HomeStrokeVol. 43, No. 11Abstracts for Canadian Stroke Conference Free AccessAbstractPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessAbstractPDF/EPUBAbstracts for Canadian Stroke Conference Originally published2 Oct 2012https://doi.org/10.1161/01.str.0000422054.44193.87Stroke. 2012;43:e115–e162Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: January 1, 2012: Previous Version 1 0.01Factors associated with economic cost after spontaneous intracerebral hemorrhage: a Canadian exampleSpecogna, AV; Patten, SB; Hill, MD; University of Calgary, Calgary, ABBackground: Previous studies have examined predictors of treatment costs after Spontaneous Intracerebral Hemorrhage (ICH) in various countries and healthcare systems. Currently however, there is no clear understanding of which factors are associated with cost in any Canadian setting and thus no way of understanding how to plan ICH healthcare spending. Methods: We used a retrospective case series study design and adjusted linear regression to investigate if an association existed between the total cost of ICH hospital care in a Canadian health centre and age (years), gender (female vs. male), Charlson Comorbidity Index (0 vs. ≥1), in-hospital mortality (alive vs. dead), and having surgery (no vs. yes). Economic, treatment, and patient data were obtained from administrative sources. Total inflation-adjusted hospital cost per discharge was estimated in Canadian Dollars and log transformed for all analyses. Results: Analyses were performed using 987 consecutive ICH discharges from 1999 to 2008. The total cost of ICH hospital care was highly variable (mean cost per discharge=$25,613.96±$36,116.33; min $446.81 to max $326,275.10). Older age (β=–0.0057; 95% CI: –0.0103 to –0.0011) and death in hospital (β=–0.4859; 95% CI: –0.6357 to –0.3342) were significantly associated with lower total hospital cost per discharge. Whereas Charlson Comorbidity Index of ≥1 (β=0.6489; 95% CI: 0.5063 to 0.7916) and having surgery (β=1.3552; 95% CI: 1.1893 to 1.5210) were significantly associated with higher total hospital cost per discharge. Gender was not significantly associated with cost (β=0.0727; 95% CI: –0.0679 to 0.2134). Conclusions: To our knowledge this is the first study to examine predictors of ICH treatment costs in Canada. In a Canadian health centre, ICH treatment costs were significantly associated with patient age, level of comorbidity, in-hospital mortality, and surgical treatment. This study provides evidence that it may be reasonable to consider these factors when planning health spending for ICH.0.02I.N.S.P.I.R.E.S.- in-patient support program in recovery from stroke- “the voice(s) of experience”Beaver, LG; Litowsky, L; Suddes, M; Neufeldt, C; Foothills Medical Centre, Calgary, ABBackground: 1400 people are admitted to hospital in Calgary with a stroke each year and are faced with a wide variety of challenges. The days and weeks after stroke are a frightening and difficult time for survivors and families. Health professionals care medically for survivors and strive to also offer education and information, however there is an increasing emphasis being placed upon the value of peer support within the context of a comprehensive stroke service. Methods: We have developed an inpatient peer support program (I.N.S.P.I.R.E.S.), supported by a standard process manual, multi-professional training program and a volunteer preparation and selection process. Volunteers, either stroke survivors or caregivers, have been recruited and trained to visit both acute and rehabilitation stroke and are recruited from our Living with Stroke self management courses and trained in partnership with our site Volunteer Services team. Results: Three cohorts of stroke survivors/family members have now been trained with a target of 20 volunteers recruited by the end of 2012. The volunteers come with ‘lived’ experience that cannot be duplicated by healthcare professionals and thus far have described very positive experiences of visiting and providing this service. A stroke survivor who was consulted during the initial pilot of the Program said “If you can meet someone who’s gone through this before, it shows there is hope; talking about it definitely helps”. Currently