Marie-Claire Smith

Background.: The TWIST tool uses clinical measures at 1 week post-stroke to predict whether patients will achieve independent walking by 4, 6, 9, 16 or 26 weeks post-stroke. TWIST was developed with 93 participants from 2 hospitals and was at least 83% accurate at all timepoints. Validation of TWIST is necessary before implementation into clinical practice. Objective.: The study aimed to perform temporal external validation of the TWIST prediction tool in a single hospital site. Methods.: Adults with stroke were recruited if they had new lower limb weakness and were unable to walk independently. A TWIST score out of 4 was calculated for each participant using age, knee extension strength and Berg Balance Test score at 1 week post-stroke. Discrimination and calibration of TWIST were calculated for each timepoint and the model was refined as required. Results.: There were 89 participants in this validation cohort (43% women, median age 69 years, 56% moderate-severe stroke). Predictions were valid for most TWIST scores at most timepoints. The TWIST tool was too optimistic for participants with TWIST scores of 0 and 2 at 16 and 26 weeks, and a TWIST score of 3 at all timepoints. The TWIST tool and clinical interpretations were refined. Discrimination of the TWIST tool is good (C-statistic > 0.8) for all timepoints. Conclusions.: TWIST is a temporally externally validated prediction tool that combines routine tests at 1-week post-stroke to predict the probability of an individual patient achieving independent walking by 4, 6, 9, 16 or 26 weeks post-stroke.

Instrumented gait analysis (IGA) has been widely used in research, but not typically in clinical practice, as it requires expertise in data analysis and interpretation. Post-stroke clinical gait assessment could be improved by integrating artificial intelligence into IGA, but previous gait assessment systems have relatively low clinical utility. This study aims to develop a clinically oriented automatic post-stroke gait assessment system based on knowledge graph (KG) to better support clinicians. A domain KG is first constructed in the field of gait analysis. This system can process IGA data to identify gait abnormalities and their potential causes based on kinematic analysis and KG. A preliminary evaluation with twenty post-stroke patients and four domain experts tested the system's performance in clinical settings, showing an average recall, precision, and F-score of 1, 0.78, and 0.89. Four clinical professionals showed high behavioural intention to use the system in clinical settings (4.33 ± 0.41 on a 5-point Likert scale based on the Technology Acceptance Model). The results depicted that this system has potential to be applied in clinical settings to provide useful supplementary insights for clinicians, which may promote the interpretation and clinical utility of IGA. The schema of this KG could be generalised or extended to gait analysis related to other diseases.

Clinical use of Virtual Reality (VR)-based treadmill training to improve post-stroke gait is limited by a lack of adherence to motor learning and control principles, and user-centred design consideration. This study aims to outline the development and evaluation of a VR-based treadmill training system for post-stroke patients, aligning with currently accepted rehabilitation concepts and principles. The proposed system involves several modules and algorithms (e.g., biomechanical model, feedback controller, adaptive difficulty controller) along with portable motion capture technology to provide individualised rehabilitation focused training. Eight physiotherapists and eleven post-stroke participants were recruited to evaluate the system’s acceptability, usability, and adherence to motor learning and control principles. The physiotherapists reported high ratings in terms of the design’s adherence to motor learning and control principles, and both physiotherapists and post-stroke patients reported positive intention to use the system. The proposed system promotes individualised and progressive practice and feedback for post-stroke gait training in VR environments. Balancing technical complexity, practical considerations, and adherence to motor learning and control principles in this system can enhance the application of VR interventions for post-stroke patients in both clinical and remote rehabilitation settings.

Background: The use of prediction tools in stroke rehabilitation research and clinical practice is increasing, but it is not clear whether these prediction tools out-perform clinician predictions. Objective: This study aimed to compare physiotherapist predictions for independent walking with the Time to Walking Independently after STroke (TWIST) prediction tool. Methods: Adults with new lower limb weakness and unable to walk independently (Functional Ambulation Category [FAC] < 4) were recruited. At 1 week post-stroke, the treating physiotherapist was asked to predict whether their patient would achieve independent walking by 4, 6, 9, 12, 16, or 26 weeks, or remain dependent. Predictions were also made using the TWIST prediction tool, but not shared. Binary logistic regressions were conducted with the time independent walking was achieved as the dependent variable and independent variables were the physiotherapist and TWIST predictions. Results: Ninety-one participants were included (median age 71 years, 36 [40%] female). Most participants (67 [74%]) were non-ambulatory (FAC = 0) at 1-week post-stroke. Thirty-seven physiotherapists were recruited. Physiotherapists made accurate predictions for time taken to achieve independent walking for 39 participants (43%). Prediction accuracy was not related to physiotherapist confidence or years of stroke-specific experience. TWIST out-performed physiotherapist predictions (Physiotherapists 76%-77%, TWIST 86%-88% accurate) for participants who achieved independent walking by 4, 6, and 9 weeks post-stroke. Accuracy of physiotherapist and TWIST predictions was similar for 16 and 26 weeks post-stroke. Conclusions: The TWIST prediction tool is more accurate than physiotherapists at predicting whether a patient will achieve independent walking by 4, 6, or 9 weeks post-stroke, but not for 16 or 26 weeks post-stroke. TWIST may be useful to inform early rehabilitation and discharge planning. Clinical Trial Registration-URL: www.anzctr.org.au Unique Identifier: ACTRN12617001434381.

