Systemic amyloidosis is a vast group of diseases defined by the presence of insoluble misfolded autologous protein deposits in various tissues and organs (1). In cardiac forms (cardiac amyloidosis, CA), amyloid fibrils accumulate within the myocardial interstitium leading to restrictive and hypertrophic cardiomyopathies often culminating in symptomatic heart failure and death. Amyloid precursor proteins involved in CA are mainly immunoglobulin light chain (AL) and transthyretin (TTR), with two distinct transthyretin amyloidosis (ATTR) types, namely hereditary or mutated (also known as ATTRv, v for variant) and wild type (ATTRwt) (2–4). Rapidly progressive forms of CA related to TTR variants V122I (p.Val142Ile) and I107V (p.Ile127Val) have been identified in populations of African descent (Afro-Americans, Afro-Caribbeans), with both variants typically associated with either predominantly cardiac or predominantly neuropathic or mixed cardiac-neuropathic phenotypes (5–13).

Central nervous system (CNS) involvement in ATTRv is more and more commonly evoked in the clinical setting. In TTR mutations, such as V30M (p.Val50Met), cerebral amyloid angiopathy (CAA), a disease of the small arteries and arterioles predominantly affecting the cortex and leptomeninges, is primarily characterized by intracerebral lobar and sub-arachnoid hemorrhage, cerebral micro-bleeds and cortical superficial siderosis (14, 15). Moreover, growing evidence now suggests that cerebral ischemic events (CIE) might be a common complication of CA in Caucasians presenting with AL amyloidosis (16, 17). In contrast, only sporadic case series have reported CIE in patients presenting with cardiac transthyretin amyloidosis (ATTR cardiomyopathy, CATTR). Even rarer are reports in patients of African descent (Afro-Americans, Afro-Caribbeans) with CATTR-V122I and CATTR-I107V (17–21).

While hereditary transthyretin amyloidosis remains a rare event (one case/100,000 Caucasians), even in expert medical centers for amyloidosis management worldwide (Sweden, Ireland, Spain, mainland France, Finland, Germany, Greece), it is assumed to be more prevalent in populations of predominantly African descent (22, 23). Our present work aims to analyze the occurrence and characteristics of cerebral ischemic events (CIE) in a real-world cohort of Afro-Caribbean patients followed for CATTR at our institution.

One-hundred and twenty Afro-Caribbean CATTR patients [N = 120; 69.2% males, mean age at CA diagnosis: 75.5 ± 7.4 years, median BMI (min-max range): 24.2 (15.9–37.0 kg/m²)] were included (Table 1, Supplementary Figure 1). Overall CA population was composed of 17 wild-type ATTR (14.2%), 73 ATTR-V122I (60.8%) and 22 ATTR-I107V (18.3%). CA genotype was undetermined for 8 patients. Patient phenotypes were predominantly cardiac for 52.5% of patients, while 57 patients (47.5%) presented mixed cardiac-neuropathic involvement (Table 1).

Thirty-six out of 120 CA patients (30.0%) presented with CIE, including three transient ischemic attacks and 33 permanent ischemic strokes. According to the ASCOD classification, the source of cerebral ischemia was cardioembolism in 25 patients (69.4%). CIE topography was mainly of the anterior circulation type (48.5% of permanent ischemic strokes) (Supplementary Table 1).

CIE was concomitant with CATTR diagnosis in 16 patients (16/36: 44.4%), while 14 patients (14/36: 38.9 %) experienced CIE over a median CATTR follow-up of 2.0 years (min-max range: 0.8–4.4 years). Six patients (6/36: 16.7%) presented with CIE before CATTR diagnosis. Eleven patients (11/36: 30.6%) experienced recurrent ischemic strokes. Overall mortality was 50.8%. Patients with concomitant diagnosis of CIE and CATTR had the worst outcome, with mean survival of 2.90 ± .5 years after established CA diagnosis, as compared to patients with CIE after CA diagnosis (mean survival: 5.60 ± .6 years) and patients with no CIE (mean survival: 4.40 ± .3 years) (Figure 1A). Table 2 further compares clinical, CA and CIE characteristics according to CA genotype, while patient characteristics with and without CIE are summarized in Table 3.