about 50 patient visits occur per month across both units. Patient and volunteer experience data is being generated and we will present this data alongside a descriptive account of the processes that have been adopted to support the work of I.N.S.P.I.R.E.S. Conclusions: Our peer support program has been successfully introduced and thus far positively evaluated. As part of the Alberta Provincial Stroke Strategy this approach is now being extended across the province.0.03Diminished cerebrovascular reserve in the unaffected hemisphere of patients with unilateral internal carotid steno-occlusive diseaseSam, K1 Poublanc, J2 Crawley, A2 Mandell, DM2 Black, SE3 Mikulis, DJ2 1. Department of Physiology, University of Toronto, Toronto, ON; 2. Division of Neuroradiology, University Health Network, Toronto, ON; 3. Division of Neurology, Sunnybrook Health Sciences Centre, Toronto, ONPurpose: To evaluate cerebrovascular reactivity (CVR) of major arterial vascular territories in patients with unilateral internal carotid artery (ICA) steno-occlusive disease. Method and Materials: In this exploratory observational study, 42 patients with unilateral severe stenosis or occlusion of the ICA and 27 healthy control subjects underwent BOLD CVR MRI at 3T (GE Healthcare, Milwaukee). Precise carbon dioxide manipulation was performed during BOLD imaging (RespirActTM, Thornhill Research, Inc) to measure flow related wash-out of deoxyhemoglobin during hypercapnea, thus enabling quantitative analysis of CVR. The patients were divided into two cohorts, 25 with right-sided and 18 with left sided unilateral ICA steno-occlusive disease. In all subjects, CVR of each major vascular territory was compared to the contralateral hemisphere. The CVR of each major vascular territory in both patient cohorts were also compared to healthy controls. Results: In both patient cohorts, a significant reduction in CVR was observed in the ipsilateral ICA territory compared to the contralateral ICA territory (right-sided patients, 0.08±0.06 vs. 0.15±0.04, P<0.005; left-sided, 0.05±0.09 vs. 0.16±0.06, P<0.005) and to controls (right-sided, 0.08±0.06 vs. 0.23±0.06, P<0.005; left-sided, 0.05±0.09 vs. 0.23±0.06, P<0.005). CVR in the unaffected hemisphere contralateral to the side of the ICA with steno-occlusive disease was reduced compared to healthy controls (P<0.01). Conclusion: Decreased CVR in the hemisphere contralateral to a unilateral ICA stenosis or occlusion was observed. This indicates that collateral blood flow support from the unaffected hemisphere to the affected hemisphere comes at a cost of reduced reserve capacity to the unaffected hemisphere. The findings suggest that there may be a reduction in functional hyperemia associated with neuronal activation, not only affecting the hemisphere ipsilateral to an occlusion, but also the unaffected hemisphere contralateral to an occlusion. It remains to be determined if “stealing” from the “rich” to support the “poor” has clinical consequences over the long term.0.04Imaging direct and indirect tract damage during motor-recovery from acute subcortical infarctsRadlinska, B1 Blunck, Y2 Leppert, IR3 Minuk, J1 Pike, B3Thiel, A1 1. Jewish General Hospital, Montreal, QC; 2. Technische Universitaet Muenchen, Munich, Germany; 3. Montréal Neurological Institute, Montreal, QCBackground: Studies in acute and chronic post-stroke neuroplasticty suggest that both morphological and molecular changes in stroke-affected fibre tracts are associated with clinical outcome. These processes can be measured noninvasily in the living human brain using diffusion tensor imaging (DTI). This paper presents prospective study results on how microstructural changes progress over time along fibres directly and indirectly affected by subcortical ischaemic stroke of the Pyramidal Tract (PT). Methods: 15 Patients with subcortical ischaemic stroke either affecting the PT (PT-group) or not (NonPT-group), as well as 6 matched controls with Transient Ischaemic Attack (TIA), underwent DTI at 3 weeks and 6 months post-stroke. From these data sets, the PT (directly affected by the stroke), Callosal Motor Fibres (CMF; indirectly affected by the stroke) and Callosal Occipital Fibres (COF; not affected by stroke) were