Association between nutritional status and food texture levels in older patients with stroke-related sarcopenia in the subacute phase: A retrospective cross-sectional study

Momoko SakuraiNorikazu Hishikawa^*Koshiro SAWADASuzuyo OhashiHiroshi MaedaYasuo Mikami

• Kyoto Furitsu Ika Daigaku Daigakuin Igaku Kenkyuka Rehabilitation Igaku Kyoshitsu, Kyoto, Japan

Background: Stroke-related sarcopenia (SRS) frequently emerges in the early phase of stroke and is associated with poor functional recovery and prolonged hospitalization. Identifying modifiable risk factors, such as nutritional status and food texture, may be important for early treatment in SRS. This study aimed to investigate the association between SRS and malnutrition diagnosed using the Global Leadership Initiative on Malnutrition (GLIM) criteria, with the primary aim of examining how the prevalence of SRS and malnutrition and the individual GLIM components vary among food texture levels. Methods: This study included 340 older adults (median age: 77.0 [66.0–83.0] years; median time since stroke onset: 22.0 [16.8–31.3] days) who were admitted to a convalescent rehabilitation ward during the subacute phase of stroke. SRS was diagnosed based on the Asian Working Group for Sarcopenia criteria, while malnutrition according to the GLIM framework. Food texture was categorized into three levels: standard, texture-modified, and tube feeding. Multivariate logistic regression was used to examine the association between SRS and GLIM-defined malnutrition, adjusting for relevant covariates. The Cochran–Armitage trend test assessed trends in the prevalence of SRS and malnutrition and in the proportions of individual GLIM components among food texture levels. Results: The prevalence of SRS and GLIM-defined malnutrition was 56.8% and 51.5%, respectively. Malnutrition was independently associated with SRS (odds ratio = 3.00; 95% confidence interval: 1.70–5.30; p < 0.001). The prevalence of both conditions increased progressively with more restrictive food textures (p for trend = 0.002 and < 0.001, respectively). Additionally, among the GLIM components, the proportions of patients with low body mass index, reduced muscle mass, and disease burden/inflammation increased with food texture restriction (q for trend = 0.001, < 0.001, and 0.007, respectively, after adjustment using the Benjamini–Hochberg false discovery rate correction). Conclusions: Older adults in the subacute phase of stroke who consume more restrictive food textures may be more prone to malnutrition, potentially due to stroke-related inflammation, which in turn may contribute to the development of SRS. Early tailored nutritional treatments that consider food texture restrictions and disease burden may help prevent SRS and enhance functional recovery in post-stroke rehabilitation.