High on treatment platelet reactivity against aspirin by ... · © Schattauer 2013 Thrombosis and Haemostasis 109.5/2013 ... anti-inflammatory drugs pharmacological mechanisms and

© Schattauer 2013 Thrombosis and Haemostasis 109.5/2013

825Theme Issue Article

High on treatment platelet reactivity against aspirin by non-steroidal anti-inflammatory drugs – pharmacological mechanisms and clinical relevanceThomas Hohlfeld; Aaruni Saxena; Karsten SchrörInstitut für Pharmakologie und Klinische Pharmakologie, Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany

SummaryInhibition of platelet function by aspirin results from irreversible in-hibition of platelet cyclooxygenase (COX)-1. While sufficient inhibition is obtained at antiplatelet doses (75–325 mg/day) in most (≥95%) treated patients, the antiplatelet effect of aspirin and subsequent car-diovascular risk reduction is much less in clinical settings and disease-dependent. Several reasons for this “high on treatment platelet reac-tivity” are known. This paper reviews the evidence for an interaction between aspirin and other COX inhibitors, namely non-steroidal anti-inflammatory drugs (NSAIDs). Numerous experimental studies dem-onstrated a pharmacodynamic interaction between aspirin and NSAIDs. This likely occurs within the hydrophobic substrate channel of platelet COX-1 and might be explained by molecular competition be-tween inhibitor drugs and substrate (arachidonic acid) at overlapping binding sites. This interaction is found with some compounds, notably ibuprofen and dipyrone (metamizole), but not with others, such as di-

clofenac and acetaminophen (paracetamol). Hence, this interaction is not a class effect of NSAIDs and/or non-steroidal analgesics but rather due to specific structural requirements which still remain to be de-fined. In vivo studies on healthy subjects and patients tend to confirm this type of interaction as well as large differences between NSAIDs and non-steroidal analgesics, respectively. These interactions may be clinically relevant and may increase the cardiovascular risk in long-term treatment for primary and secondary cardiovascular prevention in patients with chronic inflammation, such as rheumatoid arthritis. These patients have an elevated risk for myocardial infarctions and may require chronic antiplatelet treatment by aspirin in addition to treatment of inflammatory pain.

KeywordsAspirin, drug-interaction, non-steroidal anti-inflammatory drugs, dipy-rone, platelet function

Correspondence to:Prof. Dr. med. Thomas HohlfeldInstitut für Pharmakologie und Klinische PharmakologieHeinrich-Heine-Universität DüsseldorfMoorenstraße 540225 Düsseldorf, GermanyTel.: +49 211 81 12505, Fax: +49 211 81 14781E-mail: [email protected]

Received: July 31, 2012 Accepted after minor revision: November 13, 2012 Prepublished online: December 13, 2012

doi:10.1160/TH12-07-0532Thromb Haemost 2013; 109: 825–833

Introduction

The usefulness of aspirin as first-line antiplatelet treatment for pre-vention of myocardial infarctions in patients at elevated vascular risk is well established. Nevertheless, large differences exist in effi-cacy among different groups of patients including those whose platelets respond less to conventional doses of this agent than others (1). Several explanations for this “high on treatment platelet reactivity” (HPR) have been discussed: (i) variable bioavailability and/or enhanced degradation of aspirin by aspirin esterases, (ii) alterations of the pharmacological target, i.e. “bypass” of cyclo-oxygenase (COX)-1-mediated thromboxane formation via COX-2 mediated generation of thromboxane precursors, (iii) gene poly-morphisms of COX-1, (iv) residual platelet reactivity independent of COX-1, i.e. platelet stimulation by aspirin-insensitive mechan-isms and (v) negative interactions with other drugs (2-5). Thus, HPR against aspirin is clearly multifactorial in nature.

Several studies suggested that HPR may clinically result in treatment failure, i.e. insufficient protection from myocardial in-farction or ischaemic stroke. For example, Gum et al. examined a subset of stable cardiovascular patients with HPR (“aspirin resis-tance”) based on a low inhibition of arachidonic acid-induced pla-telet aggregation (6). These exhibited an increased risk of death, myocardial infarction or a cerebrovascular accident. Similar data have been reported by Chen et al. (7). HPR against aspirin, when present together with HPR against clopidogrel, also turned out as a strong risk factor for drug-eluting stent thrombosis during a six-month follow-up after percutaneous coronary intervention (8). Vice versa, another study reported that patients with drug eluting stent thrombosis exhibited a significantly greater rate of in vitro HPR against aspirin (9). A recent study in patients after off-pump coronary artery bypass grafting (CABG) suggested an association of impaired platelet laboratory response to aspirin (or clopidogrel) with an increased risk of saphenous vein graft occlusion (10).

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2017-06-19 | IP: 54.191.40.80

Thrombosis and Haemostasis 109.5/2013 © Schattauer 2013

826 Hohlfeld et al. HPR against aspirin by NSAIDs

While genetically fixed reasons for HPR cannot be modified, re-moval or treatment of acquired factors can. These include negative interactions with other drugs, such as non-aspirin non-steroidal anti-inflammatory drugs (NSAID). Ibuprofen, naproxen, diclofenac and various coxibs are among the most often prescribed drugs world-wide. Their use in individuals with chronic inflammatory diseases, who additionally require thrombosis prevention by aspirin, is highly prevalent. Co-medication of non-aspirin NSAIDs and aspirin is fa-voured by the fact that immune-mediated diseases, such as rheuma-toid arthritis, show an elevated co-morbidity of myocardial infarction and ischaemic stroke (11-13) due to pathophysiological links be-tween inflammation and atherosclerosis (14).

This article examines potential mechanisms which may cause this drug/drug interaction on a molecular basis, i.e. at the level of enzyme kinetics, summarises the existing experimental and clini-cal evidence of NSAID/aspirin interactions and gives an outlook of how to avoid this in clinical practice.

