ORIGINAL ARTICLE
Association of improved outcome in acute ischaemic stroke patients with atrial fibrillation who receive early antithrombotic therapy: analysis from VISTA A. H. Abdul-Rahima, R. L. Fultona, B. Frankb, T. Tatlisumakc, M. Paciaronid, V. Casod, H.-C. Dienerb and K. R. Leesa for the VISTA collaborators* a
EUROPEAN JOURNAL OF NEUROLOGY
Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK; bDepartment of Neurology, University Hospital Essen, Essen, Germany; cDepartment of Neurology, Helsinki University Central Hospital, Helsinki, Finland; and dStroke Unit and Division of Internal and Cardiovascular Medicine, University of Perugia, Perugia, Italy
See editorial by Kelley on page 1027.
Keywords:
antithrombotics, atrial fibrillation, intracerebral haemorrhage, stroke Received 30 April 2014 Accepted 1 August 2014 European Journal of Neurology 2015, 22: 1048–1055 doi:10.1111/ene.12577
Background and purpose: Ischaemic stroke patients with atrial fibrillation (AF) are at risk of early recurrent stroke (RS). However, antithrombotics commenced at the acute stage may exacerbate haemorrhagic transformation, provoking symptomatic intracerebral haemorrhage (SICH). The relevance of antithrombotics on the patterns and outcome of the cohort was investigated. Methods: A non-randomized cohort analysis was conducted using data obtained from VISTA (Virtual International Stroke Trials Archive). The associations of antithrombotics with the modified Rankin Scale (mRS) outcome and the occurrence of RS and SICH (each as a combined end-point of fatal and non-fatal events) at 90 days for post-stroke patients with AF were described. Dichotomized outcomes were also considered as a secondary end-point (i.e. mortality and good outcome measure at 90 days). Results: In all, 1644 patients were identified; 1462 (89%) received antithrombotics, 157 (10%) had RS and 50 (3%) sustained SICH by day 90. Combined antithrombotic therapy (i.e. anticoagulants and antiplatelets), 782 (48%), was associated with favourable outcome on ordinal mRS and a significantly lower risk of RS, SICH and mortality by day 90, compared with the no antithrombotics group. The relative risk of RS and SICH appeared highest in the first 2 days post-stroke before attenuating to become constant over time. Conclusions: The risks and benefits of antithrombotics in recent stroke patients with AF appear to track together. Early introduction of anticoagulants (2–3 days post-stroke), and to a lesser extent antiplatelet agents, was associated with substantially fewer RS events over the following weeks but with no excess risk of SICH. More evidence is required to guide clinicians on this issue.
Introduction Atrial fibrillation (AF) is the primary cause for cardioembolic stroke [1]. Patients with AF are at risk of early recurrent ischaemic stroke (RS) even after Correspondence: A. H. Abdul-Rahim, Institute of Cardiovascular and Medical Sciences, 44 Church Street, University of Glasgow, Glasgow, G11 6NT, UK (tel.: +44 141 211 1861; fax: +44 141 211 1863; e-mail: [email protected]). *VISTA collaborators are given in the Appendix.
1048
thrombolytic therapy [2]. Amongst the AF cohort, there is excellent evidence to support the use of anticoagulation for RS prevention [3,4]. Nevertheless, there is considerable uncertainty on the optimal latency after acute stroke at which anticoagulation treatment should commence in order to prevent RS without resulting in a symptomatic intracerebral haemorrhage (SICH). The European Stroke Organization and the American Stroke Association do not recommend the use of anticoagulation in
© 2014 EAN
ANTITHROMBOTICS IN RECENT STROKE WITH AF
the hyperacute period post-stroke but neither do they recommend an acceptable delay post-stroke to start anticoagulation [5,6]. Clinically, it seems reasonable to begin anticoagulation as soon as the patient is both medically and neurologically stable. This is often 2 or 3 days post-stroke. This probably achieves a therapeutic anticoagulation level by days 5–7 when the traditional vitamin K antagonist (VKA) is used. However, cardioembolic stroke is associated with an increased risk of haemorrhagic transformation in the first few days post-stroke [7,8]. Two meta-analyses suggested that early anticoagulation post-stroke is inadvisable because it offers no net benefit [9,10]. Nonetheless, there is no evidence to dispute the value of later or prolonged anticoagulation. It appears that there may be a threshold delay after stroke at which the value of anticoagulation switches from neutral to beneficial, at least amongst patients with AF. The associations of antithrombotics therapy commenced (i.e. anticoagulants or antiplatelets) on the patterns and outcomes of modified Rankin Scale (mRS), RS and SICH in a cohort of patients with recent stroke and AF were described. Dichotomized outcomes were also considered as a secondary endpoint (i.e. mortality and good outcome measure at 90 days).
