Percutaneous interventional cardiovascular medicine

Summary

This chapter describes the basic principles of the fully bioresorbable scaffold, the potential benefit of the technology and discusses the currently available preclinical and clinical data.

The fully bioresorbable coronary scaffolding device ensures temporary mechanical support to open a coronary stenosis and eventually disappears over time. Through liberating the coronary artery from the rigid cage imposed by conventional stent platforms, the vessel recovers pulsatility and becomes responsive to shear stress and to physiological cyclic strain.

The potential advantages of fully bioresorbable scaffolds (BRS) include a reduction in late/ very late stent thrombosis, return of the vessel vasomotion, adaptive shear stress, late luminal enlargement, and improvement in future treatment options such as surgical bypass grafting as well as enabling effective non-invasive imaging follow-up with multi-slice computed tomography or magnetic resonance imaging. To date, the Igaki-Tamai stent (PLLA), the absorbable metallic stent (Magnesium), the ABSORB bioresorbable everolimus-eluting PLLA scaffold, the REVA stent (tyrosine polycarbonate) and the IDEAL stent (adipic acid and salicylate) have been tested in first-in-man trials. These first-in-man trials demonstrate consistently the safety of these devices. With the exception of the ABSORB device, however, a high repeat revascularisation rate remains as a clinical issue. The second-generation BRS with drug elution (Magnesium, REVA and IDEAL) have now been developed and are currently being evaluated in clinical trials.

Introduction

The invention of balloon angioplasty as a percutaneous treatment for obstructive coronary disease by Andreas Gruntzig in 1977 was a huge leap forward in cardiovascular medicine, and undoubtedly will always be remembered as a revolution in the field of revascularization. However, this technique was plagued by multiple problems including the risk of acute vessel closure secondary to occlusive coronary dissection, sometimes necessitating emergency coronary artery bypass surgery.[1, 2, 3] Whilst late luminal enlargement and vascular remodelling could take place, more often restenosis[4, 5, 6, 7, 8] would instead occur. Essentially the restenosis would be caused by constrictive remodeling [9, 10, 11, 12, 13] and to a lesser extent by elastic recoil[14] or the neointimal hyperplastic healing response.[15, 16, 17]

The advent of bare metal stenting and the landmark BENESTENT and STRESS trials have established bare metal stenting as the second revolution in interventional cardiology.[18, 19] This technology provided a solution to acute vessel occlusion by sealing the dissection flaps and preventing recoil. The rate of subacute occlusion was reduced to 1.5%, making emergency bypass surgery a rare occurrence. Restenosis rates were further reduced from 32% to 22% at seven months, but this was still high, and neointimal hyperplasia inside the stent was even more prominent than with angioplasty, necessitating repeat treatment in numerous patients.[18]

Since the vessel was now caged with metal, late luminal enlargement and advantageous vascular remodelling could no longer occur. Another problem, namely that of late stent thrombosis, was also first described.[20] To solve the problem of in-stent restenosis, following the historical failure of brachytherapy to resolve this problem,[21, 22] drug eluting stents were introduced. The follow-up of the first 45 patients implanted with the sirolimus eluting Bx velocity stent (Cordis, Johnson & Johnson, Warren, NJ, USA) were found to have negligible neointimal hyperplasia.[23, 24] This was confirmed in the randomized RAVEL study.[25, 26] The introduction of drug eluting stents was thus dubbed the third revolution in interventional cardiology. Both large scale randomised trials and all-comer registries showed excellent results in terms of the need for repeat revascularisation. However, the early enthusiasm has been tempered in recent years following widespread concerns regarding the increased risk of late and very late stent thrombosis. [27, 28, 29, 30, 31] Registries of all comers treated with drug eluting stents showed late stent thrombosis rates of 0.53% per year, with a continued increase to 3% over four years.[32, 33] In the ARTS II trial which enrolled patients with complex multivessel disease, the rate of combined definite, probable and possible stent thrombosis was as high as 9.4% at five years, accounting for 32% of MACE events.[34] In addition, post-mortem pathological specimens of drug eluting stents revealed significant numbers of uncovered struts with evidence of a persistent inflammatory reaction around the stent struts.[35, 36, 37] Vasomotion testing demonstrated abnormal vasoconstriction responses to acetylcholine distal to the deployed stent, suggesting that the structure and function of the endothelium remained abnormal.[38, 39] All these problems promise to be solved with the advent of fully biodegradable scaffolds. This new technology, heralded as the fourth revolution in interventional cardiology, offers the possibility of transient scaffolding of the vessel to prevent acute vessel closure and recoil whilst also transiently eluting an antiproliferative drug to counteract the constrictive remodeling and excessive neointimal hyperplasia.

In the pilot ABSORB cohort A study, optical coherence tomography revealed that 100% of the scaffold struts to be fully tissue-covered and apposed.[40] After a period of two years, the stent struts were resorbed with complete integration of the scaffold into the vessel wall and infiltration by functional smooth muscle cells. In addition, it appeared that the vessel lumen enlarged and the plaque/media diminished. Vasomotion testing suggested that the endothelial structure and function were fully restored with a normal vasodilatory response to both acetylcholine and nitroglycerine within the previously scaffolded area of the vessel. Thus, this new era in interventional cardiology could be viewed as the era of Vascular Reparative Therapy (VRT), with fully bioresorbable devices ( Table 1).

FOCUS BOX 1Current concerns and possible solutions

Drug-eluting metallic stent platforms significantly reduced the amount of neointimal hyperplasia and associated clinical repeat revascularisation rate, but there are still concerns over late stent thrombosis, a persistent inflammatory reaction, incomplete endothelial coverage and impaired vasomotion distal to the stented segment, presumably due to residual drug, durable polymers and permanent metallic cage
Fully resorbable scaffolds have a potential to solve these drawbacks of drug-eluting stents

Fully bioresorbable scaffolds: the fourth revolution in interventional cardiology?

Bioresorbable scaffolds have obvious potential advantages over current metallic drug-eluting stent technology, these include:

A reduction in adverse events such as stent thrombosis (ST). As drug elution and scaffolding are temporary and are only provided by the device until the vessel has healed, no foreign material, such as non-endothelialised struts and drug polymers (potential triggers for ST) can persist long-term.
The removal, through bioabsorption, of the stented vessel’s rigid caging. This can facilitate the return of the vessel’s vasomotion, adaptive shear stress, late luminal enlargement and late expansive remodeling. Furthermore, this might also reduce the problems of jailing the ostium of side branches as seen with permanent metallic stent struts.
A reduction in bleeding complications. Once bioabsorption of the temporary scaffold has been completed, there will potentially be no requirement for long-term dual anti-platelet therapy. This is particularly pertinent given that the elderly, who are at the greatest risk of bleeding, are increasingly receiving invasive treatment for ischaemic heart disease.[41] Furthermore, early discontinuation of dual antiplatelet therapy with current metallic DES, for whatever indication, has consistently been shown to be an independent predictor of ST.[42]
An improvement in future treatment options. The treatment of complex multi-vessel disease frequently results in the use of multiple long DES; for example in the SYNTAX trial the average number of stents was four, and one third of patients had greater than 100 mm of stent implanted.[43] In such cases, repeat revascularization – either by means of percutaneous or surgical revascularisation, is potentially challenging because of the metallic cages formed by previously implanted DES. The use of a BRS would mean that there would potentially be no restriction on any future percutaneous or surgical revascularisation should they be needed.
Allowing the use of non-invasive imaging techniques such as computed tomography (CT) angiography or magnetic resonance imaging (MRI) for follow-up. Presently, metallic stents can cause a blooming effect with these imaging modalities making interpretation more difficult.[44] The PLLA scaffold should not restrict the use of CT or MRI as it is non metallic; once bioabsorption has been completed with a metallic BRS then this should also not restrict the use of CT or MRI. Non-invasive imaging follow-up could therefore become an alternative to invasive imaging follow-up.
Reservoir for the local delivery of drugs and genes. Since the duration of bioresorption is modifiable, according to the type of polymers/co-polymers, a tuned elution of multiple drugs can potentially be achievable (e.g. early elution of antiproliferative agent from a coated polymer and chronic elution of anti-inflammatory or other agent from the backbone polymer).
The elimination of the concern that some patients have at the thought of having an implant in their bodies for the rest of their lives.[45]

FOCUS BOX 2Potential advantages of fully bioresorbable scaffolds includes:

A reduction in late/very late stent thrombosis
A return of the vessel’s vasomotion, adaptive shear stress and late luminal enlargement
A reduction in bleeding
An improvement in future treatment options such as bypass grafting
Diagnostic noninvasive imaging follow-up with MSCT or MRI

BIODEGRADATION, BIOABSORPTION AND BIORESORPTION

The word «Stent», now used in many medical disciplines, derives from the name of an English dentist born in 1807, Charles Thomas Stent. Jacques Puel and Ulrich Sigwart are to be credited for introducing and popularizing this term in percutaneous coronary intervention, substituting previously used terms such as intravascular prostheses or grafts. In their landmark 1987 work[2], the researchers described the first use of intracoronary stents in humans, which at that time were called "Wallstents", after its inventor Hans Wallstén. Scaffold is a newer term that implies temporary arterial support and was introduced in the interventional cardiology field to distinguish these devices from permanent stents. It has been widely used in peer-reviewed medical literature over the last years and has become the preferred term to designate this bioresorbable coronary device.

