Clinical Outcomes and Neuroimaging Profiles in Nondisabled Patients With Anticoagulant-Related Intracerebral Hemorrhage
Abstract
Background and Purpose—
The aim of this study was to prospectively validate our prior findings of smaller hematoma volume and lesser neurological deficit in nonvitamin K oral anticoagulant (NOAC) compared with Vitamin K antagonist (VKA)-related intracerebral hemorrhage (ICH).
Methods—
Prospective 12-month observational study in 15 tertiary stroke centers in the United States, Europe, and Asia. Consecutive patients with premorbid modified Rankin Scale score of <2 with acute nontraumatic anticoagulant-related ICH divided into 2 groups according to the type of anticoagulant: NOAC versus VKA. We recorded baseline ICH volume, significant hematoma expansion (absolute [12.5 mL] or relative [>33%] increase), neurological severity measured by National Institutes of Health Stroke Scale score, 90-day mortality, and functional status (modified Rankin Scale score).
Results—
Our cohort comprised 196 patients, 62 NOAC related (mean age, 75.0±11.4 years; 54.8% men) and 134 VKA related (mean age, 72.3±10.5; 73.1% men). There were no differences in vascular comorbidities, antiplatelet, and statin use; NOAC-related ICH patients had lower median baseline hematoma volume (13.8 [2.5–37.6] versus 19.5 [6.6–52.0] mL; P=0.026) and were less likely to have severe neurological deficits (National Institutes of Health Stroke Scale score of >10 points) on admission (37% versus 55.3%, P=0.025). VKA-ICH were more likely to have significant hematoma expansion (37.4% versus 17%, P=0.008). NOAC pretreatment was independently associated with smaller baseline hematoma volume (standardized linear regression coefficient:−0.415 [95% CI, −0.780 to −0.051]) resulting in lower likelihood of severe neurological deficit (odds ratio, 0.44; 95% CI, 0.22−0.85) in multivariable-adjusted models.
Conclusions—
Patients with NOAC-related ICH have smaller baseline hematoma volumes and lower odds of severe neurological deficit compared with VKA-related ICH. These findings are important for practicing clinicians making anticoagulation choices.
Introduction
Therapeutic anticoagulation is the mainstay of treatment for secondary stroke prevention in patients with atrial fibrillation.1 Nonvitamin K Antagonist Oral Anticoagulants (NOACs) are increasingly used in the past years, as they present significant practical advantages and ease of use compared with the traditional solution of vitamin K antagonists (VKAs), namely warfarin. NOACs were approved on the basis of noninferiority to warfarin, an association that was driven mainly by their safety profile, resulting in lower rates of intracerebral hemorrhage (ICH) while the overall composite ischemic and hemorrhagic stroke rates were comparable.2–4 Despite the higher overall ICH risk with VKA, case fatality and outcomes of ICH occurring in the context of randomized Phase III trials were comparable between NOACs and VKAs.5,6 Subsequent observational studies of real-world data suggest that NOAC-related ICH have a trend toward better functional outcomes,7–9 despite lacking a safe and efficient reversal agent. The main driver behind this seems to be lower ICH volume,7,8,10 which is well established as the single most potent predictor of hematoma expansion, mortality, and functional outcome in ICH.11
In light of the former considerations, we sought to validate the findings of our prior analysis7 in a separate prospective cohort, including ICH patients without any prior disability. We hypothesize that NOAC use is independently associated with smaller ICH volume, lower likelihood of hematoma expansion and mass effect, less severe neurological deficit, lower mortality, as well as higher odds of favorable functional outcomes.
Methods
The data that support the findings of this study are available from the corresponding author on reasonable request.
Study Population
We prospectively enrolled consecutive patients with no prior disability (modified Rankin Scale (mRS) scores before the index event 0 or 1) who presented with nontraumatic, ICH and positive history of oral anticoagulant intake in the emergency rooms of the participating tertiary care stroke centers during a 1-year period (August 2016–July 2017). Active surveillance and screening of all ICH patients were conducted to ensure inclusion of all eligible patients irrespective of admission to neurosurgical, neurological, or intensive care unit service. The definition of VKA-related ICH required effective use of a VKA with an international normalized value of >1.5 on hospital admission,12 whereas the definition of NOAC-related ICH required confirmed reception of the relevant NOAC during the past 24 hours before the index event.13 NOAC use was confirmed with patients and their caregivers. Patients with major head trauma or known underlying structural or vascular cause of ICH were excluded from further evaluation. We also excluded patients with hemorrhagic transformation of ischemic infarcts and patients with pure intraventricular hemorrhage without evidence of intraparenchymal hematoma.
Baseline Characteristics and Outcomes
The following parameters were prospectively recorded for all included patients: (1) demographic characteristics (age, sex, and race); (2) history of vascular risk factors (diabetes mellitus, hypertension, current smoking, hypercholesterolemia, coronary artery disease, heart failure, and chronic kidney disease, defined as either history of kidney damage or decreased kidney function [glomerular filtration rate level <60 mL/min per 1.73 m2] for ≥3 months); (3) prior history of ischemic or hemorrhagic stroke; (4) prior medications (statins, antiplatelets, dose, and type of oral anticoagulation); (5) laboratory test values on admission (international normalized ratio [INR], activated partial thromboplastin time, and total platelet count); (6) National Institutes of Health Stroke Scale (NIHSSadm) and Glasgow Coma Scale on admission; (7) admission systolic and diastolic blood pressures, measured using automated cuffs.
