O9
Neurological recovery post ischaemic stroke: the ocular motor system as a surrogate marker for motor and cognitive recovery
Wei D1,2, Yan B1,3, Fielding J1, Millist L1, Davis S1,3, White O1,3
1Department of Neurology, Royal Melbourne Hospital, VIC, Australia; 2Department Neurology, West China Hospital, Sichuan University, China; 3University of Melbourne, VIC, Australia
Background: The standard neurological assessments of stroke patients are weighted significantly towards motor function and consistently underestimate cognitive deficits. Of note, ocular motor function has been demonstrated as a reliable quantifier of cognitive dysfunction in degenerative neurological diseases and may be a valuable tool in the assessment of neurological recovery post stroke.
Aims: To evaluate ocular motor function as a surrogate marker for cognitive function in post stroke patients.
Methodology: This was a prospective observational study with age-matched controls. We included acute stroke patients without visual field defect and gaze palsy. Patients were examined at onset, 1 month and 3 months post stroke ictus by ocular motor function, National Institutes of Health Stroke Scale (NIHSS), Modified Rankin Scale (mRS) and standard cognitive function assessments.
Results: We recruited 15 patients (Mean age: 62.7, mean NIHSS: 2.25) and 10 controls (Mean age: 62.5). Ocular motor function showed significant difference between the two groups at stroke onset as well as between first test and follow up in patients. It was more sensitive in identifying cognitive improvement compared with NIHSS and mRS.
Conclusions: Ocular motor assessment may provide improved quantifiable measurements of cognitive recovery post acute stroke.
O22
Are older patients with stroke a disadvantaged population? a systematic review
Luker J1, Edwards I1, Bernhardt J2, Wall K3 and Grimmer-Somers K4
1University of SA, Adelaide, SA, Australia; 2University of Melbourne, Melbourne, VIC, Australia; 3Flinders Medical Centre, Bedford Park, SA, Australia; 4International Centre for Allied Health Evidence, University of South Australia, Adelaide, SA, Australia
Background: Adherence to recommended clinical practices improves acute stroke outcomes with the same magnitude of effect for stroke patients in all age groups. Inferior outcomes for older stroke patients may in part be due to the provision of sub-optimal care.
Aims: To review evidence of the quality of the care provided to older patients with acute stroke compared to younger patients, as determined by adherence to recommended processes of care.
Methodology: A systematic review of the literature (1995–2009) identified publications that analysed data for specific process indicators within patient age groupings. Process indicator data was extracted and the quality of included studies was critically appraised.
Results: From 150 potential studies, eight met the inclusion criteria. Of the 54 process indicators audited in these studies 26 (48%) of indicators demonstrated significantly poorer care for older patients compared to younger ones, while younger patients received comparatively inferior care in 12 processes of care (22%). Age-related differences in quality were found in areas of stroke assessment, treatment, rehabilitation, secondary prevention and discharge planning.
Conclusions: The literature demonstrated that older stroke patients are disadvantaged in some aspects of care. The ageing population and the increasing proportion of older acute stroke sufferers make this a priority area for improvement.
O25
Dysphagia screening in acute stroke – a systematic review
Francis L, Denisenko S and Bilney B
Department of Health, Melbourne, VIC, Australia
Background: Clinical evidence supports the implementation of formal dysphagia screening (DS) programs in acute stroke. However, variability in published studies makes it difficult to identify an optimal method of clinical assessment.
Aim: To identify psychometrically robust and clinically feasible DS tools for the acute stroke population.
Method: A systematic review was conducted using OVID, CINAHL and Medline using terms including acute stroke, dysphagia and screening.
Results: Detection of aspiration by DS has been variable with sensitivity ranging from 42% to 92% and specificity between 59% and 91% [1,2] Positive predictive values for DS range from 50% to 75% and negative predictive values range from 70% to 90% [1]. Reliability is poorly investigated with variable rates reported (k = 0–1) [1]. The introduction of formal DS programs assured optimal patient outcomes, improved quality of care and proved cost effective [3]. Studies have demonstrated the relative risk reduction of pneumonia and mortality was clinically significant, ranging from 70.0% to 85.1% [3,4].
Conclusion: Current evidence does not identify a preferred DS tool for the acute stroke population. Research does demonstrate that the institution of a formal DS program results in improved patient outcomes. A coordinated, evidence based model for implementing DS for acute stroke is required.
O26
Dysphagia screening in acute stroke: a survey of existing practice
Francis L and Denisenko S
Department of Health, Melbourne, VIC, Australia
Background: Dysphagia Screening (DS) within 24 hours of acute stroke is recommended best practice in Australia [1] Audit data demonstrates 47% of patients in Victoria were not screened prior to receiving food or drink [2].
Aim: To evaluate existing DS practice in Victoria and to identify barriers to implementation of best practice.
Method: A survey assessing current service provision for DS, access to Speech Pathology Service (SPS), local barriers and issues, training and competency packages and monitoring methods was conducted.
Results: Sixty health services completed the survey (75% response rate). The majority of services (68%) do not conduct DS. Few (15%) of these have out of hours SPS. 40% of metropolitan and 29% of all regional health services conduct DS. DS methods vary widely and 58% of sites use self-generated non-validated tools. The majority of health services that conduct DS report having a training program in place, however 52% do not have a formal competency program. Barriers to implementing DS included: access to staff, time constraints, lack of access to SPS, acceptance of early NGT feeding and lack of compliance to protocols.
Conclusion: Current dysphagia screening practice in Victoria is inconsistent with poor adherence to guideline recommendations. A consistent evidence based framework to address this issue is required.
O27
The clinical predictive value of the ABCD2 score for early stroke risk is low in TIA patients presenting to an Australian tertiary hospital
Sanders L1,2, Srikanth V1,2, Psihogios H3, Ramsay D2, Wong K1 and Phan T1,2
1Stroke and Ageing Research Group, Department of Medicine, Monash Medical Centre, Monash University, Melbourne, VIC, Australia; 2Stroke Unit, Southern Health, Melbourne, VIC, Australia; 3Emergency Department, Southern Health, Melbourne, VIC, Australia
Background: Australian guidelines recommended use of the ABCD2 score in clinical decision making, however it has demonstrated varying performance in validation studies.
Aim: Determine the predictive value of ABCD2 for stroke in Australian TIA patients.
Methods: Cohort study of patients with TIA referred by the emergency department of a tertiary hospital (June 2004–November 2007). Sensitivity, specificity, predictive values, likelihood ratios and area under the curve (AUC) were calculated for stroke at 2 and 90 days.
Results: Stroke occurred in 4/292 (1.37%, 95% CI: 0.37–3.47%) and 7/289 (2.42%, 95% CI: 0.98–4.93%) confirmed TIA patients within 2 and 90 days respectively. AUCs for stroke outcome were 0.80 (CI: 0.68–0.91) and 0.62 (CI: 0.40–0.83) within 2 and 90 days respectively. At ≥ 5, ABCD2 had modest specificity (0.58–0.42), with poor positive predictive values (0.03–0.04) and positive likelihood ratios (2.40–1.71) for stroke within 2 and 90 days respectively. The score performed similarly poorly at other prespecified cut-off scores.
Conclusions: Given its poor predictive value, use of ABCD2 alone may not dependably guide clinical treatment decisions or service organisation in an Australian tertiary setting. Validation in other Australian settings is recommended before it can be applied with confidence.
O28
The perth intravenous minocycline stroke study (PIMSS)
Blacker D1,2, Prentice D3, Kohler E3, Bhangu J3 and Hankey G2,3
1Sir Charles Gairdner Hospital, WA, Australia; 2The University of Western Australia, WA, Australia; 3Royal Perth Hospital, WA, Australia
Background: The tetracycline antibiotic minocycline has properties that make it an exciting candidate for neuroprotection in ischaemic (IS) and haemorrhagic stroke (HS). There are numerous positive animal studies of minocycline and a small human study of orally administered minocycline showed promising results.
Aims: To conduct a pilot study of the feasibility, safety and efficacy of intravenous (IV) minocycline administered within 24 hours of stroke onset. PIMSS utilises the IV route to ensure absorption, and hopefully achieve more rapid bioavailability.
Methods: A multicentre prospective randomised open-label blinded endpoint evaluation (PROBE) pilot study of IV minocycline, 200 mg daily for 2 days for 40 patients with ischaemic or haemorrhagic stroke, commencing within 24 hours of symptom onset, compared with routine stroke care for 40 patients.
Results: By mid April 2010, 14 patients have been randomised; 7 to minocycline, and 7 to standard stroke unit care, including 2 concurrently treated with thrombolysis.
Discussion: This pilot study, and two other current studies of minocycline will contribute to the clinical data available on this agent, and may ‘pave the way’ for further, more definitive work. The possibility of additional studies, particularly in combination with thrombolytic agents (with the exciting potential to reduce haemorrhagic cerebral complications) are discussed.
O31
Assessment and management of cognitive impairment following stroke
McDonnell MN1, Smith AE1,2 and Bryan J3
1Sansom Institute for Health Research, University of South Australia, Adelaide, SA, Australia; 2Developmental Neuromotor and Plasticity Group, The Robinson Institute, The University of Adelaide, Adelaide, SA, Australia; 3School of Psychology, Social Work and Social Policy, University of South Australia, Adelaide, SA, Australia
Background: Cognitive dysfunction affects two-thirds of stroke patients and stroke is associated with a ten-fold increase in dementia. A recent audit revealed that Australian rehabilitation units lack a coordinated approach to the assessment and management of cognitive impairment post-stroke.
Aims: To develop a standardised cognitive assessment tool appropriate for use in rehabilitation centres and randomised controlled trials involving the stroke population.
Methods: We reviewed the literature to ascertain which neuropsychological tests may be useful for detecting changes in post-stroke cognition for intervention studies. We evaluated a battery of cognitive assessments with 17 community-dwelling stroke survivors and 13 healthy controls to confirm which tests were sensitive to detect cognitive dysfunction post-stroke.
Results: Our short cognitive assessment protocol revealed that four tests of cognition were able to detect differences in the domains of executive function, memory and information processing speed (P < 0.05) between healthy adults and stroke survivors.
Conclusion: In Australia, evidence-based recommended therapy was only provided to low numbers of stroke patients with cognitive impairments despite the high prevalence of cognitive deficits. Our data support the use of a short protocol of cognitive assessments which can be used for rehabilitation and intervention studies to evaluate changes in post-stroke cognition.
O32
Control of hypertension is poor in 10-year survivors of stroke
Kim J1,2, Gall SL3 and Thrift AG1,4
1Stroke Epidemiology, Baker IDI Heart and Diabetes Institute, Melbourne, VIC, Australia; 2Department of Physiology, Monash University, Melbourne, VIC, Australia; 3Menzies Research Institute, Hobart, TAS; 4Department of Epidemiology & Preventive Medicine, Monash University, Melbourne, VIC, Australia
Background: Control of blood pressure after stroke is important for reducing the risk of recurrent stroke.
Aim: To determine the control of hypertension in a community-based population of 10-year stroke survivors.
Methods: Cases of first-ever or recurrent stroke, excluding subarachnoid haemorrhage, from the North East Melbourne Stroke Incidence Study were interviewed at 10 years post-stroke. Individuals were classified as normotensive, controlled hypertensive, uncontrolled hypertensive, or uninformed hypertensive based on their measured BP level (cut-point 140/90 mmHg), past history of hypertension, and use of antihypertensive medications.
Results: At 10 years post-stroke, 371 (23.3%) of 1589 cases were alive. Of these, 297 (80%) had complete data on BP, antihypertensive medication use, and history of hypertension. Those with complete data were 5 years older than those without. Eighty-five percent were hypertensive; 58% had controlled hypertension, 26% had uncontrolled hypertension, and 16% were unaware that they were hypertensive. Overall, 83% of individuals classified as uncontrolled or uninformed hypertensive subjects were receiving treatment that was insufficient to achieve target BP levels. This is despite the fact that two-thirds of those on medication were taking at least two antihypertensive agents.
Conclusion: Considerable improvement can be made in the control of hypertension after stroke.
O33
Increased Cannabis use in young stroke patients
Pridmore H1, Anderson N2, Spriggs D3, Roberts S4 and Barber A5
1Neurology Stroke Fellow, Auckland University, New Zealand; 2Neurologist, Auckland City Hospital (ACH), New Zealand; 3Physician, Auckland City Hospital, New Zealand; 4Microbiologist, Labplus laboratory, Auckland City Hospital, New Zealand; 5Stroke Foundation Professor of Neurology, Auckland University, New Zealand
Introduction: There is a temporal relationship between cannabis use and ischemic stroke in case series and population based studies. We aimed to determine the prevalence of cannabis in urine drug screens in younger stroke patients.
Methods: Urine samples were obtained and tested for cannabis in ischemic stroke patients aged 18–55 years. The prevalence of positive urine drug screens was compared with that of the general population published by the New Zealand Drug Foundation.
Results: Sixty one of 90 (68%) patients aged 18–55 years presenting with stroke/TIA had urine drug screens performed between January 2009 and March 2010. Eleven of 61 patients (18%; nine men and two woman) tested positive for cannabis. This was greater than that of the NZ population in the past 12 months, with three of seven (33%) patients versus 18% of the population (P = 0.25) in 25–34 age group, one of fourteen (7%) compared with 10% in the 35–44 year age group (ns), and seven of 36 (19%) versus 5% in the 45–55 year age group (P < 0.000001).
Conclusion: We are collecting a control cohort of non-stroke patients to further confirm the relationship between cannabis and stroke. These preliminary results support the introduction of urine drug tests in younger stroke patients.
O37
Improving paramedic accuracy in activating stroke alerts
Ingall TJ1, Aguilar MI1, Demaerschalk BM1, Dodick DW1, Vargas BV1, Bobrow BJ2, Ducote J3
1Cerebrovascular Diseases Center, Mayo Clinic Hospital, Phoenix, AZ, USA; 2Department of Emergency Medicine, Maricopa Medical Center, Phoenix, AZ, USA; 3Scottsdale Fire Department, Scottsdale, AZ, USA
Background/Aims: Mayo Clinic Hospital (MCH) utilizes a Stroke Alert (SA) system to activate the Stroke Team (ST) when paramedics prenotify the emergency department (ED) that they are transporting patients identified as possibly having an acute stroke. The ST responds promptly to pre-hospital SAs to determine patient eligibility for acute stroke treatment. Since inappropriate SA activation can disrupt ST patient care, a quality improvement education process was developed to improve paramedic accuracy activating SAs.
Methods: Data on consecutive SAs activated by paramedics transporting stroke patients to MCH were collected prospectively over a 19 month period. MCH ST physicians reviewed each SA for accuracy. During the study, paramedics participated in 2-hour SA training bi-annually along with focused SA debriefing sessions. Differences in the percentages of incorrectly activated SAs over three consecutive time periods were assessed.
Results: From 1/9/2008 through 31/3/2010 there were 290 consecutive SAs activated by pre-hospital prenotification. The percentages of incorrectly activated SAs over three consecutive time periods were 26%, 13%, and 8% respectively. A Chi-Square test showed that this decline in percentages was statistically significant (P = 0.001).
Discussion: We found that on-going paramedic training, debriefing, and feedback resulted in a significant improvement in paramedic SA activation accuracy.
O38
The Australian stroke clinical registry: formative evaluation
Cadilhac D1,2, Lannin N3,4, Anderson C3,4, Levi C5, Faux S6, Price C7, Paice K1, Middleton S6,8 Donnan G1,2 on behalf of the AuSCR Consortium Partners
1National Stroke Research Institute a subsidiary of Florey Neuroscience Institutes, VIC, Australia; 2The University of Melbourne, VIC, Australia; 3Sydney Medical School, The University of Sydney, NSW, Australia; 4The George Institute for International Health, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 5University of Newcastle & Hunter Medical Research Institute, Newcastle, NSW, Australia; 6St Vincent's Hospital, Sydney, NSW, Australia; 7National Stroke Foundation, Melbourne, VIC, Australia; 8Australian Catholic University, Sydney, NSW, Australia
Background: The Australian Stroke Clinical Registry (AuSCR) was established in 2009 to provide prospective, systematic data on clinical processes and outcomes for stroke.
Aims: To describe factors that enhanced or impeded implementation of AuSCR in hospitals within the first year.
Methods: AuSCR is an online registry developed according to national operating principles and technical standards. Features include: opt-out consent; collection of an agreed minimum dataset; and 3 month outcome assessments. Formative evaluation included clinician interviews and randomly sampled medical record audits to verify data.
Results: Six hospitals participated. Data for 381 patients was provided (av. age 70 years, 51% male, 63% Australian). Feedback from clinicians was that the web-tool is simple to use and the user manuals and training were appropriate. However, clinicians desire automated data-entry methods for routine demography variables and the opt-out consent protocol was sometimes problematic. Currently, about 60 cases per month are being submitted; 67% ischaemic stroke, 15% intracerebral haemorrhage, 15% TIA and 3% undetermined aetiology. Among the cases entered, 75% were treated in a stroke unit and 4% of ischaemic strokes received intravenous thrombolysis.
Conclusion: Formative evaluation value-adds to registry development. Findings are used to improve AuSCR and demonstrate the successful implementation of this registry.
O39
Understanding the experiences of caring for someone after stroke: a qualitative study of caregivers and stroke survivors
Masry YEL1, Hackett M2 and Mullan B3
1School of Psychology, The University of Sydney, NSW, Australia; 2Neurological and Mental Health Division, The George Institute for International Health, NSW, Australia and Sydney Medical School, The University of Sydney, NSW, Australia; 3School of Psychology, The University of Sydney, NSW, Australia
Background: The majority of stroke survivors (SS) live with a permanent disability, and require some form of caregiver support. Research into the nature of stroke caregiving however, remains limited. Current literature indicates that support services and interventions for caregivers, especially in relation to managing affective symptoms, are unsatisfactory (Hackett & Anderson, 2006). The main aim of this study, therefore, was to investigate stroke caregiver experiences and needs, and their evaluation of existing support services.
Method: Twenty informal caregivers participated in individual semi-structured qualitative interviews covering all aspects of caregiving. Ten SS were also interviewed (predominantly for triangulation purposes) to discuss their views on their caregiver's experiences. Sampling continued until no new themes emerged. Data analysis and interpretation was primarily thematic, with the generation of an integrative model.
Results: Five inter-related master themes with various subcategories emerged from the data: Relationships and Support; Caregiver Factors; SS Factors; External Stressors and Positive Outcomes.
Conclusions: Caring for a SS involves a complex process of several different factors, all of which interact in different ways according to the individual. The study findings increase our understanding of caregiver experiences and needs, and inform the future development of improved and tailored support services and resources.
O40
Impact of smoking on the development of stroke and coronary heart disease in a general Japanese population: the Hisayama study
Hata J1, Doi Y2, Ninomiya T2, Fukuhara M1, Kitazono T2 and Kiyohara Y1
1Department of Environmental Medicine, Kyushu University, Fukuoka, Japan; 2Department of Medicine and Clinical Science, Kyushu University, Fukuoka, Japan
Background: Smoking is an established risk factor for stroke and coronary heart disease (CHD) in western countries. However, it has been unclear whether smoking raises the risk of stroke in Japanese.
Aim: We examined the effect of smoking on the development of stroke and CHD in a prospective cohort study of residents of the town of Hisayama in Japan.
Methods: A total of 2,421 residents of Hisayama aged 40–79 years with no history of cardiovascular disease were followed up for 14 years from 1988 to 2002.
Results: Compared with non-smokers, multivariate-adjusted relative risks (95% confidence intervals) of total stroke incidence were 1.50 (0.90–2.62) in ex-smokers, 1.89 (1.17–3.05) in current light smokers (1–19 cigarettes per day) and 2.01 (1.11–3.64) in current heavy smokers (≥20 cigarettes per day). Those of CHD were 1.10 (0.56–2.16), 1.88 (1.02–3.47) and 2.31 (1.17–4.58), respectively. In regard to stroke subtypes, current smoking was an independent and significant risk factor for ischaemic stroke and subarachnoid haemorrhage, but not for intracerebral haemorrhage. The combination of smoking and hypercholesterolemia significantly increased the risk of these diseases.
Conclusion: Smoking raised the risk of the ischaemic stroke, subarachnoid haemorrhage and CHD occurrence, and hypercholesterolemia strengthens this effect in Japanese.
O41
Delay in reaching hospital in stroke and in-patient mortality; a prospective hospital based study from southern India
Nagaraja V, Sankapithilu GB and Khan MA
Mysore Medical College and Research Institute, Mysore, Karnataka, India
Background: Stroke is one of the leading causes of death and disability in developed nations.
Aim: To determine delay in reaching hospital after stroke onset and in-hospital mortality in South Indian population.
Methods: Prospective hospital based study was conducted in a tertiary health care and academic center in South India. In stroke patients, delay in reaching hospital and in-hospital mortality was calculated and factors associated with mortality were looked for. Survival was calculated using Kaplan-Meier curve and Cox-proportional hazards model was constructed to identify independent predictors of inpatient mortality.
