Ticagrelor mitigates ischaemia–reperfusion induced vascular endothelial dysfunction in healthy young males – a randomized, single‐blinded study

Abstract

Aims

Animal data suggest that ticagrelor but not clopidogrel protects against tissue injury. It is unclear if this effect of ticagrelor is also detectable in humans. We studied the effect of ticagrelor and clopidogrel at standard clinical doses on endothelial dysfunction in an experimental model of forearm vascular ischaemia–reperfusion (IR) injury.

Methods

In a randomized, single‐blinded trial, 24 subjects underwent forearm blood flow (FBF) measurements in response to the endothelium‐dependent vasodilator acetylcholine (ACh) and to glyceryltrinitrate (GTN; endothelium‐independent) before and after a 20 min forearm ischaemia. FBF reactivity was assessed after an oral loading dose of ticagrelor or clopidogrel and after 14 days of regular intake of maintenance doses of the study medicines. In addition, the effect on platelet inhibition was evaluated using multiple electrode aggregometry.

Results

ACh‐induced vasodilation was impaired during reperfusion and not completely normalized by acute or chronic treatment with ticagrelor or clopidogrel (post‐ vs. pre‐ischaemia). However, ticagrelor mitigated endothelial dysfunction compared to clopidogrel after loading (FBF ACh_AUC ratio post‐ vs. pre‐ischaemia: 0.83 [0.70; 0.96] vs. 0.64 [0.56; 0.72]; P = 0.024) and after chronic administration (FBF ACh_AUC ratio: 0.86 [0.71; 1.00] vs. 0.66 [0.55; 0.77]; P = 0.027). As expected, GTN‐induced vasodilation was not affected by ischaemia. Ticagrelor or clopidogrel treatment inhibited platelet activation to a similar degree.

Conclusion

Our data indicate that ticagrelor treatment exerts a greater vascular salutary effect than clopidogrel during reperfusion after an acute vascular occlusion. IR‐induced vascular injury cannot be prevented completely by administration of these antiplatelet agents at standard clinical doses.

What is Already Known about this Subject

• Acute vascular occlusion leads to paradoxical ischaemia–reperfusion injury and endothelial dysfunction.

• Ticagrelor but not clopidogrel protects against tissue injury after ischaemia in an animal model.

What this Study Adds

• Ticagrelor mitigates the transient loss of endothelium‐dependent vasodilation during reperfusion compared with clopidogrel after loading and after multiple dosing in healthy subjects.

• Ischaemia–reperfusion‐induced vascular injury cannot be prevented completely in humans by administration of these antiplatelet agents at standard clinical doses.

Introduction

Rapid reperfusion of ischaemic tissue is essential to prevent cell death. However, reperfusion itself has a deleterious effect on tissue function leading to paradoxical ischaemia–reperfusion (IR) injury 1, 2. This is partly attributed to accumulation of reactive oxygen species (ROS) that emerge during reperfusion due to the incapacity of physiologic cellular antioxidant mechanisms 3, 4, 5. The endothelium has also been shown to be vulnerable to IR injury with markedly suppressed endothelial nitric oxide (NO) synthase activity during reperfusion, and combined with increased ROS, leading to excessive inactivation of endothelial‐derived NO 6, 7. This may affect blood vessel function and consequently augment tissue injury. Beside endothelial dysfunction and enhanced ROS levels, IR also induces a local inflammatory response resulting in platelet activation 8, 9, 10. Stimulated thrombocytes themselves have been considered to be important sources of ROS and thus may profoundly contribute to increased oxidative stress 11. Various antiplatelet agents such as the glycoprotein IIb/IIIa inhibitor tirofiban or the platelet P2Y₁₂ receptor antagonist clopidogrel are known to exert beneficial effects on endothelial function in populations at risk for cardiovascular events 12, 13.

