By Michael H. Crawford, MD, Editor

SYNOPSIS: In a large cohort of asymptomatic patients with hemodynamically significant chronic aortic regurgitation, volumetric left ventricular size and function measurements were equally discriminant in identifying patients at higher risk for mortality vs. traditional linear measurements.

SOURCE: Yang LT, Anand V, Zambito EI, et al. Association of echocardiographic left ventricular end-systolic volume and volume-derived ejection fraction with outcome in asymptomatic chronic aortic regurgitation. JAMA Cardiol 2020; Nov 4;e205268. doi: 10.1001/jamacardio.2020.5268. [Online ahead of print].

Current guidelines recommend using linear echocardiographic measurements of the left ventricle (LV) to identify which asymptomatic patients with chronic aortic regurgitation (AR) may benefit from valve replacement. Most echo labs measure LV volumes, too, yet little information exists about the prognostic value of volumetric measures in patients with chronic AR.

Yang et al sought to assess whether LV volumes and volume derived ejection fraction (EF) were associated with mortality in asymptomatic patients with chronic moderately severe to severe AR vs. traditional linear measurements of LV size and function. The authors retrospectively identified adult patients with moderately severe to severe chronic AR between 2004 and early 2019. They excluded patients with typical symptoms of AR, other more-than-mild valve lesions, prior valve surgery, cyanotic congenital heart disease, carcinoid heart disease, and those who underwent surgery for AR within two months of the index echo. The final population included 492 patients (mean age, 60 years; 86% men) in whom linear and volumetric LV measurements and measurements of AR severity were made. The primary outcome was all-cause death.

At a median follow-up of 5.4 years, 25% of the patients underwent aortic valve surgery (AVS) a median 14 months after their index echo. Of the patients undergoing AVS, 46% developed typical symptoms, 12% developed LVEF < 50%, 5% had LV end-systolic dimension (ESD) > 50 mm or LV end-systolic volume index (ESVi) > 25 mm/m2, and 22% developed LV end-diastolic dimension (EDD) > 65 mm (all class I or II indications). Overall mortality was 17% during follow-up. Of these 83 patients, 66 died under medical follow-up, and 17 died after AVS.

A multivariate analysis showed linear LVEF, LVESDi, LVESVi, and volumetric LVEF all were independently associated with mortality with similar C statistics (0.83-0.84). The curves for continuous risk of death started to rise at linear and volumetric LVEFs < 60%, LVESVi more than 40-45 mL/m2, and LVESDi more than 21-22 mm/m2. However, as dichotomized variables, LVESVi > 45 mL/m2 exhibited a higher risk of death (HR, 1.93; 95% CI, 1.1-3.4; P = 0.02), whereas LVESDi > 20 mm/m2 did not (P = 0.32). The authors concluded that in asymptomatic patients with chronic moderately severe to severe AR, LVESVi and volumetric LVEF were equally valuable for identifying those at higher risk of death as LVESDi and linear LVEF. Also, an LVESVi > 45 mL/m2 was significantly associated with more death.


One of the main reasons many echocardiographic labs still measure M-mode LV dimensions and their fractional change is that the 2017 AHA/ACC guidelines still recommend their use for decision-making in chronic AR patients.1 Volumetric measurements, especially of EF, could be better, but this has not been established. Thus, this study by Yang et al is of interest. Surprisingly, volumetric measurements were equal, but not better than linear measurements for predicting mortality in chronic asymptomatic AR patients. However, when cutpoints suggested by other studies were used, LVESVi > 45 mL/m2 was statistically significant at predicting mortality, whereas LVESDi > 20 mm/m2 was not. Also, LVESVi was superior to LVEDVi at predicting mortality. This suggests LVESVi is the volume measure to follow if one does not want to rely on M-mode echoes.

The results with EF were surprising in a couple of ways. First, EF by either linear estimations or volumetric calculations were equally discriminatory for mortality. The formula used for estimating EF from linear measurements was not specified in the methods. The references the authors quoted were not supportive of any of the methods that have been proposed (e.g., the Teichholz formula, Quinones equation) in comparison to volumetric measurements. Second, mortality rates started to increase at < 60%. The guidelines cutpoint for considering surgery in an asymptomatic patient is < 50%, and the lower limit of normal for two-dimensional volumetric LVEF is 55%. It seems LVEF by any means lacks precision for establishing a decision cutpoint in chronic AR. This aligns with my general suspicion of using ratios for any clinical decisions. Cardiologists often do not know whether the numerator or the denominator is driving the value.

There were limitations to this study. It was retrospective, observational, and occurred at one center. The presence or absence of symptoms was not verified by exercise testing. The authors did not exclude patients with atypical symptoms, which were present in 7% of the study population. About 10% of patients underwent only single-plane volumetric measurements because of technical difficulties. (Although < 10% were rejected for inadequate images). The study population was relatively young (mean age, 60 years), with 37% presenting with a bicuspid aortic valve. There were few deaths (17% mortality rate), which hinders the precision of the mortality risk estimates. Also, only deaths under medical surveillance were considered in the mortality risk calculations, which were 80% of the total deaths. Twenty percent of total deaths occurred within 30 days of surgery, which was performed on 121 patients (25%) during follow-up, resulting in a 14% surgical mortality. In addition, no other outcomes were considered, such as hospitalizations for heart failure or cardiac death in particular. Finally, these data may not be applicable to patients with chronic AR who are suitable candidates for transcutaneous valve replacement.


  1. Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC guideline for the management of patients with valvular heart disease: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2017;70:252-289.