Noncontrast head computed tomography (CT) scans were performed for all patients at both baseline and within 24-hour postictus, and all volumes were measured with the same method and imaging modality. CT findings were interpreted and extracted independently by either experienced neurologists or neuroradiologists in each participating institution that were unaware of each patient’s clinical data. Hematoma volume at both baseline and follow-up (within the first 24-hour) CT scans was calculated with the ABC/2 method,14 whereas hematoma expansion at 24 hours was defined as an absolute increase of >12.5 mL or a relative increase of >33% in hematoma volume at the subsequent 24-hour CT scan compared with the admission CT scan.2 Intraventricular hemorrhage was classified as present versus absent in the head CT. Repeat CT scans were also systematically evaluated and compared with the admission CT scan for the presence of edema or midline shift within the first 24 hours from the index event, crudely classified as present versus absent.
In the participating centers all eligible patients were treated according to current guidelines for the management of spontaneous ICH.15 Reversal therapies using specific antidotes (eg, idarucizumab), or coagulation factors, (fresh frozen plasma or prothrombin complex concentrates) were used according to individual hospital practices and availability but with uniform goal of restoration of INR <1.4. Intubation, surgical decompression, and external ventricular drainage were offered as per standard of care treatment in each individual hospital. All centers aimed for systolic blood pressure <140 mm Hg, using the agents of choice/standard of care in each hospital. NIHSS score at 24 hours (NIHSS24h) and at hospital discharge (NIHSSdis), as well as mRS at both discharge and at 3 months was obtained as standard of care for all patients. Disability at 3 months was defined as a score of >2 in the mRS.2 Three-month outcome was obtained as part of regular clinical follow-up in the participating sites.
The study was approved by the local institutional review board in each participating center, written informed consent was waived.
Statistical Analyses
Details on sample size estimation can be found in the online-only Data Supplement.
Continuous variables are presented as mean±SD (normal distribution) or as median with interquartile range (IQR; skewed distribution), while categorical variables are presented as percentages. Statistical comparisons between different subgroups were performed using the Pearson χ2 test, unpaired t test, and Mann-Whitney U test, where appropriate. The distributions on the mRS score among groups both at discharge and at 3-month follow-up were compared with the Cochran-Mantel-Haenszel test and univariable/ multivariable ordinal logistic regression (shift analysis). The cumulative probability of survival during the 3-month follow-up period was estimated with the Kaplan-Meier product-limit method. Comparisons between the 2 groups were performed by the log-rank test.
Multivariable regression analyses were performed for all outcomes that were found to differ (P<0.1) between NOAC-related and VKA-related groups in initial univariable analyses. In univariable models of all baseline characteristics a threshold of P<0.1 was used to identify candidate variables for inclusion in the multivariate regression models that tested statistical significance hypothesis using the likelihood ratio test with an α value of 0.05. Because of the presence of a strong linear correlation between admission NIHSS score and baseline ICH volume (regression coefficient=0.13; P<0.001), we decided a priori not to include admission NIHSS score and baseline ICH volume simultaneously as independent and dependent variables in the same multivariable models taking into account that the emergence of other important associations could be scattered because of the presence of multicollinearity. Admission NIHSS scores were included as an additional potential confounder in the univariable and multivariable analyses of 3-month follow-up outcomes. We reported all associations as linear regression coefficients in linear regression models, odds ratios (ORs) in logistic regression models and common ORs in ordinal regression models, respectively. Before univariable and multivariable linear regression analyses, baseline hematoma volume was cube root transformed for each patient to satisfy statistical assumptions about normality of the distribution.7,8 The Stata Statistical Software Release 13 for Windows (College Station, TX, StataCorp LP) was used for the aforementioned statistical analyses.
Results
Clinical, Demographic, and Laboratory Characteristics
Our final cohort comprised 196 ICH patients from 15 tertiary care stroke centers in North America (n=7), Europe (n=7), and Asia (n=1): 62 NOAC-related (mean age, 75.0±11.4 years) and 134 VKA-related (mean age, 72.3±10.5 years) ICH. Baseline characteristics are summarized in Table 1. Apixaban was the most commonly used NOAC (47%) followed by Rivaroxaban (37%) and Dabigatran (16%). VKA-related ICH patients were more likely to be male (73.1% versus 54.8%, P=0.011) and have lobar location (60.4% versus 32.2%, P<0.001). There were no differences in vascular comorbidities, CHA2DS2VASc and HAS-BLED scores (hypertension, abnormal renal and liver function, stroke, bleeding, labile INR, elderly, drugs or alcohol), antiplatelet, and statin pretreatment between the 2 groups. VKA-ICH patients had significantly higher INR (3.05±1.30 versus 1.45±0.62, P<0.001) and activated partial thromboplastin time (40.7±12.4 versus 33.4±7.9, P<0.001) without other differences in baseline laboratory values.
