Between January 2015 and November 2017, we prospectively enrolled 190 patients diagnosed with moderate-severe native valvular AS (AVA <1.5cm2). The exclusion criteria were concomitant moderate or severe aortic regurgitation or mitral valve disease, subvalvular or supravalvular AS (defined as velocity higher than 1.5 m/s), thoracic aortic diameter less than 25mm, children <18 years of age, congenital heart disease (except bicuspid aortic valve), history of valve repair or implantation, suboptimal acoustic window, and loss of follow-up. The study protocol was approved by the Ethics Committee of the province of Cadiz. All the participants gave their consent to participate in the study.

Clinical data included age, sex, hypertension, diabetes, history of smoking, hypercholesterolemia, body mass index, chronic renal failure, and coronary heart disease. Medical records were carefully screened for the presence of dyspnea, angina or syncope using the New York Heart Association functional classification. Patients manifesting angina, syncope, congestive heart failure or exercise dyspnea class ≥ 2 were classified as having symptoms attributable to AS. Blood pressure was measured at the time of the echocardiographic evaluation. BSA was calculated using the Dubois formula.8

Clinical decisions on medical management were made by the referring physician based on AS severity, left ventricular function, and symptomatic status, according to guidelines.2,3

Two-dimensional transthoracic echocardiographic and Doppler studies were obtained with clinical ultrasound machines equipped with 2.5 to 3.5MHz transducers (iE33 Phillips Medical Systems, The Best, The Netherlands). All tests were conducted by an experienced sonographer. The parasternal long-axis view with zoom was used to measure the aortic annulus diameter in early systole. The time-velocity integral was obtained by using the pulsed Doppler in the left ventricular outflow tract and placing the sample volume 1cm below the aortic valve. Stroke volume was then calculated assuming a circular shape of the left ventricular outflow tract. Continuous wave Doppler recording of flow through the valve was performed from different windows to record maximal instantaneous and mean pressure gradients across the aortic valve.

AVA was calculated using the continuity equation. An indexed AVA was estimated as AVA/BSA. Because of the uncertain role of acquired fatty tissue on indexed AVA, we also calculated AVA/height. The mean transvalvular pressure gradient was obtained with the use of the modified Bernoulli equation. A Doppler velocity index, a simplification of the continuity equation, was calculated as time-velocity integral of left ventricular outflow tract/time-velocity integral of aortic jet.

The systolic time intervals of flow through the aortic valve were measured using the velocity curve from the continuous wave Doppler recording: ejection time, acceleration time, and acceleration time/ejection time ratio, as previously described.9

A dobutamine stress echocardiogram was performed when AVA calculated by the continuity equation was less than 1.0cm2, the aortic transvalvular mean gradient was less than 40mmHg, and the left ventricular ejection fraction was less than 40%. Severe AS was suggested by an AS jet> 4 m/s or a mean gradient> 40mmHg and an AVA that did not exceed 1.0cm2 on stress echocardiography.

All measurements represent an average of 3 cardiac cycles for patients in sinus rhythm and at least 6 cycles if the patient was in a different rhythm than sinus rhythm. Estimation of extrasystolic beat was always avoided. Echocardiographic study was performed following the AHA guidelines.10 Doppler recordings were performed at a sweep speed of 150mm/s.

The primary endpoint was cardiovascular death under medical management. In patients who underwent aortic valve implantation, we used the implantation date to compute the length of follow-up, but aortic valve implantation was not part of the primary endpoint. A combined endpoint (cardiovascular death and aortic valve replacement) and global mortality were defined as secondary endpoints.

Outcomes data were obtained from patient visits or records, telephone interviews, or death certificates when applicable.

Data are expressed as mean±standard deviation for continuous variables and were compared using the unpaired t test. Categorical variables are expressed as percentages and were compared using chi-square analysis or the Fisher exact test. The relationship between continuous variables and the 3 groups was explored using 1-way ANOVA tests. A receiver-operator characteristic curve was plotted to determine the best AVA/BSA and AVA/height cutoff values for predicting cardiovascular death in patients with AS. This cutoff was determined as the value providing a balance between sensitivity and specificity, but we preferred to use a higher sensitivity cutoff value because of the high mortality of this disease. The area under the receiver operating characteristic curve was calculated.

Event rates were estimated according to the Kaplan-Meier method and compared with 2-sided log-rank tests. Univariate and multivariable analyses of time to events were performed using Cox proportional hazard models. The variables entered into the models were those that have demonstrated prognostic impact in AS: age, left ventricular ejection fraction, mean gradient, peak aortic velocity, symptomatic status, AVA/BSA, and indexed left ventricular mass.

Differences were considered significant at P values <.05. For data analysis, the statistical program SPSS version 17.0 (SPSS Inc., Chicago, Illinois, United States) was used.

Receiver operating characteristic analysis (figure 1) showed that both AVA/BSA and AVA/height could significantly predict cardiovascular death. The largest area under the curve was for AVA (area under the curve 0.75, P <.001), AVA/height (area under the curve 0.74, P <.001), and AVA/BSA (area under the curve 0.72, P <.001). Using a cutoff value of 0.50, AVA/BSA had a sensitivity of 96% and specificity of 51%, negative predictive value of 99% and positive predictive vale of 22%. The guideline-recommended cutoff value of 0.60cm2/m2 had a sensitivity of 97%, specificity of 25%, negative predictive value of 96%, and positive predictive value of 15%.