1049
were excluded: baseline National Institutes of Health Stroke Scale (NIHSS), age, medical history, concomitant medication, occurrence of adverse and serious adverse events and mRS at day 90. Only patients who were known to have a medical history of AF or were found to have AF on baseline electrocardiogram were included. Data on anticoagulant treatment were based on warfarin or other VKA treatment started post-stroke. Data on antiplatelet treatment were based on antiplatelet aggregation drug (which includes derivatives of salicylic acid, thienopyridine and dypyridamole) started post-stroke. The novel oral anticoagulants (NOACs) were not prescribed during the available period of the analysis and so were not considered. Follow-up period was 90 days poststroke. Definition of outcome events
Symptomatic RS was defined as any stroke with neurological deterioration, as indicated by NIHSS that was higher by ≥4 points than the value at baseline, or any stroke leading to death. Similarly, SICH was defined as any intracerebral haemorrhage with neurological deterioration, with worsening NIHSS ≥4 points from baseline, or any ICH leading to death. Thus, RS and SICH are each a combined end-point of fatal and non-fatal events.
Methods Statistical methods Data source
A non-randomized cohort analysis was conducted using data obtained from the Virtual International Stroke Trials Archive (VISTA, http://www.vistacollaboration. org/), specifically from the ‘VISTA Acute’ sub-section [11]. VISTA is a collaborative registry that collates and provides access to completed acute stroke trial data (from 1998 to 2008), anonymized in relation to patients’ and trials’ identity, for novel exploratory analyses. Our cohort data do not include trials that investigated thrombolysis therapy in hyperacute stroke. However, they contain data from trials in which thrombolysis was commonly used as standard therapy. Conduct and reporting of our analysis is in accordance with the STROBE guidelines for cohort studies [12]. Participants and variables
Patients who had been randomized to receive placebo or any drug now known to possess no confirmed action on stroke outcome were selected. Patients who lacked the relevant baseline or outcome information
© 2014 EAN
Descriptive statistics were recorded for recent stroke patients with AF, assessing the complete cohort and comparing those who were treated with versus without antiplatelets post-stroke and with versus without anticoagulation post-stroke. Mean (SD) or median [interquartile range (IQR)] are given for continuous variables and count (percentage) for categorical variables. Our primary outcome measures were the ordinal shift of the mRS at day 90 using the full scale and the occurrence of RS and SICH by 90 days post-stroke. For comparison with previous trials, dichotomized outcomes at day 90 (mortality, mRS 0–1, mRS 0–2 and NIHSS 0–1) were reported. The odds ratio (OR) of achieving the studied outcome(s) against the comparator, and corresponding 95% confidence intervals (95% CI), using proportional odds logistic regression (i.e. ordinal regression for mRS distribution and binary regression for dichotomized outcome) were calculated. Adjustments were made for age, baseline NIHSS and thrombolysis treatment [13]. The spread of RS and SICH over the 90-day period were plotted for patients who received anticoagulation
1050
A. H. ABDUL-RAHIM ET AL.
Figure 1 Flow chart for selection of ischaemic stroke patients with atrial fibrillation from VISTA.
treatment and patients who had antiplatelet treatment. Within the cohort who sustained RS, patients were grouped into (i) those who had RS in the absence of anticoagulation; (ii) patients who had RS within 10 days of commencing anticoagulation; and (iii) patients who had RS beyond 10 days of commencing anticoagulation. The cumulative percentage of RS and SICH on each day post-initial stroke by these groupings was plotted. Similar groupings were made for patients treated with antiplatelets and their cumulative percentage of RS and SICH was plotted. All analyses were undertaken using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA).