After the first commercially medical devices made of artificial polymers degradable in vivo (surgical sutures made from lactic acid and/or glycolic acid) were referred to as “absorbable sutures” [46], such devices were frequently dubbed as “bioabsorbable”, which in general reflects the disappearance of a compound into another substance. However, the term “bioabsorbable” is not appropriate, because “bioabsorption” does not necessarily mean “degradation”, and even less, “elimination of the polymer from the body”. Indeed, even if a “bioabsorbable” polymeric device was no longer visible as a result of degradation (“bioabsorption”), high molecular mass molecules can be still trapped between skin and mucosa without passing physiological barriers. Therefore, bioabsorption does not necessarily mean complete cleavage of macromolecules up to small molecules that can be eliminated from the body through natural pathways, namely kidney or lungs. To indicate the total elimination of polymers by excretion and assimilation, the concept of “bioresorption” was introduced. [47, 48] Bioresorption indicates the total elimination of polymers from the body via natural routes (like kidney or lungs) by dissolution, assimilation, and excretion [49]. The words “degradation” and “bio-degradation” are also confusing and should be restricted to cases of unknown or abiotic mechanisms (“degradation”) or cell-mediated in vivo mechanisms (“bio-degradation”).

FOCUS BOX 3The terms in question

“Bioabsorption” stands for the disappearance of a compound into another substance, but does not necessarily mean “biodegradation” or “elimination of the polymer from the body”
“Bioresorption” stands for total elimination of polymers from the body by excretion and assimilation after complete cleavage of macromolecules up to small molecules

Historic Development of BRS

The efforts to create BRS started approximately 30 years ago. The first experimental studies using a non-biodegradable polyethylene-terephthalate braided mesh stents in porcine animal models were published by our group in 1992.[50] In 1996, our group in collaboration with the Mayo and Cleveland Clinics[51] reported in porcine coronary arteries negative consequences after implantation of the Wiktor stent coated with five different types of biodegradable polymers (Polyglycolic acid/ polylactic acid copolymer: PGLA, polycaprolactone: PCL, Polyhydroxy-butyrate/-valerate copolymer: PHBV, Polyorthoester: POE and Polyethyleneoxide/polybutylene terephthalate: PEO/PBTP), all resulting in marked inflammation leading to neointimal hyperplasia and/or thrombus formation. Subsequently, Lincoff et al.[52] demonstrated that in a porcine model, a stent coated with high molecular weight poly-L lactic acid (PLLA 321kD) was well tolerated and effective, whilst a stent coated with low molecular weight poly-L-lactic acid (PLLA, approximately 80kD) was associated with an intense inflammatory neointimal response. The author also proved the feasibility of drug-elution from the PLLA, although no suppression of neointimal hyperplasia was reported. In 1998, Yamakawa et al.[53] reported that in the porcine model the fully biodegradable PLLA stent with tyrosine kinase inhibitor efficiently suppressed proliferative stimuli caused by balloon injury. These pioneering experiments with high-molecular-weight PLLA further supported the investigations in human. However, despite the fact that the idea of a BRS was present since the early days of stent development, the technology failed to develop at that stage for the lack of an ideal polymer and the advent of metallic DES[49] and matured long after DES had become the golden standard in percutaneous revascularization. In 2000, the Igaki-Tamai PLLA-based scaffold became the first BRS to be implanted in humans and aliphatic polyesters have dominated the field as scaffold materials ever since, despite the compromises that had to be made in various mechanical performance properties. The first investigations of magnesium alloy-based scaffolds for cardiovascular interventions started in 2003 and the first clinical trial of magnesium-based BRS for intracoronary use was published in 2007[54].

The current BRSs are composed of either a polymer or bioresorbable metal alloy. Numerous different polymers are available, each with different chemical compositions, mechanical properties and subsequent bioabsorption times ( Table 2 and Table 3). The most frequently used polymer in the current generation of BRS is poly-l-lactic acid (PLLA). This is already in widespread clinical use with applications such as resorbable sutures, soft tissue implants, orthopedic implants and dialysis media. The key mechanical traits for candidate material in coronary indications include high elastic moduli to impart radial stiffness, large break strains to impart the ability to withstand deformations from the crimped to expanded states and low yield strains to reduce the amount of recoil and overinflation necessary to achieve a target deployment. Stent developers look to increase stent strut dimensions to compensate for mechanical shortcomings of bioresorbable materials. As the thickness of these struts increases, strain levels imposed on a material increases proportionally.

Polymer

The term “polymer” originates from the Greek words «πολύ» (many or much) = poly and «μέρος» (part) = mer and refers to a macro-molecule that is composed of multiple repeated sub-units. Polymers have specific characteristics that define their mechanical properties which in turn define their suitability to be used as materials for scaffold development. The key polymer characteristics, beyond their type, include molecular weight, crystallinity and hydrophobicity, while the key mechanical properties are tensile strength, elasticity, phase behaviour and biodegradation rate. The weight-averaged molecular weight of the polymer affects its processability while the number-averaged molecular weight influences its mechanical strength and elasticity[55, 56]. Crystallinity is determined by the degree of monomers’ linear arrangement and affects both tensile strength and degradation rate; in general, higher crystallinity results in longer degradation time and higher tensile strength[57]. Hydrophobicity, from the Greek word «ύδωρ»=hydro (water) and φόβος»=phobia, is the property of repelling water rather than absorbing it or dissolving in it and is also a regulating factor of the degradation rate. Tensile strength quantifies the amount of stress that a material can endure before suffering permanent deformation. Good radial strength allows for thinner struts and lower profile, hence better deliverability. Elasticity, from the Greek word «ἐλαστός»=ductible, is defined as the ability of a body to resist a distorting/deforming influence or force and return to its original size and shape when the latter is removed. For polymers, their elastic properties are quantified by the Young’s or tensile modulus of elasticity which is highly dependent on temperature. In the case of PLLA, increasing crystallinity of the polymer is expected to increase its strength but at the expense of its elasticity[58], limiting the amount of expansion that a polymer scaffold can endure during deployment without fracturing. The phase behaviour of polymers is defined by their glass transition temperature (Tg) (=the temperature that the polymer starts to display rubbery behaviour) and melting point temperature (Tm).

FOCUS BOX 4Bioresorbable scaffold developments

Attempts to develop BRS have been made for 20 years. Despite the favourable results of the early stents, the technology failed to develop due to the inability to manufacture an ideal polymer and growing interest for the metallic DES
The current BRSs are composed of either a polymer or a bioresorbable metal alloy

BIORESORPTION PROCESS OF POLY-L-LACTIC ACID

In the PLA family of polymers, molecular weight degradation occurs in vivo predominantly through hydrolysis, which is a bimolecular nucleophilic substitution reaction that can be catalysed by the presence of either acids or bases. The schematic shown in Figure 1A describes the hydrolysis reaction in which water catalyses a chain scission event at an ester bond.

Poly(L-lactide) is a semicrystalline polymer, meaning that it is typically comprised of a mixture of ordered, densely packed crystalline phase and disordered, less dense amorphous phase ( Figure 1B). Individual polymer chains frequently span both phases; consequently, domains of crystalline polymer are linked to one another by amorphous tie chains ( Figure 1B). Since the amorphous regions are less dense, water penetrates it more readily than it does the crystalline regions.

For objects whose smallest dimension is O (100 µm), hydrolysis byproducts cannot readily diffuse out of the object and bulk degradation controls the process. The object degrades more or less from the inside out. In this case, the third order kinetics theory of Pitt et al. predicts that the hydrolysis rate depends upon the concentration of ester bonds, water, and carboxylic acid end groups, the latter of which are generated by each hydrolysis reaction.[59, 60, 61] This so-called autocatalytic model for aliphatic polyesters like PLA is based upon the third-order rate equation given by:

(1)

where [E], [COOH], and [H2O] represent the concentrations of ester bonds, carboxylic acid end groups, and water, respectively, and k is the hydrolytic degradation rate constant. Assuming that the concentrations of ester bonds and water are approximately constant throughout the degradation process and the concentration of carboxylic acid end groups is inversely proportional to the number-average molecular weight (Mn) of the polymer (i.e., [COOH] = 1/Mn), Weir and co-workers showed that: [62, 63]

(2)

where Mn(t) is the number-average molecular weight after degradation time t, and Mn(0) is the number-average molecular weight at time t=0 (prior to degradation). The assumptions inherent to the model are reasonable if mass loss does not occur, since mass loss would affect the concentrations of water and carboxylic end groups in the sample. Eq. (2) may be rewritten as:

(3)

By representing data for Mn(t)/Mn(0) versus t on a log-linear plot, one may infer the hydrolytic degradation rate from the slopes of the line connecting the points ( Figure 2A).

