321 UNMET NEEDS AFTER STROKE

Background

Unmet needs, such as lack of access to rehabilitation, support and information, are experienced by a high proportion of stroke survivors after discharge from acute and rehabilitation services. The UK guidelines on stroke (2016) recommend that all patients should be offered a structured health and social care review at 6 months post-stroke. However, no such clinics exist to date in Ireland. We aimed to explore the extent of unmet needs in the catchment area of a large teaching hospital.

Methods

A random selection of patients 6-months post-stroke were contacted by telephone and unmet needs were assessed using a validated tool, Post-Stroke Checklist. Following analysis of data, the need for a pilot clinic was identified. A separate random selection of patients discharged from acute services 6 months previously were invited to attend a review clinic. Patients were assessed using an adapted version of the Greater Manchester Stroke Assessment Tool. Assessments were conducted jointly by a physiotherapist and a speech and language therapist, and appropriate onward referrals were made.

Results

Telephone clinic: 51 patients completed the checklist. The most prevalent symptoms reported were fatigue (75%), reduced participation in hobbies and activities, decline in cognition (61%) and mobility problems (59%). Review clinic: 21 patients attended. The most prevalent symptoms reported were reduced memory/concentration (71%), low mood (71%), unintentional weight loss/gain (62%), and reduced mobility (43%). Onward referrals were made for in 16/21 (76%) cases; physiotherapy (n=12), occupational therapy (n=7), speech and language therapy (n=7), clinical nutrition (n=5), social work (n=7), psychology (n=5).

Conclusion

Data from this exploratory study supports previous research indicating a high number of stroke survivors with unmet needs. This highlights the importance of establishing post-acute stroke review clinics and pathways in the Irish setting.

317 PREVALENCE OF MOTORIC COGNITIVE RISK SYNDROME IN PATIENTS WITH MILD COGNITIVE IMPAIRMENT ATTENDING A MEMORY CLINIC

Background

Mild Cognitive Impairment (MCI) alone is not predictive of progression to dementia. However, previous research has found that the combination of MCI and slow gait speed, termed Motoric Cognitive Risk syndrome (MCR), is associated with a three-fold increased risk for the development of dementia. This study aimed to identify the prevalence of MCR among patients with MCI attending a memory clinic at a large teaching hospital.

Methods: Participants

Participants were recruited from the memory clinic at the study site. All participants matching the inclusion criteria were invited to participate in the study. Inclusion criteria: Age > 55 years, independently mobile > 50m unaided, diagnosis of MCI. All participants gave informed consent prior to testing.

Procedure

Gait speed (in cm/s) during normal pace walking was measured using a 5-metre computerised walkway with embedded sensors (GaitRite).

Data Analysis

Gait speed data was dichotomised into “normal” or “low” using previously defined gait speed cut-offs derived from age-based norms. Data was analysed using descriptive statistics on Microsoft Excel.

Results

31 participants completed testing (Males: 12; Females: 19). Mean age was 73.65 (range 63-84). Median score on the Montreal Cognitive Assessment (MoCA) was 20 (range 4-28). N = 2 (6.45%) reached criteria for MCR.

Conclusion

Although this study had small numbers, prevalence of MCR was similar to that previously reported among similar populations. This approach to identifying those at high risk for dementia may allow clinicians to target and prioritise further investigations and preventative interventions.

62 GAITKEEPER- A NOVEL ARTIFICIAL INTELLIGENCE BASED VIDEO GAIT ANALYSIS SYSTEM TO OBJECTIVELY MEASURE HEALTH STATUS

Background

Frailty is associated with increased risk of falls, dependency, disability and higher health and social care utilisation. While characterised by a slowly progressive functional deterioration, frailty can be delayed or reversed through timely intervention. Identification of those at risk of new or worsening frailty remains a challenge. How fast we walk (gait-speed) is a reliable correlate of one’s physical, psychological, and cognitive performance. Slow gait speed can discriminate those at high risk of adverse outcomes due to declining health, however current gait-speed assessment instruments are often not feasible in clinical or community settings due to cost, specialist training, and space requirements. Here we introduce GaitKeeper, a system that uses artificial intelligence (AI) to measure gait and step events from video recorded on a mobile phone, thus facilitating identification of those at risk of new or worsening frailty.

Methods

A technical evaluation, in a research setting (n = 30), compared GaitKeeper’s performance capturing accurate positional data in healthy young individuals, versus the Vicon camera motion-capture system as the verification standard.

A subsequent clinical evaluation in people with mild cognitive impairment (n = 30), within an acute hospital, compared GaitKeeper gait speed data with specialist-physiotherapist clinical observational studies, and GaitRite systems during five-meter walking under single and dual-task conditions. Comprehensive measures of frailty (SHARE-FI), cognition (RBANS, MoCA), and physical performance (MiniBEST) were recorded in this clinical evaluation.

Results

The technical evaluation provides evidence of the ability of GaitKeeper to capture accurate gait data from video (n = 30, p < 0.001, error < 3%) versus the Vicon system. The clinical evaluation provides evidence of the accuracy in comparison with the GaitRite gold standard (n = 30, P < 0.001, error < 6%) in clinical setting.

Conclusion

GaitKeeper performance was well validated in comparison to existing gait analysis practices. This innovative AI based solution provides an opportunity to longitudinally monitor and quantify ‘wellness’ in older persons with accessible, accurate gait performance measurements anytime, anywhere, by anyone.

1994 William J. Stickel Bronze Award. Amino acids, dipeptides, and leg skeletal muscle wound healing in diabetes

The authors determined the effects of active amino acids and dipeptides as anabolic agents on surgically induced wound healing in lower extremity skeletal muscles in diabetic and normal rats. In order to induce diabetes, adult male Sprague-Dawley rats (200 g; 10 to 12 animals per group) were injected with streptozotocin (65 mg/kg) 30 days before the onset of the experiment. Their blood sugars were checked at this time. Each group of rats was injected with either one stimulatory amino acid or dipeptide (150 mg/kg body weight) subcutaneously in saline for 7 days and their anabolic effects (RNA, DNA, protein, and collagen contents) on lower extremity skeletal muscle wound healing determined in both diabetic and normal control groups. It is hoped that a treatment regimen will be developed using synergistic anabolic agents locally to decrease the lower extremity muscle healing time. This will enable the diabetic patient to become mobile sooner after surgery.