Bare-Metal vs. Drug-Eluting Stents in Patients With Atrial Fibrillation Undergoing Percutaneous Coronary Intervention

– Insights From the AFCAS Registry –

Abstract

Background: We explored 12-month clinical outcomes of 929 patients with atrial fibrillation (AF) undergoing percutaneous coronary intervention (PCI) with bare-metal stents (BMS) vs. drug-eluting stents (DES) from the prospective multicenter AFCAS (Atrial Fibrillation undergoing Coronary Artery Stenting) registry.

Methods and Results: Endpoints included the first occurrence of major adverse cardiac and cerebrovascular events (MACCE), defined as a composite of all-cause death, myocardial infarction (MI), target vessel revascularization, definite/probable stent thrombosis (ST), transient ischemic attack or stroke. Bleeding events were defined according to the Bleeding Academic Research Consortium criteria. Altogether, 673 (72.4%) patients received BMS and 220 (23.7%) at least one DES. Patients treated with DES more often had diabetes and prior ischemic events, and a longer stent length (P<0.05 for all), whereas patients treated with BMS more often had heart failure and were more likely to present with acute ST-elevation MI (P<0.05 for both). At 12-month follow-up, rates and risks of MACCE and total bleeding events were comparable between the groups (22.0% with BMS vs. 19.5% with DES, P=0.51, hazard ratio (HR) 0.89, 95% confidence interval (CI) 0.63–1.25 for DES) and (19.5% vs. 15.0%, respectively, P=0.16, HR 0.75, 95% CI 0.51–1.09 for DES). Definite/probable ST was more frequent in the BMS group (1.9% vs. 0%, respectively, P=0.046).

Conclusions: In real-world patients with AF undergoing PCI, DES use was associated with outcomes comparable to those with BMS without excess bleeding complications. More ST was seen in BMS-treated patients. (Circ J 2014; 78: 2674–2681)

Approximately 5–8% of patients undergoing percutaneous coronary intervention (PCI) and stenting have an indication for long-term oral anticoagulation (OAC) because of atrial fibrillation (AF).^1,2 Although the use of drug-eluting stents (DES) has decreased the rate of restenosis in many patient populations when compared with bare-metal stents (BMS),^3–5 little is known about their performance in patients with AF undergoing PCI. Current recommendations on the optimal choice of stent type and strategy of antithrombotic therapy in this high-risk subset of patients are derived mainly from small retrospective single-center studies and expert opinion statements, providing a weak level of evidence.^6–9 Dual antiplatelet therapy (DAPT) is needed to prevent stent thrombosis (ST), but this combination is less effective than OAC alone for preventing stroke in patients with AF.¹⁰ On the other hand, OAC alone is not effective in preventing ST following PCI and stenting.¹¹

The AFCAS (Atrial Fibrillation undergoing Coronary Artery Stenting) registry is a prospective multicenter registry that enrolled patients with AF referred for PCI and stenting between December 2006 and February 2010.^12–15 In this prespecified analysis, we performed an analysis of data from the prospective AFCAS registry to explore the 12-month clinical outcomes of patients with AF undergoing PCI with DES vs. BMS implantation.

Methods

Patient Selection and Study Design

The current study is based on an analysis of data from the AFCAS registry. Because of the observational design of the study, the only exclusion criterion was unwillingness/inability to participate in the study or to provide informed written consent. At each of the 17 participating European centers, patients were treated according to the local practice, and were followed up at 3, 6, and 12 months after the index procedure. Stent selection was entirely at the operator’s discretion. BMS included those based on stainless steel (28.4%), cobalt (10.2%) and cobalt-chromium platforms (19.3%), as well as coated stents such as titanium-nitride-oxide-coated (15.0%). DES included first- and second-generation DES such as everolimus-eluting (7.2%), sirolimus-eluting (5.6%), zotarolimus-eluting (6.2%), and paclitaxel-eluting (5.6%) stents. The periprocedural and postdischarge antithrombotic medications were completely at the discretion of the treating physician.

Ethical Issues

The study was initiated by the investigators and conducted according to the ethical guidelines of the 1964 Declaration of Helsinki, as revised in 2002. Informed written consent was obtained from every patient after full explanation of the study protocol. The study protocol was approved by the ethics committees of the participating centers. The AFCAS registry is registered with ClinicalTrials.gov under the number NCT00596570.

