2.5 Assessment of the delivery system digital twin

In this phase, the whole stent was considered. It allowed appreciating the phenomena that require a full-length approach and possible aspects that were neglected during the material parameter selection where the single unit was preferred. Moreover, the production process was numerically reproduced: starting from the stress-free laser-cut stent geometry, the crimping of the stent on the balloon was simulated to obtain a realistic model of the delivery system (including residual stress in the stent). The performance assessment process was defined on the basis of the available experimental and literature data [33]. In particular, three tests were considered: free-expansion of the delivery system, tension, and radial force testing of the fully expanded stent.

The first step of the assessment process consisted of a crimping and expansion simulation using Abaqus/Explicit. The crimping step was performed following the same protocol described in section 2.4.1 and applied to the device to take into account the residual stresses. The minimum diameter was set at 1.3 mm to gain a realistic post-crimping stent diameter.

After the crimping step, the balloon was expanded by applying in 10 s an increasing uniform pressure up to 7 atm (the same as the experimental procedure) to the inner surface of the balloon. According to in vitro tests, a successive 30 s holding phase was added to permit the creep phenomenon. The balloon was then deflated by applying a decreasing pressure reaching finally a slightly negative value. Contacts were activated to account for the interactions between the different parts using the General Contact method. Given the high variability of experimental results in the dogboning phase, in agreement with the literature [16], the pressure-diameter curves obtained computationally were compared with the average curves of experimental results at the end of the dogboning phase (“post-dogboning configuration”) and when the pressure reaches the nominal value of 7 atm (“final configuration”). Moreover, the numerical results were compared with the experimental measurements also in terms of internal diameter reached at the end of the pressure maintenance phase and external diameter of the fully-expanded stent (at the end of the balloon deflation phase).

The second step of the assessment path included the simulation of the tension testing of the stent. A process of import from Abaqus/Explicit to Abaqus/Standard allowed considering the residual stresses and deformed configuration deriving from the previous steps.

The force-displacement curves obtained computationally with the “fast” and “slow” protocols and the average curves of the corresponding experimental results were compared considering the maximum force, reached during the loading phase, the force measured at the end of the relaxation phase, and the residual plasticity, namely the displacement at which a zero force is measured during the unloading phase.

A further computational test, to be compared no longer with experimental results carried out in this work (only a limited number of samples were available), but with performance declared by the manufacturer, was carried out. In the datasheet of the REVA Fantom^® Encore, the manufacturer states that its device is characterized by a radial strength of 0.22 N/mm. According to the standard guideline for this procedure [34], the radial strength is measured by subjecting the device to a recrimping test. The stent is placed inside a machine that, through the radial movement of rigid planes, progressively reduces the diameter of the stent. The test is characterized by a phase of diameter reduction followed by one in which the planes are moved radially in the opposite direction to allow the elastic recoil of the stent. The radial strength value is estimated based on what is observed in a diameter-radial force plot. In this curve, during the loading step, it is possible to identify an initial phase with a nearly constant slope that corresponds to the elastic response of the stent to the diameter reduction, followed by a phase in which the curve flattens due to the increase of plastic deformations. During the unloading step, the curve has again a roughly constant slope, whose entity has to be used in the estimation of the radial strength. The radial strength corresponds to the intersection value between the recrimping loading curve and a straight line parallel to the unloading curve, with an offset with respect to the starting point of the load curve itself that corresponds to a 15% compression [33].

This test was reproduced computationally in Abaqus/Explicit: the stent, after crimping and free-expansion, was subjected to recrimping up to reach a minimum outer diameter of 1.5 mm, using the same strategy described in section 2.4.1., without interposing the holding step between the crimping step and the release step.