Neuroimaging and clinical outcomes of oral anticoagulant-associated intracerebral hemorrhage

OBJECTIVE

Whether intracerebral hemorrhage (ICH) associated with non-vitamin K antagonist oral anticoagulants (NOAC-ICH) has a better outcome compared to ICH associated with vitamin K antagonists (VKA-ICH) is uncertain.

METHODS

We performed a systematic review and individual patient data meta-analysis of cohort studies comparing clinical and radiological outcomes between NOAC-ICH and VKA-ICH patients. The primary outcome measure was 30-day all-cause mortality. All outcomes were assessed in multivariate regression analyses adjusted for age, sex, ICH location, and intraventricular hemorrhage extension.

RESULTS

We included 7 eligible studies comprising 219 NOAC-ICH and 831 VKA-ICH patients (mean age = 77 years, 52.5% females). The 30-day mortality was similar between NOAC-ICH and VKA-ICH (24.3% vs 26.5%; hazard ratio = 0.94, 95% confidence interval [CI] = 0.67-1.31). However, in multivariate analyses adjusting for potential confounders, NOAC-ICH was associated with lower admission National Institutes of Health Stroke Scale (NIHSS) score (linear regression coefficient = -2.83, 95% CI = -5.28 to -0.38), lower likelihood of severe stroke (NIHSS > 10 points) on admission (odds ratio [OR] = 0.50, 95% CI = 0.30-0.84), and smaller baseline hematoma volume (linear regression coefficient = -0.24, 95% CI = -0.47 to -0.16). The two groups did not differ in the likelihood of baseline hematoma volume < 30cm³ (OR = 1.14, 95% CI = 0.81-1.62), hematoma expansion (OR = 0.97, 95% CI = 0.63-1.48), in-hospital mortality (OR = 0.73, 95% CI = 0.49-1.11), functional status at discharge (common OR = 0.78, 95% CI = 0.57-1.07), or functional status at 3 months (common OR = 1.03, 95% CI = 0.75-1.43).

INTERPRETATION

Although functional outcome at discharge, 1 month, or 3 months was comparable after NOAC-ICH and VKA-ICH, patients with NOAC-ICH had smaller baseline hematoma volumes and less severe acute stroke syndromes. Ann Neurol 2018;84:702-712.

Management of patients with stroke treated with direct oral anticoagulants

Since their market approval, direct oral anticoagulants (DOACs) are being increasingly used for stroke prevention in patients with atrial fibrillation. However, the management of DOAC-treated patients with stroke poses several challenges for physicians in everyday clinical practice, both in the acute setting and in long-term care. This has spurred extensive research activity in the field over the past few years, which we review here.

Fast-track versus long-term hospitalizations for patients with non-disabling acute ischaemic stroke

BACKGROUND AND PURPOSE

The aim was to assess the feasibility and safety of fast-track hospitalizations in a selected cohort of patients with stroke.

METHODS

Patients hospitalized at the Stroke Center of the University Hospital Basel, Switzerland, with an acute ischaemic stroke confirmed on magnetic resonance diffusion-weighted imaging were included. Neurological deficits of the included patients were non-disabling, i.e. not interfering with activities of daily living and compatible with a direct discharge home. Patients with premorbid disability were excluded. All patients were admitted to the Stroke Center for ≥24 h. Two study groups were compared - fast-track hospitalizations (≤72 h) and long-term hospitalizations (>72 h). The primary end-point was a composite of any unplanned rehospitalization for any reason within 3 months since hospital discharge and a modified Rankin Scale 3-6 at 3 months. Adjustment for confounders was done using the inverse probability of treatment weights (IPTW).

RESULTS

Amongst the 521 patients who met the inclusion criteria, fast-track hospitalizations were performed in 79 patients (15%). In the fast-track group, seven patients (8.9%) met the primary end-point, compared to 37 (8.4%) in the long-term group [odds ratio (OR) 1.06, 95% confidence interval (CI) 0.42-2.34, P = 0.88]. After weighting for IPTW, the odds of the primary end-point remained similar between the two arms (OR_IPTW 1.27, 95% CI 0.51-3.16, P = 0.61). The costs of fast-track hospitalizations were lower, on average, by $4994.

CONCLUSIONS

Fast-track hospitalizations including a full workup proved to be feasible, showed no increased risk and were less expensive than long-term hospitalizations.

Rivaroxaban plasma levels in acute ischemic stroke and intracerebral hemorrhage

OBJECTIVE

Information about rivaroxaban plasma level (RivLev) may guide treatment decisions in patients with acute ischemic stroke (AIS) and intracerebral hemorrhage (ICH) taking rivaroxaban.

