Ultrasonographic investigation of the mechanics of vulnerable atherosclerotic plaques: significance of the volume strain

Introduction The concept of vulnerable atherosclerotic plaque in defining the prognostic risk for plaque rupture and thromboembolic complications is an attempt to go beyond the concept of the risk as only related to the degree of the consequent stenosis. However, interventional trials still depend on the evaluation of the degree of stenosis, used as endpoint [1,2], and do not use plaque characteristics such as aspect, morphology, structure, and mechanics. Unfortunately, no consensus has yet been reached about how to define objective criteria and methodologies for assessing plaque characteristics [3–7]. In this issue of the Journal of Hypertension, Zhang et al. [8] aim at overcoming the poor accuracy of conventional ultrasonographic parameters in measuring the relative risk for plaque complication and consequent stroke at the carotid artery level and introduce a novel technique focusing on the mechanical elastic properties of the plaques. The rationale of their study involves the concept of vulnerable plaque as a function of the plaque strain due to its structural elastic compression as represented by the relative systodiastolic dimensional change. They measured the three-dimensional (3D) systodiastolic plaque volume strain as the determinant of the compression of plaques caused by blood flow pulsation and pulse wave transfer, which, by accumulating elastic potential within the plaque, may cause deformation and stress on the fibrous cap or ‘fissuration’, thus leading to plaque rupture. They introduced a novel 3D echographic technique based on a matrix ultrasonographic beam, calculating an intriguing parameter, the volume compression ratio (VCR), that is, the systodiastolic strain of an atherosclerotic carotid artery plaque at the level of its maximal intima–media thickness. Confirming the hypothesis that reactive lipid-rich plaques should be more compressible, they found that higher VCRs were related to hypodense plaques and associated with higher risk for cerebrovascular events. Furthermore, VCR was also superior to acoustic density in predicting events. Although recent technical developments of ultrasonography have led to exciting new clinical applications, including microstructural and mechanical analyses, the present contribution raises some questions as to the meaning, usefulness, and feasibility of this and similar techniques. What are the sensitivity, accuracy and reproducibility of volume compression ratio calculation? The systodiastolic plaque volumetric strain values were very small (30–58 mm3). The authors have shown that the theoretical spatial resolution of the beam (3.9 mm3) is adequate for scanning plaque volume, which is of about 100-fold higher. However, the resolution of the systodiastolic strain is less than 1° of magnitude of the plaque volume variations, and thus very close to the limits of the theoretical spatial resolution of the equipment. This problem is similar to that of the two-dimensional measurement of the arterial systodiastolic strain for calculation of the Peterson's elastic modulus, a difficulty that is overcome by echo-tracking radiofrequency analysis, but not completely by the echographic scan. Reproducibility was also close to the magnitude of the plaque volumetric strain, and thus it is difficult to understand the meaning of the small differences calculated between groups of patients such as those with or without cerebrovascular events as shown in the article. Technical bias may also occur. For example, zooming deformation, the effect of scan angle variations and the different qualities of near and far wall scans are well known intrinsic limitations of ultrasound techniques, which are well recognized in the traditional macroscopic assessment of the carotid wall [7], but may become much greater when finer ‘microscopic’ evaluations are made. What do the measured parameters mean in terms of the relationship between plaque strain and plaque risk? The authors assume that the higher the strain of the plaque, the higher its lipid content, lower the fibrous component, or both, resulting in a greater susceptibility to rupture. However, only a weak correlation was found between VCR and the acoustic density of the plaques, despite their claim that the VCR parameter should overcome the limitations of previous techniques that have used acoustic density only and have probably been responsible for the disappointing results of several previous studies [3–6]. Furthermore, higher strain occurs in an anisotropic structure both in terms of the mechanical behavior of every single volume unit and of different mechanical responses of the plaque shoulder in comparison to the plaque core. How much of the changes in VCR are due to elongation along the longitudinal axis and how much are due to the compression? In other words, can we assume that a more compressible plaque is automatically more prone to rupture? In any case, the gain of diagnostic accuracy of VCR in predicting cerebral ischemic events, although statistically significant, was modest from a clinical standpoint, as in comparison to conventional parameters, only 4% more cases (from 71 to 75%) were identified. The new technique may be a technical improvement, going one step beyond; but how long the step and how easy to make? Finally, are the results only related to the risk of the investigated plaque, or are they representative of the general risk of the patient? Several studies have attempted to determine whether the risk of cerebrovascular ischemic events is correlated to the specific investigated plaque or to a general trend in plaque complications, independent of the site being investigated. In other words, does a ‘vulnerable’ plaque forecast a future ipsilateral event, or is it a marker of a common property for all plaques in an individual patient, and thus predict any type of future ischemic event? The authors chose the latter alternative, and pooled all ischemic events independently of whether they were ipsilateral or contralateral to the site of examination. They concluded that the mechanical characteristics of the examined plaques as evaluated by VCR could be representative of all the intrapatient cerebral vascular lesions. However, we must bear in mind that many biological and mechanical factors can induce a kaleidoscope of different structural and mechanical plaque behaviors. Shear stress, which is also dependent on geometric factors with a varied distribution in the vascular tree, is one of the most important. Most of the contributions to this topic are speculative, as there is no direct demonstration that all plaques in a given patient behave in a similar way. Fisher et al. [9] studying simple macroscopic plaque characteristics such as ulceration and thrombus formation showed that in symptomatic patients plaque ulceration is unrelated to the side of carotid symptoms, whereas thrombus is associated with ipsilateral symptoms and plaque ulceration. In any case, one can argue that ulceration is the mirror of a past event, whereas the clinical interest lies in the future of plaques, not their present, or past. The main limitation of previous studies was the retrospective nature of their design, and thus it is yet unclear whether the ulceration represents the consequence of a previous event or rather a structural change prone to future complications. The role of every newly described structural or mechanical parameter should be investigated prospectively for the risk for events, in the same way as the North American Symptomatic Carotid Endarterectomy Trial, the European Carotid Surgery Trial, and the Asymptomatic Carotid Atherosclerosis Study have investigated in the past the risk associated with the degree of stenosis [10,11]. Zhang et al. [8] opened a new window for looking at the mechanics and not only at the dimensions and the aspect of atherosclerotic plaques. This intriguing technique is promising, but deserves further investigation and it cannot yet be employed extensively.