Platelet inhibition with a P2Y₁₂ receptor antagonist such as clopidogrel, prasugrel or ticagrelor in combination with acetylsalicylic acid (ASA) has become a mainstay for the treatment of patients with an acute coronary syndrome (ACS) 14, 15. In patients with ACS, ticagrelor has been shown to lower the risk of future cardiac events as compared to clopidogrel 16, 17, which might not only be explained by clopidogrel's delayed onset of action or substantial inter‐individual variability but may be due partly to ticagrelor's pleiotropic effects 18, 19, 20. Preclinical animal data has recently suggested that ticagrelor but not clopidogrel has protective effects on IR injury by inhibition of cellular adenosine uptake with downstream NO‐synthase stimulation 21. It is unclear if this action is part of the superior clinical effect of ticagrelor in humans.

This trial aimed to investigate whether treatment with ticagrelor in healthy subjects can mitigate the transient loss of endothelium‐dependent vasodilation of the vasculature following a short period of local ischaemia compared with clopidogrel at standard clinical doses for ACS. These pharmacodynamic effects were assessed after a loading dose and after 2 weeks of regular intake using a previously described model of ischaemia‐induced endothelial dysfunction of the forearm resistance vasculature 22, 23. This experimental approach was employed, as it is difficult to perform studies investigating IR injury in patients with acute ischaemia and evidence suggests that the endothelium‐dependent vasomotor response of the forearm may serve as a surrogate for coronary endothelial function 24.

Methods

The study protocol was approved by the Ethics Committee of the Medical University of Vienna, Austria (EK 1608/2015), and the national Competent Authority. It conforms to the principles outlined in the Declaration of Helsinki, including current revisions, and the ICH Good Clinical Practice Guidelines. The trial is registered at ClinicalTrials.gov (NCT02580149) and at the European Clinical Trials database (EudraCT 2015‐002763‐41). Written informed consent was obtained from all subjects before study enrolment.

Study population

Forty healthy, non‐smoking male volunteers aged 18–40 with a body mass index (BMI) of 18–27 kg m⁻² were evaluated for eligibility to randomize 24 subjects (Figure 1). Study participants were given a complete health examination at a screening visit and were excluded if they had any significant laboratory or physical finding; a history of cardiovascular disease, renal, hepatic or gastrointestinal impairment; an increased risk of bleeding or coagulation diseases; use of any medication within 2 weeks prior to inclusion. The screening visit was scheduled 3–21 days before the first study day. Subjects abstained from caffeine or xanthine‐containing beverages 24 h before the study days and from alcohol during the period of drug intake.

Figure 1.

Study flow diagram. LD, loading dose; bid, twice daily; qd, once daily

Study design and study drugs

The trial followed a prospective, randomized, controlled, single‐(investigator‐)blinded design and was conducted between October 2015 and March 2016 at the Department of Clinical Pharmacology, Medical University of Vienna, Austria. Subjects were randomly allocated to treatment with ticagrelor or clopidogrel. Investigators directly involved in the study were not aware of the randomization code. Unblinded study staff dispensed the medicines under investigation and supervised oral intake.

Subjects received either ticagrelor (Brilique^®, Astra Zeneca AB, Södertälje, Sweden) or clopidogrel (Plavix^®, Sanofi Clir SNC, Paris, France) for 14 days. All study drugs were administered at approved doses for ACS. On the first study day, a loading dose of 180 mg ticagrelor or a loading dose of 600 mg clopidogrel was administered followed by 90 mg ticagrelor twice daily (bid) or 75 mg clopidogrel once daily (qd) from Day 2 until Day 15. On the last treatment day (= Day 15), the respective medicines were dispensed once (Figure 1).

An end‐of‐study (EOS) visit was scheduled 1 week after the last treatment day.

