Reverse Left Ventricular Remodeling in a Patient With Concurrent Wild-Type Transthyretin Cardiac Amyloidosis and Heart Failure With Reduced Ejection Fraction: The Role of Conventional Heart Failure Therapy

Abstract

Wild-type transthyretin amyloid (ATTRwt) amyloidosis is a systemic, progressive disease that often affects older adults and is characterized by ATTRwt deposition in multiple organs and tissues. When the heart is involved, this deposition leads to cardiac amyloidosis (CA), which can result in life-threatening heart failure (HF) and arrhythmias. In general, as left ventricular (LV) wall thickening increases because of ATTRwt deposition, the LV cavity decreases in size. Additionally, as the disease progresses, the myocardial ATTR burden increases, and the LV ejection fraction (LVEF) decreases. Thus, at ATTRwt-CA diagnosis, the most common LV characteristics are a slightly reduced LV cavity and preserved or mildly reduced LVEF. We report the case of an 83-year-old man with HF with reduced EF (HFrEF) and ATTRwt-CA who had an LVEF of 22% and a dilated LV cavity with an end-diastolic diameter (EDD) of 58.0 mm and end-diastolic volume index (EDVI) of 86 mL/m². Additionally, the LV wall thickness was 11.3 mm, and his National Amyloidosis Centre Stage, determined using N-terminal pro-B-type natriuretic peptide and estimated glomerular filtration rate, was 1, suggesting early-stage ATTRwt-CA despite reduced LVEF and a dilated LV cavity. Conventional medications for HFrEF, including loop diuretics and neurohumoral blockers, beta-blockers, angiotensin-converting enzyme inhibitors, and mineralocorticoid receptor antagonists, were administered. Nonetheless, disease-modifying drugs for ATTRwt-CA were not administered because of dementia and frailty. Three years later, "discordant" LV remodeling was observed, with improvement in LVEF and LV dilatation, alongside inappropriate progression of LV wall thickening: LVEF increased to 36%, LVEDD decreased to 51.5 mm, and LVEDVI decreased to 66 mL/m², whereas LV wall thickness increased to 13.2 mm. Subsequently, tachycardiac atrial fibrillation and new decompensated HF developed. Early initiation of combination therapy with disease-modifying agents for ATTRwt-CA, currently the only standard therapies proven to modify disease progression, together with guideline-directed HF therapy, may be the optimal approach.

Introduction

Transthyretin, a protein primarily synthesized in the liver, forms tetramers in the bloodstream and transports thyroxine and retinol [1]. Due to aging - the exact mechanism of which remains undetermined - or genetic mutations, these tetramers can dissociate, becoming structurally unstable and forming transthyretin amyloid (ATTR) fibrils. Wild-type ATTR (ATTRwt) amyloidosis, which has no mutation in the transthyretin gene, is a systemic disease that often affects older adults and is characterized by ATTRwt deposition in the interstitium of systemic organs and tissues [2]. ATTRwt deposition in tendons, ligaments, and the heart causes carpal tunnel syndrome, spinal canal stenosis, and cardiac amyloidosis (CA) [1,2].

ATTRwt-CA has an infiltrative and relentlessly progressive nature. In ATTRwt-CA, interstitial ATTRwt deposition in the myocardium disrupts myocardial architecture and impairs systolic and diastolic functions, leading to life-threatening heart failure (HF) and arrhythmias [1]. ATTRwt-CA is a common disease accounting for approximately 10% of older patients with HF, irrespective of left ventricular (LV) ejection fraction (EF) (LVEF), and among those with LV hypertrophy, such as hypertrophic cardiomyopathy and aortic valve stenosis [3]. In general, as ventricular wall thickening increases because of ATTRwt deposition, ventricular cavity size diminishes [1]. In addition, as the disease progresses, the myocardial ATTR burden increases, and the LVEF decreases [4]. Thus, most patients exhibit normal to nearly normal LV cavity size with preserved EF (LVEF ≥ 50%) or mildly reduced EF (LVEF > 40% but < 50%) at the time of ATTRwt-CA diagnosis. However, some patients experience reduced EF (LVEF ≤ 40%) accompanied by LV enlargement [3,5]. ATTRwt-CA is often well tolerated until rhythm disturbances, including atrial fibrillation (AF), or severe diastolic dysfunction due to restrictive physiology resulting from marked wall thickening occur [1]. Nonetheless, without disease-modifying therapy for ATTR-CA, life expectancy is approximately three years after diagnosis [3]. Tafamidis, the first disease-modifying medication for ATTRwt-CA shown to improve mortality and morbidity [6], was approved in Japan and the United States in 2019, and in the European Union and other regions in 2020 [7].