METHODS

In a multicenter registry-based study (Novel Oral Anticoagulants in Stroke Patients collaboration; ClinicalTrials.gov: NCT02353585) of patients with stroke while taking rivaroxaban, we compared RivLev in patients with AIS and ICH. We determined how many AIS patients had RivLev ≤ 100ng/ml, indicating possible eligibility for thrombolysis, and how many ICH patients had RivLev ≥ 75ng/ml, making them possibly eligible for the use of specific reversal agents. We explored factors associated with RivLev (Spearman correlation, regression models) and studied the sensitivity and specificity of international normalized ratio (INR) thresholds to substitute RivLev using cross tables and receiver operating characteristic curves.

RESULTS

Among 241 patients (median age = 80 years, interquartile range [IQR] = 73-84; median time from onset to admission = 2 hours, IQR = 1-4.5 hours; median RivLev = 89ng/ml, IQR = 31-194), 190 had AIS and 51 had ICH. RivLev was similar in AIS patients (82ng/ml, IQR = 30-202) and ICH patients (102ng/ml, IQR = 51-165; p = 0.24). Trough RivLev_{(≤137ng/ml)} occurred in 126/190 (66.3%) AIS and 34/51 (66.7%) ICH patients. Among AIS patients, 108/190 (56.8%) had RivLev ≤ 100ng/ml. In ICH patients, 33/51 (64.7%) had RivLev ≥ 75ng/ml. RivLev was associated with rivaroxaban dosage, and inversely with renal function and time since last intake (each p < 0.05). INR ≤ 1.0 had a specificity of 98.9% and a sensitivity of 25.7% to predict RivLev ≤ 100ng/ml. INR ≥ 1.4 had a sensitivity of 59.3% and specificity of 90.1% to predict RivLev ≥ 75ng/ml.

INTERPRETATION

RivLev did not differ between patients with AIS and ICH. Half of the patients with AIS under rivaroxaban had a RivLev low enough to consider thrombolysis. In ICH patients, two-thirds had a RivLev high enough to meet the eligibility for the use of a specific reversal agent. INR thresholds perform poorly to inform treatment decisions in individual patients. Ann Neurol 2018;83:451-459.

A Score for Risk of Thrombolysis-Associated Hemorrhage Including Pretreatment with Statins

Background

Symptomatic intracranial hemorrhage (sICH) after intravenous thrombolysis with recombinant tissue-plasminogen activator (rt-PA) for acute ischemic stroke is associated with a poor functional outcome. We aimed to develop a score assessing risk of sICH including novel putative predictors-namely, pretreatment with statins and severe renal impairment.

Methods

We analyzed our local cohort (Berlin) of patients receiving rt-PA for acute ischemic stroke between 2006 and 2016. Outcome was sICH according to ECASS-III criteria. A multiple regression model identified variables associated with sICH and receiver operating characteristics were calculated for the best discriminatory model for sICH. The model was validated in an independent thrombolysis cohort (Basel).

Results

sICH occurred in 53 (4.0%) of 1,336 patients in the derivation cohort. Age, baseline National Institutes of Health Stroke Scale, systolic blood pressure on admission, blood glucose on admission, and prior medication with medium- or high-dose statins were associated with sICH and included into the risk of intracranial hemorrhage score. The validation cohort included 983 patients of whom 33 (3.4%) had a sICH. c-Statistics for sICH was 0.72 (95% CI 0.66-0.79) in the derivation cohort and 0.69 (95% CI 0.60-0.77) in the independent validation cohort. Inclusion of severe renal impairment did not improve the score.

Conclusion

We developed a simple score with fair discriminating capability to predict rt-PA-related sICH by adding prior statin use to known prognostic factors of sICH. This score may help clinicians to identify patients with higher risk of sICH requiring intensive monitoring.

Intravenous thrombolysis and platelet count

OBJECTIVE

To study the effect of platelet count (PC) on bleeding risk and outcome in stroke patients treated with IV thrombolysis (IVT) and to explore whether withholding IVT in PC < 100 × 10⁹/L is supported.

