug (aspirin/celecoxib) and placebo. P values in the smoking period indicate repeated measures comparison by ANOVA within each session (anti-inflammatory drug or placebo). *P b 0.05 vs. pre-smoking (3 h) within the session (paired t test followed by Bonferroni correction). Triangles indicate mean changes from baseline; bars, their 95% CIs.
− 0.03 vs 0.03 μmol/L with celecoxib and placebo respectively, P = 0.033) (Fig. 2). Comparison between the 2 substudies showed that aspirin and celecoxib had a similar effect in abrogating the smoking-induced increase in ADMA (P = NS). 4. Discussion To our knowledge, this is the first study to show that, in chronic smokers, smoking a single cigarette transiently increases arterial stiffness. A novel finding of our study is that the deterioration of arterial elastic properties that follows acute smoking is dependent primarily on COX-1 activity and secondarily on COX-2 activity. This is because aspirin, a non-selective COX-1 and COX-2 inhibitor, fully abrogated the smoking-induced increase of PWV, whereas celecoxib, a selective COX-2 inhibitor, only partially prevented the acute smoking-related adverse effect on aortic PWV. 4.1. Clinical implications It is established that chronic cigarette smoking is associated with increased arterial stiffness and wave reflections, as well as with endothelial dysfunction [2,8,9,12] which are predictors of cardiovascular risk [3, 4,7,8]. However, so far few studies only have investigated whether chronic smoking “exhausts” the capacity of arterial function to further deteriorate when a harmful intervention (such as smoking one cigarette) is added [10,11,20,21]. Our findings support that individuals who smoke on a chronic basis do not develop tolerance, so that every single cigarette transiently stiffens the aorta even further, partly in a COX (mainly COX-1)-dependent way. These changes may be clinically
meaningful, as there is evidence that even acute stimuli that impair vascular function [22] carry an increased cardiovascular risk [23]. Accordingly, our data indicate that beyond the chronic arterial damage that accompanies chronic smoking, exacerbations of vascular dysfunction after smoking a single cigarette may contribute to the increased cardiovascular risk of chronic smokers. Furthermore, our study etiologically establishes that, in chronic smokers, COX activation is an important mediator of additional acute and transient arterial damage induced by acute smoking. Our findings imply that COX-1 is perhaps more important than COX-2 in mediating these acute and transient effects. This is because high-dose aspirin, which is a potent COX-1 and COX-2 inhibitor, abrogated the detrimental effect of acute smoking on aortic PWV, whereas a selective COX-2 inhibitor such as celecoxib only partially prevented this acute smokingrelated adverse effect. Given that the low-dose aspirin used in clinical practice for primary and secondary cardiovascular protection is an efficient inhibitor of COX-1, we assume that even a lower aspirin dose (than the 1000 mg we used) likely prevents the acute additional increase of PWV induced by acute smoking in chronic smokers. However, this cannot be substantiated without a proper dose–response analysis. This vasculoprotective effect of aspirin is possibly clinically relevant, given that preservation of vascular function is associated with normal coronary vasoreactivity [24] and a lower incidence of cardiovascular outcomes [25]. As far as pure COX-2 inhibition is concerned, there is evidence that questions the cardiovascular safety of COX-2 inhibitors [26]. The net effect of a drug on the cardiovascular system is dependent on the balance between favorable and unfavorable actions. The reversal of smoking toxicity with celecoxib, which was only partial, supports the evidence that this class of drugs has a worse safety profile compared to other agents such as aspirin. This effect may not be specific to smokers though. On the other hand, even this limited beneficial action is in support of the data showing that celecoxib has a less unfavorable cardiovascular risk profile compared to other COX-2 inhibitors and may not be related to increased cardiovascular outcomes [27–29]. Nevertheless, no clear vasculoprotective effect of celecoxib can be documented in the settings similar to that of our study. 