RESEARCH
Aspirin Resistance and Compliance with Therapy Jesse Dawson,1 Terry Quinn,1 Mark Rafferty,1 Peter Higgins,1 Gautamanda Ray,2 Kennedy R. Lees1 & Matthew R. Walters1 1 Acute Stroke Unit, Division of Cardiovascular and Medical Sciences, Faculty of Medicine, University of Glasgow, Western Infirmary, Glasgow, UK 2 Acute Medicine Unit, Emergency Care Medical Services Directorate, Royal Alexandria Hospital, Paisley, UK
Keywords Antiplatelet; Aspirin; Compliance; Resistance; Stroke. Correspondence Jesse Dawson, M.D., Acute Stroke Unit, Division of Cardiovascular and Medical Sciences, Faculty of Medicine, University of Glasgow, Western Infirmary, Dumbarton Road, Glasgow, G11 6NT, UK. Tel.: +44-141-2116395; Fax: +44-141-2111863; E-mail:
[email protected]
doi: 10.1111/j.1755-5922.2010.00188.x
SUMMARY Introduction: Aspirin resistance is associated with increased cardiovascular risk in aspirin-treated patients. Poor compliance may explain many cases of “resistance,” yet few clinical studies have used objective measurement of therapy compliance. We did so in a case-controlled study. Methods: We enrolled patients within 24 h of ischemic stroke and a group of controls taking aspirin who had never suffered a vascular event on therapy. All claimed to be compliant. We assessed platelet function using platelet function analyser (PFA)-100 and rapid platelet function analyser (RPFA) devices, applying standard definitions of resistance. We used high-performance liquid chromatography for levels of aspirin metabolites in the urine to confirm compliance with therapy. We compared rates of resistance in stroke patients and controls, and performed subgroup analysis restricted to patients with objective confirmation of recent aspirin ingestion. Results: We recruited 90 cases and 90 controls. Complete platelet function tests were available in 177. Resistance rates seen in cases and controls, respectively, were: resistance on one or more test, 30 (34%) versus 21 (25%), P = 0.19; on PFA-100 testing only, 28 (32%) versus 15 (18%), P = 0.031; on RPFA testing only, 16 (18%) versus 12 (14%), P = 0.54; resistance on both tests, 12 (14%) versus 5 (6%), P = 0.037. When only patients with objective evidence of recent aspirin ingestion were considered (n = 71), rates were similar regardless of definition of resistance used. Conclusion: Aspirin resistance is common but poor compliance accounted for nearly half of cases of apparent aspirin “failure.” Objective measures to assess compliance are essential in studies of aspirin resistance.
Introduction Aspirin is the most commonly prescribed antiplatelet agent for secondary prevention after stroke and prevents up to a fifth of recurrent strokes [1]. It is now used alongside dipyridamole in the poststroke period [2], although clopidogrel monotherapy [3] is a valid alternative [4]. However, regardless of the chosen antiplatelet therapy, patients remain at risk of further cardiovascular events. It is increasingly claimed that response to antiplatelet agents varies and that some patients have a degree of “aspirin resistance.” Recent meta-analysis revealed resistance prevalence of 28% [5], with a similar rate seen in patients with a history of stroke [6]. Recent meta-analyses
c 2010 Blackwell Publishing Ltd Cardiovascular Therapeutics 29 (2011) 301–307
[5,7] also suggest resistant patients have a near 4-fold increase in risk of suffering a vascular event compared with aspirin responders. Similar findings have been seen in studies specific to the poststroke period [8,9]. Many studies will rely upon patient reports of compliance with aspirin therapy, which may not be accurate. Poor compliance predisposes to vascular events and [10] is common: perhaps as many as 10–40% of patients do not take prescribed antiplatelet tablets in the secondary prevention setting [11–15]. Compliance with therapy must therefore be adequately accounted for in studies of aspirin resistance and particularly in the acute setting where accurate reports may be more difficult to obtain.
301
J. Dawson et al.
Aspirin Resistance and Compliance with Therapy
We performed a prospective case-control study in patients with acute stroke who claimed good compliance with aspirin. We assessed rates of aspirin resistance in the group as a whole and then in a subgroup who had objective evidence of recent aspirin ingestion. Our aim was to establish whether resistance rates were higher in those with recent stroke and whether any difference was due to poor compliance with therapy.