Mechanism of acquired HPR against aspirin and NSAID/aspirin interaction

The antiplatelet action of aspirin is due to irreversible acetylation of platelet COX-1, resulting in irreversible inhibition of platelet-

dependent generation of prostaglandin endoperoxides and their subsequent conversion into thromboxane (TX) A2. This inhibition must be apparently complete (>95%) in terms of TX - forming ca-pacity to result in inhibition of platelet function. The reason is the all-or-nothing type of this response. At antiplatelet doses of 75-325 mg/day, sufficient inhibition is obtained in most (≥95%) of pa-tients (3, 15). This is much more than the about 20 % risk reduc-tion of cardiovascular events in secondary prevention (1) and sug-gests additional factors, i.e. aspirin-insensitive pathways of platelet activation or HPR as explanations.

COX-1 as the site of aspirin/non aspirin-NSAID interaction

While both aspirin and non-aspirin NSAIDs inhibit prostaglandin and TXA2 biosynthesis, non-aspirin NSAIDs cannot replace for aspirin as antiplatelet drugs. Unlike aspirin, these are reversibly acting compounds and the degree of inhibition of platelet COX-1 is determined by their plasma half-life, for most compounds in the range of 2-4 hours (h). After platelet stimulation, the local concen-tration of the substrate arachidonic acid within the binding chan-nel of COX-1 may become sufficient to displace these compounds from their binding to COX-1. This process becomes facilitated by

Figure 1: Hydrophobic channel of COX-1 as derived from crystal structure analysis. A) Empty channel without ligand; B) Binding of arachidonic acid, the natural substrate; C) Pu-tative position of salicylic acid and several non-selective NSAIDs within the hydrophobic chan-nel; D) Acetylation of Ser530 by aspirin resulting in inhibition of enzyme activity (Figure kindly provided by P. Loll).

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity



827Hohlfeld et al. HPR against aspirin by NSAIDs

clearance of the compounds, i.e. their wash-out, from plasma, and the fact that the relation between the inhibition of platelet COX-1-dependent TXA2 generation and inhibition of TXA2-dependent platelet function is non-linear (16, 17). As a net effect, NSAIDs are unable to allow for continuous circadian inhibition of platelet COX-1 and platelet function although they do inhibit platelet function and this is likely to contribute to their haemorrhagic side effects.

The molecular mechanisms of this interaction are summarised in ▶ Figure 1. Cyclooxygenases are homodimeric peroxidase enzymes associated with the cell membrane. The biological sub-strate, arachidonic acid, is released from membrane phospholipids upon cell (platelet) stimulation and enters a hydrophobic substrate binding channel towards the cytalytic site at the interior of the enzyme. The active enzyme converts arachidonic acid via a multi-step reaction into prostaglandin H2 which in platelets is further converted to TXA2 by thromboxane synthase. Repeated adminis-tration of low-dose aspirin cumulatively inactivates platelet COX-1 in the anucleate platelet (18) by acetylation of a serine530 near the catalytic site (19, 20). Arachidonic acid (▶ Figure 1 B) and non-aspirin NSAIDs, including ibuprofen, naproxen and others, are attracted by various amino acid residues in the hydrophobic channel at overlapping binding sites (▶ Figure 1 C) (21, 22). Both, acetylation by aspirin (▶ Figure 1 D) and the presence of NSAIDs (▶ Figure 1 C), hinder the access of arachidonic acid to the cata-lytic site at tyrosine385.

Competitive mechanism of interaction

Despite of these pharmacokinetic interactions, one might still ex-pect that COX-1 inhibition is synergistically amplified because both non-aspirin NSAIDs and aspirin act pharmacodynamically into the same direction. This, however, is not necessarily the case. The half-life of aspirin in blood plasma is only 15-20 minutes (min) due to rapid deacetylation to salicylate, which does not in-hibit platelet COX-1. This is much shorter than the half-life of most non-aspirin NSAIDs as outlined above and might result in inactivation of aspirin prior to its binding to the COX enzyme. This is facilitated by a relatively weak initial binding affinity of as-pirin within the hydrophobic channel, having a Ki of 27 mmol/l (23). Non-aspirin NSAIDs have Ki values several orders of magni-tude lower (usually micromolar). If present, they will prevent the access of aspirin to the site of acetylation and ´protect´ platelet COX-1 from permanent inactivation by aspirin as long as they are bound there in sufficient amounts. This is exemplified in ▶ Figure 2, showing the docking of a reversible inhibitor (ibuprofen) within the hydrophobic channel of COX-1. The carboxy group forms pu-tative hydrogen bonds with the hydroxy residues of serine530 and tyrosine385. This probably hinders the access of aspirin towards this site.

The hypothetical competion between aspirin and non-aspirin NSAIDs has been examined by enzyme kinetic modelling. One model (24) included several steps of the catalytic mechanism of COX-1 and calculated the interaction of a reversible NSAID (ce-lecoxib) with an irreversible inhibitor (aspirin). It was crossvali-

dated with in vitro experimental data on COX-1 enzyme kinetics available from the literature. The model suggested that COX-1 acetylation by aspirin was dramatically delayed in the presence of celecoxib and that this occured already at concentrations of celec-oxib which only weakly inhibit enzyme activity themselves. Unfor-tunately, this work was not extended to other common NSAIDs, which would be of interest because of the differences in inhibition kinetics among NSAIDs. A similar approach was used by Hong et al. (25) who examined the ibuprofen-induced inhibition of platelet aggregation based on experimental data generated with blood from healthy donors. While the methodology of model develop-ment was different, this simulation also concluded that a reversible COX-1 inhibitor (here ibuprofen) antagonises irreversible platelet inhibition by aspirin.

Further concepts of the molecular interaction between aspirin and NSAIDs

While a direct competitive aspirin/NSAID interaction has been clearly presented and models based on this achieve realistic results (see below), the mechanism(s) of interaction in reality may be more complex. A recent concept focusses on a ´cross-talk´ be-tween the two COX subunits (26, 27). Many NSAIDs appear to act by regulating one COX monomer, which does not contribute to catalysis. This monomer acts as a regulatory subunit controlling the activity of the partner monomer, which is catalytically active. There is evidence that only one of the two subunits of COX enzymes participates at a time in the catalytic cycle (28, 29). Hence, COX may be regarded both as a sequence homodimer and a functional heterodimer. Some NSAIDs appear to bind to the regulatory subunit and exert non-competitive inhibiton of the catalytic subunit, while others interact with the catalytic subunit, inhibiting enzyme activity by a competitive mechanism (27).