Results Baseline characteristics
Of the 10 304 ischaemic stroke patients available from VISTA, 8060 were excluded as they were in sinus rhythm. Individual patients’ data were obtained for the remaining 1644 stroke patients with AF, which formed our cohort (Fig. 1). From this
post-stroke cohort, 518 (31%) were given anticoagulation treatment alone, 162 (10%) were given antiplatelets alone and 782 (48%) received a combination of the two. One hundred and eighty-two (11%) did not receive any antithrombotic therapy. The median start time for anticoagulation treatment was at day 2 (IQR 1–3) post-stroke, whilst the median start time for antiplatelet treatment was at day 1 (IQR 1–3) post-stroke. Patients who received anticoagulation were commonly younger and had less severe baseline NIHSS. Patients who received no antithrombotic therapy tended to have more severe baseline NIHSS. Detailed baseline characteristics are given in Tables 1 and S1. The proportion of patients who were on antithrombotic treatment prior to the initial stroke (i.e. at baseline) are as follows: anticoagulant agent only, 286 (17%); antiplatelet agent only, 488 (30%); and combination of the two, 78 (5%). Outcome
In this cohort, 157 (10%) patients had RS, 50 (3%) patients sustained SICH and 390 (24%) patients died from any cause, by day 90 (Table 2). Combined antithrombotic therapy with anticoagulation and antiplatelet was associated with more favourable functional outcome across full scale mRS at 90 days after adjustment for age, baseline NIHSS and thrombolysis treatment [OR 1.79 (95% CI 1.32–2.42)], compared with no antithrombotic therapy received. The treatment effect of combined antithrombotic therapy lost statistical significance when outcome was dichotomized as mRS 0–1, mRS 0–2 and NIHSS 0–1. Combined antithrombotic therapy was associated with significantly reduced risk of RS, SICH and mortality, by day 90, compared with no antithrombotic therapy received [RS, OR 0.33 (95% CI 0.21–0.53); SICH, OR 0.18 (95% CI 0.09–0.37); and mortality,
Table 1 Baseline characteristics of the cohort according to antithrombotic treatment received following stroke Antithrombotic treatment received post-stroke
Variables Male, n (%) Age, years, mean (SD) Baseline NIHSS, median (IQR) Received thrombolysis treatment, n (%) Medical history Previous stroke, n (%) Transient ischaemic attack, n (%)
No antithrombotic (no anticoagulation, no antiplatelet), n = 182
Antiplatelet only, n = 162
Anticoagulation treatment only, n = 518
Anticoagulation treatment and antiplatelet, n = 782
85 77.1 16 71
85 75.8 15 51
243 73.4 14 203
359 74.2 14 293
(46.7%) (9.0) (11–21) (39.0%)
46 (25.3%) 11 (6.7%)
(52.5%) (10.0) (11–19) (31.5%)
54 (33.3%) 10 (6.8%)
(46.9%) (10.1) (10–18) (39.2%)
108 (20.9%) 45 (9.1%)
(45.9%) (9.8) (9–18) (37.5%)
150 (19.2%) 70 (9.3%)
n, number of observations; NIHSS, National Institutes of Health Stroke Scale; IQR, interquartile range.
© 2014 EAN
ANTITHROMBOTICS IN RECENT STROKE WITH AF
1051
Table 2 Clinical outcome at 90 days (adjusted for age, baseline NIHSS and thrombolysis treatment) No anticoagulant SICH, n/N (%) Antiplatelet 3/162 (1.9%) No antiplatelet 17/182 (9.3%) OR (95% CI) 0.24 (0.07–0.88) P value 0.032 All 20/344 (5.8%) Symptomatic RS, n/N (%) Antiplatelet 14/162 (8.6%) No antiplatelet 36/182 (19.8%) OR (95% CI) 0.43 (0.22–0.86) P value 0.016 All 50/334 (14.5%) Mortality, n/N (%) Antiplatelet 45/162 (27.8%) No antiplatelet 74/182 (40.7%) OR (95% CI) 0.60 (0.35–1.02) P value 0.061 All 119/334 (34.6%)
Anticoagulant
OR (95% CI)
P value
All
15/782 (1.9%) 15/518 (2.9%) 0.59 (0.28–1.24) 0.164 30/1300 (2.3%)
0.92 (0.25–3.37) 0.43 (0.20–0.94)
0.904 0.034
18/944 (1.9%) 32/700 (4.6%) 0.42 (0.22–0.78) 0.006
0.49 (0.26–0.92)
0.027
52/782 (6.7%) 55/518 (10.6%) 0.60 (0.40–0.91) 0.015 107/1300 (8.2%)
0.78 (0.41–1.47) 0.61 (0.37–1.01)
0.439 0.056
0.62 (0.42–0.91)
0.013
139/782 (17.8%) 132/518 (25.5%) 0.58 (0.43–0.78)
Association of improved outcome in acute ischaemic stroke patients with atrial fibrillation who receive early antithrombotic therapy: analysis from VISTA A. H. Abdul-Rahima, R. L. Fultona, B. Frankb, T. Tatlisumakc, M. Paciaronid, V. Casod, H.-C. Dienerb and K. R. Leesa for the VISTA collaborators* a
EUROPEAN JOURNAL OF NEUROLOGY
Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK; bDepartment of Neurology, University Hospital Essen, Essen, Germany; cDepartment of Neurology, Helsinki University Central Hospital, Helsinki, Finland; and dStroke Unit and Division of Internal and Cardiovascular Medicine, University of Perugia, Perugia, Italy
See editorial by Kelley on page 1027.
