Abstract
Frailty is an important prognostic factor in patients with heart failure and in transplant candidates, while cognitive impairment remains an under-recognized component that may further worsen clinical outcomes. This prospective study aimed to determine the prevalence of physical and cognitive frailty among heart transplant candidates and their association with mortality. Physical frailty was assessed using the Fried criteria; cognition was evaluated using the MoCA (cognitive screening) and WASI (general intellectual functioning); and depressive symptoms were measured using the BDI. Cognitive frailty was defined as the presence of ≥3 of 6 criteria in a modified Fried phenotype, which incorporated cognitive impairment as an additional domain. A total of 150 patients listed for heart transplantation within the previous 30 days were included at a tertiary center in Brazil. The study population was predominantly male (70%), with Chagas disease etiology (35%), and a mean age of 49.8 ± 10.4 years. Physical frailty was present in 78.7% of patients, cognitive frailty in 88.7% (MoCA) and 86.7% (WASI), and depressive symptoms in 44%, which were significantly associated with physical frailty (p = 0.021). During 1-year follow-up, 41.3% of patients died and 67.3% underwent transplantation. Physical frailty was associated with an increased risk of early mortality after listing (initial SHR = 5.47; p = 0.023), with significant attenuation of this effect over time (TVC SHR = 0.99; p = 0.008). A similar pattern was observed for cognitive frailty, measured by both MoCA (initial SHR = 1.61; p = 0.011; TVC SHR = 1.00; p = 0.011) and WASI (initial SHR = 1.57; p = 0.012; TVC SHR = 1.00; p = 0.007), indicating a reduction in its impact on mortality risk over time. However, after multivariable adjustment, neither physical nor cognitive frailty remained independently associated with overall mortality. Physical and cognitive frailty were highly prevalent among heart transplant candidates and showed a time-dependent association with early mortality risk. After adjustment for disease severity markers, this association lost significance, suggesting that frailty may, at least in part, reflect the underlying clinical status at the time of listing rather than acting as an independent prognostic factor. These findings also raise questions regarding the adequacy of the Fried phenotype as a frailty assessment tool in this specific population
Keywords: cognitive impairment, depression, physical frailty, cognitive frailty, heart failure, heart transplant
Introduction
Frailty is syndrome initially described in geriatric research that has gained increasing attention due to its impact on clinical outcomes and disease prognosis. In patients with heart failure, frailty and the disease process share a bidirectional relationship, with overlapping symptoms such as fatigue and exercise intolerance, as well as common pathophysiological mechanisms, including systemic inflammation and sarcopenia. Frailty reflects increased vulnerability to stressors resulting from reduced physiological reserve and is associated with a higher risk of decompensation, adverse events, functional decline and disability.1, 2, 3
Two main conceptual models are commonly used to assess frailty: the physical frailty phenotype proposed by Fried, based on five clinical criteria, and the Frailty Index described by Rockwood, which reflects the accumulation of deficits across medical, functional, and psychosocial domains. More recently, frailty has been linked to structural and functional brain changes and cognitive impairment, giving rise to the concept of cognitive frailty. Cognitive frailty is defined as the coexistence of physical frailty and cognitive impairment in the absence of dementia. This condition represents a potentially reversible state associated with reduced cognitive reserve and has been shown to predict mortality and adverse outcomes..4, 5, 6
In patients with advanced heart failure referred for heart transplantation, physical frailty is highly prevalent and associated with worse outcomes, including increased mortality and longer hospitalization.7 Cognitive impairment is also common but frequently underdiagnosed, despite its negative impact on self-care, treatment adherence, and transplant-related outcomes.8 Evidence to date is largely derived from studies conducted in high-income countries, where patients tend to be referred earlier and have higher educational levels, potentially limiting generalizability to middle-income settings. Given the influence of cognitive reserve and socioeconomic factors, data on cognitive frailty in transplant candidates from countries such as Brazil remain scarce.2, 3, 9, 10, 11
Although the Fried Frailty Phenotype is widely used to assess frailty in patients with heart failure, its performance in individuals with advanced disease remains uncertain. In this context, the present study aimed to estimate the prevalence of physical and cognitive frailty among patients awaiting heart transplantation, compare two cognitive assessment tools for defining cognitive frailty, and examine its association with mortality risk while on the transplant waiting list.
Methods
Study population
This study included patients with advanced heart failure referred for heart transplantation at Instituto do Coração do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HC-FMUSP) between December 2018 and May 2024. All patients who agreed to participate voluntarily, provided written informed consent, and were consecutively included were enrolled. The patient selection flowchart (Figure 1) details excluded patients. The study was approved by the Institutional Research Ethics Committee (CAAE 97526818.4.0000.0068). Participants were assessed for physical frailty, cognitive impairment, and depressive symptoms, and followed for up to 12 months or until transplantation or death. Clinical data were obtained from electronic medical records, and a database was developed using the REDCap platform.
Figure 1.
Flowchart of patient selection and follow-up.
Inclusion criteria were patients aged over 18 years, of both sexes, with a diagnosis of heart failure of various etiologies listed for heart transplantation at InCor within 30 days of inclusion on the waiting list. Exclusion criteria included clinical conditions incompatible with cognitive and depression assessments, such as somnolence, reduced level of consciousness, delirium, or acute mental confusion; incomplete assessment protocols due to clinical decompensation or refusal; and language barriers, as the neuropsychological protocol involved tasks specific to the Portuguese language.