Study Definitions and Endpoints

The primary endpoints of the current study were (1) major adverse cardiac and cerebrovascular events (MACCE) defined as a composite of all-cause death, non-fatal myocardial infarction (MI), target vessel revascularization (TVR), definite/probable ST, transient ischemic attack (TIA) or stroke, and peripheral arterial embolism, and (2) bleeding complications. Spontaneous MI was diagnosed by persistent ischemic-type chest pain with a rise of biochemical markers of myocardial necrosis (CK-MB or troponin) at least twice the upper normal limit. Periprocedural MI was diagnosed by a rise of troponin level at least 3-fold of the upper normal limit. In-hospital re-MI was diagnosed by a new rise of biochemical markers (CK-MB or troponin) at least 50% above the lowest level measured previously. TVR was defined as any repeat intervention (surgical or percutaneous) to treat a significant luminal stenosis (defined as >50% diameter stenosis by visual estimation) in the index vessel. ST was adjudicated according to the criteria of definite or probable ST described by the Academic Research Consortium.¹⁶ TIA was defined as a transient (<24 h) focal neurologic deficit adjudicated by a neurologist. Stroke was defined as a permanent focal neurologic deficit adjudicated by a neurologist, and confirmed by computed tomography or magnetic resonance imaging. Peripheral arterial embolism was defined as signs/symptoms of peripheral ischemia associated with a positive imaging test. Bleeding events were defined according to the Bleeding Academic Research Consortium (BARC) criteria as minor (BARC 2), or major (BARC 3a, 3b, 3c and 5) bleeding events.¹⁷ Secondary endpoints included the individual components of MACCE, as well as the total adverse events (a composite of MACCE plus total bleeding events).

Statistical Analysis

Continuous variables are reported as the mean±standard deviation if they were normally distributed, and as median [interquartile range] if they were skewed. Data were tested for normal distribution using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Categorical variables are described with absolute and relative (percentage) frequencies. Comparisons between study subgroups were performed using the independent samples t-test or Mann-Whitney test for continuous variables, and Pearson χ² (Fisher’s exact) test for categorical variables, as appropriate. A Cox regression hazard model was used to identify the independent predictors of ST at 12-month follow-up. Variables strongly correlated with the dependent variable by univariate analyses (P<0.1) were entered in the model as covariates. Kaplan-Meier survival curves (log-rank test) were constructed to display the time-to-event relationship for the occurrence of MACCE, and all bleeding events. Propensity score analysis was performed for risk adjustment and one-to-one propensity score matching between study groups. We estimated the propensity score for the 2 study groups by using logistic regression including all risk factors with P<0.2 in the univariate analysis in order to avoid over-fitting. Matching was performed based on an estimated caliber of 0.03 according the method suggested by Austin.¹⁸ The area under the receiver-operating characteristic (ROC) curve was used to represent the discriminatory ability of the regression models. All tests were 2-sided and statistical significance was set at 5%. Statistical analysis was performed using SPSS statistical software (SPSS v. 16.0.1, SPSS Inc, Chicago, IL, USA).

Results

Baseline Clinical and Procedural Characteristics

Of 929 patients enrolled in the AFCAS registry and followed up for 12 months, 673 (72.4%) received BMS, and 220 (23.7%) received at least one DES. First-generation DES devices were used in 86/929 (9.3%) of patients, comprising altogether 39.1% of the patients in the DES group. Additionally, 18 (1.9%) patients were treated with an endothelial progenitor cell capturing stent (Genous, Orbus Neich, Hong Kong), and 18 (1.9%) with balloon angioplasty alone (no stent), but these patients were not included in the BMS or DES group comparisons. The rate of ST with endothelial progenitor cell capturing stent was 2/18 (11.1%).

Patients treated with DES were more often diabetic (P=0.036), had a longer total stent length (P<0.001), and more often had experienced a prior MI (P=0.003), PCI (P=0.003), and coronary bypass surgery (P=0.011), whereas patients treated with BMS more often had a history of heart failure (P=0.043), venous thromboembolism (P=0.026), and were more likely to present with acute ST-elevation MI (P=0.006). Of note, the 2 stent groups were comparable for history of previous hemorrhage, peptic ulcer, risk of thromboembolism and bleeding complications, as estimated by the CHA₂DS₂-VASc and HAS-BLED scores, respectively, (P>0.05 for all) (Table 1). No differences were detected in the rate of periprocedural uninterrupted anticoagulation with oral vitamin K antagonist (VKA: 36.4% vs. 35.5%), bridging with low-molecular weight heparin (30.0% vs. 33.2%) and no oral VKA (24.8% vs. 23.2%) in patients treated with BMS and DES, respectively.

Table 1. Baseline Clinical, Angiographic, and Procedural Characteristics of the 2 Study Groups of PCI Patients