Of the 36 patients with CIE, five had CATTRwt (5/36: 13.9%), 28 had CATTRv (28/36: 77.8%) and TTR genotype was undetermined in 3 patients (3/36: 8.3%) (Table 3). There were no significant differences in terms of age, gender, body mass index and major cardiovascular characteristics (NYHA, cardiovascular risk factors, history of heart failure, conduction disorders, electrocardiographic and echographic parameters) among patients with or without CIE. Atrial fibrillation (AF) was documented in 24 of the 36 patients with CIE (66.7%) and in 41 patients without CIE (48.8%) (p = 0.19). Overall, 71 patients were on an anticoagulation regimen (59.2%). AF was the most frequent indication of anticoagulant treatment in patients with and without CIE (19/27: 70.4% and 33/44: 75.0%, respectively), with higher anticoagulation administration in patients with CIE (p = 0.02). It is to be noted that 27 patients (27/71: 38.0%) presented a CIE despite anticoagulation therapy.

CHA₂DS₂-VASc score was also higher in patients with CIE (p = 0.0005) (Table 3). Kaplan-Meier estimates of CIE risk according to CHA₂DS₂-VASc <3 or ≥ 3, LA enlargement and AF are illustrated in Figures 2A,B. Patients with a high CHA₂DS₂-VASc (≥ 3) and left atrial enlargement seemed to present with CIE more rapidly after CA diagnosis (mean survival free from event: 6.50 ± .7 years) as compared with a high CHA₂DS₂-VASc and no atrial enlargement (mean survival free from event: 9.2 ± 1.0 years) (p: 0.0012). The same tendency was observed, in the immediate time frame following CA diagnosis, for patients with a high CHA₂DS₂-VASc with or without AF (respective mean survival times free from event of 7.00 ± .7 and 7.30 ± .8 years) (p: 0.007)

Furthermore, time to CIE onset was also dependent on CA genotype. Patients with CATTRwt presented with CIE more rapidly (mean survival free from event: 0.60 ± .1 years) as compared to patients expressing the I107V variant (mean survival free from event: 5.60 ± .4 years) or the V122I variant (mean survival free from event: 7.70 ± .6 years) (p = 0.04) (Figure 1B). Risk for early CIE onset seemed to follow the same tendency when coupled to a high CHA₂DS₂-VASc: mean survival free from event of 0.60 ± .1, 5.30 ± .8, 7.00 ± .7 years respectively for ATTRwt, ATTR-V122I and ATTR-I107V (p = 0.01) (Figure 2C).

Univariate Cox regression analysis highlighted a high CHA₂DS₂-VASc score (≥3) and left atrial enlargement as potential predictors of CIE in CATTR patients, (Supplementary Tables 2 and 3).

Multivariate analysis retained only a high CHA₂DS₂-VASc score (≥3) as an independent predictor of CIE risk (all CIE and CIE with cardioembolic pattern), with Hazard Ratios (95% CI) of respectively 12.03 (1.62–89.24) and 8.24 (1.09-62.56) (Table 4).

Studies addressing the epidemiology of cerebral ischemic events in cardiac amyloidosis (CA) are scarce and mostly limited to case series and observational studies, mainly in Caucasian individuals presenting with AL amyloidosis (16–19, 30–33). Because CATTRv is a rare disease, only few patients are included, even in large multicenter studies evaluating ischemic stroke prevalence in CA patients (8, 9, 13). The present study systematically analyzes information from a large cohort of Afro-Caribbean CATTR patients presenting TTR variants commonly encountered in individuals of identifiable African descent: 60.8% V122I and 18.3% I107V (vs 14.2% ATTRwt) in our sample of 120 CATTR patients. In our series, one third of patients with CATTR experienced cerebral ischemic event (CIE) either concomitantly to CA or during its course. CIE was observed in Afro-Caribbean patients with CATTRv who displayed the typical cardiac and neurological phenotypes associated with V122I and I107V mutations, as generally described in Africans, Black Americans and Afro-Caribbeans (8, 9, 13).

Our study thus highlights that CIE may be frequent and that it may have concomitantly lead to the diagnosis of CATTR in more than 40% of CIE cases (potentially associated with worse patient outcome) in our study sample. Moreover, for 13% of CATTR patients, CIE preceded the diagnosis of systemic amyloidosis by a median interval of 4.5 years. These observations underscore the need to potentially consider early anticoagulation in CATTR patients.

