sonoda shigeru

BACKGROUND: Motor impairments not only lead to a significant reduction in patient activity levels but also trigger a further deterioration in motor function due to deconditioning, which is an issue that is particularly pronounced during hospitalization. This deconditioning can be countered by sustaining appropriate activity levels. Activities that occur outside of scheduled programs, often overlooked, are critical in this context. Wearable technology, such as smart clothing, provides a means to monitor these activities. OBJECTIVE: This study aimed to observe activity levels in patients who had strokes during the subacute phase, focusing on both scheduled training sessions and other nontraining times in an inpatient rehabilitation environment. A smart clothing system is used to simultaneously measure heart rate and acceleration, offering insights into both the amount and intensity of the physical activity. METHODS: In this preliminary cohort study, 11 individuals undergoing subacute stroke rehabilitation were enrolled. The 48-hour continuous measurement system, deployed at admission and reassessed 4 weeks later, monitored accelerometry data for physical activity (quantified with a moving SD of acceleration [MSDA]) and heart rate for intensity (quantified with percent heart rate reserve). The measurements were performed using a wearable activity monitoring system, the hitoe (NTT Corporation and Toray Industries, Inc) system comprising a measuring garment (wear or strap) with integrated electrodes, a data transmitter, and a smartphone. The Functional Independence Measure was used to assess the patients' daily activity levels. This study explored factors such as differences in activity during training and nontraining periods, correlations with activities of daily living (ADLs) and age, and changes observed after 4 weeks. RESULTS: A significant increase was found in the daily total MSDA after the 4-week program, with the average percent heart rate reserve remaining consistent. Physical activity during training positively correlated with ADL levels both at admission (ρ=0.86, P<.001) and 4 weeks post admission (ρ=0.96, P<.001), whereas the correlation between age and MSDA was not significant during training periods at admission (ρ=-0.41, P=.21) or 4 weeks post admission (ρ=-0.25, P=.45). Conversely, nontraining activity showed a negative correlation with age, with significant negative correlations with age at admission (ρ=-0.82, P=.002) and 4 weeks post admission (ρ=-0.73, P=.01). CONCLUSIONS: Inpatient rehabilitation activity levels were positively correlated with ADL levels. Further analysis revealed a strong positive correlation between scheduled training activities and ADL levels, whereas nontraining activities showed no such correlation. Instead, a negative correlation between nontraining activities and age was observed. These observations suggest the importance of providing activity opportunities for older patients, while it may also suggest the need for adjusting the activity amount to accommodate the potentially limited fitness levels of this demographic. Future studies with larger patient groups are warranted to validate and further elucidate these findings.

Practicing walking in a safety suspension device allows patients to move freely and without excessive reliance on a therapist, which requires correcting errors and may facilitate motor learning. This opens the possibility that patients with subacute stroke may improve their walking ability more rapidly. Therefore, we tested the hypothesis that overground gait training in a safety suspension device will result in achieving faster supervision-level walking than gait training without the suspension device. Twenty-seven patients with stroke admitted to the rehabilitation ward with functional ambulation categories (FAC) score of 2 at admission were randomly allocated to safety suspension-device group (SS group) or conventional assisted-gait training group (control group). In addition to regular physical therapy, each group underwent additional gait training for 60 min a day, 5 days a week for 4 weeks. We counted the days until reaching a FAC score of 3 and assessed the probability using Cox regression models. The median days required to reach a FAC score of 3 were 7 days for the SS group and 17.5 days for the control group, which was significantly different between the groups (P < 0.05). The SS group had a higher probability of reaching a FAC score of 3 after adjusting for age and admission motor impairment (hazard ratio = 3.61, 95% confidence interval = 1.40-9.33, P < 0.01). The gait training with a safety suspension device accelerates reaching the supervision-level walking during inpatient rehabilitation. We speculate that a safety suspension device facilitated learning by allowing errors to be experienced and correct in a safe environment.