	Whole cohort			Propensity score-matched pairs
	BMS (n=673)	DES (n=220)	P value	BMS (n=159)	DES (n=159)	P value
Age (years)	74.0 [10.0]	73.0 [9.0]	0.09	75.0 [10]	72 [10]	0.05
Female sex	206 (30.6)	62 (28.2)	0.55	46 (28.9)	49 (30.8)	0.71
Body mass index (kg/m2)	27.8 [6.0]	27.7 [5.7]	0.44	27.7 [5.6]	27.8 [5.4]	0.38
Periprocedural INR	1.9 [1.0]	1.9 [1.0]	0.08	1.8 [1.0]	1.9 [1.0]	0.42
Hypertension	563 (83.7)	186 (84.5)	0.83	133 (83.6)	136 (85.5)	0.64
Hypercholesterolemia	443 (65.8)	150 (68.2)	0.57	114 (71.7)	109 (68.6)	0.54
Diabetes	234 (34.8)	94 (42.7)	0.036	69 (43.4)	65 (40.9)	0.65
Current smoking	67 (10.0)	24 (10.9)	0.70	12 (7.5)	20 (12.6)	0.14
Family history of CAD	162 (24.1)	65 (29.5)	0.11	49 (30.8)	46 (28.9)	0.71
Previous TIA/stroke	116 (17.2)	32 (14.5)	0.40	22 (13.8)	23 (14.5)	0.87
Previous MI	152 (22.6)	72 (32.7)	0.003	43 (27.0)	51 (32.1)	0.33
Previous PCI	98 (14.6)	52 (23.6)	0.003	26 (16.4)	37 (23.3)	0.12
Previous coronary surgery	86 (12.8)	44 (20.0)	0.011	25 (15.7)	29 (18.2)	0.55
Previous hemorrhage	25 (3.7)	11 (5.0)	0.43	5 (3.1)	8 (5.0)	0.40
Previous peptic ulcer	33 (4.9)	11 (5.0)	1.0	8 (5.0)	8 (5.0)	1.00
Congestive heart failure	148 (22.0)	34 (15.5)	0.043	34 (21.4)	32 (20.1)	0.78
Mechanical mitral valve	6 (0.9)	2 (0.9)	1.0	1 (0.6)	1 (0.6)	1.00
Mechanical aortic valve	9 (1.3)	4 (1.8)	0.53	2 (1.3)	1 (0.6)	0.56
CHA2DS2-VASc score	4.0 [2.0]	4.0 [2.0]	0.12	4.0 [2.0]	4.0 [2.0]	0.24
HAS-BLED score	3.0 [0.5]	3.0 [0]	0.50	3.0 [1.0]	3.0 [0]	0.69
Previous venous thromboembolism	27 (4.0)	2 (0.9)	0.026	6 (3.8)	2 (1.3)	0.15
Ejection fraction (%)	50.0 [20.0]	51.0 [20.0]	0.08	54.0 [20]	50.0 [20]	0.59
Medications at discharge
Triple therapy	498 (74.0)	157 (71.4)	0.48	119 (74.8)	119 (74.8)	1.00
Dual antiplatelet therapy	112 (16.6)	49 (22.3)	0.069	26 (16.4)	36 (22.6)	0.16
VKA and clopidogrel	55 (8.2)	14 (6.4)	0.47	13 (8.2)	4 (2.5)	0.04
VKA and ASP	8 (1.2)	0	0.21	1 (0.6)	0	1.00
Statin	546 (81.1)	181 (82.3)	0.41	140 (88.1)	134 (84.3)	0.62
NSAID	5 (0.7)	0	1.0	1 (0.6)	1 (0.6)	1.00
Proton-pump inhibitor	239 (35.5)	71 (32.3)	0.47	61 (38.4)	55 (34.6)	0.76
Pattern of AF
Permanent	327 (48.6)	107 (48.6)	1.0	79 (49.7)	76 (47.8)	0.74
Persistent	83 (12.3)	22 (10.0)	0.40	20 (12.6)	16 (10.1)	0.48
Paroxysmal	258 (38.3)	91 (41.4)	0.43	60 (37.7)	67 (42.1)	0.42
Culprit vessel
Left anterior descending	272 (40.4)	103 (46.8)	0.10	71 (44.7)	77 (48.4)	0.50
Left circumflex	136 (20.2)	46 (20.9)	0.85	35 (22.0)	33 (20.8)	0.78
Right coronary artery	233 (34.6)	57 (25.9)	0.016	46 (28.9)	42 (26.4)	0.62
Left main coronary artery	19 (2.8)	17 (7.7)	0.003	9 (5.7)	12 (7.5)	0.50
Arterial graft	5 (0.7)	0	0.34	1 (0.6)	0	1.00
Saphenous vein graft	43 (6.4)	9 (4.1)	0.25	12 (7.5)	5 (3.1)	0.13
Clinical presentation						0.90
STEMI	100 (14.9)	17 (7.7)	0.006	14 (8.8)	13 (8.2)	0.84
Non-STEMI	174 (25.9)	54 (24.5)	0.72	36 (22.6)	36 (22.6)	1.00
Unstable angina	119 (17.7)	44 (20.0)	0.48	25 (15.7)	30 (18.9)	0.46
Stable angina	280 (41.6)	104 (47.5)	0.14	84 (52.8)	80 (50.3)	0.65
No. of treated vessels	1.1±0.4	1.2±0.5	0.008	1.3±0.5	1.2±0.4	0.37
Total stent length	19 [13]	24 [23]	<0.001	24 [23]	24 [19]	0.89
Femoral sheath	479 (71.2)	164 (74.5)	0.34	123 (77.4)	123 (77.4)	1.00

Continuous variables are presented as mean±SD, or median [interquartile range], whereas categorical variables are presented as frequency (percentage). AF, atrial fibrillation; ASA, aspirin; BMS, bare-metal stents; CAD, coronary artery disease; DES, drug-eluting stents; INR, international normalized ratio; MI, myocardial infarction; NSAID, nonsteroidal antiinflammatory drugs; PCI, percutaneous coronary intervention; TIA, transient ischemic attack; STEMI, ST-elevation MI; VKA, vitamin K antagonist.

