Invasively and noninvasively diagnosed patients were included. The invasive diagnosis was based on histological documentation of amyloid in endomyocardial biopsy (EMB) or extracardiac biopsy when echocardiography or magnetic resonance revealed the presence of left ventricular hypertrophy (LVH) (≥ 12mm), after the exclusion of other causes of LVH.12 To differentiate AL and ATTR, immunohistochemistry was performed using monoclonal antibodies against AA, TTR, and light chains (DAKO, Denmark). In doubtful cases or those with positive staining with multiple antibodies, mass spectrometry was performed for the definitive diagnosis of the amyloid subtype.

ATTR was noninvasively diagnosed according to published criteria13 and involved the presence of data compatible with CA from cardiac imaging techniques and a Perugini score of 2 or 3 on 99mTC-DPD scintigraphy,4 after ruling out AL using analysis of serum light chains (Freelite, The Binding Site Group, United Kingdom) and blood and urine immunofixation. A genetic study was performed in all ATTR patients involving complete sequencing of the TTR gene.

The hematological results indicated monoclonal gammopathy of undetermined significance and plasma cell dyscrasia, including multiple myeloma, as well as the need for bone marrow biopsy according to current criteria14 after hematological assessment

The study did not include patients with mutations in the TTR gene or those with AL without cardiac involvement.

The information was obtained through medical record review. Data were obtained on the first evaluation and follow-up.

Data on the final patient status were obtained from the hospital and primary care medical records. Total mortality was defined as death from any cause during follow-up.

Variable normality was studied with the Shapiro-Wilk test. Normal continuous variables are expressed as mean ± standard deviation. Categorical variables are expressed as percentages. To compare continuous variables, the 2-tailed Student t test was used for independent samples; the chi-square test or Fisher test was used for categorical data. The Kaplan-Meier curve was used to calculate survival. P < .05 was considered statistically significant.

An analysis of the time to death or heart transplantation was performed in patients with AL and with ATTR. In addition, a multivariate analysis of survival predictors was performed using a model adjusted for age, sex, CA type, and variables with P < .10 in the univariable model (table 1 of the supplementary data): heart failure (HF) at diagnosis, New York Association (NYHA) functional class III-IV at diagnosis, N-terminal pro-B-type natriuretic peptide (NT-proBNP) > 3000 ng/dL, glomerular filtration rate (continuous and < 45mL/min/1.73 m2), and left ventricular ejection fraction (LVEF) < 50%.

STATA version 12.0 (StataCorp LP, United States) was used for the statistical analysis.

Between May 2008 and September 2018, 223 biopsies were performed: 61 EMBs (34%), 71 of the bone marrow (41%), and 91 of other organs (51%). The EMBs were mainly performed in the right ventricle (56; 91.8%); only 5 (8.2%) were obtained from the left ventricle. The complication rate was low considering the invasiveness of the technique: only 2 complications (3.3%) were recorded, both nonfatal. The complications occurred after EMB of the right ventricle and comprised 1 intraprocedural ventricular arrhythmia and 1 mild effusion with tamponade due to anticoagulant therapy. The EMB yield was 98.4% and a diagnosis was made in 56% of AL patients (n = 36). In 1 patient, left ventricular EMB was performed because no amyloid was found in the right ventricular sample. Regarding extracardiac biopsies, 16 were obtained from the salivary gland, 20 from the abdominal fat, 8 from the gastrointestinal tract, 6 from the kidneys, and 1 from the peripheral nerves. The yields varied but tended to be low, with the exception of the renal and peripheral nerve biopsies, and depended on the organ targeted (19%, 40%, 60%, 80%, and 100%, respectively). In the bone marrow biopsies performed in patients suspected to have AL, a plasma cell component exceeding 10% was found in 56% of patients (36 patients) and amyloid in 81% (52 patients).

The definitive diagnosis of amyloidosis subtype was obtained using mass spectrometry in 8 patients (4.7%). The main reason for the use of this technique was double immunohistochemical staining for both TTR and one of the light chains. The proteomic results identified 2 of the patients as having AL and 6 as having ATTR, in line with the initial clinical suspicion in all patients except 1.

ATTR-CA was noninvasively diagnosed in 87 patients (75%). EMB was only required in 25 of them (22%). Four patients were diagnosed due to cardiac involvement and a positive extracardiac biopsy (3 of the abdominal fat and 1 of the gastrointestinal tract).

Although the percentage of patients with a noninvasive diagnosis in the entire series has increased in recent years, a considerable number are still diagnosed through biopsy (figure 2).

