Presence of transthyretin amyloidosis cardiomyopathy influences the prognosis of patients with acute decompensated heart failure undergoing cardiac rehabilitation

Abstract

[Purpose] This study aimed to investigate the changes in physical function during hospitalization and their impact on the outcomes of patients with transthyretin amyloidosis cardiomyopathy who experienced acute decompensated heart failure and underwent acute cardiac rehabilitation. [Participants and Methods] A matched cohort of 18 and 54 patients with and without transthyretin amyloidosis cardiomyopathy, respectively, was created and analyzed. [Results] Compared to patients without transthyretin amyloidosis cardiomyopathy, those with transthyretin amyloidosis showed similar improvements in grip strength, quadriceps isometric strength, short physical performance battery, and usual gait speed during hospitalization. However, transthyretin amyloidosis cardiomyopathy was associated with a significantly increased risk of both rehospitalization due to heart failure and all-cause mortality. [Conclusion] Although the extent of changes in physical function during hospitalization was similar in patients with and without transthyretin amyloidosis cardiomyopathy, transthyretin amyloidosis cardiomyopathy was associated with poorer outcomes.

INTRODUCTION

As the population ages, the 1-year all-cause mortality and rehospitalization due to heart failure rates in patients with acute decompensated heart failure (ADHF) have been reported to be over 25% and 45%, respectively¹⁾. The background to this is that patients with ADHF often suffer from reduced physical function, which is thought to be highly likely to affect their outcome¹⁾. Over the past decade, progress in not only drug therapy but also non-drug therapy such as cardiac rehabilitation (CR) has been reported to improve the outcome of patients with ADHF¹⁾. Kitzman et al. reported that CR improves the short physical performance battery (SPPB) as a measure of physical performance and the 6-minute walking distance (6-MWD) as a measure of exercise tolerance in patients with ADHF, which may contribute to improved outcome²⁾. In Japan, it has been reported that cardiac rehabilitation at acute care hospitals has a beneficial effect on outcome, and that initiating CR within the third day of hospitalization is important^{3, 4)}.

As the population ages, the epidemiology of heart failure in patients with ADHF is changing, with not only the proportion of heart failure with preserved ejection fraction (HFpEF) compared to heart failure with reduced ejection fraction^{5, 6)}. Furthermore, HFpEF represents a diverse range of epidemiology of heart failure with limited proven treatments⁷⁾. Among these, CR is expected to improve not only physical function and exercise tolerance but also outcome in patients with ADHF and HFpEF, and there have been an increasing number of reports demonstrating its effectiveness to date⁸⁾.

However, the existence of “non-responders” who do not respond to CR has been reported⁹⁾. One of the reasons for this is thought to be the presence of cardiac amyloidosis.

Transthyretin amyloidosis cardiomyopathy (ATTR-CM) is a progressive disease caused by the deposition of misfolded transthyretin amyloid fibrils, leading to biventricular dysfunction, vascular stiffness, and impaired heart rate regulation^{10, 11)}. Prior to cardiac symptoms, the deposition of amyloid fibrils can cause carpal tunnel syndrome, lumbar spinal canal stenosis, and hearing loss, contributing to the progression of frailty¹²⁾, and may weaken the effectiveness of CR due to the rapid decline in activities of daily living. Tafamidis, one of the therapeutic drugs, has been reported to improve the prognosis of patients with ATTR-CM and suppress the decline in exercise tolerance over time, but this does not apply to patients with New York Heart Association (NYHA) class IV heart failure; its effects are only seen in patients with NYHA class I or II¹³⁾. Therefore, even among patients with ADHF, the presence or absence of ATTR-CM may affect changes in physical function during acute phase CR and subsequent outcome.

Recent research indicates that ATTR-CM comprises 13–30% of HFpEF cases, making it a not rare disease in heart failure epidemiology^{14, 15)}. Although patients with ATTR-CM may be non-responders to acute CR, no studies to date have considered the presence or absence of ATTR-CM in acute CR studies. This study aimed to investigate whether the presence or absence of ATTR-CM affects not only changes in physical function during hospitalization but also rehospitalization due to heart failure and all-cause death in patients with ADHF who underwent cardiac rehabilitation.