| Variable | NOACs (n=62) | VKAs (n=134) | P Value |
|---|---|---|---|
| Baseline clinical characteristics | |||
| Age, y (mean±SD) | 75.0±11.4 | 72.3±10.5 | 0.107 |
| Males, % | 54.8 | 73.1 | 0.011 |
| BMI, mean±SD | 23.9±12.2 | 26.4±12.3 | 0.231 |
| Race, % | |||
| White | 78.3 | 73.1 | |
| Black | 15.0 | 17.9 | |
| Asian | 6.7 | 8.2 | |
| Hispanic | 0 | 0.8 | |
| CHA2DS2VASc score, mean±SD | 4.5±1.5 | 4.2±1.5 | 0.234 |
| HAS-BLED score, mean±SD | 2.7±1.1 | 2.6±1.1 | 0.570 |
| Hypertension, % | 93.5 | 95.5 | 0.567 |
| Diabetes mellitus, % | 35.5 | 33.5 | 0.794 |
| Hyperlipidemia, % | 61.3 | 67.9 | 0.364 |
| Heart failure, % | 25.8 | 30.6 | 0.492 |
| Current smoking, % | 11.7 | 13.9 | 0.666 |
| Coronary artery disease, % | 35.5 | 41.0 | 0.458 |
| Chronic kidney disease, % | 11.3 | 20.1 | 0.128 |
| Prior history of ischemic stroke, % | 38.7 | 29.8 | 0.219 |
| Prior history of intracerebral hemorrhage, % | 6.4 | 1.5 | 0.061 |
| Statin pretreatment, % | 57.4 | 67.2 | 0.186 |
| Antiplatelet pretreatment, % | 42.6 | 46.6 | 0.604 |
| Dual antiplatelet pretreatment, % | 8.2 | 8.9 | 0.862 |
| NOAC type, n (%) | Apixaban: 29 (46.8) | … | … |
| Dabigatran: 10 (16.1) | |||
| Rivaroxaban: 23 (37.1) | |||
| NIHSS admission, median (IQR) | 7 (2–14) | 12 (4–21) | 0.090 |
| Severe neurological deficit on admission,* % | 37.0 | 55.3 | 0.025 |
| GCS admission, median (IQR) | 14 (7–15) | 13 (7–15) | 0.458 |
| SBP admission, mm Hg (mean±SD) | 168.5±28.2 | 166.7±32.8 | 0.711 |
| DBP admission, mm Hg (mean±SD) | 88.4±16.7 | 91.0±19.2 | 0.358 |
| Baseline laboratory values | |||
| INR admission, mean±SD | 1.45±0.62 | 3.05±1.30 | <0.001 |
| aPTT admission, s (mean±SD) | 33.4±7.9 | 40.7±12.4 | <0.001 |
| Platelet count×103/μL, median (IQR) | 217 (176–267) | 201 (162–254) | 0.174 |
| CrCl on admission, mL/min (mean±SD) | 71.7±26.2 | 65.9±27.2 | 0.189 |
| Baseline CT findings | |||
| Lobar hemorrhage, % | 32.2% | 60.4% | <0.001 |
| Intraventricular hemorrhage, % | 41.9% | 47.0% | 0.507 |
| Baseline ICH volume, mL, median (IQR) | 13.8 (2.5–37.6) | 19.5 (6.6–52.0) | 0.026 |
| ICH score, median (IQR) | 2 (1–3) | 2 (1–3) | 0.987 |
| Severe ICH,† % | 35.5 | 34.6 | 0.902 |
aPTT indicates activated partial thromboplastin time; BMI, body mass index; CrCl, creatinine clearance; CT, computed tomography; DBP, diastolic blood pressure; GCS, Glasgow Coma Scale; ICH, intracerebral hemorrhage; IQR, interquartile range; INR, international normalized ratio; NIHSS, National Institutes of Health Stroke Scale; NOAC, nonvitamin K oral anticoagulant; SBP, systolic blood pressure; and VKAs, vitamin K antagonists.
*
Defined as NIHSS admission >10.
†
Defined as an ICH score of >2.
ICH severity was comparable in both groups with a mean ICH score of 2 and comparable Glasgow Coma Scale (Table 1). Although there was only a nonsignificant trend toward lower median NIHSSadm in NOAC-ICH (7 points [2–14] versus 12 [4–21]; P=0.09), there was a substantially lower proportion of NOAC-ICH patients with severe neurological deficit (37% versus 55.3%; P=0.025), although this was likely mediated by lower ICH volume in the NOAC group; ICH volume was excluded from the analysis because of multicollinearity with NIHSS as explained in the Methods section.
Neuroimaging Characteristics and Outcomes
NOAC-related ICH was associated with lower baseline hematoma volume (13.8 mL [IQR, 2.5–37.6] versus 19.5 mL [IQR, 6.6–52.0]; P=0.026; Table 1; Figure 1). Similarly, NOAC-related ICH had lower median 24-hour hematoma volume (15.9 mL [IQR, 2.5–32.7] versus 20.0 mL [IQR, 6.8–59]; P=0.021; Table 2) and were less likely to experience hematoma expansion (17% versus 37.4%, P=0.008) and midline shift (30.6% versus 47%, P=0.031). There was a nonsignificant trend toward lower likelihood for cerebral edema in NOAC-related ICH (62.9% versus 74.6%, P=0.093; Table 2). To investigate further whether hematoma expansion was influenced by the use of prothrombin complex concentrate we performed a subgroup analysis within VKA-ICH patients and found no significant difference in the rates of hematoma expansion (40.0% versus 31.4%, P=0.382) between patients treated with prothrombin complex concentrate (n=85) and those receiving no prothrombin complex concentrate treatment (n=49). No hematoma expansion occurred in any of the 2 NOAC-ICH patients receiving specific reversal agents (idarucizumab or andexanet alfa).
