Hyperattenuating lesions after mechanical thrombectomy in acute ischaemic stroke: factors predicting symptomatic haemorrhage and clinical outcomes.

AIM: To evaluate the clinical significance of hyperattenuating lesions on CT after mechanical thrombectomy for acute ischaemic stroke, and to identify imaging factors that predict symptomatic haemorrhage and unfavourable outcomes. MATERIALS AND METHODS: Seventy-eight patients with acute ischaemic stroke in the anterior circulation who underwent mechanical thrombectomy were evaluated. All patients underwent post-interventional unenhanced computed tomography (CT) within 24 h and follow-up CT or magnetic resonance imaging (MRI) within 7 days. Baseline characteristics and clinical outcomes were compared between patients with and without hyperattenuating lesions. In patients with hyperattenuating lesions, clinical and imaging factors that predict symptomatic haemorrhage and unfavourable outcomes were determined. RESULTS: Fifty-six of 78 patients (71.8%) demonstrated hyperattenuating lesions on post-interventional CT. Patients with hyperattenuating lesions showed lower Alberta Stroke Program Early CT score (ASPECTS), persistent/symptomatic haemorrhage, and unfavourable outcomes than those without. In patients with hyperattenuating lesions, larger hyperattenuating lesion volume (>21.3 ml; OR, 55.60, p<0.001) and perilesional oedema (OR, 46.04, p=0.015) were independent factors predicting symptomatic haemorrhage. Older age (OR, 1.2, p=0.006) and lower ASPECTS (OR, 0.45, p=0.046) were independent factors predicting unfavourable outcomes in patients with hyperattenuating lesions. Adding the volume of the hyperattenuating lesion to age and ASPECTS increased the predictive performance of unfavourable outcomes (area under the curve 0.874 versus 0.934, p=0.043). CONCLUSIONS: Hyperattenuating lesions on post-interventional CT are associated with increased risk of symptomatic haemorrhage and unfavourable outcomes. Larger hyperattenuating lesion volume is an independent factor of symptomatic haemorrhage and it has added predictive value for unfavourable outcomes.

Association between body mass index and stroke severity in acute ischaemic stroke with non-valvular atrial fibrillation.

BACKGROUND AND PURPOSE: The objective of this study was to investigate the association between body mass index (BMI) and both initial stroke severity at presentation and functional outcomes after acute ischaemic stroke (AIS) in patients with non-valvular atrial fibrillation (NVAF). METHODS: Patients were categorized on the basis of their BMI into underweight (BMI <18.5, n = 111), normal (18.5 ≤ BMI <25, n = 1036) and overweight to obese (BMI ≥25, n = 472) groups. Initial stroke severity was assessed using the National Institutes of Health Stroke Scale (NIHSS) score and functional outcomes were assessed using the modified Rankin Scale score at discharge. The differences in stroke severity and functional outcomes were compared between groups using robust log-linear regression with a Poisson distribution and binary logistic regression analysis. RESULTS: A total of 1619 AIS patients with NVAF from six hospitals were included. Compared with the NIHSS scores [median 5, interquartile range (IQR) 2-14] of normal-weight patients, the NIHSS scores (median 9, IQR 4-19) of underweight patients were more likely to be higher, whereas those of overweight to obese patients were lower (median 4, IQR 1-12) (P < 0.001). In terms of functional outcomes after stroke, underweight patients had a higher risk of poor functional outcomes (odds ratio 1.78, 95% confidence interval 1.09-2.56, P = 0.01) but overweight to obese patients had no significant difference in functional outcomes compared with normal-weight patients. CONCLUSION: An inverse association was found between BMI and stroke severity in AIS patients with NVAF. This suggests the presence of an obesity paradox for short-term outcomes in patients with NVAF.

Prognostic significance of volume-based 18F-FDG PET/CT parameter in patients with surgically resected non-small cell lung cancer. Comparison with immunohistochemical biomarkers.