Forearm blood flow measurements

Forearm blood flow (FBF) was assessed in both arms by strain gauge plethysmography as described previously 25, 26. Briefly, strain gauges, placed on both forearms, were connected to plethysmographs (EC‐6; D.E. Hokanson Inc., Bellevue, WA, USA) for measurement of changes in forearm volume in response to inflation of congesting cuffs on the upper arms to a supravenous pressure (45 mmHg). The distances from the elbows to the strain gauges were measured to ensure a comparable setting between study days. Measurements were recorded for 9 s at 30 s intervals during repeated inflation of the upper arm cuffs. The early linear increases of the curves were used for FBF analysis using the NIVP3 software (version 5.27, D.E. Hokanson Inc., Bellevue, WA, USA) and are expressed as ml* min⁻¹ *100 ml⁻¹ forearm volume.

Experimental protocol

After an overnight fast, FBF measurements were performed in a quiet room with an ambient constant temperature of 23°C (±1°C). FBF was assessed on Day 1 at pre‐dose (prior to the loading dose) and on both study days before and after the non‐dominant forearm was made ischaemic (Figure 2).

Figure 2.

Flow chart of the two study days that were 2 weeks apart. FBF, forearm blood flow; MEA, multiple electrode aggregometry

Subjects were in a supine position and a 27 gauge fine‐bore needle (Sol Care, Sol‐Millennium Medical Inc., Lawrenceville, GA, USA) was inserted into the brachial artery of the non‐dominant arm for intra‐arterial administration of vasoactive agents. Control (0.9% sodium chloride solution; Fresenius Kabi, Graz, Austria) FBF measurements were recorded over 5 min followed by assessing the response to the endothelium‐dependent vasodilator acetylcholine (ACh; Miochol‐E^®, Bausch & Lomb Swiss AG, Switzerland) at increasing doses of 25, 50 and 100 nmol min⁻¹ (each for 3 min), respectively. To test smooth muscle function, the endothelium‐independent dilator glyceryltrinitrate (GTN; Nitro POHL, G. Pohl‐Boskamp GmbH & Co. KG, Germany; 4, 8 and 16 nmol min⁻¹; each for 3 min) was administered after a 15 min washout period to allow restoration of control FBF conditions.

The intra‐arterial needle was removed after pre‐dose (i.e., before any study drug was administered on Day 1) and pre‐ischaemia FBF measurements. Forearm ischaemia was applied to the non‐dominant arm 2 h after administration of ticagrelor or clopidogrel for 20 min by inflating an upper arm cuff to suprasystolic values (>220 mmHg). Immediately after cuff release, an intra‐arterial cannula was reinserted followed by assessment of post‐ischaemia FBF response starting at 10 min after reperfusion onset.

Platelet inhibition

Blood for assessment of platelet function was collected by fresh venepunctures on both study days at pre‐dose (−2 h prior to ischaemia), immediately before (0 min) and after the 20 min forearm ischaemia, and at 70 min after forearm ischaemia (Figure 2). To assess possible effects of local ischaemia on platelet function, blood samples during reperfusion were taken from both forearms. Blood was collected into tubes containing recombinant hirudin (Roche, Mannheim, Germany) and analysed within 0.5–3 h after sampling using multiple aggregometry (MEA; Multiplate^® Analyser; Roche). Three hundred millilitres of hirudin blood were diluted with 0.9% saline solution at a 1:1 ratio and incubated for 3 min. After stirring at 37°C, analyses were performed with the agonist ADP and PGE₁ as inhibitor of platelet activation at final concentrations of 6.3 μM and 9.4 nM, respectively. The slope of the increase in electrical impedance was recorded for 6 min and is expressed as area under the curve (AUC) of the aggregation tracing 27, 28. The normal reference range for the AUC is 43–100 units (U).

Study objectives

The predefined primary study endpoints were to compare the ratios of the area under the vasodilator dose vs. FBF curve (AUC) to ACh (ACh_AUC) post‐ vs. pre‐ischaemia between treatment groups during acute and chronic administration of the medicines under study. Endothelium‐dependent vasodilation was considered to be of primary interest since vasodilating smooth muscle function in the forearm vasculature has been shown to be unaffected by local ischaemia in a previous own trial 23.