BACKGROUND: There is an extensive library of language tests, each with excellent psychometric properties; however, many of the tests available take considerable administration time, possibly bearing psychological strain on patients. The Short and Tailored Evaluation of Language Ability (STELA) is a simplified, tablet-based language ability assessment system developed to address this issue, with a reduced number of items and automated testing process. OBJECTIVE: The aim of this paper is to assess the administration time, internal consistency, and validity of the STELA. METHODS: The STELA consists of a tablet app, a microphone, and an input keypad for clinician's use. The system is designed to assess language ability with 52 questions grouped into 2 comprehension modalities (auditory comprehension and reading comprehension) and 3 expression modalities (naming and sentence formation, repetition, and reading aloud). Performance in each modality was scored as the correct answer rate (0-100), and overall performance expressed as the sum of modality scores (out of 500 points). RESULTS: The time taken to complete the STELA was significantly less than the time for the WAB (mean 16.2, SD 9.4 vs mean 149.3, SD 64.1 minutes; P<.001). The STELA's total score was strongly correlated with the WAB Aphasia Quotient (r=0.93, P<.001), supporting the former's concurrent validity concerning the WAB, which is a gold-standard aphasia assessment. Strong correlations were also observed at the subscale level; STELA auditory comprehension versus WAB auditory comprehension (r=0.75, P<.001), STELA repetition versus WAB repetition (r=0.96, P<.001), STELA naming and sentence formation versus WAB naming and word finding (r=0.81, P<.001), and the sum of STELA reading comprehension or reading aloud versus WAB reading (r=0.82, P<.001). Cronbach α obtained for each modality was .862 for auditory comprehension, .872 for reading comprehension, .902 for naming and sentence formation, .787 for repetition, and .892 for reading aloud. Global Cronbach α was .961. The average of the values of item-total correlation to each subscale was 0.61 (SD 0.17). CONCLUSIONS: Our study confirmed significant time reduction in the assessment of language ability and provided evidence for good internal consistency and validity of the STELA tablet-based aphasia assessment system.

Healthcare-associated COVID-19 among vulnerable patients leads to disproportionate morbidity and mortality. Early pharmacologic intervention may reduce negative sequelae and improve survival in such settings. This study aimed to describe outcome of patients with healthcare-associated COVID-19 who received early short-course remdesivir therapy. We reviewed the characteristics and outcome of hospitalized patients who developed COVID-19 during an outbreak that involved two wards at a non-acute care hospital in Japan and received short-course remdesivir. Forty-nine patients were diagnosed with COVID-19, 34 on a comprehensive inpatient rehabilitation ward and 15 on a combined palliative care and internal medicine ward. Forty-seven were symptomatic and 46 of them received remdesivir. The median age was 75, and the median Charlson comorbidity index was 6 among those who received it. Forty-one patients had received one or two doses of mRNA vaccines, while none had received a third dose. Most patients received 3 days of remdesivir. Of the patients followed up to 14 and 28 days from onset, 41/44 (95.3%) and 35/41(85.4%) were alive, respectively. Six deaths occurred by 28 days in the palliative care/internal medicine ward and two of them were possibly related to COVID-19. Among those who survived, the performance status was unchanged between the time of onset and at 28 days.

OBJECTIVES: In stroke patients, the assessment of gait ability over time is important. For quantitative gait assessment using measuring devices, the walking speed condition for measurement is generally based on the patient's preferred walking speed or the maximum walking speed at the time of measurement. However, because walking speed often increases during the convalescent stage, understanding the effects of change in walking speed on gait when comparing the course of recovery is necessary. Although several previous studies have reported the effects of change in walking speed on gait in stroke patients, the time-distance parameters described in these reports may not be generalizable because of the small case numbers. Therefore, we measured treadmill gait at the preferred walking speed (PWS) and 1.3 times the PWS (130% PWS) in 43 post-stroke hemiplegic patients and analyzed the effects of change in walking speed on time-distance parameters. METHODS: Forty-three patients with hemiplegia after a first stroke, who were able to walk on a treadmill under supervision, were recruited as subjects. Using a three-dimensional motion analysis system, treadmill gait was assessed under two conditions: PWS and 130% PWS. The primary outcome measures were the time-distance parameters, which were compared between the PWS and 130% PWS conditions. RESULTS: Cadence, stride length, and step length of the affected and unaffected lower limbs increased significantly at 130% PWS compared with at PWS. In terms of actual time, single stance time and initial and terminal double stance time in both affected and unaffected limbs decreased significantly at 130% PWS. In terms of relative time (% of the gait cycle), compared with PWS, relative single stance time increased significantly, whereas relative initial and terminal double stance times decreased significantly at 130% PWS in both the affected and unaffected limbs. CONCLUSIONS: This study on treadmill gait in patients with hemiplegia after a first stroke confirmed the effects of change in walking speed on time-distance parameters. Our results will help in the interpretation of time-distance parameters measured under different walking speed conditions.