Keywords:
antithrombotics, atrial fibrillation, intracerebral haemorrhage, stroke Received 30 April 2014 Accepted 1 August 2014 European Journal of Neurology 2015, 22: 1048–1055 doi:10.1111/ene.12577
Background and purpose: Ischaemic stroke patients with atrial fibrillation (AF) are at risk of early recurrent stroke (RS). However, antithrombotics commenced at the acute stage may exacerbate haemorrhagic transformation, provoking symptomatic intracerebral haemorrhage (SICH). The relevance of antithrombotics on the patterns and outcome of the cohort was investigated. Methods: A non-randomized cohort analysis was conducted using data obtained from VISTA (Virtual International Stroke Trials Archive). The associations of antithrombotics with the modified Rankin Scale (mRS) outcome and the occurrence of RS and SICH (each as a combined end-point of fatal and non-fatal events) at 90 days for post-stroke patients with AF were described. Dichotomized outcomes were also considered as a secondary end-point (i.e. mortality and good outcome measure at 90 days). Results: In all, 1644 patients were identified; 1462 (89%) received antithrombotics, 157 (10%) had RS and 50 (3%) sustained SICH by day 90. Combined antithrombotic therapy (i.e. anticoagulants and antiplatelets), 782 (48%), was associated with favourable outcome on ordinal mRS and a significantly lower risk of RS, SICH and mortality by day 90, compared with the no antithrombotics group. The relative risk of RS and SICH appeared highest in the first 2 days post-stroke before attenuating to become constant over time. Conclusions: The risks and benefits of antithrombotics in recent stroke patients with AF appear to track together. Early introduction of anticoagulants (2–3 days post-stroke), and to a lesser extent antiplatelet agents, was associated with substantially fewer RS events over the following weeks but with no excess risk of SICH. More evidence is required to guide clinicians on this issue.
Introduction Atrial fibrillation (AF) is the primary cause for cardioembolic stroke [1]. Patients with AF are at risk of early recurrent ischaemic stroke (RS) even after Correspondence: A. H. Abdul-Rahim, Institute of Cardiovascular and Medical Sciences, 44 Church Street, University of Glasgow, Glasgow, G11 6NT, UK (tel.: +44 141 211 1861; fax: +44 141 211 1863; e-mail: [email protected]). *VISTA collaborators are given in the Appendix.
1048
thrombolytic therapy [2]. Amongst the AF cohort, there is excellent evidence to support the use of anticoagulation for RS prevention [3,4]. Nevertheless, there is considerable uncertainty on the optimal latency after acute stroke at which anticoagulation treatment should commence in order to prevent RS without resulting in a symptomatic intracerebral haemorrhage (SICH). The European Stroke Organization and the American Stroke Association do not recommend the use of anticoagulation in
© 2014 EAN
ANTITHROMBOTICS IN RECENT STROKE WITH AF
the hyperacute period post-stroke but neither do they recommend an acceptable delay post-stroke to start anticoagulation [5,6]. Clinically, it seems reasonable to begin anticoagulation as soon as the patient is both medically and neurologically stable. This is often 2 or 3 days post-stroke. This probably achieves a therapeutic anticoagulation level by days 5–7 when the traditional vitamin K antagonist (VKA) is used. However, cardioembolic stroke is associated with an increased risk of haemorrhagic transformation in the first few days post-stroke [7,8]. Two meta-analyses suggested that early anticoagulation post-stroke is inadvisable because it offers no net benefit [9,10]. Nonetheless, there is no evidence to dispute the value of later or prolonged anticoagulation. It appears that there may be a threshold delay after stroke at which the value of anticoagulation switches from neutral to beneficial, at least amongst patients with AF. The associations of antithrombotics therapy commenced (i.e. anticoagulants or antiplatelets) on the patterns and outcomes of modified Rankin Scale (mRS), RS and SICH in a cohort of patients with recent stroke and AF were described. Dichotomized outcomes were also considered as a secondary endpoint (i.e. mortality and good outcome measure at 90 days).