METHODS

In this prospective multicenter, IVT register-based study, we compared PC with symptomatic intracranial hemorrhage (sICH; Second European-Australasian Acute Stroke Study [ECASS II] criteria), poor outcome (modified Rankin Scale score 3-6), and mortality at 3 months. PC was used as a continuous and categorical variable distinguishing thrombocytopenia (<150 × 10⁹/L), thrombocytosis (>450 × 10⁹/L), and normal PC (150-450 × 10⁹/L [reference group]). Moreover, PC < 100 × 10⁹/L was compared to PC ≥ 100 × 10⁹/L. Unadjusted and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) from the logistic regression models were calculated.

RESULTS

Among 7,533 IVT-treated stroke patients, 6,830 (90.7%) had normal PC, 595 (7.9%) had thrombocytopenia, and 108 (1.4%) had thrombocytosis. Decreasing PC (every 10 × 10⁹/L) was associated with increasing risk of sICH (OR_adjusted 1.03, 95% CI 1.02-1.05) but decreasing risk of poor outcome (OR_adjusted 0.99, 95% CI 0.98-0.99) and mortality (OR_adjusted 0.98, 95% CI 0.98-0.99). The risk of sICH was higher in patients with thrombocytopenic than in patients with normal PC (OR_adjusted 1.73, 95% CI 1.24-2.43). However, the risk of poor outcome (OR_adjusted 0.89, 95% CI 0.39-1.97) and mortality (OR_adjusted 1.09, 95% CI 0.83-1.44) did not differ significantly. Thrombocytosis was associated with mortality (OR_adjusted 2.02, 95% CI 1.21-3.37). Forty-four (0.3%) patients had PC < 100 × 10⁹/L. Their risks of sICH (OR_unadjusted 1.56, 95% CI 0.48-5.07), poor outcome (OR_adjusted 1.63, 95% CI 0.82-3.24), and mortality (OR_adjusted 1.38, 95% CI 0.64-2.98) did not differ significantly from those of patients with PC ≥ 100 × 10⁹/L.

CONCLUSION

Lower PC was associated with increased risk of sICH, while higher PC indicated increased mortality. Our data suggest that PC modifies outcome and complications in individual patients, while withholding IVT in all patients with PC < 100 × 10⁹/L is challenged.

Serum neurofilament light chain in patients with acute cerebrovascular events

BACKGROUND AND PURPOSE

Serum neurofilaments are markers of axonal injury. We addressed their diagnostic and prognostic role in acute ischemic stroke (AIS) and transient ischemic attack (TIA).

METHODS

Nested within a prospective cohort study, we compared levels of serum neurofilament light chain (sNfL) drawn within 24 h from symptom onset in patients with AIS or TIA. Patients without magnetic resonance imaging on admission were excluded. We assessed whether sNfL was associated with: (i) clinical severity on admission, (ii) diagnosis of AIS vs. TIA, (iii) infarct size on admission magnetic resonance diffusion-weighted imaging (MR-DWI) and (iv) functional outcome at 3 months.

RESULTS

We analyzed 504 patients with AIS and 111 patients with TIA. On admission, higher National Institutes of Health Stroke Scale (NIHSS) scores were associated with higher sNfL: NIHSS score < 7, 13.1 pg/mL [interquartile range (IQR), 5.3-27.8]; NIHSS score 7-15, 16.7 pg/mL (IQR, 7.4-34.9); and NIHSS score > 15, 21.0 pg/mL (IQR, 9.3-40.4) (P = 0.01). Compared with AIS, patients with TIA had lower sNfL levels [9.0 pg/mL (95% confidence interval, 4.0-19.0) vs. 16.0 pg/mL (95% confidence interval, 7.3-34.4), P < 0.001], also after adjusting for age and NIHSS score (P = 0.006). Among patients with AIS, infarct size on admission MR-DWI was not associated with sNfL, either in univariate analysis (P = 0.15) or after adjusting for age and NIHSS score on admission (P = 0.56). Functional outcome 3 months after stroke was not associated with sNfL after adjusting for established predictors.

CONCLUSIONS

In conclusion, among patients admitted within 24 h of AIS or TIA onset, admission sNfL levels were associated with clinical severity on admission and TIA diagnosis, but not with infarct size on MR-DWI acquired on admission or functional outcome at 3 months.

Abstract TP221: Untangling Prehospital Delay in Acute Ischemic Stroke: Hints on Increasing the Thrombolysis Rate - a Prospective Cohort Study

Background: Delays in the prehospital phase jeopardize the chances for stroke patients to be treated with thrombolysis. However, the causes of prehospital delay remain obscure. We assessed the causes of prehospital delay among patients with acute ischemic stroke.