delineated using deterministic tractography, and Fractional Anisotropy (FA) ratios were calculated (affected/unaffected hemisphere) for each tract. FA ratios (rFAPT; rFACMF; rFACOF) were compared within and between groups at both time-points, as well as anterograde and retrograde to the infarct, and were correlated with clinical outcome measures. Results: DTI data analyses revealed that mean rFAPT in the PT-group was significantly lower than both NonPT and TIA-groups initially and at follow-up, and correlated significantly with clinical outcome measures (Rsqr=0.781, p<0.01). PT-group rFACMF decreased over time and at follow-up (p=0.036, one-tailed), was significantly lower than PT-group rFACOF and NonPT-group rFACMF (p=0.004). PT-group rFACMF at follow-up correlated with rFAPT retrograde to the infarct (Rsqr=0.533; p=0.04). Conclusions: Overall, significant progressive changes in microstructural neuroimaging parameters were seen along fibres that have been either directly or indirectly affected by subcortical ischaemic stroke. This suggests that morphological changes in transcallosal fibres can be detected by DTI in subcortical thus providing morphological evidence for bi-hemeispheric involvement in reorganization of motor-networks.0.05Defining benchmarks s for stroke care in Canada do we hit the mark?Lindsay, P1 Kaczorowski, J2 Cote, R5 Fang, J4 Kapral, MK4 Hall, R4 Hill, MD3 1. Canadian Stroke Network, Ottawa, ON; 2. University of Montreal, Montreal, QC; 3. Calgary Stroke Program, Calgary, AB; 4. Institute for CLinical Evaluative Sciences, Toronto, ON; 5. McGill University, Montreal, QCBackground: Significant progress has been made in establishing integrated and coordinated stroke services across Canada. The Quality of Stroke Care in Canada Report provided valuable information on the current levels of care provided; however, challenges in interpreting the results arose due to lack of validated reliable benchmarks on key quality indicators. This project utilized the stroke audit data to develop benchmarks for Canada as a key driver for improvement in stroke services. Methods: A national stroke audit was conducted across Canada for all patient admissions for 2008–2009. The audit included all acute stroke admissions, and a simple random sampling method was used to select approximately 22% of all stroke patients in Canada. Data was collected on prehospital, emergency department and acute inpatient care quality indicators. Using the ABC benchmarking methodology, national, provincial and peer group benchmarks were calculated for a set of key quality indicators. Sensitivity analysis was conducted to create a range for each benchmark. Calculated benchmarks were then compared to published benchmarks from several countries. Results: At a national level, performance was close to the benchmarks on some key prevention indicators. Administration of acute thrombolysis, door to needle times, dysphagia screening and CT scan within 24 hours were all below optimal performance. Benchmarks varied between provinces likely as a result of performance on the audit, stroke volumes and resource limitations. Within peer groups, based on available stroke services, variations between individual hospitals were also significant. Conclusions: Benchmarks for some key quality indicators for stroke are available for the first time in Canada. These benchmarks will enable stroke programs to set ambitious evidence-based targets to improve stroke quality of care, and inform quality programs such as Accreditation Canada stroke distinction program.0.06Impact of onabotulinumtoxina therapy in patients with post-stroke spasticity (PSS): Findings from the BOTOX Economic spasticity trial (BEST)Sharma, S1 Wein, T2 Satkunam, L3 Bhogal, M4 Fulford-Smith, A5 Dhani, S4 1. Sunnybrook Health Sciences Centre, Toronto, ON; 2. McGill University, Montreal, QC; 3. Glenrose Rehabilitation Hospital, Edmonton, AB; 4. Allergan Inc., Markham, ON; 5. Allergan Ltd., Marlow, United KingdomBackground: BEST, a European and Canadian multicenter study, was designed to examine the effectiveness of BOTOX in achieving a patient’s individualized active functional goal and impact on quality of life (QoL). Methods: PSS patients naïve-to-toxin therapy were