AIM: We investigated the prognostic value of volume-based 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) parameters compared with other factors including several immunohistochemical biomarkers in patients with surgically resected non-small cell lung cancer (NSCLC). STUDY PARTICIPANTS: 290 patients with surgically resected and histopathologically confirmed NSCLC. The maxmum standardized uptake value (SUVmax) and metabolic tumour volume (MTV) of the primary tumour were obtained on 18F-FDG PET/ computed tomography (CT) for initial staging and Ki-67 labeling index (LI), p16, CD31 and cyclin E were evaluated in the primary tumours by immunohistochemical staining. Survival analyses for variables including PET parameters, immunohistochemical biomarker and other clinical factors were performed using the Kaplan-Meier method and Cox proportional hazards regression analysis. RESULTS: In univariate analyses, tumour stage, tumour size, and MTV were significant prognostic factors for decreased overall survival (OS) and disease-free survival (DFS). Multivariate analyses showed MTV and tumour stage were significant predictors of poor OS (MTV, hazard ratio (HR) = 1.135, p = 0.015; stage, HR = 0.644, p = 0.025) and DFS (MTV, HR = 1.128, p = 0.043; stage, HR = 0.541, p = 0.009). CONCLUSION: The MTV of primary tumours is a significant prognostic factor for survival along with tumour stage in patients with surgically resected NSCLC. The MTV can predict OS and DFS better than immunohistochemical biomarkers.

Differential impact of unrecognised brain infarction on stroke outcome in non-valvular atrial fibrillation.

There has been little information regarding the impact of unrecognised brain infarctions (UBIs) on stroke outcome in patients with non-valvular atrial fibrillation (NVAF). By using volumetric analysis of ischaemic lesions, we evaluated the potential impact of UBIs on clinical outcome according to their presence and categorised type. This study enrolled 631 patients with NVAF having no clinical stroke history. UBIs were categorised into three types as territorial, lacunar, or subcortical. We collected stroke severity, functional outcome at three months, and the total volume of UBIs and acute infarction lesions. We investigated the association between clinical outcome and the type or volume of UBI, using a linear mixed model and logistic regression analysis. UBIs were detected in 285 (45.2 %) patients; territorial UBIs were observed in 24.4 % of patients (154/631), lacunar UBIs in 25 % (158/631), and subcortical UBIs in 15.7 % (99/631). Although initial stroke severity was not different between patients with UBIs and those without, those with UBIs had less improvement during hospitalisation, leading to poorer outcome at three months. Among the three types of UBIs, only territorial UBIs were associated with poor outcome, especially in patients with relatively smaller acute infarction volume. UBIs, in particular, territorial UBIs, may be considered as predictors for poor outcome after ischaemic stroke in patients with NVAF. Our results suggest that the impact of UBIs on clinical outcome differs according to the type of UBIs and the acute stroke severity.

Role of tumour necrosis factor receptor-1 and nuclear factor-κB in production of TNF-α-induced pro-inflammatory microparticles in endothelial cells.

Tumour necrosis factor-α (TNF-α) is upregulated in many inflammatory diseases and is also a potent agent for microparticle (MP) generation. Here, we describe an essential role of TNF-α in the production of endothelial cell-derived microparticles (EMPs) in vivo and the function of TNF-α-induced EMPs in endothelial cells. We found that TNF-α rapidly increased blood levels of EMPs in mice. Treatment of human umbilical vein endothelial cells (HUVECs) with TNF-α also induced EMP formation in a time-dependent manner. Silencing of TNF receptor (TNFR)-1 or inhibition of the nuclear factor-κB (NF-κB) in HUVECs impaired the production of TNF-α-induced EMP. Incubation of HUVECs with PKH-67-stained EMPs showed that endothelial cells readily engulfed EMPs, and the engulfed TNF-α-induced EMPs promoted the expression of pro-apoptotic molecules and upregulated intercellular adhesion molecule-1 level on the cell surface, which led to monocyte adhesion. Collectively, our findings indicate that the generation of TNF-α-induced EMPs was mediated by TNFR1 or NF-κB and that EMPs can contribute to apoptosis and inflammation of endothelial cells.