Figure 2.

Percentage of patients invasively diagnosed with cardiac amyloidosis (CA) and total number of patients with CA diagnosed per year.

(0.14MB).

The TTR gene was analyzed in all individuals diagnosed with ATTR-CA and 18 patients (16%) were found to have the following mutations: 6 had Val50Met, 6 had Val142Ile, 3 had Glu109Lys, 1 had Glu109Gln, 1 had Thr80Ala, and 1 had Ala65Thr. Of the patients diagnosed with hATTR, 14 (78%) were the first members of their families diagnosed. Among the 114 patients with ATTR older than 60 years, a mutation was found in 12% (14 patients). These results enabled the genetic study of 57 first-degree relatives belonging to 16 families, and 20 other individuals (35%) were found to have mutations in the TTR gene.

99mTc-DPD scintigraphy was performed in 165 patients (92%). The majority of these patients had ATTR (112; 68%); most of these patients (91%) had a Perugini score of 3. Of the 53 patients with AL who underwent 99mTc-DPD scintigraphy, 32% (n = 17) had some degree of uptake: 7 patients had grade 3 uptake, 6 had grade 2, and 4 had grade 1. Of the 13 patients with AL and grade 2 or 3 uptake, 7 had uptake in the left ventricle alone, 5 in both ventricles, and 1 in the right ventricle alone.

The time from symptom onset to definitive diagnosis was significantly longer in the ATTR group, with a mean of 2.8 ± 4.3 vs 0.6 ± 0.7 years for AL (P < .001); 70 patients (40%) had already been diagnosed with a different cardiac condition (table 1). The previous diagnosis was obtained in 45% and 28% of patients with ATTR and AL, respectively. The most frequent diagnosis in both subtypes together was hypertensive heart disease (17 patients; 9.4%), followed by HF with preserved LVEF, which was particularly common in patients with ATTR (12; 10%).

Most of the patients diagnosed with CA were derived from cardiology (99; 55%), followed by hematology (31 patients; 17%), from where 42% of the AL patients originated, as well as 4% of the ATTR. Of the patients with ATTR, 41% met the criteria for monoclonal gammopathy of undetermined significance.

Although there is particular interest in this disease in our center and we have long had a systematic disease diagnosis system in place for patients admitted with HF and increased ventricular thickness, more than one third of patients (70; 39%) were derived from other centers (table 1). Proportionally, referrals were more frequent for patients with AL (45%) than for those with ATTR (35%).

Regarding the clinical presentation, HF/dyspnea was the most common and the initial presentation in 81% of the AL patients vs only 44% of the ATTR patients (P < .001). In contrast to AL patients, the clinical presentation prompting the diagnosis was much more variable in ATTR patients and ranged from atrial fibrillation (21% in ATTR vs 8% in AL; P = .028) and conduction disorders (6%) to electrocardiographic abnormalities (3%) or were obtained during the etiological study of stroke (5%) or as an incidental finding after scintigraphy for rheumatological or oncological reasons (6%).

CA was diagnosed during admission in 53% of patients (69% for AL, 37% for hATTR, and 45% for ATTRwt).

Up to one third of patients (61; 34%) were in NYHA III-IV functional class at diagnosis. More than half of the patients (106; 59%) had already required admission due to HF, particularly the AL subgroup (70% vs 53%; P = .021). However, the ATTR subgroup contained a higher percentage of patients with a history of atrial fibrillation (50% vs 16%; P < .001) and a higher rate of pacemakers at the first assessment (21% vs 5%; P = .004). Three patients (2%) had an implantable cardioverter-defibrillator at the first evaluation.

Regarding biomarkers, the mean NT-proBNP level was 7873 ± 11 312 pg/mL. It was significantly higher in AL than in ATTR (13 338 ± 16 064 vs 4805 ± 5492 pg/mL; P < .001) and in patients diagnosed during admission vs those diagnosed without being admitted (10 987 ± 13 704 vs 4308 ± 6068 pg/mL; P < .001). There were no significant differences in troponin I levels between AL and ATTR (0.18 ± 0.56 vs 0.20 ± 0.21μg/L; P = .843). There were also no differences in glomerular filtration rates, although patients with AL more frequently had dipstick-positive proteinuria (table 2).

