Prospects of microwave spectrometry for vascular stent monitoring. Towards a non-invasive and non-ionizing follow-up alternative

Abstract

[eng] Throughout this thesis we have assessed the prospects of microwave spectrometry (MWS) as a non-ionizing non-invasive monitoring alternative for stented patients in a very early proof-of-concept stage. In Chapter 1 we have provided a generalist retrospective medical background along with a state-of-the-art summary of existing microwave-based stent monitoring approaches. First, we have introduced cardiovascular diseases in general, and ischemic heart disease in particular. Next we have reviewed how percutaneous coronary interventions addressed the medical problem represented by atherosclerosis, giving a special emphasis to balloon angioplasty, bare-metal stenting and drug-eluting stenting. We have further exposed how the outcomes of such revolutionary strategies were compromised by the high rates of post-procedural complications, making unavoidable the invasive and ionizing follow-up of stented patients. Finally, we have summarized existing non-invasive and non-ionizing stent monitoring alternatives based in microwave techniques. In Chapter 2 we have introduced the working principle of our MWS setup. We have first presented how this arrangement can obtain the absorbance of a stent as a function of the frequency and the incidence angle of the microwave fields. We have also shown how these data are combined in a single two-dimensional chart, and how we recognize therein the characteristic resonance frequencies of stents at a glance. As an example, we have presented a typical absorbance diagram to illustrate the general features of such resonances. In particular we have highlighted that these resonances are discrete and have multi-lobed angular patterns. In Chapter 3 we have characterized many stents having a wide variety of nominal sizes to better understand their characteristic resonances in terms of microwave scattering. First, we have found that the resonance frequency obeys a reciprocal dependence on the stent length. This has allowed us to obtain an empirical expression for such relationship just by adjusting two fitting parameters. However, we have not been able to find an analogous expression for the dependence on the stent diameter. In any case, while investigating the latter, we have unexpectedly uncovered how the particular stent architecture influences the corresponding resonance frequencies. By gathering all these individual results we have finally suggested a straightforward half-theoretical half-empirical model linking the resonance frequencies of stents with their structural integrity (through their length), with their particular architecture (through the scaling factor), as well as with their surrounding medium (through the dielectric permittivity and the magnetic permeability). We have also theoretically estimated the resonance frequencies of implanted stents from their corresponding values in free space conditions, showing that in vivo resonance frequencies should be around one order of magnitude smaller than their free space counterparts. Finally, in Chapters 4 and 5 we have explored the potential diagnostic capabilities of MWS in two possible scenarios: stent fracture (SF) and in-stent neoatherosclerosis (ISNA). We have started both chapters reviewing the incidence, the medical implications, and the mechanism of these two stent-related complications. SF has been evaluated in Chapter 4 by means of two “fracture tests” consisting in a successive series of strut cuts. We have shown that MWS provides qualitative indicators for single and multiple strut fractures (downshift of the fundamental resonance frequency), and also quantitative indicators for single or multiple complete transverse linear SFs (split and upshift of that frequency). ISNA has been evaluated in Chapter 6 by means of four ``cholesterol tests'' consisting in a gradual process of increasing cholesterol deposition. We have shown that MWS provides an indicator for a growing presence of cholesterol around a stent (downshift of the fundamental resonance frequency). We have concluded this chapter calculating the theoretical evolution of the resonance frequencies along a cholesterol deposition process, estimating the upper limit for the resonance frequency displacement. Taking together the results we have reported in Chapters 5 and 6, we have shown that MWS could potentially warn about SF and ISNA.