Methods

We retrospectively analyzed 3,637 consecutive patients who had undergone percutaneous coronary intervention (PCI) at 2 institutions in the northern regions of Italy (Ospedali Riuniti di Bergamo and Ospedale Carlo Poma, Mantova) from June 2005 to March 2008. Of these, 1,379 patients (37.9%) were successfully treated with DES (707 with sirolimus-eluting stents and 672 paclitaxel-eluting stents) and constituted the study population for the present analysis. All patients were pretreated with aspirin and clopidogrel. Glycoprotein IIb/IIIa inhibitors were administered at the physician’s discretion. The patients were discharged with an indication to continue with dual antiplatelet therapy with aspirin (100 mg/day) and clopidogrel (75 mg/day) for 12 months. Aspirin was to be maintained lifelong.

Follow-up by telephone interview or outpatient clinical visits were scheduled at 1, 6, 12, 18, 24, and 36 months after the index procedure. A standard questionnaire was used to collect information on the medications the patients were taking and any adverse events. If aspirin and/or clopidogrel had been discontinued, the cause and timing when these occurred were assessed and recorded. “Early” discontinuation was defined as withdrawal of aspirin and/or clopidogrel within the first 12 months, and “late” discontinuation consisted of aspirin withdrawal after 12 months. Whenever adverse cardiac events and/or major bleeding events were reported, the patients’ charts were reviewed.

Major adverse cardiac events (MACE) were defined as death, myocardial infarction (MI), destabilizing symptoms leading to hospitalization, and nonfatal stroke. Death was considered cardiac in origin unless obvious noncardiac causes could be identified. Sudden death was defined as unexplained death in previously stable patients. MI was diagnosed if any troponin elevation with symptoms suggestive for acute coronary syndrome was detected. The presence of new pathologic Q waves on the electrocardiogram was also diagnosed as MI. Within 1 week of the index procedure, only Q-wave MI was adjudicated as MI. Peri- PCI MI was defined by a new Q wave lasting >0.03 seconds in 2 contiguous electrocardiographic leads or elevations in creatine kinase and the MB fraction of creatine kinase, including an increase in the MB fraction of creatine kinase level that was ≥3 times the local upper limit of the normal range and, when biomarkers were elevated before PCI, an additional 50% greater than baseline. The diagnosis of destabilizing symptoms was defined as the occurrence of ischemic symptoms requiring hospitalization without any biochemical evidence of myocardial necrosis. Stroke was defined as an ischemic cerebral infarction caused by an embolic or thrombotic occlusion of a major intracranial artery. ST was defined according to the Academic Research Consortium Definition. Not only sudden death, but also those deaths without enough information to exclude sudden death were regarded as possible ST. Bleeding complications were classified as major or minor according to the Thrombolysis In Myocardial Infarction criteria. Major bleeding included any intracranial bleeding or any bleeding associated with clinically overt signs associated with a decrease in hemoglobin of >5 g/dl. Minor bleeding was defined as any clinically overt sign of bleeding (including observation using imaging techniques) associated with a decrease in hemoglobin of ≥3 to ≤5 g/dl.

Continuous variables are presented as the mean ± SD or as the median and interquartile range and were compared using Student’s unpaired t test or the Mann-Whitney rank sum test, as appropriate. Categorical variables are presented as counts and percentages and were compared using the chi-square test, as appropriate (expected frequency >5). Otherwise, Fisher’s exact test was used. Patients lost to follow-up were considered at risk until the date of the last follow-up interview, at which point they were censored. Multivariate logistic regression models were applied to identify predictors of antiplatelet therapy discontinuation. The results are reported as the odds ratio and 95% confidence interval. Variables that were candidates for entry into the models included the sociodemographic and clinical factors listed in Table 1 .

The association of antiplatelet therapy discontinuation (early and late) with subsequent clinical outcomes was explored. The predictors of MACE, ST, and mortality were identified using multivariate logistic regression analysis. Survival curves were generated using the Kaplan-Meier method, and the log-rank test was used to evaluate differences between groups. Crude event rates were estimated using the Kaplan-Meier method. The MACE, ST, and death rates were investigated according to the status of aspirin and clopidogrel therapy. The analyses were stratified by the interval after the index PCI (i.e., within 30 days, 31 to 180 days, 181 to 365 days, and after 365 days) in accordance with previous reports. Receiver operating characteristic analysis was performed for an exploratory evaluation of the best cutoff point of antiplatelet discontinuation duration to predict MACE, ST, and death in our study population. The sensitivity, specificity, and positive and negative predictive values were derived using this cutoff value. Significance was set at the 2-tailed 0.05 level. Computations were performed using the Statistical Package for Social Sciences, version 15.0 (SPSS, Chicago, IL).

Results

Of the overall study population (n = 1,379), complete follow-up data were available for 1,358 patients (98.5%). The remaining patients were excluded from the present analysis, assuming a nonsignificant difference with regard to the distribution of clinical events. Of the patients who discontinued one or both antiplatelet agents, early (<12 months) and late (>12 months) discontinuation occurred in 119 (8.8%) and 65 (4.8%) patients, respectively. The baseline demographics, clinical characteristics, and discharge data of patients with (n = 184) and without (n = 1,174) antiplatelet discontinuation are listed in Table 1 . The trends in antiplatelet discontinuation over time are shown in Figure 1 .