Figure 2: Computerised docking of the putative binding of ibu-profen in the hydrophobic channel of COX-1. The green lines indicate hydrogen bond interactions. The surrounding amino acids Tyr385, Ser530, Tyr355 and Arg120 are important for the docking of aspirin to the binding pocket. Ibuprofen likely consumes the space required for aspirin to locate within the COX-1 channel (Saxena et al., manuscript in preparation).

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity




Moreover, kinetic analyses distinguished different types of NSAIDs depending on whether they act in a time-dependent manner or not. The time-dependent mode of inhibition appears to be associated with a conformational change of the enzyme. With this in mind, it may be reasoned that the interaction between COX inhibitors does not necessarily result from direct ligand competi-tion for a common binding site within the hydrophobic COX channel of the same subunit. It may (also) result from a cross-regulation between the two subunits of the dimeric COX enzyme, where binding of a non-aspirin NSAID to one monomer (the regulatory subunit) induces a conformational change. This may be transferred via amino acids located at the interface between the two monomers towards the partner monomer (catalytic subunit), which may undergo a secondary conformational change within the substrate channel. This change may subsequently slow down the rate of acetylation by aspirin. A careful analysis of the drug/drug interaction between celecoxib and aspirin based on crystallo-graphic analyses and animal experimentation suggested such a mechanism for the interaction between aspirin and celecoxib at platelet COX-1 (30).

Evidence for negative interactions in experimental and clinical studies

An interaction of reversible COX inhibitors has been demon-strated in vitro decades ago, mainly in studies on the pharmacody-namic properties of NSAIDs (31-33). These early studies did not focus on the potential clinical significance of this interaction. This has now changed, after it became apparent that non-aspirin NSAIDs and coxibs unlike aspirin, might increase the cardiovas-cular risk and possibly interact with aspirin in a negative manner in both experimental (34) and clinical (35, 36) trials. Leading health agencies, such as the European Medicines Agency (EMA)

and the Food and Drug Administration (FDA), consider these in-teractions as ´potentially important´ (37).

In vitro studies

Aspirin/NSAID interactions can be demonstrated experimentally, for example by pre-incubation of platelet suspensions with a non-aspirin NSAID followed by addition of aspirin and subsequent pla-telet stimulation by an agonist, such as arachidonic acid or low concentration-collagen, that act solely through generation of TXA2. Many non-aspirin NSAIDs partially or completely prevent platelet inhibition by aspirin in such settings. ▶ Figure 3 gives an example for the interaction of aspirin and piroxicam, a reversible long acting NSAID. There was complete inhibition of aggregation and TX synthesis by aspirin (30 µmol/l) and no interaction with piroxicam at concentrations ≤0.1 µmol/l. Higher concentrations of piroxicam, however, resulted in a concentration-dependent antag-onism of the inhibition by aspirin as seen from increased platelet aggregation and TX formation. Further increase of piroxicam con-centrations above 30 µmol/l reduced platelet activity and thromb-oxane synthesis, probably due to COX-1 inhibition by piroxicam. However, these concentrations of piroxicam are above those after analgesic dosing: the peak plasma concentration of piroxicam after one single 20 mg tablet is 3-8 µmol/l (38). This shows that pirox-icam has no antiplatelet effects by itself in clinical settings but might well reduce the antiplatelet effects of aspirin, and this for a long time, considering it extremely long (and variable) half-life of about 50 h. This collectively suggests a clinically important but not well appreciated drug interaction.

We have studied many NSAIDs in this setting, including ibu-profen, naproxen, diclofenac and others. A similar interaction with aspirin was found with many of them. ▶ Figure 4 shows a com-parison of different NSAIDs with respect to their potential for in-

Figure 3: Pharmacodynamic interaction of aspirin and piroxicam in vitro. Platelet-rich plasma from healthy volunteers (n=7) was pre-incubated with piroxicam and aspirin (30 µmol/l) subsequently added. Five min later, platelets were stimulated (1 mmol/l arachidonic acid) and aggregation recorded for 4 min. Aggregation (left) and thromboxane formation (right) (Immu-noassay for TXB2) were abolished by aspirin alone (open vs. blue symbols). The inhibition by aspirin was prevented in a concentration- dependent manner by preincubation with pirox-icam (red symbols).

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity




teraction with aspirin, as determined by platelet TX formation in the presence of NSAID and aspirin. For standardisation, all NSAIDs were added at their respective IC50, which were deter-mined separately and are also given in ▶ Figure 4. They were com-parable or lower than clinically reported plasma concentrations of these NSAIDs. For example, the mean cmax at standard analgesic doses of acetaminophen, diclofenac, ketoprofen, naproxen, ibu-profen and dipyrone were 141, 2, 10, 205, 69 and 51 µmol/l, re-spectively (39-44). Hence, these NSAID/aspirin interactions can be expected also to occur clinically.

Notably, there were remarkable differences between the com-pounds, which cannot be explained solely by potency or COX-2 selectivity. Acetaminophen, diclofenac, and ketoprofen did not in-terfere with aspirin at all. Experiments are ongoing to further examine the structural requirements.

Non-aspirin NSAIDs with high COX-2 prevalence may also in-terfere with COX inhibition by aspirin. For example, pre-exposure of COX-1 in ram seminal vesicles (45) or cultured cell lines (46) with COX-2 selective inhibitors (DuP697, NS398) attenuated COX-1 inhibition by aspirin at nanomolar concentrations. Neither compound inhibited COX-1 activity when given alone. A similar interaction was described for indomethacin (instead of aspirin) but not for ibuprofen or naproxen. The authors explained this by interaction at different binding sites. Similar results were also re-ported by Ouellet et al., who suggested that less selective coxibs are more prone to interact (23). Nimesulide, a compound with moder-ate COX-2 selectivity, also interferes with aspirin (47). Interest-ingly, interactions with aspirin have also been shown for natural compounds, such as gallic acid (48).