PARTICIPANTS AND METHODS

This study was approved by the Uwajima City Hospital Clinical Research Review Board (approval no. 2408-285) and conducted in accordance with the principles of the Declaration of Helsinki. Informed consent was not required due to the retrospective nature of the study; patients were given the opportunity to opt out.

Data from 1,063 patients with ADHF admitted to the Cardiology Department of Uwajima City Hospital between November 2018 and December 2023 were retrospectively reviewed. ADHF was diagnosed by a cardiologist in accordance with the 2017 Guidelines for the Diagnosis and Treatment of Acute and Chronic Heart Failure (Japanese Circulation Society/Japanese Heart Failure Association)¹⁶⁾. Patients who were suspected of having ATTR-CM because of conduction disorders such as left bundle branch block on electrocardiogram, atrial fibrillation, and severe left ventricular hypertrophy on echocardiogram underwent tests to diagnose ATTR-CM. The ATTR-CM diagnosis was confirmed by 99mTc pyrophosphate scintigraphy, myocardial biopsy, and genetic analysis based on the Japan Circulation Society 2020 Guidelines on Diagnosis and Treatment of Cardiac Amyloidosis¹⁷⁾. Cardiac function were comprehensively evaluated using electrocardiography, echocardiography, and coronary angiography. Admission data included demographic characteristics, comorbidities, NYHA functional classification, laboratory data, and assessment of physical function at the start of rehabilitation (grip strength, quadriceps isometric strength [QIS], SPPB, and usual gait speed). Discharge data included, single-leg standing time, and 6-MWD, in addition to the physical function assessments performed at admission. The inclusion criteria were: i) walking independently before admission, ii) evaluation at admission performed within 5 days of admission, iii) completing all physical function assessments at admission and discharge, iv) no severe dementia, and v) not taking tafamidis. The exclusion criteria were: i) missing data, ii) death during hospitalization, iii) inability to walk independently at discharge, iv) discharge or transfer within 7 days of admission, and v) plan to take tafamidis within 1 year. A physiotherapist measured each outcome.

Physical performance assessments at admission were performed at the bedside heart failure symptoms had stabilized and the first CR start date, and assessments at discharge were performed before discharge. The SPPB comprises standing balance, usual gait speed, and timed repeated chair rise, each scored on a scale of 0–4, yielding a total score of 0–12¹⁸⁾. Grip strength measurements were performed using a Smedley-type grip dynamometer (Grip-D; TAKEI, Niigata, Japan) with the patient in a natural stance with legs apart and elbows extended. Grip strength was measured twice for each hand, with the highest value used in the analysis. QIS was measured with a hand-held dynamometer (μ-Tas; ANIMA, Tokyo, Japan) fixed to a rigid bar and the patient sitting on a bed. Maximal isometric voluntary contractions of the quadriceps were recorded twice for each leg, with the knee joint angle fixed at 90° of flexion and the hip joint angle set at approximately 90° of flexion¹⁹⁾. The single-leg standing time records the time the patient was able to raise one leg with eyes open and both hands on the waist. The 6-MWD was measured according to the guidelines issued by the American Thoracic Society²⁰⁾.

Changes in physical performance from admission to discharge were shown as Δgrip strength, ΔQIS, ΔSPPB, and Δusual gait speed.

The standard CR program for heart failure was followed up by a physiotherapist with >3 years of experience in CR for ADHF. Briefly, the program was divided into the following stages: stage I, getting up from bed; stage II, 10-m walk; stage III, 40-m walk; stage IV, 80-m walk; stage V, two to three sets of 80-m walks; and stage VI, 300-m walk²¹⁾. Following completion of stage VI, aerobic exercise was initiated on an upright cycling ergometer.

The CR program for heart failure was performed for 20–40 min/day, 5 days/week.

Patients received follow-up care after discharge or as medically necessary, with a minimum of 6 months of follow-up at our hospital. Patients who were referred to the clinic were identified through telephonic contact with either the patient or their next of kin. The outcome of interest in this analysis was rehospitalization due to heart failure and all-cause death within one-year after discharge. Outcome was calculated as the number of days from the date of hospital discharge to the date of the event.

Patients without atrioventricular and bundle branch block on electrocardiogram or severe left ventricular hypertrophy on echocardiogram were defined as patients with non-ATTR-CM.