FOCUS BOX 5Bioresorption process of poly-l-lactic acid

Poly-lactide polymers are in vivo degraded through hydrolysis
After hydration of polymer, hydrolysis starts at amorphous chain. Subsequently mechanical support subsides; oligomers diffuse out into the tissue. Finally lactates are converted into CO2 and water in Krebs cycle, and is excreted from the lung and kidney

From a chemical standpoint, it is considered that resorbable implant undergo five stages which are not discrete and can overlap. [64, 65] The first stage is hydration of the polymer. After implantation of polymeric resorbable device, water in the environment diffuses into the material. In general, PLA is relatively hydrophobic and increasingly so as the molecular weight of the polymer chain increases. The chain ends, however, are hydrophilic due to the carboxylic acid end group. Importantly, chain ends cannot participate in a crystal lattice; therefore, the hydrophilic chain ends are relegated to the amorphous phase, rendering it more hydrophilic than the bulk. The second stage is characterized by random chain scission by hydrolysis ( Figure 1A), which leads to a reduction in the molecular weight. In the third stage, as hydrolysis continues, the strength of the material erodes due to the cleaving of amorphous tie chains linking the crystalline regions ( Figure 2B). At this time, structural discontinuities are normal and entirely expected, since the device is designed to degrade and be physiologically processed over time. In the fourth stage, polymer chains have been hydrolyzed to sufficiently short lengths that they are increasingly hydrophilic (and hence, soluble in an aqueous environment) and able to diffuse out of the device (loss of mass, Figure 2C). The fourth phase is assimilation or dissolution of the monomer. Phagocytes can assimilate small particles and lead to soluble monomeric anions. Lastly, in the fifth stage, the soluble oligomeric PLA molecules that have diffused out of the device ultimately hydrolyze to the lactic acid monomer, which deprotonates readily to lactate. Lactate is in turn converted to pyruvate, which eventually enters the Krebs Cycle and is further converted into carbon dioxide and water. These final products are excreted from the body through kidney or lung, which results in complete bioresorption of the implant. [66]

Due to the described properties of semi-crystalline polymers, they are used predominantly for the purposes of mechanical support (i.e. the scaffold), whereas amorphous polymers allow a more uniform dispersion of the drug and are therefore preferred for usage in controlled drug release systems (e.g. coating of the BVS system).

BIORESORPTION IN A PORCINE MODEL

Recently in a porcine coronary model, long-term histological findings after implantation of everolimus-eluting PLLA BRS have been reported.[67] Thirty-five polymeric everolimus-eluting PLLA BRS (BVS (3.0 x 12.0 mm)) were singly implanted in the main coronary arteries of 17 pigs that underwent OCT and were then euthanized either immediately (n=2), at 28 days (n=2), at 2 years (n=3), at 3 years (n=5) or at 4 years (n=5) after implantation. All BVS implanted arteries in these animals were evaluated by histology except for 5 arteries examined at 2 years with gel permeation chromatography (GPC) to assess the biodegradation of the PLLA. Fourteen arteries with BVS from an additional 6 pigs were examined by GPC at 1 (n=1), 1.5 (n=2), and 3 (n=3) years. Corresponding OCT and histology images were selected using the distal and proximal radio-opaque markers as landmarks.

OCT at 28 days showed sharply defined, bright reflection borders, best described as a box-shaped appearance, in 82% of the struts. Histologically, all struts appeared intact with no evidence of resorption ( Figure 3A-C). OCT at 2 years revealed 60 ± 20 struts to be discernible in each BVS with 80.4% strut sites showing a box-shaped appearance. Despite their similar appearances on OCT, histological analysis revealed that these structures appeared to be composed of proteoglycan, with polylactide residue being at such low level as to be no longer quantifiable by chromatography ( Figure 3D-F). OCT at 3 years demonstrated that the recognizable struts decreased to 28±9 struts per BVS. Histology shows that connective tissue cells within a proteoglycan-rich matrix replaced the areas previously occupied by the polymeric struts and coalesced into the arterial wall ( Figure 3G-I). OCT at 4 years showed only 10±6 struts per BVS were still recognizable. Histological analysis demonstrated that these struts were minimally discernible as foci of low cellular density connective tissue ( Figure 3J-L). In conclusion, still discernible struts by OCT at 2 years were compatible with complete bioresorption of the polylactic struts as demonstrated by histological and GPC analyses, whereas at three and four years, both OCT and histology confirmed complete bioresorption of the struts into the arterial wall.

FOCUS BOX 6PLLA devices in a porcine model over time in the porcine model, a PLLA device (ABSORB) degrades over 2 years. At 2 years, Alcian-Blue positive molecules replace the struts, although the struts are still visible with OCT. At 3 and 4 years, the struts become less visible on OCT, while histologically the struts are assimilated with the surrounding tissues and are hardly recognisable

Clinically tested BRS

A number of manufacturing companies have developed scaffolds that after favorable outcomes in preclinical testing have transitioned to clinical studies ( Table 4). The front-runner in this process is the Absorb scaffold, which is presently the only BRS with randomized data for clinical outcome versus a new generation DES and has an estimated number of devices implanted in the order of 200,000[68]. However, following the negative results of the clinical trials (discussed later), the manufacturer decided on a worldwide halt to sales of the Absorb scaffold as of September 14, 2017.

IGAKI-TAMAI STENT

The Igaki-Tamai PLLA coronary stent was the first fully bioresorbable stent to be implanted in humans with complete degradation taking 18-24 months. The stent had a helical zigzag design, which differed from previous knitted patterns ( Figure 4). This resulted in less vessel wall injury during implantation and therefore less initial thrombus formation and reduced intimal hyperplasia.[69] The stent was mounted on a standard angioplasty balloon and was both thermal self-expanding and balloon expandable. Self-expansion occurred in response to heating the PLLA, which was achieved by using heated contrast (up to 70 °C) to inflate the delivery balloon. Stent expansion was further optimised by inflating the delivery balloon to 6-14 atm for 30 seconds, and the stent’s nominal size was ultimately achieved by continued self-expansion at 37°C in the 20-30 minutes after stent deployment. The stent had a standard length of 12 mm, and was available in diameters of 3 mm, 3.5 mm and 4 mm; the stent strut thickness was 0.17 mm. An 8F guiding catheter was required because the stent was initially constrained by a sheath that was removed once it was across the lesion. At either end of the stent to aid visualisation were two radio-opaque cylindrical gold markers (0.6 mm high by 0.18 mm in diameter) ( Figure 4). The first in man (FIM) study of the Igaki-Tamai stent (15 patients, 19 lesions, 25 stents), demonstrated no major adverse cardiovascular events (MACE) or ST within 30 days, and one repeat PCI at 6 months follow-up. Encouragingly, the loss index (late loss/acute gain) was 0.48mm, which was comparable to BMS, and demonstrated for the first time that BRS did not induce an excess of intimal hyperplasia. Furthermore, intravascular ultrasound (IVUS) imaging demonstrated no significant stent recoil at day one, and continued stent expansion was observed in the first three months of follow-up. The mean stent cross sectional area increased from 7.42±1.51 mm² at baseline to 8.18±2.42 mm² (P=0.086) at 3 months, and 8.13±2.52 mm² at 6 months. [69] A second larger study of 50 elective patients (63 lesions, 84 stents) also showed promising results. IVUS performed at 3 years follow-up demonstrated the complete absence of stent struts, whilst angiographic analysis demonstrated a mean diameter stenosis of 25%, compared to 38%, 29% and 26% at 6, 12, and 24 months respectively. Clinical outcomes at 4-year follow-up showed rates of overall and MACE-free survival rates of 97.7% and 82.0%, respectively. [70]

At 10-year clinical follow-up, freedom from cardiac death, non-cardiac death and MACE were 98%, 87%, and 48% respectively.[71] In the limited cases with serial angiographic follow-up, the MLD was stable: the mean MLD was 2.01 mm at 1 year and 2.06 mm at 10 years. There were two ST events: 1 subacute event occurring at day 5 possibly due to inadequate heparinisation at the time of PCI, and one very late ST event occurring in the sirolimus-eluting metallic stent which was later implanted proximal to the previously placed Igaki-Tamai stent. Angiographic and OCT images of the stent at 10-year follow-up in one anecdotal case are shown in Figure 4.

Despite these impressive results, the failure of the stent to progress was primarily related to the use of heat to induce self-expansion. There were concerns that this could cause necrosis of the arterial wall leading to excessive intimal hyperplasia or increased platelet adhesion leading to ST. [72] None of these concerns were substantiated in the initial studies, however only low-risk patients were enrolled. After completion of the PERSEUS study,[73] the stent became available in Europe for peripheral use, however there are plans to review its use in coronary arteries. At present the stent has no drug elution, although preclinical studies of the polymeric stent eluting the tyrosine kinase antagonist ST 638 showed promising results.[53]

MAGNESIUM ALLOY

Magnesium (Mg) is the fourth commonest cation within the human body; the total body content is ~20g, with 350mg required daily. It is essential for the synthesis of over 300 enzymes and is a co-factor for ATPase. A high dose infusion of Mg can cause vasodilatation; the promotion and recruitment of collaterals during ischaemia; and can function as a direct inhibitor of stent thrombosis.

The absorbable metallic stent (AMS-1) (Biotronik, Berlin, Germany) is the first metallic biodegradable stent, composed of 93% magnesium (approximate weight of 3.0x10 mm is 3 mg) and 7% rare earth metals ( Figure 5A-B). The first generation AMS-1 stent, which is balloon-expandable, is available in diameters of 3.0-3.5 and lengths of 10-15 mm. It has a high mechanical strength; and has comparable properties to stainless steel stents in view of its low elastic recoil (<8%), high collapse pressure (0.8 bar) and minimal shortening after inflation (<5%).[74] The degradation of Mg produces an electronegative charge that results in the stent being hypo-thrombogenic.[75] In the porcine model, the AMS-1 has been shown to be rapidly endothelialised, and within 60 days is largely degraded into inorganic salts, with little associated inflammatory response.[76] After promising initial preclinical trials, and successful deployment in 20 patients with critical limb ischaemia,[77] the PROGRESS AMS trial was performed. This was a multicenter, non-randomized, prospective study, which assessed the efficacy and safety of the stent in 63 patients (71 stents) with single de novo lesions.