| Variable | NOACs (n=62) | VKAs (n=134) | P Value |
|---|---|---|---|
| Therapies—interventions, % | |||
| Use of reversal agents | 48.4 | 86.6 | <0.001 |
| FFP | 6.4 | 20.1 | 0.014 |
| FEIBA or PCC | 44.2 | 63.4 | 0.012 |
| Vitamin K | 9.7 | 80.6 | <0.001 |
| Idarucizumab | 1.6 | … | … |
| Andexanet alfa | 1.6 | … | … |
| Surgical decompression | 3.2 | 11.1 | 0.065 |
| External ventricular drainage | 9.7 | 17.9 | 0.137 |
| Intubation | 29.0 | 33.6 | 0.526 |
| Radiological outcomes at 24 h | |||
| ICH volume at 24 h, median (IQR) | 15.9 (2.5–32.7) | 20.0 (6.8–59) | 0.021 |
| ICH volume >30 mL at 24 h, % | 29.2 | 39.2 | 0.222 |
| Hematoma expansion, % | 17.0 | 37.4 | 0.008 |
| Cerebral edema, % | 62.9 | 74.6 | 0.093 |
| Midline shift, % | 30.6 | 47.0 | 0.031 |
| Clinical outcomes | |||
| NIHSS score at 24 h, median (IQR) | 7 (2–18) | 13 (4–23) | 0.063 |
| Days of hospitalization, median (IQR) | 7 (3–15) | 7 (3–16) | 0.587 |
| mRS score at discharge,* median (IQR) | 4 (2–6) | 4 (2–6) | 0.304 |
| In-hospital mortality, % | 27.4 | 38.0 | 0.146 |
| mRS score at 3 mo,† median (IQR) | 3 (1–6) | 4 (2–6) | 0.071 |
| Disability at 3 mo,‡ %† | 53.5 | 65.6 | 0.121 |
| Mortality at 3 mo, %† | 32.1 | 45.3 | 0.095 |
FEIBA indicates factor eight inhibitor bypassing activity; FFP, fresh frozen plasma; ICH, intracerebral hemorrhage; IQR, interquartile range; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; NOAC, nonvitamin K oral anticoagulant; and VKAs, vitamin K antagonists.
*
Calculated with the Cochran-Mantel-Haenszel test.
†
Data not available for 12 patients.
‡
Defined as mRS score of >2.
Clinical Outcomes
VKA-related ICH patients were more likely to receive coagulation factors (86.6% versus 48.4%, P<0.001), but the rates of surgical decompression or external ventricular drainage placement were similar (Table 2). There was no statistically significant difference in in-hospital mortality rate (27.4% versus 38.0%), length of hospitalization and mRS score at discharge (Table 2). NOAC-related ICH patients showed a trend toward lower 3-month mortality (32.1% versus 45.3%; P=0.095; Figure 1). Similarly, there was a trend toward better 3-month functional status in NOAC- versus VKA-related ICH patients (median 3-month mRS score of 3 [IQR, 1–6] versus 4 [IQR, 2–6]; shift analysis in mRS scores P=0.071 by Cochran-Mantel-Haenszel test; Figure 2).
Multivariable Analyses
Table 3 presents the associations of baseline characteristics, clinical parameters, neuroimaging findings, and therapeutics interventions with baseline hematoma volume (cube root transformed) on simple and multiple linear regression analyses. NOAC-related ICH was associated with lower baseline hematoma volumes independent of potential confounders (standardized linear regression coefficient =−0.415 [95% CI, −0.780 to −0.051]; P=0.026; Table 3). In multivariable logistic regression, NOAC pretreatment was independently related to lower odds of severe neurological deficit on admission (OR, 0.44; 95% CI, 0.22–0.85; P=0.016; Table I in the online-only Data Supplement).
| Simple Linear Regression | Multiple Linear Regression | |||
|---|---|---|---|---|
| Linear Regression Coefficient (95% CI) | P Value | Linear Regression Coefficient (95% CI) | P Value | |
| Age, y | 0.005 (−0.011 to 0.021) | 0.526 | … | … |
| Males | 0.142 (−0.166 to 0.451) | 0.363 | … | … |
| BMI | 0.010 (−0.004 to 0.024) | 0.189 | … | … |
| Hypertension | 0.201 (−0.567 to 0.970) | 0.606 | … | … |
| Diabetes mellitus | 0.089 (−0.268 to 0.447) | 0.623 | … | … |
| Hyperlipidemia | −0.230 (−0.587 to 0.126) | 0.205 | … | … |
| Heart failure | 0.278 (−0.095 to 0.649) | 0.144 | … | … |
| Current smoking | −0.096 (−0.597 to 0.404) | 0.705 | … | … |
| Coronary artery disease | 0.167 (−0.180 to 0.514) | 0.344 | … | … |
| Kidney failure | 0.257 (−0.189 to 0.704) | 0.258 | … | … |
| Prior history of ischemic stroke | 0.040 (−0.322 to 0.403) | 0.826 | … | … |
| Prior history of intracerebral hemorrhage | −0.716 (−1.698 to 0.264) | 0.151 | … | … |
| Statin pretreatment | −0.103 (−0.458 to 0.253) | 0.569 | … | … |
| Antiplatelet pretreatment | 0.353 (0.012 to 0.694) | 0.042 | 0.316 (−0.021 to 0.654) | 0.066 |
| Dual antiplatelet pretreatment | 0.311 (−0.292 to 0.915) | 0.310 | … | … |
| NOAC pretreatment | −0.460 (−0.820 to −0.101) | 0.012 | −0.415 (−0.780 to −0.051) | 0.026 |
| Admission SBP | 0.006 (0.001 to 0.011) | 0.026 | 0.004 (−0.002 to 0.011) | 0.197 |
| Admission DBP | 0.011 (0.001 to 0.020) | 0.020 | 0.004 (−0.007 to 0.016) | 0.474 |
BMI indicates body mass index; DBP, diastolic blood pressure; NOAC, nonvitamin K oral anticoagulant; and SBP, systolic blood pressure.