Results: Of 134 subjects studied, 73.4% were male with mean (SD) age of 53.8(18.0) years. Median delay in reaching hospital was 9-hour and only 18(13.4%) reached hospital within 3 hours and 31(20.3%) reached within 6-hour. In-hospital mortality was 26.1% and had no relation with age (P = 0.054), gender (P = 0.285), prior stroke episodes (P = 0.986), underlying hypertension (P = 0.446), and side of involvement (P = 0.209). Inpatient mortality was significantly higher in smokers (P = 0.003), diabetics (P = 0.012) and in those who could not afford Computed Tomography (P = 0.007).
Conclusion: Majority of patients did not reach hospital early enough to receive thrombolytics and had high in-hospital mortality. Diabetes and smoking were independent predictors of in-hospital mortality.
O42
Patient counting of exercise repetitions is a valid means of quantifying dosage in rehabilitation: an observational study
Scrivener K1,2, Sherrington C2, Schurr K1 and Treacy D1
1Bankstown-Lidcombe Hospital, Sydney, NSW, Australia; 2The George Institute for International Health, The University of Sydney, NSW, Australia
Background: Measuring therapy ‘dosage’ in rehabilitation is a complex task for researchers. The standard method used to quantify dosage is the time patients spend in therapy. A more accurate measure is to count each repetition of exercise the patient completes. The few published studies that have used repetitions to measure dosage used an external observer; this is a labor-intensive process that is impractical for large studies. An alternative strategy is for patients to count their own exercise repetitions whilst practicing.
Aim: To investigate the concurrent criterion-related validity of patient counting of exercise repetitions as a measure of the dosage of therapy.
Methods: An observational study of 40 rehabilitation inpatients was conducted. Participants were considered by treating physiotherapists to be able to accurately count each exercise repetition and had varied diagnoses including: stroke (43%), other neurological conditions, amputation and orthopedic conditions. Observations were completed for a 30-minute period by an external observer. Observed repetitions were compared to the participants' own tally of repetitions for that time period.
Results: The correlation between the subject and observer practice totals was 99%. Accuracy was similar among participants with neurological and non-neurological diagnoses.
Conclusion: Patient counted exercise is a valid means of quantifying therapy dosage in rehabilitation.
O43
Incidence of subarachnoid haemorrhage in China: preliminary analysis
Shiue I1, Zhang J2, Arima H1, Liu G2, Li Y2, Wang M2 and Cheng G2, Wan L2, Lv L2 and Anderson CS1 for the CHERISH Group
1The George Institute for International Health, Royal Prince Alfred Hospital and University of Sydney, Sydney, NSW, Australia; 2Department of Neurology, Baotou Central Hospital, Baotou, Inner Mongolia, China
Background and aims: There is limited epidemiological data on subarachnoid haemorrhage (SAH) in China.
Methods: The CHina Epidemiology Research In Subarachnoid Haemorrhage (CHERISH) study is a prospective, population-based, case-control study of SAH in a geographically defined segment (1.8 million) of Baotou, Inner Mongolia, China. Cases of spontaneous SAH are identified through surveillance of 11 collaborating neurology/neurosurgery hospitals, multiple smaller hospitals and health clinics, and the single city crematorium. Verbal autopsy procedures are used to ascertain probable out-of-hospital cases with rapid deaths. For each case, two non-related controls without SAH are matched by age (5-year strata), gender, and district of residence.
Results: A total of 53 SAH cases (mean age 56 ± 12 years; 70% female) were registered in the first 6 months of the planned 2 year study period from May 2009. Exposure frequencies were 30% current smokers, 42% history of hypertension, and 19% current drinkers. The crude annual incidence of SAH was estimated at 6.0 per 100,000 (3.6 males; 8.6 females), and 8.2 per 100,000 standardised to the World reference population.
Conclusions: These preliminary data indicate similar age adjusted incidence of SAH in China to most Western populations, but potentially higher gender disparities in rates.
O44
Coronary artery disease and stroke are leading causes of death in China: mortality survey results in Baotou, inner Mongolia
Zhang J1, Wang M1, Shiue I2, Cheng G1, Arima H2, Liu G1, Li Y1 and Anderson CS2 for the CHERISH Group
1Department of Neurology, Baotou Central Hospital, Baotou, Inner Mongolia, China; 2The George Institute for International Health, Royal Prince Alfred Hospital and University of Sydney, Sydney, NSW, Australia
Background and aims: Cause-specific mortality data are incomplete in China. The CHina Epidemiology Research In Subarachnoid Haemorrhage (CHERISH) study includes a comprehensive mortality survey in Baotou, the largest city in Inner Mongolia, China.
Methods: A 10% sampling (1 week every month for 6 months) of all death certificates registered with the single city crematorium were reviewed. Details of out-of-hospital deaths ascertained by verbal autopsy methods on key proxy sources, and reviewed centrally to derive a final cause of death.
Results: Overall, 44% of all deaths were due to cardiovascular disease, either coronary heart disease (36%) or stroke (8%). Cancer accounted for about one fifth of deaths. Among the out-of-hospital deaths (84% lived alone; 90% died at home), which included one quarter (13/53) of all stroke deaths, being unaware of the urgency of the illness (in 46%) or financial/personal reasons (29%) were stated as the most common explanations for avoidance of hospital care.
Conclusions: Cardiovascular disease is the leading cause of death and one quarter of all stroke deaths occur outside of hospitals in Baotou, Inner Mongolia, China. These data emphasize public health efforts to improve prevention strategies and awareness of stroke symptoms and need for urgent assessment.
O45
Blood pressure measurement and risk of subarachnoid hemorrhage: the across study
Shiue I1,2, Arima H1,2 and Anderson C1,2
1University of Sydney, NSW, Australia; 2The George Institute for International Health, Sydney, NSW, Australia
Background: Case-control studies of the risks of subarachnoid hemorrhage (SAH) are complicated by imprecise estimates of pre-morbid exposures, best exemplified by usual blood pressure levels. We determined the relationship of frequency of blood pressure (BP) measurement and SAH in a population-based case-control setting.
Methods: In 436 incident SAH cases and 473 frequency-matched community SAH-free controls, data were collected on prior measurement of BP (3 months, 4–12 months, > 1 year, or never) and ever diagnosed and treated for hypertension. Effects of ever BP measurement, hypertension and antihypertensive treatment, on risks of SAH were estimated in logistic regression models with adjustment for potential confounding variables.
Results: Half of all subjects had their BP measured in the year prior to SAH (or interview for controls). Compared to ever measured, those without any prior BP measurement had a higher risk of SAH (OR = 2.06, 95%CI = 1.04–4.11). Whereas, those with a history of hypertension, on antihypertensive treatment, but not BP measured in the last 12 months, had a higher risk of SAH (OR = 1.87, 95%CI = 1.17–3.00) as those with a history of hypertension, without antihypertensive use, but no BP measurement in the last 12 months (OR = 4.56, 95%CI = 2.06–10.09).
Conclusions: Measurement of BP was a key determinant of the risk of SAH. Regular BP monitoring to improve hypertension detection and control is relevant to reducing rates of SAH.
O46
Stroke telemedicine for Arizona rural residents –STARR
Ingall TJ1, Aguilar MI1, Demaerschalk BM1, Dodick DW1, Vargas BV1, Bobrow BJ2, Kiernan T1, Channer D1, and Collins JG1
1Cerebrovascular Diseases Center, Mayo Clinic Hospital, Phoenix, AZ, USA; 2Department of Emergency Medicine, Maricopa Medical Center, Phoenix, AZ, USA
Background/Aims: Telemedicine is an efective tool for providing medical care to patients in remote locations. Acute stroke patients in rural Arizona previously had only a 2% chance of receiving acute stroke treatment, so Mayo Clinic Hospital implemented a stroke telemedicine program in October, 2008 to improve access to acute stroke care for Arizona's rural residents. A quality measures program has monitored the effectiveness of the program.
Methods: Between October 1, 2008 and March 31, 2010, data were collected prospectively on all telemedicine consults conducted as part of the STARR stroke registry study. Specifically, data on stroke treatment provided and technical issues occuring during the telemedicine consult have been collected on all patients.
Results: Through March 31, 2010, there were 246 stroke telemedicine page activations resulting in 216 consults with 52 patients (24%) receiving thrombolytic treatment. Technical issues, both minor and major, occurred in 63% of consults, resulting in 2% of consults being abandoned.
Discussion: The STARR project has shown that although there are logistical hurdles in implementing a stroke telemedicine system, stroke telemedicine increases access to acute stroke care for patients in remote locations. Technical problems during telemedicine consults occur frequently, but rarely cause consults to be abandoned.
O48
Factors associated with poor quality of life in 7-year survivors of stroke
Leach MJ1,2, Gall SL3,4, Dewey HM4,5,6, Macdonell RAL5,6, Thrift AG1,4,7
1Stroke Epidemiology, Baker IDI Heart and Diabetes Institute, VIC, Australia; 2School of Population Health, University of Melbourne, VIC, Australia; 3Menzies Research Institute, Hobart, TAS, Australia; 4National Stroke Research Institute, VIC, Australia; 5Department of Medicine, University of Melbourne, VIC, Australia; 6Austin Hospital, VIC, Australia; 7Department of Epidemiology & Preventive Medicine, Monash University, Melbourne, VIC, Australia
Background: Little is known about long-term outcome after stroke, particularly health-related quality of life (HRQoL).
Aim: We examined factors associated with HRQoL at 7 years after stroke.
Methods: All cases of first-ever stroke from a prospective community-based stroke incidence study (excluding subarachnoid haemorrhage) were assessed 7 years after stroke. HRQoL was measured with the Assessment of Quality of Life instrument (score range −0.04 to 1.0). Proportional odds logistic regression was used to determine factors associated with HRQoL at 7 years after stroke.
Results: In total, 1321 cases were recruited, 44% were male, and the mean age (±SD) was 76.1 ± 13.2 years. Seven years after stroke, 413 (31.2%) were alive and 328 were assessed (79.4%). Those assessed were less likely to smoke cigarettes immediately prior to stroke (P < 0.01). Seventy-six survivors (23%) had a very poor HRQoL (score ≤ 0.1). Factors present at 7 years that were independently associated with better HRQoL at 7 years after stroke were younger age (P = 0.026), lesser handicap (P < 0.0001), no functional impairment (P < 0.001) and male gender (P = 0.025).
Conclusion: At 7 years after stroke, a substantial proportion of survivors were suffering from poor HRQoL. Factors such as handicap and functional status could be targeted to improve HRQoL after stroke.
O53
The financial costs of admission for transient ischaemic attack (TIA) and ABCD2 scores 4, 5 in Wellington, New Zealand
Jolliffe E, McGonigal G, Ryder A, Ioannides N and Wong L
Capital and Coast District Health Board, Wellington, New Zealand
Background: Most international guidelines for TIA management recommend that people with TIA and an ABCD2 score of ≥ 4 be admitted to hospital. There is no evidence that this improves outcome. Hospital admission carries significant risk and most people would be admitted under general physicians.
Aim: To examine the cost of this approach, if followed in the Wellington TIA service.
Methods: In the Wellington region, high risk TIA patients and referrals with ABCD2 scores of 6 or 7 are admitted. All others are assessed as urgent outpatients (OP). The nursing and hotel costs for in-patient (IP) stay were calculated.
Results: In 16 weeks, 92 people were assessed. 36 (39%) of these had an ABCD score of 4 or 5 and were assessed as OPs. All were assessed urgently and underwent appropriate treatment and investigation. Assuming an average length of stay of 36 hours and a hotel cost of NZ $548, this equates to an annualised cost of $ 96 174.00.
Discussion: This is a cost under-estimate. Non-specialists tend to over-diagnose and over investigate TIA. An OP specialist service ensures specialist assessment and avoids complications associated with hospital admission. This data can be used to support the development of OP specialist TIA services.
O55
Atrial fibrillation is associated with increased early serious cardiac adverse events following ischemic stroke
Tu HTH1, Collins M2, Kalman JM3, Lees KR4, Lyden PD5, Donnan GA6 and Davis SM1 on behalf of the Virtual International Stroke Trials Archive VISTA Collaborators.
1Department of Neurology, Royal Melbourne Hospital, University of Melbourne, VIC, Australia; 2Department of Mathematics and Statistics, University of Melbourne, VIC, Australia; 3Department of Cardiology, Royal Melbourne Hospital, University of Melbourne, VIC, Australia; 4The Western Infirmary, University Department of Medicine and Therapeutics, Glasgow, UK; 5Cedars-Sinai Medical Center, California, USA; 6Florey Neuroscience Institutes, VIC, Australia
Background: Whether early cardiac complications contribute to worse outcomes in ischemic stroke patients with atrial fibrillation (AF) remains unclear. We aimed to establish whether AF is associated with increased early serious cardiac adverse events (SCAEs) after ischemic stroke, independent of baseline stroke severity and cardiovascular risk factors.
Methods: We searched VISTA-Acute, an academic database containing standardized data for 28,131 patients from randomised-controlled acute stroke trials and registries, for imaging-confirmed placebo-treated ischemic stroke patients with complete documentation of baseline demographics, cardiovascular risk factors, presence/absence of AF, neurological impairment (NIHSS), cardiac complications and 3-month outcome (mRS). Binary logistic regression was used to determine the effect of AF on SCAEs, a composite endpoint including acute coronary syndrome, symptomatic heart failure, cardiopulmonary arrest, VT/VF and cardiac mortality.
Results: In VISTA-Acute, 2865 patients met our selection criteria, of whom 819 had AF. Compared to patients without AF, SCAEs (14 vs 6%, P < 0.001), and cardiac mortality (5 vs 3%, P = 0.003) were more frequent in AF patients. AF was an independent predictor of SCAEs (OR2.16, 95%CI: 1.62–2.89) after adjusting for all baseline imbalances.
Conclusion: AF predicts early SCAEs after ischemic stroke. Besides anticoagulation for stroke prevention, physicians should consider more-intensive cardiac evaluation in stroke patients with AF.
O56
Not all stroke units are the same: early rehabilitation practices in Melbourne, Australia and Trondheim, Norway
Purvis T1,2, Cadilhac D2,3,4 and Bernhardt J3,2
1Physiotherapy Department, Austin Health, Melbourne, VIC, Australia; 2Department of Medicine, University of Melbourne, VIC, Australia;
3National Stroke Research Institute (member Florey Neurosciences Institutes), Melbourne, VIC, Australia; 4Public Health Research Cluster, Deakin University, Melbourne, VIC, Australia
Background: Early mobilisation (EM) is not routine practice in Australia but is recommended in clinical guidelines. We explored EM practices and staff perceptions at two acute stroke units (SU): Melbourne (Australia) and a ‘gold standard’ rehabilitation SU in Trondheim (Norway).
Methods: Mixed method design with: a) retrospective audit of 50 medical records/site, extracting clinical process indicators for stroke management including early rehabilitation, b) semi-structured interviews (n = 7) with clinicians. Transcribed interviews were confirmed by interviewees prior to thematic analysis. Both sources of evidence were used to draw conclusions.
Results: The cohorts had similar demographics and care pathways from emergency to SU. More patients were mobilised < 12 hours of admission in Trondheim (Trondheim 67%, Melbourne 24%, P < 0.001). Nurses were first to mobilise most patients at both sites. Patients were mobilised more often in Trondheim, irrespective of dependency (Trondheim 89% mobilised daily by nurses, 41% daily by physiotherapists; Melbourne 28% mobilised daily by nurses, 13% daily by physiotherapist/occupational therapist). The factors associated with EM included staffing levels, training opportunities, team work, and staff attitudes.
Conclusion: EM was different in the two acute SUs. A range of factors, not just staffing levels, contributed to differences. Understanding these factors could help improve acute rehabilitation practices.
O57
Contractures after stroke – incidence and prognostic factors
Kwah LK1, Diong J1, Harvey L2 and Herbert R1
1The George Institute for International Health, University of Sydney, NSW, Australia; 2Rehabilitation Studies Unit, University of Sydney, NSW, Australia
Background: Contractures are common after stroke. However, little is known about the precise prevalence or incidence, or factors predisposing patients to contractures.
Aims: To investigate the incidence of contractures in a representative stroke cohort, and to identify prognostic factors to determine patients at high risk of developing contractures.
Methods: Consecutive patients presenting to the accident and emergency department at St George Hospital with a diagnosis of stroke are screened. Patients are eligible if they are over 18 years old, understand English, have a medically documented stroke and are able to respond to basic commands. Outcome measures are range of motion in joints. Prognostic factors include age, pre-morbid function, severity of stroke, muscle strength, spasticity, motor function and pain levels. Data are collected within 4 weeks of stroke and at 6 months.
Results: To date, 200 patients have been recruited. 135 patients have been followed up at 6 months. There have been 20 deaths and five lost to follow-up. Data collection and analysis will be finished in July 2010.
Conclusion: The results of this study will provide the first accurate incidence of contractures after stroke, and assist clinicians in prioritizing therapy for patients at high risk of contracture.
O58
Circuit class therapy for improving mobility after stroke. a Cochrane review
English C and Hillier S
University of South Australia, Adelaide, SA, Australia
Background: There is strong evidence for intensive and task-oriented rehabilitation to reduce the impact of stroke on independent function. Circuit class therapy (CCT) offers a supervised group forum for people after stroke to practice tasks, enabling increased practise time without increasing staffing.
Aim: To examine the effectiveness and safety of CCT on mobility in adults with stroke.
Method: Cochrane systematic review. Randomised or quasi-randomised controlled trials involving adult stroke participants receiving CCT.
Results: In regards walking ability, CCT was superior to the comparison intervention (6 m Walk Test: MD, fixed 76.57 m, 95%CI 38.44 to 114.70, P < 0.0001; Gait speed: MD, fixed 0.12 m/s, 95% CI 0.00 to 0.24). CCT was also effective for improving balance (Step Test: MD, fixed 3.09 steps, 95%CI 0.31 to 5.86, P = 0.03; Activities-specific Balance and Confidence: MD, fixed 7.76, 95% CI 0.66 to 14.87, P = 0.03). Length of stay showed a significant effect in favour of CCT (MD, fixed −19.63 days, 95%CI −35.16 to −4.11, P = 0.01). Risk of bias was assessed as low across the studies.
Conclusion: CCT is safe and effective in improving mobility for people after moderate stroke, and may reduce inpatient length of stay. Further research is required that investigate the differential effects of stroke severity, latency and age.
O59
Use of a GPS device for measuring outings after stroke
McCluskey A1,2,3, Ada L3, Dean C3 and Vargas J3
1Community-Based Health Care Research Unit, Faculty of Health Sciences, The University of Sydney, NSW, Australia; 2The Royal Rehabilitation Centre Sydney, NSW, Australia; 3Faculty of Health Sciences, The University of Sydney, NSW, Australia
Background: Self-report diaries are often used to measure community participation. Yet diaries may produce unreliable data and are burdensome for participants. Although direct observation is the ‘gold standard’ this method is time-consuming and costly.
Aim: The aim of this study was to investigate the accuracy of a global positioning system (GPS) device and a diary for measuring daily outings by people with stroke compared to outings observed.
Methods: In this cohort study, 20 ambulant people with stroke carried a GPS device and kept a diary for 7 days, and were observed for half a day. GPS data were downloaded and analysed with Google maps. Percent exact agreement and close agreement were calculated.
Results: Mean time post-stroke was 39.0 months (SD 19.9). Participants went on few outings (range 0 to 2) in the 3.2 hours of observation. Percent exact agreement was 83% for GPS and 89% for diaries, compared to outings observed; 100% close agreement was achieved for both methods. GPS missed two observed outings.
Conclusion: Both GPS and diaries had acceptable levels of agreement. Differences in the methods and potential uses of GPS technology in rehabilitation research will be discussed.
O60
Effects of physiotherapy interventions based on the Bobath concept versus structuredtask practice for improving the ability to walk outdoors following stroke
Brock K1, Haase G2, Cotton S3 and Thompson M1
1St. Vincent's Hospital, Melbourne, VIC, Australia; 2Kliniken Schmieder, Konstanz, Germany; 3Orygen Youth Health Research Centre, Melbourne, VIC, Australia
Background: Regaining the ability to walk safely outdoors is an important goal of stroke rehabilitation.
Aim: To compare the short term effects of two physiotherapy approaches for improving ability to walk outdoors following stroke: interventions based on the Bobath concept compared to structured task practice.
Methods: A randomised controlled trial was conducted with twenty six participants between four and sixteen weeks post stroke, at a mobility level of walking with supervision indoors. Measures included a modified six minute walk test (including steps, ramps and uneven surfaces), gait speed, stair climbing and the Berg Balance Scale. Both groups received six one hour physiotherapy sessions, with one group receiving therapy based on the Bobath concept and the other group receiving structured task practice.
Results: There was a significant difference between groups for gait speed, with participants receiving interventions based on the Bobath concept, walking faster than those in the structured task practice group; f(1,23) = 6.83, P = .02, Cohen's d = .99. No other significant differences were recorded.