The study reached the primary endpoint, a composite of cardiac death, non-fatal MI, and clinically driven TLR (=MACE) at 4 months, by achieving a rate of MACE of 23.8%; the rate of MACE at 12 months was 26.7%. The study demonstrated the safety of the AMS-1 with no reported death, MI or ST during 12 months follow-up; in addition, there was a return of vessel vaso-reactivity. Unfortunately, the rate of any TLR was disappointing: 23.8% at 4 months, and 45% at 12 months. [74] QCA and IVUS have both provided important information regarding the mechanism of this restenosis, all of which have had important implications on future stent designs. QCA showed an acute gain post stenting of 1.41 mm (SD 0.46 mm), and a reduction in the mean diameter stenosis from 61.5% pre-stenting (SD 13.1%), to 12.7% (SD 5.6%) post-stenting. At 4 months follow-up the mean diameter stenosis was 48.4% (SD 17.0%) and the in-stent late loss was 1.08 mm (SD 0.49 mm). Immediately after post-stenting balloon dilatation, which was required in 42 patients, IVUS demonstrated that the mean stent cross sectional area (6.2±1.5 mm²), and mean stent volume (116.5±40.2 mm³) were both lower than that seen with standard metallic stents deployed in similar sized vessels. These results were consistent with QCA results, and were probably related to both the lower radial force of the Mg alloy compared to stainless steel, and by immediate vessel recoil after stent implantation. At four months follow-up IVUS demonstrated that only small remnants of the original struts were visible, which were all well embedded into the intima. IVUS also showed a 42% reduction in the area delineated by the external elastic membrane, suggesting that early vessel recoil was indeed the primary cause of the high late loss, and restenosis. This vessel recoil was attributable to the loss of radial force from the early, rapid AMS-1 stent degradation, such that no stent support was available to oppose constrictive remodeling; a natural response of the vessel to injury. Exacerbating the problem was evidence to suggest that stent degradation was possibly faster than previously anticipated. A repeat IVUS in one patient only three weeks post AMS-1 implantation showed 50% of struts were already degraded.[78] Other factors besides constrictive recoil contributing to the luminal loss seen at follow-up were intra-stent tissue growth (intra-stent neointima) (41%) and tissue growth behind the stent struts (extra-stent neointima) (13.5%) ( Figure 5E). Reassuringly long-term data from angiographic and IVUS performed in the eight patients who did not experience an event at 4 months have shown that there was no evidence of either later recoil or late development of neointima. In fact, in some patients evidence was seen of neointimal regression, and/or an increase in vessel and lumen volume.[78] Of note, the AMS stent is proven to be MRI compatible[79] and furthermore, the vasodilator function after nitroglycerin injection was restored in the treated segment at follow-up.[80]

The results from the patients enrolled in studies of the AMS-1 stent demonstrated it was safe for use in both coronary and peripheral vessels.[74, 77, 81] The stent was resorbed as intended, with no undue safety concerns. The disappointingly high TLR rate compared to standard BMS and DES have lead to modifications in the future stent’s design. These aim to prolong degradation and enable drug elution, and thereby reduce restenosis that was partly due to negative remodeling, and partly due to an excessive healing response. Two new stents have been developed: AMS- 2 and AMS-3 ( Figure 5C-D). The AMS-2 stent is designed to overcome some of the problems of excessive vessel recoil seen with AMS-1. It provides prolonged mechanical stability, which has been achieved by using a different magnesium alloy, which has not only a higher collapse pressure of 1.5bar, compared to 0.8bar with AMS-1, but also a slower degradation time. In addition, the stent surface has been modified; the stent strut thickness has been reduced from 165μm to 125 μm and the shape of the strut in crosssection has been altered from rectangular to square (improving radial strength). These changes have in animal models prolonged scaffolding and stent integrity, improved radial strength and reduced neointima proliferation.

The AMS-3 stent (DREAMS 1), used a refined, WE43 alloy (93% magnesium and 7% rare earth elements) with slower bioresorption time (9-12 months), higher radial strength, 125 microns struts with rectangular shape and paclitaxel elution for 3 months. It was tested in BIOSOLVE I study, which has shown 100% procedural success with device. The device has shown good safety profile with no death or stent thrombosis at 12 months and only one MI [82]. Regarding efficacy end points, the new design has also shown considerable improvements[82] with angiographic in-stent LLL of 0.52±0.49mm at 1-year [83], while at 3 years TLR rates reached 6.6%, with no cases of CD, TV-MI or ScT[83, 84].

A newer iteration (2nd generation DREAMS, marketed as Magmaris) made of the same alloy and design but with a strut thickness of 150 µm has been recently developed. This device elutes sirolimus instead of paclitaxel, has tantalum radiopaque markers at both ends and a modified, electropolished strut cross-sectional profile. These modifications result not only in slower dismantling and resorption rate but also improved visibility, higher bending flexibility, increased deployment diameter and higher acute radial force and fracture resistance compared to the previous generations. Magmaris was evaluated in the international multicenter FIM BIOSOLVE II trial (n=123). In-scaffold LLL was 0.27±0.37mm at 6 months (primary end-point) and 0.39±0.27 mm at 12-months follow-up. TLF -a composite of CD, TV-MI, clinically indicated TLR (CI-TLR) and CABG- occurred in 4 patients (3%), consisting of 1 CD, 1 TV-MI, and 2 CI-TLR[85]. Long-term results of the Magmaris BRS have been recently published in 184 patients, including pooled follow-up data at 24 months from BIOSOLVE II and at 6 months from the pilot BIOSOLVE III trial (n=61), demonstrating TLF rates of 5.6% at 2 years[86]. Most importantly, up to now no cases of definite or probable ScT have been reported in patients treated with any of the 3 iterations of this magnesium-based BRS.

TYROSINE POLYCARBONATE: THE REVA STENT

The Reva Medical (San Diego, CA) BRS is a tyrosine poly (desaminotyrosyltyrosine ethyl ester) carbonate stent, which degrades, as summarized in Figure 6A, into water, carbon dioxide, and ethanol; in addition, tyrosine is metabo- lized by the Krebs cycle. The first two iterations of the first-generation of the Reva Medical BRS ( Figure 6B, C), showed high TLR and low technical success rates in the RESORB FIM trial (bare form)[87] and the RESTORE I trial (sirolimus-eluting version, ReZolve BRS) respectively[87]. A third iteration, ReZolve BRS 2 ( Figure 6D), is being tested in the RESTORE II trial. Preliminary results in 67 patients at 6 months showed 3 MACE cases and the trial is on-going[88]. However, the problematic deliverability of the device pushed the company to feature a conventional balloon-expandable system for its next and current iteration of the scaffold, named Fantom ( Figure 6E). The most characteristic feature of the Fantom BRS is that it contains iodine covalently bound to its desamino-tyrosine polycarbonate backbone, which makes it intrinsically radio-opaque and may decrease the need for intravascular imaging during follow-up. Furthermore, it has a wide expandability range while the time required for full reabsorption is significantly longer compared to PLLA based scaffolds (3 years), but more than 80% molecular weight loss takes place within the first year. Clinically, the Fantom BRS was initially tested in a small FIM trial[89] and subsequently in the FANTOM II trial which included 240 patients split in two cohorts. Acute technical and procedural success was observed in 95.8% and 99.1% respectively of the cohort A patients. The 6 months angiographic and clinical primary endpoints for cohort A have been formally published [90]. Recently, 12 months results for the entire cohort showing a MACE rate of 4.2% with a single ScT (subacute)[91] and 24 months results for 125 patients showing a MACE rate of 5.6% and a single very late ScT were reported[92]. In addition to the latest clinical follow up, a 25-patients’ subset in the trial underwent angiographic imaging, showing a final in-scaffold late loss of 0.25 mm, which is in the desired range of 0.2 mm to 0.4 mm. CE Mark approval for the Fantom scaffold was received in April 2017, while cohort C of the FANTOM II trial is currently enrolling patients with more complex lesions. Reva has also announced a new iteration, named Fantom Encore, with a strut thickness of 95 microns in 2.5 mm diameter; its performance results are awaited.

POLY (ANHYDRIDE ESTER) SALICYLIC ACID: THE IDEAL™ STENT

The IDEAL^TM biodegradable stent (Bioabsorbable Therapeutics, Inc. Menlo Park, CA, USA) consists of a backbone of poly-anhydride ester based on salicylic acid and adipic acid anhydride, and an 8.3μg/mm coating of sirolimus ( Figure 7), potentially giving the stent both anti-inflammatory and anti-proliferative properties. The vascular compatibility and efficacy of this biodegradable salicylate-based polymer has previously been demonstrated in the porcine model. Most notably the polymer was associated with reduced inflammation when compared to a standard BMS and Cypher stent. [93] This was very likely to be due to the anti-inflammatory properties of salicylic acid following absorption by the vessel wall after its release. Drug elution was found to be complete after 30 days, whilst complete stent degradation occurred over a 9-12 month period. The 8F compatible, balloon expandable stent is radio-opaque, and does not require any special storage. Its radial strength is greater than a BMS, but considerably less than the Cypher stent. Histological analyses from pre-clinical studies of the IDEAL stent in pigs have confirmed the absence of excessive thrombosis, or inflammatory reaction, and satisfactory healing of the vessel. Furthermore, the promising mechanical properties of the stent were confirmed with well apposed stent struts observed at follow-up. [94] In July 2009, the 11 patients enrolled in the multi-centre FIM Whisper study completed their 12 month follow-up. Primary results have shown stent safety, and confirmed structural integrity of the stent with no evidence of acute or chronic recoil. Unfortunately, insufficient neo-intimal suppression has been demonstrated.[94] This is likely to be the consequence of inadequate drug dosing, particularly when considering that the surface area dose of sirolimus is only a quarter of that found on the Cypher stent. The rapid elution of sirolimus may also be a contributing factor.

