Between August 2017 and January 2018, we prospectively included consecutive patients diagnosed with AS (peak velocity >2.5m/s) who attended echocardiography laboratory. Exclusion criteria were: age <18 years, ascending aorta less than 25mm, other moderate or severe heart valve disease, significant subaortic or supraaortic stenosis, heart valve prostheses and congenital heart disease (except aortic bicuspid valve). This protocol was approved by our institutional review board.

Cardiac physical examination was performed at the initial visit by a cardiologist who was blinded to the echocardiographic results. The following variables were evaluated using a standardized form: Levine scale for grading intensity murmur (from 1 to 6)5; murmur duration (early, middle, or late systole); second heart sound characteristics (present or absent); murmur radiation to the apex or neck; Gallavardin phenomenon (higher frequency of the systolic murmur at the apex),6 delay in carotid upstroke (0=none to 4=severe), decrease in carotid amplitude (0=none to 4=severe).

Clinical data included age, sex, arterial hypertension, diabetes, history of smoking, hypercholesterolemia, body mass index, chronic renal failure, and coronary heart disease. Patients manifesting angina, syncope, congestive heart failure or exercise dyspnea class ≥2 were classified as presenting symptoms attributable to AS. Blood pressure was measured at the time of the echocardiographic evaluation.

Two-dimensional transthoracic echocardiographic and Doppler studies were obtained with clinical ultrasound machines equipped with 2.5–3.5MHz transducers (iE33 Phillips Medical Systems, The Best, The Netherlands). All tests were conducted by an experienced sonographer. Parasternal long axis view with zoom was used for measuring the aortic annulus diameter in early systole. The time–velocity integral was obtained. Using the pulsed Doppler in the left ventricular outflow tract, 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 the 5-chambers and right parasternal windows to record maximal instantaneous and mean pressure gradients across the aortic valve.

Effective orifice area (EOA) was calculated using the continuity equation. An indexed EOA was estimated as EOA/body surface area. 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. As a measure of global left ventricular afterload, we calculated the valvuloarterial impedance (zva) as (systolic arterial pressure+mean gradient)/systolic volume index.7 We also estimated the systolic work loss with the following formula8:

Acceleration time (AT), ejection time (ET) and AT/ET ratio were also calculated.9

All measurements represent an average of three cardiac cycles for patients in sinus rhythm and at least six cycles if the patient was in a rhythm different from/other than the sinus rhythm. The estimation of extrasystolic beat was always avoided in all cases. Doppler recordings were performed at a sweep speed of 150mm/s.

AS was considered severe when EOA <1cm2, moderate with EOA between 1 and 1.5cm2, and mild if EOA >1.5cm2, according to American guidelines.10

The right brachial artery was imaged approximately 5–10cm proximal to the antecubital fossa in the longitudinal plane. All brachial artery sonograms were obtained on a Vivid T8 (General Electrics, United States) sonography unit with a linear 5.0-MHz linear transducer. A pulsed Doppler volume of 1.2mm was obtained in the middle of the brachial artery with a maximum Doppler angle of 60°. Peak systolic and diastolic velocity, time-averaged flow velocity and time from onset to peak velocity (brachial acceleration time) were calculated.

To assess interobserver variability in physical examination variables, a second cardiologist, blinded to previous physical and echocardiographic findings, performed a complete physical examination in 20 patients (10 with severe and 10 with mild or moderate AS). For the study of variability, kappa index was used.

Continuous variables are presented as mean±standard deviation and compared using the unpaired Student t-test. Categorical variables were expressed as percentages and compared using chi-square analysis or Fisher's exact test.

The comparison between quantitative variables was calculated by Pearson correlation.

We prospectively enrolled patients with AS diagnosed in our echocardiography laboratory since February 2018, which formed the validation group. The patients diagnosed between August 2017 and January 2018 constituted the derivation group.

A receiver-operator characteristic (ROC) curve was plotted in the derivation group for physical examination and brachial artery Doppler parameters to determine the best cut-off value for identifying severe AS. This cut-off value was determined as the value providing a balance between sensitivity and specificity. The area under the ROC curve was calculated. The calculated cut-off value was then tested in the validation cohort.

Differences were considered significant at P values <.05. For data analysis, the statistical program SPSS version 20.0 (SPSS Inc., United States) was used.

All subjects had an audible systolic murmur from grade 2 to 4. Murmur intensity (r=−0.32; P=.003) and duration (r=−0.54; P<.001) correlated with EOA (Fig. 1). Differences in physical examination variables between both groups are shown in Table 2.

Fig. 1.

Effective orifice area (EOA) values based on murmur intensity and duration.

(0.09MB).

The area under the curve (AUC), sensitivity, specificity, negative and positive predictive values of physical examination findings for detection of severe AS were examined (Table 3). Note that some physical examination variables, such as murmur intensity higher than or equal to 4 or a delayed carotid upstroke (pulsus tardus ≥2) have a great specificity but with low sensitivity, whilst other variables have high sensitivity, such as murmur intensity ≥3 or irradiation to the apex or the neck. However, the physical examination findings with higher AUC were second heart sound absent (AUC 0.85, P<.001), late-systole murmur duration (AUC=0.79, P<.001) and pulsus tardus ≥2 (AUC=0.77, P<.001) for severe AS.

We also analyzed 30 patients with body mass index ≥30kg/m2, with similar AUC than the general sample: second heart absent (AUC=1.0, P=.004), late-systole murmur duration (AUC=0.71, P=.22), pulsus tardus (AUC=0.70, P=.24), pulsus parvus (AUC=0.70, P=.24), and Gallavardin sign (AUC=0.80, P=.09).