Conclusion: This study demonstrated short term benefit in favour of using interventions based on the Bobath concept for improving walking ability in people with stroke.
O61
The importance of participant initiated information during an observational study of psychosocial outcomes in stroke (POISE)
O'Reilly R1, Arblaster L1, Jan S1, Glozier N2, Lindley R1 and Hackett M1
1The George Institute for International Health, Sydney, NSW, Australia; 2The University of Sydney, Sydney, NSW, Australia
Background: The impact of stroke is significant for young (< 65 years) survivors. Structured interviews may not capture all aspects of the experience of illness and recovery. It may be relevant to utilise other methods of eliciting information from patients.
Methods: POISE is an observational study to determine what psychological, social and economic factors influence younger stroke survivors returning to paid and unpaid work. Participants are interviewed 4 times during a 12 month period. Many report important areas of concern in response to a final question ‘Would you like to explain how the stroke has affected your life?’
Results: A number of themes have already emerged from use of this open-ended question. Many participants express a need for information and support following hospital discharge. Other issues include fear of another stroke, loss of independence, loss of confidence, inappropriate Centrelink payments and minimal assistance to find suitable work. Seemingly minor disabilities can have major psychological and economic impacts. On a positive note many participants report they have improved their lifestyle and outlook since their stroke, and are highly motivated to recover.
Conclusions: These data indicate that using an open-ended question has highlighted key areas of unmet need and identified targets for intervention.
O62
Death, dependency, disability and health status of stroke patients 90-days post acute stroke unit care
Jammali-Blasi A1,2, McInnes E2, Markus R3, Faux S3, O'Loughlin G3, Dale S2 and Middleton S2
1New Investigator, Bachelor of Nursing (Honours) student, St Vincent's & Mater Health Sydney & Australian Catholic University Nursing Research Institute, Sydney, NSW, Australia; 2St Vincent's & Mater Health Sydney & Australian Catholic University Nursing Research Institute, Sydney, NSW, Australia; 3St Vincent's Hospital, Sydney, NSW, Australia
Background: The burden of stroke on individuals and carers both clinically and financially is well documented in existing literature. There is currently a gap in Australian studies examining stroke outcomes at 90 days.
Aims: To investigate outcomes of death, disability, dependency and health status at 90 days post-stroke.
Methods: Prospective cohort design. Data from 54 patients consecutively admitted to an acute stroke unit were analysed to identify associations between pre-morbid risk factors, demographics, clinical and stroke characteristics; and death, disability, dependency and health status post-stroke.
Results: Within 90 days, four participants had died and 45.5% were classified as dependent (Barthel Index score ≤ 94). 56.8% were classified as disabled (modified Rankin Score ≥ 2). Ischaemic strokes were experienced by the majority of participants when compared to haemorrhagic strokes (90.7% vs. 9.3%). The SF-36 mean scores indicated that overall the cohort had less than optimal physical health (mean = 46.7, SD = 9.8) and mental health (mean = 46.4, SD = 13.1). Patients with atrial fibrillation were found to have more severe strokes (P < 0.05). Univariate analyses showed that intracerebral haemorrhage and recurrent strokes were significantly associated with lower Barthel Index scores and lower SF-36 physical component scores at 90-days (P < 0.05).
Conclusion/Discussion: The results of the study contribute information on health status outcomes that may assist with delivering appropriate levels of post-hospital care and discharge planning.
O63
Ongoing exercise opportunities to prevent falls and enhance mobility in community dwellers after stroke: the stroke club trial
Dean C1, Rissel C2, Sharkey M3, Sherrington C1, Cumming R1, Lord S4, Kirkham C3, O'Rourke S4 and Barker R5
1The University of Sydney, NSW, Australia; 2Health Promotion Unit, Sydney South West Area Health Service, NSW, Australia; 3Stroke Recovery Association, NSW, Australia; 4Prince of Wales Medical Research Institute, NSW, Australia; 5James Cook University, NSW, Australia
Background: Falls and poor mobility are major contributors to stroke-related disability. There is now good evidence that exercise can enhance mobility after stroke, yet ongoing exercise opportunities are practically non-existent for stroke survivors.
Aims: This study funded by a NSW Health Promotion Demonstration Research Grant aimed to establish and evaluate community-based sustainable exercise programs utilising the NSW Stroke Recovery Association's Stroke Club network.
Methods: A prospective randomised controlled trial in which 151 community-dwelling stroke survivors were allocated to a lower limb exercise group or a upper limb exercise group. The lower limb group received a weekly exercise class and a home exercise program aimed at preventing falls, improving walking capacity and increasing physical activity. The upper limb group received a weekly exercise class and a home program aimed at management of the affected upper limb and improving cognition. The exercise classes occurred at Stroke clubs. Falls, falls risk, mobility and physical activity were measured before and after the 12 month intervention.
Results/Discussion: Over the three year study, recruitment was challenging however, exercise classes were established in 11 stroke clubs. Data collection has recently been completed and full results will be presented at the conference.
O64
Improving walking speed and capacity using treadmill walking with body weight support in subacute non-ambulatory stroke: the mobilise trial II
Dean C1, Ada L1, Bampton J1, Morris M2, Katrak P3 and Potts S3
1The University of Sydney, NSW, Australia; 2The University of Melbourne, VIC, Australia; 3Prince of Wales Hospital, VIC, Australia
Background: Walking is frequently impaired after stroke.
Aim: The aim of this study was to determine whether treadmill walking with weight support was more effective at improving walking than current intervention for non-ambulatory stroke patients.
Methods: A prospective, randomised trial of inpatient intervention with a 6-month follow-up with blinded assessment was conducted. One hundred and twenty six stroke patients were recruited and randomly allocated to an experimental group or a control group. The experimental group undertook up to 30 minutes of treadmill walking with weight support per day while the control group undertook up to 30 minutes of overground walking. At 6 months, the 10-m and 6-min Walk Tests and walking self-rating (0–10) were collected.
Results: At 6 months, there was no difference between the groups of independent walkers in terms of speed (MD 0.10 m/s, 95% CI −0.06 to 0.26). The experimental group walked 57 m further (95% CI 1 to 113) in the 6 min walk and rated their walking 1 point (95% CI 0.07 to 1.93) higher than the control group.
Conclusion: Treadmill training with body weight support results in better walking capacity and perception of walking compared to overground walking without deleterious effects on walking quality.
O65
Is there a place for electronic decision support in transient ischaemic attack and strokemanagement?
Ranta A
MidCentral Health, Palmerston North, New Zealand
Background: Accurate diagnosis and rapid intervention are critical in Transient Ischaemic Attack (TIA) and stroke management. First point of contact is often the general practitioner (GP) who sees only one to two such patients a month. Electronic decision support (EDS) was developed to aid GPs in managing patients according to evidence-based guidelines.
Aim: To compare diagnostic accuracy and management of GPs, stroke physicians and EDS.
Methods: Twenty-three GPs and stroke physicians assessed seven cases. The cases were entered into EDS by the author and EDS generated diagnoses and management was compared to clinician responses.
Results: Diagnosis and medical management was highly consistent and guideline adherent amongst stroke physicians. GPs made an accurate and complete diagnosis only 24% of the time and only 23% initiated best medical therapy when indicated. EDS consistently agreed with expert management, including 100% of TIA patients being started on best medical therapy at first point of contact, and provided more comprehensive advice as regards life style modifications and driving restrictions than either GPs or stroke physicians.
Conclusion: This study supports the notion that there is a role for electronic decision support in the GP setting and that such a tool is able to mimic expert advice.
O66
Stroke trends – what is happening in the hunter region, australia? Events, attack rates and case fatality from 1996 to 2008
Marsden D1,2, Spratt NJ1,2,3, Walker R1, Barker D2, Attia J1,2,3, Pollack M1,2,3, Parsons M1,2,3 and Levi C1,2,3
1Hunter New England Area Health Service, Newcastle, NSW, Australia; 2Hunter Medical Research Institute, Newcastle, NSW, Australia; 3University of Newcastle, Newcastle, NSW, Australia
Background: There is limited information on long-term trends in stroke incidence and case-fatality rates across Australia's eastern states. Public hospital acute stroke admissions for adults aged 20 years and above have been prospectively registered using consistent methods since 1996 in the Hunter Region.
Aim: To examine Hunter stroke trends 1996–2008: event numbers, attack rates and case fatality.
Method: Crude, age-standardised and age-specific stroke attack rates per 100 000 population and case fatality rates were calculated. Discharge coding accuracy was determined via a medical record audit.
Results: 9796 acute stroke events were registered among 8830 individuals. Crude and age-standardised attack rates decreased consistently from 1996 to 2008. Crude rates fell from 184 to 176 per 100 000 population and age-standardised rates from 129 to 106 per 100 000 (average reduction in risk of 0.85% per year, P = 0.027; and 2.38% per year, P < 0.001, respectively). Rates for females were 27.70% (P < 0.001) lower than for males. Age-specific rates fell for each age bracket between 45 to 74 years. Case fatality rates remained constant. There was 97.5% agreement between audit and coding.
Conclusion: This study demonstrates falling stroke attack rates but stable case fatality over 13 years in a mixed urban and rural population of NSW. It suggests benefits from regional stroke prevention strategies, however also indicates that stroke remains a major disease burden in this region.
O69
Emotionalism after stroke: what is it, is it a problem and what can we do about it?
Hackett M1, Yang M1, Anderson C1, Horrocks J2 and House A3
1The George Institute for International Health, University of Sydney, NSW, Australia; 2Psychiatry and Behavioural Sciences in Relation to Medicine, University of Leeds, United Kingdom; 3Leeds Institute of Health Sciences, University of Leeds, United Kingdom
Background: People often have difficulty controlling emotions after stroke (emotionalism). We assessed variables associated with emotionalism and conducted a systematic review to determine whether treatment reduces emotionalism.
Methods: Using data from the Auckland Regional Community Stroke Study (ARCOS) we conducted generalised logistic regression to determine baseline variables associated with early (28 days), late (6 months) and persistent (28 days and 6 months) emotionalism. We searched trials registers and online databases and contacted researchers and pharmaceutical companies to identify trials. The primary endpoint was the proportion of patients who had emotionalism at the end of treatment.
Results: Variables associated with emotionalism included younger age, having diabetes, requiring some assistance with activities, and not having a close friend to confide in. Large effects of treatment were seen in five trials with 213 participants showing: 50% reduction in emotionalism, diminished tearfulness, reduction in lability, tearfulness, and scores on the pathological laughter and crying scale. However, confidence intervals were wide. No differences were seen in adverse events between groups.
Conclusions: Emotionalism may be more frequent in the younger survivor who has pre-existing diabetes, requires assistance and does not have a close friend to confide in. Antidepressants can reduce the frequency and severity emotionalism.
O70
The sleep apnea cardiovascular endpoints (SAVE) study
Heeley E1, Antic N2, Anderson C1, Huang Y6, Huang S3, Wang J4, Zhong N5 and McEvoy D2
1The George Institute for International Health, University of Sydney, NSW, Australia; 2Adelaide Institute for Sleep Health, Repatriation General Hospital, Flinders University, Adelaide, SA, Australia; 3Respiratory Medicine, Ruijin Hospital, Shanghai, China; 4Shanghai Institute of Hypertension, Ruijin Hospital, Shanghai, China; 5Respiratory Medicine, The First Affiliated Hospital of Guangzhou Medical College, Guangzhou, China; 6Neurology, Peking First University Hospital, Beijing, China
Background: There is increasing evidence of a link between obstructive sleep apnea (OSA) and cardiovascular (CV) disease.
Aims: To conduct a hard CV endpoint RCT in OSA patients.
Methods: SAVE is an international, multicentre RCT of CPAP therapy plus standard care vs standard care alone in patients with CV disease and co-existing OSA. Other design parameters are: n = 5000; a 1-week run-in phase with sham CPAP to exclude those unable to accept CPAP; follow-up of 4 years; composite CV endpoint of sudden death, MI, stroke, unstable angina, TIA and heart failure.
Results: To date 58 sites have randomised 475 patients (85 Australia/NZ, 390 China). About 72% of patients had a baseline Epworth Sleepiness Score (ESS) ≤ 10. Average recruitment during the roll-out phase has been one patient/ site/month. CPAP adherence at 1, 3 and 6 months was 5.0 ± 0.12 (n = 154), 4.9 ± 0.14 (n = 117) and 4.7 ± 0.20 (n = 63) hours/night and was the same in sleepy (ESS > 10) and non sleepy patients (4.9 vs 5.0 hours/night).
Conclusions: There has been widespread acceptance of the trial by clinicians and patients in China and Australia/ NZ. CPAP adherence is high and patient drop out and cross-over rates are low.
O73
Carotid artery anatomy and geometry as risk factor for carotid atherosclerotic disease
Phan T1, Beare R1, Das G1, Ren M1, Chong W2, Jolley D2, Srikanth V1
1Stroke and Aging Research, Monash University, Department of Radiology, Monash Medical Centre, VIC, Australia; 2Department of Diagnostic Radiology, Monash Medical Centre, Melbourne, VIC, Australia
Background and aim: The traditional risk factors for carotid atherosclerotic disease do not account for the asymmetrical nature of carotid atherosclerosis and racial difference in site of atherosclerosis. The aim is to determine if carotid artery anatomy and geometry are independently associated with carotid atherosclerotic disease.
Method: This is a retrospective study of subjects who had CTA between 2006–2007. Demographic variables (age, sex) and traditional risk factors (hypertension, diabetes, hyperlipidaemia and history of smoking) were collected. Cluster logistic regression was used to account for the paired nature of carotid artery disease.
Results: The average age was 69 ± 14 years. The prevalence of vascular risk factors was: hypertension (70%), male (65%), hyperlipidaemia (55%), smoker (34%), ischemic heart disease (34%), stroke (29%) and Diabetes (25%). In the model, the following were found to be independently associated with carotid artery stenosis: radius of internal carotid artery (OR 0.20, 95% CI: 0.14–0.29), ICA angle (OR 1.05 per degree change, 95% CI: 1.04–1.07), age (OR 1.05 per year, 95% CI: 1.03–1.07), male sex (OR 1.72, 95%CI: 1.08–2.8) and smoker (OR 1.85, 95%CI: 1.15–2.96).
Conclusion: Carotid artery anatomy and geometry are independently associated with carotid artery stenosis.
O74
Exploring the experiences of patients, and their carers, who participate in inpatient stroke rehabilitation
Edgar B1 and Bilney B2
1Ballarat Health Services, Ballarat, VIC, Australia; 2Department of Health / Ballarat Health Services, VIC, Australia
Background: Person centred care is considered to be one of the six essential aims for the 21st century health care system. A core principle underlying person centred care is to understand and engage with consumers' experiences.
Aims: To develop an understanding of the experiences of patients, and their carers, who participate in inpatient stroke rehabilitation.
Methods: Semi-structured interviews were conducted with eight patients and eight carers within three months of discharge from an inpatient stroke rehabilitation program. Data were coded and analysed thematically.
Results: The effect of suffering a stroke was profoundly negative for the patients and their carers, and was associated with fear, loss of control and loss of dignity. The subsequent rehabilitation experience was diverse, with a range of both positive and negative experiences. Communication, engagement in rehabilitation processes, and interactions with other patients emerged as important factors which determined how participants recalled their overall experience of inpatient stroke rehabilitation.
Conclusion: The manner in which healthcare professionals structure inpatient rehabilitation processes, and communicate with patients and carers can have an effect on the overall experience of inpatient stroke rehabilitation. This study highlights the need for rehabilitation processes to be flexible to the individual needs of the patient and carer.
O75
Computational modelling of circle of Willis and Leptomeningeal anastomoses: application in studying cerebrovascular occlusion syndromes
Phan TG1, Hilton J2, Beare R1, Srikanth V1, and Sinnott M2
1Stroke and Aging Research, Monash University, Department of Radiology, Monash Medical Centre, VIC, Australia; 2CSIRO, Australia
Background: We developed a computational model of the cerebral circulation to evaluate the roles of the Circle of Willis (CoW) and leptomeningeal anastomoses (LA) in intracranial artery occlusion. This model was used to estimate the effective size of the LA under simulated conditions of Middle Cerebral Artery (MCA) occlusion.
Methods: A series of experiments were simulated in which successive branches of the intracranial arteries were occluded. The diameters of anastomoses within the same arterial territory and between different arterial territories were varied to estimate the size of the LA to maintain adequate flow.
Results: The model showed reduced flow in the territory of the occluded artery. The region of reduced flow became smaller and smaller as the site of occlusion was moved from the proximal large intracranial artery to the smaller distal intracranial artery. The CoW and LA work in tandem to provide flow to the MCA territory following MCA occlusion. Simulated arterial occlusion suggested that when the effective diameter of the LA was 1 mm, blood flow in the affected MCA territory dropped below 30% (the critical ischemic threshold).
Discussion: Computational modelling provides the ability to experimentally understand the effect of cerebral arterial occlusion on regional blood flow.
O76
The effects of silent cerebral infarcts on gait in the general population
Choi P1, Phan T1, Ren M1, Beare R1, Ly J1 and Srikanth V1,2
1Stroke and Ageing Research Group, Neurosciences, Department of Medicine, Southern Clinical School, Monash Medical Centre, Monash University, Melbourne, VIC, Australia; 2Menzies Research Institute, Hobart, TAS, Australia
Background: Data on the effects of silent infarcts (SI) on gait are lacking.
Aim: To study the associations of SI with gait in a random population-based sample of people aged ≥60 in the Tasmanian Study of Cognition and Gait.
Methods: SI were identified on brain magnetic resonance images (MRI) by expert consensus as lesions >3 mm in size in the absence of a history of stroke. Gait was measured using a computerised walkway and factor analysis used to generate a summary gait score. Linear regression was used to examine associations between SI and the gait score.
Results: SI were detected in 43/342 participants. Mean gait score was lower in those with SI (0.44, SD 1.32) than those without SI (0.20, SD 0.80), P < 0.001. In multivariable regression adjusting for age, sex, and brain volume, there was a significant interaction between SI and white matter lesion volume (WMLv), β 0.02, P = 0.01 such that the presence of SI magnified the adverse effect of WMLv on gait.
Conclusion: There is an additive adverse effect for SI and WMLv on gait in the general older population, highlighting the need to study ways to minimize their occurrence.
O79
Exploring post-stroke mood changes in community-dwelling stroke survivors: a longitudinal cohort study
White J, Magin P, Attia J, Sturm J, Carter G, McElduff P and Pollack M
Hunter Stroke Service, NSW, Australia
Background: There are few longitudinal studies exploring the patterns of psychological morbidity and factors contributing to change over time. The aim of this study is to explore the predictors of anxiety and PSD and factors contributing to change in symptoms in stroke survivors over a 12 month period.
Method: This study is a prospective cohort study of 120 stroke survivors interviewed at baseline (stroke onset), 3, 6, 9, and 12 months. Outcome is assessed with measures of physical function (MRS, BI), depression and anxiety (HADS) and psycho-social function (K10, AQOL, social support, community participation).
Results: Data has been collected on 102 participants. Over time anxiety reduced (P = 0.001) and depression (HADS) and distress (K10) remained unchanged. Increased disability (MRS) was associated with higher anxiety (HADS) (P = 0.0405), depression (HADS) (P = 0.0015), distress (K10) (P = 0.0728) and reduced QOL (P ≤ 0.0001) over time
Discussion: Anxiety is heightened at baseline and reduces over time. Clinicians should be mindful of this when using screening measures that have an anxiety and depression sub-components. Increased disability impacts QOL over time and is most notable following discharge from hospital. These results support the need for greater availability of longer term community services to monitor for PSD and facilitate community participation.
O80
Relationship between haematoma growth and outcome after intracerebral haemorrhage: secondary analysis from INTERACT1
Delcourt C1, Arima H1, Anderson C1, Lindley R1, Heeley E1, Wang J2, Huang Y3, for the INTERACT2 Investigators
1The George Institute for International Health, Royal Prince Alfred Hospital, University of Sydney, NSW, Australia; 2Shanghai Institute for Hypertension, Shanghai, China; 3Peking University First Hospital, Beijing, China
Background: Haematoma volume is a key predictor of outcome in intracerebral haemorrhage (ICH).
Aim: INTERACT1 data were used to quantify associations of haematoma growth on the death and dependency outcome cluster.
Method: INTERACT1 (n = 404) ICH patients (<6 hours of onset) with elevated systolic BP (150–220 mmHg) and no definite indication/contraindication to treatment were randomly assigned to rapid lowering or standard management of blood pressure (BP). The primary efficacy was change in haematoma volume at 24 hours. Associations of absolute and relative haematoma growth, and death and dependency at 90 days, were assessed.