Methods: In this prospective cohort study, we included patients admitted to the Stroke Center of the University Hospital Basel, Switzerland, with an acute ischemic stroke. A corresponding positive Diffusion Weighted Imaging MR was required to be included in the study, i.e. patients with no or negative DWI-MR were excluded. Trained study nurses interviewed all patients or their proxies in person at the bedside. The study period spanned from September, 2015 to July, 2017.

Results: Overall, 321 patients were included. The median NIHSS score was 3 (interquartile range [IQR]: 1-5). Median delay between symptom onset and call for help was 55 minutes (IQR: 10-420). The mode of transportation to the Stroke Center was with paramedics (ambulance or helicopter) for 60 % of patients, without in 40% (private car: n=72; taxi: n=31; public transportation, walk-in or other: n=25). NIHSS was higher in patients transported with paramedics (4 [IQR 2-6] vs. 1.5 [0-3], P <0.001). A family physician was consulted immediately prior to hospitalization by 11% of patients (n=35), 89% were admitted directly to the hospital, and the difference in NIHSS was not statistically significant.

The thrombolysis ratio was 25%. After adjusting for NIHSS, the odds of thrombolysis were higher among patients transported with paramedics (adjusted odds ratio [aOR]: 2.4, 95%-CI: 1.2-4.7, P=0.01), and lower among patients with prior family physician consultation (aOR: 0.10, 95%-CI: 0.01-0.77, P =0.03).

Conclusions: One out four patients wait 7 hours or more before calling for help, a considerable delay. Transport with paramedics was associated with increased chances of thrombolysis. One in ten patients consulted a family physician immediately prior to hospitalization, reducing by 90 percent points the chances of thrombolysis. Identifying and informing family physicians who see patients with symptoms of an acute stroke may be a so far underestimated, cost-effective measure to increase the rate of thrombolysis.

Intravenous Thrombolysis in Patients with Stroke Taking Rivaroxaban Using Drug Specific Plasma Levels: Experience with a Standard Operation Procedure in Clinical Practice

Background and Purpose

Standard operating procedures (SOP) incorporating plasma levels of rivaroxaban might be helpful in selecting patients with acute ischemic stroke taking rivaroxaban suitable for IVthrombolysis (IVT) or endovascular treatment (EVT).

Methods

This was a single-center explorative analysis using data from the Novel-Oral-Anticoagulants-in-Stroke-Patients-registry (clinicaltrials.gov:NCT02353585) including acute stroke patients taking rivaroxaban (September 2012 to November 2016). The SOP included recommendation, consideration, and avoidance of IVT if rivaroxaban plasma levels were <20 ng/mL, 20‒100 ng/mL, and >100 ng/mL, respectively, measured with a calibrated anti-factor Xa assay. Patients with intracranial artery occlusion were recommended IVT+EVT or EVT alone if plasma levels were ≤100 ng/mL or >100 ng/mL, respectively. We evaluated the frequency of IVT/EVT, door-to-needle-time (DNT), and symptomatic intracranial or major extracranial hemorrhage.

Results

Among 114 acute stroke patients taking rivaroxaban, 68 were otherwise eligible for IVT/EVT of whom 63 had plasma levels measured (median age 81 years, median baseline National Institutes of Health Stroke Scale 6). Median rivaroxaban plasma level was 96 ng/mL (inter quartile range [IQR] 18‒259 ng/mL) and time since last intake 11 hours (IQR 4.5‒18.5 hours). Twenty-two patients (35%) received IVT/EVT (IVT n=15, IVT+EVT n=3, EVT n=4) based on SOP. Median DNT was 37 (IQR 30‒60) minutes. None of the 31 patients with plasma levels >100 ng/mL received IVT. Among 14 patients with plasma levels ≤100 ng/mL, the main reason to withhold IVT was minor stroke (n=10). No symptomatic intracranial or major extracranial bleeding occurred after treatment.

Conclusions

Determination of rivaroxaban plasma levels enabled IVT or EVT in one-third of patients taking rivaroxaban who would otherwise be ineligible for acute treatment. The absence of major bleeding in our pilot series justifies future studies of this approach.

Outcome of intracerebral hemorrhage associated with different oral anticoagulants

Objective:

In an international collaborative multicenter pooled analysis, we compared mortality, functional outcome, intracerebral hemorrhage (ICH) volume, and hematoma expansion (HE) between non–vitamin K antagonist oral anticoagulation–related ICH (NOAC-ICH) and vitamin K antagonist–associated ICH (VKA-ICH).