Comparison of stent retriever and intra-arterial fibrinolysis in patients with acute ischaemic stroke.

BACKGROUND AND PURPOSE: Although the stent retriever (SR) has shown a better reperfusion rate and clinical outcome than the older generation mechanical clot retrieval device, it is uncertain whether the SR is superior to intra-arterial fibrinolysis (IAF). METHODS: Ischaemic stroke patients who were treated with SR or IAF as initial endovascular treatment modality for unilateral arterial occlusion in the anterior circulation were included. Successful reperfusion was defined as Thrombolysis in Cerebral Infarction grade ≥2b. A favourable clinical outcome was defined as a modified Rankin Score ≤2 at 3 months. RESULTS: Between January 2009 and May 2012, 55 patients were treated with SR and 50 patients were treated with IAF. The baseline characteristics were similar between the two groups except for the occlusion site and rescue treatment. In binary logistic regression analysis adjusted for the occlusion site and rescue treatment, SR was independently associated with increased successful reperfusion [82.0% vs. 47.3%; odds ratio (OR) 5.21; 95% confidence interval (CI) 1.92-14.14) and a more favourable clinical outcome at 3 months (54.0% vs. 43.6%; OR 3.40; 95% CI 1.31-8.84). The frequency of symptomatic intracranial haemorrhage and mortality at 3 months was not different between the two groups. CONCLUSIONS: Stent retriever was as safe as and more effective than IAF. Our findings suggest that SR may be considered as an initial modality rather than IAF in acute stroke patients who undergo endovascular treatment.

The distribution of cerebral microbleeds determines their association with arterial stiffness in non-cardioembolic acute stroke patients.

BACKGROUND AND PURPOSE: Increased arterial stiffness causes vessel damage in the end-organs. Therefore small vessels in the brain may be susceptible to increased arterial stiffness. Cerebral microbleeds (CMBs) are topographically or pathophysiologically categorized as deep or infratentorial type and strictly lobar type. Whether the presence and location of CMBs are associated with brachial-ankle pulse wave velocity (baPWV) which represents a measure of arterial stiffness was investigated. METHODS: Between June 2006 and January 2012, 1137 consecutive patients diagnosed with non-cardioembolic acute ischaemic stroke and who underwent baPWV measurement and brain gradient-echo imaging were enrolled. CMBs were classified as deep or infratentorial or strictly lobar according to their location. Severity of leukoaraiosis was determined using the Fazekas scoring system. RESULTS: CMBs were found in 30.7% of the included patients. These patients were older than those without CMBs. Mean baPWV was higher in patients with CMBs than in those without (20 ± 5 m/s vs. 19 ± 5 m/s; P = 0.001). When comparing baPWV according to the location of the CMB, it was higher in the deep or infratentorial CMB group than in the strictly lobar CMB group (22 ± 5 m/s vs. 20 ± 5 m/s; P = 0.001). In univariate and multivariate multinomial logistic regression analyses, baPWV was found to be independently associated with deep or infratentorial CMBs. CONCLUSIONS: Arterial stiffness was independently associated with deep or infratentorial CMBs but not lobar CMBs. These findings suggest a pathophysiological association between arterial stiffness and CMBs in the deep or infratentorial region.

The association between asymptomatic coronary artery disease and CHADS2 and CHA2 DS2 -VASc scores in patients with stroke.

