The relationship between stroke-related sarcopenia and 3-month neurological outcome in stroke patients: a prospective cohort study.

BACKGROUND: Sarcopenia was initially defined as an age-related decline in muscle mass and function. As sarcopenia has been observed at any age and secondary to many diseases, extending the classical meaning of sarcopenia beyond the age-related phenomenon to the 'disease-related' sarcopenia has been debated recently. Stroke-related sarcopenia occurs secondary to the brain injury and is age-independent, potentially worsening patients' neurological outcomes. We aimed to explore the relationship between stroke-related sarcopenia and 3-month neurological outcome in stroke patients, as well as to evaluate the predictive value of the two main indicators of stroke-related sarcopenia-muscle mass and muscle strength-for adverse neurological outcomes. METHODS: A single-centre prospective cohort study of 228 individuals with ischaemic stroke who were admitted for routine rehabilitation treatment, followed up for neurological outcome for 3 months following the onset of the stroke. The muscle mass was expressed by the ratio of appendicular skeletal muscle mass (ASM) to height (Ht) (ASM/Ht2), and the muscle strength was assessed by the grip strength. The neurological outcome was assessed using modified Rankin Scale (mRS) scores three months after the stroke. Statistical analyses, including correlation coefficient tests, multivariate regression analyses, and receiver operating characteristic (ROC) analyses, were applied in this study. RESULTS: The prevalence of sarcopenia was 28.9% in all included patients. Compared with the good outcome group, the poor outcome group showed a significantly lower level of ASM/Ht2, grip strength, and a higher level of mRS scores (all P < 0.05). Both ASM/Ht2 and grip strength were associated with mRS scores (r = -0.549 and - 0.592, respectively; both P < 0.01). Both ASM/Ht2 and grip strength were independent protective factors for the neurological outcome (OR = 0.525 [0.399-0.690] and 0.684 [0.605-0.773], respectively; both P < 0.01). The AUCs of grip strength and ASM/Ht2 were 0.884 and 0.833, respectively. The optimal cutoff values for grip strength to predict neurological outcome in the total patients, the male patients, and the female patients were 18.66, 26.69, and 18.66, respectively, with sensitivities of 79.10%, 90.48%, and 95.65% and specificities of 90.68%, 72.04%, and 77.94%, respectively. The optimal cutoff values for ASM/Ht2 to predict neurological outcome in the total population, the male patients, and the female patients were 6.01, 6.19, and 6.01, respectively, with sensitivities of 85.07%, 71.43%, and 93.48% and specificities of 73.91%, 86.02%, and 54.41%, respectively. CONCLUSIONS: Stroke-related sarcopenia is closely related to worse neurological function at the convalescence stage. Higher ASM/Ht2 and grip strength at baseline were independent protective factors for neurological outcomes. ASM/Ht2 and grip strength could be of superior predictive value for neurological outcomes, with different weights and gender differences in their predictive efficacy.