Aortic regurgitation (AR) is a common valvular lesion associated with increased mortality once the left ventricle enlarges significantly or develops systolic dysfunction (ejection fraction < 50%). Valve guidelines recommend aortic valve repair or replacement (AVR) for left ventricular (LV) linear end-systolic dimension ≥ 50 mm or end-diastolic dimension ≥ 65 mm. However, chamber quantification guidelines recommend using LV volume for LV size determination because linear measurements may not accurately reflect LV remodeling. The aim of this study was to evaluate the correlation of LV volumes with linear dimensions, interobserver variability in the estimation of volumes, and the association of volumes with outcomes in patients with AR. A total of 1,100 consecutive patients with chronic moderate to severe and severe AR on echocardiography between 2004 and 2019 were retrospectively analyzed. The modified Simpson disk summation method was used for LV volume estimation. The primary outcome was all-cause mortality; the secondary outcome was mortality censored at AVR. Patients' age was 60 ± 17 years, and 198 were women (18%). Volumes were measured using the biplane method in 939 patients (85%) and the monoplane method in 161 (15%); end-systolic volume was normal in 169 (11%). Correlations between volumes and linear dimensions were 0.5 for end-diastolic volume and 0.6 for end-systolic volume. At median follow-up of 5.4 years (interquartile range, 2.4-10.0 years), 216 patients had died and 539 had undergone AVR. Indexed LV end-systolic volume (iLVESV) and indexed left ventricular end-systolic dimension were both associated with mortality and symptoms, but the association of iLVESV was stronger. iLVESV, age, male gender, Charlson comorbidity index, New York Heart Association functional class III or IV, and time-dependent AVR were independently associated with all-cause mortality. Interobserver variability in the estimation of LV volumes in 200 patients included intraclass coefficients of 0.94 (95% CI, 0.92-0.95) for end-diastolic volume and 0.88 (95% CI, 0.78-0.93) for end-systolic volume. Patients with iLVESV ≥ 45 mL/m2 had lower survival and a higher prevalence of symptoms than those with volumes < 45 mL/m2. Echocardiographic LV volume assessment had good reproducibility in patients with moderate to severe and severe AR. The correlation between linear dimensions and volumes was limited. Both iLVESV and indexed left ventricular end-systolic dimension were associated with worse outcomes, but the association of iLVESV was stronger. iLVESV ≥ 45 mL/m2 was associated with worse outcomes. Copyright © 2020 American Society of Echocardiography. Published by Elsevier Inc. All rights reserved.