Studies in healthy subjects

Several studies on healthy subjects have shown that NSAIDs can block or weaken the antiplatelet action of aspirin. For example, Ca-tella-Lawson et al. have found that ibuprofen interferes with as-pirin in terms of platelet aggregation and serum thromboxane formation if ibuprofen was given (i) once daily 2 h before aspirin

or (ii) by multiple daily dosing (49). Another small trial reported similar effects also for naproxen (50) which was confirmed in a larger randomised trial (51). Six NSAIDs have been studied by Gladding et al. in healthy subjects in a randomised placebo con-trolled design (52). The authors examined capillary closure times by the PFA-100 analyser and found that the prolongation of clo-sure time by aspirin was significantly antagonised by ibuprofen, indomethacin, naproxen and tiaprofenic acid, but not by sulindac and celecoxib. All the above studies noted that the NSAIDs inter-acted with aspirin at sub-inhibitory doses, which agrees with the experimental data summarised above.

A significant portion of patients (~75 %) subjected to CABG surgery may not respond to low-dose aspirin with the expected in-hibition of platelet function (53). A more detailed analysis for pos-sible pharmacological reasons of this HPR showed that blood samples from aspirin non-responders contained metabolites of a routinely administered analgesic (dipyrone or metamizol), while samples from responders did not. A subsequent in vitro study re-vealed that the active dipyrone metabolite, 4-metylaminoantipy-rine, very effectively inhibited aspirin-induced inhibition of arachidonic acid- or collagen-induced platelet aggregation, TXA2 formation and P-selectin expression (54). Analgesic doses of dipy-rone did not change platelet activity directly but prevented in-hibition by aspirin. The effect was specific since dipyrone did not interfere with platelet inhibition by TXA2 receptor antagonists or inhibitors of TX - synthase. In agreement with this, others re-ported that HPR after CABG surgery can be transferred by plasma to previously responsive platelets (55). ▶ Figure 5 exemplifies this interaction between dipyrone and aspirin in a healthy individual.

Studies in patients

Whether HPR against aspirin by NSAIDs also occurs in clinical settings is less clear, but some indications exist. For example, a small study identified 18 patients receiving aspirin for second-ary stroke prevention and concomitant therapy with ibuprofen or naproxen (56). At 27 months follow-up, all patients showed

Figure 4: NSAIDs largely differ in their interaction with aspirin at platelet COX-1. Platelets were pre-incubated with different non-aspirin NSAIDs, each at its respective IC50 (right, separately measured in platelets). Aspirin (30 µmol/l) was added for 5 min and platelets then stimulated with 1 mmol/l arachidonic acid. TXB2 was determined in the supernatants by immunoassay and is shown as % of control (no aspirin or NSAID). Acetaminophen, diclofenac, and ketoprofen did not prevent the antiplatelet effect of aspirin. All other NSAIDs (red bars) reduced the aspirin effect with large differences between the compounds. Bars represent mean values ± SEM (n=4-6) (Saxena et al., manuscript in preparation).

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity




largely normal arachidonic acid and collagen-stimulated whole blood platelet aggregation, similar to drug-free controls. Then, NSAIDs were either discontinued or aspirin taken at different time schedule (see section Sequential dosing) in order to mini-mise an interaction. Re-testing after 2-4 weeks revealed that this had largely restored platelet inhibition by aspirin. This suggests that an appropriate modification of therapy can indeed avoid HPR. In another study on patients with osteoarthritis celecoxib did not appear to interfere with the antiplatelet effect of low-dose aspirin (57), although some experimental studies support an interaction (23, 30).

A prejudice of the above studies is that they examined labora-tory parameters without proven prognostic relevance for clinical outcome. Although meta analyses (58, 59) and some studies in dif-ferent clinical settings (6, 8-10, 60) suggested that HPR to aspirin may be related to worse clinical outcome, there is also recent data in patients with stable angina suggesting that HPR does not relate to clinical outcome (61).

It is still a matter of debate whether and if so which platelet function assay has (the best) predictive value for clinical outcome in aspirin-treated patients. Two principal laboratory assays are available to examine platelet inhibition by aspirin ex vivo – the measurement of platelet function (e.g. aggregation, P-selectin ex-pression, secretion) and the determination of platelet-derived TXA2 formation. The analysis of platelet aspirin sensitivity is based on inhibition of platelet function after stimulation by a par-ticular agonist. Such assays may not capture the full antiplatelet re-sponse to aspirin and ignore indirect effects on coagulation and anti-inflammatory processes. Assay sensitivity to antiplatelet ef-fects of aspirin also depends on the agonist used. Arachidonic acid-induced aggregation in platelet rich plasma is probably the most aspirin-sensitive functional assay and well suitable to dem-onstrate the interaction between aspirin and NSAIDs in vitro (▶ Figure 3). Collagen, ADP and thrombin are less appropriate as agonist to study aspirin effects because they stimulate platelet re-sponses largely by thromboxane-independent pathways. Serum- or arachidonic acid-induced thromboxane formation in whole blood or platelet-rich plasma are also appropriate to examine as-pirin/NSAID interaction in vitro. Whole blood assays, such as ar-achidonic acid-induced impedance aggregometry (e.g. Multiplate Aspi test), Verify Now and the PFA-100 assay, are available as point-of-care assays for monitoring of aspirin antiplatelet effects. These assays have not been systematically tested for their ability to detect aspirin/NSAID interactions. While normal ranges have been established here, they appear to produce divergent clinical re-sults in patients (62). We have recently discussed the available test-

Figure 5: HTPR against aspirin after oral dipyrone (metamizol). Re-peated recordings of in vitro platelet aggregation (platelet-rich plasma, light transmission) and thromboxane (TX) formation at repeated times over a 17 day protocol in a subject who received low dose aspirin (starting at day 1, blue bar) and a 7-day protocol of dipyrone tablets (500 mg/day, starting at day 7, red bar). Dots mark platelet stimulation with arachidonic acid (1 mmol/l). Aspirin was taken in the morning (8 am) and dipyrone in the evening (8 pm).