The initial univariable associations between NOAC pretreatment and lower likelihood of hematoma expansion (Table II in the online-only Data Supplement), cerebral edema (Table III in the online-only Data Supplement), and midline shift (Table IV in the online-only Data Supplement) were attenuated in multivariable analyses, including baseline hematoma volume as a confounder and did not reach the level of statistical significance. Similarly, NOAC pretreatment was not associated with 3-month functional improvement (decrease by 1 point across all mRS scores; Table V in the online-only Data Supplement) and lower odds of 3-month mortality (Table VI in the online-only Data Supplement) on multivariable ordinal and binary logistic regression models, when baseline hematoma volume and NIHSSadm were included as confounders. Increasing baseline hematoma volume was independently related to lower odds of functional improvement (common OR per 1 mL increase: 0.97; 95% CI, 0.95–0.99; P<0.001) and increased likelihood of 3-month mortality (OR per 1 mL increase: 1.03; 95% CI, 1.01–1.05; P<0.001).
Discussion
In this prospective multicenter cohort study of patients with nontraumatic anticoagulant-associated ICH without prior history of disability, we validated our prior findings suggesting a more favorable neuroimaging and clinical profile in patients with NOAC-related ICH compared with those with VKA-related ICH. Specifically, NOAC pretreatment was independently associated with lower baseline hematoma volume and lower odds of severe neurological deficit on admission. We also detected a trend toward more favorable 3-month functional outcomes in NOAC-pretreated patients.
The association of lower hematoma volume with NOAC pretreatment is the most salient of our findings as ICH volume has been well established as the most potent predictor of mortality and adverse functional outcomes.2,16 This observation is in line with the findings of our previous multicenter cohort7 and with animal data.17 The exact reason is not clear but different pharmacological properties and specifically a more selective coagulation inhibition mechanism in NOACs might partially account for this. In the current cohort, VKA-pretreated patients not only had significantly higher INR than NOAC-pretreatred patients, but the median INR of 3.05 was above the upper margin of the generally accepted as therapeutic INR range, suggesting that many VKA-treated patients had markedly supratherapeutic INR resulting in more severe disruption of the coagulation cascade.18 Our findings are in line with a recently published meta-analysis documenting a favorable trend toward lower ICH volume in NOAC-associated ICH (standard mean difference: −0.24; 95% CI, −0.52 to 0.04; P=0.093).10
The results of our study are not without controversy as another multicenter cohort study found contradictory results, primarily noting no difference in admission hematoma volume between NOACs- and VKA-related ICHs.13 Three marked differences may explain this. First, in the study by Wilson et al,13 many centers report nonconsecutive patient selection which might have introduced substantial bias. Second, ICH volume measurement was not consistent among contributing centers: both the ABC/2 method and semiautomated segmentation methods were used, which may have introduced discrepancies in ICH volume estimation. Finally, patients with prior disability before the OAC-related ICH were not excluded in the study by Wilson et al.
Additionally, NOAC pretreatment was associated with lower odds of significant hematoma expansion and midline shift at the 24-hour follow-up CT scan. The incidence of hematoma expansion in NOAC-related hemorrhage was 17%, comparable to our previous cohort (23%)7 but markedly different compared with the study by Wilson et al13 (40%). A possible explanation is that in our study, we used a more stringent cutoff (>12.5 mL) compared with a lower threshold (>6 mL) used by the study by Wilson et al,13 which could explain the lower incidence of hematoma expansion in our cohort. However, an alternative explanation is that likely the hematoma expansion occurred in a smoother, more homogeneous fashion in NOACs, that is, many NOAC-ICHs increased by small amounts as opposed to fewer but more dramatic expansions in VKA-ICH. This potential explanation is further supported by the increase of the upper 95% CI in the follow-up CT compared with baseline CT in the VKA-ICH group (19.5 mL [6.6–52.0] to 20.0 mL [6.8–59]) in contrary to the NOAC-ICH group (13.8 mL [2.5–37.6] to 15.9 mL [2.5–32.7]). Furthermore, the rate of hematoma expansion in VKA-related ICH in our cohort (37%) is comparable to both the study by Wilson et al13 and a recent meta-analysis (34%),10 despite differences in cutoff limits. Therefore, the significant difference in hematoma expansion in these 2 cohorts is driven by our study’s substantially lower rates in the NOAC group. This is unsurprising given that ICH volume is the most potent predictor of all of these imaging outcomes.19 Indeed there was collinearity between NOAC treatment and ICH volume, rendering the former insignificant in multivariable-adjusted models and suggesting that the association between NOAC treatment and more favorable neuroimaging outcomes is mediated by the lower hematoma volume. However, the marked discrepancy between the reported studies is of interest and merits more in-depth research, as hematoma expansion is strongly associated with functional outcome and mortality.