Propensity score matching (PSM) was performed to reduce potential confounding factors in the presence of ATTR-CM. The propensity model was performed using logistic regression analysis, with the presence or absence of ATTR-CM as the dependent variable and age, sex, BMI, NT-proBNP, LVEF, 1-year history of hospitalization for heart failure, CR start date after admission, atrial fibrillation, diabetes, hypertension, chronic kidney disease, dyslipidemia, and assessment of physical function at the start of rehabilitation (grip strength, QIS, SPPB, and usual gait speed) as independent variables to estimate the propensity score. Matching accuracy was evaluated using an area under the curve >0.7 value based on the receiver operating characteristic curve. Using the calculated propensity score, we used fixed ratio matching because this study had an imbalance in the number of samples in each group. Because Bottigliengo et al.²²⁾ reported that matching at a ratio of 1:3 or more results in a large relative bias, this analysis used 1:3 matching with a caliper value of 0.2. After propensity score matching, covariate balance was assessed using standardized mean differences, with values >0.1 considered indicative of residual imbalance. For covariates showing residual imbalance, additional adjustment was made in the Cox regression models to account for potential confounding.

The Shapiro–Wilk test was used to test for normality. Continuous variables are presented as means ± standard deviations for normally distributed data and medians (interquartile ranges) for nonnormally distributed data; categorical variables are presented as numbers (percentages).

Kaplan–Meier curves and log-rank tests were performed for 1-year rehospitalization due to heart failure and all-cause mortality events with or without ATTR-CM. Cox proportional hazards analyses were then performed for the presence or absence of ATTR-CM on each event, adjusting for gender. The results of Cox proportional hazard models were presented as hazard ratios and 95% confidence intervals. Concordance was used to evaluate the model’s predictive accuracy, and the likelihood ratio test was used to evaluate the regression coefficients. A post hoc power analysis was conducted using the “powerSurvEpi” package in R (version 4.3.2) for key outcomes with limited event numbers, particularly for all-cause mortality, to assess the robustness of the observed associations.

Statistical significance was set at p<0.05. All analyses were performed using JMP^® software version 8.0.2 (SAS Institute, Cary, NC, USA).

RESULTS

Of the 1,063 patients hospitalized for ADHF between November 2018 and December 2023, 67 (6.3%) were diagnosed with ATTR-CM. Of the 67 patients with ATTR-CM, 26 had missing data (12 not assessed at admission, 10 not assessed at discharge, and four not assessed at either time), four died in-hospital, eight were unable to walk before admission, and two were discharged or transfered within 7 days. Of the 936 patients with no-ATTR-CM, 566 were excluded due to missing data (n=309; 44 not assessed at admission, 26 not assessed at discharge, and 133 not assessed at either time), walking inability (n=75), severe dementia (n=68), in-hospital death (n=84), and discharge or transfer within 7 days of admission (n=30). Finally, 430 patients with no-ATTR-CM and 27 patients with ATTR-CM were included in the analysis (Fig. 1). After propensity score matching, 54 patients with no-ATTR-CM and 18 patients with ATTR-CM were required. The area under the curve was 0.862.

Fig. 1.

Screening and data analysis.

Cardiac rehabilitation (CR) was prescribed for all patients. *age, sex, BMI, NT-proBNP, LVEF, hospitalization due to heart failure within 1-year, CR start date after admission, atrial fibrillation, diabetes, hypertension, chronic kidney disease, dyslipidemia, and assessment of physical function at the start of rehabilitation (grip strength, QIS, SPPB, and usual gait speed). BMI: body mass index; LVEF: left ventricular ejection fraction; ATTR-CM: transthyretin amyloidosis cardiomyopathy; NT-proBNP: N-terminal pro-brain natriuretic peptide; QIS: quadriceps isometric strength; SPPB: short physical performance battery.

Patients’ characteristics and physical function assessment results before and after PSM are presented in Table 1. In terms of patient characteristics, the standardized mean difference for after PSM was ≤0.1 (the target value) for only age, diabetes, and dyslipidemia, but all items were within ≤0.25.

Table 1. Patient characteristics.

