This is the largest registry in Brazil to describe the demographic, genetic, and clinical imaging characteristics and treatment of patients diagnosed with ATTR. We observed that there was a greater proportion of the variant form, contrasting with other studies [2, 6]. The most frequently associated mutation in this study was Val50Met, followed by Val142Ile, in line with previous studies [8, 9]. There were also other mutations that have been reported in other areas of the world, such as Thr80Ala, Ala39Asp, Phe64Ser, Glu109Lys [10–15]. The variety of mutations found might be explained by the fact that São Paulo is the most populated state in Brazil, the largest country in South America, housing population groups from other regions of the globe.
The median age was 65 years, which is consistent with other studies that showed a higher incidence in the population > 63 years of age [6, 16–18]. We observed that patients with ATTRv were younger (median of 54 years of age) than patients with ATTRwt, in line with some studies [19], though others have reported a median age of 73 years in ATTVv patients [20]. This is possibly because these patients were from different geographic regions where the epigenetic factors to which the population is exposed differ.
Of the whole sample, 62.8% were male, which coincides with the findings of a systematic review that compared 4,669 patients with ATTR by sex, revealing a male predominance of 83% [21]. This could be due to low clinical suspicion, which reduces the likelihood of diagnosis in women, in addition to the likely cardioprotection caused by estrogens. Although this is a hypothesis, the study by Prasad et al. demonstrated that the identification of this disease in females has increased, with the percentage of women diagnosed with ATTR-CA increasing from 5.9% before 2019 to 16.7% from 2019 to 2022 [22].
In this study, the majority of patients were white (74.5%), as in the study by Porcari et al., in which there was a higher prevalence of amyloidosis in white patients (80.4%). This predilection is notable, but it could be due to the population base, as these patients were from the United Kingdom [20]. The most associated comorbidities were arterial hypertension, CKD and dyslipidemia, which were also associated with ATTR according to previous studies [20, 23–25].
We determined that among medical specialties, cardiologists most frequently diagnosed ATTR, a finding also obtained by other researchers [25]. In this study, this could have been because the centers mostly specialized in cardiology.
The time from the onset of symptoms to the diagnosis of amyloidosis is essential, as it has been directly related to survival in patients [25]. In this study, the duration of pregnancy was 1.853 days, which is in agreement with previous studies [9, 25]. The initial symptoms were neurological and were mainly referred to as paresthesia in the extremities and cardiological symptoms, such as dyspnea. This clinical heterogeneity was due to the systemic involvement of the disease.
In this study, the neurological phenotype was present in 51.1% of patients. There was also evidence of this phenotype predominating in geographic regions such as South America and Asia and in European countries such as Spain and Portugal, according to the Transthyretin Amyloidosis Outcomes Survey (THAOS) [26]. In a study from Spain, the neurological phenotype was mainly associated with the variant form, similar to the results of this study [27].
The cardiological phenotype was found in 53.9% of the patients in the registry. In patients with ATTRwt, there was a predominance of the cardiological phenotype (89.9%) compared to ATTRv (43.9%). Notably, the incidence of the cardiological phenotype in ATTRv was much greater than that reported in the Italian registry (13.5%) [1]. It is possible that this difference is due to the greater presence of the Val142Ile mutation in our registry than the Italian registry, in which cardiological impairment is predominant [6, 7]. On the other hand, it is also plausible to assume a selection bias of patients with a cardiological phenotype in our registry since most of the centers involved were centers specializing in cardiology.
Cardiac amyloidosis is a progressive disease that is one of the etiologies of HF and requires costs and resources for the follow-up and treatment of patients. According to the NYHA classification, most patients were classified as CF II, which is consistent with previous studies [27, 29].
In the past, there was no standardized therapeutic approach for heart failure and no specific treatment for the disease. With the positive results of multicenter studies on medications that stabilize transthyretin protein, improvements in quality of life and survival were demonstrated in this population. In our study, diuretics were the most commonly used medications, and TTR stabilizing therapy (tafamidis) was the most frequently reported among the specific treatments for this disease, although it was used by a very low proportion.
One of the red flags is the low voltage of the QRS complex on the ECG and the increased thickness of the ventricular wall on the echocardiogram, which raises the suspicion of amyloid cardiomyopathy [30]. Another suggestive sign on the ECG is apical longitudinal strain sparing, which was present in 31.3% of the patients in this study. This finding suggests the use of an easily recognizable and reproducible method to distinguish ATTR-CM from other causes of LV hypertrophy, as described in the study by Phelan et al. [31]. Given the relatively high sensitivity and specificity of these imaging findings compared to those of other echocardiographic parameters, the British Society of Echocardiography Guideline recommends that all patients with increased LV wall thickness undergo strain imaging, not just those patients with suspected cardiac amyloidosis. This approach encourages suspicion of and facilitates the diagnosis of these entities [32].
On the other hand, after excluding AL amyloidosis, positive cardiac scintigraphy is also suggestive of ATTR-CM, the 1-hour image being more sensitive and the 3-hour image being more specific for the diagnosis [33]. The heart/contralateral uptake ratio (H/CL) is calculated as the ratio of the mean count from the heart region of interest to the mean count from the contralateral chest. An H/CL ratio > 1.5 on a 1-hour image and an H/CL > 1.3 on a 3-hour image are highly suggestive of ATTR-CM [34].
In this study, CMR showed increased T1 map and ECV values, in addition to subendocardial late gadolinium enhancement, which is characteristic of patients with ATTR-CM and is also in agreement with a previous study [27].
The present study had some limitations. Some patients were excluded because they did not have genetic data to classify them as ATTRv or ATTRwt. Another limitation was that the examinations were not done in all patients, and we analyzed the data when they were available at the participating centers.