Early diagnosis, disease stage and prognosis in wild-type transthyretin amyloid cardiomyopathy: The DIAMOND study

Abstract

Aims: Disease staging and prognostic scoring in wild-type transthyretin-related cardiac amyloidosis (ATTRwt-CA) can be captured by two systems (NAC and Columbia scores). However, uncertainty remains as epidemiology of the disease is evolving rapidly. We evaluated features associated with staging systems across ATTRwt-CA patients from different diagnostic pathways, and their association with prognosis.

Methods: We performed an analysis on DIAMOND patients with available data to evaluate NAC and Columbia score. DIAMOND was a retrospective study from 17 Italian referral centres for CA, enrolling 1281 patients diagnosed between 2016 and 2021, and aimed at describing characteristics of pathways leading to ATTRwt-CA diagnosis. Of the original cohort, 811 patients were included in this analysis. Each patient had NAC and Columbia score calculated. Patients were grouped according to NAC and Columbia scoring classes. We described characteristics of patients according to staging classes and diagnostic pathways at diagnosis. Prevalence of early diagnoses, defined as NAC Ia, NYHA class I, no use of diuretics, no history of heart failure (HF) hospitalizations nor of atrial fibrillation prior to diagnosis, was investigated. Finally, prognostic variables were tested alone and grouped as NAC or Columbia scores in Cox univariate and multivariate regression analyses. Prognosis was investigated as all-cause mortality, in the whole population and dividing patients in HF versus other diagnostic pathways.

Results: Only 1% of the study population had an early ATTRwt-CA diagnosis. Distribution of prognostic variables and of NAC and Columbia classes was heterogeneous across diagnostic pathways. The prevalence of NAC III and Columbia III was higher in the HF diagnostic pathway, but all NAC and Columbia classes were present in all pathways. Both NAC and Columbia scores were associated with all-cause mortality at univariate Cox regression analysis in the whole population, in patients from the HF diagnostic pathway and in those from other pathways. At multivariate analysis, Columbia score remained significantly associated with the outcome, together with age at diagnosis, left ventricular ejection fraction and maximal wall thickness.

Conclusions: In this contemporary nationwide cohort, an ATTRwt-CA early diagnosis was very rare. Disease staging with NAC and Columbia scoring systems determined classes of patients with heterogeneous features. Both scores were significantly associated with mortality, but other variables also had prognostic significance.

Keywords: Cardiac amyloidosis; Disease stage; NAC score; Prognostic scoring; Wild‐type transthyretin cardiac amyloidosis.

Figure 1

Proportion of early ATTRwt‐CA diagnoses in the DIAMOND cohort. An ‘early diagnosis’ was defined as a NAC score Ia and a Columbia score of 1, in the absence of a history of HF hospitalizations or of atrial fibrillation prior to diagnosis. Early diagnoses accounted for only 1% of the population. Notably, NAC Ia accounted for 5% of the population.

Figure 2

Distribution of prognostic variables in patients diagnosed in the heart failure pathway versus in patients diagnosed in other pathways.

Figure 3

Kaplan–Meier curves for survival free from all‐cause mortality according to Columbia classes in patients diagnosed in the heart failure pathway and in those diagnosed in other pathways. Solid lines represent heart failure pathway; dashed lines represent pathways other than heart failure. No statistical difference was observed between diagnostic pathways in each Columbia class: P _log for Columbia I: 0.88, II: 0.29, III: 0.88.