	Before PSM (n=457)			After PSM (n=72)
	no-ATTR-CM (n=430)	ATTR-CM (n=27)	SMD	no-ATTR-CM (n=54)	ATTR-CM (n=18)	SMD
Age (years)	77.1 ± 11.8	80.7 ± 5.7	0.391	79.39 ± 10.9	80.44 ± 6.2	0.10
Male n (%)	265 (61.6%)	22 (81.5%)	0.451	45 (83.3%)	14 (77.8%)	0.14
BMI (kg/m2)	24.3 ± 5.1	23.7 ± 4.0	0.121	24.0 ± 3.6	23.4 ± 2.5	0.19
Risk factors n (%)
Atrial fibrillation	232 (54.0%)	22 (81.5%)	0.616	42 (77.8%)	15 (83.3%)	0.141
Diabetes	147 (34.2%)	4 (14.8%)	0.462	5 (9.3%)	2 (11.1%)	0.061
Hypertension	199 (46.3%)	12 (44.4%)	0.037	26 (48.1%)	7 (38.9%)	0.188
Chronic kidney disease	134 (31.2%)	11 (40.7%)	0.201	14 (25.9%)	7 (38.9%)	0.250
Dyslipidemia	98 (22.8%)	4 (14.8%)	0.205	6 (11.1%)	2 (11.1%)	<0.001
Hospitalization due to heart failure within 1 year	145 (33.7%)	10 (37.0%)	0.069	16 (29.6%)	7 (38.9%)	0.196
NYHA class
Ⅲ/Ⅳ	399 (93.0%)	25 (92.6%)	0.016	50 (92.6)	16 (88.8)	0.128
Echocardiographic findings
LVEF (%)	51.1 ± 16.3	52.4 ± 11.5	0.093	52.6 ± 16.2	49.6 ± 11.9	0.211
Laboratory findings
NT-proBNP (pg/mL)	3,822.5 (1,961.0 to 7,294.5)	5,584.0 (2,926.0 to 14,571.0)	0.398	4,460.0 (2,648.8 to 7,587.3)	4,802.0 (2,344.5 to 13,152.8)	0.144
CR start date after admission (days)	2.0 (1.0 to 3.0)	2.0 (1.0 to 3.0)	0.205	2.00 (1.0 to 2.8)	2.0 (1.0 to 3.0)	0.134
Physical function assessment
Grip strength (kg)	22.6 ± 9.5	21.8 ± 8.0	0.098	23.8 ± 8.8	22.7 ± 7.6	0.134
QIS (kgf)	21.2 ± 10.6	22.4 ± 10.2	0.113	24.7 ± 11.5	22.0 ± 10.2	0.240
Total SPPB (score)	7.2 ± 3.3	7.0 ± 3.4	0.050	7.3 ± 3.5	6.7 ± 3.0	0.215
Usual gait speed (m/s)	0.65 ± 0.2	0.65 ± 0.2	0.011	0.67 ± 0.3	0.60 ± 0.3	0.250

PSM: propensity score matching; ATTR-CM: transthyretin amyloidosis cardiomyopathy; SMD: standardised mean differences; BMI: body mass index; NYHA: New York Heart Association; LVEF: left ventricular ejection fraction; NT-proBNP: N-terminal pro-brain natriuretic peptide; CR: cardiac rehabilitation; QIS: quadriceps isometric strength; SPPB: short physical performance battery.

Table 2 shows the progress of CR, length of hospital stay, medication status, and physical function assessment at the time of discharge and outcomes, while Table 3 shows the amount of change in physical function during hospitalization. No significant differences in angiotensin-converting enzyme inhibitor and angiotensin II receptor blocker use were noted after PSM. No significant difference was observed in physical function at the time of discharge either before or after PSM, and the amount of change in physical function obtained from CR during hospitalization was similar.

Table 2. Comparison of medication and physical function at discharge, and outcome between groups.