4.2. Mechanisms Our findings are in line with previous studies showing that certain aspects of arterial function are impaired during a 30-min period after cigarette smoking [11,13,21]. ADMA levels increased after smoking, indicating a lower endothelial NO bioavailability. This finding confirms previous experimental studies showing that smoke products increase acutely the endothelial production of ADMA, an endogenous inhibitor of endothelial NO synthase, in chronic smokers [30]. As NO is also a determinant of arterial elastic properties [31], impaired NO availability may have contributed to the deterioration of PWV after smoking. However, this association seems to be of partial importance in our study, as both aspirin and celecoxib abrogated the rise of ADMA after smoking, yet only aspirin fully prevented the smoking-related deterioration of PWV. Chronic smoking is a risk factor [1,2] that is associated with an increase of COX activity [16]. Both COXs are upregulated in smokers and may contribute, through an excess production of COX-1 and COX-2derived vasoconstrictor and platelet pro-aggregatory prostanoids, to the smoking-related cardiovascular injury [16–18,32]. Our study provides evidence for a further rapid increase of both COX-1 and COX-2 activities after smoking a single cigarette, given that celecoxib partly and aspirin fully protected against the deleterious effects of smoking on aortic stiffness. Existing evidence suggests that smoking increases the levels of proinflammatory cytokines leading to impaired arterial function [33] in acute conditions, through a COX-2-dependent mechanism [34] and an altered balance of prostanoids [19,32,35–37]. Interestingly, smoking
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
297
Fig. 2. Box-and-whisker plots of levels of hsCRP, IL-6 and ADMA before and after smoking, according to treatment. The centreline of the box denotes the median value; the extremes of the box, the interquartile range; and the bars, the upper and lower limits of 95% of the data. The circles represent outlying data. Probability values by paired t-test of the changes (after minus before) between the inflammatory drug (aspirin/celecoxib) and placebo sessions.
can increase the arterial COX-2 expression even after a few minutes, and this effect is facilitated by the presence of pro-inflammatory cytokines [38]. It seems that CRP and IL6 do not play a major role in this setting, as we observed no change in their levels. Our findings suggest a role of acute smoking-induced induction of COX-2 for arterial stiffening, as celecoxib partially prevented the increase of PWV after smoking. On the other hand, cigarette smoking activates COX-1 in young smokers, while it is well known that COX-1 mediates the synthesis of the vasoconstrictor thromboxane A2 (TXA2) [18,39]. Selective COX-2 inhibition leaves the COX-1 induced formation of thromboxane unopposed, which may have adverse vascular effects. Furthermore, selective COX-2 inhibition is known to inhibit the formation of the vasodilator prostacyclin [40]. On the other hand, the formation of thromboxane is almost abolished with high-dose aspirin [41]. These data may explain our finding of a more favorable effect of aspirin on arterial stiffness compared with celecoxib. In conclusion, this is the first study to demonstrate that in chronic smokers, smoking a single cigarette impairs arterial stiffness and produces a systemic endothelial activation. In addition, the deterioration of aortic stiffness that follows acute smoking depends primarily on COX-1 activity and partly on COX-2 activity given that aspirin, a nonselective COX-1 and COX-2 inhibitor, fully abrogated the acute
smoking-related adverse effect on aortic PWV, whereas celecoxib, a selective COX-2 inhibitor, only partially prevented this effect. These findings may have clinical implications given the prognostic importance of arterial stiffness for cardiovascular performance and risk. Conflict of interest None. Source of funding None. Disclosures None. References [1] D.G. Yanbaeva, M.A. Dentener, E.C. Creutzberg, G. Wesseling, E.F. Wouters, Systemic effects of smoking, Chest 131 (2007) 1557–1566. [2] J.A. Ambrose, R.S. Barua, The pathophysiology of cigarette smoking and cardiovascular disease: an update, J. Am. Coll. Cardiol. 43 (2004) 1731–1737.