The 3-month mortality was lower in the NOAC-pretreated patients (32% versus 45%) although this did not reach statistical significance. The difference of ≈13% is similar to the findings from our prior cohort7 although in absolute terms the rates are lower (22% versus 36%). In the study by Wilson et al,13 NOAC-related ICH mortality was similar to our cohort’s (33%) but VKA-ICH mortality was substantially lower (31%). The reason for the different outcomes in VKA-ICH between the 2 cohorts is not clear; it should be noted that in the study by Wilson et al,13 VKA-treated patients had lower premorbid mRS score (0 versus 1) and had lower rates of early palliation indicating that there might have imbalances in prescription practices with a higher overall risk population overrepresented in the NOAC group.
A similar pattern was observed in-hospital mortality, with a trend toward lower rates in the NOAC-pretreated group (27% versus 38%). This observation is corroborated by the results of a recent meta-analysis reporting 10% lower in-hospital mortality rates in ICH patients pretreated with NOACs (17% versus 27%).10 A recent study by Inohara et al9 evaluating ≈140 000 patients hospitalized with ICH reported significantly lower rates of in-hospital mortality in ICH patients pretreated with NOACs (26.5%) than with VKAs (32.6%). Although the difference in the rates of in-hospital mortality did not reach the level of statistical significance in our cohort, this may be attributed to inadequate power. Indeed, in the large sample studied by Inohara et al9 (15 036 cases of VKA-related ICH and 4918 cases with NOAC-related ICH), NOAC pretreatment was independently related to lower risk of in-hospital mortality, with an adjusted risk difference of −5.7% that was accentuated to −12% in the subgroup of patients with supratherapeutic INR. We think that, in similar fashion to neuroimaging outcomes, the trends toward better clinical outcomes in the NOAC group are mediated by the smaller ICH volume.
It should also be noted that the above outcomes occurred, despite significantly more frequent coagulation factor use in VKA patients (87% versus 48%), an intervention with established effect on limitation of hematoma expansion that may also translate into lower rates of in-hospital mortality.12 Despite the lack of targeted anticoagulation reversal agent for NOAC in our cohort (1 case treated with idarucizumab, 1 with andexanet alfa), all neuroimaging and clinical outcomes were in favor of this group, and the rates of significant hematoma expansion were low. It would be reasonable to expect that NOAC-specific reversal agents will have a beneficial effect, further limiting hematoma expansion and improving clinical outcomes.20 Although idarucizumab was approved by the Food and Drug Administration in October 2015, there was a lag between its approval and widespread adoption and availability in-hospital pharmacies. Andexanet alfa Food and Drug Administration approval was delayed till March 2018 after the request for additional information by the Food and Drug Administration after the initial review in August 2016. Therefore, it was not available for widespread use during the period that the study was conducted. However, the only Phase 3 trial examining the efficacy of a NOAC-specific reversal agent did not include comparator control group and, in the ICH subgroup, there was neither assessment of time to bleeding cessation nor hematoma expansion on follow-up head CT. The relatively benign natural history of NOAC-ICH emerging from our findings raises the question whether the efficacy and safety of NOAC reversal agents needs to be better demonstrated before widespread use or whether there are specific subgroups that might benefit more, such as elderly, patients with renal or hepatic dysfunction.
Our study has limitations: The time between symptom onset and brain CT was not recorded, which did not allow us to account for the effect of this variable on hematoma expansion.21 Assignment of anticoagulant group was not randomized, which might have introduced unmeasured confounding. Practical experience from the United States suggests that affordability of NOACs is often a decisive factor which might have resulted in socioeconomic imbalances that could not be accounted for. However, there were no differences in age or race between the 2 groups. Specific laboratory assessment to assess to confirm NOAC use was not performed routinely in all centers and, therefore, we relied on historical confirmation of NOAC use. Biological confirmation of NOAC use would have been optimal as this might have introduced patients with history of NOAC prescription but not active use at the time of ICH. Further, the relatively small number of patients per contributing center did not allow for subgroup analyses according to geographical location, and the uniformity of hospital setting (tertiary center) did not allow for exploration of differences according to hospital type. Accordingly, there was inevitable heterogeneity in management choices, as highlighted by the range practices in coagulation factors and reversal agents use. Similarly, the small number of NOAC-related hemorrhage did not allow subgroup analyses according to specific NOAC use. Perihematoma edema is challenging to define radiographically and measure accurately.22 Therefore, our reporting on this particular imaging outcome is inevitably crude. A widely accepted operational definition with well-defined imaging parameters would be important, and accurate measurement of it should ideally be performed with the use of semiautomated quantitative imaging software. Last, similar to our prior study, magnetic resonance imaging was not systematically performed in our study leaving the potential moderating effect of cerebral microbleeds uncertain.
However, potential strengths of our study include a large sample size for a prospective cohort, predefined outcomes of interest and statistical analysis plan, participation of multiple centers across United States, Europe, and Asia, limiting the risk of regional practice bias. In contrary to all prior relevant studies, we limited our analyses to functionally independent patients before the index events, resulting in a more homogeneous cohort and eliminating the effect of premorbid disability on mortality and functional outcomes. Furthermore, all patients included in the present report can be expected that they had been anticoagulated, because only patients with INR >1.5 were included in the VKA-ICH group, while last intake <24 hours was mandatory in the NOAC-ICH group. The median INR value of 1.5 in the NOAC-ICH group indicates that most of the patients of this group may have been at peak NOAC levels when ICH occurred.