Clinical Outcome at 12-Month Follow-up

At 12-month follow-up, the primary endpoint of MACCE was comparable between patients who received BMS and those who received DES (22.0% vs. 19.5%, respectively, P=0.51). Similarly, total bleeding events (BARC 2, 3a, 3b, 3c, 5) were comparable between the 2 stent groups (19.5% vs. 15.0%, respectively, P=0.16). The rate of definite or probable ST was higher in the BMS group (1.9% vs. 0%, respectively, P=0.046). The 2 groups were comparable for the rates of all-cause death, non-fatal MI, TVR, stroke/TIA/peripheral arterial embolism, minor bleeding events (BARC 2), major bleeding events (BARC 3a, 3b, 3c, 5), and total adverse events (P>0.05 for all) (Table 2). When analyzed separately for those treated with triple therapy, DAPT or VKA+clopidogrel, the rates of MACCE (21.7% vs. 20.9% P=0.91; 22.3% vs. 16.3%, P=0.53; 18.2% vs. 14.3%, P=1.0) and its derivatives of mortality, MI, TVR, ST and TIA/stroke (data not shown) were comparable in patients treated with BMS vs. DES, respectively.

Table 2 Clinical Outcomes at 12-Month Follow-up in the 2 Study Groups of PCI Patients

Outcome	Whole cohort			Propensity score-matched pairs
	BMS (n=673)	DES (n=220)	P value	BMS (n=159)	DES (n=159)	P value
MACCE	148 (22.0)	43 (19.5)	0.51	36 (22.6)	35 (22.0)	0.89
All-cause mortality	78 (11.6)	21 (9.5)	0.46	20 (12.6)	16 (10.1)	0.48
TVR	52 (7.7)	18 (8.2)	0.89	12 (7.5)	15 (9.4)	0.55
ST	13 (1.9)	0	0.046	5 (3.1)	0	0.06
MI	34 (5.1)	17 (7.7)	0.18	7 (4.4)	14 (8.8)	0.11
Stroke/TIA/peripheral arterial embolism	25 (3.7)	4 (1.8)	0.20	4 (2.5)	3 (1.9)	1.00
Total bleeding events	131 (19.5)	33 (15.0)	0.16	27 (17.0)	24 (15.1)	0.65
Minor (BARC 2)	57 (8.5)	13 (5.9)	0.25	17 (10.7)	11 (6.9)	0.24
Major (BARC 3a, 3b, 3c, 5)	75 (11.1)	20 (9.1)	0.45	11 (6.9)	13 (8.2)	0.67
BARC 3a	32 (4.8)	7 (3.2)	0.45	4 (2.5)	5 (3.1)	1.00
BARC 3b	23 (3.4)	10 (4.5)	0.42	3 (1.9)	7 (4.4)	0.34
BARC 3c	10 (1.5)	0	0.13	3 (1.9)	0	0.25
BARC 5	9 (1.3)	3 (1.4)	1.0	0	1 (0.6)	1.00
Total (MACCE+bleeding) events	238 (35.4)	67 (30.5)	0.19	57 (35.8)	52 (32.7)	0.56

Variables are presented as frequency (percentage). BARC, Bleeding Academic Research Consortium; MACCE, major adverse cardiac and cerebrovascular events; ST, stent thrombosis; TVR, target vessel revascularization. Other abbreviations as in Table 1.

Figures 1 and 2 show the Kaplan-Meier curves for MACCE-free survival and BARC bleeding-free survival for the 2 stent groups. DES were associated with similar outcome for MACCE (hazard ratio (HR) 0.89, 95% confidence interval (CI) 0.63–1.25, P=0.49) and any bleeding event (HR 0.75, 95% CI 0.51–1.09, P=0.13) in the Cox regression model compared with BMS.

Figure 1.

Kaplan-Meier curve for 12 months’ survival free of major adverse cardiac and cerebrovascular events (MACCE) after the index percutaneous coronary intervention (PCI). BMS, bare-metal stent; DES, drug-eluting stent.

Figure 2.

Kaplan-Meier curve for 12 months’ survival free of [Bleeding Academic Research Consortium (BARC)] bleeding after the index percutaneous coronary intervention intervention (PCI). BMS, bare-metal stent; DES, drug-eluting stent.

Patients with first-generation DES experienced similar rates of MACCE, ST and repeat revascularization compared with those implanted with newer generation DES; however, the rate of MI was higher (15.1% vs. 3.0%, P<0.001, respectively). The frequency of patients on triple, DAPT or VKA+clopidogrel was similar in patients with first-generation vs. newer generation DES at hospital discharge (data not shown). Mean clopidogrel duration was longer in patients with first-generation vs. newer generation DES (9.0±3.2 months vs. 8.3±4.2 months, P=0.018).