298
C. Vlachopoulos et al. / International Journal of Cardiology 189 (2015) 293–298
[3] C. Vlachopoulos, K. Aznaouridis, C. Stefanadis, Clinical appraisal of arterial stiffness: the Argonauts in front of the Golden Fleece, Heart 92 (2006) 1544–1550. [4] A.A. Quyyumi, Prognostic value of endothelial function, Am. J. Cardiol. 91 (2003) 19H–24H. [5] S. Laurent, J. Cockcroft, L. Van Bortel, European Network for Non-invasive Investigation of Large Arteries, et al., Expert consensus document on arterial stiffness: methodological issues and clinical applications, Eur. Heart J. 27 (2006) 2588–2605. [6] E. Agabiti-Rosei, G. Mancia, M.F. O'Rourke, et al., Central blood pressure measurements and antihypertensive therapy: a consensus document, Hypertension 50 (2007) 154–160. [7] C. Vlachopoulos, K. Aznaouridis, C. Stefanadis, Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and metaanalysis, J. Am. Coll. Cardiol. 55 (2010) 1318–1327. [8] C. Vlachopoulos, K. Aznaouridis, M.F. O'Rourke, M.E. Safar, K. Baou, C. Stefanadis, Prediction of cardiovascular events and all-cause mortality with central haemodynamics: a systematic review and meta-analysis, Eur. Heart J. 31 (2010) 1865–1871. [9] K. Aznaouridis, C. Vlachopoulos, A. Protogerou, C. Stefanadis, Ambulatory systolic– diastolic pressure regression index as a predictor of clinical events. A metaanalysis of longitudinal studies, Stroke 43 (2012) 733–739. [10] C. Stefanadis, E. Tsiamis, C. Vlachopoulos, et al., Unfavorable effect of smoking on the elastic properties of the human aorta, Circulation 95 (1997) 31–38. [11] C. Vlachopoulos, F. Kosmopoulou, D. Panagiotakos, et al., Smoking and caffeine have a synergistic detrimental effect on aortic stiffness and wave reflections, J. Am. Coll. Cardiol. 44 (2004) 1911–1917. [12] D.S. Celermajer, K.E. Sorensen, D. Georgakopoulos, et al., Cigarette smoking is associated with dose-related and potentially reversible impairment of endothelium dependent dilatation in healthy young adults, Circulation 88 (1993) 2149–2155. [13] J. Lekakis, C. Papamichael, C. Vemmos, K. Stamatelopoulos, A. Voutsas, S. Stamatelopoulos, Effect of acute cigarette smoking on endothelium-dependent arterial dilatation in normal subjects, Am. J. Cardiol. 81 (1998) 1225–1228. [14] O.T. Raitakari, M.R. Adams, R.J. McCredie, K.A. Griffiths, D.S. Celermajer, Arterial endothelial dysfunction related to passive smoking is potentially reversible in healthy young adults, Ann. Intern. Med. 130 (1999) 578–581. [15] N.A. Jatoi, P. Jerrard-Dunne, J. Feely, A. Mahmud, Impact of smoking and smoking cessation on arterial stiffness and aortic wave reflection in hypertension, Hypertension 49 (2007) 981–985. [16] B.F. McAdam, D. Byrne, J.D. Morrow, J.A. Oates, Contribution of cyclooxygenase-2 to elevated biosynthesis of thromboxane A2 and prostacyclin in cigarette smokers, Circulation 112 (2005) 1024–1029. [17] M. Hermann, H. Krum, F. Ruschitzka, To the heart of the matter: coxibs, smoking, and cardiovascular risk, Circulation 112 (2005) 941–945. [18] W.M. Loke, K.L.M. Sing, C.Y.J. Lee, et al., Cyclooxygenase-1 mediated platelet reactivity in young male smokers, Clin. Appl. Thromb. Hemost. 20 (2014) 371–377. [19] C. Stefanadis, C. Vlachopoulos, E. Tsiamis, et al., Unfavorable effects of passive smoking on aortic function in men, Ann. Intern. Med. 128 (1998) 426–434. [20] C. Papamichael, K. Karatzi, E. Karatzis, et al., Combined acute effects of red wine consumption and cigarette smoking on haemodynamics of young smokers, J. Hypertens. 24 (2006) 1287–1292. [21] K. Karatzi, C. Papamichael, E. Karatzis, et al., Acute smoke-induced endothelial dysfunction is more prolonged in smokers than in non-smokers, Int. J. Cardiol. 120 (2007) 404–406. [22] C. Vlachopoulos, I. Dima, K. Aznaouridis, et al., Acute systemic inflammation increases arterial stiffness and decreases wave reflections in healthy individuals, Circulation 112 (2005) 2193–2200. [23] L. Smeeth, S.L. Thomas, A.J. Hall, R. Hubbard, P. Farrington, P. Vallance, Risk of myocardial infarction and stroke after acute infection or vaccination, N. Engl. J. Med. 351 (2004) 2611–2618.