A second-generation stent has been developed, with a higher dose of sirolimus, and a slower drug release pattern. Furthermore, the stent design has been optimized; which has resulted in a reduced crossing profile (6F compatible), and thinner struts (175μm). Although the program was on hold in early 2009, the program was resumed and the new device is currently undergoing preclinical evaluation [95].

Novolimus-eluting PLLA scaffold: The DESolve scaffold

The DESolve family of PLLA-based BRS (Elixir Medical, Sunnyvale, CA, US) includes the 1st generation device with 150 μm strut thickness and the 2nd generation device which has a strut thickness of 120 μm and elutes novolimus, an active metabolite of sirolimus. In both iterations, the scaffold is coated with a matrix of polylactide-based biodegradable polymer and antiproliferative drug, which is applied using a proprietary technique[96]. The important features of the DESolve BRS are intrinsic self-correcting deployment properties that become operative in the event of minor strut malapposition and ability to over-expand across a wide range of diameters without risk of strut fracture[97]. The first iteration was initially tested as a myolimus eluting device, in the small (n=16), FIM DESolve I trial, which showed encouraging imaging and clinical results at 6 and 12 months respectively[98]. Subsequently, the novolimus eluting version was tested in the larger (n=126), single arm DESolve NX trial, in which the primary end point of in-stent LLL at 6 months was 0.21±0.32 mm (113 patients with paired analysis) with 98.7% struts coverage at the same time point. At 5 years, the rates of MACE, TV-MI, CD and clinically indicated TLR were 9%, 1.6%, 3.3% and 4.1%, respectively. No cases of definite ScT were observed [99]. A single, retrospective, comparative analysis between this device and the Absorb BRS using propensity-score matching is also available, showing similar outcomes between the two devices with regards to 1-year rates of TLF, TLR, CD and definite ScT[100]. The second iteration was tested in the relatively small (n=50), single arm DESolveCx trial, showing in-scaffold LLL of 0.19±0.25 mm at 6 months and no MACE at 12 months’ follow-up[101]. Both iterations have CE mark while the company is currently developing a 3^rd generation device (Desolve NXT plus) with 120 strut thickness and a contoured strut design for enhanced acute performance.

Arterial remodelling technologies (ART) PLLA stent

The ART stent (Noisy le Roi, France) is a poly-lactic acid (PLA) BRS which has a unique combination of L and D isomers resulting in a highly biocompatible and haemocompatible stent. It is produced using molecular weight preserving technology, which preserves its mechanical properties and avoids premature degradation. The balloon expandable 6F compatible ART stent provides suitable mechanical strength for 5- 7 months to resist vessel recoil, and complete monomer resorption occurs within 18 months. In preclinical studies positive remodelling (vessel enlargement) has been demonstrated to occur between 3 and 6 months. The device has no antiproliferative drug and the developers believe that it is not necessary as the positive remodelling will accommodate intervention-induced intimal hyperplasia and in addition intimal hyperplasia reaches a peak at about 6 months then regresses [102]. Absence of antiproliferative coating may also permit more rapid return of normal endothelial coverage, function and maintain an efficient endothelial barrier limiting neoatherosclerosis [103, 104]. The first-in-man trial, ART-DIVA is undergoing to evaluate the safety of the ART scaffold for the treatment of patients with single de novo lesion of a native coronary artery with mandatory balloon predilatation. The scaffold received CE Mark in May 2015, but was never marketed in Europe and in February 2017 Terumo announced the termination of the product’s co-development with ART.

FOCUS BOX 7First-in-man trials

Since 1999, Igaki-Tamai stent (PLLA), absorbable metallic stent (Magnesium), REVAstent (tyrosine polycarbonate) and IDEAL stent (adipic acid and salicylate) have been tested in first-in-man trials
While these first-in-man trials consistently showed the safety of these devices, high revascularisation rates remained a clinical issue. The second-generation BRSs with drug elution (Magnesium, REVA and IDEAL) have been developed and are being/will be tested in clinical trials

EVEROLIMUS-ELUTING PLLA scaffold: the ABSORB BVS scaffold

The BVS stent design is characterized by a crossing profile of 1.4 mm with circumferential hoops of PLLA. The struts are 150 microns thick and are either directly joined or linked by straight bridges ( Figure 8). Both ends of the stent have two adjacent radio-opaque platinum markers. The radial strength, measured in a water bath at 37°C using IVUS and by flat plate compression of 10, 15 and 25 % is 0.048±0.007 N/mm2, 0.070±0.008 N/mm2 and 0.106±0.009 N/mm2, while comparative values for a contemporary bare-metal stent (Vision coronary stent, Abbott Vascular, Santa Clara, CA, USA) is 0.073±0.011 N/mm2, 0.114±0.012 N/mm2 and 0.155±0.012 N/mm2, respectively.[105]

The backbone of BVS device is made of semicrystalline polymer called Poly-L-lactic acid.[105] The coating consists of poly D,L-lactide, which is a random copolymer of D- and L-lactic acid with lower crystallinity than the BVS backbone. The coating contains and controls the release of the anti-proliferative drug, everolimus. Both PLLA and PDLLA are fully bioresorbable. During bioresorption, the long chains of PLLA and PDLLA are progressively shortened as ester bonds between lactide repeat units are hydrolyzed, and small particles less than 2 µm in diameter are phagocytosed by macrophages. Ultimately, PLLA and PDLLA degrade to lactic acid, which is metabolized via the Krebs cycle. In a porcine coronary artery model, mass decreased with time; 30% at 12 months increasing to 60% at 18 months and to 100% at 24 months post implantation.

ABSORB COHORT A AND B TRIALs

The ABSORB BRS has a PLLA-based backbone with a PDLLA coating. The later contains everolimus with a coating to drug ratio of 1:1 and controls its release in a purely diffusion-related fashion[96]. There have been two iterations of this device. The first-generation, 1.0, was tested in the FIM ABSORB A trial (n = 30), demonstrating very late lumen enlargement (from 6 months to 2 years), restoration of vasomotion and endothelial function at 2 years[40, 106], with a MACE rate of 3.4% and no ScT at 5 years[107]. At 6-month follow-up, the angiographic in-stent late lumen loss (LLL) was 0.44 mm, attributed mainly to a mild reduction of the stent area (-11.8%) as measured by IVUS (chronic recoil). The neointimal area was small (0.30 mm²), with a minimal area obstruction of 5.5%, demonstrating effective suppression of restenosis by everolimus[105]. In contrast to the radio-opaque metallic stents that hinder in-stent luminal assessment with MSCT because of blooming artefacts, the polymeric BVS stent is radiolucent except for two metallic markers located at both extremities of the stent that facilitates the luminal and length assessment of the scaffold with MSCT ( Figure 8). Due to the non-invasive nature of MSCT, 25 patients underwent MSCT imaging 18 months after the index procedure, which represents a larger number of patients as compared to the patients who underwent conventional coronary angiography at 24 months (n=19). Out of the 25 patients who underwent MSCT, quantitative analysis was feasible in 24. According to MSCT measurements, the mean luminal area was 5.2 ± 1.3mm2, the minimal lumen area was 3.6 ± 0.9 mm2 and the mean area stenosis was 34 ±15%. The calculated mean diameter stenosis was 19 ± 9 % and was in fact not too much at variance with the results of invasive quantitative coronary angiography (% diameter stenosis, 27±11%).

At 5 years, 18 patients underwent MSCT angiography. All scaffolds were patent with an median minimal lumen area of 3.25 mm2 (interquartile range [IQR] 2.20, 4.33). Non-invasive FFR analysis was feasible in 13 of the 18 scans, which yielded a median distal FFR of 0.83 [IQR: 0.81, 0.94] ( Figure 9). [107] The feasibility and accuracy of the use of MSCT in analyzing radiolucent biodegradable stents with a possibility of functional assessment may therefore usher in a new era of non-invasive evaluation of patients treated with radiolucent stents.[40, 108]

The underlying principle of palpography is that softer tissue is more readily deformed than harder or scaffolded tissue when pulsatile arterial pressure is applied.[109, 110, 111] The rationale of this analysis in the present study was to detect some subtle changes in strain resulting from scaffolding and late bioresorption of the stent. The deformability of a vessel wall is quantified using back-scattering radiofrequency analysis of signals at different diastolic pressure levels. This allows for the reconstruction of a color-coded ‘strain’ image in which harder (low strain in blue color) and softer (high strain in yellow color) regions of the coronary arteries can be identified, with radial strain values ranging between 0% and 2%.

At 2-year follow-up, the cumulative strain values showed no significant interval changes between 6 months and 2 years follow-up (0.28± 0.12 vs. 0.31± 0.17, p=0.80). These findings suggest that even 2 years after implantation of BVS, plaque deformability remains diminished and less prone to rupture. ( Figure 10)

In the ABSORB trial, multiple imaging modalities/analyses were used in an attempt to evaluate the bioabsorption of the polymeric struts [40, 105, 112]. From pre- to post-stenting, there was an increase in the mean “DC” (9.8% vs. 25.4%, p<0.001) and “necrotic core (NC)” (15.5% vs. 30.5%, p=0.001) (n=13).[113] However, at 6-month follow-up, VH showed a relative 30% decrease in “DC” (29.7% vs. 21.2%, p<0.001) and a nearly 20% decrease in “NC” (26.9% vs. 21.9%, p=0.005) (n=27). Between 6 months and 2-year follow-up, IVUS-VH assessments demonstrated no significant differences in percentage of each plaque component ( Figure 10).[40] Serial OCT data obtained immediately after stent implantation, at 6 months and 2-year follow-up were available in 7 patients from the ITT population[40]. One of the main findings was a reduction in the number of apparent struts over time. The total number of apparent struts decreased from 403 at baseline to 368 at 6 months follow-up and to 264 at 2-year follow-up (35% reduction over two years). Strut appearance at 2 years is shown in Figure 11. The main observation provided by Grey-scale IVUS was the significant increase in minimal luminal area and average luminal area/volume together with a significant decrease in plaque area/volume between 6 months and 2-year follow-up. With the exception of the minimal luminal area (decreasing from 5.09 to 4.35 mm2, p=0.034), there were no apparent differences in the vessel area, average luminal area, plaque area as well as lumen area stenosis between the immediate post-procedure and 6-month follow-up measurements. Of note, the vessel area/volumes remained constant during follow-up suggesting the absence of significant remodeling; late enlargement of the lumen was however observed in OCT analysis (n=7).

