Association of Aspirin Dosage to Clinical Outcomes after PCI: Observations from the Ottawa Heart Institute PCI Registry
- Volume 21 - Issue 3 - March, 2009
- Posted on: 3/10/09
- 0 Comments
- 8888 reads
From the University of Ottawa Heart Institute, Ottawa, Ontario and *Harvard School of Public Health, Boston, Massachusetts.
Disclosures: Michel Le May has received speaker honoraria from and is a consultant to Sanofi and has received research grants from Schering. Derek So has received speaker honoraria from Sanofi-Aventis and travel grants from Astra-Zeneca.
Manuscript submitted July 15, 2008, provisional acceptance given August 19, 2008, manuscript accepted November 10, 2008.
Address for correspondence: Derek So, MD, FRCPC, University of Ottawa Heart Institute, Cardiology, 40 Ruskin Street, Ottawa, Ontario K1Y 4W7, Canada. E-mail: email@example.com
ABSTRACT: Background. Dual antiplatelet therapy, with aspirin and a thienopyridine, is the accepted treatment after percutaneous coronary intervention (PCI). No clear evidence exists regarding the ideal dosage of aspirin. Recent guidelines recommend higher-dose aspirin because of the possible decrease in stent thrombosis. The purpose of this study was to test the hypothesis that high-dose aspirin of 325 mg decreases death and myocardial infarction (MI) compared to a lower dose of 81 mg in patients undergoing PCI. Methods. An observational cohort study of 1,840 consecutive patients who underwent PCI was conducted. Patients who did not survive to discharge were excluded. The primary endpoint was a composite of all-cause mortality and MI at 1 year. Results. Nine-hundred and thirty patients (50.5%) were discharged on 325 mg of aspirin and 910 (49.5%) were discharged on 81 mg. The risk of all-cause mortality or MI was not significantly different between patients: low-dose 5.49% (50/910) vs. high-dose 4.19% (39/930); adjusted odds ratio [OR], 1.16; 95% confidence interval [CI], 0.73–1.85). In a multivariable analysis, the Charlson comorbidity score (OR, 1.37; 95% CI, 1.18–1.58) and urgent PCI (OR, 1.75; 95% CI, 1.03–3.00) were associated with increased death or MI. Among patients with drug-eluting stents, the use of low-dose aspirin did not predispose them to death or MI (adjusted OR, 1.12, 95% CI, 0.53–2.34). Conclusions. In this large contemporary analysis of PCI patients, no differences in death or MI were observed at 1 year between patients discharged on low-dose aspirin 81 mg compared to patients on a higher dose.
J INVASIVE CARDIOL 2009;21:121–127
Key words: aspirin dosage, percutaneous coronary intervention
Percutaneous coronary intervention (PCI) is the most common revascularization procedure, with over 1 million performed each year worldwide.1 Despite advances in PCI technology, it is well recognized that the utilization of appropriate adjunctive pharmacotherapy is imperative to long-term success after PCI.2
Antiplatelet therapy is the cornerstone in adjunctive PCI pharmacotherapy, with clinical efficacy proven from the peri-PCI period and extending to 1 year.3,4 Aspirin has been universally accepted as therapy in PCI. Its use was first demonstrated in the balloon angioplasty era, where it was shown to reduce acute ischemic outcomes.5,6 Although aspirin has been universally accepted as essential therapy for patients undergoing PCI, its optimal dosing has not been extensively studied in this setting. Prior studies of aspirin dosing in various cardiovascular disorders revealed benefits of low-dose aspirin of 75–160 mg in preventing adverse clinical outcomes.7–10 Low doses may be advantageous in decreasing side effects such as gastrointestinal bleeding.11 None of these studies were based on patients after PCI. Many contemporary studies have focused on the dosing and timing of the antiplatelet drug clopidogrel, an adenosine diphosphate (ADP)-receptor antagonist. These studies have demonstrated the important impact of clopidogrel on clinical outcomes in patients undergoing PCI.3,4,12,13
Recently, the American College of Cardiology (ACC)/American Heart Association (AHA) released an update on guidelines for patients undergoing PCI, which emphasized the importance of at least 162–325 mg of aspirin for 1 month in patients receiving bare-metal stents (BMS), instead of the often prescribed lower doses.14 The guidelines further suggested 3 months and 6 months of higher-dose aspirin therapy for those receiving drug-eluting stents (DES) coated with sirolimus and paclitaxel, respectively. The evidence for recommendation was based on the dosage and duration of therapy of aspirin in several large DES trials15,16 and not on specific studies looking at dosage and duration of antiplatelet treatment. In fact, several smaller DES trials did use lower doses of ≥ 75 mg, but this did not alter the recommendations of the guidelines.17–21 Concerns of an increased risk of subacute stent thrombosis because of the delayed endothelialization may have prompted these recommendations.22,23 Although concurrent ADP-receptor antagonist use has been suggested to be the more important factor in preventing stent thrombosis, the dosage of aspirin may also play an integral role.
