The advances in the technology have greatly improved the success and reduced the complications of the procedure. Currently, a successful procedure (angiographic success), defined as a reduction of the stenosis to less than a 20% diameter narrowing, occurs in 95–99% of patients. The success is dependent upon the coronary anatomy, with lower success rates in patients with tortuous, small, or calcified vessels or chronic total occlusions. Chronic total occlusions have the lowest success rates and their recanalization is usually not attempted unless the occlusion is recent (within 3 months) or there are favorable anatomic features. Improvements in equipment and technique have increased the success rates of recanalization of chronic total occlusions.
Serious complications are rare but include a mortality rate of 0.1–0.3% for elective cases, a large myocardial infarction occurs in less than 3%, and stroke in less than 0.1%. Patients who are elderly (>65 years), undergoing an emergent or urgent procedure, have chronic kidney disease, present with an ST-segment elevation myocardial infarction (STEMI), or are in shock have significantly higher risk. Scoring systems can help to estimate the risk of the procedure, although no perfect scoring system has yet been developed.
Myocardial infarction during PCI can occur for multiple reasons including an acute occluding thrombus, severe coronary dissection, embolization of thrombus or atherosclerotic material, or closure of a side branch vessel at the site of angioplasty. Most myocardial infarctions are small and only detected by a rise in the creatinine phosphokinase (CPK) or troponin level after the procedure. Only those with significant enzyme elevations (more than three times the upper limit of normal) are associated with a less favorable long-term outcome. Coronary stents have largely prevented coronary dissections due to the scaffolding effect of the stent. Metallic stents are also prone to thrombotic occlusion (1–3%), either acute (<24 h) or subacute (1–30 days), which can be ameliorated by greater attention to full initial stent deployment and the use of dual antiplatelet therapy [aspirin, plus a platelet P2Y12-receptor blocker (clopidogrel or prasugrel)]. Late (30 days–1 year) and very late stent thromboses (>1 year) occur very infrequently with stents but are slightly more common with drug-eluting stents, necessitating dual antiplatelet therapy with these stents for up to 1 year or longer. Premature discontinuation of dual antiplatelet therapy particularly in the first month after implantation is associated with a significantly increased risk for stent thrombosis (three- to ninefold greater). Stent thrombosis results in death in 10–20% and a myocardial infarction in 30–70% of patients. Elective surgery that requires discontinuation of antiplatelet therapy after drug-eluting stent implantation should be postponed until after 6 months and preferably after 1 year, if at all possible.
Restenosis, or renarrowing of the dilated coronary stenosis, is the most common complication of angioplasty and occurs in 20–50% of patients with balloon angioplasty alone, 10–30% of patients with bare metal stents, and in 5–15% of patients with drug-eluting stents. The fact that stent placement provides a larger acute luminal area than balloon angioplasty alone reduces the incidence of subsequent restenosis. Drug-eluting stents further reduce restenosis through a reduction in excessive neointimal growth over the stent. If restenosis does not occur, the long-term outcome is excellent (Fig. 246-3). Clinical restenosis is recognized by recurrence of angina or symptoms within 9 months of the procedure. Most commonly, patients with clinical restenosis present with worsening angina (60–70%), but patients can present with non-ST-elevation myocardial infarction (10%) or ST-elevation myocardial infarction (5%) as well. Clinical restenosis requires confirmation of a significant stenosis at the site of the prior PCI, with repeat PCI or coronary artery bypass grafting (CABG). This is termed target lesion revascularization (TLR) or target vessel revascularization (TVR). By angiography, the incidence of restenosis is significantly higher than clinical restenosis (TLR or TVR) because many patients have mild restenosis that does not result in a recurrence of symptoms. The management of clinical restenosis is usually to repeat the PCI with balloon dilatation and, placement of a bare metal or a drug-eluting stent. Rarely, intracoronary brachytherapy using beta radiation is used. Once a patient has had restenosis, the risk of a second restenosis is further increased. The risk factors for restenosis are diabetes, long lesions, small-diameter vessels, and suboptimal initial PCI result.
The mechanism of restenosis is similar to that of wound healing, with inflammation and the migration and proliferation of smooth muscle cells that create a thick neointima (scar) that narrows the lumen at the site of dilatation (Fig. 246-2). The neointima is covered with endothelium, but it remains dysfunctional. The primary cause of restenosis in balloon angioplasty is adverse vessel remodeling with constriction of the vessel relative to the adjacent nondilated vessel. This change in remodeling can be appreciated by intravascular ultrasound but not by angiography since the latter only shows the lumen and not the entire vessel size. In addition to remodeling, excessive growth of the neointima further narrows the lumen. Stents prevent this unfavorable constrictive remodeling, and drug-eluting stents not only prevent this constriction but reduce the excessive neointimal growth as well. Common risk factors for atherosclerosis such as hyperlipidemia, hypertension, or cigarette smoking do not increase the risk of restenosis, although diabetes mellitus does.