Methods:

We compared all-cause mortality within 90 days for NOAC-ICH and VKA-ICH using a Cox proportional hazards model adjusted for age; sex; baseline Glasgow Coma Scale score, ICH location, and log volume; intraventricular hemorrhage volume; and intracranial surgery. We addressed heterogeneity using a shared frailty term. Good functional outcome was defined as discharge modified Rankin Scale score ≤2 and investigated in multivariable logistic regression. ICH volume was measured by ABC/2 or a semiautomated planimetric method. HE was defined as an ICH volume increase >33% or >6 mL from baseline within 72 hours.

Results:

We included 500 patients (97 NOAC-ICH and 403 VKA-ICH). Median baseline ICH volume was 14.4 mL (interquartile range [IQR] 3.6–38.4) for NOAC-ICH vs 10.6 mL (IQR 4.0–27.9) for VKA-ICH (p = 0.78). We did not find any difference between NOAC-ICH and VKA-ICH for all-cause mortality within 90 days (33% for NOAC-ICH vs 31% for VKA-ICH [p = 0.64]; adjusted Cox hazard ratio (for NOAC-ICH vs VKA-ICH) 0.93 [95% confidence interval (CI) 0.52–1.64] [p = 0.79]), the rate of HE (NOAC-ICH n = 29/48 [40%] vs VKA-ICH n = 93/140 [34%] [p = 0.45]), or functional outcome at hospital discharge (NOAC-ICH vs VKA-ICH odds ratio 0.47; 95% CI 0.18–1.19 [p = 0.11]).

Conclusions:

In our international collaborative multicenter pooled analysis, baseline ICH volume, hematoma expansion, 90-day mortality, and functional outcome were similar following NOAC-ICH and VKA-ICH.

Intravenous Thrombolysis in Patients Dependent on the Daily Help of Others Before Stroke

Background and Purpose—

We compared outcome and complications in patients with stroke treated with intravenous thrombolysis (IVT) who could not live alone without help of another person before stroke (dependent patients) versus independent ones.

Methods—

In a multicenter IVT-register–based cohort study, we compared previously dependent (prestroke modified Rankin Scale score, 3–5) versus independent (prestroke modified Rankin Scale score, 0–2) patients. Outcome measures were poor 3-month outcome (not reaching at least prestroke modified Rankin Scale [dependent patients]; modified Rankin Scale score of 3–6 [independent patients]), death, and symptomatic intracranial hemorrhage. Unadjusted and adjusted odds ratios (ORs) with 95% confidence intervals (OR [95% confidence interval]) were calculated.

Results—

Among 7430 IVT-treated patients, 489 (6.6%) were dependent and 6941 (93.4%) were independent. Previous stroke, dementia, heart, and bone diseases were the most common causes of preexisting dependency. Dependent patients were more likely to die (OR unadjusted , 4.55 [3.74–5.53]; OR adjusted , 2.19 [1.70–2.84]). Symptomatic intracranial hemorrhage occurred equally frequent (4.8% versus 4.5%). Poor outcome was more frequent in dependent (60.5%) than in independent (39.6%) patients, but the adjusted ORs were similar (OR adjusted , 0.95 [0.75–1.21]). Among survivors, the proportion of patients with poor outcome did not differ (35.7% versus 31.3%). After adjustment for age and stroke severity, the odds of poor outcome were lower in dependent patients (OR adjusted , 0.64 [0.49–0.84]).

Conclusions—

IVT-treated stroke patients who were dependent on the daily help of others before stroke carry a higher mortality risk than previously independent patients. The risk of symptomatic intracranial hemorrhage and the likelihood of poor outcome were not independently influenced by previous dependency. Among survivors, poor outcome was avoided at least as effectively in previously dependent patients. Thus, withholding IVT in previously dependent patients might not be justified.

Serum Neurofilament Light Chain Levels Are Associated with Clinical Characteristics and Outcome in Patients with Cervical Artery Dissection

BACKGROUND

Serum neurofilament light chain (sNfL) levels represent a promising marker of neuroaxonal injury. They are elevated in several neurological conditions, but their importance in cerebrovascular diseases remains unclear. In a proof of concept study, we compared sNfL levels with clinical characteristics and outcome in patients with cervical artery dissection (CeAD).

METHODS

A total of 49 non-traumatic CeAD patients were included. sNfL levels were measured by high-sensitivity electrochemiluminescence immunoassay. Levels were compared with regard to (i) type of presenting symptoms (local symptoms only (n = 8), transient ischemic attack (TIA; n = 10) or ischemic stroke (n = 31)), (ii) stroke severity quantified by National Institute of Health Stroke Scale (NIHSS), (iii) time interval between onset of symptoms and blood sampling and (iv) 3-month outcome as measured by the modified Rankin Scale score. Analyses were performed using univariate and multivariate linear and ordinal regression models.