BACKGROUND: The number of wheelchair users is increasing as the population ages. However, there is still lack of tools to objectively assess posture during wheelchair propulsion. OBJECTIVE: The purpose of this study was to develop a system to measure trunk and pelvis ROM (range of motion) while driving a wheelchair and to assess the reproducibility of the measured posture changes and cyclic fluctuations during propulsion. METHODS: Motion recorders were attached to the sternum and pelvis of fifteen healthy adults, and a pressure sensor was attached to the right heel. Subjects drove a standard wheelchair using their right leg and trunk and pelvis ROM was measured in two separate sessions. To detect the trend of postural changes during propulsion, a trend curve was computed from the measured ROM using a low-pass filter and then subtracted from the measured ROM waveforms. The resulting curves were normalized and averaged, and the maximum, minimum, and amplitude of the cyclic variation during propulsion were assessed for reproducibility. RESULTS: Intraclass correlation coefficients of the maximum, minimum, and amplitude ranged from 0.65 to 0.84. CONCLUSIONS: The proposed method can estimate the posture change and the periodic fluctuation during wheelchair propulsion with high reproducibility.

OBJECTIVES: Falls can occur daily in stroke patients and appropriate independence assessments for fall prevention are required. Although previous studies evaluated the short physical performance battery (SPPB) in stroke patients, the relationship between SPPB and fall prediction and walking independence remains unclear. Therefore, we aimed to verify whether SPPB is a predictor of walking independence. MATERIALS AND METHODS: The present study included 105 hemiplegic stroke patients who were admitted to the rehabilitation ward and gave consent to participate. Cross-sectional physical function and functional independence measure cognitive (FIM-C) evaluation were conducted in hemiplegic stroke patients. Logistic regression analysis using the increasing variable method (likelihood ratio) was performed to extract factors for walking independence. Cutoff values were calculated for the extracted items using the receiver operating-characteristic (ROC) curve. RESULTS: Among 86 participants included in the final analysis, 36 were independent walkers and 50 were dependent walkers. In the logistic regression analysis, SPPB and FIM-C were extracted as factors for walking independence. The cutoff value was 7 [area under the curve (AUC), 0.94; sensitivity, 0.83; specificity, 0.73)] for SPPB and 32 (AUC, 0.83; sensitivity, 0.69; specificity, 0.57) for FIM-C in ROC analysis CONCLUSIONS: SPPB and FIM-C were extracted as factors for walking independence. Although SPPB alone cannot determine independent walking, combined assessment of SPPB with cognitive function may enable more accurate determination of walking independence.

<jats:p>Background. The difference in the effects of combined therapy with repetitive facilitative exercise (RFE) and neuromuscular electrical stimulation (NMES) on stroke upper limb paralysis was only reported by a pilot study; it has not been investigated in many patients. Objective. We investigated the effect of combined therapy with RFE and NMES on stroke patients with severe upper paresis. Methods. This study included 99 of the very severe paresis stroke patients with scores of zero or 1a in the Finger-Function test of the Stroke Impairment Assessment Set (SIAS). We randomly divided the patients into four groups, namely, NMES, RFE, RFE under NMES, and conventional training (CT) groups. A total of 20 min of group-specific training in addition to 40 min of conventional exercise per day, seven times a week for 4 weeks after admission, was performed. The upper extremity items of the Fugl-Meyer Assessment (FMA) were evaluated before and after the training period. Results. The total score gains of the FMA, FMA wrist item, and FMA finger item were significantly larger in the RFE under NMES group than those in the CT group (<jats:inline-formula> <math xmlns="http://www.w3.org/1998/Math/MathML" id="M1"> <mi>p</mi> <mo>&lt;</mo> <mn>0.05</mn> </math> </jats:inline-formula>). Conclusion. The combination of voluntary movement and electrical stimulation may promote the activation of paralyzed muscles and improve distal function for very severe paralyzed upper limbs.</jats:p>