Prior to patient approach, a thorough review of medical records was conducted, along with discussion with the multidisciplinary team responsible for patient care, in order to exclude individuals with diagnosed dementia or delirium. In addition, all participants were assessed by a single psychologist researcher, who conducted an initial semi-structured interview before administering the cognitive tests, ensuring that signs of dementia or delirium were systematically evaluated and ruled out.
Physical frailty assessment
Physical frailty was assessed using Fried’s criteria. In this classification, after evaluation in five domains (unintentional weight loss, exhaustion, reduced gait speed, low physical activity level, decreased grip strength), patients were classified as non-frail if no criteria were present (0 points), pre-frail with 1 or 2 points, and frail with 3 or more points out of these five criteria.5 Table S1 briefly shows the domains and frailty classification measures used in this study.
Cognitive assessment
Cognition was assessed using the MoCA screening test and a more robust tool, the WASI. Most studies on frailty use cognitive screening scales; nonetheless, some researchers emphasize the requirement for in-depth cognitive exploration. Therefore, the WASI was chosen to provide solid data and refined analysis of patients’ cognitive performance. The use of the instrument in the hospital setting and its validation for the Brazilian population provided the necessary scope to strengthen the cognitive variable in this study.
MoCA is a brief cognitive screening instrument shown to be practical and effective in distinguishing performance between cognitively normal adults and those with cognitive decline. The tool assesses eight domains (executive functions, visuospatial abilities, memory, working memory, attention, language, and orientation) with a maximum score of 30 points. Scores < 26 indicate the presence of cognitive impairment, whereas scores ≥ 26 indicate the absence of cognitive impairment.12
WASI is an individually administered intelligence assessment instrument covering ages 6–89 years. The battery provides information on total intelligence quotient (TIQ), performance IQ (PIQ), and verbal IQ (VIQ) from four subtests (vocabulary, block design, similarities, and matrix reasoning) assessing various cognitive aspects, including verbal knowledge, visuospatial organization and processing, memory, spatial and nonverbal reasoning, fluid and crystallized intelligence, lexical competencies and language integration, concept formation and categorization, mental flexibility, and planning. The test generates total IQ, performance IQ, and verbal IQ scores classified into seven levels: very superior (> 130 points), superior (120–129 points), high average (110–119 points), average (90–109 points), low average (80–89 points), borderline (70–79 points), and delayed (< 69 points). The same cutoff as MoCA was used for cognitive impairment classification: scores below average (< 90 points) indicated impairment and scores ≥ 90 indicate the absence of cognitive impairment13, 14
Studies involving cognitive assessment rarely rely exclusively on screening tests, as these instruments are understood as initial and simplified measures of cognitive functioning. Thus, screening tests should be regarded as an initial step and complemented by more comprehensive instruments that allow a deeper characterization of participants’ cognitive profiles. In this context, the use of the Wechsler Abbreviated Scale of Intelligence (WASI) in the present study provides a more robust and nuanced evaluation of cognitive functioning. Furthermore, several studies recommend investigating the cognitive profile of patients with heart failure.
Cognitive frailty assessment
Cognitive frailty (CogF) was defined as the coexistence of physical frailty and cognitive impairment in the absence of dementia. CogF was operationalized using a modified Fried frailty phenotype, in which a cognitive measure (MoCA or WASI) was incorporated as a sixth criterion, consistent with previous literature. Cognitive impairment contributed one point to the frailty score, whereas its absence contributed zero points. Participants were classified as cognitively frail when ≥3 of the six criteria were present and as non–cognitively frail when <3 criteria were met (Table S1).2, 3
Depressive symptoms assessment
Depressive symptoms were assessed using the Beck Depression Inventory (BDI), a common tool used for evaluating depressive symptoms in patients with cardiac diseases, correlating with quality of life and clinical prognosis. This is a self-administered instrument composed of 21 items whose objective is to measure the presence and intensity of depressive symptoms. Patients answer questions regarding sadness, pessimism, feelings of failure, lack of satisfaction, guilt, feelings of punishment, self-deprecation, self-accusations, suicidal ideation, crying spells, irritability, social withdrawal, indecisiveness, distorted body image, inhibition at work, fatigue, somatic preoccupation, and changes in sleep, appetite, weight, and libido. Results up to 13 points indicate the absence of depressive symptoms; 14–19 points, symptoms of mild depression; 20–28 points, symptoms of moderate depression; and 29–63 points, symptoms of severe depression.15
Statistical analysis
Continuous variables were presented as mean ± standard deviation (one decimal place), along with the number of participants available for analysis. Between-group comparisons were conducted using the nonparametric Wilcoxon-Mann-Whitney test for independent samples. Categorical variables were expressed as absolute and relative frequencies (percentages, one decimal place) and compared using Fisher’s exact test. Cutoff points for defining physical and cognitive frailty were based on previously validated criteria in the literature. All variables collected at baseline were assessed in univariate analyses; those with p < 0.20 and clinical relevance were included in the multivariable model. In the final model, only variables with p < 0.05 were retained and considered independently associated with the outcome. A two-sided p-value < 0.05 was considered statistically significant. All analyses were performed using R software (version 4.3).
Given the presence of a competing event (transplantation), a competing risks analysis was performed, considering death as the event of interest and transplantation as the competing event. Associations between covariates and mortality were evaluated using the Fine–Gray subdistribution hazards model, which estimates the effect of covariates on the cumulative incidence of the event of interest in the presence of competing risks. Multivariable models were adjusted for age, sex, and IABP use, and were intentionally parsimonious given the limited number of events to avoid overfitting. Results are presented as subdistribution hazard ratios (SHR) with 95% confidence intervals. The proportional hazards assumption was assessed by testing interactions between covariates and time; when violations were identified, time-dependent effects were modeled using interaction terms with follow-up time, and results are reported as SHR for the baseline effect and for the time-varying component (TVC SHR).