randomised to BOTOX + standard care (B) or placebo + SC (P) for up to 2 treatment cycles. The study period was 52 weeks (a 24 week double-blind phase followed by an open-label phase). The primary endpoint was the attainment of an active functional goal based on the goal attainment scale at week 24. Secondary goals (active and passive) were also set. Here, we compare outcomes for the Canadian cohort(Cc) compared to the overall study population. Results: Of 273 total patients, 74% were treated for lower limb (LL) spasticity vs 92% in Cc (n=22). Median BOTOX dose was 340U in all patients vs 485U in Cc. Primary functional goal was achieved by 40.3% vs 32.8% (B vs P; p=0.251) of all patients, compared to 54.5% vs 62.5.0% (B vs P) of Cc (p=0.049 Canada vs all patients). Secondary goal was achieved by 51.6% vs 40.7% (B vs P; p=0.079) of all patients, compared to 72.7% vs 25.0% of Cc (B vs P) (p=0.607 Canada vs all patients). Patient-perceived benefit, assessed by the SF-12 physical component score, was higher in Cc vs all patients (B: 8.529±11; P: 5.85±5.75 vs B:3.85±8.2; P: 2.3±8.2, respectively). Conclusion: In Canada, median BOTOX doses were higher than the overall study population, perhaps due to higher proportion of patients treated for LL spasticity. No significant difference was observed in the rate of achievement of the active functional goal, although there was a trend to greater achievement of secondary goals in BOTOX-treated Canadian patients, possibly contributing to higher QoL scores reported.0.07Longitudinal mapping of sensorimotor cortex after targeted stroke in miceHarrison, TC; Silasi, G; Murphy, TH; University of British Columbia, Vancouver, BCBackground: Recovery from stroke is hypothesized to involve the reorganization of surviving cortical areas. In the motor cortex this phenomenon has been investigated with intracortical stimulating electrodes, but this technique is difficult to perform longitudinally. To study the organization of sensorimotor cortex before and after stroke, we have developed techniques for light-based motor mapping of Channelrhodopsin-2 (ChR2) transgenic mice. Methods: Brief pulses of light stimulation are targeted to an array of cortical points, and the motor activity evoked by stimulation is recorded at each of these points using non-invasive motion sensors. Intrinsic signal imaging is used to generate maps of the forelimb sensory cortex. In mice implanted with cranial windows, the resulting sensory and motor maps can be repeatedly generated before and after photothrombotic infarcts are targeted to either forelimb motor or sensory cortex. Results: Infarcts targeted to sensory or motor forelimb areas caused reorganization of those areas peaking in the initial 2 weeks after stroke. Sensory maps targeted by stroke tended to shift anteromedially, into forelimb motor cortex. When the infarct was placed within forelimb motor cortex, motor maps shifted posteromedially into somatosensory cortex. After 2 months, sensory and motor maps typically remapped toward their original locations, occupying territory in the peri-infarct cortex. Interestingly, shifts in the post-stroke positions of sensory and motor areas were uncorrelated. We also observed a trend toward larger movements being produced after strokes in the motor cortex, suggesting increased cortical excitability after stroke. Discussion: The rapid recovery of sensory and motor maps is an encouraging demonstration of the impressive capacity for spontaneous reorganization of cortical circuits. These results suggest that sensorimotor cortical neurons are adept at vicariously performing the function of stroke-damaged circuits. Future experiments will attempt to identify means of enhancing cortical reorganization after stroke.0.08The iScore predicts efficacy and risk of bleeding in the NINDS t-PA stroke trialSaposnik, G1 Demchuk, A2 Tu, JV3 Johnston, CS4 1. Department of Medicine, University of Toronto, Toronto, ON; 2. University of Calgary, Calgary, AB; 3. ICES, Toronto, ON; 4. UCSF, San Francisco, CA, USABackground and Purpose: The iScore is a validated tool to estimate outcomes after an acute ischemic stroke. A previous study showed the iScore can predict clinical response and risk of intracerebral