Secondary outcome parameters included GTN‐induced vasodilation during reperfusion and assessment of additive effects of ticagrelor or clopidogrel on FBF reactivity in the absence of an ischaemic event (pre‐ischaemia vs. pre‐dose). Other secondary objectives were the inhibitory effect of ticagrelor and clopidogrel on whole blood platelet aggregation using MEA, and safety.

Statistical analysis and sample size estimation

SPSS 23 (IBM Cooperation, New York, NY, USA) and Microsoft Excel 2011 for Macintosh (Microsoft Corporation, Redmond, WA, USA) were used for statistical analysis. All data were examined for normal distribution with the Kolomogorov–Smirnov test and log transformed for further analysis if they were non‐normally distributed.

FBF was calculated as ml*min⁻¹*100 ml⁻¹ forearm volume, percent changes, AUC or ratios before unblinding of the study. The effects of ACh or GTN (= mean of six measurements at each dose level) on FBF are expressed as percent changes of ratios from intervention vs. control forearm with individual basal FBF ratios (= mean of 10 measurements during saline infusion) defined as 100%. FBF‐AUC were derived from the dose‐effect curves according to the linear trapezoidal rule restricted to the range of active dosing consisting of three data points (i.e. 25, 50, 100 nmol*min⁻¹ for ACh, and 4, 8, 16 nmol*min⁻¹ for GTN, respectively). The effects of acute or chronic treatment with ticagrelor or clopidogrel on individual ACh or GTN dose levels and FBF‐AUC during reperfusion (post‐ vs. pre‐ischaemia), a possible additive effect of the study medicines (pre‐ischaemia vs. baseline [= before loading dose]), as well as the degree of platelet inhibition, were analysed by a repeated measures ANOVA. Time point and dose level were considered as within‐subject factors, treatment as a between‐group factor and, if appropriate, pre‐dose (baseline) levels were used as covariate in the ANOVA model. Ratios of AUC were analysed using the two‐sample t test and for the primary study objectives (ACh_AUC ratios post‐ vs. pre‐ischaemia) the Bonferroni–Holm procedure was applied to adjust for multiple comparisons. All other (secondary) study objectives were analysed in an exploratory fashion. An (adjusted) two‐sided P value <0.05 was considered to be statistically significant. Values are given as mean and 95% confidence interval (CI) unless indicated otherwise.

Sample size calculation was based on the predefined primary outcome parameter ACh_AUC ratio (post‐ vs. pre‐ischaemia). Previous data indicate that the within‐group SD for the percent FBF changes at different ACh doses is approximately 20% 29. Due to a lack of data for the expected impact of ticagrelor vs. clopidogrel on ACh‐induced vasodilation, a difference of 25% between groups was considered to be clinically relevant. Thus, with a sample size of 12 subjects per group, a mean difference of 25% in the ACh_AUC ratios (ticagrelor vs. clopidogrel) could be detected with 80% power using a two‐sample t test at a significance level of 5% (two‐sided). The calculated sample size of 12 per group included drop‐out of one subject.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 30, and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 31.

Results

Study population

Demographic and laboratory data of randomized study participants are presented in Table 1.

Table 1.

Characteristics of study participants. Data are presented as mean and S.E.M.

	Ticagrelor (n = 12)	Clopidogrel (n = 12)
Age, years	27 (2)	27 (1)
Body mass index, kg m −2	22.5 (0.5)	22.5 (0.6)
Platelet count, G l −1	219 (13)	233 (10)
Prothrombin time, %	85 (4)	89 (4)
Activated partial thromboplastin time, s	34.7 (1.0)	36.1 (0.8)
Serum creatinine, mg dl −1	0.9 (0.0)	1.0 (0.0)
Aspartate aminotransferase, U l −1	25 (2)	27 (2)
Alanine aminotransferase, U l −1	24 (2)	25 (3)
γ‐Glutamyl transferase, U l −1	16 (1)	16 (1)
Systolic blood pressure, mmHg	128 (3)	127 (2)
Diastolic blood pressure, mmHg	76 (3)	77 (2)