1049
were excluded: baseline National Institutes of Health Stroke Scale (NIHSS), age, medical history, concomitant medication, occurrence of adverse and serious adverse events and mRS at day 90. Only patients who were known to have a medical history of AF or were found to have AF on baseline electrocardiogram were included. Data on anticoagulant treatment were based on warfarin or other VKA treatment started post-stroke. Data on antiplatelet treatment were based on antiplatelet aggregation drug (which includes derivatives of salicylic acid, thienopyridine and dypyridamole) started post-stroke. The novel oral anticoagulants (NOACs) were not prescribed during the available period of the analysis and so were not considered. Follow-up period was 90 days poststroke. Definition of outcome events
Symptomatic RS was defined as any stroke with neurological deterioration, as indicated by NIHSS that was higher by ≥4 points than the value at baseline, or any stroke leading to death. Similarly, SICH was defined as any intracerebral haemorrhage with neurological deterioration, with worsening NIHSS ≥4 points from baseline, or any ICH leading to death. Thus, RS and SICH are each a combined end-point of fatal and non-fatal events.
Methods Statistical methods Data source
A non-randomized cohort analysis was conducted using data obtained from the Virtual International Stroke Trials Archive (VISTA, http://www.vistacollaboration. org/), specifically from the ‘VISTA Acute’ sub-section [11]. VISTA is a collaborative registry that collates and provides access to completed acute stroke trial data (from 1998 to 2008), anonymized in relation to patients’ and trials’ identity, for novel exploratory analyses. Our cohort data do not include trials that investigated thrombolysis therapy in hyperacute stroke. However, they contain data from trials in which thrombolysis was commonly used as standard therapy. Conduct and reporting of our analysis is in accordance with the STROBE guidelines for cohort studies [12]. Participants and variables
Patients who had been randomized to receive placebo or any drug now known to possess no confirmed action on stroke outcome were selected. Patients who lacked the relevant baseline or outcome information
© 2014 EAN
Descriptive statistics were recorded for recent stroke patients with AF, assessing the complete cohort and comparing those who were treated with versus without antiplatelets post-stroke and with versus without anticoagulation post-stroke. Mean (SD) or median [interquartile range (IQR)] are given for continuous variables and count (percentage) for categorical variables. Our primary outcome measures were the ordinal shift of the mRS at day 90 using the full scale and the occurrence of RS and SICH by 90 days post-stroke. For comparison with previous trials, dichotomized outcomes at day 90 (mortality, mRS 0–1, mRS 0–2 and NIHSS 0–1) were reported. The odds ratio (OR) of achieving the studied outcome(s) against the comparator, and corresponding 95% confidence intervals (95% CI), using proportional odds logistic regression (i.e. ordinal regression for mRS distribution and binary regression for dichotomized outcome) were calculated. Adjustments were made for age, baseline NIHSS and thrombolysis treatment [13]. The spread of RS and SICH over the 90-day period were plotted for patients who received anticoagulation
1050
A. H. ABDUL-RAHIM ET AL.
Figure 1 Flow chart for selection of ischaemic stroke patients with atrial fibrillation from VISTA.
treatment and patients who had antiplatelet treatment. Within the cohort who sustained RS, patients were grouped into (i) those who had RS in the absence of anticoagulation; (ii) patients who had RS within 10 days of commencing anticoagulation; and (iii) patients who had RS beyond 10 days of commencing anticoagulation. The cumulative percentage of RS and SICH on each day post-initial stroke by these groupings was plotted. Similar groupings were made for patients treated with antiplatelets and their cumulative percentage of RS and SICH was plotted. All analyses were undertaken using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA).