OBJECTIVE: To examine the effect of onset to admission interval (OAI) and stroke type on activities of daily living (ADL) outcome. MATERIALS AND METHODS: Stroke patients (n=3112) admitted to and discharged from comprehensive inpatient rehabilitation wards at Nanakuri Memorial Hospital were classified into 8 OAI segments and by stroke type [intracerebral hemorrhage (ICH) and cerebral infarction (CI)]. Motor subscore of the Functional Independence Measure (FIM-M) on admission, FIM-M at discharge, FIM-M gain, length of stay (LOS), and FIM-M efficiency in the ICH and CI group matched by OAI segment were compared using the Wilcoxon test. Multiple comparisons using the Steel-Dwass test of FIM-M on admission, FIM-M at discharge, FIM-M gain, LOS, and FIM-M efficiency by OAI segments were performed. RESULTS: FIM-M on admission was lower in the ICH group than the CI group in matched OAI segments. However, FIM-M improvement was greater in the ICH group than the CI group, resulting in no difference in FIM-M between groups at discharge. In both groups, the longer the OAI, the lower the FIM-M on admission and at discharge. The distribution pattern of significant differences among OAI segments differed between the groups. LOS tended to be longer and FIM-M efficiency tended to be higher in the ICH group than in the CI group. CONCLUSIONS: The brain mass effect at the time of admission was larger and took longer to decrease in the ICH group than in the CI group. These results may improve prediction of outcomes in comprehensive inpatient rehabilitation wards.

BACKGROUND: The International Classification of Functioning, Disability, and Health (ICF) Generic-30 (Rehabilitation) Set is a tool used to assess the functioning of a clinical population in rehabilitation. The ICF Generic-30 consists of nine ICF categories from the component "body functions" and 21 from the component "activities and participation". This study aimed to develop a rating reference guide for the nine body function categories of the ICF Generic-30 Set using a predefined, structured process and to examine the interrater reliability of the ratings using the rating reference guide. METHODS: The development of the first version of the rating reference guide involved the following steps: (1) a trial of rating patients by several raters; (2) cognitive interviews with each rater to analyze the thought process involved in each rating; (3) the drafting of the rating reference guide by a multidisciplinary panel; and (4) a review by ICF specialists to confirm consistency with the ICF. Subsequently, we conducted a first field test to gain insight into the use of the guide in practice. The reference guide was modified based on the raters' feedback in the field test, and an inter-rater reliability test was conducted thereafter. Interrater agreement was evaluated using weighted kappa statistics with linear weights. RESULTS: The first version of the rating reference guide was successfully developed and tested. The weighted kappa coefficient in the field testing ranged from 0.25 to 0.92. The interrater reliability testing of the rating reference guide modified based on the field test results yielded an improved weighted kappa coefficient ranging from 0.53 to 0.78. Relative improvements in the weighted kappa coefficients were observed in seven out of the nine categories. Consequently, seven out of nine categories were found to have a weighted kappa coefficient of 0.61 or higher. CONCLUSIONS: In this study, we developed and modified a rating reference guide for the body function categories of the ICF Generic-30 Set. The interrater reliability test using the final version of the rating reference guide showed moderate to substantial interrater agreement, which encouraged the use of the ICF in rehabilitation practice.

発病から15年の慢性脳血管障害者で介護保険利用の14名と非利用の14名を対象とし、能力障害(ADL)と社会参加障害(IADL)の推移を調査した。内訳は男性20名、女性8名、年齢は67.0歳±9.7であり、診断名は脳出血10名、脳梗塞16名、その他2名で、麻痺11例、片麻痺17例であった。発病から15年間のADLとIADLを追跡調査した。その結果、介護保険利用者と非利用者におけるADLの推移に特徴がみられた。本研究から抽出された地域包括ケアシステムの課題として、ADLの自立支援や介護予防、就労支援で適切な介入時期を個々に発見すること、生活期リハビリテーションにおける訪問リハビリテーションやデイケアで適切な介入プログラムを設定して社会参加を促すプランを作り、積極的に解決に向けて粘り強く進めること、があげられた。

<p>Background and Purpose: The aim of this study was to clarify the relationship between the characteristics of aphasia and hematoma type in patients with left putaminal hemorrhage who were admitted to comprehensive inpatient rehabilitation wards. Methods: Age, gender period from onset to hospitalization, hematoma type and volume, neurological symptoms, and language functions were retrospectively investigated from the medical records of 92 patients. We examined the relationship between hematoma type and score of the Standard Language Test of Aphasia (SLTA) and its subitems of writing, speech, and comprehension. Results: Aphasia in patients with type I hematoma was mild and that in patients with types IV and V hematomas was severe. There was no difference in the severity of aphasia between types II and III hematomas. There was only a decrease in the writing score in type I hematoma. Writing and speech were impaired in type IV hematoma. Type V hematoma had lower scores for all modalities. Conclusion: The effect of hematoma volume on aphasia severity and all language modality by hematoma type was significant. We suggested that white matter damage caused by hematoma extension may be involved in aphasia severity.</p>