Forearm blood flow

Basal FBF of the interventional and control arm during intra‐arterial saline infusion are presented in Table 2. FBF at pre‐dose (before administration of ticagrelor or clopidogrel) on Day 1 was significantly higher in both the interventional and control arm as compared to the basal pre‐ or post‐ischaemic FBF measurements (P = 0.005 [interventional arm] and P < 0.001 [non‐interventional arm], ANOVA). However, FBF at corresponding time points was comparable between study cohorts and arms, respectively (P > 0.05 for group differences and interventional vs. control arm, ANOVA).

Table 2.

Absolute forearm blood flow (ml*min⁻¹*100 ml⁻¹) of the interventional and control arm at baseline (BL) before study drug administration and during treatment with ticagrelor or clopidogrel before (PRE) and after (POST) forearm ischaemia (n = 12 per group). Data are presented before and after acute dosing on Day 1 and after multiple dosing on Day 15 as mean and 95% confidence interval

	Day 1
	Ticagrelor		Clopidogrel
	Interventional arm	Control arm	Interventional arm	Control arm
BL	5.7 (4.2; 7.1)	5.8 (4.0; 7.6)	5.1 (3.3; 6.9)	4.7 (2.6; 6.9)
PRE	4.2 (2.8; 5.6)*	4.2 (2.9; 5.6)*	3.8 (2.8; 4.9)*	3.5 (2.5; 4.5)*
POST	3.8 (3.1; 4.6)*	3.2 (2.5; 4.0)*	3.7 (2.5; 4.9)*	2.7 (2.1; 4.3)*

	Day 15
	Ticagrelor		Clopidogrel
	Interventional arm	Control arm	Interventional arm	Control arm
PRE	4.1 (3.0; 5.3)*	4.7 (3.3; 6.0)*	3.9 (2.6; 5.3)*	3.9 (2.4; 5.5)*
POST	4.2 (3.1; 5.2)*	3.8 (2.8; 4.7)*	3.7 (2.4; 5.0)*	3.2 (2.2; 4.2)*

^* P < 0.05 vs. BL (ANOVA).

Endothelium‐dependent (ACh‐induced) response

Figure 3 shows the percent changes of ACh‐induced vasodilation of the intervention vs. control arm (FBF ratio) at different dose levels and time points, respectively. ACh infusion caused a dose‐dependent increase in FBF ratio on both study days (P < 0.001 for dose levels, ANOVA). FBF reactivity is summarized in Table 3, and Figure 4 shows the FBF ratio calculated as ACh_AUC at pre‐dose and during acute or repeated dosing. There was a small increase in FBF reactivity after the loading dose and after multiple dosing of ticagrelor or clopidogrel, which was not statistically significant compared to baseline or between treatment groups (Table 3; Figure 4). At 10 min after reperfusion onset, FBF response was impaired and could not be completely normalized by ticagrelor or clopidogrel (P < 0.001 post‐ vs. pre‐ischaemia ACh_AUC Day 1 and Day 15, ANOVA). However, vasodilation to ACh was substantially less mitigated during ticagrelor loading as compared to clopidogrel (0.83 vs. 0.64 for ACh_AUC ratios post‐ vs. pre‐ischaemia; P = 0.024, two‐sample t test). After chronic administration, a similar response was observed on ACh_AUC ratios with 0.86 during ticagrelor and 0.66 during clopidogrel intake, respectively (P = 0.027, two‐sample t test). Differences in ACh_AUC between groups resulted from a less impaired FBF reactivity at all ACh dose levels in subjects who received ticagrelor (Figure 3, C and E).

Figure 3.

