Background

Intravascular elastography of coronary arteries in patients is challenging since the intravascular ultrasound catheter will move a lot due to the contraction of the heart muscle. If this motion is in the imaging plane, correction is feasible [1][2]. However, out of plane motion cannot be corrected and therefore needs to be avoided or minimised. Since no active control of the motion is possible, the phase of the heart cycle with minimal motion needs to be determined.

Elastography

Preliminary acquisitions were performed in patients during percutaneous transluminal coronary angioplasty (PTCA) procedures [3]. Data were acquired in patients (n=12) with an EndoSonics InVision echo apparatus equipped with an rf-output. For obtaining the rf-data, the machine was working in ChromaFlo mode resulting in images of 64 angles with unfocussed ultrasound data. The systemic pressure was used to strain the tissue. This strain was determined using cross-correlation analysis of sequential frames. A likelihood function was determined to obtain the frames with minimal motion of the catheter in the lumen, since motion of the catheter prevents reliable strain estimation (see figure 1). Minimal motion was observed near end-diastole. Reproducible strain estimates were obtained within one pressure cycle and over several pressure cycles. Validation of the results was limited to the information provided by the echogram. Strain in calcified material (0.20%) was lower (p<0.001) than in non-calcified tissue (0.51%).

Figure 1: Acquisition protocol for acquiring elastograms in vivo. The likelihood curve (green curve) shows high similarity between frames at end-diastole (pressure= blue curve, ECG=red curve). Echoframes acquired in this phase of the heart cycle are used to determine elastograms. The figure also demonstrates high reproducibility over several heart beats. Five elastograms are acquired at end-diastole in sequential heart cycles. The elastograms show similar features in all the elastograms.

High-resolution elastograms were acquired using an JOMED InVision echo apparatus. The beam-formed image mode (512 angles) ultrasound data (fc = 20 MHz) was acquired with a PC based acquisition system. Frames acquired at end-diastole with a pressure difference of approx. 5 mmHg were taken to determine the elastograms. Elastograms of soft, calcified and stented plaques were determined. The elastogram of a soft plaque, as identified from the deformation during the pressure cycle, reveals strain values up to 2% with increased strain regions at the shoulders of the plaque (figure 2). Calcified material, as identified from the echogram, shows strain values of 0-0.2%. An elastogram of a plaque before intervention shows mixed tissue in the central part of the plaque and high strain values at the shoulders. After stenting, the elastogram revealed very low strain values, except for the two shoulder regions: these are imaged between the stent struts and stille reveal the soft material at thiese locations (figure 3).

Figure 2 : Intravascular echogram and elastogram of a coronary artery obtained in vivo in a patient. The echogram suggests a calcified region between 12 and 3 o'clock. The elastogram reveals low strain values in this region corroborating this finding. High strain values were found at the shoulders of this eccentric plaque.


Figure 3 : Intravascular echogram (A) and elastogram (B) of a coronary artery obtained before intervention. A mixed plaque compostion can be observed in the central part of this plaque. High strain values were found at the shoulders. After stenting, the strain values turn to almost zero due to the initial deformation of the material by the inflation and the shielding of the rigid stent struths (D). However, high strain can still be observed at the shoulder regions. These regions are images between two struths of the stent.

Automated Palpography

Intravascular ultrasonic palpation is a one-dimensional elasticity imaging technique [4]. Experiments were conducted on patients whom were referred for percutaneous coronary intervention with either stable or unstable angina pectoris. High resolution beam formed radio frequency (RF) echo frames were captured continuously for 4 seconds at full frame rate (30 Hz) from a commercially available InVision ultrasonic imaging system (JOMED) using a fast data acquisition system. In a heart-cycle, all frames with a high likelihood are used to calculate a compund palpogram. Since the contour and the likelihood are determined automatically, this method is fully automated [5]. Figure 4 shows a representative example of a strain palpogram that was created.

Figure 4: Echogram and strain palpogram of human coronary artery acquired in vivo. The echogram reveals a calcified area between 12 and 3 o'clock. The strain palpogram identifies this region as being hard. High strain values were found at the shoulders of the plaque.

References

[1] C. L. de Korte, E. I. Céspedes, A. F. W. van der Steen. "Influence of catheter position on estimated strain in intravascular elastography" IEEE Transactions on Ultrasound Ferroelectrics and Frequency Control 46(3): 616-625; 1999
[2] C. R. M. Janssen, C. L. de Korte, M. S. van der Heiden, A. F. W. van der Steen, C. Wapenaar. Angle matching in intravascular elastography." Ultrasonics 38(1-8): 417-423; 2000
[3] C. L. de Korte, S. G. Carlier, F. Mastik, M.M. Doyley, A. F. W. van der Steen and N. Bom. "Morphologic and mechanic information of coronary arteries obtained with intravascular elastography: a feasibility study in vivo”. European Heart Journal 23(5): 405-413; 2002.
[4] E.I. Céspedes, C.L. de Korte, A.F.W. van der Steen, "Intraluminal ultrasonic palpation: assessment of local and cross-sectional tissue stiffness" Ultrasound in Medicine and Biology 26(3):385-396; 2000
[5] M. M. Doyley, F. Mastik, C. L. de Korte, S. G. Carlier, E. I. Céspedes, P. W. Serruys, N. Bom, A. F. W. van der Steen " An automated approach for clinical intravascular ultrasound palpation. " Ultrasound in Medicine and Biology. 27(11): 471-480; 2001.