Results: In patients (n = 386) with available baseline and 24 hours CT scans, absolute and relative haematoma growth were strongly associated with poor outcome. The associations remained significant after adjustment with age, sex, antithrombotic use, NIHSS ≥ 14, baseline haematoma volume, location and intraventricular extension, and study treatment: OR 1.76 (95%CI 1.24–2.52) per 1SD in absolute change and OR 1.72 (95% CI 1.28–2.33) per 1SD in relative change in haematoma growth.
Conclusion: Haematoma growth is a strong predictor of outcome in ICH. Treatment strategies to control haematoma growth should improve recovery of patients with ICH.
O81
Return to driving after stroke: a qualitative study
McCluskey A1 and White J2
1Community Based HealthCare Research Unit, Faculty of Health Sciences, University of Sydney and Royal Rehabilitation Centre, Sydney, NSW, Australia; 2Hunter Stroke Service, NSW, Australia
Background: About two-thirds of stroke survivors do not attempt a driving assessment nor return to driving. Loss of this important role significantly affects community participation.
Aim: To explore attitudes and experiences of stroke survivors who return to driving or cease driving.
Methods: A qualitative study design was employed. A sample of 38 stroke survivors were recruited within 12 months of stroke from two metropolitan cities. The majority of participants were interviewed at least twice. Data were analysed for categories and themes using ground theory methods.
Results: About 32% of participants resumed driving within 6 months following stroke. 45% had resumed by 12 months. Participants received inconsistent advice regarding return to driving and driving legislation. Reluctance to attempt driving tests was common due to fear of failure and the financial burden of the test fee. Lack of confidence and limited professional support influenced the decision to return to driving. Ongoing driving was characterised by self imposed restrictions and outings.
Conclusion: Stories from these stroke survivors suggest a need for improved information about return to driving and support to resume and maintain driving skills. Findings suggest a need for proactive rehabilitation programs addressing community participation and return to driving early after stroke.
O85
Circuit class therapy for rehabilitation after stroke. Protocol of the CIRCIT trial
English C1, Hillier S1, Crotty M2, Segal L1, Bernhardt J3 and Esterman A1
1University of South Australia, Adelaide, SA, Australia; 2Flinders University, Adelaide, SA, Australia; 3National Stroke Research Institute, Melbourne, VIC, Australia
Background: There is strong evidence for a dose-response relationship between physical therapy in the first 6 months after stroke and recovery of function. The optimal method of maximising therapy time within finite health care resources is not known.
Aims: To determine the effectiveness and cost effectiveness of two alternative models of physical therapy service delivery (7-day week therapy or group circuit class therapy [CCT] 5 days a week) compared to standard care for people receiving rehabilitation after stroke.
Method: A multicentre, 3-armed randomised controlled trial with blinded assessment of outcomes. 282 people admitted to rehabilitation facilities after stroke will be randomised to receive either; standard care therapy 5 days a week, standard care therapy 7 days a week or CCT 5 days a week. The primary outcome measure is the six-minute walk test at 4 weeks. Economic analysis will include length of hospital stay and resource costs related to therapists' time and equipment use. Secondary outcome measures include functional walking ability, ability to perform activities of daily living, arm function, quality of life and participant satisfaction.
Conclusion: This trial will provide evidence of the relative effectiveness and cost-effectiveness of 7-day week therapy and CCT for people receiving rehabilitation after stroke.
O86
Loss of muscle mass after stroke. A systematic review
English C1, McLennan H1, Thoirs K1, Coates A1 and Bernhardt J2
1University of South Australia, Adelaide, SA, Australia; 2National Stroke Research Institute, Melbourne, VIC, Australia
Background: Loss of muscle mass after stroke has potentially devastating health consequences. It impacts on function as well as the body's ability to break down glucose which may lead to an increased risk of recurrent stroke.
Aim: To determine the current level of evidence about the rate and magnitude of loss of muscle mass after stroke.
Method: Systematic review and (where appropriate) meta-analyses. Studies involving direct measurement of muscle size on adult humans after stroke were included.
Results: Fourteen studies from nine countries were included. Nine studies compared muscle mass in the paretic and non-paretic limbs in participants later after stroke. Meta-analyses showed significantly less muscle in the paretic leg (342 g, 95%CI 247–437 g) and in the paretic arm (281 g, 95% CI 213–350 g). The five longitudinal studies varied in participant characteristics and timing of assessments, but change over time is likely related to stroke severity.
Discussion: There is strong evidence that people later after stroke have less muscle mass in their paretic limbs. Little is known about the rate and timing of muscle loss or the relationship between muscle loss, metabolic dysfunction and recurrent stroke. Early exercise has the potential to slow the rate of muscle loss.
O87
Assessing cognitive function in acute stroke
Cumming T1, Brodtmann A1, Darby D2 and Bernhardt J1
1National Stroke Research Institute, Melbourne, VIC, Australia; 2University of Melbourne, Melbourne, VIC, Australia
Background: Assessing cognitive function in acute stroke patients is important for identifying deficits and allows tracking of subsequent improvement. Assessment, however, is challenging due to heterogeneity in arousal state and focal neurological deficits, and also premorbid factors including education and language. We report preliminary experience with a reliable, brief, language-neutral and engaging computerised cognitive battery.
Aim: To evaluate usability of the CogState battery within two weeks of acute stroke.
Methods: Patients presenting to the Austin Hospital with confirmed stroke underwent 4 computerised CogState screening tasks assessing psychomotor processing, visual attention, working memory and executive ability.
Results: By April 2010, five patients (64–86 years; four infarct/one haemorrhage; three left-sided lesion/two right-sided lesion) were assessed between 1 and 12 days post-stroke. Testing was well tolerated, though 1 patient could not perform the executive task (which required maze learning). Speed and accuracy data varied markedly from chance to high normal scores and appeared to accord with other clinical parameters of severity. Further patients are currently being recruited.
Conclusion: Early testing indicates that this novel computerized battery is promising as a tool to track performance in acute stroke. Correlation with stroke severity and detailed neuropsychological assessment will provide important information for future clinical utilisation.
O88
The ABCD2 score and transient ischaemic attack (TIA) – the area under the curve (AUC) is not the whole story
Srikanth V1, Sanders L1, Phan T1 and Blacker D2
1Stroke and Ageing Research Group, Southern Clinical School, Department of Medicine, Monash Medical Centre, Monash University and Southern Health, Melbourne, VIC, Australia; 2Department of Neurology, Sir Charles Gairdner Hospital, Nedlands, WA, Australia
Background: The ABCD2 score is recommended in Australian Guidelines for decision-making regarding urgency of tests or admission for TIA patients. A recent systematic review (Stroke, 2010. 41:667–673) reported a pooled AUC of 0.72 for the score in predicting stroke at 7 days, concluding a ‘good’ predictive value. However, an AUC is only a pre-test estimate of accuracy, and does not assist in clinical prediction for an individual patient.
Aim: To derive post-test measures from the above review, namely positive predictive value (PPV) and positive likelihood ratio (PLR), which are more informative in the clinical setting.
Methods: Data were extracted to assist in the generation of PPVs and PLRs from individual studies in the review for a cut-off ABCD2 score >4 for ‘high’ risk of stroke recommended in Australian Guidelines.
Results: Data were available in 16/20 studies. For ABCD2 cut-off >4, the PPVs and PLRs for stroke at 7 days were poor, ranging from 7.9%–25.5%, and 0.67–1.61 respectively. Results were similar for other cut-off scores.
Conclusions: Given such low estimates of PPV and PLR, the clinical predictive value of the ABCD2 score is highly questionable. The costs of misclassification include delays to key investigations and interventions, or inappropriate use of hospital-beds.
O93
Finding a valid screening tool for cognitive impairment after stroke
Cumming T
National Stroke Research Institute, Melbourne, VIC, Australia
Background: There is a need for cognitive assessment tools that are feasible to administer and have acceptable sensitivity and specificity in stroke populations. The Mini-Mental State Examination (MMSE), while widely used, has well-documented shortcomings in stroke.
Aim: To determine whether the Montreal Cognitive Assessment (MoCA) or the Rowland Universal Dementia Assessment Scale (RUDAS) have better validity than the MMSE in patients with stroke.
Methods: Patients with confirmed stroke undertook two testing sessions at 3 months post-stroke. Scores on the three screening tools (MMSE, MoCA, RUDAS) were validated against a ‘gold standard’ classification of cognitive impairment, derived from an extended neuropsychological battery.
Results: By April 2010, 34 patients (mean age = 70.7) had been tested and 22 (65%) were classified as cognitively impaired. Mean scores were lowest on MoCA (20.5/30), followed by MMSE (23.4/30), then RUDAS (24.7/30). The MoCA was the best predictor of cognitive impairment, with an area under the curve of 0.93, followed by the MMSE (AUC = 0.87) and the RUDAS (AUC = 0.76).
Conclusion: The MoCA – scored out of 30 and taking approximately 10 minutes, like the MMSE – appears to be a better cognitive screening tool than the MMSE in stroke.
O95
A population based study of thrombolysis for acute stroke in South Australia
Leyden J, Chong K, Kleinig T, Lee A, Field J and Jannes J
Queen Elizabeth Hospital, Adelaide, SA, Australia
Background: Thrombolysis has been proven to be a safe effective treatment for acute ischaemic stroke in clinical trials and voluntary registries. The rates of thrombolysis and subsequent haemorrhage are unknown in the broader community.
Aim: To report the rate and complications of thrombolysis for acute stroke in South Australia over a 2 year period from October 2007 to October 2009.
Methods: Observational Population based retrospective review of case notes and imaging, using multiple case ascertainment methods in all hospitals, public, private urban and rural in South Australia covering a population of 1.5 million people over two years. Contraindications for thrombolysis administration were sought according to the 2007 National Stroke Foundation Guidelines. Incidence of haemorrhage and misdiagnosis were recorded Population thrombolysis rates were calculated according to distance from an acute stroke unit.
Results: A total of 158 instances of thrombolytic therapy for suspected acute ischaemic stroke were identified. 15 patients (9.5%) had symptomatic intracerebral haemorrhage of which eight (5.1%) died as a consequence. five cases (3%) were administered thrombolysis meeting absolute exclusion criteria. People living nearer stroke units were more likely to receive thrombolysis.
Conclusions: Rates of symptomatic haemorrhage after thrombolysis were not significantly different from voluntary registries. It is estimated that less than 2% of acute strokes are treated with thrombolysis in South Australia.
O96
Does a multi-factorial falls prevention program reduce falls in people with stroke returning home after rehabilitation? A randomised controlled trial
Batchelor F1,2, Hill K1,3, Mackintosh S4, Said C2,5 and Whitehead C6
1National Ageing Research Institute, Parkville, VIC, Australia; 2The University of Melbourne, Melbourne, VIC, Australia; 3LaTrobe University & Northern Health, Bundoora, VIC, Australia; 4University of South Australia, Adelaide, SA, Australia; 5Austin Health, Heidelberg, VIC, Australia; 6Flinders University, Adelaide, SA, Australia
Background: Up to 75% of people with stroke fall in the year after hospital discharge. To date, there have been no published randomised controlled trials (RCTs) evaluating multi-factorial falls prevention programs for this high risk group.
Aims: To evaluate the effectiveness of a multi-factorial falls prevention program for people with stroke.
Methods: People with stroke, returning home and at high risk of falls (n = 156) were randomised to receive usual care or targeted falls prevention strategies incorporating home exercise, education and referral. Outcomes were falls rate and proportion of fallers, monitored prospectively for one year.
Results: Falls rate in the intervention group was 1.89 falls/person-year and in the control group 1.76 falls/person-year, with no significant difference between the two groups (negative binomial regression, IRR = 1.10, 95%CI: 0.63–1.90). The proportion of fallers in the intervention group was lower but this was not significant.
Conclusion: This study is of one of the first large RCTs evaluating an intervention specifically aimed at reducing falls in people with stroke. Falls prevention strategies that are effective in the community-dwelling older population were not effective in this population. Findings from this study will guide future research into appropriate interventions.
O97
Stroke continuity nurse led clinic – fundamental to comprehensive post stroke management
Shum E and Hui V
Acute Stroke Unit, Prince of Wales Hospital, Shatin, Hong Kong
Background: Post stroke management required on-going review to match the patient's need. Moreover, the synergy of multidisciplinary collaborations would not appear in the absence of good coordination. Service gap was sometimes the result of defective communication amongst the care providers.
Aim: The aim of the study was to check if stroke continuity nurse led clinic could bridge the service gap effectively
Method: Follow up minor stroke patients in 3 weeks, 3 months and 6 months intervals. Check the amount of essential care missed. Remedy and determined the percentage of these items eventually rectified by nurse clinic.
Result: Seventy patients were recruited in 6 months. Eleven items of missed care were identified. They were lack of referrals to life style modifications such as smoking cessation, weight reduction, appropriate diet and exercise. Insufficient patient empowerment programs for disease management. Three patients ran out of medications and required immediate refilled. Many patients forgot their medical follow up schedule. 80–100% of these items were rectified in the clinic
Conclusion: The service gaps identified are critical for secondary stroke prevention and enhancement of patient's quality of life. The study intends to recruit mild disable stroke patient, the result can be more alarming if not. This is because demand for care is directly proportion to disability. Stroke continuity nurse led clinic proved to be effective in bridging thee service gaps.
O99
Stroke patients do not need to be inactive in the first 2 weeks after stroke: results from a stroke unit focused on early rehabilitation
Askim T1, Bernhardt J3 and Indredavik B1,2
1Department of Neuroscience, Norwegian University of Science and Technology, Norway; 2Stroke Unit, Department of Medicine, Trondheim University Hospital, Trondheim, Norway; 3National Stroke Research Institute, Heidelberg, VIC, Australia
Background: Stroke patients have been found to be ‘inactive and alone’, even though rehabilitation is an important aspect of acute stroke unit care.
Aim: This study aimed to determine the activity levels of an unselected group of acute stroke patients treated in a stoke unit focused on early, active rehabilitation.
Methods: Patients admitted to the Stroke Unit at Trondheim University Hospital, were observed (1–14 days post stroke) using a standardised method of observation of motor activity at 10-minute intervals from 8:00 AM to 5:00 PM over a single day.
Results: One hundred and seventeen patients (56.4% male) were included. Mean(SD) age; 78.7(9.2), days from onset; 6.7(3.7) and NIHSS score; 8.2(7.0). 42.7% had moderate to severe stroke (NIHSS > 7). The patients spent 30.1% of the time in bed, 46.3% of the time sitting out of bed and 19.8% of the time in higher motor activities (transferring, standing or walking). Subgroup analysis showed that patients with mild, moderate and severe stroke spent 79.3%, 60.0% and 31.1% of the time out of bed or engaged in higher motor activity, respectively.
Conclusion: This study shows that it is possible for acute stroke patients to spend most of the daytime out of bed and engaged in higher motor activity.
O100
Results from the carotid revascularization endarterectomy vs. stenting trial (CREST)
Howard V1, Cohen S2, Voeks J1, Roubin G3, Moore W4, Lal B5, Meschia J6, Sheffet A7, Howard G1 and Brott T6 for the CREST Investigators
1University of Alabama, Birmingham, Alabama, USA; 2Sunrise Hospital and Medical Center, Las Vegas, Nevada, USA; 3Lenox Hill Hospital, New York, USA; 4University of California, Los Angeles, Californai, USA; 5University of Maryland Medical Center, Baltimore, Maryland, USA; 6Mayo Clinic, Jacksonville, Florida, USA; 7University of Medicine and Dentistry of New Jersey-New Jersey Medical School, Newark, New Jersey, USA
Background: Carotid artery stenting (CAS) and carotid endarterectomy (CEA) are options for treating carotid artery stenosis.
Aims: We compared CAS versus CEA in patients with symptomatic (≥50%) or asymptomatic (≥60%) extracranial carotid stenosis.
Methods: Subjects were randomized to CAS or CEA. The composite endpoint was any stroke, myocardial infarction (MI), or death during the peri-procedural period or any ipsilateral stroke up to 4 years.
Results: For 2,502 subjects, there was no difference in the 4-year rate of endpoints (7.2 vs. 6.8%; HR = 1.11; P = 0.51). There was no differential treatment effect by symptomatic status (P = 0.84) or sex (P = 0.34); however, we observed a differential effect by age (P = 0.02) with CAS superior at young ages, but CEA superior at older. Peri-procedurally, there were treatment differences in the endpoint components with higher stroke rates in CAS (4.1 versus 2.3%; P = 0.012), but higher MI rates for CEA (1.1 vs. 2.3%, P = 0.032). After the peri-procedural period, the incidence of ipsilateral stroke was low for both treatments (2.0 vs. 2.4%, P = 0.85).
Conclusions: For patients with symptomatic or asymptomatic carotid stenosis, the primary outcome did not differ between CAS and CEA; peri-procedurally there was higher stroke risk with CAS and higher MI risk with CEA.
O102
Lower treatment blood pressure is associated with greatest reduction in haematoma growth after acute intracerebral haemorrhage: the interact trial
Arima H1, Anderson C1, Wang JG2, Huang Y3, Heeley E1, Delcourt C1 and Chalmers J1 for INTERACT Investigators
1The George Institute for International Health, University of Sydney, Sydney, NSW, Australia; 2Centre of Epidemiological Studies and Clinical Trials, Riu Jin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China; 3Peking University First Hospital, Beijing, China
Background: The pilot phase of the INTERACT trial showed that rapid blood pressure (BP) lowering can attenuate hematoma growth in acute intracerebral hemorrhage (ICH).
Aims: To define the systolic BP level associated with greatest attenuation of hematoma growth.
Methods: INTERACT included 404 patients with CT-confirmed ICH and elevated systolic BP (150–220 mmHg). CT was done at baseline and at 24 hours using standardized techniques, with digital images analyzed centrally. Associations of baseline and achieved on-treatment (mean during the first 24 hours) systolic BP levels with increase in hematoma volume were explored.
Results: There were 346 patients with duplicate CT scans. There was no significant association between baseline systolic BP levels and either the absolute or proportional growth in hematoma volume (P = 0.26 and 0.12 for trend, respectively). By contrast, achieved on-treatment systolic BP levels in the first 24 hours were clearly associated with both absolute and proportional hematoma growth (both P = 0.03 for trend). Maximum reduction in hematoma growth was obtained among the one third of participants who achieved the lowest on-treatment systolic BP levels (median 135 mmHg).
Conclusion: Intensive BP reduction to lower levels between 130 and 140 mmHg during the first 24 hours after ICH is likely to provide the maximum protection against hematoma growth.
O103
Changes in organisation of Australian stroke care, the past 10 years
Cadilhac DA2,3, Harris D1, Kilkenny M2,3, Ritchie E1, Price C1 and Lalor E1 On behalf of the National Stroke Foundation Advisory Committee: National Stroke Audit Acute Services and the National Stroke Audit Collaborative
1National Stroke Foundation, Melbourne, VIC, Australia; 2National Stroke Research Institute of Florey Neuroscience Institutes, Melbourne, VIC, Australia; 3University of Melbourne, Melbourne, VIC, Australia
Background: Over the past ten years, significant advances in health care have occurred. In 1999, the National Stroke Foundation carried out the first national survey of organisational services for stroke care. A similar survey was undertaken in 2004 prior to establishment, in 2007, of regular biennial surveys of the organisation of acute stroke services.
Aim: To describe and compare the organisational resources available for stroke, such as access to stroke units, among Australian hospitals between 1999 and 2009.
Methods: Cross-sectional surveys (mainly public hospitals) covering respondent details, hospital characteristics e.g. range of services, emergency department care, clinical management of patients, stroke team configuration, use of clinical tools and access to rehabilitation. Categorical variables were analysed using the χ2 test.
Results: Response rates improved from 79% in 1999 to 92% in 2009. Fewer doctors completed the most recent surveys (1999 71% vs. 2009 10%). An absolute increase in dedicated Stroke Units (35 in 1999 to 68 in 2009); outpatient clinics for transient ischeamic attack (1999 5% vs. 19% 2009), and same day CT scanning (1999 50% vs. 82% 2009) was found.
Discussion: Significant improvements in stroke services have occurred since 1999. These results are descriptive since the sampling strategies and questionnaires differed.
O106
Improving organisational performance by measuring what matters
Harris D1, Cadilhac DA2,3, Kilkenny M2,3, Price C1, Ritchie E1 and Lalor E1
National Stroke Foundation, Melbourne, VIC, Australia Institutions:
1National Stroke Foundation, Melbourne, VIC, Australia; 2National Stroke Research Institute of Florey Neuroscience Institutes, Melbourne, VIC, Australia; 3University of Melbourne, Melbourne, VIC, Australia
Background: Quality of stroke services is affected by availability of hospital-based organisational resources, including stroke units (SUs) and multidisciplinary clinicians. The UK Sentinel Audit has used an organisational scoring system to monitor and plan improvements in stroke care. A similar approach may be useful for Australia.
Aim: To develop an Australian organisational stroke services ‘quality’ score for hospitals.