Cardiac remodeling, characterized by changes in cardiac geometry and function leading to HF decompensation and poor prognosis, can occur in response to various myocardial diseases with or without myocyte loss and pressure/volume overload, such as systemic hypertension [8]. During this process, molecular, cellular, biochemical, structural, and functional changes occur. Key pathological mechanisms include alterations in LV geometry, myocardial hypertrophy, myocardial fibrosis, inflammation, neurohormonal activation in the sympathetic nervous system and the renin-angiotensin-aldosterone system, as well as changes in natriuretic peptides [9]. Several therapies targeting these maladaptive processes, which play a central role in the development of adverse remodeling, have been shown to delay or even reverse ventricular dilation and dysfunction [8]. This process is known as "reverse remodeling" and is associated with improved prognosis. In the pharmacological management of HF without ATTRwt-CA, the choice of medication depends on LVEF [10]. In cases of HF with reduced EF (HFrEF) - where the LV is typically dilated - beta-blockers, angiotensin-converting enzyme inhibitors (ACEi)/angiotensin II receptor blockers (ARB)/angiotensin receptor-neprilysin inhibitors (ARNI), mineralocorticoid receptor antagonists (MRA), and sodium-glucose cotransporter 2 inhibitors (SGLT2i), as well as loop diuretics, are used. In HF with preserved EF (HFpEF), only MRA and SGLT2i are recommended in addition to loop diuretics. Disease-modifying medications for ATTR-CA merely slow disease progression rather than reversing the disease course [7,11]. Therefore, they do not improve LVEF when used alone [12]. Recently, the importance of conventional HF medications in addition to disease-modifying therapy has been reported, as improved prognostic effects have been observed [13]. Nonetheless, reports on changes in LV remodeling remain scarce [14].

We report a case of HFrEF and early-stage ATTRwt-CA with an enlarged LV cavity and mildly increased LV wall thickness. Conventional HF medications, including neurohormonal blockade but not disease-modifying drugs for ATTRwt-CA, were administered. Subsequently, reversal of LV remodeling occurred, characterized by improved LVEF and reduced LV chamber size. However, LV wall thickness increased, and persistent AF with decompensated HF developed.

Case presentation

In 2021, an 83-year-old Japanese man (height: 158.5 cm, weight: 60.1 kg, body mass index (BMI): 23.9 kg/m²) was admitted to our hospital for the first time, presenting with a two-day history of productive cough, fatigue, and exertional dyspnea. He reported no history of alcohol consumption and a 25-year smoking history of 20 cigarettes per day; however, he had quit smoking at 45 years of age. He had undergone surgery for benign prostatic hypertrophy 25 years before presentation and had no history of conditions associated with a high suspicion of ATTR-CA, such as carpal tunnel syndrome, lumbar spinal stenosis, biceps tendon rupture, rotator cuff tear, or autonomic dysfunction.

Physical examination in the emergency department revealed a body temperature of 36.3 °C, pulse rate of 106 beats/min, systemic blood pressure of 143/106 mmHg, and 91% oxygen saturation on room air (measured using a pulse oximeter). A third heart sound and moist rales were audible on auscultation. The liver was palpable at a one-finger breadth along the right midclavicular line below the costal margin, and moderate pretibial edema was observed. Blood tests revealed hypoalbuminemia; elevated liver enzymes likely caused by congestive hepatopathy; elevated plasma B-type natriuretic peptide (BNP), high-sensitivity cardiac troponin I (hs-cTnI), and D-dimer levels; and a decreased estimated glomerular filtration rate (eGFR) (Table 1). Arterial blood gas analysis on room air revealed hypoxemia and metabolic acidosis due to lactic acidosis resulting from hypoxemia or/and hypoperfusion, with appropriate respiratory compensation.