Individual characteristics of ST events are presented in Table 3. Half (50%) of the ST events occurred early (<30 days). The median CHA₂DS₂-VASc score was significantly higher in patients with ST vs. those without (5.0 [1.0] vs. 4.0 [2.0], P=0.004). Compared with those without ST, patients who developed ST more often had a HAS-BLED score ≥3 (100% vs. 75% P=0.029). Increasing age (HR 1.10, 95% CI 1.02–1.19, P=0.023), diabetes (HR 3.80, 95% CI 1.45–12.5, P=0.009), and previous gastrointestinal bleeding episode (HR 7.78, 95% CI 1.75–34.6, P=0.007) remained independent predictors of definite or probable ST at 12-month follow-up in the Cox regression model including also acute coronary syndrome as an indication for PCI, congestive heart failure, and preprocedural anemia (hemoglobin <13 g/dl for males and <12 g/dl for females) as covariates. Patients with an early BARC (2, 3a, 3b, 3c or 5) bleeding episode within 30 days after the index PCI had non-significantly more early ST events compared with those without any bleeding events (1.9% vs. 0.6%, P=0.18, respectively).

Table 3. Characteristics of Individual Cases of ST Among the Study PCI Patients

Case no.	Age (years)	Sex	Indication for index PCI	ST days after index PCI	Antithrombotic medication at the time of ST	Remarks
Early ST
1	78	F	NSTEMI	13	LMWH+ASA+CLOP
2	77	M	STEMI	3	VKA+ASA+CLOP	INR at event 1.7
3	78	M	NSTEMI	0	VKA+ASA	ST event before postprocedural CLOP loading
4	82	M	NSTEMI	7	VKA+ASA+CLOP
5	77	M	NSTEMI	9	(VKA)+ASA	Stopped CLOP because of poor adherence, INR at event 1.5
6	78	M	Unstable AP	8	VKA+ASA+CLOP
Late ST
7	50	F	Unstable AP	50	ASA+CLOP
8	69	M	Stable AP	35	LMWH+ASA+CLOP
9	76	F	Unstable AP	193	VKA+ASA+CLOP
10	81	F	NSTEMI	356	VKA+CLOP
11	80	M	Stable AP	212	VKA+ASA+CLOP
12	90	M	Unstable AP	151	VKA	INR at event 1.9

AP, angina pectoris; CLOP, clopidogrel; LMWH, low-molecular weight heparin; NSTEMI, non-STEMI. Other abbreviations as in Tables 1,2.

Antithrombotic Medications

At hospital discharge, the 2 groups were comparable for the rates of prescription of triple therapy (OAC, clopidogrel and aspirin), and OAC plus clopidogrel. The prescription of DAPT tended to be more frequent in the DES group (22.3% vs. 16.6%, respectively, P=0.069). Very few patients (1.2%) in the BMS group were prescribed VKAs and aspirin at discharge. The median and mean durations of clopidogrel therapy were significantly shorter in patients treated with BMS vs. those treated with DES (1 [5] vs. 12 [6] months, P<0.001; and 4.1±4.4 vs. 8.9±4.0 months, respectively).

Propensity Score Analysis

We estimated the propensity score for the 2 study groups by using logistic regression including all risk factors with P<0.2 in the univariate analysis (Hosmer-Lemeshow’s test: P=0.501, area under the ROC curve 0.72, 95% CI 0.67–0.76). One-to-one propensity matching resulted in 159 pairs with similar baseline characteristics (Table 1). The analysis confirmed the comparable outcome observed in the overall series (Table 2). After propensity score-matched analysis, there was a trend towards more frequent definite or probable ST in the BMS group (P=0.06).

Discussion

The present study demonstrates that in real-world patients with AF referred for PCI, BMS were more frequently used. Reflecting the clinical practice guidelines,¹⁹ patients who received BMS were less often diabetic, had fewer prior ischemic events but more heart failure or prior venous thromboembolism, and were more likely to present with acute ST-elevation MI. In contrast, patients treated with DES had features that may be markers for more advanced coronary artery disease (CAD) and higher restenosis rates such as higher prevalence of prior MI, prior PCI, and prior CABG. At 12-month follow-up, the rates of MACCE and bleeding events were comparable between the BMS and DES groups; however, patients who received BMS more frequently experienced definite or probable ST. The type of stent did not seem to affect the choice of antithrombotic medications at discharge, but the duration of clopidogrel therapy was significantly longer in patients treated with DES. Despite the prolonged duration of clopidogrel treatment, no excess in bleeding complications was observed when using DES.

To the best of our knowledge, the current stent-based analysis is derived from the largest series of AF patients undergoing PCI and stenting with mid-term follow-up. Recent reports have suggested a paradigm shift in the occurrence of ST after DES implantation as evidenced by lower ST rates in patients with newer generation DES compared with those of BMS.^20,21 In line with this, the rate of definite/probable ST was significantly higher with BMS vs. DES in the AFCAS registry, although overall 12-month ST rates were relatively low (1.5%). Approximately 9% of patients received a first-generation stent device. The predilection by the operators to use DES in patients with lower-risk baseline characteristics may have contributed to this difference: more patients presenting with ST-elevation MI, and more patients with heart failure received BMS than DES. In AFCAS trial patients who developed early bleeding events (within 30 days after the index PCI) tended to have more ST vs. those without such an event. Bleeding episodes may have led to premature discontinuation of DAPT, and thus predisposed to ST. Half of the ST events were early (<30 days after the index PCI). At the time of the event, most of the patients (4/6) with early ST had ongoing triple therapy; 1 patient had discontinued both VKA and clopidogrel at home; and 1 patient had periprocedural ST before the clopidogrel loading dose was given. In addition, half of the patients with late ST were on triple therapy (3/6); 1 on aspirin+clopidogrel; 1 on VKA only, and 1 on VKA+clopidogrel.