Methods: National Stroke Audit 2009 Organisational Survey responses were grouped into four domains: Hyper-acute; Acute; Multidisciplinary team; Discharge processes and communication. Each question (min 8, max 16 per domain) was assigned a grade (range 0–2), based on the Acute Stroke Services Framework 2008 recommended standards. A weighted score out of 100 was estimated for Category A (n = 48) and B hospitals (n = 38).
Results: Organisational scores plausibly discriminated between hospitals. The median score was 64 (inter-quartile range [IQR] 48–78). Scores for hospitals with a SU (71 IQR 58–80) were higher than hospitals without SUs (40 IQR 35–45). Category A hospital scores (72 IQR 61–81) were greater than Category B hospitals (52 IQR 40–64). Scores for urban (median 64) and rural (median 67) hospitals were similar.
Conclusion: An Australian organisational score may be useful for distinguishing high quality services and focusing improvements in stroke care.
O109
Mild cognitive impairment after stroke is associated with activity participation in an Australian cohort
Spitzer J1,2, Tse T1,2, Baum C3 and Carey L1,2
1National Stroke Research Institute, Division of Neurorehabilitation and Recovery, VIC, Australia; 2La Trobe University, School of Occupational Therapy, Melbourne, VIC, Australia; 3Washington University School of Medicine, St. Louis, Missouri, USA
Background: The onset of long-term disability following stroke can severely impact activity participation and quality of life.
Aims: To investigate the association between cognition, mood and activity participation in stroke survivors living in the community.
Methods: Participation in household, social/educational and leisure activities were quantified using the recently modified Activity Card Sort Australia (ACS-Aus)
1
. Cognition was measured using the Montreal Cognitive Assessment (MoCA) and mood using the Centre for Epidemiological Studies Depression Scale (CES-D). Non-parametric correlation analyses quantified the presence and strength of association between variables. Differences for those with mild cognitive impairment or with depressive symptoms were investigated.
Results: Thirty stroke survivors participated. Those with mild cognitive impairment had significantly reduced participation in all activity domains: total participation (U = 7.00; P = 0.001), high-demand leisure (U = 12.00; P = 0.002), social/educational activities (U = 25.50; P = 0.013); low-demand leisure (U = 21.00; P = 0.007) and household activities (U = 25.50; P = 0.015). Significant differences in activity participation were not found with mood, although relatively few were identified as being depressed.
Conclusion: Using a quantitative measure of participation we identified mild cognitive impairment as important for occupational therapists to consider when setting participation goals post-stroke.
O110
Clinical feasibility and utility of multimodal CT for acute ischemic stroke in a tertiary hospital
Campbell B1, Weir L1, Desmond P2, Hand P1, Yan B1, Parsons M3 and Davis S1
1Department of Neurology, Royal Melbourne Hospital, University of Melbourne, Parkville, VIC, Australia; 2Department of Radiology, Royal Melbourne Hospital, University of Melbourne, Parkville, VIC, Australia; 3Department of Neurology and Hunter Medical Research Institute, John Hunter Hospital, Newcastle, NSW, Australia
Background: Multimodal CT (mmCT) including noncontrast, CT perfusion (CTP), CT angiography (CTA) is technically possible on most current CT scanners. It has potential to inform treatment decisions in acute stroke.
Aim: To report experience with introducing mmCT at Royal Melbourne Hospital.
Methods: mmCT was performed routinely (Siemens 16 slice, 2 × 24 mm CTP, arch-vertex CTA) on all acute ischemic stroke patients presenting to RMH within 9 hours unless renal dysfunction was known or suspected (diabetics without recent eGFR). CTP was omitted for posterior circulation syndromes. Clinical data and discharge diagnosis were prospectively recorded.
Results: Between January 2009 and April 2010, 209 patients presented <9 hours from ischemic stroke onset. mmCT was performed in 127 (61%; exclusions: 21 known eGFR < 60 mL/min, 13 diabetic and unknown eGFR at time of scan, 12 posterior circulation syndromes, 36 clinician decision not specified). Dual-slab CTP was technically adequate in 102 (80%; five bolus failures, nine bolus mistimed, 11 excessive movement). Non-contrast CT showed evidence of stroke in 45/102 (44%) with additional diagnostic yield of 44/102 (43%) using CTP. CTP was falsely normal despite technically adequate scans in 12.7% [lacunar (four) and reperfused (nine) strokes].
Conclusions: mmCT significantly increases diagnostic confidence in acute stroke and is feasible in routine clinical practice.
O112
Diffusion lesion reversal in acute ischemic stroke is uncommon in the 3–6hr treatment window
Campbell B1,2, Purushotham A3, Christensen S2, Desmond P2, Nagakane Y4, Parsons M5, Lansberg M3, Mlynash M3, Straka M3, De Silva D6, Olivot JM3, Bammer R3, Albers G3, Donnan G4 and Davis S1 for the EPITHET-DEFUSE Investigators.
1Departments of Medicine and Neurology, The Royal Melbourne Hospital, University of Melbourne, Parkville, VIC, Australia; 2Department of Radiology, The Royal Melbourne Hospital, University of Melbourne, Parkville, VIC, Australia; 3Departments of Neurology and Neurological Sciences and the Stanford Stroke Center, Stanford University Medical Center, Stanford, California, USA; 4Florey Neuroscience Institutes, The University of Melbourne, Parkville, VIC, Australia; 5Department of Neurology and Hunter Medical Research Institute, John Hunter Hospital, University of Newcastle, NSW, Australia; 6Singapore General Hospital Campus, National Neuroscience Institute, Singapore
Background: The accuracy of perfusion-diffusion mismatch in estimating ischemic penumbra is challenged by reports of diffusion lesion reversal (DLR).
Aim: To investigate the frequency and implications for mismatch classification of DLR using imaging from the EPITHET and DEFUSE studies.
Methods: Follow-up T2 images were co-registered to acute diffusion and the lesions manually outlined in diffusion space. DLR was defined as voxels of acute diffusion lesion that corresponded to normal brain at follow-up (i.e. final infarct, leukoaraiosis and CSF voxels were excluded from consideration). The appearance of DLR was visually checked for artifact, the volume calculated and the impact of adjusting baseline diffusion lesion volume for DLR volume on perfusion-diffusion mismatch analysed.
Results: In 119 patients available for analysis (83 treated with IV tPA), median DLR volume reduced from 4.4 mL to 1.5 mL after excluding CSF/leukoaraiosis. There were 4/119 (3.4%) with DLR > 10 mL and >10% of baseline diffusion volume however three cases were atrophy-related. Visual inspection verified 7/119 (5.9%) with true DLR, median volume 3.2 mL. Subtracting DLR from acute diffusion volume altered perfusion-diffusion mismatch (Tmax > 6sec, ratio 1.2) in 3/119 (2.5%) patients.
Conclusions: DLR is uncommon and rarely alters perfusion-diffusion mismatch. The acute diffusion lesion is generally an accurate reflection of infarct core.
O114
Trials the tribulations and the truths: recruiting hospitals to a very early rehabilitation trial (AVERT)
Ellery F, Borschmann K and Bernhardt J
National Stroke Research Institute, Melbourne, VIC, Australia
Background: While recruitment of participants to trials is often discussed, there is little attention given to the time and costs associated with the selection of recruiting hospitals to clinical trials.
Aims: To describe the time required to reach important milestones between first contact with a potential hospital, to recruitment of their first patient to AVERT.
Methods: Hospitals were selected after feasibility questionnaires, phone contact and face to face meetings with potential staff. Records of all contact were maintained throughout.
Results: Of 76 hospitals indicating preliminary interest, 36 proceeded to ethics approval and 25 hospitals have recruited patients. The time between initial hospital contact and selection to participate ranged from 4 to 694 days. Full ethics approval took between 20 and 562 days from submission. Training completion after ethics approval ranged from 1 to 405 days. Once trained, it took between 1 and 816 days to recruit the first patient. In total, the time from initial contact to recruitment of first patient ranged from 139 to 1460 days. We will describe major reasons for variation.
Discussion: When planning a clinical trial, the time (and cost) associated with selecting and starting up hospitals needs careful consideration. We hope our experiences aid other trialists.
O115
Recovery trajectories and prognostic indicators for stroke in China
Wei J1, Heeley E1, Wang JG2, Huang Y3, Wong L4, Li Z5, Heritier S1, Arima H1 and Anderson C1, for the ChinaQUEST Investigators
1The George Institute for International Health, Royal Prince Alfred Hospital and University of Sydney, Sydney, NSW, Australia; 2Centre for Epidemiological Studies and Clinical Trials, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China; 3Department of Neurology, Peking University First Hospital, Beijing, China; 4Prince of Wales Hospital, Chinese University of Hong Kong, Hong Kong, China; 5Xian Traffic University Medical College First Hospital, Xian, China
Background: Stroke is a leading cause of mortality and morbidity in China.
Aims: To assess predictors of, and recovery patterns for, stroke.
Methods: Outcome data for patients (n = 6354) in the ChinaQUEST study. Logistic regression models to determine factors associated with outcome.
Results: Early recovery was greater for haemorrhagic stroke (ICH), however ischaemic stroke (IS) patients were twice as likely to experience a good outcome by 12 months post-stroke (OR 1.98, CI 1.76–2.24). In IS patients, diabetes and atrial fibrillation were strongly associated with poor outcome at 12 months post-stroke, after adjustment for confounders including age and severity; whilst use of antiplatelet and lipid-lowering therapy post-stroke were associated with improved outcome. For ICH patients, low education and atrial fibrillation were associated with poor outcome following adjustments, and antihypertensive use was associated with improved outcome. Despite benefits of secondary prevention strategies, only 66% and 17% of IS patients were on antiplatelet and lipid-lowering therapy, respectively; and only 80% of ICH patients were on antihypertensives by 12 months post-stroke.
Conclusion: In China, ICH and IS patients have different recovery patterns but share similar prognostic factors. To maximise odds of a good prognosis, uptake of evidence based secondary prevention therapies should be improved.
O116
Cost determinants of acute stroke care in China
Wei J1, Heeley E1, Jan S1, Huang Y2, Wang JG3 and Anderson C3, for the ChinaQUEST Investigators
1The George Institute for International Health, Royal Prince Alfred Hospital and University of Sydney, Sydney, NSW, Australia; 2Department of Neurology, Peking University First Hospital, Beijing, China; 3Centre for Epidemiological Studies and Clinical Trials, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China
Background: Stroke costs are projected to rise in China due to ageing and demographic transitions.
Aims: To determine factors predictive of acute stroke care costs.
Methods: Data was collected using the ChinaQUEST study. Multilevel mixed regression analysis was used to determine factors associated with hospital costs.
Results: Overall, mean cost of hospitalisation was 11153 CNY (≍US $1593) per patient, equating to over half a year's wage for the average person in China. Variations in this cost were largely attributable to stroke severity and length of stay (LOS), with effects varying by hospital level. Notably, in large Level 3 hospitals, high annual household income (≥20000 CNY [≍US $2857]) and in-hospital complication(s) were associated with a 10% and 17% increase in cost, respectively; whereas in smaller Level 2 hospitals, health insurance and the hospital being located in a rich province were each associated with a 17% and 134% increase in costs, respectively. Moreover, model forecasts showed that reduction of LOS from the mean to one week afforded cost reductions of approximately 30% and 20% for Level 3 and 2 hospitals, respectively.
Conclusion: Changes in hospital policies to minimise hospital stay may help reduce of acute stroke care costs in China.
O117
Temporal trends and clinical characteristics of spontaneous intracerebral haemorrhage in the Waikato region of New Zealand
Irwin J, Reeve P and Wright P
Waikato hospital, Hamilton, New Zealand
Background: Epidemiological studies have documented a temporal increase in intracerebral haemorrhage (ICH) incidence. No such published data exist for Australasia.
Aims: To determine the incidence of ICH in the Waikato, New Zealand between 1999 and 2008. To analyze clinical and patient parameters, and to correlate these with outcome.
Methods: A retrospective analysis of records was performed on patients presenting to Waikato Hospital with ICH during the study period.
Results: A total of 686 ICHs were identified. The annual incidence per 100,000 per year was 17.9 (16.6–19.3, 95%CI). This increased from an average of 15.6 (13.3–18.2) between 1999 and 2001 to 22.9 (20.2–25.8) between 2006 and 2008 (P < 0.0001). Two hundred and fifty eight (37.6%) patients died within 30 days of the sentinel bleed. Intraventricular extension of bleed on CT (OR 5.76, P < 0.001), warfarin (OR 1.96, P < 0.001), and age ≥60 (OR 1.47, P = 0.02) increased mortality within 30 days. ICH occurred seasonally (rate ratio, 95% CI; winter 1.63 (1.30–2.03), spring 1.43 (1.14–1.81), autumn 1.18 (0.92–1.50), summer 1.00 (reference).
Conclusion: ICH has increased in incidence over the past 10 years. Increasing age and increased use of warfarin contributed to this increase. Radiological evidence of extension of intraventricular bleed, the use of anticoagulation and increasing age correlated with poorer survival. ICH was more common in winter.
O119
Sexual adaptation in people experiencing physical impairment from stroke and other conditions
Hamam N1, McCluskey A2
1Faculty of Health Sciences, The University of Sydney, NSW, Australia; 2Community-Based Health Care Research Unit, Faculty of Health Sciences, The University of Sydney and Royal Rehabilitation Centre Sydney, NSW, Australia
Background: Stroke and other conditions often result in physical impairment which can interfere with sexual activities. Research to date has primarily focused on the negative biomedical and psychosocial affects, rather than strategies that promote participation. Through adaptation, many people successfully carry out their preferred sexual activities.
Aim: This study aimed to investigate the practices, equipment, positions and adaptive strategies used by adults with physical impairment and their partners during their sexual activities.
Methods: A qualitative study design, semi-structured interviews and framework analysis were used. The sample included 18 adults with physical impairment and 13 partners of adults with physical impairment (including five stroke survivors and five partners). Questions focused on sexual activities, adaptive techniques, equipment used, and advice received from health professionals.
Results: Participants described sexual adaptation strategies to overcome six areas of concern: movement impairment, pain, fatigue, sensation loss, genital dysfunction and incontinence. Strategies that helped included: returning to sex early; exploring different sexual activities; talking openly about the problems; and checking with doctors about the effect of medications.
Discussion: Successfully resuming sexual activities is possible despite varying degrees of physical impairment. A range of practical strategies described by participants will be presented and the implications for health professionals will be discussed.
O120
Are rehabilitation and/or care co-ordination interventions delivered in the community effective in reducing depression, facilitating participation, and improving quality of life after stroke?
Graven C1,2, Joubert L1 and Brock K2
1The University of Melbourne, VIC, Australia; 2St. Vincent's Hospital Melbourne, VIC, Australia
Aim: To conduct a systematic review to explore the effectiveness of community-based rehabilitation interventions delivered by allied health professionals and/or nursing staff in reducing depression, facilitating participation, and improving health-related quality of life (HRQoL) post inpatient stroke rehabilitation.
Method: A search was conducted in the databases of MEDLINE, PEDro, CINAHL, and the Cochrane Library. Publications were classified into intervention categories. Best Evidence Synthesis and meta-analysis were utilised to determine the level of evidence.
Results: Fifty-four studies were included in the review, and divided into nine broad intervention categories. Meta-analysis demonstrated a significant benefit for exercise interventions in depression (n = 137; Effect estimate SMD −2.03, 95% CI −3.22, −0.85). Community-based interventions targeting participation and leisure domains showed moderate evidence for improvement in global participation measures and HRQoL. Comprehensive rehabilitation demonstrated limited evidence for depression and participation, and strong evidence for HRQoL.
Conclusion: There is limited to moderate evidence supporting some rehabilitation interventions in affecting the outcomes of depression, participation, and HRQoL post-stroke. The overall heterogeneity of the studies made evidence synthesis a difficult task. Further consideration needs to be given to the type and timing of outcome measures selected to represent the domains of participation and HRQoL.
O121
Assessing the validity of the Barwon health 121 dysphagia screening tool within an acute stroke population
Robertson J and Grimley R
Nambour General Hospital, QLD, Australia
Background: Multiple validated screening tools are available for dysphagia in acute stroke, with no evidence to clearly recommend any one. The Barwon Health 121 Dysphagia Screening Tool (121 screen) is appealing as it is fast, intuitively easy to follow and performed well in initial validation studies [1].
Aims: To assess the validity of the 121 screen against speech pathologist swallow assessment in a representative acute stroke population.
Methods: During a 6 month introduction period, consecutive patients admitted with a diagnosis of acute ischaemic or haemorrhagic stroke were assessed prospectively with both the 121 screen by trained nursing and medical staff and formal bedside swallow assessment by a speech pathologist.
Results: One hundred and ten of one sixty one (66.26%) patients were screened of whom 107 received speech pathology swallow assessment. Screening rates improved from 50.6% (in the first 3 months) to 81.2% (in the second 3 months). The sensitivity and specificity were 87.1% and 92.1% respectively. Predictive value positive was 81.8% and predictive value negative 94.6%. All seven patients with pneumonia failed the screen.
Discussion: The 121 Screen demonstrated good uptake and excellent validity in a representative acute stroke population.
O122
Does intensive blood pressure lowering affect brain volume measures? INTERACT1 trial results
Fullard K, Delcourt C, Hata J, Heeley E, Arima H and Anderson C, for the INTERACT I Investigators.
The George Institute for International Health, Royal Prince Alfred Hospital and the University of Sydney, Sydney, NSW, Australia
Background: The INTERACT1 study suggests that rapid blood pressure (BP) lowering affects haematoma growth in intracerebral haemorrhage (ICH), but the effects on other brain volume parameters remain uncertain.
Aim: To determine the effects of study treatment on midline shift volume (MSV) and Evan's index (EI).
Methods: MSV and EI were measured in 100 INTERACT1 patients' baseline and 72 hours CT scans. The assessor was blinded to the study treatment. Mistar software was used to carry out volumetric analysis. Treatment effects of rapid BP lowering on absolute and proportional changes in MSV and EI were estimated using a multivariate logistic regression model with adjustment for baseline haematoma volume and time to CT.
Results: Mean systolic BP (1–24 hours from randomisation) was 17.7 mmHg lower in the intensive group than the control group. Adjusted mean difference in absolute MSV change over 72 hours was −0.2 ml (95%CI: −1.8 to 1.4 ml, P = 0.77) and the equivalent proportional MSV change was −2.4% (95%CI: −15.2 to 10.4%, P = 0.71). Similarly, adjusted mean difference in absolute EI change over 72 hours was 0.0 ml (95%CI: −0.01 to 0.01 ml, P = 0.73) and proportional EI change was 0.7% (95%CI: −3.6 to 4.9%, P = 0.76).
Conclusion: Early, intensive BP lowering has no appreciable effect on MSV or EI in ICH.
O123
Importance of growth of midline shift volume in intracerebral haemorrhage (ICH): INTERACT1 trial results
Fullard K, Delcourt C, Hata J, Heeley E, Arima H and Anderson C, for the INTERACT I Investigators.
The George Institute for International Health, Royal Prince Alfred Hospital and the University of Sydney, Sydney, NSW, Australia
Background: Mass effect secondary to intracerebral haemorrhage (ICH) predicts a poor outcome. Midline shift volume has been little explored, but may represent a sensitive prognostic indicator.
Aim: Using the INTERACT1 dataset, to determine if midline shift volume (MSV) predicts death or dependency at 90 days.
Methods: CT scans from 89 patients (average age 63 years; 67% male) were analysed. MSV values (baseline and 72 hours, ml) were obtained by a single, blinded reporter, with high inter- (n = 23) and intra-reporter (n = 10) reliability established. The relationships between absolute MSV difference (MSV 72-0%) and relative MSV difference (MSV 72-0%), and death or dependency (modified Rankin Scale 3 to 5) at 90 days were estimated using a multivariate logistic regression model.
Results: Significant association was shown between MSV 72-0 and MSV 72-0%, and death or dependency. These associations remained significant even after controlling for age, sex, randomised treatment and log transformed baseline haematoma volume (OR: 1.68 per 1SD increase of MSV 72-0, 95%CI: 1.00 to 2.82, P = 0.049 and OR: 1.98 per 1SD increase of MSV 72-0%, 95%CI: 1.12 to 3.50, P = 0.02).
Conclusions: MSV 72-0 and MSV 72-0% are novel prognostic markers of poor outcome in ICH, independent of baseline haematoma volume.
O124
Home or clinic: where is the best place to receive stroke rehabilitation for people living in the community?
Luker J1 and Inglis-Jassiem G2
1Centre for Allied Health Evidence, University of South Australia, SA, Australia; 2Faculty of Health Sciences, Stellenbosch University, South Africa
Background: People who have left the inpatient or hospital setting after stroke often require ongoing rehabilitation. It has been hypothesised that rehabilitation delivered in the person's home may be more effective than a traditional outpatient clinic, because of the more realistic setting and the potentially increased involvement of carers.
Aim: To evaluate the effectiveness of stroke rehabilitation for community-dwelling people with stroke, delivered in their own home compared with an outpatient clinic or day hospital setting, on measures of activity, carer issues, cost and other benefits.