Forearm blood flow response to acetylcholine in the intervention vs. control arm (FBF ratio) at pre‐dose (A), before a 20 min forearm ischaemia that was applied at 2 h after loading with ticagrelor or clopidogrel (B) and at 10 min of reperfusion (C). Panels D and E show the FBF response to acetylcholine after 14 days intake of ticagrelor or clopidogrel before (D) and at 10 min of reperfusion after forearm ischaemia. During reperfusion, reactivity to acetylcholine was less impaired in subjects receiving ticagrelor as compared to clopidogrel. FBF ratio is expressed as mean ± S.E.M.; n = 12 per group; *P < 0.05 for corresponding acetylcholine dose levels between groups; two‐sample t test

Table 3.

Ratios of the area under the vasodilator dose vs. forearm blood flow curve to acetylcholine (ACh_AUC ratio) or glyceryltrinitrate (GTN_AUC ratio) at baseline (BL) before study drug administration and during treatment with ticagrelor or clopidogrel before (PRE) and after (POST) forearm ischaemia (n = 12 per group). Data are presented before and after acute dosing on Day 1 and after multiple dosing on Day 15 as mean and 95% confidence interval

	Day 1		Day 15
	Ticagrelor	Clopidogrel	Ticagrelor	Clopidogrel
ACh AUC ratio
PRE vs. BL	1.16 (1.00; 1.32)	1.18 (0.87; 1.48)	1.07 (0.89; 1.24)	1.03 (0.79; 1.27)
POST vs. PRE	0.83 (0.70; 0.96)*	0.64 (0.56; 0.72)	0.86 (0.71; 1.00)*	0.66 (0.55; 0.77)

	Day 1		Day 15
	Ticagrelor	Clopidogrel	Ticagrelor	Clopidogrel
GTN AUC ratio
PRE vs. BL	1.17 (0.91; 1.44)	1.12 (0.83; 1.41)	1.00 (0.87; 1.14)	1.18 (0.93; 1.43)
POST vs. PRE	1.06 (0.70; 1.43)	1.05 (0.83; 1.28)	1.05 (0.88; 1.23)	1.08 (0.81; 1.34)

^* P < 0.05 between treatment groups (two‐sample t test)

Figure 4.

Area under the vasodilator dose vs. forearm blood flow curve in the interventional vs. control arm to acetylcholine (ACh_AUC) at pre‐dose (baseline) and after loading or multiple dose treatment with ticagrelor or clopidogrel over 14 days before and after ischaemia reperfusion (IR). (n = 12 per group). Data are presented as mean and 95% confidence interval; *P < 0.05 before vs. after IR; paired t test

Endothelium‐independent (GTN‐induced) response

The dose‐dependent increases of endothelium‐independent vasodilation were similar across study cohorts at all FBF assessments (P < 0.001 for dose levels, P = 0.75 for group differences, ANOVA). FBF ratios calculated from GTN_AUC are presented in Table 3. Administration of ticagrelor or clopidogrel did not significantly affect GTN‐induced vasodilation and the response to GTN during reperfusion was similar to that seen prior to ischaemia after loading and after 2 weeks of regular intake of ticagrelor or clopidogrel.

Platelet inhibition

The impact of the medicines under study on MEA is summarized in Table 4. Administration of ticagrelor or clopidogrel induced a rapid and sustained platelet inhibition, which was comparable between treatments (P < 0.001 vs. pre‐dose, P = 0.41 for group differences, ANOVA). At 2 h after loading, (pre‐ischaemia) reductions to 16 U (ticagrelor) and 22 U (clopidogrel) were noted. Platelet activation remained significantly attenuated during ticagrelor and clopidogrel (18 U) treatment after ischaemia, respectively. At pre‐dose after 2 weeks of regular intake, the inhibitory effect was similar between ticagrelor and clopidogrel (20 U vs. 22 U) with maximum reductions to 16 U for ticagrelor or 14 U for clopidogrel after dosing. Forearm ischaemia on the interventional arm did not alter MEA responses as compared with the control arm (P > 0.05 interventional vs. control arm, paired t test; Table 4).

Table 4.

