Abstracts Stroke 2018 Conference, 7–10 August 2018, Sydney, Australia: Abstracts Oral presentations 2 Predictors of severity of hypertensive response in acute intracerebral haemorrhage Xia Wang 1 , Else Sandset 2 , Guojuan Chen 1 , Cheryl Carcel 1 , Candice Delcourt 1 , John Chalmers 1 and Craig Anderson 1 1 The George Institute For Global Health, Sydney, Australia; 2 Oslo University Hospital, Department of Neurology, Oslo, Norway Background and aims: Early elevation in blood pressure (BP) known as acute hypertensive response is common in patients with spontaneous intracerebral hemorrhage (ICH), and a known predictor of a poor outcome. We aimed to identify the independent predictors. Methods: We pooled INTERACT1 (n ¼ 404) and INTERACT2 (n ¼ 2839) of acute ICH patients (<6 hrs of onset) with elevated systolic BP (SBP, 150–220 mmHg) who were randomized to intensive (target SBP < 140 mmHg) or guideline-recommended (target SBP < 180 mmHg) management. BP at randomisation was measured at least twice and at least two minutes apart using the non-paretic arm (or right arm in situations of coma or tetraparesis) with the patient supine. Multivariable linear regression was used to determine associations. Results: Among 3233 patients, the mean age was 63.4 (12.8) years, mean SBP 179.3 (17.1) mmHg and 36.8% were female. History of hypertension (estimate 2.295, SE 0.691), admission gly- cemia > 6.5 mmol/lt (3.132, 0.618), elevated heart rate (1.238, 0.222 per 10-bmp increase), and higher NIHSS (0.154, 0.047 per 1-point increase) were significantly associated with hypertensive response at baseline. The use of antithrombotics (3.742, 0.954) and time from onset to random- ization (0.992, 0.255 per 1-hour increase) were inversely associated with hypertensive response. Conclusions: For patients with acute ICH, if they are with history of hypertension, elevated blood glucose and heart rate, severe neurological deficit, not on antithrombotics, and presented earlier, more frequent BP measurement is required and more intensive BP treatment may be needed. 4 Increasing intensity of practice after stroke using apps, internet and sensors to connect patients and therapists remotely: A feasibility study Dawn Simpson 1 , Matthew Schmidt 2 , Marie-Louise Bird 3 , Stuart Smith 4 and Michele Callisaya 1 1 Menzies Institute For Medical Research, Hobart, Australia; 2 School of Health Sciences, University of Tasmania, Hobart, Australia; 3 GF Strong Rehabilitation Research Laboratory, University of British Colombia, Vancouver, Canada; 4 Southern Cross University, Coffs Harbour, Australia Background: Intensity of task practice after stroke is important to improve function, yet adherence to complete exercise programs can be challenging once home. We aimed to determine whether using a chair sensor, tablet application and internet connection could motivate and provide feedback on progress of a sit to stand exercise at home. Methods: Ten participants with stroke completed a 4-week sit-to-stand exercise in the community. Participants learnt how to use the app and chair sensor that a therapist installed in their home. A therapist remotely monitored the exercise program, updated exercise targets, and provided perso- nalised feedback via the app. Feasibility measures included adherence to the prescribed exercise session frequency and number of exercise repetitions (%), and participant satisfaction measures (enjoyment, usability and perceived benefit questionnaires). Results: Participants (mean age 73.6 years [SD 9.9 years], 50% male, mean gait speed 0.57 m/s [SD 0.31 m/s] at baseline) performed 125% of the exercise sessions prescribed over the 4 week period. There was a mean exercise repetition adherence of 104% [range 97% to 111%]. Participants rated the system usability as high (78%), enjoyment as high (70%) and rated perceived benefit of the system positively (80%). Conclusions: It was feasible to prescribe, monitor and progress exercise by connecting participants and therapists remotely using an app and sensor-based system. Exercise session and repetition adher- ence was high, with positive satisfaction reported by participants. A definitive trial is now required to determine if use of such technology may facilitate greater exercise participation and improve function after stroke. 11 Cognitive function on admission predicts motor change in sub-acute rehabilitation: an observational study Ingrid Li 1 , Tram Bui 2 and Katharine Scrivener 3 1 MQ Health, Macquarie University, Australia; 2 Royal Rehab Private, Ryde, Australia; 3 Department of Health Professions, Macquarie University, Macquarie University, Australia Background: Exercise dose is commonly recorded as therapy time. Repetitions of exercise under- taken during therapy more accurately reports the amount of exercise completed. The relationship between exercise dose, measured as repetitions, and functional gains, in sub-acute rehabilitation is under researched. Aims: To compare the types and repetitions of exercise performed by neurological and general rehabilitation groups during physiotherapy intervention