BACKGROUND AND PURPOSE: CHADS2 and CHA2 DS2 -VASc scores are measurement tools that stratify thromboembolic risk in patients with non-valvular atrial fibrillation, and are predictive of cerebral atherosclerosis, fatal stroke and ischaemic heart disease. Patients with higher CHADS2 and CHA2 DS2 -VASc scores are more likely to have had an akinetic/hypokinetic left ventricular segment or a recent myocardial infarction, all of which are associated with coronary artery disease (CAD). Most of the CHADS2 score components are also risk factors for atherosclerosis. Thus, CHADS2 and CHA2 DS2 -VASc scores may be predictive of CAD. METHODS: In all, 1733 consecutive patients with acute ischaemic stroke who underwent multi-slice computed tomography coronary angiography were enrolled. The association of CHADS2 and CHA2 DS2 -VASc scores with the presence and severity of CAD was investigated. RESULTS: Of the 1733 patients, 1220 patients (70.4%) had any degree of CAD and 576 (33.3%) had significant CAD (≥ 50% stenosis in at least one coronary artery). As the CHADS2 and CHA2 DS2 -VASc scores increased, the presence of CAD also increased (P < 0.001). The severity of CAD was correlated with CHADS2 score (Spearman coefficient 0.229, P < 0.001) and CHA2 DS2 -VASc score (Spearman coefficient 0.261, P < 0.001). In multivariate analysis, after adjusting for confounding factors, CHADS2 and CHA2 DS2 -VASc scores ≥2 were independently associated with CAD. The CHA2 DS2 -VASc score was a better predictor of the presence of CAD than the CHADS2 score on area under the curve analysis. CONCLUSION: CHADS2 and CHA2 DS2 -VASc scores were predictive of the presence and severity of CAD in patients with stroke. When a patient has high CHADS2 or CHA2 DS2 -VASc scores, physicians should consider coronary artery evaluation.

The frequency of cerebral microbleeds increases with CHADS(2) scores in stroke patients with non-valvular atrial fibrillation.

BACKGROUND AND PURPOSE: Cerebral microbleeds (CMBs) are extravasations of blood from lipohyalinized or amyloid angiopathic cerebral arterioles, and the presence and numbers of CMBs are significantly associated with the development of oral anticoagulation (OA)-related intracranial haemorrhage (ICH). The aim of this study was to investigate whether there is a difference in CMBs burden according to CHADS(2) scores or CHA(2) DS(2) -VASc scores in non-valvular atrial fibrillation (NVAF) patients. METHODS: We included 550 ischaemic stroke patients who had NVAF and who had undergone brain magnetic resonance imaging (MRI) with gradient-recalled echo (GRE) T2 sequences from our prospective stroke registry between January 2005 and November 2011. We calculated CHADS(2) scores and CHA(2) DS(2) -VASc scores for all patients based on their underlying cardiovascular diseases. The presence, location and number of CMBs were assessed in each patient. We also investigated whether the CMBs were actually associated with the development of ICH during follow-up. RESULTS: The mean patient age was 70.4 ± 10.5 years, and 324 (58.9%) patients were men. One-hundred and seventy-three patients (31.5%) had CMBs detected on GRE MRI. Higher CHADS(2) scores or CHA(2) DS(2) -VASc scores were strongly associated with the presence and number of CMBs. During follow-up of median 3.1 ± 1.6 years, the presence of CMBs was independently associated with the development of ICH, whilst the CHADS(2) scores or CHA(2) DS(2) -VASc scores were not. CONCLUSIONS: Considering the positive association between the presence of CMBs and OA-related ICH, our results suggest that the increase in ICH in high-risk groups during OA may be related to an increased burden of CMBs.

Magnesium-doped zinc oxide electrochemically grown on fluorine-doped tin oxide substrate.

Nanostructures of magnesium (Mg) doped Zinc oxide (ZnO) were successfully deposited on conducting fluorine-doped tin oxide (FTO) coated glass plates by cathodic electrochemical deposition method at different potentials and temperature conditions. The deposited samples were characterized by XRD and SEM techniques to confirm their structures, morphologies and optical properties. These measurements show that Mg doped ZnO has a wurtzite structure and that the strongest intensity of the (002) peak is found at 60 degrees C and -1.0 V. Tunable transmittance of Mg doped ZnO has a band gap energy from 3.45 eV to 3.82 eV, which is the direct evidence of doping.

Low ankle-brachial index is a predictive factor for initial severity of acute ischaemic stroke.

