Co-existence of Possible Sarcopenia and Dysphagia is Associated with Poor Functional Outcomes in Patients with Acute Stroke: A 1-Year Follow-up Study
Article information
Abstract
Background
This study aimed to clarify the association between the co-existence of possible sarcopenia (PS) and dysphagia in patients with acute stroke and the functional outcomes 12 months after stroke.
Methods
This multicenter retrospective cohort study was conducted on patients with stroke admitted to two acute care hospitals in Japan between November 2020 and October 2023. PS was defined based on grip strength (males <28 kg, females <18 kg) and calf circumference (males <34 cm, females <33 cm) on admission. Dysphagia was defined as a Functional Oral Intake Scale score of less than 4. The outcome was defined as a composite of either poor functional outcome (modified Rankin Scale [mRS] score of 3–6) or failure to recover to the premorbid mRS at 12 months after stroke onset. Modified Poisson regression analysis was used to examine the association between the co-existence of PS and dysphagia and poor functional outcomes.
Results
This study included 604 patients (median age 76 years; 367 men). Compared with the non-PS and non-dysphagia group, the PS and dysphagia group had a higher proportion of poor functional outcomes. Subsequently, modified Poisson regression analysis was performed. The interaction between PS and dysphagia was significantly associated with poor functional outcomes (risk ratio=4.348, 95% confidence interval 2.683–7.046).
Conclusion
In patients with acute stroke, the co-existence of PS and dysphagia was associated with poor functional outcomes 12 months after stroke. Future multicenter prospective interventional studies are required to clarify the effectiveness of multifaceted intervention programs in high-risk patients.
INTRODUCTION
Advances in stroke treatment have reduced post-stroke mortality.1) However, stroke remains a leading cause of functional disability worldwide, imposing significant burdens on the affected individuals, caregivers, and healthcare systems.2,3) Furthermore, functional impairments after stroke often lead to reduced quality of life, increased care dependency, and higher institutionalization rates.4-6) Therefore, understanding the factors that negatively affect long-term functional outcomes after stroke is essential for optimal patient recovery and improvement of overall societal health.
Sarcopenia profoundly affects post-stroke recovery.7,8) Characterized by the systemic loss of skeletal muscle mass and strength, sarcopenia is frequently observed in patients with stroke.9) Its prevalence increases from onset through the acute phase to the recovery phase.10) Previous studies have reported an association between sarcopenia and poor outcomes in patients with stroke,7,11) making it an important factor in preventing severe disability in these patients.
Dysphagia also markedly affects the disability trajectory of patients with stroke. It can lead to serious complications such as aspiration pneumonia, malnutrition, and dehydration, which compromise overall health and rehabilitation outcomes.12,13) The profound relationship between sarcopenia and dysphagia is increasingly recognized not only in patients with stroke but also in various older and chronically ill populations.14-17) For example, sarcopenia-related muscle weakness and reduced physical activity can exacerbate swallowing difficulties, while dysphagia-induced malnutrition may conversely accelerate muscle loss.15,18) Individually, both sarcopenia and dysphagia are established risk factors for various adverse outcomes, including increased mortality, prolonged hospital stays, and poorer functional recovery in patients with stroke and other medical conditions.8,19-21) However, in the context of post-stroke recovery, while sarcopenia and dysphagia are individually recognized as risk factors for poor outcomes, the interactive effects of their co-existence and subsequent impact on functional prognosis have not been sufficiently investigated. The possibility that these risk factors not only have additive effects but also interact to exert adverse effects has not yet been investigated. A detailed understanding of how the co-existence of sarcopenia and dysphagia in the acute phase of stroke influences chronic functional outcomes 12 months after onset is indispensable for developing appropriate intervention strategies for high-risk individuals. Therefore, this study aimed to investigate the association between the co-existence of sarcopenia and dysphagia in patients with acute stroke and their functional outcomes 12 months after stroke onset.