In summary, the present prospective multicenter cohort study confirms our prior findings suggestive of lower hematoma volume, lower likelihood of hematoma expansion, and lower odds of severe neurological deficit in ICH patients pretreated with NOACs in comparison to VKA-related ICH. These findings highlight NOACs as an attractive treatment option for patients in need of anticoagulation, especially those in high risk for ICH. Continued surveillance and further validation of these findings are larger, well-defined prospective cohorts are necessary, to confirm and further refine these findings.
Supplemental Material
File (str_stroke-2018-021979_supp1.pdf)
- Download
- 321.58 KB
References
1.
Tereshchenko LG, Henrikson CA, Cigarroa J, Steinberg JS. Comparative effectiveness of interventions for stroke prevention in atrial fibrillation: a network meta-analysis. J Am Heart Assoc. 2016;5:e003206.
2.
Anderson CS, Heeley E, Huang Y, Wang J, Stapf C, Delcourt C, et al; INTERACT2 Investigators. Rapid blood-pressure lowering in patients with acute intracerebral hemorrhage. N Engl J Med. 2013;368:2355–2365. doi: 10.1056/NEJMoa1214609
3.
Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al; ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883–891. doi: 10.1056/NEJMoa1009638
4.
Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al; ARISTOTLE Committees and Investigators. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981–992. doi: 10.1056/NEJMoa1107039
5.
Hart RG, Diener HC, Yang S, Connolly SJ, Wallentin L, Reilly PA, et al. Intracranial hemorrhage in atrial fibrillation patients during anticoagulation with warfarin or dabigatran: the RE-LY trial. Stroke. 2012;43:1511–1517. doi: 10.1161/STROKEAHA.112.650614
6.
Hankey GJ, Stevens SR, Piccini JP, Lokhnygina Y, Mahaffey KW, Halperin JL, et al; ROCKET AF Steering Committee and Investigators. Intracranial hemorrhage among patients with atrial fibrillation anticoagulated with warfarin or rivaroxaban: the rivaroxaban once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and embolism trial in atrial fibrillation. Stroke. 2014;45:1304–1312. doi: 10.1161/STROKEAHA.113.004506
7.
Tsivgoulis G, Lioutas VA, Varelas P, Katsanos AH, Goyal N, Mikulik R, et al. Direct oral anticoagulant- vs vitamin K antagonist-related nontraumatic intracerebral hemorrhage. Neurology. 2017;89:1142–1151. doi: 10.1212/WNL.0000000000004362
8.
Wilson D, Charidimou A, Shakeshaft C, Ambler G, White M, Cohen H, et al; CROMIS-2 Collaborators. Volume and functional outcome of intracerebral hemorrhage according to oral anticoagulant type. Neurology. 2016;86:360–366. doi: 10.1212/WNL.0000000000002310
9.
Inohara T, Xian Y, Liang L, Matsouaka RA, Saver JL, Smith EE, 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;319:463–473. doi: 10.1001/jama.2017.21917
10.
Boulouis G, Morotti A, Pasi M, Goldstein JN, Gurol ME, Charidimou A. Outcome of intracerebral haemorrhage related to non-vitamin K antagonists oral anticoagulants versus vitamin K antagonists: a comprehensive systematic review and meta-analysis. J Neurol Neurosurg Psychiatry. 2018;89:263–270. doi: 10.1136/jnnp-2017-316631
11.
Davis SM, Broderick J, Hennerici M, Brun NC, Diringer MN, Mayer SA, et al; Recombinant Activated Factor VII Intracerebral Hemorrhage Trial Investigators. Hematoma growth is a determinant of mortality and poor outcome after intracerebral hemorrhage. Neurology. 2006;66:1175–1181. doi: 10.1212/01.wnl.0000208408.98482.99
12.
Kuramatsu JB, Gerner ST, Schellinger PD, Glahn J, Endres M, Sobesky J, et al. Anticoagulant reversal, blood pressure levels, and anticoagulant resumption in patients with anticoagulation-related intracerebral hemorrhage. JAMA. 2015;313:824–836. doi: 10.1001/jama.2015.0846
13.
Wilson D, Seiffge DJ, Traenka C, Basir G, Purrucker JC, Rizos T, et al; And the CROMIS-2 Collaborators. Outcome of intracerebral hemorrhage associated with different oral anticoagulants. Neurology. 2017;88:1693–1700. doi: 10.1212/WNL.0000000000003886
14.
Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke. 1996;27:1304–1305.
15.
Hemphill JC, Greenberg SM, Anderson CS, Becker K, Bendok BR, Cushman M, et al; American Heart Association Stroke Council; Council on Cardiovascular and Stroke Nursing; Council on Clinical Cardiology. Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2015;46:2032–2060. doi: 10.1161/STR.0000000000000069
16.
Broderick JP, Brott TG, Duldner JE, Tomsick T, Huster G. Volume of intracerebral hemorrhage. A powerful and easy-to-use predictor of 30-day mortality. Stroke. 1993;24:987–993.
17.
Lauer A, Pfeilschifter W, Schaffer CB, Lo EH, Foerch C. Intracerebral haemorrhage associated with antithrombotic treatment: translational insights from experimental studies. Lancet Neurol. 2013;12:394–405. doi: 10.1016/S1474-4422(13)70049-8
18.