BACKGROUND: A low ankle-brachial index (ABI) is predictive of peripheral arterial disease (PAD). For unknown reasons, patients with PAD demonstrate higher vascular mortality during follow-up than do those without. Initial stroke severity is a strong predictor of long-term outcome and may be different between patients with and without PAD. Thus, we investigated whether a low ABI was associated with severe stroke presentation. METHODS: We enrolled 1147 first-ever ischaemic stroke patients who underwent ABI measurements during hospitalization. Patients were categorized into the normal (≥ 0.90) or the abnormal (<0.90) ABI group. Baseline characteristics and initial National Institutes of Health Stroke Scale (NIHSS) scores were compared between the groups. We further analysed components of the NIHSS subscales in these groups. RESULTS: Ankle-brachial index was abnormal in 85 (7.4%) patients. Mean initial NIHSS score was higher in the abnormal ABI group (6.61 ± 6.56) than in the normal ABI group (4.36 ± 4.90) (P = 0.003). A low ABI was independently associated with higher NIHSS score in a multivariate analysis. In the abnormal ABI group, leg weakness was more severe than it was in the normal ABI group, and the contribution of leg weakness to the initial NIHSS score was higher. CONCLUSIONS: Patients with low ABI values presented with more severe ischaemic stroke. Contribution of pre-existing PAD to leg weakness may play a role in the initial severity of stroke in patients with PAD. Our findings suggest that poor clinical outcomes in patients with PAD may be partially explained by their increased likelihood for severe stroke.

Prediction of thrombus resolution after intravenous thrombolysis assessed by CT-based thrombus imaging.

The degree of thrombus resolution directly indicates the effectiveness of a thrombolytic drug. We investigated the degree of thrombus resolution and factors associated with thrombus resolution after intravenous (IV) recombinant tissue plasminogen activator (rt-PA) using thin-section noncontrast computed tomography (NCT). Thin-section NCTs were performed before and immediately after IV rt-PA infusion in acute stroke patients. The thrombus volume and Hounsfield unit were measured using three-dimensional imaging software. Immediate recanalisation was assessed immediately after IV rt-PA infusion using CT angiography. During a three-year study period, 130 patients were prospectively enrolled. On baseline thin-section NCT, no thrombi were found in 30 patients (23%). Among the 100 patients with confirmed thrombus, the median volume decreased by 20% on the follow-up NCT. The thrombus was completely resolved in 8%. Of note, an increase in thrombus volume was observed in 20 patients. Independent predictors of thrombus resolution were total rt-PA dose, thrombus location in the M2 segment of the middle cerebral artery, and time from baseline to follow-up NCT. Thrombus resolution increased by 9% per each 10-mg increase in rt-PA (p = 0.045). Immediate complete recanalisation was achieved in 12% of patients. Total dose of rt-PA was independently associated with complete recanalisation [odds ratio [OR] 4.52, 95% confidence interval [CI] 1.345-15.184) and good functional outcome at three months (modified Rankin scale score <3, OR 2.34, 95% CI 1.104-4.962). In conclusion, rt-PA dose was associated with the degree of thrombus resolution, immediate complete recanalisation, and good outcome at three months. CT-based thrombus imaging may be helpful in determining thrombolysis effectiveness.

Use of a handheld, computerized device as a decision support tool for stroke classification.