To study vasomotion, either the endothelium independent vasoconstrictor methylergonovine maleate (methergin), or the endothelium dependent vasoactive agent, acetylcholine were administered at the time of 2-year angiographic follow-up. Mean lumen diameters were measured by QCA after baseline saline infusion and after administration of methergine/acetylcholine. Both tests were terminated by intracoronary administration of 200 micrograms of nitroglycerin. In the methergin group (n=7), there was significant vasoconstriction in proximal (pre 2.70±0.43mm vs. post Met 2.49±0.46mm, p=0.02) and scaffolded segments (pre 2.64±0.22mm vs. post Met 2.44±0.33mm, p=0.03). In the acetylcholine group (n=9), 5 patients exhibited vasodilation in the scaffolded segment ( Figure 12). These results suggested that there was restoration of vasomotor function in the scaffolded segment, an observation that has obviously never been made after metallic stent implantation.

At 5 years, clinical follow-up was obtained from 29 out of 30 enrolled patients.[67], [114, 115, 116] There was only one non-Q wave myocardial infarction (peak troponin 2.21ng/ml) related to the treatment of a non-flow-limiting stenosis (QCA DS 42%) in a patient implanted with the BVS 46 days earlier. Furthermore, this patient experienced a single episode of angina at rest without any electrocardiographic evidence of ischemia.

Otherwise, there were no new MACE events between 6 months and 5 years, and no instances of stent thrombosis as defined by the protocol or ARC definitions. In total, the MACE rate at 5 years was 3.4%.[116]

During the ABSORB cohort A trial, the mechanical properties of the polymeric stent were assessed. The acute recoil was evaluated in angiography as the difference between the mean luminal diameter of the scaffolded vessel and the diameter at maximal balloon inflation. The mean percent acute recoil was 6.85±6.96% for BVS in the Absorb cohort A trial, while it was 4.27 ±7.08% for Xience V stent (Abbott Vascular, Santa Clara, USA) in the SPIRIT I trial. [117] This suggests that the scaffolding properties of the BVS are slightly weaker than in the Xience V stent.

The late recoil assessed by means of IVUS was defined as a reduction of the stent area from post procedure to 6-month follow-up.[118] At 6 months, the lumen area was reduced by 16.6%, while the late recoil was 11.7%. This suggested that approximately two-thirds of the luminal area reduction was caused by late recoil. ( Figure 13) To enhance the mechanical strength of the struts and to reduce acute and late recoil[78], the strut design and the manufacturing process of the polymer were modified in the revised version: BVS 1.1. Firstly, the new design has in-phase zigzag hoops linked by bridges that allow a more uniform strut distribution, reduce maximum circular unsupported surface area (MCUSA) and provide more uniform vessel wall support and drug transfer ( Figure 8).[119] Secondly, a modified manufacturing process has resulted in a slower hydrolysis (in vivo degradation) rate of the polymer, thus preserving its mechanical integrity for a longer period of time [120].

The BVS revision 1.1 was tested in 101 patients of the ABSORB Cohort B study. This cohort was subdivided in two subgroups of patients: the first group (B1) had to undergo invasive imaging with QCA, IVUS, IVUS-VH and OCT at 6 and 24 months whereas the second group (B2) underwent invasive imaging at 12 and will repeat at 36 months.[107], [121, 122, 123] Between one and three years, late luminal loss remained unchanged (6 months: 0.19 mm, 1 year: 0.27 mm, 2 years: 0.27 mm, 3 years: 0.29 mm) and the in-segment angiographic restenosis rate for the entire cohort B (n=101) at three years was 6%. On IVUS, mean lumen, scaffold, plaque and vessel area showed enlargement up to two years. Mean lumen and scaffold area remained stable between two and three years whereas significant reduction in plaque behind the struts occurred with a trend toward adaptive restrictive remodelling of EEM ( Figure 14). Hyperechogenicity of the vessel wall, a surrogate of the bioresorption process, decreased from 23.1% to 10.4% with a reduction of radiofrequency backscattering for dense calcium and necrotic core (p<0.001). At three years, the count of strut cores detected on OCT increased significantly, probably reflecting the dismantling of the scaffold; 98% of struts were covered. In the entire cohort B (n=101), the three-year major adverse cardiac event rate was 10.0% without any scaffold thrombosis. [120] Between 6 months/1 year and 5 years angiographic luminal late loss (LLL) remained unchanged; B1 0.14±19mm vs 0.13±0.33mm, p=0.7953, B2 0.23±0.28mm vs 0.18±0.32mm, p=0.5685. When patients with a TLR were included LLL was 0.15±0.20 mm vs 0.15±0.24 mm, p=0.8275 for B1 and 0.30±0.37 mm vs 0.32±0.48 mm, p=0.8204 for B2. At 5 years, in-scaffold and in-segment binary restenosis was 7.8% (5/64) and 12.5% (8/64).

On IVUS the minimum lumen area of B1 decreased from 5.23±0.97 mm² at 6 months to 4.89±1.81mm²at 5 years, p=0.04 but remained unchanged in B2 (4.95±0.91mm² at 1 year to 4.84±1.28 mm²at 5 years, p=0.5). At 5 years the struts were no longer discernable by OCT and IVUS. On OCT the minimum lumen area (MLA) in B1 decreased from 4.51±1.28mm²at 6 months to 3.65±1.39mm²at 5years, p=0.01, but remained unchanged in B2, 4.35±1.09mm²at 1 year and 4.12±1.38mm² at 5 years, p=0.24. Overall, the 5-year Major Adverse Cardiac Event (MACE) rate was 11.0% without any scaffold thrombosis [124].

Randomized trials and registries

The promising results of this second-generation bioresorbable drug-eluting scaffold (BVS 1.1) constitute the proof of concept that this device can adequately revascularize coronary vessels and prevent restenosis. After the encouraging results of ABSORB B, a number of large registries were initiated [125, 126, 127] to evaluate results in real world patients, but most importantly, a series of randomized controlled trials (RCTs) comparing the Absorb BRS vs a CoCr-based, everolimus eluting stent (Xience, Abbot Vascular) were conducted, with on-going follow-up. Currently, 2-3 years’ interim analysis results are available for all of these trials and for most even longer follow-up results have been presented, although some of them have not been formally published in peer reviewed journals. The planned follow-up extends to 5 years for all ABSORB RCTs, except for ABSORB Japan (4 years) and TROFI II (3 years). In general, these trials included relatively simple lesions –i.e. excluded moderately or heavily calcified or thrombotic lesions and excessive vessel tortuosity- while intravascular imaging guidance was used only in 23.9% of the BRS-treated patients[128] with no collection of the manner in which it was applied to guide device implantation[129]. The latest available results of these trials are presented in Table 5. In ABSORB China, enrolling 480 patients (1:1) with a primary end-point of angiographic in-segment LLL at 12 months, Absorb BRS was non-inferior to Xience with regards to the primary end-point, while at 3 years there were no statistically significant differences between the two devices with respect to all clinical end-points[130]. In ABSORB Japan, enrolling 400 patients (2:1) with a primary end-point of target lesion failure (TLF) at 12 months [a composite of CD, target-vessel MI (TV-MI), and ID-TLR], Absorb BRS was demonstrated to be non-inferior to Xience [131], while TLF rates were not significantly different between the two devices at both 2 and 3 years of follow-up [132, 133]. ABSORB II, enrolled 501 patients in a 2:1 fashion and the primary endpoint was superiority of the Absorb BRS versus the Xience DES in angiographic vasomotor reactivity at 3 years with a co-primary endpoint being the non-inferiority of angiographic late luminal loss. The trial, which is the longest randomized comparison to date, did not meet its co-primary endpoints. More troubling, however, was the fact that although the study was not powered for clinical endpoints, the Absorb BRS demonstrated significantly higher rates of the secondary TLF endpoint at 3 years (10% vs 5%, p=0.04), driven by increased rates of TV-MI (6% vs 1%; p=0.01)[134], despite similar rates of this end-point at shorter follow-ups: 1[135] and 2[136] years. Patient-oriented composite end-point and definite ScT rates were not statistically different between the two devices, but rates of definite or probable ScT were higher for Absorb (3% vs 0%, p=0.03)[134]. Interestingly, the recently published 4 years’ data in 428 patients, documented non significantly different rates of TLF (11.1% vs. 6.0%, p=0.05), while no new events of ScT were observed in the landmark analysis from 3 to 4-years[137]. ABSORB III is the largest RCT reported today and the pivotal trial of US premarket approval of the Absorb BRS. It enrolled 2008 patients randomized in a 2:1 fashion and its primary end-point was TLF (CD, TV-MI or ID-TLR) at 1 year. Although Absorb was non-inferior to Xience in both intention-to-treat and as treated analyses with regards to the primary end-point[138], at 2 years, TLF rates were significantly higher for Absorb BRS (11.0% vs. 7.9%, p=0.03) and this result was driven by the higher rates of TV-MI (7.3% vs. 4.9%, p=0.04)[139]. At 3 years, TV-MI rates remained significantly higher for Absorb BRS and additionally, the numeric higher tendency in device thrombosis against Absorb observed at 2 years (1.9% vs. 0.8%) became a statistically significant difference (2.3% vs. 0.7%; p = 0.01). However, TLF rates were non significantly higher both through 3 years (13.4% vs 10.4%, p = 0.06) and between 1 and 3 years (7.0% vs 6.0%, p = 0.39)[140]. Patients with thrombotic events in both treatment periods were taking dual antiplatelet therapy (DAPT) at the time of the events, the majority of which occurred in appropriately sized vessels. However, in BRS-assigned patients, treatment of vessels with diameter <2.25 mm was an independent predictor of 3-year TLF and ScT, driven by the relationship at 1 year. The AIDA investigator-initiated trial enrolled 1845 patients in a 1:1 fashion and was designed to evaluate the non-inferiority of Absorb vs. Xience at 2 years with regards to TLF (a composite of CD, TV-MI or TLR). The data and safety monitoring board of the trial recommended early reporting of the study results, due to safety concerns, after a median follow-up of 707 days. By then, no significant differences in the rates of the primary end point were observed (11.7% vs. 10.7%, p=0.43) and this was also the case for CD and TLR. However, a higher incidence of TV-MI (5.5% vs. 3.2%, p = 0.04) was observed, driven by increased rates of definite or probable ScT for the Absorb BRS (3.5% vs. 0.9%, p<0.001). Furthermore, no major predictors of ScT were found[141]. EVERBIO II had the most liberal inclusion criteria in the ABSORB RCT family. In this single-center study, a total of 240 patients were randomized in a 1:1:1 fashion to Absorb BRS, everolimus-eluting persistent polymer DES (EES) and biolimus-eluting bioabsorbable polymer DES (BES) with the primary endpoint being angiographic in-device late loss at 9 months. The rates of the primary endpoint were similar for BVS vs. EES/BES, while a post-hoc non-inferiority analysis showed non-inferiority (p<0.001) of the BVS for the same end point. Overall, event rates were very low and the study was underpowered regarding clinical end-points. However, in the comparison between BRS and BES, device related adverse events at 2 years were significantly higher for Absorb[142]. TROFI II was another relatively small RCT (n=191), whose importance lies on the fact that it included only patients with STEMI undergoing primary percutaneous coronary revascularization. The primary endpoint of the 6-month optical frequency domain imaging healing score (HS) was based on the presence of uncovered and/or malapposed stent struts and intraluminal filling defects and was lower in the Absorb arm (P_{non-inferiority} < 0.001)[143]. At 2 years, the rates of all clinical secondary end-points were similar between the two devices[144]. Again, event rates were too low to allow any meaningful statistical comparisons or clinical correlations.