Methods: Systematic review of the literature with meta-analysis of common outcome measure/s.
Results: Eleven trials were found and results pooled for the Barthel Index, the measure of functional independence used most consistently. There was a significant effect in favour of home-based rehabilitation at 6 weeks (P = 0.03) and 3–6 months (P = 0.01). Individual studies reported cost benefits and increased carer satisfaction in favour of home-based rehabilitation.
Conclusion: The provision of stroke rehabilitation for people living in the community should trend towards home-based, requiring a major shift in service provision policy. Further research is required into adverse events and the experiences of all stakeholders.
O126
Acute stroke care: a hospital audit of mobilisation practices
Sheedy R1,2, Shields N2 and Bernhardt J2,3
1Barwon Health, Geelong, Victoria, Australia; 2LaTrobe University, VIC, Australia; 3National Stroke Research Institute, Melbourne, VIC, Australia
Background: Early mobilisation (EM) is recommended in many acute stroke management guidelines. However, whether EM is actually undertaken as standard practice is unclear and national auditing does not capture EM.
Aim: To establish current standard practice for the timing and frequency of mobilisation on an acute stroke unit.
Methods: The medical records of 100 consecutive patients with a confirmed diagnosis of stroke and admitted to an acute stroke unit were retrospectively audited. Information about mobilisation over the first 14 days of admission was collected.
Results: On admission 42.6% (n = 40) of patients were able to walk and 25.5% (n = 24) walked without any assistance. The initial mobilisation of 9.6% (n = 9) of patients was recorded within 12 hours of admission while 38.3% (n = 36) of patients were mobilised within 24 hours and 40.4% (n = 38) of patients were mobilised over 24 hours after admission. Physiotherapists recorded 61.4% (n = 51) of initial mobilisations while nurses and doctors recorded 35.9% (n = 29) and 3.6% (n = 3) respectively. Medical restrictions prevented mobilisation in 5.3% (n = 5) of patients.
Conclusion: Despite the number of patients assessed as being able to walk on admission, only half of the patients had mobilisation documented within the first 24 hours of admission.
O127
TIA risk stratification: can this impact on management?
Bilney B, Denisenko S and Ermel S
Department of Health, VIC, Australia
Background: Risk stratification tools after Transient Ischaemic Attack (TIA) may be used to guide clinical decision making regarding admission, investigations and treatment.
Aim: To identify if retrospective ABCD2 scores were related to decisions to admit, frequency of investigations and the initiation of secondary prevention therapy.
Method: Data were collected via a retrospective file audit of 389 patients with TIA who were admitted or discharged from the Emergency Department (ED) of nine Victorian hospitals. The ABCD2 score was extracted from the medical file and categorized as low -moderate risk (≤4) or high risk of stroke (≥5). Data were analysed using Chi Square tests.
Results: Overall, 51.4% of the sample were categorized as high risk of stroke. High risk patients were more likely to be admitted than discharged directly from ED (P = 0.004), to receive Carotid Doppler (P = 0.023) and to be discharged with antihypertensive (P = 0.44) and antithrombotic therapy (P = 0.03). High risk of stroke was not associated with increased frequency of MRI, echocardiography or lipid lowering therapy.
Conclusion: This audit demonstrates some association between risk stratification factors and frequency of admission, investigation and treatment; however it is likely that other factors influence clinical decision making.
O128
TIA management: exploring variations in practice
Bilney B, Denisenko S and Ermel S
Department of Health, VIC, Australia
Background: There is some evidence that Australian patients admitted after Transient Ischaemic Attack (TIA) have a lower frequency of stroke and TIA at 28 days than those discharged from the Emergency Department (ED) [1]. Kehdi et al. suggest this may be related to ‘delay or omissions of appropriate investigations and treatment’ (p. 11).
Aims: To identify differences between admitted and discharged patients with TIA in the frequency and timing of investigations, and discharge on secondary prevention medications.
Methods: A retrospective audit of 189 patients discharged from ED and 200 patients admitted with a primary diagnosis of TIA was conducted at nine Victorian hospitals. Data were analysed with Chi-Square tests.
Results: The admitted TIA patient group were significantly more likely than the ED group to receive CT (96% vs. 76.2%), MRI (20.5% vs. 4.8%), Carotid Doppler (72% vs. 27.5%) and Echocardiography (35.5% vs. 11.1%) P = <0.001. Admitted patients were also more likely to be discharged with antihypertensive (71.5% vs. 43.9%), antithrombotic (90.9% vs. 70.4%) and lipid lowering therapy (56% vs. 34.4%) P = <0.001. There was considerable within group variation for time to investigations.
Conclusion: There is significant variation in the initiation of investigations and secondary prevention management for admitted and discharged TIA patients.
O134
Does advanced CT imaging in acute stroke influence decision-making or prognostication?
Chung KK, Levi C, Spratt N, Quain D and Parsons M
Neurology Department, John Hunter Hospital, Newcastle, NSW, Australia
Background: Non contrast computer tomography (NCCT) is the commonest imaging modality employed in management of acute ischaemic stroke. In some centres advanced CT imaging techniques including CT angiography (CTA) and CT perfusion (CTP) are part of emergent imaging.
Aims: We conducted a retrospective questionnaire study to ascertain whether advanced CT imaging added to NCCT altered thromobolysis decision-making or outcome prediction.
Method: Patients were randomly selected from the John Hunter CTP imaging database. Three stroke neurologists were given a questionnaire with pertinent patient data and reviewed anonymised imaging data; NCCT first, followed by CTA and finally CTP. At each step, the neurologists predicted likely outcome and whether they would give thrombolytic treatment.
Result: Treatment decision based on clinical info and NCCT (CN) was changed in 32.5% of patients after CTA added (CNA) and after both CTA and CTP added(CNAP). Interobserver agreement for (CN) was 75%, (CNA) 50% and (CNAP) 75%. Correct 3 month outcome prediction were (CN) 45%, (CNA) 50% and (CNAP)70%. Confidence level with treatment decision with (CN) high 3%, medium 35% and low 62%, (CNA) high 7.5%, medium 71% and low 21.5%, (CNAP)high 72%, medium 28%.
Conclusion: This retrospective questionnaire study shows that though advanced CT imaging probably does not alter thrombolysis decision in most patients, it certainly improves patient outcome prediction, and increases treating neurologists' confidence in their decisions about thrombolysis and also their in prognostication.
O136
Self-management programs for stroke survivors: results of a Phase II, single blind randomised controlled trial.
Cadilhac D1,3,6, Kilkenny M1,6, Hoffmann S2, Osborne R3, Lindley R4, Lalor E2 and Battersby M5, on behalf of the National Stroke Foundation.
1National Stroke Research Institute of Florey Neuroscience Institutes, Melbourne, VIC, Australia; 2National Stroke Foundation, Melbourne, VIC, Australia; 3Deakin University, Melbourne, VIC, Australia; 4University of Sydney, NSW, Australia; 5Flinders University, Adelaide, SA, Australia; 6The University of Melbourne, VIC, Australia
Background: We are unclear whether stroke-specific chronic disease self-management programs are feasible and effective.
Aims: To assess whether a chronic disease self-management program designed for patients with stroke (SSMP: 8 weeks) is safe and feasible compared to standard care (SC) or a generic program (GCDP: 6 weeks).
Methods: Participant recruitment from South Australia occurred by letter from seven hospitals or via indirect approaches (e.g. newspapers). Eligible participants were randomised 1:1:1 ratio. Sample size: 50 per group. Primary outcomes: recruitment, participation and adverse events. Secondary outcomes: change in baseline self-efficacy, quality of life and mood at 6 months.
Results: A total of 315 screened, 235 eligible and 143 randomised (48 SSMP, 47 GCDP, 48 SC): 89% from direct approach. Mean age 69 (SD 11); 59% female. Age, gender and education levels were similar between groups. 57% accessed the interventions and completion of >50% of sessions: 52% SSMP and 38% GCDP (P 0.18). 32 participants reported adverse events (7 SC, 12 GCDP, 13 SSMP P 0.3; 34% severe) and none were attributable to the interventions.
Conclusions: Access to the interventions was a limiting factor. Feasibility of the SSMP may be better than for a generic program. These issues should be considered when designing efficacy trials.
O137
Have the national stroke foundation's public awareness campaigns increased ambulance dispatches for stroke in Melbourne?
Bray JE, Bailey M, Mosley I, Barger B and Bladin C
Ambulance Victoria, Eastern Health, National Stroke Foundation, Monash University, Melbourne, VIC, Australia
Introduction: The National Stroke Foundation (NSF) stroke awareness campaigns have occurred annually during Stroke Week (September) since 2004. From 2007 the campaign has used FAST (Face, Arm, Speech, Time) and has specified calling an ambulance.
Aim: To explore the impact of these campaigns on Melbourne ambulance dispatches for stroke (Advanced Medical Priority Dispatch Card 28).
Methods: Cross-sectional study examining monthly proportions of emergency dispatches for stroke between 1999 and 2009. Seasonal variation was examined. The proportions of dispatches for stroke were statically compared for the month prior to Stroke Week (August) and the month after Stroke Week (October) for each year.
Results: Between 1999 and 2009, the annual proportion of dispatches for stroke increased from 2.1% (n = 4327) to 2.85% (n = 9208). From 2005, the peak season for stroke calls shifted from winter to spring. When stroke dispatches in August were compared to those in October, a significant increase in October was only detected for 2007 (2.62% to 3.00%, P = 0.006), 2008 (2.62% to 3.05%, P = 0.003), and 2009 (2.70% to 3.09%, P = 0.007).
Conclusions: Since 2007, when calling an ambulance was added to the FAST message, there have been significant increases in ambulance calls for stroke in the month following the National Stroke Foundation's annual Stroke Week.
O138
An examination of factors related to patient's delay in presenting to hospital for stroke symptoms
Bray JE, Cui J, O'Connell B, Gilligan A, Livingston P and Bladin C
Eastern Health, Deakin University and Monash University, Melbourne, VIC, Australia
Introduction: Previous research has identified significant delays when stroke patients are involved in prehospital decision-making, with most of this time spent between deciding to seek medical attention to presenting to hospital. This study aimed to examine factors related to delay during this period of time.
Methods: Stroke patients (n = 100) admitted to two hospitals were interviewed. Cox regression analysis identified factors related to delays in arriving at hospital after deciding to seek medical attention (utilization delay > 45 minutes) and total prehospital delay > 3.5 hours.
Results: The majority of patients had mild strokes (88% National Institute Health Stroke Scale < 8). Utilization delay > 45 minutes was related to: thinking they could manage symptoms themselves (HR = 0.20, 95%CI:0.07–0.60, P = 0.004); wanting to see a family doctor (HR = 0.07, 95%CI:0.01–0.49, P = 0.008); no tertiary education (HR = 0.37, 95%CI:0.16–0.81, P = 0.014); and not knowing a stroke occurred in the brain (HR = 0.27, 95%CI:0.08–0.89, P = 0.03). A total delay of >3.5 hours was related to wanting to see a family doctor (HR = 0.20, 95%CI:0.10–0.40, P < 0.001); choosing to ignore symptoms (HR = 0.37, 95%CI:0.21–0.64, P < 0.001); not considering symptoms serious (HR = 0.44, 95%CI:0.26–0.72, P = 0.001); and not being aware of the FAST campaign (HR = 0.48, 95%CI:0.24–0.96, P = 0.036).
Conclusions: Awareness of the FAST campaign was associated with presentation in time to receive thrombolytic therapy. Future campaigns need to address the preference of patients seeking medical attention from a family doctor instead of seeking emergency care.
O139
Do GP triage protocols address stroke as a medical emergency?
Bray J, Wright M, Gilligan A, Burke M and Bladin C
Eastern Health, Monash University, Melbourne, VIC, Australia
Introduction: Up to 30% of patients contact their General Practitioner (GP) for acute stroke symptoms and the majority of these have long delays in presenting to hospital. GP receptionists, through screening and triaging, have an important role to play in reducing this delay.
Aim: The aim of this study was to determine the current triage practice for patients calling with stroke symptoms.
Methods: A self-administered questionnaire was mailed out to receptionists working in 130 practices in eastern Melbourne. Data collected included: demographic characteristics, triage training; existing triaging protocols; action taken for specific presenting symptoms.
Results: Surveys were received from 100 receptionists from 39 (30%) clinics. The majority of receptionists (72%) worked part time, median of 7 years experience (range 3 weeks–40 years). Triage protocols were received from 17 clinics. The majority of these did not address stroke or its symptoms (58%), 29.4% prioritised stroke as an emergency but did not describe specific symptoms. Calling an ambulance or immediate triage by a nurse or GP was the action nominated by receptions for 85% of patients presenting with arm weakness, 85% droopy mouth, 89% slurred speech, and 79% leg weakness.
Conclusion: The majority of practice triage protocols do not reflect stroke as a medical emergency. However, most receptionists stated they would treat patients presenting with stroke symptoms as an emergency.
O142
Comparison of response rates and completeness of mail versus telephone outcome assessment: a randomised evaluation of a stroke registry
Lannin N1,2, Cadilhac D3,4, Anderson C2, Price C5, Lim J2, Hung YT2, Faux S6, Levi C7 and Donnan G3,4, on behalf of the AuSCR Consortium Partners
1Rehabilitation Studies Unit, The University of Sydney, NSW, Australia; 2The George Institute for International Health, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 3National Stroke Research Institute a subsidiary of Florey Neuroscience Institutes, Heidelberg Heights, VIC, Australia; 4The University of Melbourne, VIC, Australia; 5National Stroke Foundation, Melbourne, VIC, Australia; 6St Vincent's Hospital, Sydney, NSW, Australia; 7University of Newcastle & Hunter Medical Research Institute, Newcastle, NSW, Australia
Background: There is uncertainty over the most effective and efficient method of patient follow-up in disease registries.
Aim: To determine the most effective method of outcome assessment of patients registered in the Australian Stroke Clinical Registry (AuSCR).
Methods: Among participants (n = 437) registered in AuSCR due for a 3-month follow-up assessment, they were randomly allocated (1:1 ratio) to receive either a questionnaire (covering any re-admission to hospital, stroke recurrence, and current health-related quality of life) by mail or telephone. Responses by time, resource use and completeness were compared. Sociodemographic characteristics, health and disease factors considered to be associated with the outcomes were explored.
Results: Participants (mean [SD] age 68 [16] years; 52% male) had equal proportions of follow-up method. The telephone group provided more complete data than mail group (68% vs. 49%; OR 2.21 95% CI 1.50–3.24; P < 0.0001). Mean (SD). time for completion was 33 (35) days for telephone compared to 43 (29) days for mail (95%CI 2–18, P = 0.009).
Discussion: Telephone interview provided the most complete response in the shortest timeframe. However, the manpower costs of the telephone method are likely to be greater on a national scale. Therefore, mail follow-up maybe a more feasible compromise for full roll-out of AuSCR.
O143
Establishing walking using treadmill walking with body weight support in subacute non-ambulatory stroke: the mobilise Trial I
Ada L1, Dean C1, Morris M2, Simpson J1 and Katrak P3
1The University of Sydney, NSW, Australia; 2The University of Melbourne, VIC, Australia; 3Prince of Wales Hospital, Sydney, NSW, Australia
Background: Many people are unable to walk after stroke.
Aim: To determine whether treadmill walking with weight support was more effective at establishing independent walking more often and earlier, than current intervention for non-ambulatory stroke patients.
Methods: A prospective, randomised trial with a 6 month follow-up and blinded assessment was conducted. 126 stroke patients who were unable to walk independently early after stroke were randomly allocated to receive up to 30 minutes of treadmill walking with partial weight support via an overhead harness per day or up to 30 minutes of overground walking. The primary outcome was the proportion of participants achieving independent walking each week.
Results: The proportion of experimental participants who achieved independent walking was 37% compared with 26% of the control group at 1 month, 66% compared with 55% at 2 months, and 71% compared with 60% at 6 months (P = 0.13). The experimental group walked 2 weeks earlier, at 5 weeks compared with 7 weeks for the control group. Also, 14% (95% CI −1 to 28) more of the experimental group were discharged home.
Conclusion: Treadmill walking with body weight support appears feasible, safe and tends to result in more people walking independently and earlier after stroke.
O144
Mechanically assisted walking with body weight support versus assisted overground walking in subacute non ambulatory stroke: a systematic review
Ada L, Dean C, Ennis S and Vargas J
The University of Sydney, NSW, Australia
Background: Mechanically assisted walking is becoming a common intervention after stroke.
Aim: To determine whether mechanically assisted walking with body weight support is more effective at establishing independent walking than assisted overground walking in non-ambulatory stroke patients, and whether it is detrimental to speed or capacity of walking.
Methods: A systematic review with meta-analysis of randomised trials was conducted. Databases were electronically searched and the reference lists of retrieved papers were hand searched. The outcomes were the proportion of participants achieving independent walking, walking speed measured during the 10-m walk test and walking capacity measured as distance during the 6-minute walk test.
Results: Four studies comprising 470 participants were included. Mechanically assisted walking with body weight support resulted in more people walking independently at 4 weeks (RD 0.24, 95% CI 0.16 to 0.32) and at 6 months (RD 0.24, 95% CI 0.13 to 0.35); faster walking at 6 months (MD 0.13 m/s, 95% CI 0.02 to 0.24), and further walking at 6 months (MD 55 m, 95% CI 15 to 96).
Conclusion: Mechanically assisted walking with body weight support is more effective than overground walking at increasing independent walking and is not detrimental to walking speed or capacity in subacute non-ambulatory stroke.
O145
Stroke following cardiopulmonary transplantation: rehabilitation issues and outcomes
Bowman M and Faux S
Sacred Heart rehabilitation Unit, St Vincent's Hospital, Sydney, NSW, Australia
Background: Stroke is not an uncommon complication of heart / lung transplantation, occurring in up to 10% of patients. Transplant patients are usually unable to receive usual stroke care, including thrombolysis or stroke unit admission.
The Sacred Heart Rehabilitation Unit is a specialised unit in the areas of stroke and heart/lung transplantation.
Aims: To report on the management and rehabilitation outcomes of this condition.
Methods: Case series. Outcome data was compared with all other stroke patients admitted during the same time period. Outcomes included Functional Independence Measure (FIM), length of stay (LOS), and FIM efficiency.
Results: A total of eight patients with stroke following cardiopulmonary transplantation were admitted for inpatient rehabilitation. For stroke post transplant, the mean LOS was 26 days with a FIM change of 0.7/day. For all other stroke, the mean LOS was 41 days with a FIM change 0.8/day.
Discussion: Patients with stroke following transplant have delayed transfer to rehabilitation, due to factors associated with the transplant. The admission FIM was higher. Rehabilitation was complicated by immunosuppression, rejection/infection and muscle weakness from steroids. Interruptions to rehabilitation were higher. However, a comparable rate of functional change was noted between transplant and non-transplant stroke patients.
O147
Does access to videofluoroscopy have an impact on how clinicians manage patients with acute swallowing issues following stroke?
Smith-Tamaray M1, Wilson L1 and McAllister L2
1Charles Sturt University, Albury, NSW, Australia; 2University of Queensland, Brisbane, QLD, Australia
Background: The utilisation of videofluoroscopy as the gold standard of swallowing assessment is well recognized. Despite this, there is no Australian data related to availability of this service and subsequent impact on clinical care.
Aims: This paper will present data related to availability and accessibility of videofluoroscopy in non-metropolitan NSW and Victoria. The impact of restrictions in videofluoroscopy access on patient care following stroke will then be presented and discussed.
Methods: Data from two phases of a larger mixed methodological study will be presented. Phase one consisted of a speech pathology representative from every non-metropolitan acute public health facility in NSW and Victoria being invited to participate in a structured telephone survey. A response rate of 87% was achieved. Quantitative data were presented through the use of descriptive statistics and geographical mapping of service availability. Phase two consisted of semi-structured interviews conducted with eight clinicians. Data were analysed and presented thematically.
Results: Access to videofluoroscopy was identified as variable and limited by many factors. Clinicians reported differences in management of swallowing function based on these restrictions.
Conclusion/discussion: The differences in practice were of concern from quality of care and medico-legal perspectives. These findings have implications for equity of care and potential outcomes post-stroke.
O149
Curb negotiation following stroke
Evans K1, Said C1,2, Hill K3,4,5, Mackintosh S6, Whitehead C7 and Batchelor F1,5
1The University of Melbourne, VIC, Australia; 2Austin Health, Melbourne, VIC, Australia; 3LaTrobe University, Melbourne, VIC, Australia; 4Northern Health, Melbourne, VIC, Australia; 5National Ageing Research Institute, Melbourne, VIC, Australia; 6University of South Australia, Adelaide, SA, Australia; 7Flinders University, Adelaide, SA, Australia
Background: Curb negotiation is a potentially hazardous walking task, often avoided by people with disability including stroke.
Aims: To quantify spatial and temporal characteristics of curb negotiation following stroke.
