Patients with paradoxical low-flow (LF) severe aortic stenosis (AS) despite preserved left ventricular ejection fraction (LVEF) appear distinct from normal-flow (NF) patients, showing worse prognosis, more concentric hypertrophy and smaller left ventricular (LV) cavities. The left ventricular remodelling index (LVRI) has been demonstrated to reliably discriminate between physiologically adapted athlete's heart and pathological LV remodelling. We studied patients with index echocardiographic diagnosis of severe AS (aortic valve area <1 cm2) with preserved LVEF (>50%). The LVRI was determined by the ratio of the LV mass to the end-diastolic volume, as previously reported, and was compared between patients with LF and NF AS. Patients were prospectively followed up for at least 3 years, and clinical outcomes were examined in association with LVRI. Of the 450 patients studied, 112 (24.9%) had LF AS. While there were no significant differences in baseline clinical profile between LF and NF patients, LVRI was significantly higher in the LF group. Patients with high LVRI (>1.56 g/mL) had increased all-cause mortality (log-rank 9.18, P = 0.002) and were more likely to be admitted for cardiac failure (log-rank 7.61, P = 0.006) or undergo aortic valve replacement (log-rank 18.4, P < 0.001). After adjusting for the effect of age, hypertension, aortic valve area and mean pressure gradient on multivariate Cox regression, high LVRI remained independently associated with poor clinical outcomes (hazard ratio 1.64, 95% confidence interval 1.19-2.25, P = 0.002). Pathological LV remodelling (increased LVRI) was more common in patients with LF AS, and increased LVRI independently predicts worse clinical outcomes.