RESULTS

CeAD patients presenting with stroke had significantly higher sNfL levels (median 108.9 pg/ml, interquartile range (37.8-427.7)) than patients with TIA (16.4 pg/ml (8.7-36.3), p = 0.002) or local symptoms (23.4 pg/ml (17.8-30.8), p = 0.0007). Among stroke patients, sNfL levels were positively associated with both NIHSS (p = 0.0002) and time between stroke onset and serum sampling (p = 1.9 × 10-6). Higher sNfL levels were associated with unfavorable outcome at 3 months (OR 4.67, 95% CI 1.69-12.95, p = 0.003). However, this association lost significance after adjustment for NIHSS. The highest sNfL level was observed in a TIA patient who had ischemic stroke 1 day after serum sampling for sNfL measurement.

CONCLUSION

sNfL levels were increased in CeAD patients presenting with stroke, correlated with clinical severity and were influenced by the time point of blood sampling. The prognostic meaning of sNfL in CeAD deserves further testing.

Recanalization Therapies in Acute Ischemic Stroke Patients

Background—

We explored the safety of intravenous thrombolysis (IVT) or intra-arterial treatment (IAT) in patients with ischemic stroke on non-vitamin K antagonist oral anticoagulants (NOACs, last intake <48 hours) in comparison with patients (1) taking vitamin K antagonists (VKAs) or (2) without previous anticoagulation (no-OAC).

Methods and Results—

This is a multicenter cohort pilot study. Primary outcome measures were (1) occurrence of intracranial hemorrhage (ICH) in 3 categories: any ICH (ICH any ), symptomatic ICH according to the criteria of the European Cooperative Acute Stroke Study II (ECASS-II) (sICH ECASS-II ) and the National Institute of Neurological Disorders and Stroke (NINDS) thrombolysis trial (sICH NINDS ); and (2) death (at 3 months). Cohorts were compared by using propensity score matching. Our NOAC cohort comprised 78 patients treated with IVT/IAT and the comparison groups of 441 VKA patients and 8938 no-OAC patients. The median time from last NOAC intake to IVT/IAT was 13 hours (interquartile range, 8–22 hours). In VKA patients, median pre-IVT/IAT international normalized ratio was 1.3 (interquartile range, 1.1–1.6). ICH any was observed in 18.4% NOAC patients versus 26.8% in VKA patients and 17.4% in no-OAC patients. sICH ECASS-II and sICH NINDS occurred in 2.6%/3.9% NOAC patients, in comparison with 6.5%/9.3% of VKA patients and 5.0%/7.2% of no-OAC patients, respectively. At 3 months, 23.0% of NOAC patients in comparison with 26.9% of VKA patients and 13.9% of no-OAC patients had died. Propensity score matching revealed no statistically significant differences.

Conclusions—

IVT/IAT in selected patients with ischemic stroke under NOAC treatment has a safety profile similar to both IVT/IAT in patients on subtherapeutic VKA treatment or in those without previous anticoagulation. However, further prospective studies are needed, including the impact of specific coagulation tests.

Abstract W P16: Comparing Outcome Prediction Scoring Systems in the Interventional Management of Stroke (IMS) III Trial: No Clear Winner

Introduction: Several prognostic scores have been developed to predict patient outcome after ischemic stroke. We aimed to compare eight scores in the IMS -III dataset.

Hypothesis: Prediction of good and poor outcome using different scores is possible in IMS-III.

Methods: We retrospectively calculated area under the curve (AUC), sensitivity, and specificity for poor (mRS 4-6) and good (mRS 0-2) 90-day outcome in all IMS-III participants with complete data. The scoring systems analyzed were: Houston Intra-arterial Therapy 2 (HIAT2), Ischemic Stroke Predictive Risk Score (iScore), Totaled Health Risks in Vascular Events (THRIVE), DRAGON, Hemorrhage after Thrombolysis (HAT), Alberta Stroke Programme Early CT Score (ASPECTS), Stroke-Thrombolytic Predictive Instrument (TPI), and Simple Variables Model (SVM). Those receiving IV rtPA followed by endovascular therapy (treatment) and those receiving IV rtPA only (control) were analyzed separately.

Results: Among 656 IMS III subjects, 651 had complete data, including 429 treatment and 222 control subjects. Overall, all scoring systems showed good prediction abilities (AUC range 0.60 to 0.78). AUC comparisons did not clearly favor a single scoring system.