BACKGROUND: The Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification has been widely used to determine etiology of ischemic stroke. However, interrater reliability is known to be modest. The complexity of abstraction and the interpretation of various clinical and laboratory data might limit the accuracy of the TOAST classification. In this study, we developed a computerized clinical decision support system for stroke classification that can be used in a handheld device and tested whether this system can improve diagnostic accuracy and reliability. METHODS: Based on the TOAST classification, a logical algorithm was developed and implemented on a handheld device, named iTOAST. After answering six questions using the touch interface, the stroke subtype result is displayed on the screen. Four neurology residents were randomly assigned to classify stroke subtypes using iTOAST or the conventional method (cTOAST). Using a crossover design, they classified the stroke subtypes of 70 patients. The standard subtypes were determined by three stroke experts. Correlated kappa coefficients using iTOAST compared with cTOAST were determined. RESULTS: The kappa (SE) value of iTOAST [0.790 (0.041), 95% CI: 0.707-0.870] was higher than that of cTOAST [0.692 (0.046), 95% CI: 0.600-0.782] (P<0.001). Neither sequence (P=0.857) nor period effect (P=0.999) was observed. CONCLUSIONS: The stroke classification tool using a handheld, computerized device was easy, accurate, and reliable over the conventional method. It may have additional benefit because a handheld, computerized device is accessible anytime and anywhere.

Stroke mechanism in patients with non-valvular atrial fibrillation according to the CHADS2 and CHA2 DS2 -VASc scores.

BACKGROUND AND PURPOSE: The CHADS(2) and CHA(2) DS(2) -VASc scores are useful to stratify embolic risks in patients with non-valvular atrial fibrillation (NVAF) and to identify patients eligible for anticoagulation. Although the risk of stoke increases in patients with higher CHADS(2) or CHA(2) DS(2) -VASc scores, it is uncertain why the stroke rate increases in them. Concomitant potential cardiac sources of embolism (PCSE) may be more frequent in patients with higher CHADS(2) or CHA(2) DS(2) -VASc scores because stroke risks increase when concomitant PCSE is present in Atrial fibrillation (AF). On the other hand, atherothrombosis may be the cause when considering that most components of the CHADS(2) and CHA(2) DS(2) -VASc scores are risk factors for atherosclerosis. METHODS: Amongst 5493 stroke patients who were prospectively registered with the stroke registry for 11years, 860 consecutive patients with NVAF were included for this study. We investigated the mechanisms of stroke according to the CHADS(2) /CHA(2) DS(2) -VASc score in stroke patients with NVAF. RESULTS: Amongst 860 patients, concomitant PCSE were found in 334 patients (38.8%). The number of PCSE increased as the CHADS(2) /CHA(2) DS(2) -VASc score increased (P<0.001). Of individual PCSE, akinetic left ventricular segment, hypokinetic left ventricular segment and myocardial infarction <4weeks were associated with the CHADS(2) /CHA(2) DS(2) -VASc score. The presence of possible atherothrombotic mechanism, in addition to AF, was suggested in 27.3%. The proportion of patients with concomitant presence of possible atherothrombosis was increased as the CHADS(2) /CHA(2) DS(2) -VASc score increased (P<0.001). CONCLUSIONS: Increased frequency of concomitant PCSE and that of the atherothrombotic mechanism may explain the high risk of stroke in patients with higher CHADS(2) /CHA(2) DS(2) -VASc score.

Early neurological outcomes according to CHADS2 score in stroke patients with non-valvular atrial fibrillation.

BACKGROUND AND PURPOSE: A higher CHADS(2) score or CHA(2)DS(2)-VASc score is associated with an increased risk of ischaemic stroke in patients with non-valvular atrial fibrillation (NVAF). However, there are no data regarding early neurological outcomes after stroke according to the risk levels. METHODS: In this study, a total of 649 stroke patients with NVAF were enrolled and categorized into three groups: low-risk (CHADS(2) score of 0-1), moderate-risk (CHADS(2) score 2-3), or high-risk group (CHADS(2) score ≥4). CHA(2)DS(2)-VASc score was divided into four groups including 0-1, 2-3, 4-5, and ≥6. We investigated whether there were differences in initial stroke severity, early neurological outcome, and infarct size according to CHADS(2) score or CHA(2)DS(2)-VASc score in stroke patients with NVAF. RESULTS: The initial National Institutes of Health Stroke Scale (NIHSS) score was highest in high-risk group [9.5, interquartile range (IQR) 4-18], followed by moderate-risk (8, IQR 2-17) and low-risk group (6, IQR 2-15) (P=0.012). Likewise, initial stroke severity increased in a positive fashion with increasing the CHA(2)DS(2)-VASc score. During hospitalization, those in the high-risk group or higher CHA(2)DS(2)-VASc score had less improvement in their NIHSS score. Furthermore, early neurological deterioration (END) developed more frequently as CHADS(2) score or CHA(2)DS(2)-VASc score increased. Multivariate analysis showed being in the high-risk group was independently associated with END (OR 2.129, 95% CI 1.013-4.477). CONCLUSIONS: Our data indicate that patients with NVAF and higher CHADS(2) score or CHA(2)DS(2)-VASc score are more likely to develop severe stroke and a worse clinical course is expected in these patients after stroke presentation.