Akoudad S, Darweesh SK, Leening MJ, Koudstaal PJ, Hofman A, van der Lugt A, et al. Use of coumarin anticoagulants and cerebral microbleeds in the general population. Stroke. 2014;45:3436–3439. doi: 10.1161/STROKEAHA.114.007112
19.
Brouwers HB, Greenberg SM. Hematoma expansion following acute intracerebral hemorrhage. Cerebrovasc Dis. 2013;35:195–201. doi: 10.1159/000346599
20.
Pollack CV, Reilly PA, van Ryn J, Eikelboom JW, Glund S, Bernstein RA, et al. Idarucizumab for dabigatran reversal - full cohort analysis. N Engl J Med. 2017;377:431–441. doi: 10.1056/NEJMoa1707278
21.
Dowlatshahi D, Demchuk AM, Flaherty ML, Ali M, Lyden PL, Smith EE; VISTA Collaboration. Defining hematoma expansion in intracerebral hemorrhage: relationship with patient outcomes. Neurology. 2011;76:1238–1244. doi: 10.1212/WNL.0b013e3182143317
22.
Krishnan K, Mukhtar SF, Lingard J, Houlton A, Walker E, Jones T, et al. Performance characteristics of methods for quantifying spontaneous intracerebral haemorrhage: data from the Efficacy of Nitric Oxide in Stroke (ENOS) trial. J Neurol Neurosurg Psychiatry. 2015;86:1258–1266. doi: 10.1136/jnnp-2014-309845
Information & Authors
Information
Published In
Copyright
© 2018 American Heart Association, Inc.
Versions
You are viewing the most recent version of this article.
History
Received: 26 April 2018
Revision received: 6 July 2018
Accepted: 27 July 2018
Published online: 30 August 2018
Published in print: October 2018
Keywords
Subjects
Authors
Disclosures
Dr Tsivgoulis reports receiving speaker fees, consulting fees, and travel grants from Boehringer Ingelheim, Daichii Sankyo, and Bayer and has received no funding related to this project. Dr Varelas has been in the Advisory Board of Portola and has received honoraria. Dr Paciaroni received honoraria as a member of the speaker bureau of Aspen, Sanofi-Aventis, Boehringer Ingelheim, Bayer, Bristol Meyer Squibb, Daiichi Sankyo, and Pfizer. The other authors report no conflicts.
Sources of Funding
Dr Lioutas work has been partially supported by the Greek Diaspora Fellowship Grant from Stavros Niarchos Foundation. Dr Selim reports receiving grant support from the National Institute of Neurological Disorders and Stroke (Grant No U01NS 074425) and American Heart Association (Grant No 15CSA24540001).
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
- Trends in prior antithrombotic medication and risk of in-hospital mortality after spontaneous intracerebral hemorrhage: the J-ICH registry, Scientific Reports, 14, 1, (2024).https://doi.org/10.1038/s41598-024-62717-5
- Intracerebral haemorrhage in patients taking different types of oral anticoagulants: a pooled individual patient data analysis from two national stroke registries, Stroke and Vascular Neurology, 9, 6, (640-651), (2024).https://doi.org/10.1136/svn-2023-002813
- A Combination of Ex Vivo and In Vivo Strategies for Evaluating How Much New Oral Anticoagulants Exacerbate Experimental Intracerebral Bleeding, TH Open, 07, 03, (e195-e205), (2023).https://doi.org/10.1055/s-0043-1770782
- Is a high chest CT severity score a risk factor for an increased incidence of long-term neuroimaging findings after COVID-19?, Journal of Stroke and Cerebrovascular Diseases, 32, 2, (106920), (2023).https://doi.org/10.1016/j.jstrokecerebrovasdis.2022.106920
- Neuroimaging in Common Neurological Diseases Treated by Anticoagulants, Neurophysiology - Networks, Plasticity, Pathophysiology and Behavior, (2022).https://doi.org/10.5772/intechopen.105128
- Inhibitors of Early Mobilization in the Acute Phase of Intracerebral Hemorrhage: A Retrospective Observational Study, Journal of Stroke and Cerebrovascular Diseases, 31, 8, (106592), (2022).https://doi.org/10.1016/j.jstrokecerebrovasdis.2022.106592
- Management of stroke in patients on antithrombotic therapy: Practical issues in the era of direct oral anticoagulants, Revue Neurologique, 178, 3, (185-195), (2022).https://doi.org/10.1016/j.neurol.2021.07.021
- Decision-Making Process for the Management of Acute Stroke in Patients on Oral Anticoagulant: From Guidelines to Clinical Routine, Frontiers in Neurology, 12, (2022).https://doi.org/10.3389/fneur.2021.794001
- Predictors of hemorrhagic stroke in older persons taking nonsteroidal anti-inflammatory drugs: Results from the Food and Drug Administration Adverse Event Reporting System, Journal of the American Association of Nurse Practitioners, 33, 11, (959-966), (2021).https://doi.org/10.1097/JXX.0000000000000489
- Trends in incidence of oral anticoagulant-related intracerebral hemorrhage and sales of oral anticoagulants in Capital Region of Denmark 2010–2017, European Stroke Journal, 6, 2, (143-150), (2021).https://doi.org/10.1177/23969873211008770
- See more
Loading...
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
Purchase this article to access the full text.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.