Frailty and sarcopenia in dialysis patients: a double burden with clinical impact

Frailty is a progressive, age-related decline in physiological capacity leading to diminished functional reserves, increasing vulnerability to stressors, and a higher risk of adverse outcomes, including loss of independence and death. While closely linked, sarcopenia—characterized by loss of muscle strength and mass—remains a distinct clinical condition. Both conditions are common in chronic kidney disease (CKD), particularly among dialysis patients.1–3 Despite their clinical significance, routine assessment is uncommon, and no consensus exists on the most appropriate frailty assessment tool for CKD. This gap is clinically important, as early identification of frailty and sarcopenia enables timely intervention that may prevent functional decline, reduce hospitalizations, and improve survival. The present study aimed to investigate the relationship between frailty and sarcopenia, to compare the predictive performance of different frailty tools, and to explore the combined effects of frailty and sarcopenia on mortality outcomes. A prospective observational study, SPARSE (Sarcopenia in Patients with CKD), was conducted at a tertiary academic renal center.4 Maintenance hemodialysis patients aged ≥50 years were eligible for inclusion in this study. Frailty was assessed using three instruments: the Clinical Frailty Scale (CFS >4), the Edmonton Frailty Scale (EFS >7), and a modified Fried Phenotype (FP >2). Sarcopenia was defined according to the European Working Group on Sarcopenia in Older People 2 and the 2019 Asian Working Group for Sarcopenia criteria, including probable, confirmed, and severe sarcopenia. Mortality risk was analyzed using Cox proportional hazard models. Detailed methodology is provided in the Supplementary Methods. Eighty-one patients were included; 64% were male, mean age was 73 years. Sarcopenia was present in 74% of patients. Frailty prevalence varied widely, from 12% to 19% depending on the frailty tool applied. Considerable overlap was observed between sarcopenia and frailty (Figure 1A), underscoring the burden of these combined geriatric syndromes in CKD. Over a median follow-up of 28.5 months, 20 patients died. Causes of death included withdrawal from dialysis, infection, cardiovascular event, cancer, or unknown (Figure S1). Deceased patients had lower post-dialysis diastolic blood pressure (P < .05), longer Timed Up and Go and 5-Times Chair Stand test times (P < .01 and P = .03), and lower serum creatinine (P = .04). Frailty was more common in this group, though this was not statistically significant (see Table S1). There were no significant differences between the living and deceased groups in terms of age, Charlson comorbidity index, self-reported physical activity level, dialysis vintage, handgrip strength, appendicular skeletal muscle index, prevalence of sarcopenia, hemoglobin, C-reactive protein, indoxyl sulfate, p-cresyl sulfate, or serum albumin levels. (A) The relationship between sarcopenia and frailty in maintenance hemodialysis patients. (B–D) Kaplan-Meier mortality-free survival curves among maintenance hemodialysis patients, stratified by frailty diagnostic tools: Clinical Frailty Scale (CFS), Edmonton Frailty Scale (EFS), and Fried Phenotype (FP). Panels (E–G) illustrate the combined effect of coexisting sarcopenia and frailty on mortality. Abbreviations: CFS, Clinical Frailty Scale; CS, confirmed sarcopenia; EFS, Edmonton Frailty Scale; FP, Fried Phenotype; NSF, non-sarcopenic but frail; NSNF, non-sarcopenic and non-frail; PS, probable sarcopenia; SF, sarcopenic and frail; SNF, sarcopenic but non-frail; SS, severe sarcopenia. Associations between frailty and mortality are presented in Figure 1B–D. For the CFS, median survival was 42 months (95% CI, 13-not reached) in the frail group, while survival for the non-frail group was not reached, with ∼70% patients alive at follow-up. One-year survival was 98% in the non-frail group versus 85% in the frail group (log-rank χ² = 3.18, P = .07). Using the EFS, median survival in the frail group was 33 months (95% CI, 5-not reached), with stronger discrimination (χ² = 6.96, P < .01). In contrast, FP did not distinguish survival. In Cox regression analysis, EFS >7 was significantly associated with mortality (HR = 3.60 [95% CI, 1.17–9.36], P = .03). CFS >4 showed an increased hazard ratio, although this did not reach statistical significance (HR = 2.64 [95% CI, 0.82–5.90], P = .10). After adjusting for age and sex, EFS >7 remained independently predictive of mortality (aHR = 3.50 [95% CI, 1.13–9.13], P = .03), while CFS >4 was strengthened (aHR = 2.74 [95% CI, 0.99–7.57], P = .08). FP >2 showed no association. When combined with sarcopenia, the EFS revealed a clear survival gradient (log-rank trend, P = .04): patients with both frailty and sarcopenia had the poorest outcomes (Figure 1E–G). Cox regression confirmed higher mortality with both conditions versus neither (HR of 5.57 [95% CI, 1.46–22.81, P = .01]), which remained significant after adjustment (see Tables S2 and S3). Our study is among the first in CKD to demonstrate that coexisting frailty and sarcopenia predicts the worst survival. This mirrors findings in other populations: Petermann-Rocha et al. reported in >300 000 UK Biobank participants that combined sarcopenia and frailty increased cardiovascular, respiratory, and all-cause mortality.5 Similar associations were seen in gastrointestinal cancer6 and hepatobiliary-pancreatic surgery.7 These findings underscore the importance of systematically identifying patients who exhibit these vulnerability syndromes before they progress to advanced stages. By integrating simple, pragmatic assessment tools such as the CFS and EFS into routine renal practice, clinicians can more reliably detect frailty state at an earlier point in the patient journey. These tools are quick to administer, require minimal training, and can be readily embedded into standard clinic workflows without additional resource burden. This will eventually allow for timely implementation of targeted interventions—such as nutritional optimization, exercise programs, multidisciplinary support, and risk-stratified care planning—ultimately improving patient outcomes and informing clinical decision-making. Interestingly, the EFS appeared to perform better than both the CFS and FP in predicting mortality outcomes. While FP remains one of the most widely used methods to assess physical frailty, it did not show a significant association with mortality in our cohort. This may reflect the variability in cutoffs for components such as handgrip strength and gait speed, which may lead to under- or over-estimation of risk. The lack of statistical significance between the FP and mortality should be interpreted with caution, especially given the limited statistical power for this analysis with the low incidence of death. Our study has several limitations. First, the generalizability of our findings is limited by the relatively small sample size and the single-center study design. Second, frailty was assessed once at baseline, without accounting for longitudinal changes. Lastly, muscle mass estimation by bioelectrical impedance might have been affected by fluid status, a persistent challenge in dialysis. In conclusion, the coexistence of sarcopenia and frailty was associated with highest mortality. Integrating simple assessments of these conditions into routine renal care could enable earlier intervention and improve outcomes in this vulnerable population. Limy Wong: Conceptualization; Methodology; Formal analysis; Investigation; Writing—original draft, Writing—review & editing. Emily Schembri: Formal analysis, Writing—review & editing. Lawrence P. McMahon: Supervision, Writing—review & editing. L.W. received support for this study from the 2021 Eastern Health Foundation Research and Innovation Grant, the Royal Australasian College of Physicians Foundation (Jacquot Research Establishment Fellowship, 2021REF00021 and 2022REF00084). The funders had no role in study design, data collection, analysis, reporting, or the decision to submit for publication. All data generated or analyzed during this study are available in the article and its supplementary information files. Ethics approval was obtained from Eastern Health Human Research Ethics Committee in Victoria, Australia (reference number E20-025-67026). Full written informed consents were obtained from all study participants. Supplementary material is available at Journal of Nephrology online.