hemorrhage (ICH) after tPA. Methods: We applied the iScore (www.sorcan.ca/iscore) to participants in the NINDS tPA stroke trial to evaluate its ability to predict clinical response and risk of ICH after thrombolysis. Based on previous study, patients were stratified according to the iScore in <200 and ≥200. The main outcome measures included ICH (any type) and a global favorable outcome (defined as a modified Rankin scale of 0 or 1, NIHSS <1, Barthel index≥ 95 or GCS <1) at 3 months. Univariate and multivariable logistic regression analyses were used for the analysis. Results: Of the 624 patients enrolled in the trial, the iScore was calculated in all 624 patients. The mean iScore in the entire cohort was 161.3±44.7 (for the tPA group: 159.4±47.8; for the placebo 163.2±41.5; p=0.28). There were 507 (81%) patients with an iScore <200 and 117 (19%) with an iScore > 200. An iScore ≥200 was associated with greater risk of ICH in the tPA compared to the placebo group (30.8% vs. 11.5%; p=0.014, NNH=5). Case fatality at 3 month among those with ICH was greater in patients with an iScore ≥200 (69.2% vs. 23.8%; p<0.001; NNH 2). Participants with an iScore<200 in the tPA group had higher global favorable outcome at 3 months compared to placebo (58.7% vs 41.9%; p<0.001, NNT: 6). However, tPA administration among those patients with an iScore≥200 was not associated with significantly better outcomes (15.4% vs 13.4%; p=0.77). Conclusion: The iScore is a validated tool to estimate the clinical response and risk of hemorrhagic complications after tPA for acute ischemic stroke. Results in the NINDS trial validate those of observational studies.0.10Reperfusion and patient’s clinical parameters are stronger predictors of functional outcome than recanalization in ischemic strokeEilaghi, A1 Jakubovic, R1 d’Esterre, C2 Lee, T4,3 Aviv, R1 1. Sunnybrook Health Sciences Centre, Toronto, ON; 2. Robarts Research Institute, London, ON; 3. Lawson Health Research Institute, London, ON; 4. Robarts Research Institute, University of Western Ontario, London, ONPurpose: To compare the ability of reperfusion index and recanalization to predict the functional outcome, infarct volume, infarct growth and penumbra salvage volume in acute ischemic. Methods: Baseline computed tomography perfusion (CTP) within 4.5 hours of acute anterior circulation stroke, follow up CTP within 24 hours, and 5–7 day MRI were obtained for 114 patients. A good functional outcome was defined as 90-day mRS score <2. Reperfusion indices were derived from the admission and 24 hours post CTP maps. Univariate and multivariate logistic regression were used to investigate the association of clinical and imaging parameters with the 90-day mRS, infarct volume, infarct growth and penumbra savaged. Results: Cerebral Blood Flow (CBF), Cerebral Blood Volume (CBV), Mean Transit Time (MTT) and Tmax perfusion parameters were significantly different in patients who recanalized than those who did not (p<0.001). On univariate analysis reperfusion, recanalization and patient’s clinical and radiological parameters showed significant association with outcome (p<0.05). Only reperfusion (p=0.004), National Institutes of Health Stroke Scale (NIHSS) (p=0.001), age (p=0.006), hyperglycaemic status (p=0.047) and recombinant tissue plasminogen activator (rtPA) treatment (p=0.049) retained significance in the multivariate model. All reperfusion indices were strongly correlated with outcome but Tmax index >60% showed highest predictive value. Patients without recanalization and positive reperfusion status had significantly lower total infarct volume (p=0.012), infarct growth (p=0.042) and higher salvaged penumbra volume (p=0.028). Conclusions: Reperfusion index is more strongly associated with good stroke outcome than recanalization.0.11Leptomeningeal collaterals: Role in infarct growth reduction in patients with acute ischemic strokesNambiar, VK1 Sohn, S2 Quazi, E1 Quazi, A1 Kosior, J1 Goyal, M1 Hill, MD1 Demchuk, AM1 Menon, BK1 1. University of Calgary, Calgary, AB; 2. Keim Yong University, Keimyong, KoreaIntroduction: We explore the relationship between baseline infarct volume, time from baseline imaging to recanalization and leptomeningeal collateral status in predicting