Given the advanced clinical status of the study population, particular attention was paid to avoid overadjustment. Variables included in multivariable models were selected based on clinical relevance and statistical criteria, acknowledging the potential overlap between frailty measures and markers of disease severity, which may share common underlying pathways.
Results
General characteristics of the population
The study population was predominantly male (70%) with self-reported black race (67.3%). The mean age of participants was 49.8 years (± 10.4), and the mean education level was 8.9 years (± 4.8). Concerning the etiology of heart failure, Chagas disease was the most prevalent (35%), followed by dilated cardiomyopathy (32%) and ischemic heart disease (12%). In our sample, 56.7% of patients were admitted to the intensive care unit (ICU) and 39.3% to the ward. Many were prioritized for transplantation due to dependence on vasoactive drugs (88.7%) or an intra-aortic balloon pump (31.3%). The remaining characteristics are described in Table 1.
Table 1.
General Characteristics of the Population
| Values (N=150) | |
|---|---|
| Age (years) | 49.8±10.4 |
| Male | 105 (70) |
| Race Black | 101 (67.3) |
| Education (years) | 8.9±4.8 |
| Married | 77 (51.3) |
| Monthly income (n° minimum wages)* | 1.8±1.8 |
| Etiology | |
| Chagasic | 53 (35.3) |
| Dilated | 49 (32.7) |
| Risk factors | |
| Comorbidities | 87(58) |
| Hypertension | 53(60.9) |
| Diabetes | 27(31) |
| Dyslipidemia | 37(42.5) |
| Renal failure | 32(36.8) |
| Smoking | 54(36) |
| Alcoholism | 61(40.7) |
| Mental health history | 82(54.7) |
| BMI classification | |
| Underweight | 25(16.7) |
| Normal | 84(56) |
| Main medications | |
| ACE inhibitors | 42(28) |
| ARB’s | 9(6) |
| Hydralazine | 71(47.3) |
| Isosorbide | 44(29.3) |
| Spironolactone | 111(74) |
| Beta-blockers | 24(16) |
| Furosemide | 130(86.7) |
| Sacubitril-valsartan | 5(3.3) |
| Digoxin | 7(4.7) |
| Vasoactive drugs | 133(88.7) |
| Dobutamine | 131(98.5) |
| Milrinone | 15(11.3) |
| Sodium nitroprusiate | 43(32.3) |
| BIA | 47(31.3) |
| INTERMACS | |
| II | 53(35.3) |
| III | 92(61.3) |
| BDI Classification | |
| No symptoms | 84(56) |
| Mild | 38(25.3) |
| Moderate | 16(10.7) |
| Severe | 12(8) |
| Exams | |
| Hemoglobin (g/dL) | 10.8±1.9 |
| Urea (mg/dL) | 57.3±30 |
| Creatinine (mg/dL) | 1.4±0.6 |
| RV function | |
| Mild | 41±27.3 |
| Moderate | 61±40.7 |
| LVEF (%) | 25±8.1 |
Categorical variables are expressed as absolute numbers and percentages (n) and numerical variables as means and standard deviations (mean ± SD). ACE inhibitors = Angiotensin-converting enzyme inhibitors; ARBs = Angiotensin II receptor blockers; ISGLT2 = Sodium glucose cotransporter 2 inhibitors; BMI = Body mass index; IABP = Intra-aortic balloon pump; BDI = Beck Depression Inventory; RV = Right ventricle; LVEF = Left ventricular ejection fraction.
*The Brazilian minimum wage corresponds to R$1620, and the mean income of the study participants was 1.8 times the minimum wage, which is equivalent to approximately US$ 540 per month.
Cognitive analyses
In MoCA classification, 133 (86.7%) patients were classified as cognitively impaired. Lower scores were observed in memory (2.5 ± 1.5 points), attention (3.2 ± 1.7 points), and abstraction (1.0 ± 0.8 points) among the assessed domains. Orientation was the most preserved domain (5.6 ± 0.6), followed by naming (2.6 ± 0.6), language (2.0 ± 0.8), and visuospatial skills (2.7 ± 1.5).
In WASI classification, 130 (84.7%) patients were classified as cognitively impaired. The mean Full-Scale IQ (QIT) was 73.4 (±14.9). The mean PIQ was 73.6 (±14.9), and the mean VIQ was 78.8 (±15.6). Thus, most patients scored below the population average on verbal and performance tasks (76% and 84%, respectively), indicating global cognitive impairment. The PIQ revealed poorer performance on fluid intelligence tasks involving abstract reasoning, novel problem solving, and focused attention. The VIQ showed relative preservation of linguistic skills and vocabulary, reflecting aspects of crystallized intelligence.
Agreement analysis between the two cognitive assessment tests showed a high level of concordance, as indicated by the unweighted Kappa coefficient (k = 0.840; 95% confidence interval: 0.773–0.908; p < 0.001), having an agreement rate of 88%.
Prevalence of physical, cognitive frailty and depressive symptoms
Table 2 summarizes the prevalence of physical frailty according to Fried Frailty Phenotype across its five assessed components, as well as data on cognitive frailty measured by Montreal Cognitive Assessment (MoCA) and Wechsler Abbreviated Scale of Intelligence (WASI), and depressive symptoms assessed using the Beck Depression Inventory (BDI).
Table 2.