[24] T.J. Anderson, A. Uehata, M.D. Gerhard, et al., Close relation of endothelial function in the human coronary and peripheral circulations, J. Am. Coll. Cardiol. 26 (1995) 1235–1241. [25] B. Williams, P.S. Lacy, S.M. Thom, CAFE Investigators, Anglo-Scandinavian Cardiac Outcomes Trial Investigators, CAFE Steering Committee and Writing Committee, et al., Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study, Circulation 113 (2006) 1213–1225. [26] J.S. Bennett, A. Daugherty, D. Herrington, P. Greenland, H. Roberts, K.A. Taubert, The use of nonsteroidal anti-inflammatory drugs (NSAIDs). A science advisory from the American Heart Association, Circulation 111 (2005) 1713–1716. [27] M. Hermann, F. Ruschitzka, Cardiovascular risk of cyclooxygenase-2 inhibitors and traditional non-steroidal anti-inflammatory drugs, Ann. Med. 39 (2007) 18–27. [28] S.E. Kimmel, J.A. Berlin, M. Reilly, et al., Patients exposed to rofecoxib and celecoxib have different odds of nonfatal myocardial infarction, Ann. Intern. Med. 142 (2005) 157–164. [29] P. McGettigan, D. Henry, Cardiovascular risk and inhibition of cyclooxygenase. A systematic review of the observational studies of selective and nonselective inhibitors of cyclooxygenase 2, JAMA 296 (2006) 1633–1644. [30] W.Z. Zhang, K. Venardos, J. Chin-Dusting, D.M. Kaye, Adverse effects of cigarette smoke on NO bioavailability. Role of arginine metabolism and oxidative stress, Hypertension 48 (2006) 278–285. [31] I.B. Wilkinson, S.S. Franklin, J.R. Cockcroft, Nitric oxide and the regulation of large artery stiffness: from physiology to pharmacology, Hypertension 44 (2004) 112–116. [32] I. Ikonomidis, J. Lekakis, G. Vamvakou, F. Andreotti, P. Nihoyannopoulos, Cigarette smoking is associated with increased circulating proinflammatory and procoagulant markers in patients with chronic coronary artery disease: effects of aspirin treatment, Am. Heart J. 149 (2005) 832–839. [33] K. Bhagat, P. Vallance, Inflammatory cytokines impair endothelium-dependent dilatation in human veins in vivo, Circulation 96 (1997) 3042–3047. [34] R.K. Kharbanda, B. Walton, M. Allen, et al., Prevention of inflammation-induced endothelial dysfunction: a novel vasculo-protective action of aspirin, Circulation 105 (2002) 2600–2604. [35] W. Qi, J.X. Wei, I. Dorairaj, R.P. Mahajan, V.G. Wilson, Evidence that a prostanoid produced by cyclo-oxygenase-2 enhances contractile responses of the porcine isolated coronary artery following exposure to lipopolysaccharide, Br. J. Anaesth. 98 (2007) 323–330. [36] J. Blanco-Rivero, V. Cachofeiro, V. Lahera, et al., Participation of prostacyclin in endothelial dysfunction induced by aldosterone in normotensive and hypertensive rats, Hypertension 46 (2005) 107–112. [37] J. Blanco-Rivero, I. Márquez-Rodas, F.E. Xavier, et al., Long-term fenofibrate treatment impairs endothelium-dependent dilation to acetylcholine by altering the cyclooxygenase pathway, Cardiovasc. Res. 75 (2007) 398–407. [38] S.S. Barbieri, B.B. Weksler, Tobacco smoke cooperates with interleukin-1β to alter βcatenin trafficking in vascular endothelium resulting in increased permeability and induction of cyclooxygenase-2 expression in vitro and in vivo, FASEB J. 21 (2007) 1831–1843. [39] J. Toivanen, O. Ylikorkala, L. Viinikka, Effects of smoking and nicotine on human prostacyclin and thromboxane production in vivo and in vitro, Toxicol. Appl. Pharmacol. 82 (1986) 301–306. [40] S. Roy, Effects of smoking on prostacyclin formation and platelet aggregation in users of oral contraceptives, Am. J. Obstet. Gynecol. 180 (1999) S364–S368. [41] P. Borgdorff, M.L. Handoko, Y.Y. Wong, G.T. Tangelder, COX-2 inhibition by use of rofecoxib or high dose aspirin enhances ADP-induced platelet aggregation in fresh blood, Open Cardiovasc. Med. J. 4 (2010) 198–205.