in a sub-acute rehabilitation setting and explore the relationship between exercise dose and functional gain. Methods: Data collected from two previous prospective observation studies at Royal Rehab Private were retrospectively examined. Predictors of Functional Independence Measure (FIM) motor score change were investigated using univariate linear regression models. Results: Data were collected from 33 participants. Those with a stroke diagnosis made up approxi- mately half of the neurological group. Exercise dose was similar between groups – 243 (SD 127) versus 245 (SD 115) repetitions in neurological and general rehabilitation groups respectively. Strengthening exercises were performed more than functional exercises in both groups. Only 16% of the variance in FIM motor change was explained by exercise dose. Cognitive FIM on admission accounted for 97% of the variance in FIM motor change in the neurological group. Every point increase in initial cognitive FIM was associated with a 3.7 point decrease in FIM motor change. Conclusion: There was no difference in exercise dose between neurological and general rehabilita- tion groups. Exercise dose does not appear to predict improvements in motor function. In a neuro- logical population, those with poorer cognition on admission may make greater motor gains during rehabilitation. 14 Sensory retraining of the lower limb after stroke: A systematic review Fenny Chia 1,2 , Nancy Low Choy 1,3 and Suzanne Kuys 1 1 School of Physiotherapy, Faculty of Health Sciences, Australian Catholic University, Brisbane, Australia; 2 Tasmanian Health Service, Hobart, Australia; 3 Queensland Health Service, Brisbane, Australia Background: Lower limb somatosensory impairment has detrimental effects on balance and gait. Despite this, little is known regarding the effectiveness of somatosensory retraining in the lower limb following stroke. Aims: This systematic review aimed to investigate the effects of lower limb somatosensory retraining after stroke on balance, gait, and somatosensory impairment. Methods: All types of quantitative studies incorporating somatosensory interventions that addressed the lower limb after stroke were retrieved through a systematic search. Databases searched included Cochrane Library, PubMed, Medline, CINAHL, EMBASE, PEDro, PsycINFO, and Scopus. Reference lists of relevant publications were also manually searched. The Quality Assessment Tool for Quantitative Studies was used for quality appraisal. Standardised mean differences were calculated and meta-analyses were performed using pre-constructed Microsoft Excel spreadsheets. Results: The search yielded 15 studies, comprising of 428 participants using a diverse range of somatosensory interventions. The majority of the studies had a weak quality rating, mainly due to the use of somatosensory measures that have not been evaluated for its psychometric properties. A significant positive summary effect size (SES) was found for balance (SES 0.80; 95% CI 0.40–1.20; I 2 ¼ 72%). Gait SES was not significant. A positive SES was found for somatosensory impairment (SES 0.60; 95% CI 0.09–1.11; I 2 ¼ 76%). Conclusions: This review suggests that lower limb somatosensory interventions post-stroke signifi- cantly improve balance and somatosensory impairment but not gait. Future research should focus on establishing the psychometric properties of available lower limb somatosensory assessment tools, and developing standardised lower limb somatosensory intervention methods in stroke rehabilitation. 16 Telehealth: Enhancing access to secondary prevention education for stoke clients in rural and remote Western Australia Ruth Warr 1 , Prue Matthews 1 and Temika Allen 1 1 WA Country Health Service, Geraldton, Australia Background: The ‘‘Moving on from Stroke’’ education series aims to ensure all clients with stroke or transient ischaemic attack in the Midwest region of Western Australia (WA) are assessed and informed of their risk factors for recurrent stroke. Clients are provided with strategies to identify risk factors and modify behavior. As the Midwest health region of WA covers a geographical area of over 470,000 square kilometres, it is often a challenge to ensure people who have experienced stroke are able to receive stroke care in line with the National Stroke Foundation (NSF) Clinical Guidelines (2017). Telehealth has been identified as a way to decrease the barriers associated with geographical isolation. Aims: To provide comprehensive education on stroke and secondary prevention education to clients in regional WA, irrespective of their geographical proximity to stroke units or rehabilitation centres. Methods: The ‘‘Moving on from Stroke’’ group consists of three sessions covering; what is stroke, risk factors, medications, emotional changes and grief, physical activity and lifestyle modifications. Clients International Journal of Stroke 2018, Vol. 13(1S) 3–48 ! 2018 World Stroke Organization Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1747493018778666 journals.sagepub.com/home/wso International Journal of Stroke, 13(1S)