Prevalence of Physical Frailty, Cognitive Frailty, and Depressive Symptoms
| Value (n=150) | |
|---|---|
| Physical frailty score (0–5 points) | 3.3±1.1 |
| Physical frailty componentes - FRIED | |
| Unintentional weight loss | 84 (56.0) |
| Exhaustion | 70 (46.7) |
| Reduced handgrip strength | 66 (44.0) |
| Slow gait speed | 127 (84.7) |
| vLow physical activity level | 149 (99.3) |
| Physical frailty classification - FRIED | |
| Non-frail (0 points) | 0 (0.0) |
| Pre-frail (1–2 points) | 32 (21.3) |
| Frail (≥ 3 points) | 118 (78.7) |
| Cognitive frailty WASI (0–6 points) | 4.2±1.2 |
| CogF classification – WASI | |
| Non-frail (< 3 points) | 20 (13.3) |
| Frail (≥ 3 points) | 130 (86.7) |
| Cognitive frailty MoCA (0–6 points) | 4.2±1.3 |
| CogF classification – MoCA | |
| Non-frail (< 3 points) | 17 (11.3) |
| Frail (≥ 3 points) | 133 (88.7) |
| Beck Depression Inventory (BDI) total score | 14.6±9.0 |
| BDI classification | |
| No depressive symptoms | 84 (56.0) |
| Mild symptoms | 38 (25.3) |
| Moderate symptoms | 16 (10.7) |
| Severe symptoms | 12 (8.0) |
Categorical variables are expressed as absolute number and percentage [n (%)], and continuous variables as mean ± standard deviation. CogF= Cognitive frailty; MoCA = Montreal Cognitive Assessment; WASI = Wechsler Abbreviated Scale of Intelligence; BDI = Beck Depression Inventory.
According to Fried’s physical frailty criteria, 118 (78.7%) patients were classified as frail, 32 (21.3%) as pre-frail, and none as non-frail. The mean total points from the five Fried criteria were 3.3 (±1.1), with 99.3% of the sample showing low physical activity, 84.7% reduced gait speed, 56% unintentional weight loss, 46.7% exhaustion, and 44% decreased grip strength.
The prevalence of cognitive frailty measured by WASI (CogF WASI) was 86.7% (130), while cognitive frailty by MoCA (CogF MoCA) reached 88.7% (133). The mean CogF MoCA score was 4.2 (±1.3), and the CogF WASI was 4.2 (±1.2). The scoring criteria are numerically similar; however, they vary between scales, and detailed analysis revealed that not all patients classified as frail by one measure were similarly classified by the other. This discrepancy was also reflected in differences in the categorical distributions between groups across the two scales.
Among the 150 patients included in the study, 118 (78.7%) were consistently classified as frail across all three frailty models evaluated (PF, CogF MoCA and CogF WASI). When considering physical frailty specifically, none of the patients classified as physically frail were categorized as cognitively non-frail, regardless of the cognitive instrument used.
Regarding depressive symptoms analysis, the mean total score on BDI was 14.6 ± 9.0. Overall, 56% of participants (n = 84) presented no depressive symptoms, while 44% exhibited some degree of depressive symptomatology. Mild depressive symptoms were observed in 25.3% (n = 38), moderate symptoms in 10.7% (n = 16), and severe symptoms in 8% (n = 12) of the cohort. An association was observed between depressive symptoms and physical frailty, with frail patients exhibiting higher scores than pre-frail patients (15.4 ± 9.5 vs. 11.4 ± 6.0, p = 0.021). In contrast, no association was identified between depressive symptoms measured by the BDI and cognitive frailty as assessed by the MoCA (p = 0.497) or the WASI (p = 0.659) [Table S2].
Baseline characteristics stratified by frailty criteria
Physically frail patients had lower education (8.4 ± 4.8 vs. 10.7 ± 4.7 years; p = 0.011) than pre-frail patients. Similarly, cognitive frailty by WASI was associated with lower education (8.6 ± 4.7 vs. 11.1 ± 5.1 years; p = 0.018). There was no difference in physical (p=0.666) and cognitive frailty, assessed by MoCa (p=0.816) or WASI (p=0.767) according to etiology. The remaining characteristics are described in Table 3.
Table 3.
