Pre-diagnostic red flags and referral pathways in transthyretin cardiac amyloidosis: correlation with disease stage and outcomes

Background

Diagnostic delay remains a major barrier to timely treatment in transthyretin cardiac amyloidosis (ATTR-CM)1,2. Although current staging systems describe disease severity at diagnosis3,4, they do not capture the temporal dimension of preceding disease manifestations. The prognostic relevance of early, domain-specific red flags and their cumulative duration before diagnosis remains unclear.

Objectives

To reconstruct pre-diagnostic trajectories of ATTR-CM, quantify the median cumulative temporal burden of red-flag manifestations and evaluate its association with cardiovascular and all-cause mortality.

Methods

We retrospectively included consecutive ATTR-CM patients from two referral centers in Italy. All pre-diagnostic red flags across cardiologic, neurologic, orthopedic, renal, and incidental domains were identified and time-mapped. A cumulative temporal red-flag burden was computed as the sum of all pre-diagnostic intervals; due to right-skewed distribution, the median cumulative burden was selected as the primary exposure. Burden quartiles were examined, merging Q3–Q4 due to similar behavior. Outcomes included cardiovascular (CV) death and all-cause death, analyzed using Fine–Gray competing-risk models and Cox regression adjusted for age, sex, NAC stage, and tafamidis use. Diagnostic trajectories were visualized using a Sankey diagram.

Results

A total of 208 patients were included (median age 80 years; 84% male; 97% ATTRwt). Red flags frequently emerged several years before diagnosis, with orthopedic and neurologic manifestations representing the earliest clues, whereas cardiologic findings clustered closer to diagnosis. Diagnostic pathways were heterogeneous but converged predominantly through cardiology referral.

Higher median cumulative red-flag burden was independently associated with increased CV mortality (p = 0.015; adjusted HR 1.10, 95% CI 1.02–1.19), while the association with all-cause mortality was weaker and borderline significant (p = 0.047). Discrimination for CV mortality was high (time-dependent AUC 0.81), with no interaction by age (<80 vs ≥80 years).

Cumulative incidence analyses showed a significantly higher incidence of CV death in the highest burden category (Gray’s test p = 0.036; Figure 1), whereas differences in all-cause mortality were not significant. Sankey analysis revealed a progressive shift toward cardiologic referral pathways despite frequent early involvement of non-cardiac domains (Figure 2).

Conclusions

The median cumulative red-flag burden provides a novel, time-based descriptor of the pre-diagnostic phase of ATTR-CM and represents an independent predictor of cardiovascular mortality beyond established clinical variables. Earlier recognition of non-cardiac red flags may shorten diagnostic delay, enable timelier initiation of disease-modifying therapies and improve clinical outcomes.CIF of CV death across median quartilesFor image description, please refer to the figure legend and surrounding text.Sankey from first domain to referralFor image description, please refer to the figure legend and surrounding text.