Platelet inhibition in response to ticagrelor or clopidogrel loading (Day 1) and after repeated (multiple) dosing (Day 15) using multiple electrode aggregometry (n = 12 per group). Data are presented as mean and 95% confidence interval. No significant differences between treatment groups or arms were detectable

	Multiple Electrode Aggregometry (Units)
	Ticagrelor		Clopidogrel
	Control arm	Interventional arm	Control arm	Interventional arm
Day 1
Pre‐dose (−120 min)	60 (51; 71)	N.A.	54 (47; 62)	N.A.
Pre‐ischaemia (0 min)	16 (12; 20)*	N.A.	22 (16; 28)*	N.A.
Post‐ischaemia (+20 min)	18 (12; 24)*	18 (14; 22)*	18 (12; 24)*	20 (14; 26)*
Post FBF (+70 min)	17 (9; 25)*	21 (11; 31)*	21 (13; 29)*	21 (13; 28)*
Day 15
Pre‐dose (−120 min)	20 (12; 28)*	N.A.	22 (15; 29)*	N.A.
Pre‐ischaemia (0 min)	17 (13; 22)*	N.A.	15 (11; 19)*	N.A.
Post‐ischaemia (+20 min)	16 (11; 22)*	16 (12; 20)*	14 (10; 19)*	15 (11; 20)*
Post FBF (+70 min)	17 (11; 24)*	18 (12; 25)*	15 (10; 20)*	15 (7; 22)*

^* P < 0.05 vs. pre‐dose (ANOVA). FBF = forearm blood flow; N.A. = not assessed.

Safety

The medicines under study were well tolerated and the incidence of adverse events (AE) was low. Seventeen AEs were reported (ticagrelor: n = 10, clopidogrel: n = 7) that were mild in severity and resolved without sequelae. Six AEs were classified as probably related to antithrombotic treatment (haematoma [n = 1, ticagrelor; n = 3, clopidogrel], gingival bleeding [n = 1, ticagrelor], dyspnoea [n = 1, ticagrelor]), two AEs were caused by arterial needle insertion with transient paraesthesia and nine were considered to be unrelated to pharmacological treatment or trial‐related interventions (pruritus, sore throat, fatigue, headache, vertigo, nausea, common cold, tonsillitis, increased thirst).

Discussion

This study demonstrates that IR‐induced forearm vascular endothelial dysfunction could not be fully prevented by treatment with oral ticagrelor or clopidogrel at therapeutic doses. However, the transient loss of ACh‐induced resistance to artery vasodilation was less impaired during ticagrelor compared with clopidogrel administration. After acute loading and after repeated dosing over 2 weeks, vasodilatory smooth muscle function was unaffected by IR. Both antiplatelet drugs induced a comparable and sustained platelet inhibition. These findings suggest that ticagrelor may have greater effects on resistance artery endothelial function than clopidogrel at an equivalent degree of platelet inhibition.

Consistent with previous data, our results confirm that an intermittent arterial vascular occlusion causes severe endothelial dysfunction, whereas the endothelium‐independent vasodilator response is not markedly altered 23. The present study employed an active control using clopidogrel as comparator and did not assess the impact of a transient ischaemia on the resistance forearm vasculature alone. A previous trial 23 in healthy participants demonstrated a similar vascular impairment during placebo administration as observed in our study with clopidogrel treatment, indicating that there is little if any effect of clopidogrel on ACh‐induced vasodilation during reperfusion in healthy subjects. Further, the effect of prasugrel was not assessed in this study and it remains unclear whether this P2Y₁₂ receptor antagonist affects the endothelium‐dependent FBF response. However, available data suggests that prasugrel does not improve vascular function substantially 32, 33.

It has been proposed that ticagrelor's pleiotropic effects are mediated by locally increased adenosine concentrations and endothelial NO synthase activation that is considered, among others, to play a key role in preventing endothelial dysfunction and IR injury 16, 21, 34, 35, 36. In our study orally administered ticagrelor at standard clinical doses could not fully restore ACh‐stimulated vasodilation in the forearm vasculature. We cannot conclude if ticagrelor's limited benefit is due to exceeding its capacity of NO synthase stimulation with subsequent imbalance of NO and ROS as these parameter were not assessed. However, Campo et al. reported increased NO and mitigated ROS levels during ticagrelor treatment, which also induced higher adenosine plasma concentrations in cardiac patients 34, 37.