infarct growth and final clinical outcome in patients with acute ischemic stroke. Methods: Data is from a prospective study of consecutive acute stroke patients (2005–2009) from Keimyung University, South Korea analysed at University of Calgary. Only patients with M1 MCA± intracranial ICA on baseline CT-angio, known stroke onset time with MR DWI at baseline and follow-up were included for analyses. Baseline infarct volume on DWI was calculated using Quantomo. Infarct growth was calculated as the difference in DWI volume between 24 hrs and baseline and collaterals at baseline CT-angio assesed using a previously published scale. Results: Of 84 patients [mean age 65.2±13 yrs, 52.4% male, median NIHSS 14 IQR=8.5, median time from onset of stroke symptom to baseline MR 164 mins, IQR 100.5) in the study, 59.5% achieved TIMI 2–3 recanalization and 35.7% good clinical outcome (mRS 0–2). Median baseline DWI volume was 31.6ml (IQR 75); median infarct growth 29.8ml (IQR 64.1). We noted significant correlation (spearman’s r=–0.68, p<0.001) between baseline DWI volume and collateral status and between infarct growth and collateral status (r=–0.38, p<0.01). No correlation (r=0.1, p=0.34) was noted between infarct growth and baseline MR to recanalization time. In patients with baseline DWI volume <=18ml (group 1), 61.3% of patients achieved good clinical outcome when compared to 26.7% in patients with volume 18–80ml (group 2) and 13% with volume >80ml (group 3). Median infarct growth was lowest in group 1 (5.9ml IQR 52.7) followed by group 2 (30.5ml IQR 70.6) and group 3 (63ml IQR 108.1). Conclusion: Leptomeningeal collateral status not only determines baseline volume of infarct but also extent of infarct growth until recanalization is achieved.0.12Skill maintenance of dysphagia screening on the TOR-BSST: a randomized controlled trialMartino, R1 Breadner, B2 Wellman, L3 Kamitomo, G3 Mitchell, A4 O’Connor, L5 Durkin, L6 Churchward, K3 Powell-Vinden, B7 Diamant, N8 1. University of Toronto, Toronto, ON; 2. Georgian Bay General Hospital, Midland, ON; 3. Alberta Health Services, Calgary, AB; 4. Royal Alexandra Hospital, Edmonton, ON; 5. Lourdes Hospital, Binghamton, NY, USA; 6. University Health Network, Toronto, ON; 7. Heart and Stroke Foundation of Ontario, Toronto, ON; 8. Queen’s University, Kingston, ONScreeners must be competent in identifying dysphagia risk. Our objectives were to determine screening skill performance over time and to assess the benefit of independent web-based practice. This study first developed an interactive independent web-based dysphagia-screening refresher, followed by a prospective randomized controlled trial enrolling screeners with and without previous training by speech-language pathology (SLP) on the TOR-BSST© screen. Using concealment, screeners were randomly assigned to no access or full access to the new web-based refresher. For screeners in both arms, screening skill was assessed at 5 time-points: baseline, after live SLP training, 12-weeks, 24-weeks, and 36-weeks. Logistic regression modeling assessed for effect of randomization arm, training and study week. Across 4 North American stroke centers, 98 screeners (58 with and 38 without previous SLP training) were enrolled and equally randomized. Between arms, no differences were identified in screener demographic and practice variables or performance on final judgments. Screeners with full access who practiced more outperformed those who practiced less on water swallow (OR=1.51, 95% CI 1.16,1.95) and voice quality (OR=1.34, 95% CI 1.03,1.75) items. Across all screeners, those with previous SLP training outperformed screeners without such training at baseline only (p<.02); final judgment performance increased over time (p<.001) and outperformed single item judgments at all times (p<.001). Live SLP TOR-BSST© training improves screening performance to highly accurate level. All trained screeners maintained high accuracy over time regardless of access or amount of practice with web refresher. Performance on total score was consistently more accurate than any item alone.0.13Implementing a clustered acute stroke unit at a community hospital improves patient careDykes, LJ; Small Sekeris, A; Mason, C; Maxfield, S; Matchett, D; Roger, S; Steeves, K; Bluewater