Table 1. Laboratory test values of the patient upon admission and after treatment.

ALT, alanine aminotransferase; BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; HCO₃^-, bicarbonate ion; hs-cTnI, high-sensitivity cardiac troponin I; PCO₂, partial pressure of carbon dioxide; PO₂, partial pressure of oxygen.

*Determined through arterial blood gas analysis in room air.

Blood test parameter	Result			Reference range
	On the first admission	30 months after the first admission	On the second admission (36 months after the first admission)
Albumin (g/dL)	3.7	3.5	2.8	3.8-5.3
ALT (U/L)	39	12	10	0-40
Creatine kinase (U/L)	65		32	≤190
eGFR (mL/min per 1.73 m2)	46	43	43	90-120
Potassium (mEq/L)	4.3	4.8	4.0	3.6-5.0
hs-cTnI (pg/mL)	128.1		272.7	≤18.4
BNP (pg/mL)	526.1	244.7	703.8	≤18.4
Glycated hemoglobin A1c (%)	6.0	5.3	6.0	4.6-6.2
White blood cells (/μL)	8,700	4,900	7,400	4,000-8,500
Hemoglobin (g/dL)	15.0	10.7	10.0	11.0≤
Platelet (/μL)	19.6 × 104	18.6 × 104	25.6 × 104	14-40 × 104
D-dimer (μg/mL)	6.4	0.3		≤1.0
pH	7.441*			7.35-7.45
PCO2 (mmHg)	30.2*			35-45
PO2 (mmHg)	57.9*			80-100
HCO3- (mmol/L)	20.1*			24.2-29.8
Base excess (mmol/L)	-2.7*			-2.5-2.5
Anion gap (mmol/L)	14.9*			10-18
Lactic acid (mmol/L)	2.0*			0.4-1.4

Chest radiography and computed tomography (CT) revealed pulmonary congestion, pleural effusion, emphysema, and interstitial fibrosis (Figure 1A, Figure 2).

Figure 1. Plain film chest radiography images obtained on the day of the first admission (A), before discharge (B), and three years later at another hospital (C).

(A and B) Chest radiography images obtained in the standing position. (C) Chest radiography image obtained in the sitting position. Panels A and C show pulmonary congestion and pleural effusion (arrows). Panel B shows improvement of pulmonary congestion and pleural effusion.

Figure 2. Chest computed tomography images acquired during the first admission showing the coronal view (A) and axial views (B, upper dotted line; C, lower dotted line).

Pulmonary congestion, pleural effusion (asterisks), emphysema, and interstitial fibrosis are observed.

Electrocardiography revealed sinus rhythm, with a heart rate of 105 beats/min, low voltage in the limb leads, and premature supraventricular contractions (Figure 3A).

Figure 3. Electrocardiograms obtained during the first admission (A) and three years later (B).

Panel A shows sinus rhythm with a heart rate of 105/min, low voltage in the limb lead, and a premature supraventricular beat. Panel B shows atrial fibrillation, premature ventricular beat, low voltage in the limb lead, poor R-wave progression, decrease in the R-wave amplitude in leads V4-V6, and ST-T abnormalities in leads II, III, aVF, V5, and V6 likely caused by the digitalis effect (arrows).

Transthoracic echocardiography revealed mild LV wall thickening with an interventricular septal thickness of 11.3 mm, LV dilation with an LVEDD of 58.0 mm, an end-diastolic volume index (EDVI) of 86 mL/m², and reduced wall motion throughout the LV (Figure 4, upper row). The LVEF was 22%, and the LV mass index was 162 g/m². Moderate mitral regurgitation caused by mitral leaflet tethering, apical sparing, and markedly deteriorated LV global longitudinal strain of -7.2 was also observed.

Figure 4. Transthoracic echocardiograms obtained during the first admission (A-C) and three years later (D-F).

