Discordant low-gradient aortic stenosis: a phenotype-first, likelihood-tiered multimodality imaging framework

Discordant low-gradient aortic stenosis (AS), commonly defined by the aortic valve area (AVA) ≤1.0 cm 2 with mean gradient <40 mmHg and peak velocity <4.0 m/s, is reported in up to 40% of patients evaluated for AS in selected cohorts. Current guidelines and consensus documents outline echocardiographic quality control, haemodynamic phenotype assignment, dobutamine stress echocardiography (DSE), CT aortic valve calcium scoring (CT-AVC), and integrated assessment, yet uncertainty persists when haemodynamic, structural, symptomatic, and myocardial findings diverge. This narrative review proposes a phenotype-first, likelihood-tiered multimodality framework that adds four operational elements to current guidance: routine documentation of transvalvular flow rate, phenotype-specific interpretation of DSE and CT-AVC, separation of valve severity adjudication from myocardial risk refinement, and structured reporting of residual uncertainty. DSE with projected AVA is favoured for flow-dependent severity assessment in classical low-flow low-gradient AS, whereas CT-AVC likelihood categories are emphasised in preserved-ejection-fraction low-gradient phenotypes and after inconclusive DSE. Cardiovascular magnetic resonance, global longitudinal strain, extracellular volume fraction, and cardiac damage staging are used as prognostic and counselling adjuncts, not stand-alone triggers for valve intervention. Indeterminate cases are reported as unresolved discordance with the basis of uncertainty stated explicitly. The approach is intended to improve Heart Team communication and requires prospective validation before use as a treatment algorithm.