Health, Sarnia, ONBackground: Stroke unit care was a priority recommendation of Canadian Best Practice Recommendations for Stroke Care in 2010. While acute stroke units have been established at large centres and teaching hospitals, smaller patient volumes and varied resources challenge community hospitals to make this recommendation a reality. Methods: Since 2008, various models for implementing acute stroke unit care were considered and trialled at a District Stroke Centre community hospital in Southwestern Ontario. A stand alone unit was not feasible due to fluctuating patient volumes and resource requirements. Clustering stroke patients on a large general medical unit was trialled. This model was unsuccessful due to staffing models, resource neutral funding, isolation requirements, patient volumes, and hospital processes. After re-evaluation, on February 1, 2011, a clustered acute stroke unit was opened within an existing 20 bed telemetry medicine unit. Results: In the first year of operation, 192 of the 231 patients admitted with a diagnosis of stroke received care on the Acute Stroke Unit. With consistent clustering of patients to one location, the interprofessional team reported improved communication and collaboration. Patient flow improved; the time from stroke onset to admission to inpatient rehabilitation was reduced from 8 days (median) in FY 2009–10 to 6 days in FY 2011–12 (Q1-Q3). Atrial fibrillation was diagnosed in 10 stroke patients in the emergency department while acute stroke unit monitoring lead to an additional 15 diagnoses, an increase of 150%. Endarterectomy consultation increased from 3 in the previous year to 10. Conclusions: Acute stroke unit care is possible and effective within a community hospital setting and can be implemented in a resource neutral environment. A clustered care model in a telemetry medicine unit can improve processes and patient outcomes.0.14The economic cost of spontaneous intracerebral hemorrhage over one decade in a Canadian centreSpecogna, AV; Patten, SB; Hill, MD; University of Calgary, Calgary, ABBackground: Spontaneous Intracerebral Hemorrhage (ICH) is a devastating condition which affects many Canadians each year. Currently, there is no clear understanding of the cost to treat ICH in Canada, and thus no way of understanding ICH resource utilization, how to budget for ICH, or how any new treatments would impact ICH health spending. Methods: We used a retrospective case series study design to estimate the direct and indirect cost of ICH hospital care from 1999 to 2008 in one Canadian centre. Inflation-adjusted costs per discharge were estimated in Canadian Dollars for ambulatory care, audiology and speech therapy, cardiodiagnostic lab services, clinical lab services, clinical nutrition services, diagnostic investigation services, nursing unit care, occupational therapy, pharmacy services, physiotherapy, recreation therapy, respiratory therapy, social work, and surgery for every calendar year using administrative data. Results: Cost data were available for 987 consecutive discharges. The total cost of ICH hospital care was highly variable (mean cost per discharge for all years combined=$25,613.96±$36,116.33; min $446.81 to max $326,275.10). Overall, direct costs represented the highest proportion of total cost per discharge (mean=$21,076.27±$29,930.40; min $344.50 to max $271,907.40), whereas indirect costs represented the lowest (mean=$4,532.67±$6,317.04; min $68.62 to max $54,367.66). Specifically, direct nursing unit cost represented the highest proportion of total cost per discharge (mean=$15,047.49±$22,312.86; min $21.92 to max $197,085.20), whereas indirect day surgery cost represented the lowest (mean=$0.29±$5.28; min $0 to max $118.28). There was no significant difference in mean inflation-adjusted total cost between 1999 to 2003 and 2004 to 2008 (p=0.459). Conclusions: To our knowledge this study provides the first ever report of the cost of ICH hospital care in Canada. The cost of care per discharge was highly variable but not different across a previous decade.0.15SWI is more reliable than T2*-weighted GRE MRI for detection of microbleeds in cerebral amyloid angiopathyCheng, A; McCr