MATERIALS AND METHODS
Participants and Setting
This retrospective cohort study was conducted at two acute care hospitals in Niigata and Hyogo, Japan. The inclusion criteria were admission between November 2020 and October 2023 with cerebral infarction or hemorrhage within 48 hours of onset. The presence of stroke was confirmed in all enrolled patients by using computed tomography or magnetic resonance imaging and diagnosis by a medical doctor. The exclusion criteria were missing data, a premorbid modified Rankin Scale (mRS) score of 5, presence of coronavirus disease 2019 (COVID-19) at the time of admission, and refusal to participate in the study. At the hospitals where this study was conducted, a postal survey at 12 months is routinely performed as part of usual clinical care, and the data had already been collected at the time the study was conceived, representing secondary use of routine clinical data.
Rehabilitation during Hospitalization
The rehabilitation program (up to 3 hr/day) was tailored to accommodate individual patients’ functional abilities and disabilities, such as paralyzed limb facilitation, range of motion exercises, basic movement training (mainly for the legs), walking training, resistance training, aerobic exercises using an ergometer aimed at improving endurance, activities of daily living training, and dysphagia rehabilitation.22) These interventions followed the Japan Stroke Society Guidelines for the Treatment of Stroke.23) There was no difference in this rehabilitation concept between the two hospitals included in the study.
Data Collection
Data regarding age, sex, body mass index (BMI), stroke type, lesion side, stroke severity (National Institutes of Health Stroke Scale [NIHSS] score), comorbidities, primary treatment, premorbid and discharge mRS scores, and length of hospital stay were collected from medical records.
Measurement of Possible Sarcopenia
Possible sarcopenia (PS) was defined according to the Asian Working Group for Sarcopenia (AWGS) 2019 criteria, by using calf circumference for case finding (males <34 cm, females <33 cm) and low grip strength for case identification (males <28 kg, females <18 kg).24) Grip strength on the non-paretic side was measured twice by using a Smedley-type hand dynamometer (TKK 5401; Takei Scientific Instruments Co., Ltd., Tokyo, Japan), and the maximum value was recorded. In cases of bilateral paresis, the maximum values from both sides were used. Patients who were unable to undergo grip strength measurements because of impaired consciousness were excluded from the analysis because of missing data. Calf circumference was measured to the nearest 0.1 cm by using a tape measure. This measurement was performed on the non-paretic leg, with the patient in the supine position and the knee flexed to 90° and the ankle relaxed. If no paresis was present, the measurement was performed on both sides, and the maximum value was recorded. Patients with pitting edema were excluded from the analysis because of missing data. All measurements were performed by experienced physical therapists trained in standardized protocols and conducted within 5 days of admission, considering acute-phase treatment.
Assessment of Dysphagia
Swallowing function was assessed using the Functional Oral Intake Scale (FOIS),25) following an initial dysphagia screening performed by speech-language pathologists. The FOIS is a clinical tool used to evaluate a patient’s ability to consume food and liquids. Scores of 1–3 indicate dependence on tube feeding, ranging from no oral intake to minimal oral intake with supplemental tube feeding. Scores of 4–6 reflect partial oral intake with varying dietary restrictions, whereas a score of 7 represents full oral intake without restrictions and is considered to indicate functional normalcy. Based on a previous study,26) dysphagia in this study was defined as a FOIS score of less than 4, indicating a tube feeding-dependent state. A tube feeding–dependent state represents a criterion that objectively identifies severe dysphagia requiring intensive interventions during acute management. Because previous studies26,27) have defined post-stroke dysphagia as a FOIS score of less than 4, the same definition was adopted in this study to ensure comparability with existing evidence. This assessment was conducted within 5 days of admission, considering acute-phase treatment.