Results Baseline characteristics
Of the 10 304 ischaemic stroke patients available from VISTA, 8060 were excluded as they were in sinus rhythm. Individual patients’ data were obtained for the remaining 1644 stroke patients with AF, which formed our cohort (Fig. 1). From this
post-stroke cohort, 518 (31%) were given anticoagulation treatment alone, 162 (10%) were given antiplatelets alone and 782 (48%) received a combination of the two. One hundred and eighty-two (11%) did not receive any antithrombotic therapy. The median start time for anticoagulation treatment was at day 2 (IQR 1–3) post-stroke, whilst the median start time for antiplatelet treatment was at day 1 (IQR 1–3) post-stroke. Patients who received anticoagulation were commonly younger and had less severe baseline NIHSS. Patients who received no antithrombotic therapy tended to have more severe baseline NIHSS. Detailed baseline characteristics are given in Tables 1 and S1. The proportion of patients who were on antithrombotic treatment prior to the initial stroke (i.e. at baseline) are as follows: anticoagulant agent only, 286 (17%); antiplatelet agent only, 488 (30%); and combination of the two, 78 (5%). Outcome
In this cohort, 157 (10%) patients had RS, 50 (3%) patients sustained SICH and 390 (24%) patients died from any cause, by day 90 (Table 2). Combined antithrombotic therapy with anticoagulation and antiplatelet was associated with more favourable functional outcome across full scale mRS at 90 days after adjustment for age, baseline NIHSS and thrombolysis treatment [OR 1.79 (95% CI 1.32–2.42)], compared with no antithrombotic therapy received. The treatment effect of combined antithrombotic therapy lost statistical significance when outcome was dichotomized as mRS 0–1, mRS 0–2 and NIHSS 0–1. Combined antithrombotic therapy was associated with significantly reduced risk of RS, SICH and mortality, by day 90, compared with no antithrombotic therapy received [RS, OR 0.33 (95% CI 0.21–0.53); SICH, OR 0.18 (95% CI 0.09–0.37); and mortality,
Table 1 Baseline characteristics of the cohort according to antithrombotic treatment received following stroke Antithrombotic treatment received post-stroke
Variables Male, n (%) Age, years, mean (SD) Baseline NIHSS, median (IQR) Received thrombolysis treatment, n (%) Medical history Previous stroke, n (%) Transient ischaemic attack, n (%)
No antithrombotic (no anticoagulation, no antiplatelet), n = 182
Antiplatelet only, n = 162
Anticoagulation treatment only, n = 518
Anticoagulation treatment and antiplatelet, n = 782
85 77.1 16 71
85 75.8 15 51
243 73.4 14 203
359 74.2 14 293
(46.7%) (9.0) (11–21) (39.0%)
46 (25.3%) 11 (6.7%)
(52.5%) (10.0) (11–19) (31.5%)
54 (33.3%) 10 (6.8%)
(46.9%) (10.1) (10–18) (39.2%)
108 (20.9%) 45 (9.1%)
(45.9%) (9.8) (9–18) (37.5%)
150 (19.2%) 70 (9.3%)
n, number of observations; NIHSS, National Institutes of Health Stroke Scale; IQR, interquartile range.
© 2014 EAN
ANTITHROMBOTICS IN RECENT STROKE WITH AF
1051
Table 2 Clinical outcome at 90 days (adjusted for age, baseline NIHSS and thrombolysis treatment) No anticoagulant SICH, n/N (%) Antiplatelet 3/162 (1.9%) No antiplatelet 17/182 (9.3%) OR (95% CI) 0.24 (0.07–0.88) P value 0.032 All 20/344 (5.8%) Symptomatic RS, n/N (%) Antiplatelet 14/162 (8.6%) No antiplatelet 36/182 (19.8%) OR (95% CI) 0.43 (0.22–0.86) P value 0.016 All 50/334 (14.5%) Mortality, n/N (%) Antiplatelet 45/162 (27.8%) No antiplatelet 74/182 (40.7%) OR (95% CI) 0.60 (0.35–1.02) P value 0.061 All 119/334 (34.6%)
Anticoagulant
OR (95% CI)
P value
All
15/782 (1.9%) 15/518 (2.9%) 0.59 (0.28–1.24) 0.164 30/1300 (2.3%)
0.92 (0.25–3.37) 0.43 (0.20–0.94)
0.904 0.034
18/944 (1.9%) 32/700 (4.6%) 0.42 (0.22–0.78) 0.006
0.49 (0.26–0.92)
0.027
52/782 (6.7%) 55/518 (10.6%) 0.60 (0.40–0.91) 0.015 107/1300 (8.2%)
0.78 (0.41–1.47) 0.61 (0.37–1.01)
0.439 0.056
0.62 (0.42–0.91)
0.013
139/782 (17.8%) 132/518 (25.5%) 0.58 (0.43–0.78)