Demographic Data Stratified by Frailty Measures
|
Physical Frailty |
CogF MoCA |
CogF WASI |
|||||||
|---|---|---|---|---|---|---|---|---|---|
| Variable | Pre-frail (n=32) | Frail (n=118) | p- valor | Non-frail (n=17) | Frail (n=133) | p- valor | Non-frail (n=20) | Frail (n=130) | p- valor |
| Age (Years) | 52.8±10 | 49±10.4 | 0.103 | 50.7±9.7 | 49.7±10.5 | 0.693 | 51.5±9.2 | 49.6±10.5 | 0.487 |
| Male | 25(78.1) | 80(67.8) | 0.286 | 12 (70.6) | 93 (69.9) | >0.999 | 16(80) | 89(68.5) | 0.432 |
| Race | |||||||||
| Black | 16(50) | 85(72) | 0.049 | 9(52.9) | 92(69.2) | 0.262 | 10(50) | 91(70) | 0.195 |
| Education (Years) | 10.7 (4.7) | 8.4 (4.8) | 0.011 | 10.6±5.3 | 8.7±4.8 | 0.096 | 11.1±5.1 | 8.6±4.7 | 0.018 |
| Married | 14(43.8) | 63(53.4) | 0.251 | 9(52.9) | 68(51.1) | 0.638 | 10(50) | 67(51.5) | 0.334 |
| Monthly income | 2.6±2.8 | 1.6±1.4 | 0.161 | 2.7±3.3 | 1.7±1.5 | 0.625 | 3.1±3 | 1.6±1.5 | 0.060 |
| Etiology | |||||||||
| Chagasic | 9(28.1) | 44(37.3) | 0.666 | 6(35.3) | 47(35.3) | 0.816 | 5(25) | 48(36.9) | 0.767 |
| Ischemic | 3(9.4) | 15(12.7) | 1(5.9) | 17(12.8) | 315) | 15(11.5) | |||
| Dilated | 15(46.9) | 34(28.8) | 8(47.1) | 41(30.8) | 9(45) | 40(30.8) | |||
| Risk factors | |||||||||
| Comorbidities | 22(68.8) | 65(55.1) | 0.226 | 10(58.8) | 77(57.9) | >0.999 | 12(60) | 75(57.7) | >0.999 |
| Hypertension | 16 (72.7) | 37(56.9) | 0.310 | 5(50) | 48(62.3) | 0.497 | 8(66.7) | 45(60) | 0.761 |
| Diabetes Melitus | 4(18.2) | 23(35.4) | 0.183 | 1(10) | 26(33.8) | 0.161 | 2(16.7) | 25(33.3) | 0.325 |
| Dyslipidemia | 6(27.3) | 31(47.7) | 0.133 | 3(30) | 34(44.2) | 0.504 | 4(33.3) | 33(44) | 0.542 |
| Renal failure | 7(31.8) | 25(38.5) | 0.616 | 4(40) | 28(36.4) | >0.999 | 4(33.3) | 28(37.3) | >0.999 |
| Smoking | 15(46.9) | 39(33.1) | 0.154 | 9(52.9) | 45(33.8) | 0.178 | 12(60) | 42(32.3) | 0.023 |
| Alcoholism | 16(50) | 45(38.1) | 0.232 | 9(52.9) | 52(39.1) | 0.303 | 12(60) | 49(37.7) | 0.086 |
| Mental health history | 18(56.3) | 64(54.2) | 0.999 | 11(64.7) | 71(53.4) | 0.445 | 11(55) | 71(54.6) | >0.999 |
| BMI classification | |||||||||
| Underweight | 2(63) | 23(19.5) | 0.197 | 1(5.9) | 24(18) | 0.608 | 2(10) | 23(17.7) | 0.754 |
| Normal | 18(56.3) | 66(55.9) | 10(58.8) | 74(55.6) | 11(55) | 73(56.2) | |||
| Medications in use | |||||||||
| ACE inhibitors | 12(37.5) | 30(25.4) | 0.189 | 7(41.2) | 35(26.3) | 0.251 | 7(35) | 35(26.9) | 0.436 |
| ARB | 3(9.4) | 6(5.1) | 0,403 | 2(11.8) | 7(5.3) | 0.270 | 2(10) | 7(5.4) | 0.343 |
| Hydralazine | 15(46.9) | 56(47.5) | >0.999 | 9(52.9) | 62(46.6) | 0.797 | 10(50) | 61(46.9) | 0.815 |
| Isosorbide | 11(34.4) | 33(28) | 0.515 | 7(41.2) | 37(27.8) | 0.267 | 8(40) | 36(27.7) | 0.295 |
| Spironolactone | 21(65.6) | 90(76.3) | 0.258 | 10(58.8) | 101(75.9) | 0.147 | 12(60) | 99(76.2) | 0.169 |
| Beta-blocker | 7(21.9) | 17(14.4) | 0.292 | 4(23.5) | 20(15) | 0.479 | 5(25) | 19(14.6) | 0.321 |
| Furosemide | 28(87.5) | 102(86.4) | >0.999 | 16(94.1) | 114(85.7) | 0.472 | 17(85) | 113(86.9) | 0.732 |
| Sacubitril-valsartan | 1(3.1) | 4(3.4) | >0.999 | 1(5.9) | 4(3) | 0.457 | 1(5) | 4(3.1) | 0.516 |
| ISGLT2 | 0(0) | 2(1.7) | >0.999 | 0(0) | 2(1.5) | >0.999 | 0(10) | 2(1.5) | >0.999 |
| Digoxin | 1(3.1) | 6(5.1) | >0.999 | 1(5.9) | 6(4.5) | 0.577 | 1(5) | 4(4.6) | >0.999 |
| Vasoactive drugs | 25(78.1) | 108(91.5) | 0.054 | 12(70.6) | 121(91) | 0.027 | 14(70) | 119(91.5) | 0.013 |
| Dobutamine | 25(100) | 106(98.1) | >0.999 | 12(100) | 119(98.3) | >0.999 | 14(100) | 117(98.3) | >0.999 |
| Milrinone | 0(0) | 15(13.9) | 0.041 | 0(0) | 15(12.4) | 0.220 | 0(0) | 15(12.6) | 0.222 |
| Sodium nitroprusiate | 5(20) | 38(35.2) | 0.079 | 1(8.3) | 42(34.7) | 0.042 | 2(14.3) | 41(34.5) | 0.062 |
| BIA | 4(12.5) | 43(36.4) | 0.010 | 1(5.9) | 46(34.6) | 0.023 | 1(5) | 46(35.4) | 0.004 |
| INTERMACS | |||||||||
| II | 7(21.9) | 46(39) | 0.109 | 3(17.6) | 50(37.6) | 0.052 | 3(15) | 50(38.5) | 0.038 |
| III | 23(71.9) | 69(58.5) | 12(70.6) | 80(60.2) | 15(75) | 77(59.2) | |||
| IV | 2(6.3) | 3(2.5) | 2(11.8) | 3(2.3) | 2(10) | 3(2.3) | |||
| BDI Classification | |||||||||
| No symptoms | 23(71.9) | 61(51.7) | 11(64.7) | 73(54.9) | 0.659 | 13(65) | 71(54.6) | 0.497 | |
| Mild | 7(21.9) | 31(26.3) | 0.178 | 5(29.4) | 33(24.8) | 6(30) | 32(24.6) | ||
| Exams | |||||||||
| Hemoglobin (g/dL) | 11.8±1.9 | 10.5±1.9 | 0.001 | 11.8±2.1 | 10.7±1.9 | 0.050 | 11.8±2.2 | 10.6±1.9 | 0.025 |
| Urea (mg/dL) | 61.1±35.9 | 56.3±28.3 | 0.872 | 61.5±32.1 | 56.8±29.8 | 0.561 | 66.3±35.4 | 55.9±29 | 0.241 |
| Creatinine(mg/dL) | 1.6±0.7 | 1.4±0.6 | 0.215 | 1.6±0.8 | 1.4±0.6 | 0.383 | 1.7±0.8 | 1.4±0.6 | 0.142 |
| RV function | |||||||||
| Mild | 6(18.8) | 35(29.7) | 0.499 | 4(23.5) | 37(27.8) | 0.918 | 4(20) | 37(28.5) | 0.632 |
| Moderate | 16(50) | 45(38.1) | 8(47.1) | 53(39.8) | 11(55) | 50(38.5) | |||
| LVEF | 22.9±8.1 | 25.6±8.1 | 0.002 | 23.6±10.5 | 25.2±7.8 | 0.030 | 23.8±9.6 | 25.2±7.9 | 0.052 |
Categorical variables are expressed as absolute numbers and percentages (n) and numerical variables as means and standard deviations (mean ± SD).