(A, Aʹ, B, D, Dʹ, and E) Apical long-axis views at end-diastole (A and D), mid-systole (B and E), and end-systole (Aʹ and Dʹ). (C and F) Bull’s eye map (apex at the center of the color-coded map) showing segmental longitudinal left ventricular (LV) peak systolic strain values based on the 16-segment model. Diffusely hypokinetic LV wall motion with an ejection fraction (EF) of 22% (asterisk in C), LV dilation with LV end-diastolic diameter (LVEDD) of 58.0 mm and end-diastolic volume index (EDVI) of 86 mL/m², and moderate mitral regurgitation (MR) caused by mitral valve tethering (arrow in B) at admission are shown. Mild interventricular septum thickness (IVST) of 11.3 mm, increased LV mass index of 162 g/m², and reduced LV global longitudinal peak systolic strain (GLPS) of -7.2 (hashtag in C) are also shown. Three years later, an EF of 37% (asterisk in F), mild MR (arrow in E), LVEDD of 51.5 mm, LVEDVI of 66 mL/m², IVST of 13.2 mm, LV mass index of 180 g/m², and GLPS of -6.5 (hashtag in F) are shown. The apical-to-basal strain ratio, which was calculated as the apical septal longitudinal strain divided by the average of the basal septal (anteroseptal and inferoseptal) longitudinal strains in (C), was 4.3, indicating apical sparing (cut-off > 2.1). However, the apical-to-basal strain ratio in (F) was 1.3, indicating the absence of apical sparing.

AA, ascending aorta; LA, left atrium

Therefore, the patient was diagnosed with acute decompensated HFrEF with mild LV wall thickening. Enalapril maleate (2.5 mg/day) and spironolactone (12.5 mg/day) were initiated. Additionally, intravenous furosemide was initiated and was later changed to oral azosemide. The signs and symptoms of acute decompensated HF gradually improved. On day six of hospitalization, bisoprolol fumarate (0.625 mg/day) was initiated and was later increased to 1.25 mg/day. Paroxysmal AF was observed during hospitalization; therefore, anticoagulants (edoxaban tosilate hydrate 30 mg once daily) were administered. However, because the patient already had interstitial lung disease (Figure 2), the administration of amiodarone for the prevention of paroxysmal AF associated with HFrEF was withheld [15].

We investigated potential causes of HFrEF. First, coronary angiography showed no significant atherosclerotic disease. Second, technetium-99m-pyrophosphate (Tc-99m-PYP) scintigraphy for the detection of ATTR-CA, performed because of LV wall thickening and an apical sparing pattern on echocardiography, demonstrated tracer uptake in the oblique muscles, as well as in the myocardium with a semi-quantitative visual grade of 3 (myocardial uptake greater than rib uptake) and a heart-to-contralateral lung ratio of 1.886 (Figure 5, Figure 6A).

Figure 5. Single-photon emission computed tomography/computed tomography fusion images of technetium-99m-labeled pyrophosphate (Tc-99m-PYP) scintigrams obtained 2 hours after injecting 740 megabecquerel radiotracer.

The images show the axial plane at the level of the heart (A) and the short axial plane of the heart (B). Tc-99m-PYP accumulation was observed in the left ventricular myocardium, including the interventricular septum (white arrows) and right ventricular myocardium (yellow arrows).

Figure 6. Single-photon emission computed tomography (SPECT)/computed tomography (CT) fusion images of technetium-99m-labeled pyrophosphate (Tc-99m-PYP) scintigraphy obtained two hours after injecting 740 megabecquerel radiotracer, and Tc-99m-PYP imaging-based CT-guided core needle biopsy of the right internal oblique muscle.

SPECT/CT fusion axial image of the abdomen (A) shows Tc-99m-PYP uptake in the oblique muscles (white arrows). The biopsy target was the site with the highest Tc-99m-PYP uptake in the right internal oblique muscle. CT images were obtained after the patient was placed in the supine oblique position on the CT scanner table, and the CT biopsy grid was applied to the skin at the biopsy site (light green arrows) (B). The introducer stylet and cannula of an 18-gauge spring-loaded semi-automatic biopsy needle (orange arrow) were advanced into the right internal oblique muscle through a small skin incision to obtain muscle specimens.

