A method for determining local pulse wave velocity in human ascending aorta from sequential ultrasound measurements of diameter and velocity

Abstract

Background: Pulse wave velocity (PWV) is an indicator of arterial stiffness, and predicts cardiovascular events independently of blood pressure. Currently, PWV is commonly measured by the foot-to-foot technique thus giving a global estimate of large arterial stiffness. However, and despite its importance, methods to measure the stiffness of the ascending aorta are limited.

Objective: To introduce a method for calculating local PWV in the human ascending aorta using non-invasive ultrasound measurements of its diameter (D) and flow velocity (U).

Approach: Ten participants (four females) were recruited from Brunel University students. Ascending aortic diameter and velocity were recorded with a GE Vivid E95 equipped with a 1.5-4.5 MHz phased array transducer using M-mode in the parasternal long axis view and pulse wave Doppler in the apical five chamber view respectively. Groups of six consecutive heartbeats were selected from each 20 s run based on the most similar cycle length resulting in three groups for D and three for U each with six waveforms. Each D waveform was paired with each U waveform to calculate PWV using ln(D)U-loop method.

Main results: The diastolic portions of the diameters or velocities waveforms were truncated to allow the pairs to have equal length and were used to construct ln(D)U-loops. The trimmed average, excluding 10% of extreme values, resulting from the 324 loops was considered representative for each participant. Overall mean local PWV for all participants was 4.1(SD = 0.9) m s^-1.

Significance: Local PWV can be measured non-invasively in the ascending aorta using ultrasound measurements of diameter and flow velocity This should facilitate more widespread assessment of ascending aortic stiffness in larger studies.

Figure 1.

(a) Concatenated frames of a cine-loop of a M-mode recording used to determine D waveforms in a volunteer. (b) ECG signal as extracted from the concatenated image. The horizontal black line is the R-wave peak determined as the time of the maximum of the ECG following the crossing with the threshold of 0.8 * ECG height above which R-wave peaks are detected. The detection of the peaks is highlighted by the vertical lines.

Figure 2.

(a) A representative ultrasound image (M-mode) of the ascending aorta for one cardiac cycle. The tracing achieved by off-line analysis on the inner arterial walls is superimposed in red between two consecutive R-peaks as detected by the analysis and shown as vertical white lines. (b) The resulting diameter waveform calculated by subtraction of the two walls traced in (a). The waveform is shown between the same R-wave peaks (depicted by the black vertical lines). Time is shown in pixels where 1 pixel  =  3.5 ms.

Figure 3.

(a) A representative ultrasound image of the PW Doppler velocity for one cardiac cycle in the apical five chamber view (A5CH) view of the ascending aorta. The tracing achieved by off-line analysis (red points) is superimposed between two consecutive R-wave peaks. (b) The resulting velocity waveform extracted from the ultrasound image between the same two consecutive R-wave peaks.

Figure 4.

Representative examples of (a) a diameter waveform (b) a velocity waveform and (c) a ln(D)U-loop for the diameter and velocity shown in (a) and (b). The red line indicates the linear fit from which the pulse wave velocity (PWV) is calculated. In this example it is 4.1 m s⁻¹. The red dots in (a) and (b) represent the onset of upstroke of the diameter and velocity waveforms used to align the start of the ln(D)U-loop in order to estimate PWV. Note that the ln(D)U loop is plotted for the entire cardiac cycle, although the determination of PWV depends only on the slope of the curve during early systole.