ACEIs = Angiotensin-converting enzyme inhibitors; ARBs = Angiotensin II receptor blockers; ISGLT2 = Sodium glucose cotransporter 2 inhibitors; BMI = Body mass index; IABP = Intra-aortic balloon pump; BDI = Beck Depression Inventory; RV = Right ventricle; LVEF = Left ventricular ejection fraction; CogF = Cognitive frailty.
Outcomes and competing risks analysis
In our cohort, during 1-year follow-up, 62 deaths (41.3%) and 101 transplants (67.3%) were recorded. Overall, the majority of events, both deaths and transplants, were concentrated within the first six months of the follow up.
In the competing risk analysis, assuming heart transplantation as a competing event, physical frailty showed a strong initial association with mortality risk (initial SHR = 5.47; 95% CI: 1.27–23.67; p = 0.023), with a significant attenuation of this effect over time (TVC SHR = 0.99; 95% CI: 0.97–1.00; p = 0.008), consistent with non-proportional hazards. Similarly, cognitive frailty, defined by either MoCA (initial SHR = 1.61; 95% CI: 1.12–2.32; p = 0.011) or WASI (initial SHR = 1.57; 95% CI: 1.27–23.67; p = 0.023) demonstrated an early association with mortality that progressively attenuated during follow-up in MoCA (TVC SHR = 1.00; 95% CI: 0.99–1.00; p = 0.011) and WASI (TVC SHR = 1.00; 95% CI: 0.99–1.00; p = 0.007), as showed in Table 4.
Table 4.
Physical and Cognitive Frailty in Relation to All-cause Mortality in a Competing-risk Analysis
| Variable | SHR | 95% CI | p- value | Adjusted SHR* | 95% CI | p-value |
|---|---|---|---|---|---|---|
| Physical frailty FRIED | ||||||
| Baseline | 5.47 | 1.27 – 23.67 | 0.023 | 3.39 | 0.76 – 15.11 | 0.110 |
| TVC | 0.99 | 0.97 – 1.00 | 0.008 | 0.99 | 0.97 – 1.00 | 0.061 |
| Cognitive frailty MoCA | ||||||
| Baseline | 1.61 | 1.12 – 2.32 | 0.011 | 1.27 | 0.88 – 1.85 | 0.210 |
| TVC | 1.00 | 0.99 – 1.00 | 0.011 | 0.99 | 0.99 – 1.00 | 0.145 |
| Cognitive frailty WASI | ||||||
| Baseline | 1.57 | 1.11 – 2.23 | 0.012 | 1.23 | 0.85 – 1.78 | 0.293 |
| TVC | 1.00 | 0.99 – 1.00 | 0.007 | 1.00 | 1.00–1.01 | 0.195 |
SHR, subdistribution hazard ratio; CI, confidence interval; IABP, intra-aortic balloon pump. Montreal Cognitive Assessment; WASI, Wechsler Abbreviated Scale of Intelligence. Heart transplantation was treated as a competing event. Physical and cognitive frailty was analyzed as a baseline and time-varying covariate (TVC). Multivariable models were adjusted for age, sex, and IABP use.
After adjustment for clinically relevant variables (age, sex, and IABP use) the initial association between physical frailty and mortality was attenuated and no longer statistically significant (initial SHR = 3.39; 95% CI, 0.76–15.11; p = 0.110). Likewise, the early associations between cognitive frailty, defined by MOCA (initial SHR = 1.27; 95% CI, 0.88–1.85; p = 0.210) or WASI (initial SHR = 1.23; 95% CI, 0.85–1.78; p = 0.293), and mortality were no longer significant after multivariable adjustment (Table 4). A time-dependent effect was observed, with higher subdistribution hazard ratios at baseline and progressive attenuation over time across all frailty measures.
The overall study design, conceptual framework and results are summarized in Central Figure (Figure 2).
Figure 2.
Central figure.