Figure 6: Relation between compliance to aspirin, use of NSAIDs and risk of a first myocardial infarction (white). Insufficient adherence to aspirin intake (orange) as well as frequent co-medication of NSAIDs (red) antag-onised the protective effect of aspirin in aspirin-adherent subjects to a similar extent as in non-aspirin-adherent individuals. Figure is based on data from the Physician´s Health Study (68). Re-produced from (75) with permission by the auth-or.

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity




ing methods for antiplatelet effects of aspirin in more detail, in-cluding flow cytometry and analysis of urinary thromboxane metabolites (3).

The available data on clinical outcome in conjunction with HPR due to possible NSAID/aspirin interactions are also ambigu-ous. For example, two studies (63, 64) examined the association between NSAID and aspirin use according to the issued prescrip-tions in the UK General Practice Research Database. They found no indication for an adverse NSAID/aspirin interaction, but the results may have been biased by non-compliance and undocu-mented “over-the-counter” (OTC) use of aspirin and non-pre-scriptional analgesics, including ibuprofen and diclofenac among others.

Other trials, however, support the notion that NSAIDs may ad-versely interact with aspirin. For example, a case-control study with 1,055 cases of first non-fatal myocardial infarction and 4,153 controls showed that aspirin alone reduced cardiovascular events but was without favourable effect when given combined with NSAIDs (65). This applied particularly to ibuprofen and became even more evident with more frequent NSAID use. Also, the retro-spective Tayside trial on 7,107 patients with established cardiovas-cular disease observed an about two-fold increase in all-cause and cardiovascular mortality when low-dose aspirin was taken to-gether with ibuprofen, as opposed to aspirin alone (66). Interest-ingly, patients on diclofenac plus aspirin had mortality rates simi-lar to aspirin alone.

The Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET) randomised patients to ibuprofen, naproxen or lumiracoxib. It also included patients with high cardiovascular risk on low-dose aspirin. In a post-hoc substudy, patients on ibuprofen exhibited a higher cardiovascular event rate than those on lumir-acoxib (2.14 vs. 0.25 %, p=0.038), whereas event rates were similar for naproxen vs. lumiracoxib (67). This suggests that ibuprofen, but not naproxen, reduces cardiovascular prevention by aspirin. A similar conclusion can be drawn from a post-hoc analysis of the US Physician´s Health Study (PHS), a randomised controlled trial in 22,071 healthy physicians (68), which analysed whether co-ther-apy of low-dose aspirin and NSAIDs modified the effect of low-dose aspirin on cardiovascular outcome during the five-year treat-ment period. Participants taking NSAIDs together with aspirin for >60 days/year behaved similar to aspirin non-adherent subjects, i.e. missed any protection from myocardial infarction in contrast to those who regularly took the drug (▶ Figure 6) (69). Though the particular NSAIDs which were taken have not been identified, the data indicate that NSAID/aspirin interactions are clinically im-portant. The overall benefit of the total population was signifi-cantly driven by individuals who regularly took aspirin (every other day) and avoided regular intake of NSAIDs.

Thus, there is evidence that ibuprofen and other - but not all - NSAIDs may counteract the antiplatelet effects of aspirin in clini-cal practice. More detailed information is highly desirable but might be difficult to obtain from randomised prospective trials be-cause of ethical reasons.

Options for avoiding HPR against aspirin by NSAIDs

Since platelet inhibition by aspirin is irreversible and long-lasting, single doses of an interacting NSAID taken for acute pain relief are unlikely to interact with aspirin: In the original experiment shown in ▶ Figure 5, it took more than three days of regular co-medi-cation of dipyrone in a conventional analgesic dose until the re-covery of TXA2 formation was sufficient to allow for normal arachidonic acid-induced platelet aggregation.

Sequential dosing

A critical determinant of the interaction of NSAIDs with aspirin at platelet COX-1 is the actual NSAID plasma concentration. Since immediate release aspirin reaches cmax after about 1 h and is rapidly deacetylated to salicylic acid, NSAIDs, if necessary, should not be given within 2 h after aspirin intake. Accordingly, an appro-priate time interval between NSAID and aspirin intake might re-store platelet inhibition in patients with HPR against aspirin (56). It has been recommended to take a single dose of ibuprofen at least 30 min after immediate release aspirin and not less than 8 h before aspirin (37). In contrast, repeated daily dosing of short acting NSAIDs or single doses of long acting NSAIDs (e.g. piroxicam) may permanently interact with aspirin. Similar effects may also be expected for individuals with a poor metaboliser phenotype of NSAID-metabolising enzymes (e.g. CYP2C9) (70). Whether NSAIDs may also form ´active´ metabolites causing a longer du-ration of interaction with aspirin is not known.

NSAID selection

An interesting aspect is that NSAIDs differ in their interaction po-tential with aspirin (▶ Figure 4). Therefore, if combined therapy is required, NSAIDs should be selected that do not show this interac-tion. These include acetaminophen, diclofenac (49, 71) and others (52). Data on celecoxib are contradictory (30, 52, 57). Naproxen has been recommended for use in patients with cardiovascular dis-ease taking low-dose aspirin (72), but several studies found that it impairs platelet inhibition by aspirin (50-52). The ongoing PRECI-SION trial (Prospective Randomised Evaluation Of Celecoxib In-tegrated Safety vs Ibuprofen Or Naproxen), which compares chronic treatment with ibuprofen, naproxen or celecoxib with re-spect of cardiovascular outcome in patients with arthritis at elev-ated cardiovascular risk, will provide more information and will be completed in 2014.

The competitive mechanism of aspirin/NSAID interactions at platelet COX-1 suggests that both, reducing the dosage of NSAIDs as well as increasing the dose of aspirin, may also overcome this interaction. In human platelets, the prevention of aspirin effect by dipyrone could be restored in the presence of dipyrone by increas-ing the concentration of aspirin (54). A possible explanation may be the interaction kinetics of aspirin with COX-1: aspirin, like many non-aspirin NSAIDs, initially binds reversibly to Arg120 at the COX channel entrance before entering the acetylation site at of Ser530 (▶ Figure 1 C) (73). During this initial reversible step aspirin

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity




may become displaced by other NSAIDs in a competitive manner. No clinical data appear to exist on this issue.