In our analysis, 1,840 patients were followed for a 1-year period after the index PCI procedure to document the occurrence of death, myocardial infarction (MI) and repeat target vessel revascularization. We hypothesized that patients who received a low dose (81 mg/day) of aspirin would have more cardiac adverse events compared to those receiving a higher dose of 325 mg.
Setting and design. The Ottawa Heart Institute PCI Registry is a prospective registry of all patients undergoing PCI at the University of Ottawa Heart Institute, a tertiary cardiac center with a referral base of 30 peripheral hospitals servicing a population of over 1 million. The present analysis included patients who underwent PCI from December 1, 2003 to November 30, 2004. Patients were identified retrospectively from the registry and were then followed prospectively over a 1-year period. The study was approved by the Human Research Ethics Review Board of the University of Ottawa Heart Institute (Ottawa, Ontario, Canada).
Data collection. Trained cardiac registered nurses collected data from hospital charts of patients who underwent PCI during the study period. One year after the index PCI procedure, a telephone interview was conducted. If admission to the hospital for cardiac symptomatology occurred within 1 year of PCI, the records from the admitting hospital were obtained if consent was given. For patients with repeat angiography or PCI, subsequent procedural reports and in-hospital charts were evaluated for the presence of outcomes. For patients with > 1 PCI in the study period, the first procedure was counted as the entry date into the study. Explicit definitions for data elements were predetermined. Recurrent MI was defined as ischemic symptoms with recurrent elevation of ST-segments in ≥ 2 contiguous leads or the elevation of cardiac biomarkers to > 2 times the upper limit of normal. Target vessel revascularization (TVR) was defined as unplanned repeat coronary revascularization by PCI or coronary artery bypass graft surgery (CABG) involving the artery of the index PCI procedure. The Charlson score, a validated measure of comorbidity, was calculated for each patient.24 Urgent PCI was defined as any PCI performed on in-hospital patients. Stent thrombosis included cases confirmed by angiography and autopsy up to 1 year after the index PCI procedure (ARC-definite).
The dose of aspirin prescribed was at the operator’s discretion post PCI. Patients were divided into two groups based on dosage of aspirin at the time of hospital discharge. Those with a dose of ≥ 325 mg per day of aspirin were deemed to be in the high-dose group, while those with < 325 mg per day were in the low-dose group.
Patients were excluded from the study if they were referred for PCI, but no coronary intervention was performed, or if they underwent a failed PCI followed by emergent CABG or if they died during their index hospital admission. Patients with an aspirin allergy or who were not discharged on aspirin were also excluded from the analysis.
Outcome measures. The prospectively chosen primary outcome was a combined endpoint of all-cause mortality and MI at 1 year. Secondary outcomes included a composite of all-cause mortality, MI and repeat revascularization at 1 year, as well as each of the individual components of the combined endpoint. For patients with > 1 MI or repeat revascularization, only the first event was counted as an endpoint. Planned staged PCI procedures were not considered as repeat revascularization.
Statistical analysis. Statistical analysis for the study was conducted using SAS software, version 9.1 (SAS Institute, Inc., Cary, North Carolina). The primary hypothesis that those with low-dose aspirin would have a higher event rate for the primary outcome compared to those with high-dose aspirin was tested using Fisher’s exact test with adjusted odds ratios (OR). All p-values were reported as two-tailed with an accepted significance of ≤ 0.05. Odds ratios were reported with 95% confidence intervals. Characteristics between the 81 mg and 325 mg groups were compared using the Fisher’s exact test for binary variables and x2 test for other categorical variables. For continuous variables, the student’s t-test was used for normally distributed variables and the Wilcoxon’s Rank Sum test for non-normal data.
A logistic regression model to determine predictors for the primary outcome was created. Covariates included in the model were: dosage of aspirin, age, gender, diabetes mellitus, hypercholesterolemia, hypertension, smoking, family history of premature coronary artery disease, history of MI, previous PCI, history of CHF, history of transient ischemic accident or stroke, nonelective PCI, acute coronary syndromes, concurrent lipid-lowering therapy, use of DES, multivessel coronary disease and the Charlson comorbidity score. Multiple separate logistic regression models were constructed using the above variables to determine the impact of low-dose aspirin in several subgroups of interest.
An additional analysis was conducted by the creation of a propensity score to predict the probability of receiving a low aspirin dose (a propensity score) from a logistic regression model as a function of the clinical variables in the original logistic model. This analysis approach allows comparisons of treatment groups in observational studies where selection bias may exist, which predisposes patient allocation into one treatment group over another. Conditioning on the propensity score in an adjusted analysis provides an alternative mechanism for controlling confounding variables. A stratified analysis using the Mantel-Haenszel weighted formula was performed with subjects assigned to strata defined by subranges of the propensity score.