Discussion
In this cohort of patients with advanced heart failure listed for heart transplantation, we observed a very high prevalence of physical frailty (78.7%) and cognitive frailty, identified in more than 86% of patients using either MoCA or WASI. Depressive symptoms were also frequent, affecting approximately 44% of the sample, predominantly at mild to moderate levels. Both physical and cognitive frailty were associated with all-cause mortality at baseline; however, these associations attenuated over time and lost statistical significance after multivariable adjustment (age, sex, and IABP use).
A notable characteristic of our cohort was the high proportion of patients with Chagas disease (35%), which reflects the epidemiological relevance of Chagas cardiomyopathy among patients with advanced heart failure and those undergoing heart transplantation in Brazil. The 2025 Brazilian Transplant Registry (RBT) reported 473 heart transplants performed nationwide but did not provide the distribution of cardiomyopathy etiologies among transplant recipients [28]. Previous Brazilian transplant-center experiences have reported Chagas disease in approximately one-quarter of heart transplant recipients, including 26.2% in a major Brazilian heart transplant center [29,30]. However, the prevalence of Chagas disease is not homogeneous across Brazil, with a substantial concentration of cases in the state of São Paulo, where our transplant center is located, which may partly explain the relatively high proportion observed in our cohort.
Evidence directly addressing the association between Chagas disease and frailty remains scarce. However, available evidence suggests that Chagas disease may be associated with systemic manifestations potentially relevant to frailty, including sarcopenia and cognitive impairment [31,32]. Proposed mechanisms include chronic inflammatory activation, oxidative stress, autonomic dysfunction, and musculoskeletal alterations, which may contribute to reduced physiological reserve. However, in our cohort, heart failure etiology was not associated with the presence of physical or cognitive frailty, regardless of the instrument used to assess cognition, as shown in Table 3. Thus, despite the potential systemic effects of Chagas disease and the limited evidence suggesting a possible relationship with conditions related to frailty, our findings do not support an association between Chagas etiology and frailty in patients with advanced heart failure awaiting heart transplantation. Further studies are needed to clarify whether Chagas disease contributes to frailty through specific biological or clinical pathways.
The time-dependent nature of the association between frailty and mortality represents a key finding of this study. Rather than exerting a constant effect over time, frailty appeared to identify a period of early vulnerability following waitlist inclusion, during which patients were more likely to experience adverse outcomes. This dynamic behavior suggests that frailty may be particularly relevant for short-term risk stratification, while its prognostic value diminishes as clinical trajectories evolve.
One of the novelties of this study lies in the simultaneous assessment of physical and cognitive frailty in an understudied population of heart transplant candidates. The prevalence of physical frailty was higher than that reported in most international cohorts, likely reflecting differences in referral patterns and transplant logistics. In Brazil, heart transplantation is predominantly performed in hospitalized priority patients requiring intensive care and vasoactive support, a context that likely contributes to restricted mobility and reduced physical activity.
Cognitive impairment was highly prevalent and consistently identified across two distinct assessment approaches. MoCA and WASI demonstrated similar impairment patterns. Although the WASI provides a more comprehensive neuropsychological evaluation, it did not confer additional prognostic advantage over MoCA, supporting the latter as a practical screening tool in this setting.
Physical and cognitive frailty were predominantly associated with early mortality risk, with attenuation over time. After adjustment for IABP use, frailty was no longer independently associated with mortality, suggesting that frailty may partly reflect underlying disease severity and acute clinical instability. Given that IABP support identifies patients with more advanced hemodynamic compromise, overlap between frailty measures and disease severity is plausible and may influence risk estimates in this population. In this context, distinguishing between frailty as a biological construct and frailty as a marker of acute illness severity becomes particularly challenging.
A central finding of this study is the limited discriminatory capacity of the Fried Frailty Phenotype in patients with advanced heart failure awaiting transplantation. Although widely used, this instrument relies heavily on physical performance measures such as gait speed and physical activity, which are profoundly influenced by acute clinical status in this population. In our cohort, the near-universal presence of low physical activity (99.3%) and the absence of non-frail individuals suggest a ceiling effect, markedly reducing the ability of the instrument to differentiate risk across patients. In this context, components of the Fried phenotype may reflect acute hemodynamic compromise, hospitalization, and therapeutic restrictions rather than intrinsic biological frailty, leading to potential misclassification and attenuation of prognostic associations. This limitation is particularly relevant in critically ill populations, in whom functional impairment may be driven more by acute clinical constraints than by baseline physiological reserve.
Taken together, these findings highlight the multidimensional vulnerability of transplant candidates in a high-acuity inpatient setting and suggest that the prognostic relevance of frailty may be context-dependent and dynamic. They also underscore the need to develop and validate frailty assessment strategies better tailored to advanced heart failure, integrating cognitive, functional, and clinical parameters.16 Incorporating multidimensional frailty assessment into pre-transplant evaluation may improve early risk stratification and guide targeted interventions. Future multicenter studies are warranted to refine transplant-specific frailty constructs and evaluate whether interventions during this early vulnerability window can improve outcomes. Importantly, frailty alone should not preclude transplant eligibility, supporting a risk-based rather than exclusionary approach to candidate selection.
Limitations
Considering this is a single-center study conducted at a quaternary care hospital, some limitations should be highlighted. First, we included patients with high clinical severity, often managed in the intensive care unit, limiting generalizability to less severe ambulatory populations, which may limit external validity. Second, the small sample size of the study may suggest it was underpowered to detect independent effects after adjustment. Finally, the low educational level and the high prevalence of frailty in this cohort may have reduced discrimination between groups and attenuated the observed associations with outcomes.
