Novel quantitative assessment of myocardial perfusion by harmonic power Doppler imaging during myocardial contrast echocardiography

Abstract

Objective: To test the hypothesis that the power of the received signal of harmonic power Doppler imaging (HPDI) is proportional to the bubble concentration under conditions of constant applied acoustic pressure, and to determine whether a new quantitative method can overcome the acoustic field inhomogeneity during myocardial contrast echocardiography (MCE) and identify perfusion abnormalities caused by myocardial infarction.

Methods: The relation between Levovist concentration and contrast signal intensity (CI) of HPDI was investigated in vitro under conditions of constant acoustic pressure. MCE was performed during continuous infusion of Levovist with intermittent HPDI every sixth cardiac cycle in 11 healthy subjects and 25 patients with previous myocardial infarction. In the apical views myocardial CI (CI(myo)) was quantified in five myocardial segments. The CI from the left ventricular blood pool adjacent to the segment was also measured in dB and subtracted from the CI(myo) (relative CI (RelCI)).

Results: CI had a logarithmic correlation and the calculated signal power a strong linear correlation with Levovist concentration in vitro. Thus, a difference in CI of X dB indicates a microbubble concentration ratio of 10(X/10). In normal control subjects, CI(myo) differed between the five segments (p < 0.0001), with a lower CI(myo) in deeper segments. However, RelCI did not differ significantly between segments (p = 0.083). RelCI was lower (p < 0.0001) in the 39 infarct segments (mean (SD) -18.6 (2.8) dB) than in the 55 normal segments (mean (SD) -15.1 (1.6) dB). RelCI differed more than CI(myo) between groups.

Conclusions: The new quantitative method described can overcome the acoustic field inhomogeneity in evaluation of myocardial perfusion during MCE. RelCI represents the ratio of myocardium to blood microbubble concentrations and may correctly reflect myocardial blood volume fraction.

Figure 1

Harmonic power Doppler image of Levovist solution. The contrast signal intensity (CI) decreased in proportion to depth of the solution due to ultrasound attenuation by the presence of microbubbles. The region of interest (ROI) was placed just below the border between the jerry block and the solution to measure CI under conditions of constant applied acoustic pressure.

Figure 2

Relations between Levovist concentration and CI in dB with mechanical index (MI) of (A) 0.6 and (B) 1.0, and between the concentration and CI in squared acoustic units (AU²) with MI of (C) 0.6 and (D) 1.0. In A and B, the concentration is expressed in logarithmic scale where the value in dB was calculated as 10 × log(concentration). In addition, CI in dB can be regarded as the log converted value of the power of the ultrasound signal as 10 × log(power).

Figure 3

Representative results of measurement of relative CI (RelCI). Images obtained (A) at the higher Doppler gain of 65% and (B) at the lower Doppler gain of 45% were analysed. The RelCIs were calculated as −15.9 dB in the basal septal segment, −17.3 dB in the mid-septal segment, −16.6 dB in the apical segment, −16.0 dB in the mid-lateral segment, and −16.1 dB in the basal lateral segment.

Figure 4

(A) Myocardial CI (CI_myo) and (B) RelCI in normal control subjects. Apical, apical segment; Basal L, basal lateral segment; Basal S, basal septal segment; Mid L, mid-lateral segment; Mid S, mid-septal segment. ANOVA, analysis of variance; NS, not significant.

Figure 5

(A) CI_myo and (B) RelCI in normal and infarct segments.