	Before PSM (n=457)			After PSM (n=72)
	no-ATTR-CM (n=430)	ATTR-CM (n=27)	p-value	no-ATTR-CM (n=54)	ATTR-CM (n=18)	p-value
CR progression: Gait distance on the Seventh day of hospitalization (m)	50.0 (21.3 to 100.0)	50.0 (32.5 to 100.0)		70.0 (20.0 to 150.0)	50.0 (21.3 to 100.0)
Length of hospital stay (days)	21.5 ± 10.9	20.1 ± 9.4		20.6 ± 9.8	20.4 ± 9.0
Medication n (%)
β blokcer	298 (69.3%)	13 (48.1%)	*	39 (72.2%)	7 (38.9%)	*
ACE-i/ARB	190 (44.2%)	5 (18.5%)	*	1 (1.9%)	2 (11.1%)
Loop diuretic	336 (78.1%)	23 (85.2%)		44 (81.5%)	15 (83.3%)
Aldosterone antagonist	271 (63.0%)	17 (63.0%)		31 (57.4%)	13 (72.2%)
Tolvaptan	141 (32.8%)	16 (59.3%)	**	18 (33.3%)	13 (72.2%)	**
Sacubitril/Valsartan	44 (21.4%)	3 (15.8%)		6 (25.0%)	1 (8.3%)
Sodium-glucose cotransporter 2	104 (50.2%)	10 (52.6%)		11 (45.8%)	4 (33.3%)
Physical function assessment
Grip strength (kg)	23.3 ± 9.4	21.9 ± 7.4		24.5 ± 9.0	21.8 ± 8.0
QIS (kgf)	23.9 ± 11.0	23.6 ± 9.4		26.5 ± 10.5	23.2 ± 10.4
Total SPPB (score)	9.4 ± 3.0	8.8 ± 2.9		8.9 ± 2.9	8.3 ± 2.4
Usual gait speed (m/s)	0.87 ± 0.34	0.83 ± 0.21		0.85 ± 0.25	0.78 ± 0.23
Single-leg standing time (s)	3.7 (0.7 to 14.1)	1.8 (0.4 to 4.6)	*	3.4 (0.6 to 16.1)	1.0 (0.4 to 3.4)
6-MWD (m)	270.0 (160.0 to 371.5)	244.0 (166.5 to 299.0)		253.5 (170.0 to 398.3)	234.5 (181.3 to 263.5)
1-year incidence of events n (%)
Rehospitalization due to heart failure	184 (42.8%)	14 (51.9%)		19 (35.2%)	12 (66.7%)	**
All-cause mortality	39 (9.1%)	8 (29.6%)	**	1 (1.9%)	7 (38.9%)	***

*p<0.05, **p<0.01, ***p<0.001.

PSM: propensity score matching; ATTR-CM: transthyretin amyloidosis cardiomyopathy; ACE-i: angiotensin-converting enzyme inhibitors; ARB: angiotensin II receptor blocker; SPPB: short physical performance battery; 6-MWD: 6-minute walk distance.

Table 3. Comparison of changes in physical function between groups.

	Before PSM (n=457)			After PSM (n=72)
	no-ATTR-CM (n=430)	ATTR-CM (n=27)	p-value	no-ATTR-CM (n=54)	ATTR-CM (n=18)	p-value
Δ Grip strength (kg)	+0.1 (−2.0 to 2.7)	−0.6 (−5.0 to −2.1)		+0.1 (−1.6 to −2.8)	−0.8 (−3.2 to −1.5)
ΔQIS (kgf)	+2.2 (−0.7 to −5.6)	+1.2 (−2.0 to −3.3)		+2.2 (−1.0 to −4.1)	+1.3 (−1.9 to −4.6)
ΔTotal SPPB (score)	+2.0 (1.0 to 3.0)	+2.0 (1.0 to 3.0)		+1.0 (0.3 to −3.0)	+2.0 (1.0 to −2.8)
ΔUsual gait speed (m/s)	+0.2 (0.1 to −0.3)	+0.2 (0.06 to −0.3)		+0.2 (0.0 to 0.3)	+0.2 (0.0 to −0.3)

PSM: propensity score matching; ATTR-CM: transthyretin amyloidosis cardiomyopathy; QIS: quadriceps isometric strength; SPPB: short physical performance battery.

Before PSM, 184 (42.8%) and 39 (9.1%) patients in the no-ATTR-CM group and 14 (51.9%) and eight (29.6%) patients in the ATTR-CM group experienced rehospitalization due to heart failure and all-cause mortality, respectively. After PSM, 19 (35.2%) and one (1.9%) patients in the no-ATTR-CM group and 12 (66.7%) and seven (38.9%) patients in the ATTR-CM group experienced rehospitalization due to heart failure and all-cause mortality, respectively (Table 2).

Before PSM, Kaplan–Meier curves showed no significant difference in the rate of rehospitalization due to heart failure in the ATTR-CM group (Fig. 2A), but there was a significant difference in the rate of all-cause mortality (Fig. 2B). However, after PSM, the ATTR-CM group was also associated with a significantly increased risk of rehospitalization due to heart failure (Fig. 2C, 2D).