Therefore, we suspected ATTR-CA. Monoclonal protein studies, including serum and urine immunofixation electrophoresis and serum-free light-chain assays, revealed no evidence of plasma cell dyscrasia. Tc-99m-PYP imaging-based CT-guided core needle biopsy of the internal oblique muscle was performed without complications (Figure 6) [16]. An endomyocardial biopsy was not performed.

Congo red staining and immunohistochemical staining of biopsy specimens, evaluated at a specialized amyloid center in Japan, confirmed ATTR deposition (Figure 7). Transthyretin gene sequencing revealed no variants.

Figure 7. Histopathological images of specimens obtained from a biopsy of the right internal oblique muscle.

Congo red staining revealed amyloid deposits showing red-orange under light microscopy (A) and apple-green birefringence under cross-polarized light microscopy (B) (arrows). Immunohistochemical staining using a panel of type‐specific antibodies against the most common amyloidogenic proteins causing cardiac amyloidosis, including anti‐transthyretin 115-124 (C), anti‐kappa light chain 116-133 (D), anti‐lambda light chain 118-134 (E), and anti‐amyloid A (F), showed positivity for anti‐transthyretin 115-124 (C, arrows), while antibodies against other amyloid proteins were negative (D-F). The scale bars in all panels indicate 500 μm.

Therefore, the patient was diagnosed with ATTRwt amyloidosis and classified as National Amyloidosis Centre (NAC) Stage 1, based on eGFR and N-terminal pro-BNP (NT-proBNP), which was calculated using the validated conversion formula from plasma BNP values, of 2,988 pg/mL (log NT-proBNP = 1.21 + 1.03 × log BNP − 0.009 × BMI − 0.007 × eGFR) [17,18]. Although the patient had ATTRwt-CA, we determined that standard treatment for HFrEF was necessary and decided to continue MRA, half the standard dose of ACEi, and one-quarter the standard dose of beta-blockers. The patient was discharged on day 21 of hospitalization. Chest radiography performed immediately before discharge indicated the disappearance of pulmonary congestion and pleural effusion (Figure 2B).

Treatment with tafamidis was planned; nonetheless, three months after discharge, he developed a cerebral embolism, likely of cardiac origin, resulting in right-sided hemiplegia, although electrocardiography revealed sinus rhythm (not shown). Subsequently, cognitive impairment occurred, and frailty progressed. Edoxaban tosilate hydrate was continued to prevent systemic embolization, and conventional HF treatment was continued. Tafamidis administration was no longer considered.

At 30 months post discharge, blood tests obtained at an outpatient visit revealed anemia and decreased plasma BNP level, with little change in eGFR (Table 1). The plasma D-dimer level was within the normal range. Electrocardiography demonstrated sinus rhythm (not shown).

Tachycardiac AF and acute decompensated HF developed 36 months after discharge (Figure 1C), and the patient was admitted to another hospital. The loop diuretic dose was temporarily increased, and pulmonary congestion improved. Dapagliflozin propylene glycolate hydrate (10 mg once daily) and digoxin (0.125 mg once daily) were also initiated at another hospital, in addition to bisoprolol fumarate (1.25 mg once daily) and spironolactone (12.5 mg once daily). Nevertheless, enalapril maleate was withheld because of hypotension. The serum digoxin concentration measured seven days after digoxin initiation was 1.1 ng/mL (therapeutic concentration: 0.8-2.0 ng/mL), and the digoxin dose was reduced to 0.07 mg once daily. On day 21 after admission to the other hospital, the patient was transferred to our hospital.

On admission to our hospital, blood tests revealed hypoalbuminemia and anemia (Table 1). Additionally, plasma BNP and hs-cTnI levels increased further compared with the test results at the first admission. Pulmonary congestion had subsided, and an electrocardiogram showed persistent AF (Figure 3B). Compared with the findings at the first admission (Figure 4, upper panel), echocardiography revealed a decreased LVEDD of 51.5 mm and LVEDVI of 66 mL/m², improved LV wall motion with an EF of 36%, and improved mitral regurgitation, despite increased LV wall thickness with an interventricular septal thickness of 13.2 mm and an LV mass index of 180 g/m² (Figure 4, lower panel). The same medications prescribed at the other hospital were administered at our hospital.