The rates of TVR were comparable between the 2 groups at 12 months, which might be related to selection bias towards the use of DES in patients at a higher risk of restenosis (ie, higher prevalence of diabetes, and longer total stent length in the DES group). Moreover, the older age and high prevalence of comorbidity (including AF) in the current series might have favored medical treatment over re-intervention in patients with recurrent angina after PCI, thus reducing the difference in TVR between the 2 groups. Additionally, no angiographic follow-up was performed, and all TVR procedures were ischemia-driven. It is known that ischemia-driven TVR tends to decrease the absolute risk difference in TVR between stent groups, and thus underestimates the actual rates of in-stent restenosis.

The major clinical implication of this report is that a fear of using DES in frail patients receiving multiple antithrombotic medications may be overemphasized, given the low ST rates in patients with DES without significant increase in bleeding events compared with that in the patients with BMS. Of note, the risk of bleeding as estimated by the HAS-BLED score, was comparable between the 2 groups, suggesting that it did not affect stent choice. The frequency of triple-therapy prescription at discharge was also not affected by stent type. The fact that over 70% of patients in both groups were prescribed triple therapy at discharge might explain the high bleeding rates observed. The WOEST randomized controlled trial demonstrated that an antithrombotic regimen of clopidogrel plus OAC, compared with triple therapy, reduced bleeding events at 1-year follow-up, with no increase in thrombotic events in patients with stable CAD.²² In another non-randomized series of DES, both ischemic and major bleeding complications were comparable between DAPT and triple therapy at 2-year follow-up.²³ In the latter study, the choice of antithrombotic regimen was based on clinical and echocardiographic criteria of thrombotic risk. In a retrospective analysis of patients on warfarin therapy referred for PCI, the incidences of stroke and ST were lowest with triple therapy; the stroke rate was highest with DAPT alone, and the ST rate was highest with a regimen of aspirin plus OAC.²⁴

Patients with AF need long-term OAC to reduce the risk of stroke and other thromboembolic events.¹⁰ On the other hand, DAPT is needed following PCI and stent implantation, especially DES, in order to prevent ST.^25,26 Triple therapy (combination of DAPT and OAC) increases the incidence of major bleeding 3.2–6.6-fold, compared with DAPT.^1,27 Similarly, triple therapy increases the risk of hospital admission for bleeding complications in post-MI patients 3-fold, compared with OAC alone.²⁸ Evidence suggests that bleeding complications predict long-term mortality in patients undergoing PCI.²⁹ Therefore, patients with AF have been considered as a special “high-risk” group that is usually excluded from randomized controlled trials comparing different stent designs. In the current study, although the rates of bleeding at 12-month follow-up were high in both stent groups, no significant difference was found in major, minor, or total bleeding events. Our results are supported by a propensity score-matched study in which DES implantation was not associated with increased major bleeding or ischemic/thrombotic complications.³⁰ The same study reported that ST was rare after cessation of a short course of triple therapy, suggesting that “brief” triple therapy following DES implantation in this patient subset may be sufficient to prevent ST, while reducing bleeding complications.³⁰ In a pooled analysis, the incidence of major bleeding increased from 4.6% to 10.3% when the duration of triple therapy increased from 1 month to 6–12 months or more.¹ This underscores the importance of the duration of triple therapy, which should be minimal based on careful evaluation of the individual patient’s risk of ischemic and bleeding complications. In a smaller retrospective cohort, DES implantation in patients receiving OAC significantly reduced TVR, compared with BMS, with no increase in major or minor bleeding events.⁸ In contrast, another propensity score-matched cohort demonstrated that DES implantation was associated with a higher rate of major bleeding, vs. BMS, with similar rates of ischemic events.⁶

Study Limitations

The current study has all the inherent limitations of an observational study design, including individual risk-based decision making in treatment choices, which may introduce selection bias. Another important limitation is the heterogeneity of the AF population among the participating centers, and some differences in the periprocedural routines. The use of several different DES creates heterogeneity in the DES group in this study and represents a limitation of this analysis. Nevertheless, it is unlikely that this constitutes major bias, as the rate of ST was 0% with DES in the 12-month follow-up. Times in therapeutic range values after hospital discharge were not routinely collected by any of the study sites. Nevertheless, the INR level at the time of the first bleeding event was 1.8±0.6 (range 1.0–3.0), reflecting that bleeding events were generally not caused by supratherapeutic VKA treatment.³¹ This also likely reflects that INR levels were modestly controlled in patients receiving triple antithrombotic therapy. The strength of the study is the enrolment of consecutive patients with the only exclusion criterion being unwillingness/inability to participate in the study. In this regard, the study population well represents real-world patients with AF referred for PCI.

Conclusions

In real-world patients with AF undergoing PCI, the use of DES was less frequent and associated with an outcome comparable to that with BMS. More ST was seen in patients who received BMS. No excess in bleeding complications was observed despite a longer duration of combination antithrombotic treatment in the DES group, although both groups had comparable intensity of antithrombotic medications prescribed at discharge. The fear of increased risk of thrombotic and bleeding events after DES implantation may be overemphasized.