Treatment group: For predicting poor outcome, the TPI score outperformed ASPECTS (AUC 0.73 vs 0.61; P <0.05). For predicting good outcome, the HAT score was significantly better than ASPECTS (0.73 vs. 0.64; P <0.05).

Control group: For poor outcome, the TPI score was significantly better than ASPECTS and THRIVE (0.78 vs. 0.62 and 0.71, respectively; P <0.05). For good outcome, TPI performed better than ASPECTS and THRIVE (0.73 vs. 0.61 and 0.67, respectively; P <0.05).

Conclusions: Outcome prediction using different scores in IMS III is possible. HIAT2, iSCORE, DRAGON, HAT, TPI, and SVM all performed comparably. Prognostic performances were comparable in both treatment arms.

Abstract 62: GRASPS Score to Predict Risk of Clinically Relevant Symptomatic Intracranial Hemorrhage after Intravenous Tissue Plasminogen Activator

Background: The NINDS Study definition of symptomatic hemorrhage (sICH) after intravenous tissue plasminogen activator (tPA), requiring only minimal early worsening and minimal petechial hemorrhage, is now widely recognized as overly inclusive. Clinically relevant sICH is better defined by at least moderate worsening and parenchymal hematoma. The 6-item GRASPS scale for predicting sICH was derived from the national US Get with the Guidelines - Stroke registry based on the NINDS Study definition. For use in clinical practice, it is desirable to recalibrate the GRASPS scale to predict more stringently defined symptomatic hemorrhage.

Methods: We merged prospectively collected data of patients with consecutive ischemic stroke who received tPA in 7 stroke centers. We applied the GRASPS (glucose at presentation, race [Asian], age, sex [male], systolic blood pressure at presentation, and severity of stroke at presentation [NIH Stroke Scale]) to predict Safe Implementation of Thrombolysis in Stroke (SITS)-defined sICH. To derive a monotonic predictive model, we used a generalized additive model framework and fit 6 different transformations of the GRASPS score: linear scale, log scale, without and with splines, and without and with local smoothing (“loess”) to identify any non-parametric patterns.

Results: The final cohort comprised 5274 eligible patients, of whom 143 (2.7%) had symptomatic ICH per SITS criteria. Based on favorable residual deviance scores and Akaike's information criterion scores, the linear transformation of raw GRASPS scores provided the best fit. With this model, the area under the curve for predicting SITS sICH was 0.68 (95% CI 0.63-0.72). Risk score values for cardinal GRASPS scale points included 0.6% for GRASPS 50, 2.4% for GRASPS 70, and 9.5% for GRASPS 90.

Conclusions: The GRASPS scale showed moderate performance in predicting SITS-defined symptomatic intracerebral hemorrhage after IV tPA. With the predictive values available from this study, the GRASPS score can be used to assess risk of clinically relevant symptomatic hemorrhage following thrombolytic therapy.

Dose-related effects of statins on symptomatic intracerebral hemorrhage and outcome after thrombolysis for ischemic stroke

BACKGROUND AND PURPOSE

The aim of our study was to assess whether statins have dose-dependent effects on risk of symptomatic intracerebral hemorrhage (sICH) and outcome after intravenous thrombolysis for ischemic stroke.

METHODS

We pooled data from 2 European intravenous thrombolysis registries. Statin doses were stratified in 3 groups according to the attainable lowering of cholesterol levels (low dose: simvastatin 20 mg or equivalent; medium dose: simvastatin 40 mg or equivalent; and high dose: simvastatin 80 mg or equivalent). sICH was defined according to the European Cooperative Acute Stroke Study. Modified Rankin Scale score 0 to 2 at 3 months was considered a favorable outcome.

RESULTS

Among 1446 patients analyzed (median age, 75 years; median initial National Institutes of Health Stroke Scale score, 11; 54% men), 317 (22%) used statins before intravenous thrombolysis. Of them, 120 patients had low-dose, 134 medium-dose, and 63 high-dose statin therapy. sICH occurred in 4% of patients (n=53). Frequency of sICH was 2%, 6%, and 13% in patients with low-, medium-, and high-dose statin treatment, respectively (P<0.01). Adjusted odds ratio (OR) for sICH was 2.4 (95% confidence interval [CI], 1.1-5.3) and 5.3 (95% CI, 2.3-12.3) for patients with medium- and high-dose statins compared with non-statin users. Statin users more often achieved favorable outcome compared with non-statin users (58% versus 51%; P=0.03). An independent association of statin use with favorable outcome was detected (adjusted OR, 1.8; 95% CI, 1.3-2.5). The association was maintained when stratifying for statin dose, although it was not significant in the high-dose group anymore (OR, 1.7; 95% CI, 0.9-3.2).