OBJECTIVE: Although studies on the efficacy of the rehabilitation robot are increasing, there are few reports using the robot for gait training in the actual clinical setting. This study aimed to investigate the effectiveness of gait training using Welwalk in hemiparetic stroke patients in a real clinical setting. MATERIALS AND METHODS: This prospective study included 36 hemiparetic stroke patients who underwent gait training using Welwalk. We examined the walking ability improvement efficiency using Functional Independence Measure (FIM)-walk as the primary outcome, which was compared with that of 36 patients (matched control group) who underwent conventional rehabilitation. Other outcomes were the actual gait training period using Welwalk, raw FIM-walk score, lower extremity motor functions score in Stroke Impairment Assessment Set at discharge, and duration from stroke onset until discharge. RESULTS: The improvement efficiency of the FIM-walk was significantly higher in the Welwalk group than in the matched control group (control 0.48 ± 0.31, Welwalk 0.80 ± 0.38, p-value < 0.001). The mean gait training period using Welwalk was 5 weeks. No significant differences were found in other outcomes between the Welwalk group and the matched control group. CONCLUSION: This study demonstrated the effectiveness of gait training using Welwalk on the improvement efficiency of the FIM-walk in hemiparetic stroke patients in an actual clinical setting.

在宅慢性脳卒中者における心身機能維持・改善のためのリハビリテーション(リハ)練習について、その目的や練習結果等について発病から10年間にわたり在宅調査を行った。初発脳卒中で発病半年以内に入院しリハ医療を行った40歳以上の脳卒中者125名のうち、発病10年時に在宅調査が可能であった36名(男性27名、女性9名)を対象とした。本調査は前向きコホート研究であり、同一調査員が一貫して調査し、原則在宅訪問で行った。心身機能の10年間の推移を調べたところ、入院中にADL(FIM)は有意に改善し、退院後はほぼ維持されていた。IADL(FAI)の改善は乏しく、低値で推移した。リハ練習(自主トレーニングを含む)は、36人全員が「必要」としていたが、発病2年時から「必要なし」が2人から4人とでてきた。リハ練習意欲は「十分にある」が減り、「少しある」が増え、「全くない」が発病4年時から1人、2人と出てきたが、10年時では少数にとどまった。リハ練習は、1年時は全員が行っていたが、「十分に行っている」が年々減り続け、逆に「行っていない」が徐々に増えてきた。その理由は、「良くなったから」が増える反面、「諦めた」および「面倒くさい」が増え、「その他」として仕事をあげた人が多かった。自主リハ練習は、発病1年時で83%が行っていたが、発病10年時では75%に減少した。リハ練習の結果、主観的な改善度は、発病1年時では66%が改善、34%が不変と答えていたが、10年時には、改善が30%、不変が59%、悪化が11%と答えた。

BACKGROUND: The World Health Organization developed the International Classification of Functioning, Disability, and Health (ICF) in 2001 and has been in the process of implementing it in clinics since then. Current international efforts to implement ICF in rehabilitation clinics include the implementation of ICF Core Sets and the development of simple, intuitive descriptions for the ICF Generic-30 Set (also called Rehabilitation Set). The present study was designed to operationalize these ICF tools for clinical practice in Japan. This work included 1) the development of the Japanese version of the simple, intuitive descriptions for the ICF Generic-30 Set, 2) the development of a rating reference guide for Activity and Participation categories, and 3) the examination of the interrater reliability of rating Activity and Participation categories. METHODS: The Japanese version of the simple, intuitive descriptions for the ICF Generic-30 Set was developed following the process employed to develop the Chinese and Italian versions. For further operationalization of this ICF Set in practice, a rating reference guide was developed. The development of the rating reference guide involved the following steps: 1) a trial of rating patients by several raters, 2) cognitive interviewing of the raters to analyse the thinking process involved in rating, 3) drafting of the rating reference guide, and 4) review by ICF specialists to confirm consistency with the original ICF concepts. After the rating reference guide was developed, interrater reliability of the rating with the reference guide was determined. Interrater reliability was examined using weighted kappa statistics with linear weight. RESULTS: Through the pre-defined process, the Japanese version of the simple, intuitive descriptions for 30 categories of the ICF Generic-30 Set and the rating reference guides for 21 Activity and Participation categories were successfully developed. The weighted kappa statistics ranged from 0.61 to 0.85, showing substantial to excellent agreement of the ratings between raters. CONCLUSIONS: The present study demonstrates that ICF categories can be translated into clinical practice. Collaboration between clinicians and researchers would further enhance the implementation of the ICF in Japan.