Acknowledgments

Grants: This study was supported by unrestricted grants from Novartis^®, Germany, and Sanofi-Aventis^®, Germany, and by grants from the Finnish Foundation for Cardiovascular Research, Helsinki, Finland (K.E.J.A. and T.K.), and a scholarship from the Finnish Cardiac Society, Finland (T.K.).

We thank our study coordinator Tuija Vasankari, RN, for her input to the data management.

References

• 1.    Rubboli  A,  Halperin  JL,  Juhani Airaksinen  KE,  Buerke  M,  Eeckhout  E,  Freedman  SB, et al. Antithrombotic therapy in patients treated with oral anticoagulation undergoing coronary artery stenting: An expert consensus document with focus on atrial fibrillation. Ann Med 2008; 40: 428–436.
• 2.    Rubboli  A,  Colletta  M,  Valencia  J,  Capecchi  A,  Franco  N,  Zanolla  L, et al. Periprocedural management and in-hospital outcome of patients with indication for oral anticoagulation undergoing coronary artery stenting. J Interv Cardiol 2009; 22: 390–397.
• 3.    Stone  GW,  Ellis  SG,  Cox  DA,  Hermiller  J,  O’Shaughnessy  C,  Mann  JT, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004; 350: 221–231.
• 4.    Suzuki  S,  Ishii  H,  Matsudaira  K,  Okumura  N,  Yoshikawa  D,  Hayashi  M, et al. Long-term outcome of drug-eluting vs. bare-metal stents in patients with acute myocardial infarction: Subgroup analysis of the Nagoya Acute Myocardial Infarction Study (NAMIS). Circ J 2013; 77: 2024–2031.
• 5.    Kubo  S,  Kadota  K,  Shimada  T,  Ozaki  M,  Ichinohe  T,  Eguchi  H, et al. Seven-year clinical outcomes of unprotected left main coronary artery stenting with drug-eluting stent and bare-metal stent. Circ J 2013; 77: 2497–2504.
• 6.    Ruiz-Nodar  JM,  Marín  F,  Sánchez-Payá  J,  Hurtado  JA,  Valencia-Martín  J,  Manzano-Fernández  S, et al. Efficacy and safety of drug-eluting stent use in patients with atrial fibrillation. Eur Heart J 2009; 30: 932–939.
• 7.    Ruiz-Nodar  JM,  Marín  F,  Hurtado  JA,  Valencia  J,  Pinar  E,  Pineda  J, et al. Anticoagulant and antiplatelet therapy use in 426 patients with atrial fibrillation undergoing percutaneous coronary intervention and stent implantation: Implications for bleeding risk and prognosis. J Am Coll Cardiol 2008; 51: 818–825.
• 8.    Rogacka  R,  Chieffo  A,  Michev  I,  Airoldi  F,  Latib  A,  Cosgrave  J, et al. Dual antiplatelet therapy after percutaneous coronary intervention with stent implantation in patients taking chronic oral anticoagulation. JACC Cardiovasc Interv 2008; 1: 56–61.
• 9.    Camm  AJ,  Kirchhof  P,  Lip  GYH,  Schotten  U,  Savelieva  I,  Ernst  S, et al. Guidelines for the management of atrial fibrillation: The Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010; 33: 2719–2747.
• 10.    Connolly  S,  Pogue  J,  Hart  R,  Pfeffer  M,  Hohnloser  S,  Chrolavicius  S, et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): A randomised controlled trial. Lancet 2006; 367: 1903–1912.
• 11.    Leon  MB,  Baim  DS,  Popma  JJ,  Gordon  PC,  Cutlip  DE,  Ho  KKL, et al. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. N Engl J Med 1998; 339: 1665–1671.
• 12.    Rubboli  A,  Schlitt  A,  Kiviniemi  T,  Biancari  F,  Karjalainen  PP,  Valencia  J, et al. One-year outcome of patients with atrial fibrillation undergoing coronary artery stenting: An analysis of the AFCAS registry. Clin Cardiol 2014; 37: 357–364.
• 13.    Kiviniemi  T,  Karjalainen  P,  Rubboli  A,  Schlitt  A,  Tuomainen  P,  Niemelä  M, et al. Thrombocytopenia in patients with atrial fibrillation on oral anticoagulation undergoing percutaneous coronary intervention. Am J Cardiol 2013; 112: 493–498.
• 14.    Schlitt  A,  Rubboli  A,  Lip  GYH,  Lahtela  H,  Valencia  J,  Karjalainen  PP, et al; AFCAS (Management of patients with Atrial Fibrillation undergoing Coronary Artery Stenting Study Group). The management of patients with atrial fibrillation undergoing percutaneous coronary intervention with stent implantation: In-hospital-data from the Atrial Fibrillation undergoing Coronary Artery Stenting study. Cathet Cardiovasc Interv 2013; 82: E864–E870, doi:10.1002/ccd.25064.