Amaranth

The Amaranth scaffold family (Amaranth Medical, CA) includes BRS with unique polymer production features that result in enhanced radial force and over-expansion capabilities with increased fracture resistance. The first generation device in this group is the Fortitude scaffold (150 microns), which was initially tested as a non-drug eluting version in the MEND I (n=13) trial and subsequently as a sirolimus-eluting version with encouraging results in the RENASCENT-I, MEND II[145] and FORTITUDE (n=63) trials[146, 147], with the latter showing 5.3% TLF and 1.8% ScT rates at 2 years. It was then further miniaturized, leading to two newer iterations, the Aptitude and Magnitude scaffolds, with strut thickness of 115 and 98 microns respectively. The single arm RENASCENT-II trial enrolled 60 patients treated with the Aptitude scaffold and has reported on its primary endpoints of safety and efficacy at 9 months[148]. In-scaffold LLL was 0.34±0.36 mm, TLF rates were 3.4% driven by 2 cases of TV-MI, while no ScT were observed. Clinical success was 98.3%, while OCT demonstrated 97% strut coverage and low rate of malapposition. CE mark approval has been submitted. The on-going FIM RENASCENT-III trial with complete enrolment of 70 patients will address the outcomes of the Magnitude scaffold, the world’s first clinically tested sub-100-micron BRS, and the results regarding its 9-month primary end-point are expected in May 2018. Preliminary results at 30 days for the entire cohort have shown 3 TVF cases and no ScT[147]

MeRes 100

MeRes BRS (Meril Life Sciences, Vapi, Gujarat, India) is a PLLA-based, sirolimus-eluting BRS with hybrid geometry structure that provides high radial strength, thin struts and tri-axial radiopaque markers. MeRes Ι FIM study (n=108) was performed in 16 medical centres of India [58] and showed in-scaffold LLL of 0.15±0.23 mm with “virtually complete" strut coverage (99.3%) at 6 months and very low MACE rates at 1 year (0.93%, 1 ischemia-driven TLR with no ST). Based on the impressive results of the FIM trial, the company has initiated an ambitious programme of further trials being performed in non-US centres, but with worldwide participation, including MeRes-I extend and MeRes-100 China, a pivotal RCT of MeRes100 vs Xience (1:1) with a goal of enrolling 470 patients.

Mirage

Mirage (Manli Cardiology Singapore) is a PLLA-based sirolimus-eluting scaffold, that incorporates a unique helix coil design for high flexibility, has relatively short bioresorption time, high scaffold dislodging force and high radial strength, while it can be stored at room temperature (as long as it is below 25^oC) and has been shown to be “MR conditional”[149]. Clinically, it was evaluated in a single blinded, randomized clinical trial of 60 patients who were randomized to either the Mirage (n=31) or the Absorb (n=29) BRS. The primary end-point of in-scaffold LLL at 12 months (0.48±0.49 mm for Mirage), as well as the rates of all clinical end-points were similar between the two study groups, despite the fact that diameter stenoses 1 year after implantation on angiography and OCT were significantly higher with Mirage[150]. Clinical follow-up of this trial is scheduled yearly for 5 years, but the company will begin a new study, incorporating small changes to further enhance the performance of this device.

Firesorb

Firesorb (Shanghai MicroPort Medical) is a sirolimus-eluting BRS, with similar structure with the Absorb BRS but significantly lower strut thickness and antiproliferative drug dose (60%). It has been tested in the FIM FUTURE I trial, a single-center Chinese study in 45 patients randomly assigned to 2 cohorts (2:1), undergoing multimodality intravascular imaging at 6 months and 1 year respectively. TLF and ScT rates were 0% both at 30 days (primary end-point) and at 1 year, with only 1 patient undergoing revascularization for a non-target MI the day after the index procedure. At 6 months, in scaffold LLL was 0.15±0.11 mm and struts coverage 98.4%. At 12 months, the latter was significantly increased to 99.0%, while there were no other significant differences between the two cohorts in terms of imaging findings[151]. In September 2017, the manufacturing company announced the enrolment of the first patient in the FUTURE II RCT which is designed to enrol 610 patients and test the Firesorb BRS against the Xience stent (NCT02890160).

Xinsorb

The Xinsorb BRS (Shandong Huaan Biotechnology Co., Ltd., Hangzhou, Zhejiang, People’s Republic of China) is a sirolimus-eluting BRS, with a backbone composed of a mixture of PLA, PCL and PGA and a coating of PDLLA mixed with PLLA. The device is stored at 4°C [152]. The manufacturing company has completed a small (n=30) FIM trial in China, in which imaging findings at 6 months were favourable[153] and no MACE were observed, but extended clinical follow-up to 30 months demonstrated a MACE rate of 3.3%, with one confirmed case of ScT and 2 cases of ID-TLR[154]. Based on the encouraging initial results of the FIM trial, a RCT versus a biodegradable-polymer DES of 400 patients and a single-arm registry of 800 patients were subsequently initiated. Both of these trials have completed enrolment in June 2016 and are currently in the data collection phase.

NeoVas

This sirolimus-eluting, PLLA-based BRS with relatively thick struts (NeoVas, Lepu Medical, Beijing, China) was initially tested in an on-going follow-up FIM study of 31 patients that at 6 months reported TLF rates of 3.2%, no ScT and angiographic in-scaffold LLL of 0.26±0.32 mm with 95.7% strut coverage[155]. Most importantly, the NeoVas BRS was subsequently tested in a 1:1 design RCT of 560 patients versus the Xience stent, that has completed one-year clinical and angiographic follow-up in 99.8% and 87.9% of the patients respectively[156] and upon announcement of the results will become the first BRS to present controlled data after ABSORB.

Iron based alloy

Iron was the first metal to be considered for use in a BRS. In 2001 Peuster et al. developed a corrodible iron stent (NOR-1) containing 41mg of pure iron, which represents a month’s normal oral iron intake in a common human diet[157]. The stents which have a degradation period of >1 year were implanted into 16 rabbits, all of which experienced no adverse events or thromboembolic complications during 6-18 months follow-up. Importantly there was no evidence of any toxic effects of iron or iron related organ damage. Furthermore, no significant neointimal proliferation, or inflammatory response was seen on histological examination.[157] A further study in 29 mini pigs compared the iron alloy stent with a BMS, and demonstrated comparable neointimal proliferation at 12 months with no evidence of toxicity.[158] This device was tested in 2009 in the FIM WHISPER study (n=11). Although the trial has not been fully reported, IVUS and OCT showed higher-than-expected intimal hyperplasia that was attributed to the too rapid elution and the too low surface area dose of the antiproliferative drug[159]. A new iteration has been developed with higher dose and slower release kinetics of the drug and also thinner struts. It is currently undergoing pre-clinical evaluation.