Outcomes
The outcome was defined as a composite of either poor functional outcome (mRS score 3–6) or failure to recover to the premorbid mRS at 12 months after stroke onset.28) The mRS score at 12 months was assessed using a mailed self-administered questionnaire or medical record review. The mRS score at admission was assessed by a stroke-trained physician experienced in mRS application according to the following scale, where a score of 0, 1, 2, 3, 4, 5 and 6. Scores of 0 to 6 indicate the following conditions: 0, no symptoms; 1, symptoms are present but there is no significant disability and the patient is able to carry out all usual duties and activities; 2, slight disability, with inability to carry out all previous activities but able to look after own affairs without assistance; 3, moderate disability, requiring some help but able to walk without assistance; 4, moderately severe disability, unable to walk without assistance and unable to attend to own bodily needs without assistance; 5, severe disability, bedridden, incontinent, and requiring constant nursing care and attention; and 6, death.29)
Additionally, the number of adverse events was investigated in a supplementary analysis. These included major adverse cardiovascular events (MACEs), defined as rehospitalization or death due to stroke or cardiovascular disease up to 12 months after onset, and non-MACEs. For non-MACEs, information on pneumonia, other medical conditions, falls, and other relevant events was collected. Furthermore, data on all-cause mortality encompassing both MACE and non-MACEs were collected. Outcomes were monitored via mailed, self-administered questionnaires or through a retrospective review of medical records.
Ethics Approval
This retrospective cohort study was approved by the ethics committee of the study center (Approval No. E2024000601) and conducted in accordance with the Declaration of Helsinki. The requirement for informed written consent was waived because of the retrospective design and the use of anonymized data. Instead, the participants were offered the opportunity to withdraw from the study at any time.
Statistical Analysis
Continuous and categorical data are presented as medians (interquartile range) and number (%) of individuals, respectively. The patients were categorized into four groups based on the combination of the presence or absence of PS and dysphagia: non-PS and non-dysphagia, non-PS and dysphagia, PS and non-dysphagia, and PS and dysphagia groups. Analysis of variance and chi-square tests were used for comparisons between the groups. To assess the potential for attrition bias in this study, clinical characteristics were compared between patients included in the final analysis and those lost to follow-up using the Mann–Whitney U test and the chi-square test. A modified Poisson regression analysis was conducted with the combinations of PS and dysphagia as independent variables and poor functional outcome as the dependent variable to examine the association between each combination and the outcome. The covariates included age, sex, BMI, NIHSS score, stroke type, hypertension, previous stroke, diabetes mellitus, dyslipidemia, atrial fibrillation, heart failure, and premorbid disability (premorbid mRS score of >2). These variables were selected based on previous studies reporting stroke outcomes, considering their clinical relevance.28,30-32) The p-value for trend was obtained by incorporating an ordinal variable (coded 1–4 according to PS and dysphagia status) into the fully adjusted modified Poisson regression model including all covariates. Furthermore, a modified Poisson regression analysis, including dysphagia, PS, and their interaction terms (PS×dysphagia), was performed simultaneously to evaluate the interaction between sarcopenia and dysphagia. This analysis specifically aimed to determine whether the combined effect of PS and dysphagia on poor functional outcomes was additive, or if there was an additional synergistic effect beyond the sum of their individual impacts. As sensitivity analyses, the same analyses were performed in the following subgroups: (1) after excluding patients with bilateral paralysis, and (2) using an alternative definition of dysphagia as a FOIS score of 6 or less, which included patients with mild dysphagia. Statistical analyses were performed using Statistical Package for the Social Sciences version 28.0 (IBM Corp., Tokyo, Japan). A p-value of less than 0.05 was considered statistically significant.