CONCLUSIONS

We observed an association between increasing dose of statin use and risk of sICH after intravenous thrombolysis. Nevertheless, there was an overall beneficial effect of previous statin use on favorable 3-month outcome.

IV thrombolysis and renal function

OBJECTIVE

To investigate the association of renal impairment on functional outcome and complications in stroke patients treated with IV thrombolysis (IVT).

METHODS

In this observational study, we compared the estimated glomerular filtration rate (GFR) with poor 3-month outcome (modified Rankin Scale scores 3-6), death, and symptomatic intracranial hemorrhage (sICH) based on the criteria of the European Cooperative Acute Stroke Study II trial. Unadjusted and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Patients without IVT treatment served as a comparison group.

RESULTS

Among 4,780 IVT-treated patients, 1,217 (25.5%) had a low GFR (<60 mL/min/1.73 m(2)). A GFR decrease by 10 mL/min/1.73 m(2) increased the risk of poor outcome (OR [95% CI]): (ORunadjusted 1.20 [1.17-1.24]; ORadjusted 1.05 [1.01-1.09]), death (ORunadjusted 1.33 [1.28-1.38]; ORadjusted 1.18 [1.11-1.249]), and sICH (ORunadjusted 1.15 [1.01-1.22]; ORadjusted 1.11 [1.04-1.20]). Low GFR was independently associated with poor 3-month outcome (ORadjusted 1.32 [1.10-1.58]), death (ORadjusted 1.73 [1.39-2.14]), and sICH (ORadjusted 1.64 [1.21-2.23]) compared with normal GFR (60-120 mL/min/1.73 m(2)). Low GFR (ORadjusted 1.64 [1.21-2.23]) and stroke severity (ORadjusted 1.05 [1.03-1.07]) independently determined sICH. Compared with patients who did not receive IVT, treatment with IVT in patients with low GFR was associated with poor outcome (ORadjusted 1.79 [1.41-2.25]), and with favorable outcome in those with normal GFR (ORadjusted 0.77 [0.63-0.94]).

CONCLUSION

Renal function significantly modified outcome and complication rates in IVT-treated stroke patients. Lower GFR might be a better risk indicator for sICH than age. A decrease of GFR by 10 mL/min/1.73 m(2) seems to have a similar impact on the risk of death or sICH as a 1-point-higher NIH Stroke Scale score measuring stroke severity.

Validation of the DRAGON Score in 12 Stroke Centers in Anterior and Posterior Circulation

Background and Purpose—

The DRAGON score predicts functional outcome in the hyperacute phase of intravenous thrombolysis treatment of ischemic stroke patients. We aimed to validate the score in a large multicenter cohort in anterior and posterior circulation.

Methods—

Prospectively collected data of consecutive ischemic stroke patients who received intravenous thrombolysis in 12 stroke centers were merged (n=5471). We excluded patients lacking data necessary to calculate the score and patients with missing 3-month modified Rankin scale scores. The final cohort comprised 4519 eligible patients. We assessed the performance of the DRAGON score with area under the receiver operating characteristic curve in the whole cohort for both good (modified Rankin scale score, 0–2) and miserable (modified Rankin scale score, 5–6) outcomes.

Results—

Area under the receiver operating characteristic curve was 0.84 (0.82–0.85) for miserable outcome and 0.82 (0.80–0.83) for good outcome. Proportions of patients with good outcome were 96%, 93%, 78%, and 0% for 0 to 1, 2, 3, and 8 to 10 score points, respectively. Proportions of patients with miserable outcome were 0%, 2%, 4%, 89%, and 97% for 0 to 1, 2, 3, 8, and 9 to 10 points, respectively. When tested separately for anterior and posterior circulation, there was no difference in performance ( P =0.55); areas under the receiver operating characteristic curve were 0.84 (0.83–0.86) and 0.82 (0.78–0.87), respectively. No sex-related difference in performance was observed ( P =0.25).

Conclusions—

The DRAGON score showed very good performance in the large merged cohort in both anterior and posterior circulation strokes. The DRAGON score provides rapid estimation of patient prognosis and supports clinical decision-making in the hyperacute phase of stroke care (eg, when invasive add-on strategies are considered).