• 15.    Lahtela  H,  Rubboli  A,  Schlitt  A,  Karjalainen  PP,  Niemelä  M,  Vikman  S, et al. Heparin bridging vs. uninterrupted oral anticoagulation in patients with atrial fibrillation undergoing coronary artery stenting: Results from the AFCAS registry. Circ J 2012; 76: 1363–1368.
• 16.    Cutlip  DE,  Windecker  S,  Mehran  R,  Boam  A,  Cohen  DJ,  van Es  G, et al. Clinical end points in coronary stent trials. Circulation 2007; 115: 2344–2351.
• 17.    Mehran  R,  Rao  SV,  Bhatt  DL,  Gibson  CM,  Caixeta  A,  Eikelboom  J, et al. Standardized bleeding definitions for cardiovascular clinical trials: A consensus report from the Bleeding Academic Research Consortium. Circulation 2011; 123: 2736–2747.
• 18.    Austin  PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat 2011; 10: 150–161.
• 19.    Lip  GYH,  Huber  K,  Andreotti  F,  Arnesen  H,  Airaksinen  JK,  Cuisset  T, et al. Antithrombotic management of atrial fibrillation patients presenting with acute coronary syndrome and/or undergoing coronary stenting: Executive summary: A Consensus Document of the European Society of Cardiology Working Group on Thrombosis, endorsed by the European Heart Rhythm Association (EHRA) and the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J 2010; 31: 1311–1318.
• 20.    Palmerini  T,  Biondi-Zoccai  G,  Riva  DD,  Stettler  C,  Sangiorgi  D,  D’Ascenzo  F, et al. Stent thrombosis with drug-eluting and bare-metal stents: Evidence from a comprehensive network meta-analysis. Lancet 2012; 379: 1393–1402.
• 21.    Palmerini  T,  Biondi-Zoccai  G,  Della Riva  D,  Mariani  A,  Genereux  P,  Branzi  A, et al. Stent thrombosis with drug-eluting stents: Is the paradigm shifting? J Am Coll Cardiol 2013; 62: 1915–1921.
• 22.    Dewilde  WJ,  Oirbans  T,  Verheugt  FW,  Kelder  JC,  De Smet  BJ,  Herrman  J, et al. Use of clopidogrel with or without aspirin in patients taking oral anticoagulant therapy and undergoing percutaneous coronary intervention: An open-label, randomised, controlled trial. Lancet 2013; 381: 1107–1115.
• 23.    Sarafoff  N,  Ndrepepa  G,  Mehilli  J,  Dörrler  K,  Schulz  S,  Iijima  R, et al. Aspirin and clopidogrel with or without phenprocoumon after drug eluting coronary stent placement in patients on chronic oral anticoagulation. J Intern Med 2008; 264: 472–480.
• 24.    Karjalainen  PP,  Vikman  S,  Niemelä  M,  Porela  P,  Ylitalo  A,  Vaittinen  M, et al. Safety of percutaneous coronary intervention during uninterrupted oral anticoagulant treatment. Eur Heart J 2008; 29: 1001–1010.
• 25.    Iakovou  I,  Schmidt  T,  Bonizzoni  E,  Ge  L,  Sangiorgi  GM,  Stankovic  G, et al. Incidence, predictors, and outcome of thrombosis after successful implantation of drug-eluting stents. JAMA 2005; 293: 2126–2130.
• 26.    Levine  GN,  Bates  ER,  Blankenship  JC,  Bailey  SR,  Bittl  JA,  Cercek  B, et al. 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Society for Cardiovascular Angiography and Interventions. J Am Coll Cardiol 2011; 58: e44–e122, doi:10.1016/j.jacc.2011.08.007.
• 27.    Bernard  A,  Fauchier  L,  Pellegrin  C,  Clementy  N,  Saint Etienne  C,  Banerjee  A, et al. Anticoagulation in patients with atrial fibrillation undergoing coronary stent implantation. Thromb Haemost 2013; 110: 560–568.
• 28.    Sørensen  R,  Hansen  ML,  Abildstrom  SZ,  Hvelplund  A,  Andersson  C,  Jørgensen  C, et al. Risk of bleeding in patients with acute myocardial infarction treated with different combinations of aspirin, clopidogrel, and vitamin K antagonists in Denmark: A retrospective analysis of nationwide registry data. Lancet 2009; 374: 1967–1974.
• 29.    Romaguera  R,  Wakabayashi  K,  Laynez-Carnicero  A,  Sardi  G,  Maluenda  G,  Ben-Dor  I, et al. Association between bleeding severity and long-term mortality in patients experiencing vascular complications after percutaneous coronary intervention. Am J Cardiol 2012; 109: 75–81.
• 30.    Annala  A,  Karjalainen  PP,  Biancari  F,  Niemelä  M,  Ylitalo  A,  Vikman  S, et al. Long-term safety of drug-eluting stents in patients on warfarin treatment. Ann Med 2012; 44: 271–278.
• 31.    Kiviniemi  T,  Puurunen  M,  Schlitt  A,  Rubboli  A,  Karjalainen  P,  Vikman  S, et al. Performance of bleeding risk-prediction scores in patients with atrial fibrillation undergoing percutaneous coronary intervention. Am J Cardiol 2014; 113: 1995–2001.