Among patients with AL, 11 (17%) were in Mayo stage III at diagnosis and 48 (75%) in stage IV. Among patients with ATTR, 39 (34%) were in Gillmore stage II and 25 (22%) in stage III.15

In the electrocardiograms, one fourth of the series had atrial fibrillation at the first assessment, with a significantly higher rate for ATTR patients (36% vs 9%; P < .001). In the AL forms, the classic low-voltage pattern was found in 34 patients (54%) and the pseudoinfarct pattern in 32 (50%), which comprised the most frequent electrocardiographic pattern in the entire cohort (table 3).

Regarding the echocardiographic evaluation, the mean septal thickness of the cohort was 16.8 ± 3.4mm and 20% of the patients had eccentric LVH. No significant differences in LVEF were seen between ATTR and AL, although there was a tendency for worse LVEF in AL (56.2% ± 12.9% vs 52.6% ± 12.7%; P = .077). In contrast, the AL forms had lower left atrial dilatation (44.7 ± 6.4 vs 42.4 ± 6.0mm, P = .025; 36.5 ± 13.1 vs 31.0 ± 9.2mL/m2, P = .001) but a significantly higher lateral E/E’ ratio (15.2 ± 6.5 vs 18.9 ± 10.4cm/s; P = .013).

Of the 67 patients who underwent cardiac magnetic resonance (37%), more than 80% showed the characteristic gadolinium kinetics comprising faster myocardial nulling vs that of the blood pool. In addition, late gadolinium enhancement was seen in almost the entire series (63; 96.9%), with a global subendocardial pattern in more than half of the patients (18; 67%), particularly in the AL forms (28; 80%).

During follow-up, 9 patients (6%) required pacemaker implantation (6 ATTR patients and 3 AL); 2 patients underwent a defibrillator implantation in primary prevention (one to a patient with AL and restrictive phenotype who received a resynchronization device in another center after documentation of complete block and another to a patient with ATTRwt with severe ventricular dysfunction and syncope).

There was clinical evidence of amyloid deposition in the connective tissue in more than half of the patients with ATTR, mainly in the form of carpal tunnel syndrome (48 patients; 41%) or lumbar spinal stenosis (11; 10%). AL was also associated with carpal tunnel syndrome, although at a lower frequency (14%; P < .001). Neurological and gastrointestinal (vomiting and diarrhea) symptoms were relatively frequent (25% and 33%, respectively). Neurological involvement was evident in 13 of the 18 patients with hATTR (72%); peripheral neuropathy symptoms were also found in some of the ATTRwt group (12%).

During the study period, 18 patients (10%) participated in some type of clinical trial with specific antiamyloid molecules.

In this period, 15 patients with ATTR (13%) received some type of targeted therapy. Most (12 patients) received a stabilizing therapy (tafamidis in 10 patients and diflunisal in 2). Three patients with hATTR and neurological involvement were treated with gene silencing (patisiran) designed to suppress TTR production after disease progression during stabilizing therapy (2 patients) or from onset (1 patient). Finally, 2 patients underwent treatment with epigallocatechin-3-gallate, a polyphenol from green tea.

Most patients with AL (72%) received antineoplastic agents, 44 with bortezomib combinations and 2 with alkylating agents. Six patients underwent autologous hematopoietic cell transplantation after a previous treatment, and 18 were unable to receive treatment and received just a single cycle due to their death.

During the period analyzed, no patient with hATTR underwent liver transplantation. In contrast, 3 patients with AL underwent heart transplantation (table 4).

During a mean follow-up of 1.6 ± 1.7 years, the total mortality of the cohort was 38% (68 patients); it was significantly higher in AL patients (48% vs 32% in ATTR; P = .028) (table 4). In 60 of those who died (88%), the cause of death was attributable to the amyloidosis.

The overall 1-year survival was 76%. The 12-, 24-, and 36-month survival rates were 65%, 52%, and 43% in AL patients and 83%, 73%, and 59% in ATTR patients, respectively (figure 3).

Figure 3.

Survival according to cardiac amyloidosis subtype. AL, amyloid light-chain amyloidosis; ATTR, transthyretin amyloidosis.

(0.09MB).

In the multivariable analysis, the only factors significantly associated with death or heart transplantation were AL subtype (hazard ratio [HR] = 6.16; 95% confidence interval [95%CI], 1.56-24.3; P = .010), female sex (HR = 2.35; 95%CI, 1.24-4.46; P = .009), and NYHA class III-IV at diagnosis (HR = 2.07; 95%CI, 1.11-3.89; P = .023) (table 5).

The present data were derived from a descriptive and retrospective study performed in a single center and subject to selection and survival biases. Notably, about 80% of the patients were diagnosed in the last 5 years, which could also have influenced the results.