Method: Spatial and temporal data were collected using 3D-motion analysis as 15 stroke participants and 15 gender, age and height matched controls stepped up a curb. Controls performed walking trials at two speeds: self-selected (SSS) and matched (MS).
Results: Preliminary results are available from 12 participants. All participants stepped up with their unaffected limb first, but only eight participants stepped up with the affected limb first. MANOVA and post hoc t-tests revealed significant (P < .05) differences between strokes and controls (SSS) for five spatial variables (unaffected lead limb pre-step distance; affected and unaffected lead and trail post-step distance,) and stride time. Additionally, unaffected lead limb pre-step distance and unaffected trail limb post-step distance were significantly reduced following stroke compared with controls at MS (P < .05). Toe clearance did not differ.
Conclusions: Slow gait speed following stroke accounts for some but not all of the differences observed when ascending a curb. Stroke participants have difficulty leading with their affected limb and partial lead foot placement on the curb may compromise stability.
O150
Routine practices for fever, hyperglycaemia and dysphagia management in NSW acute stroke units
Drury P1, Levi C, Griffiths R, Ward J, Grimshaw J, D'Este C, Cheung W, McElduff P, Hardy J, McInnes E1 and Middleton S1
1Nursing Research Institute, St Vincent's & Mater Health Sydney & Australian Catholic University, NSW, Australia; 2University of Newcastle, Priority Centre for Brain & Mental Health Research, NSW, Australia; 3University of Ottawa, Department of Epidemiology and Community Medicine, Canada; 4Ottawa Health Research Institute, Canada; 5University of Western Sydney, School of Nursing and Midwifery, NSW, Australia; 6The University of Newcastle, Centre for Clinical Epidemiology and Biostatistics, NSW, Australia; 7Westmead Hospital and University of Sydney, Department of Diabetes and Endocrinology, NSW, Australia; 8The University of Sydney, Sydney Nursing School, NSW, Australia
Background: Few studies have investigated clinician adherence to acute stroke guidelines and evidence-based recommendations surrounding the management of fever, hyperglycaemia and dysphagia.
Aim: To report routine practices for fever, hyperglycaemia and dysphagia management for the pre-intervention Quality in Acute Stroke Care (QASC) study on acute stroke patients in NSW.
Method: A prospective, medical record audit was conducted between July 2005 and September 2007. All instances of fever (temperature > 37.5°C) and hyperglycaemia (finger prick blood glucose level (FP-BGL) ≥ 11 mmol/l) occurring in the first 72 hours following admission to an Acute Stroke Unit (ASU); and swallowing screening data were recorded.
Results: Nineteen NSW ASUs and 719 patients participated. 24% (n = 170) of patients had fever during the first 72 hours. 43% (n = 73) of febrile patients received paracetamol. For any instance of fever, the median time to administration of paracetamol was 30 minutes. 60% (n = 430) had a FP-BGL recorded during the first 72 hours, of which, 22% (n = 95) were hyperglycaemic. For hyperglycaemic patients, 33% (n = 31) received insulin. The median time to administration of insulin was 62 minutes. 24% (n = 174) of all patients underwent a swallow screen within 24 hours of admission to hospital.
Conclusion: This multi-centre audit provides evidence of variable adherence to the existing evidence-based recommendations for the management of fever, hyperglycaemia and dysphagia within the first crucial 72 hours following a stroke.
O152
Five-year mortality and prognostic factors in an ethnically diverse cohort of elderly stroke patients
Cordato D1, Whiting R2, Shen Q2, Hung WT3 and Chan D2
1Liverpool Hospital, Sydney, NSW, Australia; 2Bankstown-Lidcombe Hospital, Sydney, NSW, Australia; 3University of Technology, Sydney, NSW, Australia
Background: Few observational studies have examined long-term outcome after stroke in the elderly.
Aims: Determine predictors for survival length in elderly stroke patients.
Methods: 186 consecutive acute stroke patients aged ≥65 years admitted to Bankstown-Lidcombe Hospital between 03/2002 and 03/2003 and followed up in 2007/8. Patients censored if lost to follow-up or still alive at end of study (censoring date 27/6/2007). Kaplan-Meier Survival Analysis examined distribution of ‘time to death’. Cox Regression used to model time to death by list of covariates, including stroke type, severity, age and pre-stroke dependence.
Results: Ninety-six (51.6%) patients died before censoring date. Mortality rate was similar between English-speaking and non-English-speaking background patients (55.5% vs. 43.1%, P = 0.15). Median survival was 1109 days (range 1–1925). Time to death was significantly associated with stroke type (P = 0.006); median number of days to death for haemorrhage and infarct 377 and 1688, respectively. Predictors in final model of survival length: haemorrhagic stroke, older age, pre-stroke accommodation other than home, lower GCS, inability to walk and atrial fibrillation (OR = 2.06, 1.05, 1.77, 0.89, 1.67, 3.57, respectively, P < 0.05 for all).
Conclusion: The study identified six predictors for shorter survival length in a cohort of elderly patients followed up to 5 years post-stroke.
O153
Sense: study of the effectiveness of neurorehabilitation on sensation; individual patient characteristics that predict favourable outcomes
Carey L1,2, Matyas T1,2,3, Walker J1,2,4 and Macdonell R5
1National Stroke Research Institute, Florey Neuroscience Institutes, Heidelberg, VIC, Australia; 2La Trobe University, Bundoora, VIC, Australia; 3The University of Melbourne, Parkville, VIC, Australia; 4Deakin University, Geelong, VIC, Australia; 5Department of Neurology, Austin Health, Heidelberg, VIC, Australia
Background: Ability to benefit from rehabilitation may be influenced by individual patient characteristics.
Aim and hypotheses: To quantify the impact of individual patient characteristics on ability to benefit from a novel somatosensory discrimination training program. We hypothesised that of side of lesion and initial severity of sensory loss would be associated with outcome success, but that prior duration of stroke and age would not.
Methods: Fifty stroke survivors with impaired sensation received sensory training within the SENSe randomised control trial. Primary outcome was a combined index of somatosensory discrimination capacity across functional measures of texture discrimination, wrist position sense and tactile object recognition. Impact of individual patient characteristics was investigated using regression analyses.
Results: There was no evidence of differences in outcome based on side of lesion, interval between stroke and onset of training, or age. Improvement tended to be larger with milder initial deficit (<−50 standardised-deficit-units) provided the participant was male, or had no prior stroke, or had non-dominant side impairment; but was not statistically significant.
Conclusion: Stroke survivors improve following sensory discrimination training, irrespective of side of lesion, time post-stoke and age. Severity of initial deficit may impact, such that those with milder deficits show larger improvement.
O155
Pilot validation of the stroke rehabilitation motivation scale
Cordato D1, White G2, Mentis RL2, Ghia D1 and Chan D2
1Liverpool Hospital, Sydney, NSW, Australia; 2Bankstown-Lidcombe Hospital, Sydney, NSW, Australia
Background: There is no validated instrument to measure motivation in stroke patients.
Aims: To study a novel stroke rehabilitation motivation scale (SRMS) performed in two major metropolitan hospitals.
Methods: Validation cohort, in two stages. Stage one: 28-item scale tested for inter-rater reliability in 18 patients. The seven most reliable items were selected to form a 7-item SRMS. Stage two: 7-item SRMS tested for reliability in a further 13 patients; results combined. The SRMS was adapted from Sports Motivation Scale and divided into seven subscales measuring extrinsic and intrinsic motivation.
Results: 19 male and 12 female patients included; 21 had normal/high motivation group and 10 had low motivation [Motivation score: (mean Extrinsic + mean Intrinsic)/2 – Amotivation]. Good internal and inter-rater reliability were demonstrated. SMRS was reduced from 28 to 7 items to make it more clinically practical. Discriminate predictive validity suggested that motivation was related to anxiety, depression and stress; stress was correlated significantly with walking speed (P = 0.04). The higher motivation group had greater, but non-significant, walking speed and sit-to-stand height improvement rates.
Conclusions: The Stroke Rehabilitation Motivation Scale showed good inter-rater reliability and internal consistency, as well as associations between motivation and depression, anxiety and stress.
O156
Improvement in risk factor management stroke survivors exposed to an integrated model of care
Joubert J1,5, Joubert L2, Jackson D1, Wilson A7, Pearce C8, Reid C9, Levi C3, Hankey G4, Ames D1, Davis S5 and Donnan G6
1National Ageing Research Institute; 2Melbourne School of Health Sciences, University of Melbourne, VIC, Australia; 3John Hunter Hospital, NSW, Australia; 4Royal Perth Hospital, WA, Australia; 5Melbourne Royal Melbourne Hospital, VIC, Australia; 6St Vincent's Hospital, NSW, Australia; 7Howard Florey Research Institute, VIC, Australia; 8Department of General Practice, University of Melbourne, VIC, Australia; 9Monash Centre of Cardiovascular Research, Monash, VIC, Australia
Background: Despite acknowledgement that risk factor management is effective in reducing risk of recurrent cerebrovascular disease, there is little evidence that integrated care models are effective in the reduction of risk factors in stroke survivors.
Aim: To compare risk factor management in stroke survivors exposed to ‘usual care’ to those in an Integrated Care Model.
Methods: Stroke survivors were randomised to either the treatment (integrated care) or control (usual care) group and were followed up over 12 months. The model of integrated care involved collaboration between a specialist stroke service, a hospital coordinator, and the patient's general practitioner.
Results: Twelve months post stroke, systolic blood pressure (sBP) was reduced in the treatment group but increased in controls. The group difference was significant (P = 0.04). Treatment patients also exhibited better modification of body mass index (P = 0.007) and number of walks taken (P < 0.001) than controls. Rankin scores indicated significantly reduced disability in treatment patients relative to controls in the year post stroke (P = 0.003). We will report on the results of 240 patients exposed to this consumer focused, carer inclusive secondary risk prevention program.
Conclusion: The integrated care model was found to be effective in modifying risk factors and reducing disability in stroke survivors.
O157
Combination therapy in experimental stroke
O'Collins V1,3, Macleod M4, Donnan G2,3 and Howells D1,3
1Department of Medicine, University of Melbourne, VIC, Australia; 2Florey Neuroscience Institutes, Parkville, VIC, Australia; 3National Stroke Research Institute, VIC, Australia; 4Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom
Background: Stroke is a complex, heterogeneous disease which may benefit from multi-agent therapy. Additional therapies might raise efficacy by extending time-windows for treatment, broadening the range of targets, reducing side-effects and assisting with the drug delivery.
Aims: We sought to a) determine whether therapies in combination were superior to their individual constituents in experimental stroke models; b) to identify the most promising combinations and c) to identify factors moderating combination efficacy.
Methods: We conducted a systematic review and meta-analysis of controlled experiments administering single and multi-agent therapies in ischemic stroke models using infarct size as an outcome.
Results: Ninety-nine therapies were tested in combination in 287 experimental contrasts deploying 5,831 animals. Combination therapy yielded an additional therapeutic benefit of 17% over single treatments. Superior combinations targeted multiple mechanisms and used multiple models of delivery. Evidence also suggested a ceiling on salvageability, affected foremost by the agents, but also by presurgery fasting, hypertension and the therapeutic target.
Conclusion: The findings are encouraging to those seeking to apply combination therapy in acute stroke as a strategy to improve outcome. Whether clinical trials might also be subject to ceiling effects is a matter of speculation due to the paucity of clinical data.
O158
Stretch for contractures: a Cochrane systematic review
Katalinic O1, Harvey L1, Herbert R2, Moseley A2, Lannin N1 and Schur K3
1Rehabilitation Studies Unit, Sydney, NSW, Australia; 2The George Institute for International Health, Sydney, NSW, Australia; 3Bankstown Hospital, Sydney, NSW, Australia
Background: Contractures are a common complication after stroke. Stretch is widely used for the management of contractures. However, it is unclear whether stretch is effective.
Aim: To determine the effectiveness of stretch for the treatment and prevention of contractures.
Methods: A Cochrane systematic review was undertaken. Electronic searches were conducted for randomised controlled trials of stretch interventions. Outcomes of interest included joint mobility, pain, spasticity, activity limitation, participation restriction and quality of life. Meta-analyses were conducted.
Results: Thirty-five studies with 1391 participants were included. Twenty-four studies involved patients with neurological conditions and 12 studies involved stroke populations. No studies administered stretch for more than 7 months. There is high quality evidence that stretch has no clinically important effects on joint mobility in people with stroke (mean difference 0 degrees; 95% CI −4 to 3). There is little or no effect of stretch on pain, spasticity, activity limitation, participation restriction or quality of life.
Conclusion: Stretch does not produce lasting increases in joint mobility. The routine use of stretch for the management of contractures has no or little benefit if performed for less than 7 months. The effects of stretch performed for more than 7 months have not been investigated.
O159
The Lazarus phenomenon after IV TPA in stroke: prognostic factors and outcome
Machumpurath B1, Davis SM2 and Yan B2
1University of Melbourne, VIC, Australia; 2Department of Neurology, The Royal Melbourne Hospital, Melbourne, VIC, Australia
Background: Treatment with tissue plasminogen activator (tPA) is associated with improved outcome in ischemic stroke. A proportion of patients demonstrate rapid and significant neurological recovery within 24 hours (the ‘Lazarus phenomenon’).
Aim: To examine the incidence of the ‘Lazarus phenomenon’, the predictive factors and correlation with clinical outcomes.
Methods: We included 161 patients with ischemic stroke who received intravenous tPA within 4.5 hours. ‘Lazarus phenomenon’ was defined as an improvement of at least 50% within 24 hours as measured by the National Institute of Health Stroke Score (NIHSS). Long term outcomes were assessed by 3 month modified Rankin Scale (mRS).
Results: ‘Lazarus phenomenon’ was present in 44 of 161 patients (27.3%). This correlated with favorable outcomes at 3 months (P < 0.0005). Lower Baseline NIHSS (P = 0.006), mild (NIHSS < 12) versus severe stroke (NIHSS ≥ 12) (P = 0.002), normal serum glucose levels on admission (P = 0.015) and younger age (P = 0.043) predicted the ‘Lazarus phenomenon’.
Conclusion: The ‘Lazarus phenomenon’ defines a rapid responder population and was demonstrated in a quarter of IV tPA-treated patients. It strongly predicts a good clinical outcome. Further research should involve earlier assessment of the phenomenon after IV tPA, which could aid in the selection of patients for subsequent catheter-based therapy.
O160
Systematic review and meta-analysis of the efficacy of training in animal models of stroke. Can it reveal the optimal dose and time for commencement?
Janssen H1,2, Speare S3, Spratt N1,2 and Bernhardt J3
1University of Newcastle, Newcastle, NSW, Australia; 2Hunter Stroke Service, Newcastle, NSW, Australia; 3National Stroke Research Institute, Melbourne, VIC, Australia
Background: Despite evidence from human studies associating better functional outcomes with a greater amount and intensity of rehabilitation, the ideal dosage and time to commencement of post stroke training remains unknown. Translation from experimental models to human rehabilitation studies is relatively underdeveloped, however, there is potential for these models to inform clinical practice.
Aim: Conduct a systematic review and meta-analysis of experimental stroke models to determine study quality and the effect of training on: 1) sensorimotor function and 2) learning, infarct size, stress and mortality.
Methods: Systematic review of controlled studies of training. Training was defined as physically based interventions aimed at improving skilled movement. Data extracted was analysed using a weighted mean difference meta-analysis, with random effects standardized method employed for pooled sensorimotor data.
Results: Quality (CAMARADES checklist) was moderate (5, IQR: 4–6). Skilled movement training (7 studies, n = 166), cardiovascular exercise (10 studies, n = 290), and constraint induced movement (8 studies, n = 170) were the most frequently investigated models of training in experimental stroke.
Discussion: Reporting of quality of studies was fair, with a relatively small pool of animals studied for each training model. Effects of training on stroke outcome and the clinical implication of these results will be discussed.
O161
‘Ramp time’: delays from emergency department (ED) arrival to triage assessment among acute stroke patients presenting by ambulance
Mosley I, Nicol M, Donnan G and Dewey H
National Stroke Research Institute, Melbourne, VIC, Australia
Background: Rapid care protocols to reduce delays for acute stroke patients are now common place in the clinical setting. These protocols frequently include pre-hospital and in-hospital pathways. However, little is known about delays in the transition from the ambulance to the ED.
Aims: To investigate factors associated with delay times from ambulance arrival to triage assessment. (Ramp Time)
Methods: For 6 months in 2004, all ambulance-transported stroke or TIA patients arriving from a geographically defined region in Melbourne (Australia) to one of three hospital emergency departments were assessed. Ambulance records and hospital medical records were analysed.
Results: 198 patients were included in the study. 187 Ambulance Patient Care Records were complete and available for analysis. Median ramp time was 5 minutes (IQ 2–8 minutes). Factors associated with ‘Ramp Time <5 minutes’ were: Presentation to Austin Hospital (P = 0.011) and paramedic reported ‘Time Critical Emergency’ (P = 0.012). Paramedic hospital pre-notification was not associated with rapid triage assessment. However, these cases were less likely to experience prolonged ramp times >9 min. (P = 0.081).
Conclusions: ‘Ramp Time’ can be included in rapid care pathways. Key pre-hospital triggers for rapid triage assessment like pre-notification may further reduce time to treatment for stroke patients.
O162
Hypertension among acute stroke patients presenting to the emergency department by ambulance
Mosley I, Nicol M, Donnan G and Dewey H
National Stroke Research Institute, Melbourne, VIC, Australia
Background: Raised blood pressure (BP) is a potent risk factor and predictive of both first ever and recurrent stroke. Risk factor management including antihypertensive treatment is recommended for both primary and secondary prevention of stroke.
Aims: To investigate the prevalence of hypertension in stroke patients presenting to the emergency department (ED) by ambulance.
Methods: For 6 months in 2004, all ambulance-transported stroke or TIA patients arriving from a geographically defined region in Melbourne (Australia) to one of three hospital emergency departments were assessed. Patients were interviewed, ambulance records and hospital medical records were analysed.
Results: 198 patients were included in the study. 187 Ambulance Patient Care Records were complete and available for analysis. Median systolic blood pressure (SBP) was 165 (IQ 140–180). 137 patients (73%) presented with raised SBP (>140). 99 patients (53%) were being treated for hypertension while 39 patients presented with undiagnosed hypertension. Past History (P < 0.01) and Family History (P = 0.03) of stroke were associated with hypertension on presentation.
Conclusions: The vast majority of patients in this study presented with hypertension including those in high-risk populations. This raises important issues for the provision of risk factor management and treatment strategies to prevent stroke.
O163
Five-year relative survival of admitted patients with transient ischaemic attack (TIA): an analysis of linked hospitalisation and death data.
Worthington JM1,2,3, Gattellari M4,5 and Garden F6
1Liverpool Health Service, Sydney, NSW, Australia; 2Manly and Mona-Vale Hospitals, Sydney, NSW, Australia; 3South Western Sydney Clinical School, The University of New South Wales, Sydney, NSW, Australia; 4School of Public Health and Community Medicine, The University of New South Wales, Sydney, NSW, Australia; 5Centre for Research, Evidence Management and Surveillance, Sydney South West Area Health Service, Sydney, NSW, Australia; 6Sydney School of Public Health, The University of Sydney, NSW, Australia
Background: The risk of stroke following Transient Ischaemic Attack (TIA) is well known. However, a likely increase in the risk of death is less clearly described.
Aims: To estimate age and sex-adjusted relative survival for hospitalised TIA.
Methods: 22,885 patients admitted to hospital with a primary diagnosis of TIA (ICD-10 G45.x), June 2000–July 2006, were identified from the Admitted Patient Data Collection, a census of NSW hospital separations. Data were linked to NSW death registrations to December 31, 2007. ABS statistics on NSW deaths, by sex and age, were used to calculate expected survival estimates.
Results: One year after hospitalisation for TIA, 91.0% of patients had survived, which was lower than expected (95.2%). The relative survival ratio was 95.5% (95% CI = 95.1–95.9). The relative difference between observed (66.6%) and expected survival (77.2%) was greater five years after hospitalisation (Relative Survival Ratio = 86.2%, 95% CI 85.3%–87.2%). Observed survival decreased relative to expected survival with increasing age. In patients aged 50–64 years, the relative survival was 94.7% of the expected rate, decreasing to 81.1% in patients aged 75–84 years at five years.
Conclusions: TIA is associated with both short and longer-term excess mortality. These data may justify intensification of ongoing risk reduction.
O165
Stroke, death and myocardial infarction after emergency department (ED) presentation of transient ischaemic attack (TIA): an analysis of 12,000 presentations
Worthington JM1,2,3, Goumas C4 and Gattellari M4
1Liverpool Health Service, Sydney, NSW, Australia; 2South Western Sydney Clinical School, The University of New South Wales, Sydney, NSW, Australia; 3Manly and Mona Vale Hospitals, Sydney, NSW, Australia; 4School of Public Health and Community Medicine, The University of New South Wales, Sydney, NSW, Australia
Background: There is heterogeneity in TIA management.