BACKGROUND: In stroke rehabilitation, the most important concern of the patients and their families is whether the patients can walk independently and whether they need braces after discharge. AIM: This study aims to investigate the relationship between several types of putaminal hemorrhage and walking independence and orthotic therapy in patients with hemiplegia. DESIGN: Observational study. SETTING: Inpatients rehabilitation department, Fujita Health University Nanakuri Memorial Hospital, Japan. POPULATION: Total 264 patients with putaminal hemorrhage admitted to our hospital. METHODS: Neurological and cognitive functions were examined as per the stroke scale of the National Institutes of Health and the Mini-mental state examination, respectively. The hematomas were classified into five types, and the volume was measured using computed tomography (CT). Walking ability was evaluated by Functional Ambulation Category (FAC), and walking independence was defined as FAC ≥4. The relationship between the types of hematomas and walking independence and orthotic therapy in patients with hemiplegia with putaminal hemorrhage was also analyzed. RESULTS: We observed differences within the hematoma types in volume, neurological symptoms, and cognitive function but not in age, sex, and lesion side aspects of these patients-143 of whom could walk independently (FAC≥4) and 121 non-independently. Walking independently and the need for orthosis were closely related to the type of hematoma. CONCLUSIONS: CT imaging at stroke onset can provide useful information when examining walking independence and indicate necessity for an orthosis at the time of discharge to the rehabilitation ward. CLINICAL REHABILITATION IMPACT: This study might help to better understand the role of neuroimaging in stroke rehabilitation.

慢性脳卒中患者の総合的追跡調査から、退院時のHealth Locus of Control(HLC)と退院後の在宅での発病5年までの保健行動との関連性について検討した。脳卒中で入院した211名のうち、在宅で発病5年までの保健行動が聴取できた36名(男性30人、女性6人、発病時平均57.8±8.4歳)を対象とした。退院時のHLCの評価得点をもとにInternals群17名(11〜21点、I群)、Moderate群6名(中間値22点、M群)、Externals群13名(23〜33点、E群)の3群に分けた。M群ではQOLで有意な向上が示され、I群では脳卒中後の機能の改善と手の実用度が有意に改善、E群では手の実用度のみ有意に改善していた。保健行動の内訳に関して、「外来リハビリテーション」の利用はどの群でも80%以上を示していたが、「自主練習」ではI群が10人(77%)で、E群9人(59%)を上回っていた。また、「食事への配慮」はI群では13人全員が行っていたが、E群では12人(70%)にとどまっていた。

PURPOSE: This trial aimed to validate the effectiveness of using the Gait Exercise Assist Robot (GEAR) in patients with hemiplegia after primary stroke. METHODS: The study design was open-label randomized controlled trial. Twenty-six patients with hemiplegia after primary stroke admitted to the comprehensive inpatient rehabilitation wards were enrolled and randomized to a group using GEAR in gait training and a control group. The intervention period was 4 weeks. Evaluations were conducted at admission, during intervention period, 8 weeks from start of intervention, and at discharge. Primary outcome measure was improvement efficiency of Functional Independence Measure (FIM)-walk score (FIM-walk improvement efficiency) that was calculated at the time of achieving FIM-walk score 5 (supervision level) during the intervention period or as weekly gain in FIM-walk score during 4 weeks for those who did not achieve score 5. RESULTS: FIM-walk improvement efficiency was .7 ± .4 in GEAR group and .4 ± .3 in control group, and was significantly higher in GEAR group (P = .01). The FIM-walk score gain after 4 weeks was significantly higher in the GEAR group (P = .01), but there were no significant differences between 2 groups after 8 weeks and at discharge. CONCLUSIONS: Gait training using GEAR for 4 weeks improved walking ability of subacute stroke patients. GEAR contributes to early improvement of walking ability probably by the knee flexion assist during swing phase on the paralyzed side thereby increasing the volume of training, and by the finely adjustable stance/swing assist mechanism for the paralyzed limb which optimizes the training difficulty level.