Pulsus tardus >2 was associated with higher aortic peak velocity (4.6±0.5 vs 3.6±0.7m/s; P<.001), mean gradient (52.8±11.1 vs 33.2±14.1mmHg; P<.001), acceleration time (104±25 vs 84±22ms; P=.002), AT/ET ratio (0.35±0.05 vs 0.27±0.06; P<.001), indexed left ventricular mass (141±20 vs 120±30g/m2; P=.003) and lower EOA (0.76±0.14 vs 1.1±0.32cm2; P<.001) and indexed EOA (0.43±0.08 vs 0.60±0.18cm2/m2; P<.001).

Brachial artery AT was higher in more severe AS. We also found a non-significant trend to lower peak systolic velocity in patients with more severe AS (Table 4). A ROC curve was plotted to evaluate the optimal cut-off point in brachial AT for detecting severe AS (Fig. 2): AT had an AUC of 0.71 (0.55–0.83, P<.001), and a cut-off value of 84ms had the better balance between sensitivity and specificity.

Fig. 2.

Receiver-operator characteristic (ROC) curves to differentiate severe aortic stenosis from other stages based on measurement of brachial acceleration time.

(0.08MB).

Validation group (58 patients) was composed by 32 patients (55%) with non-severe AS and 26 patients (45%) with severe AS. There were no significant differences in clinical basal characteristics with derivation group.

In the validation group, using a cut-off value of 84ms, it had a sensitivity of 46%, specificity of 87%, positive predictive value of 75% and negative predictive value of 67%. A cut-off value of 100ms had a specificity of 100% for severe AS. Using a cut-off value of 50ms in brachial arterial AT, the negative predictive value was 100% for severe AS.

A significant correlation was found between brachial AT and echocardiographic AT (r=0.46, P=.001), AT/ET ratio (r=0.47, P=.001) and semi-quantitative pulsus tardus (r=0.64, P<.001).

There was an acceptable correlation between the two observers in physical examination variables (Table 5). Physical findings with higher diagnostic accuracy for detecting severe AS showed a good interobserver variability: second heart sound absent (kappa index 0.83, P=.001), pulsus tardus (kappa index 0.81, P=.003) and murmur duration (kappa index 0.82, P=.002).

Although other authors have used peak velocity as reference parameter of AS severity,14–16 in our study severe AS was defined as an EOA <1cm2 since peak velocity and mean gradient are flow-dependent and may be unrepresentative of the grade of AS at extremes of physiologic flow,18 according to current echocardiographic recommendations.10

However, it must be taken into account that there is not an ideal measure of AS severity because of the lack of a standard of reference in this heart valve disease.19,20

The systolic murmur of AS is generated when blood passing through a narrowed orifice becomes turbulent. The characteristics of the murmur are determined by the velocity of the jet (that depends on AS severity), but also by direction of the jet and variable transmission of the sound through the chest wall.21

Absence of the second heart sound could be produced by a prolonged left ventricular ET in severe AS or severe calcification of aortic cusps leading to an inaudible aortic closure sound. Nevertheless, a split second heart sound could be heard in patients with severe AS and pulmonary hypertension, right ventricle conduction delay or non-severe calcification of aortic cusps.22

A pulsus parvus et tardus is a well-recognized clinical finding in severe AS.23 The prolonged carotid peak intensity may be due to delayed closing of aortic valve during ejection in severe AS24; this is in agreement with the correlation found in our study between AT and pulsus tardus. Pulsus parvus is due to a decreased stroke volume and it is a typical finding of severe AS.25

Although our aim was to demonstrate the diagnostic accuracy of physical examination in AS, these parameters have a subjective component. For assessing pulse alterations in an objective way, we decided to perform an artery Doppler study.

Although carotid pulse waveform may be affected by artery compliance, impedance, and reflected pressure waves, so coexisting hypertension or atherosclerosis may result in an early peak and normal amplitude carotid upstroke despite severe AS.26 Our data show that AT in brachial artery is higher as the AS is more severe, and there was a non-significant trend to a decreased peak systolic velocity in severe AS, similar to “subjective” physical examination findings.

One of the most important finding of our study was that an AT cut-off of 84ms in brachial artery Doppler has a high specificity for determining severe AS. To our knowledge, there are no studies that were able to establish a cut-off value for severe AS using a peripheral artery Doppler with an acceptable accuracy; that is, we could say that it is the first time that pulsus tardus has a cut-off value.

In our opinion, it could be useful to begin with the highest sensitivity parameters (murmur intensity ≥3 and neck irradiation) to discriminate patients who could be assess by echocardiography. On the other hand, we believe that high specificity parameters (pulsus tardus ≥2 or prolonged AT in brachial artery) could help to discriminate severe AS in case of discrepancies.

Small sample size must be underlined, so this could make difficult to reach another statistical differences. Another limitation is the short follow-up time, since there were few cardiovascular events to be able to find outcome predictors.

It could be argued that some parameters are subjective, but we demonstrated an acceptable interobserver correlation in physical examination findings. There is a wide confidence interval in all parameters because of the small sample size.

Our hospital is a reference center for echocardiography evaluation, so it is possible that a higher proportion of severe AS could be found in our sample. Therefore, caution should be taken in generalizing the results.

We decided to use brachial artery Doppler despite the fact that the central pressure curves differ from the peripheral artery,27 since carotid atherosclerosis is present in most patients with AS.28