Examination of Rehabilitation Intensity According to Severity of Acute Stroke: A Retrospective Study

• Published in: Journal of stroke and cerebrovascular diseases 2021-09, Vol.30 (9), p.105994-105994, Article 105994
• Main Authors: Fujino, Yuji, Fukata, Kazuhiro, Inoue, Masahide, Okawa, Shinsuke, Okuma, Katsunobu, Kunieda, Yota, Miki, Hiroshi, Matsuda, Tadamitsu, Amimoto, Kazu, Makita, Shigeru, Takahashi, Hidetoshi, Fujiwara, Toshiyuki
• Format: Article
• Language: English
• Subjects: Acute; Intensity; Rehabilitation; Severity; Stroke
• ISSN: 1052-3057; 1532-8511
• DOI: 10.1016/j.jstrokecerebrovasdis.2021.105994

•This study used cluster analysis to classify stroke patients into three groups.•Neurological severity and trunk performance were significantly correlated.•Acute rehabilitation in the mild severity group showed a ceiling effect.•Rehabilitation of the moderate severity group was the most intensive and effective.•In the group with severe severity, rehabilitation had a limited effect. To investigate the intensity and effectiveness of rehabilitation in acute stroke patients according to the severity of functional impairments in them. This retrospective cohort study included 294 patients with acute hemispheric stroke admitted to three acute-care hospitals who subsequently underwent an inpatient rehabilitation program. Stroke severity was classified according to neurological deficits and trunk dysfunction. The following data were obtained from medical records: age, sex, stroke type, lesion side, hospitalization duration, initial functional status determined using the National Institutes of Health Stroke Scale, rehabilitation start date, first day out of bed after admission, total treatment duration, total number of treatment sessions, rehabilitation implementation rate between start of rehabilitation and discharge, trunk control test and Barthel Index score on the first day out of bed after admission and discharge, and post-discharge outcomes. Hierarchical cluster analysis was performed with clusters categorized using the National Institutes of Health Stroke Scale and trunk control test scores. Variables were compared using the Kruskal–Wallis test, and Dunn's nonparametric comparison test was performed for post-hoc analysis to determine differences between clusters. The National Institutes of Health Stroke Scale and trunk control test showed a significant correlation (r = −0.816, p