RESULTS
The study flowchart is shown in Fig. 1. During the study period, 1,124 patients with stroke were admitted to the hospital. Of these, 253 were excluded: 222 with missing data, 29 with a premorbid mRS score of 5, and two with COVID-19. Additionally, 250 patients did not return the questionnaire after 12 months or could not be followed up by using their medical records, and 17 who declined to participate. The final analysis included 604 patients. Table 1 shows the characteristics of the patients in each group. The median age (interquartile range) of the cohort was 76 (67–84) years, and 367 (60.8%) patients were men. The number of patients in each group was as follows: non-PS and non-dysphagia group, 281 (46.5%); non-PS and dysphagia group, 61 (10.1%); PS and non-dysphagia group, 187 (31.0%); and PS and dysphagia group, 75 (12.4%). Supplementary Table S1 shows a comparison of baseline characteristics between the 604 patients included in the final analysis and the 250 patients excluded due to loss to follow-up. Compared with patients included in the analysis, those excluded were significantly older, had higher NIHSS score, higher mRS score both premorbid and at discharge, and had lower handgrip strength and smaller calf circumference.
Flowchart of the study. mRS, modified Rankin Scale.
Patient characteristics
Fig. 2 shows the number and proportion of patients with poor functional outcomes in each group. Poor functional outcomes were observed in 22 (7.8%), 16 (26.2%), 74 (39.6%), and 63 (84.0%) patients in the non-PS and non-dysphagia, non-PS and dysphagia, PS and non-dysphagia, and PS and dysphagia groups, respectively. Fig. 3 shows the results of the modified Poisson regression analysis for poor functional outcomes. When the non-PS and non-dysphagia group was used as the reference, the risk ratios (RRs) and 95% confidence interval (CI) for each group were as follows: non-PS and dysphagia group, 2.096 (95% CI 1.160–3.789); PS and non-dysphagia group, 2.830 (95% CI 1.725–4.643); and PS and dysphagia group, 3.868 (95% CI 2.224–6.725). In this analysis, the dispersion parameter (φ) was 0.811. The Akaike Information Criterion (AIC) was 658.544 for the fully adjusted model and 672.078 for the crude model. When the groups were treated as ordinal variables, the p-value for trend was found to be significant (p<0.001). Fig. 4 shows the results of the modified Poisson regression analysis including dysphagia, PS, and the interaction term (PS×dysphagia) simultaneously. The RR for each variable was as follows: dysphagia, 1.449 (95% CI 1.133–1.854); PS, 2.322 (95% CI 1.531–3.523); and PS×dysphagia, 4.348 (95% CI 2.683–7.046). In this analysis, the φ was 0.815. The AIC was 661.548 for the fully adjusted model and 686.358 for the crude model. Sensitivity analysis results are presented in the Supplementary Figs. S1–S4. Across analyses excluding patients with bilateral paralysis and redefining dysphagia as the FOIS score of 6 or less, the PS and dysphagia group consistently showed significantly higher RRs compared with the non-PS and non-dysphagia group, and the PS×dysphagia interaction terms were statistically significant. All subgroup analyses demonstrated trends consistent with the main results.
The number and proportion of patients with poor functional outcomes in each group. Numbers within the bar graph indicate the number of patients. Each group was normalized to 100%. PS, possible sarcopenia.
The modified Poisson regression analysis for poor functional outcomes. Dots represent risk ratios for fully adjusted model including all covariates, and error bars indicate 95% confidence intervals. PS, possible sarcopenia.
The modified Poisson regression analysis of the interaction term between possible sarcopenia and dysphagia for poor functional outcomes. Dots represent risk ratios for fully adjusted model including all covariates, and error bars indicate 95% confidence intervals. PS, possible sarcopenia.
Table 2 shows adverse events up to 12 months after onset for each group. There were no significant differences in the incidence of MACEs among the groups (recurrent stroke, p=0.166; other cardiovascular diseases, p=0.197). For non-MACEs, significant intergroup differences were observed for pneumonia and other medical conditions (both p<0.001). Specifically, the incidence of pneumonia was as follows: non-PS and non-dysphagia group, three patients (1.1%); non-PS and dysphagia group, six patients (9.8%); PS and non-dysphagia group, seven patients (3.7%); and PS and dysphagia group, nine patients (12.0%). The incidences of other medical conditions were as follows: non-PS and non-dysphagia group, five patients (1.8%); non-PS and dysphagia group, six patients (9.8%); PS and non-dysphagia group, 18 patients (9.6%); and PS and dysphagia group, nine patients (12.0%). All-cause mortality was also significantly different among the groups (p<0.001): non-PS and non-dysphagia group, two patients (0.7%); non-PS and dysphagia group, five patients (8.2%); PS and non-dysphagia group, 19 patients (10.2%); and PS and dysphagia group, 15 patients (20.0%).