Fig. 2.

Kaplan–Meier curves of the risk of outcome.

In Kaplan–Meier curves analysis, before propensity score matching, ATTR-CM was not associated with rehospitalization due to heart failure (log-rank: A, p=0.17 and B, p<0.001, respectively). After propensity score matching, ATTR-CM was significantly associated with rehospitalization due to heart failure and all-cause mortality (log-rank: C, p<0.01 and D, p<0.001, respectively). ATTR-CM: transthyretin amyloidosis cardiomyopathy.

In the Cox hazard model for the effect of ATTR-CM on days to event, only all-cause mortality was associated with ATTR-CM before PSM (rehospitalization due to heart failure; HR 1.5, 95% CI 0.8–2.5, p=0.18, concordance=0.513, likelihood ratio test=0.2 and all-cause mortality; HR 3.7, 95% CI 1.7–8.0, p<0.001, concordance=0.559, likelihood ratio test=0.003), but after PSM, ATTR-CM was also associated with rehospitalization due to heart failure rate (rehospitalization due to heart failure; HR 2.7, 95% CI 1.3–5.6, p<0.01, concordance=0.668, likelihood ratio test=0.004 and all-cause mortality; HR 27.1, 95% CI 3.3–221.1, p<0.001, concordance=0.884, likelihood ratio test ≤0.001) (Table 4).

Table 4. Impact of ATTR-CM on rehospitalization due to heart failure and all-cause mortality before and after PSM.

	Before PSM (n=457)			After PSM (n=72)
	HR	95% CI	p-value	HR	95% CI	p-value
Rehospitalization due to heart failure	1.5	0.8 to 2.5		2.7	1.3 to 5.6	**
All-cause mortality	3.7	1.7 to 8.0	***	27.1	3.3 to 221.1	***

**p<0.01, ***p<0.001.

ATTR-CM: transthyretin amyloidosis cardiomyopathy; PSM: propensity score matching; HR: heart rate.

DISCUSSION

This study investigated the change in physical function due to CR and its impact on prognosis in patients with or without ATTR-CM in ADHF. CR during hospitalization was associated with similar changes in physical function regardless of whether patients had ATTR-CM or not. However, at 1-year follow-up, rehospitalization due to heart failure and all-cause mortality were significantly associated with increased risk in the ATTR-CM group.

Dyspnea in acute heart failure is caused not only by pulmonary congestion but also by a wide range of other factors, including sympathetic nervous system activity, arterial insufficiency with microcirculatory disorders, and neurohormonal activity²³⁾. Given this background, outcomes of CR in patients with ADHF are often based on ADL and movement ability^24,25,26). Among these, physical performance, as represented by the SPPB, is strongly related to prognosis and has the advantage of being able to be used for continuous evaluation during hospitalization. It has been reported that an SPPB score of 7 or less at the time of discharge is associated with a poor prognosis²⁷⁾. In this study, the SPPB score at discharge was 8 or higher in both groups, and the median change during hospitalization was +1.0–2.0. In previous studies, during acute hospitalization, the SPPB scores improved by +2.2 ± 2.4 score in those in their 80s^{28, 29)}, similar results were observed in the ATTR-CM group. In ATTR-CM, muscle metabolic reflexes are activated more than in other types of heart failure, leading to the perception of fatigue and reduced skeletal muscle function due to sympathetic nerve activation³⁰⁾, and this may lead to the progression of disuse syndrome after hospitalization. Considering that the decline in physical function contributes to the prognosis of patients with ADHF¹⁾, it makes sense that start of CR within 3 days of hospitalization could improve the prognosis⁴⁾. Because all patients in this study underwent CR, it is not possible to compare the effects of CR; however, we believe that, at the very least, by performing acute phase CR, the decline in physical function may help prevented. Therefore, in this study, changes in physical function were similar regardless of whether or not patients had ATTR-CM, which is believed to be due to the early start of CR, the similar progress of CR, and the similar prevention of disuse during hospitalization.