Conclusion
In this cohort of heart transplant candidates, physical and cognitive frailty were highly prevalent and demonstrated a time-dependent association with early mortality risk. However, these associations were attenuated after adjustment for markers of disease severity, suggesting that frailty may partly reflect acute clinical status in high-acuity settings. These findings highlight important limitations of traditional frailty instruments in advanced heart failure and support the need for more tailored, multidimensional approaches to risk assessment in transplant candidates.
Data availability statement
The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request, in accordance with institutional and ethical regulations.
Funding
This work was carried out with support from the Coordination for the Improvement of Higher Education Personnel – Brazil (CAPES) – Funding Code 001.
CRediT authorship contribution statement
Erika Tiemi Ikeda, Juliana Araújo Nascimento: Research design, Article writing, Analytical tools, Data analysis, Luis Fernando B. da Costa Seguro, Mônica Samuel Avila, Victor Sarli Issa: Article writing, Research execution, Data analysis, Iascara W. de Campos, Fábio Antônio Gaiotto: Research design, Article writing, Research execution, Data analysis, Miguel Morita: Article writing, Research execution, Marcus Vinícius B. Santos, Maria Ignez Zanetti Feltrim, Edimar Alcides Bocchi, Silvia Helena Lage, Fernando Bacal: Article writing, Research execution, Analytical tools, Data analysis, Sandrigo Mangini, Fabiana G. Marcondes-Braga: Article writing
Declaration of Competing Interest
The authors declare no conflicts of interest.
Footnotes
Supplemental data associated with this article can be found in the online version at doi:10.1016/j.jhlto.2026.100680.
Appendix A. Supplemental material
Supplementary material
References
- 1.Pandey A., Kitzman D., Reeves G. Frailty is intertwined with heart failure: mechanisms, prevalence, prognosis, assessment, and management. JACC. 2019;73:1001–1011. doi: 10.1016/j.jchf.2019.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jha S.R., Ha H.S., Hickman L.D., et al. Frailty in advanced heart failure: a systematic review. Heart Fail Rev. 2015;20:553–560. doi: 10.1007/s10741-015-9493-8. [DOI] [PubMed] [Google Scholar]
- 3.Jha S.R., Hannu M.K., Gore K., et al. Prevalence and prognostic significance of frailty in patients with advanced heart failure referred for heart transplantation. Transplantation. 2016;100:429–436. doi: 10.1097/TP.0000000000000991. [DOI] [PubMed] [Google Scholar]
- 4.Rockwood K., Song X., MacKnight C., et al. A global clinical measure of fitness and frailty in elderly people. CMAJ. 2005;173:489–495. doi: 10.1503/cmaj.050051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fried L.P., Tangen C.M., Walston J., et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56:M146–M156. doi: 10.1093/gerona/56.3.m146. [DOI] [PubMed] [Google Scholar]
- 6.Azzopardi R.V., Beyer I., Gülistan F., et al. Increasing use of cognitive measures in the operational definition of frailty: a systematic review. Ageing Res Rev. 2018;43:10–16. doi: 10.1016/j.arr.2018.01.003. [DOI] [PubMed] [Google Scholar]
- 7.MacDonald P.S., Jepson N., Currie M., et al. Impact of frailty on mortality after heart transplantation. J Heart Lung Transplant. 2021;40:87–94. doi: 10.1016/j.healun.2020.11.007. [DOI] [PubMed] [Google Scholar]
- 8.Feola M., Garnero S., Palumbo D., et al. Correlation between cognitive impairment and prognostic parameters in patients with congestive heart failure. Arch Med Res. 2007;38:234–239. doi: 10.1016/j.arcmed.2006.10.004. [DOI] [PubMed] [Google Scholar]
- 9.Leto L., Feola M. Cognitive impairment in heart failure patients. J Geriatr Cardiol. 2014;11:316–328. doi: 10.11909/j.issn.1671-5411.2014.04.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Butts B., Gary R. Coexisting frailty, cognitive impairment, and heart failure: implications for clinical care. J Clin Outcomes Manag. 2015;22:38–46. [PMC free article] [PubMed] [Google Scholar]
- 11.Kobashigawa J., Dadhania D., Bhorade S., et al. Frailty in heart transplantation: report from the heart workgroup of a consensus conference on frailty. Am J Transplant. 2021;21:636–644. doi: 10.1111/ajt.16207. [DOI] [PubMed] [Google Scholar]
- 12.Nasreddine Z.S., Phillips N.A., Bédirian V., et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695–699. doi: 10.1111/j.1532-5415.2005.53221.x. [DOI] [PubMed] [Google Scholar]
- 13.Wechsler D. Wechsler abbreviated scale of intelligence – second edition (WASI-II) J Psychoeduc Assess. 2011;31(3):337–341. [Google Scholar]
- 14.Yates D.B., Schneider A., Viana M.E. Apresentação da escala de inteligência Wechsler abreviada (WASI) Aval Psicol. 2006 Dec;5(2):227–233. [Google Scholar]
- 15.Beck A.T., Ward C.H., Mendelson M., Mock J., Erbaugh J. An inventory for measuring depression. Archives of General Psychiatry. 1961;4(6):561–571. doi: 10.1001/archpsyc.1961.01710120031004. [DOI] [PubMed] [Google Scholar]
- 16.Denfeld Q.E., Jha S.R., Fung E., et al. Assessing and managing frailty in advanced heart failure: an International Society for Heart and Lung Transplantation consensus statement. J Heart Lung Transplant. 2023;42:1492–1505. doi: 10.1016/j.healun.2023.09.013. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary material
Data Availability Statement
The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request, in accordance with institutional and ethical regulations.