Subsequently, the patient died on day 10 of hospitalization because of profound bradyarrhythmia. However, the details were unclear. The patient's clinical course, changes in echocardiographic findings, and prescribed medications are shown in Figure 8. Serum digoxin concentration was not measured during the second hospital admission.

Figure 8. Summary figure.

The patient experienced an initial episode of acute decompensated heart failure (1st ADHF), and treatment was initiated with loop diuretics, mineralocorticoid receptor antagonists (MRAs), beta-blockers, and angiotensin-converting enzyme inhibitors (ACEis). During this hospitalization, atrial fibrillation was first detected on electrocardiography (FD-AF), and anticoagulants were prescribed. Treatment with loop diuretics, MRAs, beta-blockers, and ACEis was continued; however, a second episode of ADHF (2nd ADHF) occurred, and persistent atrial fibrillation (P-AF) was also noted three years after the 1st ADHF. At the time of the 2nd ADHF, the ACEis were discontinued, and digoxin (D) and sodium-glucose cotransporter-2 inhibitors (Ss) were initiated instead. The patient died one month after the onset of the 2nd ADHF. Owing to conventional heart failure treatment, the left ventricular ejection fraction (LVEF) increased from 22% to 36%, and the left ventricular end-diastolic volume index (LVEDVI) decreased from 86 mL/m² to 66 mL/m².

Discussion

We encountered a case of an unusual presentation of concurrent early-stage ATTRwt-CA and HFrEF with an enlarged LV cavity. The cause of HFrEF in our patient was considered to be non-amyloid cardiomyopathy, although the underlying myocardial disease responsible for HFrEF remained unclear. Conventional HF medications, including beta-blockers, ACEi, and MRAs, were administered, whereas disease-modifying drugs for ATTRwt-CA were not. This treatment led to "discordant LV remodeling" characterized by improved LVEF and reduced LV cavity size, but increased LV wall thickness and the occurrence of persistent AF.

In this case, a reduced LVEF of 22% and a dilated LV cavity with an LVEDD of 58.0 mm and LVEDVI of 86 mL/m² were observed at the time of diagnosis. In general, patients with ATTRwt-CA may exhibit varying degrees of LVEF reduction due to systolic dysfunction; across the overall patient population, the average LVEF is approximately 50% [5,12,19]. Approximately 10% of older patients with HFrEF have ATTR-CA [3]. Achten et al. [5] reported that 31% of patients with ATTR-CA presented with HFrEF, and among them, 17% (5% of the total) exhibited severe LV dilation (LVEDD > 58 mm or LVEDVI ≥ 75 mL/m²), as seen in this case. Ioannou et al. [19] also reported that 22% of patients had an LVEF of 40% or less, although they did not report the size of the LV cavity. In a report by Sheikh et al. [4] examining the relationship between amyloid burden assessed by extracellular volume on cardiac magnetic resonance (CMR) and echocardiographic findings, the results showed that, as amyloid burden increases, LV wall thickness increases and LVEF decreases; however, LV cavity size initially decreases and then increases. The mean interventricular septal wall thickness in cases of HFrEF reported by Achten et al. [5] and Ioannou et al. [19] was 16-17 mm, which is considerably thicker than the value observed in our case (11.3 mm at diagnosis). This finding is consistent with cases showing low myocardial ATTR burden reported by Sheikh et al. [4]. Thus, we considered that early-stage ATTRwt-CA coincidentally coexisted with HFrEF in this case. However, given the incidence of ATTRwt-CA [3], such cases are expected to increase in the future.

