The pathology of cardiovascular interventions and devices for coronary artery disease, vascular disease, heart failure, and arrhythmias

Mechanisms of luminal expansion

The mechanisms of luminal expansion in coronary balloon angioplasty and its consequences have been characterized in animal models, postmortem human hearts, specimens obtained from patients who have undergone PTCA, and by in vivo intravascular ultrasound. The five possible mechanisms of luminal expansion are (1) plaque fracture, (2) medial dissection, (3) stretching of plaque-free arterial segments, (4) plaque stretching, and (5) plaque compression and redistribution ( Figs. 18.1 and 18.2 and Table 18.1 ) .

In PTCA, the progressive and substantial expansile force induced by a balloon inflated at 8–12 atm of pressure causes a radial force that induces an acute and complex arterial injury consisting of endothelial denudation, plaque fracture, intimal-medial separation accompanied by an intramural dissection, and stretching of the media and adventitia, which causes the essentially nondistensible plaque to split . The site of plaque disruption (its weakest point) is not necessarily that which is most severely involved with atherosclerosis. The split extends at least to the intimal-medial border and often into the media, with consequent circumferential and longitudinal dissection . Dissimilar plaques respond differently to balloon dilation, and the composition and configuration of the original atherosclerotic lesion play a key role in the outcome of angioplasty . Immediate success is probably enhanced in arteries having eccentric plaques with large lipid-rich necrotic cores and/or calcification, in contrast to concentric fibrotic lesions. For eccentric atheromas, balloon-induced splits most commonly involve the junction between the plaque and the disease-free portion of the arterial wall.

Plaque fracture, medial dissection, and stretching of the media beyond the dissection are accompanied by local flow abnormalities and generation of new thrombogenic blood-contacting surfaces ( Figs. 18.3 and 18.4 ). This supports the concept that an atherosclerotic plaque after angioplasty has many features of a spontaneously disrupted plaque, namely those associated with the acute coronary syndromes. The immediate postangioplasty healing process in either arteries or bypass grafts is not well understood, but dissolution of soft atheromatous material, retraction of the split plaque, thrombus formation, and intimal healing with re-endothelialization likely contribute.

Comparison of percutaneous transluminal coronary angioplasty with thrombolytic therapy

PTCA has largely replaced primary thrombolytic therapy for the treatment of acute MI in patients reaching the hospital . Primary PTCA is beneficial in reducing the composite end point of death, recurrent MI, or stroke, but this benefit is attenuated at 6 months (owing to proliferative restenosis; see below). Thus primary PTCA may ultimately achieve a higher patency rate, but thrombolytic therapy may actually open more arteries very early (before 60 min). However, the risk for early reocclusion is nearly twofold higher for patients receiving thrombolysis than for patients receiving primary PTCA. Today, low-dose thrombolytic therapy is being used as initial therapy in the field while patients are being transported to the hospital where PCI is instituted . Ongoing research is aimed at developing adjunctive therapy to reduce the amount of reperfusion injury that frequently develops with either thrombolytic therapy or PCI .

Complications

Major ischemic complications of PTCA that prompt emergency bypass surgery occur in less than 1% of patients treated by PTCA . Features associated with increased risk for PTCA complications include advanced age, female gender, congestive heart failure (HF) or left ventricular dysfunction, recent thrombolytic therapy, and complex coronary lesional anatomy. Periprocedural MI owing to side branch occlusion or embolization of thrombus, platelet aggregates, or plaque occurs in 5%–20% of patients. The most important complications are abrupt closure (i.e., an early and sustained reduction of flow in the treated vessel) and restenosis (i.e., late loss of the acute luminal gain at the treated site) .

Late restenosis

Progressive restenosis accompanied by recurrent ischemic signs and symptoms limits the long-term success of angioplasty ( Fig. 18.5 ). Clinically significant restenosis occurs in ~30%–50% of patients after coronary balloon angioplasty, most frequently within the first 4–6 months. Restenosis probably represents a fundamental but complex vascular healing response that achieves clinical significance only in some patients. The risk of restenosis depends on definition used, patient population, and lesion complexity. For example, the long-term outcome is better among patients with single-vessel disease than those with multivessel disease. Many therapeutic approaches have been directed toward prevention of restenosis. Endovascular stents have reduced the frequency of this complication (see later). PTCA has also been used in coronary bypass grafts, but the late success rates are less than in native vessels.

Although vessel wall recoil and organization of thrombus likely contribute, the major process leading to restenosis is excessive smooth muscle proliferation as a response to angioplasty-induced injury ( Fig. 18.5 ). Medial smooth muscle cells migrate to the intima where, along with existing plaque smooth muscle cells, they proliferate and secrete abundant extracellular matrix, consisting predominantly of collagen and glycosaminoglycans . This yields lesions that have a loose myxoid matrix in which numerous stellate smooth muscle cells are haphazardly arranged. Over time, the extracellular matrix becomes more densely collagenous, and the neointima becomes less cellular, with scattered spindle-shaped cells in a more laminar arrangement. The process of restenosis has mechanistic similarities to both atherosclerosis and vascular graft healing. Moreover, after prominent thrombus formation, or in a hyperlipidemic patient, the site of healing may also contain foam cells and lipid crystals, resembling a mature fibrofatty atheroma. There is considerable interest in locally delivered pharmacologic and molecular therapies to mitigate restenosis. However, despite the efficacy of various approaches in animal models, success in humans has not been effective. In general, the largest acute luminal diameter provides better tolerance of subsequent intimal proliferation before hemodynamically significant renarrowing results at the treatment site.