Methods The study was performed in the University Division of Cardiovascular and Medical Sciences at the Western Infirmary, Glasgow. The study was approved by the West Medical Research Ethics Committee. All participants had a single baseline assessment.
Study Population Cases were aged >18 years and they (or their next of kin) reported compliance with prescribed aspirin therapy and had a clinical diagnosis of ischemic stroke or transient ischemic attack. We required that recruitment and platelet function analysis be performed within 24 h of onset of symptoms and that participants would be able to submit a urine sample. Patients with known poor therapy compliance, those taking clopidogrel or anticoagulant therapy or who had received thrombolytic therapy within the past 2 weeks and those with brain imaging, which suggested intracerebral hemorrhage were excluded. Compliance was assessed by direct questioning. Cases were identified on admission to the Acute Stroke Unit and gave full written informed consent or witnessed oral consent. Where potential participants were unable to consent, assent from the nearest relative or welfare guardian was accepted. Controls comprised individuals aged over 18 years of age who reported compliance with prescribed aspirin therapy. Treatment duration was at least 1 year and they must never have suffered a cardiovascular event on aspirin. Exclusion criteria were the same as for cases (with the exception of the need for brain imaging). Compliance was assessed by direct questioning. All gave full written informed consent. Controls were mostly identified during outpatient attendance at cardiovascular risk factor clinics at the Western Infirmary.
Measurement of Platelet Function and Aspirin Responsiveness Approximately 12 mL of blood was drawn using a vacutainer system and a 21 G needle into 3.8% citrate
302
blood tubes. Well-validated point-of-care cartridge-based platelet analyzers (PFA-100, Dade-Behring, Miami, FL [16,17] and Verify Now, Accumetrics [18,19]) were used to assess platelet function. Samples were analyzed at between 30 min and 4 h of being taken. Further details of these techniques are described elsewhere [16–19]. Using the PFA-100 analyzer, aspirin resistance was defined as a closure time below the in-house upper limit of the normal range for healthy non-aspirin-treated individuals (550 aspirin reaction units (Accumetrics Inc. 2002).
Confirmation of Recent Aspirin Ingestion Participants submitted a single urine sample (the next urine sample as soon as possible following recruitment and before ingestion of further antiplatelet treatment). We determined urine concentrations of salicylic acid and salicyluric acid levels using an established reversephase high performance liquid chromatography (HPLC) method, the details of which have been described elsewhere [20]. Limited data exist to inform levels of salicyluric acid or salicylic acid at intervals after aspirin ingestion or in patients not taking aspirin. In a small sample (n = 12) in our laboratory, levels of 0 to 5.25 μg/mL of salicyluric acid have been seen in patients taking no aspirin whereas levels of >5 μg/mL were seen 24 h after a single ingestion of 75 mg aspirin. These results are broadly similar to those outlined by others [21] and we therefore used a salicyluric acid of >5 μg/mL to represent evidence of recent aspirin ingestion.
Statistical Analysis We planned to study 90 cases and 90 controls. Our primary endpoint was the proportion of individuals showing resistance to aspirin on any test. Secondary endpoints were resistance defined as resistance to both tests, to the PFA-100 test and to the RPFA test and we also assessed agreement between the different platelet function tests. We compared the rates of primary and secondary endpoints between cases and controls in the whole group and then only in those with objective evidence of recent aspirin ingestion. We estimated that the rate of aspirin resistance would be approximately 30% in the whole group but that it would be higher, at approximately 40%, in cases [8] and lower in controls. We estimated that approximately 20% would have no evidence of recent aspirin ingestion [11–15]. After exclusion of these patients from the secondary analyses, 90 patients per group would still allow detection of a ∼20% difference in aspirin resis-
c 2010 Blackwell Publishing Ltd Cardiovascular Therapeutics 29 (2011) 301–307
J. Dawson et al.
tance rate between cases and controls with 80% power and a significance of 0.05 (for example 40% in cases vs. 20% in controls). Differences between cases and controls were assessed by either the Student’s paired t-test or nonparametric equivalent for continuous variables and the McNemar [chi-square] or Fisher’s exact test for proportions. Agreement between the platelet function tests was assessed using kappa statistics. A P-value of