Aim: To characterise TIA outcomes by admission status.
Methods: TIA in the NSW Emergency Department Data Collection 2005–07 was linked to death, stroke and myocardial infarction (MI) to June 2008. We compared outcomes by admission status. Multivariate analyses accounted for clustering by hospital and adjusted for age, co-morbidity, admission status and year of presentation.
Results: There were 12,850 patients; 8,402 (65.4%) were admitted. Admitted patients had greater age (P's < 0.05) and co-morbidity. Adjusted rates of death, total stroke or MI in admitted and discharged patients by 7-, 30-, 90- and 365-days were 4.2% and 3.1% (P = 0.03), 6.6% and 5.4% (P = 0.11), 10.8% and 8.3% (P = 0.006), and 21.1% and 16.5% (P < 0.001). Discharge diagnoses in admitted patients were stroke (27.3%), TIA (44.5%) or mimic (28.2%). Compared with discharged patients, 7- and 30-day event rates were significantly greater in patients with stroke (HR = 2.53, 95% CI 2.04–3.13; HR = 2.01, 95% CI 1.67–2.42). Final diagnoses of TIA had lower rates (HR = 0.65, 95% CI 0.48–0.87; HR = 0.73, 95% CI 0.57–0.92).
Conclusion: Higher event rates in admitted patients may relate to greater age and co-morbidity and better diagnosis of stroke and MI. Admitted patients with a final diagnosis of TIA had lower event rates than discharged patients.
O167
Occupation-based outcomes associated with sensory retraining post-stroke
Mastos M and Carey L
National Stroke Research Institute, Heidelberg Heights, VIC, Australia
Background: Sensory impairment is found in 50%–85% of stroke survivors and can have a significant impact on the performance of daily tasks, safety, spontaneous hand use and quality of life. Whilst there have been a growing number of studies exploring the value of occupation-focused practices, none have done so in the context of sensory retraining.
Aims: The purpose of the proposed project was to investigate occupational outcomes associated with sensory retraining post-stroke.
Methods: A quantitative exploratory design was employed which incorporated two studies. Study One explored the impact of a generalization discrimination sensory retraining program on occupational outcomes for 10 stroke participants using the Canadian Occupational Performance Measure, This study was conducted in the context of a larger randomised control study known as SENSe: Study of the Effectiveness of Neurorehabilitation on Sensation. In Study Two, an occupation-based approach that applied the Generalized Discrimination Training (GDT) principles was developed and tested on eight participants.
Results: Both the SENSe and Occupation-based groups demonstrated an improvement in sensory capacity and occupational performance post-treatment which was maintained at 6 weeks follow-up. The magnitude of change in occupational performance outcomes was greater for the occupation-based group.
Conclusions: Implementation of a client-centred occupation-based program that applies the GDT principles for stroke clients at the post-acute phase is feasible. Clinically meaningful improvements in occupational performance outcomes may be achieved which are maintained over time.
O168
Transient ischaemic attack (TIA) emergency department presentations to nsw stroke unit hospitals: hospital and patient predictors of admission to hospital
Gattellari M1, Worthington J2,3,4 and Goumas C1
1School of Public Health and Community Medicine, The University of New South Wales, Sydney, NSW, Australia; 2Liverpool Health Service, Sydney, NSW, Australia; 3South Western Sydney Clinical School, The University of New South Wales, Sydney, NSW, Australia; 4Manly and Mona Vale Hospitals, Sydney, NSW, Australia
Background: NSW stroke units have different admission policies for emergency TIA presentations.
Aim: To determine influence of hospital and patient characteristics on TIA admissions across 22 NSW metropolitan stroke unit hospitals.
Methods: We identified patients with TIA recorded on the NSW Emergency Department Data Collection, 2005–2007. Data were linked to prior hospitalisations in the previous 4.5 years to obtain co-morbidities.
Results: 7,584 patients had an emergency TIA diagnosis. Seventy percent (n = 5,230) were admitted. Admitted patients were significantly older and more likely to have a history of heart failure (75.0% vs. 68.5%), hypertension (70.7% vs. 68.0%), diabetes (73.5% vs. 68.2%), ischaemic heart disease (72.1% vs. 68.3%) and atrial fibrillation (73.0% vs. 68.3%) (P's < 0.05). Prior history of stroke, myocardial infarction and carotid stenosis was not correlated with admission. The proportion of admitted patients varied significantly across sites from 39.0% to 90% (Median 66%, IQR = 58.5%–78.5%). Principal referral hospitals admitted a higher proportion of patients than non-principal referral hospitals (70.6% vs. 66.3%) (P < 0.001). Hospital and age were the only independent, significant predictors of admission (P < 0.001).
Conclusions: Variation in hospital admission rates for TIA emergency presentations across metropolitan stroke units in NSW TIA are less influenced by patient co-morbidities than by age and hospital practice.
O169
Delivering stroke prevention for patients with atrial fibrillation: a cluster randomised controlled trial in a primary health care setting – results from the stop-stroke in AF expert panel
Gattellari M1, Worthington J2,3,4, Blacker D5,6 and Leung D2,3, for the Stop-Stroke in AF expert panel.
1School of Public Health and Community Medicine, The University of New South Wales, Sydney, NSW, Australia; 2Liverpool Health Service, Sydney, NSW, Australia; 3South Western Sydney Clinical School, The University of New South Wales, Sydney, NSW, Australia; 4Manly and Mona Vale Hospitals, Sydney, NSW Australia; 5School of Medicine and Pharmacology, The University of Western Australia, WA, Australia; 6Sir Charles Gairdner Hospital, Perth, WA, Australia
Background: DESPATCH is an ongoing trial randomising 70 practices to expert feedback (Stop-Stroke in AF panel) about atrial fibrillation (AF).
Aim: To identify barriers to the wider uptake of anticoagulation.
Methods: GPs participated in peer-led academic detailing sessions identifying patients where antithrombotic choice had been difficult. Academic detailers collected information using standardised patient proformas. Summarised proformas were sent to one of nine specialists. Feedback was relayed by letter.
Results: Thirty-eight GPs referred 81 patients (Mean age = 76.9 -years) to the Stop-Stroke panel. Of 58 patients currently not receiving warfarin, 79% had a CHADS2 score of 2 or more. Patients not on warfarin received aspirin (48%) or clopidogrel (28%). Almost 25% were not receiving any antithrombotic medication. Cardiologist recommendations against warfarin represented a common barrier to warfarin prescribing (20%) as did previous haemorrhage (10%) or falls risk (10%). Warfarin was judged suitable for the majority of patients, except in cases of unsupervised dementia, alcohol abuse, unmanaged mental illness, and intracranial haemorrhage. GPs requested information about warfarin use after cardioversion and insertion of drug-eluting stents.
Conclusions: GPs were reluctant to contradict specialist recommendations, despite GPs' own wishes to prescribe warfarin. The majority of patients not on warfarin were judged appropriate for anticoagulation.
O171
Professionals' opinions on early mobilisation after stroke
Sjöholm A1, Skarin M1, Linden T1,2 and Bernhardt J2
1Institute of Neuroscience and Physiology, Sahlgrenska University Hospital, Gothenburg, Sweden; 2National Stroke Research Institute, Melbourne, VIC, Australia
Background: Early mobilisation after stroke may be important for a good outcome and is currently recommended in various international guidelines. The evidence base is however limited and no commonly accepted definition of what constitutes early mobilisation is present.
Aim: To explore professionals' opinion about 1) when after stroke first mobilisation should take place, 2) whether early mobilisation may affect patients' final outcome and 3) what level of evidence they would need to change their mobilisation practice.
Methods: A 9 item questionnaire was used to interview stroke care professionals during the combined Stroke Society of Australasia and Smart Strokes Australasian Nursing and Allied Health Conference 2008.
Results: Among 202 stroke care professionals interviewed, 40% were in favour of mobilising both ischemic and hemorrhagic stroke patients within 24 hours of stroke onset. There was no clear agreement in the optimal time point beyond 24 hours. Most professionals (57–76%) thought that patients' final outcome depends on being mobilised early. Only 19% required a large randomised controlled trial or a systematic review to change practice.
Conclusion: The diverging opinions highlight the absence of clearly defined guidelines and evidence base in this important area of stroke recovery and rehabilitation, which highly motivates further research.
O175
Assessing goal achievement in focal spasticity management
Baguley I1, Turner-Stokes L2, De Graaf S3, Katrak P4, Davies L5, McCrory P6 and Hughes A7
1Westmead Hospital, Sydney, NSW, Australia; 2King's College London School of Medicine, London, United Kingdom; 3Caulfield General Medical Centre, Melbourne, VIC, Australia; 4Prince of Wales Hospital, Sydney, NSW, Australia; 5Royal Prince Alfred Hospital, Sydney, NSW, Australia; 6Box Hill Hospital, Melbourne, VIC, Australia; 7Austin & Repatriation Medical Centre, Melbourne, VIC, Australia
Background: Post-stroke spasticity produces a variable degree of functional impairment, modified by the severity of muscle over-activity and the degree of retained voluntary motor control. The wide range of potential goals thus makes measurement of improvement with interventions such as botulinum toxin A (BoNT-A) difficult to measure.
Methods: This paper presents data from a secondary analysis of a multi-centre double-blind, placebo-controlled randomized clinical trial. The trial tested the impact of BoNT-A in the management of post-stroke upper limb spasticity against a primary outcome measure of improvement in quality of life. Individualized goal attainment scaling (GAS), spasticity and other person-centred measures (pain, mood, quality of life and global benefit) were assessed.
Results: Ninety subjects completed 2 cycles of treatment/placebo per protocol. A significant treatment effect was observed with respect to goal attainment (Mann-Whitney z = −2.33, P ≤ 0.02). GAS T-scores were highly correlated with reduction in spasticity (rho = 0.36, P = 0.001) and global benefit (rho = 0.45, P < 0.001), but not with other outcome measures. Passive goals (for example, opening an involuntarily clenched fist for cleaning) were achieved more frequently than those requiring active function.
Conclusion: Goal-attainment scaling provided a responsive measure for evaluating focal intervention for upper limb spasticity across a wide range of functional activity, not otherwise identifiable using standardized measures.
O176
Perfusion CT definition of the penumbra in acute stroke
Bivard A, Spratt N, Levi C and Parsons M
University Of Newcastle, NSW, Australia
Introduction and Aims: Computerised tomography perfusion (CTP) imaging in acute stroke may identify penumbra and infarct core. However, this technique has undergone limited study, and validation of perfusion thresholds that can delineate tissue pathophysiology is required.
Method: Acute CTP (0–6 hours after stroke onset) and infarct volume on coregistrered sub acute MRI were assessed in 76 hemispheric ischaemic stroke patients in order to define the entire of extent of the perfusion lesion (penumbra plus infarct core. A pixel-based Receiver Operator Characteristic curve analysis was undertaken to calculate the Sensitivity, Specificity, Positive Predictive Value and the Area Under the Curve to assess the ability of CTP at detecting the acute perfusion lesion.
Results: A Tmax threshold of >2 seconds compared to contralateral hemisphere was found to be the most accurate threshold for defining the acute perfusion lesion (AUC 0.82, sensitivity 0.89, specificity 0.78, PPV 0.79). The second best predictor of the acute perfusion lesion was an MTT of 135% (AUC 0.786, Sensitivity 0.787, specificity 0.785, PPV 0.746).
Discussion: Tmax thresholds on CTP are the most accurate at identifying the entire extent of tissue at risk of infarction.
O177
Can more stroke patients be recruited into an ongoing trial of very early rehabilitation (AVERT)?
Collier J
National Stroke Research Institute (Florey Neuroscience Institutes), VIC, Australia
Background: Recruiting all eligible patients into a Very Early Rehabilitation Trial (AVERT) would optimise trial completion timelines.
Aims: To evaluate non-modifiable and modifiable factors for patients not recruited.
Methods: Study design: Mulitcentre, international, Phase III RCT using nurse/physiotherapist clinicians as recruiters. Inclusion criteria: Patients <24 hours of stroke with physiological parameters within set limits. Exclusion Criteria: Patients with severe premorbid disability, severe comorbidities or needing palliative care. For all admitted patients Logs of age, gender, stroke type, severity and reasons for patients not recruited are recorded. Modifiable exclusion factors: No recruiting staff and administrative. Non modifiable exclusion factors: Admitted >24 hours post stroke, premorbid mRS > 2, medically unwell, admitted to another trial or refused. Factors were not mutually exclusive.
Results: 19 hospitals from July 2006–2009 screened 7143 patients: 489 recruited; 6654 not recruited. Non modifiable: 17.2% were disabled prior to stroke, 18.6% were medically unwell. Few patients were recruited to other intervention trials (1.3%) or refused (1.6%). Delay in hospital admission after stroke prevented 46.4% of patients from being eligible. Modifiable: 29.5% of eligible patients were not recruited.
Conclusions: Stroke clinicians have limited recruitment flexibility. Dedicated trials nurses could improve recruitment rates.
O178
Recurrent sensory symptoms from likely amyloid angiopathy related acute and chronic subarachnoid haemorrhage.
Gerraty R1,2 and Trost N2,3
1Neuroscience Department, Monash University, Alfred Hospital, Melbourne, VIC, Australia; 2Department of Medicine, Monash University, Alfred Hospital, Melbourne, VIC, Australia; 3Medical Imaging Department, St Vincent's Hospital, Department of Radiology, Melbourne, VIC, Australia; 4University of Melbourne, Fitzroy VIC, Australia
Background: Acute and chronic subarachnoid haemorrhage (SAH) may contribute to recurrent transient sensory symptoms mimicking TIA. Susceptibility weighted imaging (SWI) increases the sensitivity of MRI for haemosiderin over gradient echo imaging (GRE).
Methods: We present 2 cases who presented with sensory TIAs investigated with MRI including SWI.
Results: A 68 year old woman with recurrent sensory attacks affecting the right arm and face was found to have left central sulcus acute subarachnoid haemorrhage on FLAIR MRI. Four years later she had the same problem on the other side, with similar MRI findings. GRE MRI showed only very subtle SAH on the second occasion, but SWI was strongly positive for old SAH. A 71 year old man had recurrent sensory attacks affecting the left face and arm, and had acute SAH in the left (sic) parietal region on FLAIR, and extensive old SAH on SWI overlying both hemispheres.
Conclusion: Transient sensory attacks may be due to subarachnoid blood, acute or chronic. Brain CT may show focal subarachnoid bleeding, but is often normal. FLAIR demonstrates the acute blood, but SWI is superior to GRE in demonstrating old subarachnoid blood. MRI with SWI should be performed in older patients with recurrent transient sensory symptoms.
O180
Frequency and type of neurosyphilis admissions to a major teaching hospital in a culturally diverse population between 2005 and 2009 inclusive
Taneja S, Djekic S, Cordato D, McDougall A, Beran R, Cappelen-Smith C, Griffith N, Hanna I, Hodgkinson S and Worthington J
Liverpool Hospital, Sydney, NSW, Australia
Background: Syphilis is a recognised but rare cause of neurovascular events including stroke and cognitive decline.
Aim: To determine the prevalence of Neurosyphilis including stroke related presentations in a culturally diverse population admitted to a major teaching hospital.
Methods: ICD-10-AM diagnostic coding was used to identify all patients admitted to Liverpool Hospital with a primary or secondary diagnosis of Neurosyphilis or late/ secondary Syphilis between 1st January, 2005 to 31st December, 2009.
Results: Twenty-seven patients were identified. Sixteen had an ICD-10-AM diagnosis of Neurosyphilis, seven late Syphilis, two cardiovascular, one congenital and one secondary Syphilis. Five of the subjects presented with a stroke (three with infarction and two intracerebral haemorrhages). The demographics, clinical features and laboratory investigation results will be discussed.
Conclusion: Syphilis is a rare but important cause of serious neurological events in our culturally diverse population
O181
Hospital management of urinary incontinence in patients with acute stroke and the influence of stroke units
Jordan L-A1 and Cadilhac D2,3 On behalf of Stroke Services New South Wales Coordinating Committee a Network of The Agency for Clinical Innovation New South Wales Health, Sydney, NSW. Australia.
1Hunter Stroke Service, Newcastle, NSW, Australia; 2National Stroke Research Institute a subsidiary of Florey Neuroscience Institutes, Heidelberg Heights, VIC, Australia; 3The University of Melbourne, VIC, Australia
Background: Evidence suggests that up to 60% of patients experience urinary incontinence (UI) as a consequence of stroke.
Aim: To describe continence management practices in Australian stroke units compared with general medical wards.
Methods: Retrospective cohort study of consecutive eligible admissions from 32 hospitals in the New South Wales Stroke Audit project between 2003 and 2007 in Australia.
Results: Data for 2,231 patients were included (mean age 73.8 yrs, SD 14, males 50.3%) and 969 (44%) were managed in a SU during their admission. Among patients incontinent within 72 hours of admission more were female (57% v 43% male P < 0.001) and 64% had either a total anterior or partial anterior infarct. Patients managed in SUs had greater odds of having a clinical management plan to avoid UI and complications (OR 0.41 95% CI 0.33 to 0.53). Interestingly, 6.7% of patients in SUs were diagnosed with moderate/severe UTIs compared to 3.2% in general wards (P < 0.001). Patients with a moderate/severe UTI and about 72% did not have a continence care plan regardless of whether treated on a SU or general ward.
Discussion: Stroke teams need to be aware of factors associated with UI such as UTI and implement plans to improve management.
O184
Hospital management and outcomes of stroke in young indigenous australians: evidence from the 2009 national stroke audit-acute services
Kilkenny M2,3, Harris D1, Ritchie E1 and Price C1 and Cadilhac D2,3
1National Stroke Foundation, Melbourne, VIC, Australia; 2National Stroke Research Institute of Florey Neuroscience Institutes, Melbourne, VIC, Australia; 3University of Melbourne, Melbourne, VIC, Australia
On behalf of the National Stroke Foundation Advisory Committee: National Stroke Audit Acute Services and the National Stroke Audit Collaborative
Background: No national review of hospital stroke care and outcomes have been reported for Indigenous Australians.
Aim: To compare adherence to clinical processes and outcomes among Indigenous and non-Indigenous patients admitted to hospital with acute stroke.
Methods: Hospitals participating in the 2009 National Stroke Audit that provided data for ≥ one Indigenous patient (aged 18 to 64 years) were included. Differences in death/dependency (modified Rankin Score 3–6) at discharge were determined using two level (hospital and patient factors) multivariate analysis.
Results: Among 305 eligible patients (64% male), 18% were Indigenous. Baseline risk factors, ischaemic stroke type (Indigenous 57% v non-Indigenous 78%, P = 0.001), stroke unit care (Indigenous 28% v non-Indigenous 44%, P = 0.03), received aspirin <=48 hours, if ischaemic stroke (Indigenous 57% v non-Indigenous 77%, P = 0.03) and outcome status (Indigenous 72%; non-Indigenous 49%, P = 0.005) varied. After adjusting for hospital site, age, risk factors (e.g. diabetes, smoking, alcohol), prognostic variables and stroke subtype, Indigenous patients had a greater odds of dying or being dependant at discharge compared with non-Indigenous Australians (aOR 2.69 95% CI 1.10–6.59).
Conclusions: Differences in the clinical management and outcomes between Indigenous and non-Indigenous patients with stroke treated in Australia were found. Further research to explain these differences is needed.
O185
A comprehensive, specialist assessment service for those with transient ischaemic attack (TIA) in Wellington, New Zealand
Ryder A, McGonigal G, Ioannides N, Jolliffe E and Wong L
Capital and Coast District Health Board, Wellington, New Zealand
Background: Current UK guidelines for TIA management conclude that a practical strategy is to limit specialist TIA assessment to those with ABCD2 scores >3. In 2009, a specialist TIA service was established for all referrals with possible TIA in the Wellington Region.
Aims: To examine the practicality and effectiveness of providing such a service; assessing the need for urgent investigation in those with ABCD2 score <4 and measuring the rate of pharmacological and non-pharmacological intervention.
Results: In 16 weeks, 92 people were assessed: 69 (75%) as OPs. 55 (60%) had TIA, 8 (9%) disabling stroke, and 29 (32%) non-cerebrovascular disease. 28/92 (30%) referrals had ABCD2 scores<4, 5(18%) of whom needed urgent investigation. In those diagnosed with TIA, 20/55 (36%) delayed >2 days to see Primary Care. 42/55 (76%) had pharmacological intervention. 25/36 (69%) with Anterior Circulation TIA had urgent carotid imaging, 5 proceeding to Carotid Endarterectomy. Further performance data will be presented.
Conclusions: It is practical and effective to develop comprehensive TIA services. Approximately 1:5 of those with ABCD2 scores <4 required urgent assessment. Recorded delays in presentation highlight a lack of public awareness of TIA symptoms. In a substantial minority of patients further intervention or imaging was deemed inappropriate. This important group has been under-reported in the literature and is discussed.