Adverse events up to 12 months post-onset for each group
DISCUSSION
This study investigated the relationship between the co-existence of PS and dysphagia and functional prognosis in patients with acute stroke. The results revealed a significant association between the co-existence of PS and dysphagia and poor functional outcomes at 12 months after stroke, with a significant interaction observed. This suggests that the two risk factors do not merely exert additive effects but mutually augment their adverse impacts.
The co-existence of PS and dysphagia is associated with poor functional outcomes and exhibits an interaction effect. While previous studies have reported sarcopenia alone11,33,34) and dysphagia alone35-37) as independent risk factors for poor outcomes after stroke, this study is unique in demonstrating for the first time that these two conditions do not merely exacerbate functional prognosis independently, but amplify their effects, leading to a more severe disability trajectory. Several mechanisms can explain this interaction. One pathway involves the difficulty in achieving adequate nutritional intake due to dysphagia, which accelerates the progression of sarcopenia. Malnutrition suppresses muscle protein synthesis and promotes muscle breakdown, thereby worsening sarcopenia.38-40) Severe dysphagia often necessitates the use of a tube feeding, which, combined with reduced physical activity, can lead to further loss of muscle mass and strength. Conversely, systemic sarcopenia implies generalized muscle weakness, including in the swallowing muscles, which can delay the recovery of the swallowing function and lead to more severe dysphagia.41,42) It should also be noted that, given the study design, sarcopenia may reflect a pre-existing condition prior to stroke onset. Therefore, when interpreting the interactions in this study, it may be natural to consider a temporal framework: the recovery trajectory is modified when dysphagia—a direct functional consequence of stroke—is superimposed on a pre-existing state of reduced physical reserve, such as sarcopenia. It is plausible that these vicious cycles interact, diminishing the effectiveness of rehabilitation, and consequently leading to a substantial deterioration in long-term functional prognosis.
This study found a potential association between the co-existence of PS and dysphagia and the incidence of medical conditions including pneumonia. This may be another potential factor contributing to poor functional outcomes. Dysphagia is the primary factor that increases the risk of aspiration pneumonia.43) Sarcopenia is associated with impaired immunity and reduced pulmonary function.44-46) Therefore, when both conditions coexist, a systemic state prone to severe complications may be formed, which can lead to increased medical interventions, interruptions in rehabilitation, and further physical decline, thus potentially contributing to poor functional outcomes at 12 months after stroke.
The results of this study indicate the importance of early screening for PS and dysphagia in patients with acute stroke in order to identify high-risk patients with coexisting conditions. Implementing multifaceted interventions that combine early and intensive swallowing rehabilitation, nutritional management, and exercise therapy could potentially break the vicious cycle of sarcopenia and dysphagia, prevent complications such as pneumonia, and ultimately improve functional outcomes. Given the demonstrated interaction, it is crucial not only to treat each condition individually but also to implement a dual therapeutic cycle of rehabilitation and nutritional treatment that considers their mutual relationship.47) For example, chair-stand exercises have been shown to improve not only sarcopenia but also swallowing function in patients with stroke.48,49) Furthermore, frequent and individualized nutritional support can lead to improvements in nutritional status, activities of daily living, and swallowing function after stroke.50) Future prospective studies should investigate the potential of multifaceted interventions targeting both sarcopenia and dysphagia. Such studies should also evaluate the dose of rehabilitation after admission and explore potential mitigating factors, such as the interaction between rehabilitation intensity and the risk of complications (e.g., reduced pneumonia risk with a higher rehabilitation dose in patients with dysphagia), to further optimize patient outcomes.