BRS technologies in preclinical phase

ArterioSorb

The ArterioSorb™ scaffold (Arterius Ltd, Leeds, United Kingdom) is manufactured using a patented die-drawing processing technique that results in improved radial strength and a thin strut thickness (95µm). The poly-L-lactic acid (PLLA) that is processed further to manufacture ArterioSorb™ using a novel technology enables the cost-effective production with high strength and stiffness and greater flexibility. The test device ArterioSorbÔ (Arterius Ltd., Leeds, UK) is composed of an 8-cell open-cell design with smaller cells at the centre to improve radial strength and cell connectors distributed in a spiral design. The scaffold is coated with a bioabsorbable (PDLA) polymer containing the active drug sirolimus. Bench testing has shown good radial strength of ArterioSorbÔ following IRIS testing of the radial force of the expanded ArterioSorb™.

HARTSORB™ stent

HARTLON (Wilmington, Delaware, USA) has developed a laboratory-scale process for manufacturing a bioresorbable scaffold that is different than previous bioresorbable scaffolds because the HARTSORB stent has struts that are comprised of a novel microporous, fibrillated polymeric material instead of a solid polymeric material. The microporous, fibrillated struts offer superior resistance against vessel recoil and strut fracture. The stronger fibrillated material offers the potential to reduce the strut thickness thereby maximizing arterial lumen capacity. The struts include very small fibrils of poly-L-lactic acid (PLLA) oriented in multiple axes between nodes of PLLA that increase the radial strength of the scaffold. The void space between the fibrils increases the toughness of the scaffold thereby reducing the potential of strut fracture during deployment or scaffold overlapping. However, as there is a limitation that fibrillated PLLA loses orientation when forming network during heating, the company has shifted to the development of ultra-high molecular weight PLLA polymer to form a scaffold with high tensile strength and ductility.

FROM INITIAL TRIALS AND PROMISES TO CURRENT STATUS AND FUTURE OUTLOOK

Seventeen years after the first implantation of a BRS in human, there are still few data to support the theoretical advantages of these devices. Although improvements in vasomotion within the treated coronary segment were documented using physiological testing in observational studies[83, 124], in ABSORB B at 5 years the improvement in vasomotion did not reach the functionality of the native vessels[160] and the randomized controlled ABSORB II trial failed to show this improvement at 3 years for the Absorb BRS against the Xience stent[134]. The latter trial on the other hand demonstrated that expansive vessel wall remodelling was more frequent and intense with Absorb at 3 years[161], supporting: (i) previous, non-randomized observations that showed that at 6 to 12 months after BRS implantation coronary geometry tends to revert to its pre-implant level [162], and (ii) recent observations showing persistence of in-scaffold late lumen enlargement with normalization of the FFR_CT up to 6 years[163]. It should be noted however, that LLL was generally larger in the Absorb compared to the Xience group. Although the altered vessel geometry and biomechanics can result in chronic irritation and flow disturbances that may contribute to neointimal proliferation and adverse events[164, 165], the clinical impact of these observations remains uncertain. This also is the case for the scarce data showing late increases in scaffold-free ostial area in jailed side branches[166] -despite initial higher post-procedural incidence of side-branch occlusion compared to DES[167]- and facilitation of non-invasive surveillance imaging with CT[168]. On the contrary, sufficient mid-term clinical data are now available to support the fact that Absorb BRS is inferior to new generation DES[134, 139, 141] and both FDA[169]and ESC-EAPCI[68] have issued documents reflecting on this position. Paradoxically, the greatest promise turned out as the Achilles foot of this emblematic first generation BRS. In all ABSORB RCTs, published rates of device thrombosis were numerically higher for Absorb, regardless of the statistical significance of this difference, while a meta-analysis of the first six trials reported a two-fold increase in the risk of device thrombosis with Absorb at just 1 year of follow-up, despite similar rates of all other clinical end-points[170]. Moreover, several recent meta-analyses summarizing mid-term outcomes at 2 years[128], [171, 172, 173, 174, 175, 176] ( Figure 15) and one also in 3 years[177] ( Figure 16) reported concordant findings, with a higher risk of both thrombosis and TLF for the Absorb BRS, the latter driven mainly by increased rates of TV-MI and ID-TLR. Current evidence also suggests no late advantage in terms of clinical efficacy, including relief of angina pectoris.

It should be emphasized that the advantages of BRS are not fully demonstrated until bioresorption is complete and long-term results, i.e. > 3 years, are still needed to complete the picture. In this regard, the 4 years results of ABSORB II although they don’t represent the beginning of a clinical benefit, may be interpreted at least as the end of the increased mid-term thrombotic risk. The ultimate judge of the Absorb BRS will probably be the on-going ABSORB IV trial, that has just recently reported 30 days’ results in 2604 patients[178]. Since ABSORB III and IV will be pooled together for a landmark clinical endpoint analysis between 3 and >5 (7-10) years, we still need to wait many years before understanding if a long-term clinical benefit can come from BRS. However, even if such a final benefit can be reached for Absorb, it will have to pass through the “Symplegades” of increased mid-term thrombotic risk. This risk was concordant across the early (<30 days), late (30 days to 1 year), and very late (>1 year) periods[171], [173, 174, 175], an observation in favour of an inherent thrombogenicity of the scaffold. The latter is partially supported by the results of the AIDA trial and seems to originate from the rather thick struts of this first generation device, although issues inherent to the mechanical disintegration of the scaffold during the resorption process may also contribute[171, 173, 179]. Whether the risk of ScT can be mitigated by a specific implantation technique, by avoiding very small vessels, or by prolonging dual antiplatelet therapy (DAPT) is subject to debate. Extension of DAPT duration until BRS resorption is completed is widely recommended, however this is based more on expert’s consensus[180] rather than on solid evidence [181] and is subject to a potential offset of its beneficial effect by the accompanying increased bleeding risk. Regarding the effects of meticulous adherence to good implantation techniques, briefly summarized in the acronym PSP (pre-dilate, sizing, post-dilate)[129], there are indeed data suggesting improved outcomes[127, 182, 183]. However, these data were derived from post-hoc analyses and have not been prospectively validated. In ABSORB-IV, where pre and post-dilatation were performed in 99.8% and 82.6% of the lesions respectively, overall rates of device thrombosis at 30 days were markedly lower compared to ABSORB III at the same time point (0.6% vs 1.1% for Absorb BRS) and Absorb was non-inferior to Xience with regards to the primary end-point of TLF, but again with numerically higher rates of ScT. Furthermore, it has been estimated that only half of patients with late or very late ScT might benefit from meticulous implantation techniques[184], which in addition result in a significant lengthening of the procedure[185]. Finally, small vessel diameter, which was an exclusion criterion in the ABSORB RCTs series, has consistently shown to be a predictor of adverse events[128, 183], but the effect of this observation in the results of these trials can only be speculated. It is conceivable that even if the aforementioned solutions become accepted as a necessary short-term price to pay in order to achieve the potential long-term benefits of BRS, it will be a Pyrrhic victory for Absorb in an era where DES are both “operator-friendly” in terms of implantation and related with very short DAPT durations. Reflecting on this, Abbott has discontinued the sales of Absorb in both USA and Europe and restricted its use in Europe in registries, maintaining a rigorous follow-up of the on-going RCTs in the hope of more favorable long-term results[186]. For the BRS technology overall however, at present hope seems to lie more on the development of second generation devices, and even Abbott is developing a new generation, sub-100 μm scaffold named Falcon[186]. Currently, the distinction between first and second-generation devices is based predominately on strut thickness, although a number of other characteristics still need improvement to achieve an ideal BRS platform, including over-expansion capability, visibility, deliverability and storage conditions. A number of second generation devices have shown encouraging results in clinical trials[187] and coupled with clear differences in material technology from the archetypical Absorb BRS challenge the class effect of its so far unfavourable results. The evaluation of these devices will not be possible until randomized data become available and many experts doubt if an appropriate challenger is currently available to go against a mature technology with solid results, like DES[185]. In spite of this, a number of companies worldwide have initiated ambitious RCTs for their products, that will determine if BRS evolutional history will follow the pattern of DES, in terms of success through further development, or another scenario will prevail.

FOCUS BOX 8Fully bioresorbable, everolimus-eluting scaffold

Fully bioresorbable, everolimus-eluting scaffold (ABSORB 1.0) has been tested in the ABSORB cohort A study since 2006
The first-in-man ABSORB trial, with multi-modality imaging showed:
1. favourable clinical outcomes (MACE rate 3.4% at 4 years)
2. occurrence of bioresorption (IVUS-VH, echogenicity and OCT),
3. late lumen enlargement (IVUS and OCT)
4. restoration of vasomotion in the scaffolded segment
At that time, elimination of scaffold shrinkage at 6 months remained a challenge

Conclusion

FOCUS BOX 11Future assessments

In the future, the clinical advantage of BRS technology over the currently available drug eluting stents needs to be further investigated
Larger studies with specific endpoints such as angina status, functional exercise testing or fractional flow reserve (FFR), radio-isotopic investigation flow and myocardial metabolism may be necessary
A potential drawback or “Achilles heel” of this emerging technology is strut fracture. The clinical significance of this needs to be further elucidated, and strut fracture should be avoided by respecting the nominal size of the scaffold

Acknowledgements

The author would like to acknowledge the work of Drs. Michel Vert, Michiel van Alst and James Oberhauser. Some of their basic research statement have been quoted but rephrased. Finally the author would like to thank the investigators of ABSORB A and B, who have contributed by their investigation to the progression of the field.

Bioresorbable scaffolds