Our data show that acute or chronic treatment with ticagrelor (or clopidogrel) had no effect on FBF reactivity in the absence of vascular ischaemia. It may be hypothesized that vascular NO bioactivity is not augmented by platelet inhibition, at least in healthy subjects under resting conditions. It is unlikely that the higher absolute FBF during saline infusion at pre‐dose (Table 2) has affected this observation, as ratios of interventional vs. control arm were analysed and absolute changes of vasodilation in response to ACh did not differ between assessments (pre‐ischaemia vs. pre‐dose). The higher absolute baseline FBF prior to loading with the study medicines may be explained by a shorter resting period before the pre‐dose FBF that was assessed shortly after admission to our department.

Pharmacodynamic responses obtained with MEA indicate that both ticagrelor and clopidogrel induced a comparable and sustained platelet inhibition as noted in a previous healthy volunteer study 38, which suggests that the observed group differences in FBF reactivity did not result from individual resistance to clopidogrel 27. Mitigated vascular ACh reactivity resulting from a delayed onset of clopidogrel's action after loading can be ruled out as a similar ACh‐induced vasodilation was observed at steady state after 2 weeks of regular intake. However, in at‐risk populations, the response on platelet inhibition is more pronounced with ticagrelor compared to clopidogrel and it is possible that the clinically relevant differences in trials with ACS patients are related to the greater inhibitory properties of ticagrelor than to its pleiotropic effects 16, 39.

In patients with ACS, a P2Y₁₂ receptor antagonist in combination with acetylsalicylic acid (ASA) is recommended for secondary prevention and statin and angiotensin‐converting enzyme (ACE) inhibitor therapy are prescribed 40. Animal data has shown that the tissue protective effect of ticagrelor is not abrogated by co‐administration of ASA (up to 5 mg kg⁻¹ BW) 36. In addition, statins and ACE inhibitors have also been suggested to ameliorate endothelial dysfunction and thus may modulate ticagrelor's effect in cardiac patients with co‐medication 41, 42, 43.

Limitations

The trial was conducted in healthy volunteers and an experimental vascular model of the forearm was used to assess IR injury. Thus it remains unclear whether our results can be extrapolated to a population at risk such as patients with ACS or peripheral artery disease with pre‐existing endothelial dysfunction and acute ischaemia in a different vascular bed with thrombus formation. Further, the more potent P2Y₁₂ receptor antagonist prasugrel was not tested and markers of oxidative stress (ROS, NO) or adenosine concentrations were not assessed.

Conclusion

The impairment of ACh‐induced vasodilation during reperfusion after acute vascular occlusion was less pronounced after ticagrelor administration compared to clopidogrel. This was detectable following a loading dose and after 2 weeks of regular intake of ticagrelor or clopidogrel, despite a similar degree of platelet inhibition achieved.

Competing Interests

There are no competing interests to declare.

We are grateful for the administrative work and assistance of Claudia Eder RN.

Contributors

S.W.: drafting of the manuscript, acquisition of data, statistical analysis and interpretation of data; B.L.: statistical analysis and interpretation of data, critical revision of the manuscript for important intellectual content; M.E. and E.L.H.: acquisition of data, critical revision of the manuscript for important intellectual content; M.W.: study concept and design, interpretation of data, critical revision of the manuscript for important intellectual content.

Weisshaar, S. , Litschauer, B. , Eipeldauer, M. , Hobl, E. L. , and Wolzt, M. (2017) Ticagrelor mitigates ischaemia–reperfusion induced vascular endothelial dysfunction in healthy young males – a randomized, single‐blinded study. Br J Clin Pharmacol, 83: 2651–2660. doi: 10.1111/bcp.13378.