For many years, the cornerstone of treatment for ATTR-CA symptoms has been careful fluid balance management using loop diuretics [19]. This approach is unlikely to change significantly in the near future. Furthermore, since disease-modifying drugs became available for prescription in 2019, the primary treatment goal for patients with ATTR-CA is to initiate these drugs as early as possible [6]. Nonetheless, their high cost - 155,464 Japanese yen per day as of 2021 - restricts use, particularly in patients with dementia and frailty, and they have not yet been approved for ATTR-CA treatment in many countries [19]. Additionally, these drugs only slow disease progression and do not improve LVEF or LV remodeling, or suppress the onset of AF, although all-cause mortality, cardiovascular-related hospitalization, and quality of life improve [6,7,11]; therefore, existing HF treatments are equally important. In our patient, neurohormonal drugs for HFrEF and loop diuretics for volume management were prescribed. However, tafamidis for ATTRwt-CA was not prescribed because of dementia and frailty. Despite the lack of a standardized definition for reverse remodeling [8], a relative percentage change in LVEDVI of 23% and an absolute increase in LVEF of 14%, with improvement in mitral valve tethering, were observed in our case and were considered to reflect reverse remodeling induced by conventional HF treatment. In HFrEF without ATTR-CA, ACEi/ARBs are reported to improve LVEF by 1%-4%, beta-blockers by 4%-12%, and MRAs by 4% [8]. The 14% increase observed in this case may reflect the combined effect of these three medications, although beta-blockers and ACEi were administered at doses lower than the standard. The isolated effects of ACEi, ARBs, MRAs, ARNIs, and SGLT2i on reverse remodeling remain controversial [20]. In contrast, LV wall thickness increased three years after treatment. The reduction in LV cavity size observed in this case may be attributable to both ATTRwt deposition and the effects of HF therapy [1,4,12].

Currently, SGLT2i and MRAs are recommended for HF in patients with ATTRwt-CA, irrespective of EF [13]. Additionally, low-dose beta-blockers may be considered for patients with HFrEF when the dose is tailored on a case-by-case basis according to the patient's individual response to treatment; however, routine use is not recommended for patients with HF with preserved or mildly reduced EF [13,19]. In the setting of a fixed stroke volume due to restrictive physiology in ATTR-CA [1], a relatively high heart rate is required to maintain cardiac output. Therefore, medications with negative chronotropic effects often need to be discontinued. ACEi, ARBs, and ARNIs are not recommended for patients with ATTR-CA because no studies have clarified the benefits of these medications [13,19]. If the increase in cardiac output is insufficient to counteract the vasodilatory effects of drugs such as ACEi and ARBs, hypotension occurs, compromising tolerance to these medications [19]. MRAs can be used without discontinuation because of their limited effects on blood pressure and possible diuretic effects. MRAs may exhibit synergistic effects when used in combination with loop diuretics and may increase potassium reabsorption, which is often necessary when high-dose loop diuretics are administered [19]. However, there is still insufficient evidence to support the use of SGLT2i or MRAs in patients with ATTR-CA, and periodic reassessment of the need for continued administration, as well as monitoring for adverse effects, is necessary. This is particularly important when these medications are used in older patients with impaired renal function.

In our patient, paroxysmal AF was identified just after HFrEF treatment was initiated; nevertheless, antiarrhythmic drug therapy was not administered. Persistent AF subsequently developed three years later. The management of AF in patients with ATTR-CA presents particular challenges because a long-term sinus rhythm is difficult to maintain. This difficulty stems from ATTR deposition itself and restrictive physiology because of ATTR infiltration, leading to progressive atrial remodeling [13,21]. Evidence regarding rate and rhythm control in patients with ATTR-CA is limited and controversial, and whether rate- or rhythm-control therapy should be used has not yet been determined. Although AF is associated with a higher prevalence and incidence of HF [13], no differences in survival have been demonstrated between these therapies. Few studies have investigated the efficacy and safety of antiarrhythmic drugs used as part of rhythm-control strategies. Amiodarone is the drug of choice for most clinicians [13], although amiodarone-induced pulmonary toxicity occurs in approximately 2%-5% of patients and is more common in older individuals and in those with underlying lung disease [22]. In addition, the risk increases approximately threefold for every 10-year increase in age in patients over 60 years. Compared with nonusers, patients who use SGLT2i may experience a significant reduction in AF development [23]. Data on catheter ablation for AF are scarce and controversial; therefore, catheter ablation is usually not advised [13], although in carefully selected cases, recurrence rates may be low [24]. Rate-control medications are often poorly tolerated and associated with risks including hypotension, low cardiac output, HF decompensation, and conduction disorders [13]. Non-dihydropyridine calcium channel blockers are generally avoided, whereas low doses of beta-blockers and digoxin can be used for rate control, with close monitoring of hemodynamics and signs of toxicity. Drug-level monitoring is also advised for patients treated with digoxin. Even so, toxicity can occur at therapeutic serum levels [13,25,26]. After digoxin was prescribed at another hospital as part of a rate-control strategy for tachycardic AF, the patient died suddenly because of intractable bradyarrhythmia. Although we could not prove a causal relationship between digoxin administration and mortality, we inferred a significant influence. Administering two types of atrioventricular nodal blockers, beta-blockers and digoxin, is likely inappropriate [27]. Measurement of serum digoxin concentrations and continuous monitoring would have been preferable after the patient was transferred to our hospital. The Japanese Circulation Society guidelines recommend maintaining serum digoxin concentrations below 0.9 ng/mL when administering digoxin to patients with chronic HF caused by systolic dysfunction [25]. Nonetheless, this recommendation does not apply to patients with ATTR-CA. After initiating digoxin therapy or adjusting the dose, blood samples should be collected after the serum digoxin concentration reaches a steady state to allow therapeutic drug monitoring. This typically occurs at approximately day seven in patients with normal renal function and slightly later in those with impaired renal function.