Effect of warfarin withdrawal on thrombolytic treatment in patients with ischaemic stroke.

BACKGROUND AND PURPOSE: Abruptly discontinuing warfarin may induce a rebound prothrombotic state. Thrombolytic agents may also paradoxically induce prothrombotic conditions, which include platelet activation and thrombin generation. Therefore, prothrombotic states may be enhanced by withdrawing warfarin in patients under thrombolytic treatment. This study was aimed to determine whether patients with warfarin withdrawal have different clinical outcomes from those without warfarin use after thrombolytic treatment. METHODS: A total of 148 consecutive patients with atrial fibrillation who were not on anticoagulants at admission and who received thrombolysis were included in this study. We compared the outcomes between a warfarin withdrawal group and a no-warfarin group. RESULTS: Fourteen patients (9.5%) were included in the warfarin withdrawal group. Although baseline National Institute of Health Stroke Scale (NIHSS) scores, recanalization rates, and hemorrhage frequencies did not differ between the groups, the warfarin withdrawal group showed poorer outcomes. Increased NIHSS scores during the first 7days were more frequent in the warfarin withdrawal group (57.1% vs. 26.9%, P=0.029). The median percent improvement in NIHSS scores at 24h after thrombolysis was also lower in the warfarin withdrawal group. After adjusting for covariates, warfarin withdrawal was a strong predictor of poor functional outcome at 3months (modified Rankin score≥3) (odds ratio, 17.067, 95% CI 2.703-107.748). CONCLUSIONS: Discontinuing warfarin was associated with early neurologic deterioration and poor long-term outcomes after thrombolytic treatment.

Failure of complete recanalization is associated with poor outcome after cardioembolic stroke.

BACKGROUND: Recanalization is strongly associated with outcomes after thrombolytic treatment. Cardiac emboli are known as better response to fibrinolytic agents because they are fibrin-rich; however, cardioembolic stroke itself is associated with poor outcomes and high mortality. Completeness of recanalization may therefore affect the outcome of cardioembolic stroke. We investigated whether degree of recanalization influences outcomes following fibrinolytic therapy in cardioembolic stroke. METHODS: Consecutive stroke patients with relevant artery occlusions on baseline CT angiography who had received thrombolytic treatment were enrolled. Completeness of recanalization was assessed by the Thrombolysis in Myocardial Infarction (TIMI) grade, which was compared between patients with and without cardiac sources of embolism (CSE). We also investigated independent predictors of poor outcome (modified Rankin scale score 3-6) at 3 months. RESULTS: Of the 127 patients enrolled, 65 (51%) had one or more CSE. Although the overall recanalization rates (TIMI 2 or 3) in patients with CSE (65%) and patients without CSE (68%) were similar (P=0.710), patients with CSE were less likely to show complete recanalization (TIMI 3) compared with those without CSE (19% vs. 39%, P=0.011). Multivariate analysis revealed that CSE was associated with failure of complete recanalization (OR 2.809, 95% CI 1.097-7.192) and was an independent predictor of poor outcome at 3months (OR 3.629, 95% CI 1.205-8.869). CONCLUSIONS: In cardioembolic strokes, failure of complete recanalization following thrombolytic therapy was frequent and was associated with poor outcome after thrombolysis.