Doppler Echo Risk Stratification in Aortic Stenosis

By Michael H. Crawford, MD, Editor

SOURCES: Rusinaru D, et al. Relation of dimensionless index to long-term outcome in aortic stenosis with preserved LVEF. JACC Cardiovasc Imaging 2015;8:766-775.

Zoghbi WA. Velocity acceleration in aortic stenosis revisited. JACC Cardiovasc Imaging 2015;8:776-778.

The accurate estimation of the severity of aortic stenosis (AS) by Doppler echocardiography is hampered by the need for left ventricular outflow tract (LVOT) area in the continuity equation. LVOT area is estimated from LVOT diameter, which is difficult to measure and assumes a circular subvalvular channel, which it probably isn’t. The dimensionless index (DI) is the ratio of LVOT time velocity integral (VTI) to the aortic VTI, with severe AS defined as a DI of < 0.25. It is basically the continuity equation without LVOT area. The purpose of this study was to test the hypothesis that DI predicts outcome after AS diagnosis. They enrolled 488 patients from two tertiary hospitals in France and defined more than mild AS as having a valve area of < 2 cm² and a left ventricular ejection fraction (LVEF) > 50%. Patients with more than mild left heart valve regurgitation, non-native valve AS, angina, heart failure, or syncope were excluded. Patients were managed by their own physicians. The mean age of the population was 76 years and 57% were men. All had no or minimal symptoms and were followed for a mean of 32 months (range 8-58). The primary outcome was all-cause death or aortic valve replacement (AVR). The secondary outcome was cardiac death. Receiver operating curve analysis showed that a DI of 0.25 was the best cutoff for predicting a peak aortic jet velocity of > 4 m/s and a DI of 0.20 for a jet velocity of > 5 m/s. There were 241 primary events (117 death and 124 AV replacements). The 5-year event-free survival was 56% for a DI > 0.25, 41% for a DI between 0.20-0.25, and 22% for a DI < 0.20. Multivariate analysis showed that at DIs > 0.25 there was no increase in risk with decreasing DI, but with DIs < 0.25 decreasing DI showed increased risk. Also, using DI as compared to AV area and peak jet velocity improved the prediction of events (C-statistic 0.714 vs 0.674 and vs 0.667, respectively, P = 0.002). DI was not associated with all-cause death, but did predict cardiac death (hazard ratio, 2.08; 95% confidence interval, 1.06-4.11; P = 0.034). The authors concluded that DI is an accurate indicator of AS severity of clear prognostic value, and a cutoff of < 0.25 should be used for decision making.

COMMENTARY

The determination of the severity of AS by Doppler echo is the standard approach, but it does have pitfalls. Using peak velocity can underestimate the severity if the Doppler beam isn’t within 20 degrees of the jet or if the patient has a low-flow state, such as reduced LV function. It can also overestimate the severity of AS if the ascending aorta is small because of the phenomenon of pressure recovery. Since the traditional gold standard is the invasive determination of valve area by the Gorlin equation, Doppler echo has been used to calculate aortic valve area by the continuity of flow equation. One must remember that flow is the product of the velocity multiplied by the cross-sectional area of the flow channel. The difficulty in determining the cross-sectional area of the LVOT flow accurately has hampered the estimation of aortic valve area by Doppler echo. Hence, there has been interest in using the continuity equation without the LVOT area, which becomes the ratio of the LVOT VTI divided by the aortic valve VTI or the DI. The DI has not been used extensively because of a paucity of data, especially with regard to outcomes.

This study is the first large study of the utility of the DI to predict death or AVR in mainly asymptomatic patients with more than mild AS and LVEF > 50%. In other words, these patients had significant AS, but did not meet criteria for immediate AVR. They found that DI in this population predicts AVR or cardiac death during follow-up, regardless of other co-morbidities. Additionally, there is an abrupt change in the slope of this relationship at a DI < 0.25 or when the aortic flow velocity is more than four times the LVOT flow velocity. They propose that a DI < 0.25 defines severe AS. This cutoff is in agreement with previous studies. Also, the DI is not affected by LVOT diameter, body surface area, or stroke volume. However, these were patients with relatively normal stroke volumes. DI does not solve the problem of low-flow, low-gradient AS patients. In addition, as Dr. Zoghbi’s editorial points out, DI could be affected by extremes in LVOT size as might be seen in those with very dilated LVs or very small LVs due to extreme hypertrophy. In the average patient with an LVOT diameter around 2 cm (1.8-2.2 cm) there should be no problem. However, both the authors and Dr. Zoghbi recommend a multi-measurement approach considering valve anatomy, peak velocity, AV area, and the DI in patients with normal LV function to estimate the severity of AS.