Bare-metal stents, drug-eluting stents, and new-generation stents

The first coronary stents developed for clinical use were bare-metal stents (BMS). The proliferative restenosis that occurs after PTCA alone was found not to be prevented by stenting with BMS, and the tissue between the stent and the blood (called a neointima ) may progressively thicken. With the goal of eliminating this intimal thickening, drug-eluting stents (DESs) were developed . DESs markedly reduced restenosis compared with BMS, and consequently, they have been widely used in the treatment of coronary artery disease. The first generation of DES, composed of stents coated with a polymer containing either of the drugs, sirolimus or paclitaxel, had significant problems, including hypersensitivity reactions, delayed healing, and restenosis, often with superimposed atherosclerosis . More recently, stenting has progressed to second-generation drug-eluting metal stents, with features intended to improve implantation and biocompatibility, such as thinner struts, improved designs that maintain radial strength, despite improved deployability, biodegradable drug-containing polymers, and new drugs and dosing regimens. Notwithstanding these improvements, rigid metallic stents (BMS and DES) inhibit coronary vasomotion owing to splinting, interfere with further interventions such as coronary artery bypass grafting, and can cause prolonged tissue reactions.

Therefore there is considerable interest in developing drug-eluting biodegradable/resorbable/stents (BRSs) composed of biomaterials that degrade and disappear over time . In contrast to the permanent presence of a foreign body in BMS and DES, BRSs provide a scaffold to support vessel expansion, avoid restenosis, and then ultimately disappear completely by resorption of the foreign material. BRSs also ultimately permit more versatility in subsequent therapies, and do not interfere with the diagnostic evaluation by noninvasive imaging such as cardiac magnetic resonance and CT. The most studied BRS is the Absorb bioresorbable vascular scaffold (Absorb BVS, Abbott Vascular). However, conflicting data regarding the safety and efficacy of RBS and efficacy in preclinical and clinical studies, particularly related to the possibility of stent thrombosis, have slowed down their introduction into the market .

Complications

The complications and limitations of stenting with either BMS or DES include initial failure, early thrombosis, and late restenosis ( Fig. 18.8 ). Subacute stent thrombosis (ST) occurs in 1%–3% of patients, usually within 7–10 days of the procedure, and results in occlusion of the stented vessel with platelet-rich thrombus and associated MI or death. Antiplatelet treatment minimizes the risk of subacute ST. With BMS, the most significant complication associated with long-term stent placement is in-stent restenosis caused by intimal thickening . DESs greatly reduce the risk of restenosis due to intimal proliferation compared with BMS; however, late ST emerged as a major safety concern with first-generation DES early after their adoption in clinical practice, requiring prolonged dual antiplatelet therapy . Biomarkers are being utilized to predict complications of stent placement .

Pathologic findings in humans

The histologic changes induced in human atherosclerotic coronary arteries by a wide variety of commercial stents have been described . In general, implantation is accompanied by damage to the endothelial lining and stretching of the vessel wall, stimulating early adherence and accumulation of platelets and leukocytes . Stent wires are initially covered with a variable platelet-fibrin coating. They may eventually become completely covered by a neointima, whereby the wires are embedded in a layer of intimal thickening consisting of α-actin positive smooth muscle cells in a collagenous matrix, in some cases with foam cells, extracellular lipid, and lipid crystals predominantly near the stent wires ( Fig. 18.8 ). Ultimately, there is a neointima lined by endothelium. Most stent struts are in direct contact with atherosclerotic plaque; in some cases, medial damage is associated with struts.

One study reported findings in 55 stents at a mean of 39 days after implantation, and in 10 patients later than 30 days . Fibrin, platelets, and neutrophils were associated with stent struts early after deployment. In stents implanted 3 days or less, inflammatory cells were sparse when associated with struts in contact with fibrous plaque but were prominent with struts embedded in either lipid core or damaged media. Neointimal growth in stents implanted longer than 30 days was greatest at strut sites when medial laceration or rupture was present compared with struts in contact with plaque or with an intact media. Medial destruction and lipid core penetration increased inflammation, and neointimal growth increased as the ratio of stent area to reference lumen area increased. The presence of increased inflammation associated with stent struts in the vicinity of damaged media and increased neointimal thickness at struts associated with medial damage suggests that deployment strategies that reduce severe arterial injury during catheter-based interventions with stents can have a beneficial effect by lowering the frequency of in-stent restenosis. An autopsy study of DES suggested that although DES reduced intimal proliferation, they impaired endothelial healing, suggesting a potential mechanism for late thrombosis . The intimal lesions that do develop in some stents may undergo secondary atheromatous change, that is, neoatherosclerosis . Distinctive features of the various first, second, and current generation stents have been characterized . As mentioned above, early clinical use of fully bioresorbable stents suggests a high degree of thrombosis . Certainly, further understanding of mechanisms of acute and chronic vascular reactions and the role of stent biomaterials and design factors will be crucial to understand and potentially eliminate deleterious interactions of various stent types . In vivo imaging studies may be a useful adjunct to observe such interactions longitudinally.