Change of the antiplatelet compound

A final option for prevention of aspirin/NSAID interactions is to replace aspirin by another antiplatelet agent. One alternative is clopidogrel which was comparable to or modestly better than as-pirin for secondary prevention in the CAPRIE trial (74). So far, there is no clinical evidence supporting the replacement of aspirin by clopidogrel to avoid HPR by NSAIDs against aspirin. However, one should keep in mind that all patients in all clinical trials on dual antiplatelet therapy with P2Y12-ADP-receptor antagonists were on aspirin. This was no “placebo” treatment and the possible contribution of HPR to aspirin for the overall clinical outcome is unknown.

Conflicts of interestNone declared.

References1. Antithrombotic Trialist´s Collaboration. Collaborative meta-analysis of rando-

mised trials of antiplatelet therapy for prevention of death, myocardial infarc-tion, and stroke in high risk patients. Br Med J 2002; 324: 71-86.

2. Tantry US, Mahla E, Gurbel PA. Aspirin resistance. Prog Cardiovasc Dis 2009; 52: 141-152.

3. Schrör K, Huber K, Hohlfeld T. Functional testing methods for the antiplatelet effects of aspirin. Biomark Med 2011; 5: 31-42.

4. Grinstein J, Cannon CP. Aspirin Resistance: Current Status and Role of Tailored Therapy. Clin Cardiol 2012; epub ahead of print.

5. Rocca B, Petrucci G. Variability in the responsiveness to low-dose aspirin: phar-macological and disease-related mechanisms. Thrombosis 2012; epub ahead of print.

6. Gum PA, Kottke-Marchant K, Welsh PA, et al. A prospective, blinded determi-nation of the natural history of aspirin resistance among stable patients with cardiovascular disease. J Am Coll Cardiol 2003; 41: 961-965.

7. Chen WH, Cheng X, Lee PY, et al. Aspirin resistance and adverse clinical events in patients with coronary artery disease. Am J Med 2007; 120: 631-635.

8. Gori AM, Marcucci R, Migliorini A, et al. Incidence and clinical impact of dual nonresponsiveness to aspirin and clopidogrel in patients with drug-eluting stents. J Am Coll Cardiol 2008; 52: 734-739.

9. Pinto Slottow TL, Bonello L, Gavini R, et al. Prevalence of aspirin and clopido-grel resistance among patients with and without drug-eluting stent thrombosis. Am J Cardiol 2009; 104: 525-530.

10. Mannacio VA, Di Tommaso L, Antignano A, et al. Aspirin plus clopidogrel for optimal platelet inhibition following off-pump coronary artery bypass surgery: results from the CRYSSA (prevention of Coronary arteRY bypaSS occlusion after off-pump procedures) randomised study. Heart 2012; epub ahead of print.

11. Radovits BJ, Popa-Diaconu DA, Popa C, et al. Disease activity as a risk factor for myocardial infarction in Rheumatoid Arthritis. Ann Rheum Dis 2008; 68: 1271-1276.

12. Nadareishvili Z, Michaud K, Hallenbeck JM, et al. Cardiovascular, rheumato-logic, and pharmacologic predictors of stroke in patients with rheumatoid ar-thritis: a nested, case-control study. Arthritis Rheum 2008; 59: 1090-1096.

13. Zöller B, Li X, Sundquist J, et al. Risk of subsequent ischemic and haemorrhagic stroke in patients hospitalized for immune-mediated diseases: a nationwide fol-low-up study from Sweden. BMC Neurol 2012; 12: 41.

14. Hansson GK, Hermansson A. The immune system in atherosclerosis. Nat Im-munol 2011; 12: 204-212.

15. Frelinger AL, III, Furman MI, Linden MD, et al. Residual arachidonic acid-in-duced platelet activation via an adenosine diphosphate-dependent but cyclo-

oxygenase-1- and cyclooxygenase-2-independent pathway: a 700-patient study of aspirin resistance. Circulation 2006; 113: 2888-2896.

16. Di-Minno G, Silver MJ, Murphy S. Monitoring the entry of new platelets into the circulation after ingestion of aspirin. Blood 1983; 61: 1081-1085.

17. Reilly IA, FitzGerald GA. Inhibition of thromboxane formation in vivo and ex vivo: implications for therapy with platelet inhibitory drugs. Blood 1987; 69: 180-186.

18. Pedersen AK, FitzGerald GA. Dose-related kinetics of aspirin. Presystemic acetylation of platelet cyclooxygenase. N Engl J Med 1984; 311: 1206-1211.

19. DeWitt DL, el-Harith EA, Kraemer SA, et al. The aspirin and heme-binding sites of ovine and murine prostaglandin endoperoxide synthases. J Biol Chem 1990; 265: 5192-5198.

20. Loll PJ, Picot D, Garavito RM. The structural basis of aspirin activity inferred from the crystal structure of inactivated prostaglandin H2 synthase. Nat Struct Biol 1995; 2: 637-643.

21. Llorens O, Perez JJ, Palomer A, et al. Structural basis of the dynamic mechanism of ligand binding to cyclooxygenase. Bioorg Med Chem Lett 1999; 9: 2779-2784.

22. Selinsky BS, Gupta K, Sharkey CT, et al. Structural analysis of NSAID binding by prostaglandin H2 synthase: time-dependent and time-independent inhibitors elicit identical enzyme conformations. Biochemistry 2001; 40: 5172-5180.

23. Ouellet M, Riendeau D, Percival MD. A high level of cyclooxygenase-2 inhibitor selectivity is associated with a reduced interference of platelet cyclooxygenase-1 inactivation by aspirin. Proc Natl Acad Sci USA 2001; 98: 14583-14588.

24. Goltsov A, Maryashkin A, Swat M, et al. Kinetic modelling of NSAID action on COX-1: focus on in vitro/in vivo aspects and drug combinations. Eur J Pharm Sci 2009; 36: 122-136.