Background: Noninvasive brain stimulation (NIBS) is a promising technique for improving upper limb motor performance post-stroke. Its application has been guided by the interhemispheric competition model and typically involves suppression of contralesional motor cortex. However, the bimodal balance recovery model prompts a more tailored application of NIBS based on ipsilesional corticomotor function. Objective: To review and assess the application of repetitive transcranial magnetic stimulation (rTMS) protocols that aimed to improve upper limb motor performance after stroke. Methods: A PubMed search was conducted for studies published between 1st January 2005 and 1st November 2022 using rTMS to improve upper limb motor performance of human adults after stroke. Studies were grouped according to whether facilitatory or suppressive rTMS was applied to the contralesional hemisphere. Results: Of the 492 studies identified, 70 were included in this review. Only 2 studies did not conform to the interhemispheric competition model, and facilitated the contralesional hemisphere. Only 21 out of 70 (30%) studies reported motor evoked potential (MEP) status as a biomarker of ipsilesional corticomotor function. Around half of the studies (37/70, 53%) checked whether rTMS had the expected effect by measuring corticomotor excitability (CME) after application. Conclusion: The interhemispheric competition model dominates the application of rTMS post-stroke. The majority of recent and current studies do not consider bimodal balance recovery model for the application of rTMS. Evaluating CME after the application rTMS could confirm that the intervention had the intended neurophysiological effect. Future studies could select patients and apply rTMS protocols based on ipsilesional MEP status.

Background: This observational study examined whether lower limb (LL) motor-evoked potentials (MEPs) 1 week post-stroke predict recovery of independent walking, use of ankle-foot orthosis (AFO) or walking aid, at 3 and 6 months post-stroke.Methods: Non-ambulatory participants were recruited 5 days post-stroke. Transcranial magnetic stimulation was used to determine tibialis anterior MEP status and clinical assessments (age, National Institutes of Health Stroke Scale (NIHSS), ankle dorsiflexion strength, LL motricity index, Berg Balance Test) were completed 1 week post-stroke. Functional Ambulation Category (FAC), use of AFO and walking aid were assessed 3 months and 6 months post-stroke. MEP status, alone and combined with clinical measures, and walking outcomes at 3 and 6 months were analysed with Pearson χ²and multivariate binary logistic regression.Results: Ninety participants were included (median age 72 years (38–97 years)). Most participants (81%) walked independently (FAC ≥ 4), 17% used an AFO, and 49% used a walking aid 3 months post-stroke with similar findings at 6 months. Independent walking was better predicted by age, LL strength and Berg Balance Test (accuracy 92%, 95% CI 85% to 97%) than MEP status (accuracy 73%, 95% CI 63% to 83%). AFO use was better predicted by NIHSS and MEP status (accuracy 88%, 95% CI 79% to 94%) than MEP status alone (accuracy 76%, 95% CI 65% to 84%). No variables predicted use of walking aids.Conclusions: The presence of LL MEPs 1-week post-stroke predicts independent walking at 3 and 6 months post-stroke. However, the absence of MEPs does not preclude independent walking. Clinical factors, particularly age, balance and stroke severity, more strongly predict independent walking than MEP status. LL MEP status adds little value as a biomarker for walking outcomes.

Background: The likelihood of regaining independent walking after stroke influences rehabilitation and hospital discharge planning. Objective: This study aimed to develop and internally validate a tool to predict whether and when a patient will walk independently in the first 6 months post-stroke. Methods: Adults with stroke were recruited if they had new lower limb weakness and were unable to walk independently. Clinical assessments were completed one week post-stroke. The primary outcome was time post-stroke by which independent walking (Functional Ambulation Category score ≥ 4) was achieved. Cox hazard regression identified predictors for achieving independent walking by 4, 6, 9, 16, or 26 weeks post-stroke. The cut-off and weighting for each predictor was determined using β-coefficients. Predictors were assigned a score and summed for a final TWIST score. The probability of achieving independent walking at each time point for each TWIST score was calculated. Results: We included 93 participants (36 women, median age 71 years). Age < 80 years, knee extension strength Medical Research Council grade ≥ 3/5, and Berg Balance Test < 6, 6 to 15, or ≥ 16/56, predicted independent walking and were combined to form the TWIST prediction tool. The TWIST prediction tool was at least 83% accurate for all time points. Conclusions: The TWIST tool combines routine bedside tests at one week post-stroke to accurately predict the probability of an individual patient achieving independent walking by 4, 6, 9, 16, or 26 weeks post-stroke. If externally validated, the TWIST prediction tool may benefit patients and clinicians by informing rehabilitation decisions and discharge planning.

Background. The PREP2 algorithm combines clinical and neurophysiological measures to predict upper-limb (UL) motor outcomes 3 months poststroke, using 4 prediction categories based on Action Research Arm Test (ARAT) scores. The algorithm was accurate at 3 months for 75% of participants in a previous validation study. Objective. This study aimed to evaluate whether PREP2 predictions made at baseline are correct 2 years poststroke. We also assessed whether patients’ UL performance remained stable, improved, or worsened between 3 months and 2 years after stroke. Methods. This is a follow-up study of 192 participants recruited and assessed in the original PREP2 validation study. Participants who completed assessments 3 months poststroke (n = 157) were invited to complete follow-up assessments at 2 years poststroke for the present study. UL outcomes were assessed with the ARAT, upper extremity Fugl-Meyer Scale, and Motor Activity Log. Results. A total of 86 participants completed 2-year follow-up assessments in this study. PREP2 predictions made at baseline were correct for 69/86 (80%) participants 2 years poststroke, and PREP2 UL outcome category was stable between 3 months and 2 years poststroke for 71/86 (83%). There was no difference in age, stroke severity, or comorbidities among patients whose category remained stable, improved, or deteriorated. Conclusions. PREP2 algorithm predictions made within days of stroke are correct at both 3 months and 2 years poststroke for most patients. Further investigation may be useful to identify which patients are likely to improve, remain stable, or deteriorate between 3 months and 2 years.