Figure 1.

Receiver operating characteristics curves for predicting cardiovascular mortality. AVA, aortic valve area; BSA, body surface area; ROC, receiver operating characteristic.

(0.11MB).

For AVA/height, a cutoff of 0.49, it showed a sensitivity of 96%, specificity of 52%, negative predictive value of 95% and positive predictive value of 25%. A cutoff value of 1.0cm2 for AVA showed a sensitivity of 96%, specificity of 38%, negative predictive value of 98%, and positive predictive value of 18%.

We divided the patients into 3 groups according to calculated (0.50cm2/m2) and guideline-recommended (0.60cm2/m2) cutoff values for AVA/BSA. Among the 190 patients included, 41 (22%) patients had AVA/BSA> 0.60cm2/m2, 40 patients (21%) between 0.50 and 0.60cm2/m2, and 109 patients (57%) had AVA/BSA less than 0.50cm2/m2. The baseline characteristics of patients are compared in table 1. Patients with AVA/BSA <0.50 showed more severe echocardiographic parameters (table 2).

According to AVA/height, we divided the patients into 3 groups: 73 patients (38%) with AVA/height> 0.6cm2/m, 44 patients (22%) with AVA/height between 0.49 and 0.6cm2/m, and 73 patients (38%) AVA/height <0.49cm2/m. The clinical and echocardiographic characteristics of the patients according to AVA/height are shown in table 3 and table 4.

Complete follow-up was achieved in 100% of the patients. Mean follow-up was 247±190 days, without significant differences according to the AVA/BSA group (226±150 vs 245±262 vs 256±172 days, P=.70) and AVA/height group (235±220 vs 220±134 vs 275±184 days, P=.26).

There were 36 deaths during the follow-up, of which 24 were from cardiovascular causes. Aortic valve replacement was performed in 51 patients (27%). The combined endpoint was reached by 75 patients (39%).

Survival according to AVA/BSA is presented in figure 2. Cardiovascular mortality was higher in patients with lower AVA/BSA (2.5% vs 0% vs 21%, P <.001). There were more cardiovascular combined events in AVA/BSA <0.50 (10% vs 12% vs 61%, P <.001) but global mortality did not reach statistical differences (7% vs 17% vs 24%, P=.07). One-year survival was 91±5% for AVA/BSA> 0.60cm2/m2, 100±0% for AVA/BSA between 0.50 and 0.60cm2/m2, and 75±5% for AVA/BSA <0.50cm2/m2 (log-rank P=.001). Moreover, 2-year survival was 95±5% in patients with AVA/BSA> 0.50cm2/m2 and 37±5% in patients with AVA/BSA ≤ 0.50cm2/m2P <.001).

Figure 2.

Kaplan-Meier plots showing risk for cardiovascular death according to AVA/BSA. AVA, aortic valve area; BSA, body surface area.

(0.11MB).

We did not find different survival rates in patients with AVA/BSA> 0.6cm2/m2 and AVA/BSA between 0.5 and 0.6cm2/m2 (P=.39).

Regarding AVA/height, there were more cardiovascular deaths in patients with lower AVA/height (1% vs 11% vs 25%, P <.001), more cardiovascular events (7% vs 57% vs 62%, P <.001) and global mortality (14% vs 11% vs 29%, P=.02). One-year survival (figure 3) was 95±5% for AVA/height> 0.6cm2/m, 87±5% in patients with AVA/height between 0.6 and 0.49cm2/m, and 72±5% for AVA/height <0.49cm2/m (log-rank P=.001). Two-year survival was 91±4% in patients with AVA/height ≥ 0.49cm2/m and was 36±4% in AVA/height <0.49cm2/m (P=.001).

Figure 3.

Kaplan-Meier plots showing risk for cardiovascular death according to AVA/height. AVA, aortic valve area.

(0.12MB).

We also found significantly different survival rates after excluding patients with left ventricular dysfunction: 1-year survival was 81±4% in patients with AVA/BSA ≤ 0.50cm2/m2 and 94±5% in AVA/BSA> 0.5cm2/m2 (P=.003). One-year survival was also higher in AVA/height ≥ 0.49cm2/m than in <0.49cm2/m (92±4% vs 78±6%, P=.006) after exclusion of patients with left ventricular systolic dysfunction.

There were only 41 asymptomatic patients with left ventricular ejection fraction ≥ 50%, of which 1 patient had cardiovascular death (AVA/BSA <0.5cm2/m2).

The excess risk of cardiovascular mortality in univariate analysis for AVA/BSA <0.5cm2/m2 (HR, 16.3; 95% confidence interval [95%CI], (2.2-120.9), P=.001) was also observed after adjustment by age, sex, left ventricular ejection fraction, symptomatic status, aortic peak velocity, and indexed left ventricular mass (HR, 10.9; 95%CI, (1.2-103.7]), P=.037) (table 5). However, a higher cardiovascular mortality in univariate analysis for patients with AVA/height <0.49cm2/m (HR, 4.3; 95%CI, (1.7-11.0), P=.002) was not significant in multivariate analysis (HR, 2.0; 95%CI, (0.6-6.0), P=.22).