25. Hong Y, Gengo FM, Rainka MM, et al. Population pharmacodynamic model-ling of aspirin- and Ibuprofen-induced inhibition of platelet aggregation in healthy subjects. Clin Pharmacokinet 2008; 47: 129-137.

26. Sidhu RS, Lee JY, Yuan C, et al. Comparison of cyclooxygenase-1 crystal struc-tures: cross-talk between monomers comprising cyclooxygenase-1 homo-dimers. Biochemistry 2010; 49: 7069-7079.

27. Smith WL, Urade Y, Jakobsson PJ. Enzymes of the cyclooxygenase pathways of prostanoid biosynthesis. Chem Rev 2011; 111: 5821-5865.

28. Yuan C, Rieke CJ, Rimon G, et al. Partnering between monomers of cyclooxyge-nase-2 homodimers. Proc Natl Acad Sci USA 2006; 103: 6142-6147.

29. Prusakiewicz JJ, Duggan KC, Rouzer CA, et al. Differential sensitivity and mechanism of inhibition of COX-2 oxygenation of arachidonic acid and 2-arac-hidonoylglycerol by ibuprofen and mefenamic acid. Biochemistry 2009; 48: 7353-7355.

30. Rimon G, Sidhu RS, Lauver DA, et al. Coxibs interfere with the action of aspirin by binding tightly to one monomer of cyclooxygenase-1. Proc Natl Acad Sci USA 2009; 107: 28-33.

31. Vargaftig BB. The inhibition of cyclo-oxygenase of rabbit platelets by aspirin is prevented by salicylic acid and by phenanthrolines. Eur J Pharmacol 1978; 50: 231-241.

32. Humes JL, Winter CA, Sadowski SJ, et al. Multiple sites on prostaglandin cyclo-oxygenase are determinants in the action of nonsteroidal antiinflammatory agents. Proc Natl Acad Sci USA 1981; 78: 2053-2056.

33. Cerletti C, Livio M, De Gaetano G. Non-steroidal anti-inflammatory drugs react with two sites on platelet cyclo-oxygenase. Evidence from "in vivo" drug interaction studies in rats. Biochim Biophys Acta 1982; 714: 122-128.

34. Kaber G, Kaiser B, Baumgartel-Allekotte D, et al. Antagonism of the antithrom-botic and anti-atherosclerotic actions of aspirin by rofecoxib in the cholesterol-fed rabbit. Br J Pharmacol 2011; 164: 561-569.

35. Fosbol EL, Folke F, Jacobsen S, et al. Cause-specific cardiovascular risk associ-ated with nonsteroidal antiinflammatory drugs among healthy individuals. Circ Cardiovasc Qual Outcomes 2010; 3: 395-405.

36. Trelle S, Reichenbach S, Wandel S, et al. Cardiovascular safety of non-steroidal anti-inflammatory drugs: network meta-analysis. Br Med J 2011; 342:c7086.

37. Ellison J, Dager W. Recent FDA warning of the concomitant use of aspirin and ibuprofen and the effects on platelet aggregation. Prev Cardiol 2007; 10: 61-63.

38. Yuksel N, Karatas A, Ozkan Y, et al. Enhanced bioavailability of piroxicam using Gelucire 44/14 and labrasol: in vitro and in vivo evaluation. Eur J Pharm Bioph-arm 2003; 56: 453-459.

39. Critchley JA, Critchley LA, Anderson PJ, et al. Differences in the single-oral-dose pharmacokinetics and urinary excretion of paracetamol and its conjugates between Hong Kong Chinese and Caucasian subjects. J Clin Pharm Ther 2005; 30: 179-184.

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity




40. Raz I, Hussein Z, Samara E, et al. Comparative pharmacokinetic analysis of a novel sustained-release dosage form of diclofenac sodium in healthy subjects. Int J Clin Pharmacol Ther Toxicol 1988; 26: 246-248.

41. Rolf C, Engstrom B, Beauchard C, et al. Intra-articular absorption and distribu-tion of ketoprofen after topical plaster application and oral intake in 100 pa-tients undergoing knee arthroscopy. Rheumatology 1999; 38: 564-567.

42. Zhou D, Zhang Q, Lu W, et al. Single- and multiple-dose pharmacokinetic com-parison of a sustained-release tablet and conventional tablets of naproxen in healthy volunteers. J Clin Pharmacol 1998; 38: 625-629.

43. Pargal A, Kelkar MG, Nayak PJ. The effect of food on the bioavailability of ibu-profen and flurbiprofen from sustained release formulations. Biopharm Drug Dispos 1996; 17: 511-519.

44. Suarez-Kurtz G, Ribeiro FM, Estrela RC, et al. Limited-sampling strategy mod-els for estimating the pharmacokinetic parameters of 4-methylaminoantipyrine, an active metabolite of dipyrone. Braz J Med Biol Res 2001; 34: 1475-1485.

45. Rosenstock M, Danon A, Rubin M, et al. Prostaglandin H synthase-2 inhibitors interfere with prostaglandin H sythase-1 inhibition by nonsteroidal anti-inflam-matory drugs. Eur J Pharmacol 2001; 412: 101-108.

46. Rosenstock M, Danon A, Rimon G. PGHS-2 inhibitors, NS-398 and DuP-697, attenuate the inhibition of PGHS-1 by aspirin and indomethacin without alter-ing its activity. Biochim Biophys Acta 1999; 1440: 127-137.

47. Burde T, Rimon G. On the interaction of specific prostaglandin H synthase-2 inhibitors with prostaglandin H synthase-1. Eur J Pharmacol 2002; 453: 167-173.

48. Crescente M, Cerletti C, De Gaetano G. Gallic acid, a dietary polyphenolic com-ponent, blunts the inhibition of platelet COX-1 by aspirin: preliminary in-vitro findings. Thromb Haemost 2007; 97: 1054-1056.

49. Catella-Lawson F, Reilly MP, Kapoor SC, et al. Cyclooxygenase inhibitors and the antiplatelet effects of aspirin. N Engl J Med 2001; 345: 1809-1817.