Trends in antiplatelet discontinuation over time. Estimates of cumulative incidence were obtained with the Kaplan-Meier method.

The causes of antiplatelet therapy discontinuation are summarized in Table 2 . Bleeding events represented the most frequent cause of early discontinuation and the second-most frequent cause of late discontinuation. Surgery was the leading cause of late discontinuation and the second-most frequent cause of early discontinuation. The predictors of antiplatelet therapy discontinuation are summarized in Table 3 . Early discontinuation of one or both antiplatelet agents within the first year was predicted by in-hospital major bleeding, oral anticoagulant use at discharge, and the lack of statin prescription. A history of previous stroke was the only independent predictor of late discontinuation. During the overall study period, the discontinuation of aspirin was predicted by the prescription of oral anticoagulants at discharge. In contrast, diabetes mellitus, serum creatinine at admission, statins at discharge, and oral anticoagulants at discharge predicted thienopyridine discontinuation.

The cumulative incidence of MACE and ST was 14.9%, and 3.8%, respectively. Patients with early discontinuation of antiplatelet therapy experienced a greater incidence of MACE (28.6% vs 13.7%, p <0.001) and ST (7.6% vs 3.4%, p = 0.038) compared to those who did not ( Table 4 ). Overall mortality (13.4% vs 4.7%, p <0.001) and cardiovascular death (5% vs 1.2%, p = 0.007) were significantly more frequent among patients with early discontinuation. Also, the incidence of MI and destabilizing symptoms requiring hospitalization was greater, albeit not statistically significant, in patients with early discontinuation. In patients with late discontinuation, a nonstatistically significant increase in MACE and ST and a significantly greater incidence of MI and nonfatal stroke were seen. No significant differences were found in cardiovascular death and overall death between patients with and without late discontinuation.

Table 4

Short and long-term events of patients with versus without antiplatelet therapy discontinuation

	Early Discontinuation of Dual Antiplatelet Therapy			Late Discontinuation of Aspirin
	No (n = 1,239)	Yes (n = 119)	p Value	No (n = 1,174)	Yes (n = 65)	p Value
In-hospital major adverse cardiac events	28 (2.3%)	3 (2.5%)	0.749	25 (2.1%)	3 (4.6%)	0.19
In-hospital major bleeding	10 (0.8%)	7 (6.0%)	<0.001	9 (0.8%)	1 (1.7%)	0.469
In-hospital minor bleeding	40 (3.4%)	11 (9.5%)	0.004	37 (3.3%)	3 (5.0%)	0.467
Cumulative major adverse cardiac events	169 (13.7%)	34 (28.6%)	<0.001	156 (13.3%)	13 (20.0%)	0.128
Any death	58 (4.7%)	16 (13.4%)	<0.001	57 (4.9%)	1 (1.5%)	0.361
Myocardial infarction	52 (4.2%)	9 (7.6%)	0.09	46 (3.9%)	6 (9.2%)	0.05
Unstable angina leading to hospitalization	51 (4.1%)	9 (7.6%)	0.08	48 (4.1%)	3 (4.6%)	0.747
Nonfatal stroke	8 (0.6%)	0 (0%)	1.00	5 (0.4%)	3 (4.6%)	0.006
Cumulative death	58 (4.7%)	16 (13.4%)	<0.001	57 (4.9%)	1 (1.5%)	0.361
Cumulative cardiac death	15 (1.2%)	6 (5.0%)	0.007	15 (1.3%)	0 (0%)	1.00
Cumulative major bleeding	23 (1.9%)	22 (18.6%)	<0.001	15 (1.3%)	8 (13.1%)	<0.001
Cumulative minor bleeding	73 (6.1%)	34 (29.1%)	<0.001	59 (5.2%)	14 (22.6%)	<0.001
Cumulative definite stent thrombosis	31 (2.8%)	8 (6.7%)	0.027	27 (2.3%)	4 (6.2%)	0.075
Cumulative definite or probable stent thrombosis	32 (2.8%)	8 (6.7%)	0.048	28 (2.4%)	4 (6.2%)	0.083
Cumulative definite, probable or possible stent thrombosis	42 (3.4%)	9 (7.6%)	0.038	38 (3.2%)	4 (6.2%)	0.275

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Relation Between the Incidence, Location, and Extent of Thoracic Aortic Atherosclerosis Detected by Transesophageal Echocardiography and the Extent of Coronary Artery Disease by Angiography
2. Comparison of Impact of Mortality Risk on the Survival Benefit of Primary Percutaneous Coronary Intervention Versus Facilitated Percutaneous Coronary Intervention
3. Different Clinical Outcome of Paravalvular Leakage After Aortic or Mitral Valve Replacement
4. Effect of Pre-Eclampsia–Eclampsia on Major Cardiovascular Events Among Peripartum Women in Taiwan

Stay updated, free articles. Join our Telegram channel

Join