This study had several limitations. First, it was conducted at only two acute care hospitals in Japan, which may limit the generalizability of the results. Second, a considerable number of participants were excluded because of missing data, potentially leading to selection bias. Comparison of baseline characteristics between patients excluded during follow-up and those included in the analysis showed that excluded patients were older, had more severe stroke, and poorer physical function. Therefore, the possibility of attrition bias among patients with these characteristics cannot be ruled out. Third, owing to the retrospective design of the study, it was difficult to establish causality. Fourth, we could not fully adjust for the influence of confounding factors such as genetic, social, and clinical variables, which could contribute to poor functional outcomes. In particular, premorbid mRS differed across the groups. Although this factor was adjusted for in the statistical models, potential residual differences in baseline patient characteristics cannot be ruled out. Thus, pre-existing functional limitations and related vulnerability may have contributed to the observed outcomes. Fifth, the robustness of the definitions of sarcopenia and dysphagia may have been limited. According to the AWGS criteria, a definitive diagnosis of sarcopenia requires body composition analysis, in addition to grip strength and calf circumference.24) Furthermore, detailed dysphagia assessments, such as videoendoscopic or videofluoroscopic swallowing studies, were not performed; therefore, the accuracy of the FOIS classification in fully capturing dysphagia severity may be limited. In addition, owing to the retrospective design, the reasons underlying clinicians’ decisions to initiate tube feeding could not be fully ascertained, and the possibility of selection bias cannot be excluded. Future studies incorporating objective assessments such as videoendoscopic and/or videofluoroscopic swallowing studies are warranted.
In conclusion, this study demonstrated that the co-existence of PS and dysphagia in patients with acute stroke is associated with poor functional outcomes at 12 months after stroke onset and that a significant interaction exists between the two. These findings suggest that early screening for sarcopenia and dysphagia, along with integrated intervention strategies that consider their interaction, is indispensable for improving the long-term prognosis after stroke. Future studies should include prospective interventional studies to validate the effectiveness of multifaceted intervention programs for high-risk patients.
Notes
We express our deepest gratitude to the Rehabilitation Center of Uonuma Kikan Hospital and the Konan Medical Center for their support.
CONFLICT OF INTEREST
The researchers claim no conflicts of interest.
FUNDING
None.
AUTHOR CONTRIBUTIONS
Conceptualization, YS, TA, KO, AS, KY; Data curation, YS, TA, KO, AS, KY; Formal analysis, YS; Writing_original draft, YS; Writing_review & editing, YS, YK, ST, KI, MN.
SUPPLEMENTARY MATERIALS
Supplementary materials can be found via https://doi.org/10.4235/agmr.26.0002.
Comparison of baseline characteristics of patients included in the study and those excluded at follow-up
The modified Poisson regression analysis for poor functional outcomes in the subgroup excluding patients with bilateral paralysis. Dots represent risk ratios for fully adjusted model including all covariates, and error bars indicate 95% confidence intervals. PS, possible sarcopenia.
The modified Poisson regression analysis of the interaction term between possible sarcopenia and dysphagia for poor functional outcomes in the subgroup excluding patients with bilateral paralysis. Dots represent risk ratios for fully adjusted model including all covariates, and error bars indicate 95% confidence intervals. PS, possible sarcopenia.
The modified Poisson regression analysis for poor functional outcomes in patients with dysphagia defined as the Functional Oral Intake Scale score 6 or less. Dots represent risk ratios for fully adjusted model including all covariates, and error bars indicate 95% confidence intervals. PS, possible sarcopenia.
The modified Poisson regression analysis of the interaction term between possible sarcopenia and dysphagia for poor functional outcomes in patients with dysphagia defined as the Functional Oral Intake Scale score 6 or less. Dots represent risk ratios for fully adjusted model including all covariates, and error bars indicate 95% confidence intervals. PS, possible sarcopenia.