50. Capone ML, Sciulli MG, Tacconelli S, et al. Pharmacodynamic interaction of naproxen with low-dose aspirin in healthy subjects. JACC 2005; 45: 1295-1301.

51. Anzellotti P, Capone ML, Jeyam A, et al. Low-dose naproxen interferes with the antiplatelet effects of aspirin in healthy subjects: recommendations to minimize the functional consequences. Arthritis Rheum 2011; 63: 850-859.

52. Gladding PA, Webster MW, Farrell HB, et al. The antiplatelet effect of six non-steroidal anti-inflammatory drugs and their pharmacodynamic interaction with aspirin in healthy volunteers. Am J Cardiol 2008; 101: 1060-1063.

53. Zimmermann N, Wenk A, Kim U, et al. Functional and biochemical evaluation of platelet aspirin resistance after coronary artery bypass surgery. Circulation 2003; 108: 542-547.

54. Hohlfeld T, Zimmermann N, Weber AA, et al. Pyrazolinone analgesics prevent the antiplatelet effect of aspirin and preserve human platelet thromboxane syn-thesis. J Thromb Haemost 2008; 6: 166-173.

55. Anger K, Kempfert J, Kostelka M, et al. Platelet hyporesponsiveness to acetylsali-cylic acid can be transferred by plasma in humans. Pharmacology 2010; 85: 241-247.

56. Gengo FM, Rubin L, Robson M, et al. Effects of ibuprofen on the magnitude and duration of aspirin's inhibition of platelet aggregation: clinical consequences in stroke prophylaxis. J Clin Pharmacol 2008; 48: 117-122.

57. Renda G, Tacconelli S, Capone ML, et al. Celecoxib, ibuprofen, and the antipla-telet effect of aspirin in patients with osteoarthritis and ischemic heart disease. Clin Pharmacol Ther 2006; 80: 264-274.

58. Snoep JD, Hovens MC, Eikelboom CJ, et al. Association of Laboratory-defined aspirin resistance with a higher risk of recurrent cardiovascular events. Arch In-tern Med 2007; 167: 1593-1599.

59. Krasopoulos G, Brister SJ, Beattie WS, et al. Aspirin "resistance" and risk of car-diovascular morbidity: systematic review and meta-analysis. Br Med J 2008; 336: 195-198.

60. Niccoli G, Giubilato S, Leo A, et al. Predictors of thromboxane levels in patients with non-ST-elevation acute coronary syndromes on chronic aspirin therapy. Thromb Haemost 2012; 108: 133-139.

61. Pettersen A-AR, Seljeflot I, Abdelnoor M, et al. High on-aspirin platelet reactiv-ity and clinical outcome in patients with stable coronary artery disease: Results from ASCET (Aspirin Nonresponsiveness and Clopidogrel Endpoint Trial) . J Am Heart Ass 2012; epub ahead of print.

62. Gurbel PA, Bliden KP, DiChiara J, et al. Evaluation of dose-related effects of as-pirin on platelet function: results from the Aspirin-Induced Platelet Effect (AS-PECT) study. Circulation 2007; 115: 3156-3164.

63. Garcia Rodriguez LA, Varas-Lorenzo C, Maguire A, et al. Nonsteroidal antiin-flammatory drugs and the risk of myocardial infarction in the general popu-lation. Circulation 2004; 109: 3000-3006.

64. Fischer LM, Schlienger RG, Matter CM, et al. Current use of nonsteroidal anti-inflammatory drugs and the risk of acute myocardial infarction. Pharmacother-apy 2005; 25: 503-510.

65. Kimmel SE, Berlin JA, Reilly M, et al. The effects of nonselective non-aspirin non-steroidal anti-inflammatory medications on the risk of nonfatal myocardial infarction and their interaction with aspirin. J Am Coll Cardiol 2004; 43: 985-990.

66. MacDonald TM, Wei L. Effect of ibuprofen on cardioprotective effect of aspirin. Lancet 2003; 361: 573-574.

67. Farkouh ME, Greenberg JD, Jeger RV, et al. Cardiovascular outcomes in high risk patients with osteoarthritis treated with ibuprofen, naproxen or lumiracox-ib. Ann Rheum Dis 2007; 66: 764-770.

68. Hennekens CH, Schneider WR, Hebert PR, et al. Hypothesis formulation from subgroup analyses: nonadherence or nonsteroidal anti-inflammatory drug use explains the lack of clinical benefit of aspirin on first myocardial infarction at-tributed to "aspirin resistance". Am Heart J 2010; 159: 744-748.

69. Ridker PM, Manson JE, Gaziano JM, et al. Low-dose aspirin therapy for chronic stable angina. A randomized, placebo-controlled clinical trial. Ann Intern Med 1991; 114: 835-839.

70. Rodrigues AD. Impact of CYP2C9 genotype on pharmacokinetics: are all cyclo-oxygenase inhibitors the same? Drug Metab Dispos 2005; 33: 1567-1575.

71. Schuijt MP, Huntjens-Fleuren HW, de Metz M, et al. The interaction of ibu-profen and diclofenac with aspirin in healthy volunteers. Br J Pharmacol 2009; 157: 931-934.

72. Mackenzie IS, Coughtrie MW, MacDonald TM, et al. Antiplatelet drug interac-tions. J Intern Med 2010; 268: 516-529.

73. Mancini JA, Riendeau D, Falgueyret JP, et al. Arginine 120 of prostaglandin G/H synthase-1 is required for the inhibition by nonsteroidal anti-inflammatory drugs containing a carboxylic acid moiety. J Biol Chem 1995; 270: 29372-29377.

74. CAPRIE Steering Committee. A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet 1996; 348: 1329-1339.

75. Schrör K. Acetylsalicylsäure. 2 ed. Frechen: Dr. Schrör Verlag, 2011.

Hig

h on

-tre

atm

ent

plat

elet

rea

ctiv

ity


High on treatment platelet reactivity against aspirin by ... · © Schattauer 2013 Thrombosis and Haemostasis 109.5/2013 ... anti-inflammatory drugs pharmacological mechanisms and

Documents