Although exercise tolerance assessed by the 6-MWD, an important prognostic indicator, was similar to the mean value for patients in their 80s (221.0 m)²⁹⁾ in both groups (no-ATTR-CM 253.5 m, and ATTR-CM 234.5 m), rehospitalization due to heart failure and all-cause mortality was significantly higher in the ATTR-CM group (rehospitalization due to heart failure; 35.2 vs. 66.7%, p=0.028 and all-cause mortality; 1.9 vs. 38.9%, p<0.001). Also, patients with ATTR-CM had higher than the heart failure rehospitalization and mortality rates reported by Liu et al., which were 25% and 45%, respectively¹⁾. Ingle et al. has reported that survival rates are reduced in patients with chronic heart failure with 6-MWD ≤240m³¹⁾. Nicol et al. has reported that a high-risk group for event occurrence in ATTR-CM using cutoff values ​​of Peak VO₂ ≤ 13 mL/kg/min and NT-proBNP ≥1,800 ng/L¹¹⁾. Using the formula of Cahalin et al. {Peak VO₂ = 6-MWD (m) × 0.03 + 3.98}, a Peak VO₂ of 13 mL/kg/min corresponds to a 6-MWD of 300 m³²⁾. The ATTR-CM group in this study had higher NT-proBNP and lower 6-MWD(4802.0 pg/mL and 234.5 m, respectively) than in previous studies^{11, 31)}, and therefore is considered to be at high risk of events.

Berthelot et al.³³⁾ reported that patients with ATTR-CM presenting with acute heart failure have a poor prognosis, and Wernhart et al.³⁴⁾ reported that the presence of ATTR-CM is associated with a worsening of various indices in cardiopulmonary exercise testing. Furthermore, in patients with ATTR-CM, progression of diastolic dysfunction, biventricular contractile dysfunction, and accumulation of amyloid protein in the pulmonary arteries causes further reduction in exercise tolerance and effective ventilation³⁵⁾. We also reported that in patients with ATTR-CM treated with tafamidis, despite good exercise tolerance at baseline, the VE/VCO₂ slope, an indicator of sudden death, was poor, and after one year, AT VO₂, Peak VO₂, oxygen pulse, and ΔVO²/WR decreased by 20% as frailty progressed³⁶⁾. Broussier et al.¹²⁾ reported that the prognosis worsens as frailty in ATTR-CM progresses. This suggests that the ATTR-CM group experienced a significant decline in physical function and exercise tolerance over time, even when they had good physical function and exercise tolerance at the time of discharge, which may have contributed to the decreased survival rate. Future studies should track changes in physical function over time and investigate associations with survival.

This study has several limitations. First, it was a single-center, retrospective study with a small sample size. Second, patients receiving tafamidis treatment to investigate the prognostic impact of acute phase CR in patients with ATTR-CM were excluded. It has been reported that tafamidis can suppress the decline in exercise tolerance over time. Therefore, the medication status may affect the prognosis. Third, although this study used propensity score matching, there may have been many confounding factors that could not be adjusted for. The prognosis for ATTR-CM may be particularly influenced by the long-term disease and the presence of frailty. Fourth, the number of all-cause deaths was limited (n=8), which may lead to imprecision in the estimation of the hazard ratio. A post hoc power analysis using the powerSurvEpi package in R revealed that the statistical power for detecting the observed hazard ratio of 27.1 was approximately 25.6%. This indicates a high risk of type II error and suggests that the observed association should be interpreted with caution. Fifth, the possibility of potential bias cannot be excluded because not all patients with no-ATTR-CM were diagnosed with ATTR-CM. In particular, in the case of very elderly patients, scintigraphy may be performed but invasive evaluations such as myocardial biopsy are often not performed. Finally, due to the regional characteristics of our hospital, outpatient CR was not performed for all patients in this study. Because continued outpatient CR from the acute phase has been reported to improve prognosis³⁷⁾, it should be investigated whether the implementation of outpatient CR changes the outcomes of ATTR-CM.

This study reported the change in physical function after CR and its impact on prognosis depending on the presence or absence of ATTR-CM in ADHF. These results suggest that ATTR-CM is associated with a poor prognosis, even though physical function and exercise tolerance are similar to those of no-ATTR-CM. Given the poor outcome associated with delayed diagnosis, early identification of ATTR-CM is essential. When precursor symptoms such as carpal tunnel syndrome or lumbar spinal canal stenosis are present, clinicians should consider ATTR-CM and promptly refer patients to a specialized center for further evaluation.

Funding

This study received no grants from public, commercial, or not-for-profit funding agencies.

Conflicts of interest

The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.