Consistently with previous reports (7, 9, 13), CA patients with V122I TTR variant had more frequently heart failure history and worse myocardial function as compared to CATTRwt and other CATTRv patients. Whereas history of atrial fibrillation and atrial enlargement appeared more frequent in CATTR-V122I and CATTRwt patients, prevalence of CIE was similar independently of CATTR genotype. Survival free from CIE was however shorter in CATTRwt patients, when compared with those with CATTR-I107V and CATTR-V122I.

In our study, CATTR patients who experienced CIE had atrial enlargement and frequently presented at least one episode of permanent AF in their clinical course. Yet, comparison with CATTR patients who did not experience CIE failed to reach statistical significance (p = 0.07). Moreover, even if AF was more frequent in patients with CIE, one-third of the latter occurred in CATTR patients with sinus rhythm. These findings are consistent with previous reports attributing these observations to intracardiac thrombi formation along with CA-induced atrial myopathy (14, 18, 19). Proposed mechanisms contributing to intra-cardiac thrombus in CA include blood stasis associated intracavitary turbulence, focal amyloid deposition impairing cardiac wall motion, and amyloid infiltration of coronary arteries leading to myocardial ischemia (14, 18, 19, 34–38). In this context, several studies in CA patients have shown that atrial myopathy can explain thrombi formation in patients with sinus rhythm and/or despite adequate anticoagulation (19, 34, 35). Systemic involvement of the disease, advanced age, multiple cardiovascular risk factors represent other predisposing factors for thromboembolic events.

Overall, our study also readdresses the question of the optimal strategy of anticoagulation therapy in CATTR patients. If anticoagulation must be started as soon as possible in CATTR patients with AF (regardless of CHA₂DS₂-VASc score) (35, 36), anticoagulation in CATTR patients in sinus rhythm remains an open question. Chronic anticoagulation is associated with a significantly lower risk for intracardiac thrombosis, with a similar trend observed in CA patients with no AF (35). Effective anticoagulation might thus reduce cerebral ischemic events, which are significant contributors to mortality in CA patients (17). In contrast, anticoagulation may also exacerbate hemorrhagic tendency, which is a well-known complication of amyloidosis attributed to amyloid deposits in vessel walls, potential coexisting coagulopathy or underlying cerebral amyloid angiopathy (CAA) (1).

Furthermore, our results underline that a CHA₂DS₂-VASc score ≥ 3 may represent a strong and independent predictive factor of CIE in CATTR patients. In our study sample, this might be linked to age distribution (age range: 55–97 years) as well as a high prevalence of cardiovascular risk factors (hypertension, diabetes) and heart failure. A large series of CA patients has previously identified a likewise association between CIE and a high CHA₂DS₂-VASc score, but failed to report its prognostic value in multivariate models due to the small number of CATTR patients in the study sample (18). Overall, a CHA₂DS₂-VASc score ≥ 3 may identify CATTR patients with a high risk of CIE, even beyond the presence of AF. Although the benefit of anticoagulation in CATTR patients with no AF remains controversial, a CHA₂DS₂-VASc score ≥ 3 could trigger the search for cardiac intracavity thrombi and motivate anticoagulation when applicable.

Transthyretin amyloid heart disease could potentially be a full-fledged cause of cerebral ischemic events (mainly cardioembolic). The present study describes one of the largest series of CIE in Afro-Caribbean cardiac amyloidosis patients presenting with V122I and I107V transthyretin variants. We found that CIE occurred in one-third of CATTR patients, and that it seemed to be of worse prognosis when it manifested concomitantly to CATTR diagnosis. A significant proportion of events occurred in patients who had no AF or despite anticoagulation therapy. As reported by Ballantyne et al., thrombotic risk should be better characterized in CATTR patients with integration of echocardiographic parameters and transesophageal echocardiography (14). In addition, our study confirms that a CHA₂DS₂-VASc score ≥3 may also be a strong independent predictive factor of CIE in CATTR patients. Further studies are needed to assess the efficacy and timeliness of anticoagulation in CATTR patients, independently of AF.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The study was approved by the institutional review board of the University Hospital of Martinique. All patients were managed in accordance with the amended Declaration of Helsinki (https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/). Written informed consent from the patients/participants or patients/participants' legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and the institutional requirements.

RB, T-HT, AM, AS, and JI collected patient data. RB, AS, JI, and RN analyzed and interpreted patient data. RB and RN wrote the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.