Anticoagulation is advised for patients with ATTR-CA and AF, regardless of their CHA2DS2-VASc scores [28]. Embolic events are common in ATTR-CA, particularly among patients with AF who are not receiving anticoagulation therapy, and they can occur even in patients with sinus rhythm. Additionally, the risk of thromboembolism despite appropriate anticoagulation therapy has been emphasized in patients with ATTR-CA complicated by AF [13]. Atrial electromechanical dissociation caused by atrial ATTR deposition results in loss of normal atrial contraction, even in sinus rhythm [21,29]. Together with the low cardiac output in ATTR-CA, this promotes blood stasis in the left atrium; therefore, thrombus formation is more likely, even during anticoagulation therapy. While CMR imaging is suitable for assessing direct ATTR infiltration into the left atrium, strain analysis via echocardiography is suitable for evaluating left atrial function, and transesophageal echocardiography is highly sensitive for detecting the presence of left atrial thrombi and blood stasis, there are no guidelines for initiating anticoagulation therapy in patients with sinus rhythm [29]. Anticoagulation therapy cannot be considered in isolation from the risk of hemorrhagic complications, and given that ATTRwt-CA is common in older patients with impaired renal function, there is hesitation regarding the initiation of unnecessary treatment. In this case, anticoagulant therapy was initiated after AF was confirmed; however, the patient developed a cerebral embolism despite receiving anticoagulant therapy. There are also no treatment guidelines for the management of cardioembolic events that occur during anticoagulation therapy. Left atrial appendage occlusion may be an option [13].

This report has several limitations. First, because an endomyocardial biopsy was not performed, the myocardial ATTR burden could not be determined, and it was not possible to confirm the presence of coexisting cardiac disease, which was highly likely in this patient with ATTRwt-CA. Second, we suspected that the cause of death may have been profound bradyarrhythmia resulting from the combined use of two types of drugs - digoxin and beta-blockers - that block the atrioventricular node; however, the exact cause remains unclear. Third, the NAC staging was determined using a validated conversion formula derived from plasma BNP levels, rather than direct measurement of plasma NT-proBNP levels.

Conclusions

Our case of HFrEF demonstrated that, even in the presence of concomitant ATTRwt-CA, if ATTRwt-CA is in its early stage, reverse remodeling - defined as improvement in decreased LVEF and a dilated LV cavity - can occur with standard treatment for HFrEF alone, without the need for disease-modifying therapy for ATTRwt-CA. However, since ATTR deposition was not suppressed by disease-modifying therapy in this case, LV wall thickness and LV mass increased, indicating that the so-called "honeymoon period" -during which an improvement in decreased LVEF and LV dilatation was observed - was likely temporary. The key is to initiate treatment as early as possible with disease-modifying drugs, which are the only standard therapies proven to alter the natural course of ATTR-CA, together with conventional HF medications. In cases of HFrEF and ATTRwt-CA at a more advanced stage, it is also necessary to accurately determine whether the current hemodynamics are primarily due to the former or the latter and to adjust HF medication accordingly; furthermore, since these conditions may change over time, treatment must be carefully tailored.