The American College of Cardiology (ACC)/American Heart Association (AHA) guidelines extensively review the indications for PCI in patients with stable angina, unstable angina, non-ST-elevation, and ST-elevation myocardial infarction and should be referred to for a comprehensive discussion of the indications. Briefly, the two principal indications for coronary revascularization in patients with chronic stable angina (Chap. 243) are (1) to improve anginal symptoms in patients who remain symptomatic despite adequate medical therapy and (2) to reduce mortality rates in patients with severe coronary disease. In patients with stable angina, who are well controlled on medical therapy, older studies and the more current Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) and Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trials have shown that revascularization does not lead to better outcomes and can be safely delayed until symptoms worsen or evidence of severe ischemia on noninvasive testing occurs. Randomized trials done in the 1960s and 1970s showed that CABG reduced mortality rates in patients with severe three-vessel or left main coronary disease when compared with medical therapy alone regardless of the degree of symptoms. Whether PCI also confers the same degree of protection is not known as trials of PCI versus medical therapy in patients with three-vessel disease have not been conducted, but randomized trials comparing CABG and PCI have shown equal rates of death and myocardial infarction (MI) rates over 5–10 years of follow-up. Consistently these studies have also shown that PCI, despite the use of stents, is associated with a 10–30% need for repeat PCI during the first year after the procedure due largely to restenosis, although drug-eluting stents have decreased this rate. This contrasts with a need for PCI or repeat CABG in bypass patients of 2–5%.
When revascularization is indicated, the choice of PCI or CABG depends upon a number of clinical and anatomic factors (Fig. 246-4). A subgroup analysis from the Bypass Angioplasty Revascularization Investigation (BARI) randomized trial showed that patients with treated diabetes mellitus and multivessel disease fared better with CABG; however, registry experiences suggest that PCI can be done in selected diabetic patients with less-severe multivessel disease with good long-term outcome. The Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) trial compared PCI with the paclitaxel drug-eluting stent to CABG in 1800 patients with three-vessel coronary disease or left main disease. The study found no difference in death or myocardial infarction at 1 year, but repeat revascularization was significantly higher in the stent-treated group (13.5 vs. 5.9%), while stroke was higher in the surgical group (2.2 vs. 0.6%). The primary endpoint of death, MI, stroke, or revascularization was significantly better with CABG due to the higher rate of revascularization in the drug-eluting stent group. Only 1 year of results are currently available and longer follow-up is needed to assess fully these two revascularization strategies in patients with severe coronary disease.
In patients requiring revascularization, several factors need to be considered in choosing between bare metal stents, drug-eluting stents, or coronary artery bypass surgery. ACS, acute coronary syndrome; BMS, bare metal stent; CABG, coronary artery bypass grafting; DES, drug-eluting stent; IVUS, intravascular ultrasound; STEMI, ST-segment elevation myocardial infarction. (From AA Bavry, DL Bhatt: Circulation 116:696, 2007; with permission.)
The choice of PCI versus CABG is also related to the anticipated procedural success and complications of PCI and the risks of CABG. For PCI, the characteristics of the coronary anatomy are critically important. The location of the lesion in the vessel (proximal or distal), the degree of tortuosity, and size of the vessel are considered. In addition, the lesion characteristics including the degree of the stenosis, the presence of calcium, lesion length, and presence of thrombus are assessed. The most common reason to decide not to do angioplasty is that the lesion felt to be responsible for the patient's symptoms is not treatable. This is most commonly due to the presence of a chronic total occlusion (>3 months in duration). In this setting the historical success rate has been low (30–70%) and complications are more common. A lesion classification to characterize the likelihood of success or failure of PCI has been developed by the ACC/AHA. Lesions with the highest success are called type A lesions (such as proximal noncalcified subtotal lesion) and those with the lowest success or highest complication rate are type C lesions (such as chronic total occlusions). Intermediate lesions are classified as type B1 or B2 depending on the number of unfavorable characteristics. Approximately 25–30% of patients will not be candidates for PCI due to unfavorable anatomy, whereas only 5% of CABG patients will not be candidates for surgery due to coronary anatomy. The primary reason for being considered inoperable is the presence of severe comorbidities such as advanced age, frailty, severe chronic obstructive pulmonary disease (COPD), or poor left ventricular function. Another consideration in choosing a revascularization strategy is the degree of revascularization. In patients with multivessel disease, bypass grafts can usually be placed in all vessels with significant stenosis, while PCI may be able to treat only some of the lesions due to the presence of unfavorable anatomy. The decision to do PCI versus CABG will then depend upon the importance of complete revascularization in the patient. Given the multiple factors that need to be considered in choosing the best revascularization for an individual patient with multivessel disease, it is optimal to have a discussion between the cardiac surgeon and interventional cardiologist and the physicians caring for the patient to properly weigh the choices.
Patients with acute coronary syndrome are at excess risk of short- and long-term mortality. Randomized clinical trials have shown that PCI is superior to intensive medical therapy in reducing mortality rate and myocardial infarction, with the benefit largely confined to those patients who are high risk. This includes patients with refractory ischemia, recurrent angina, positive cardiac-specific enzymes, new ST-segment depression, low ejection fraction, severe arrhythmias, or a recent PCI or CABG. PCI is preferred over surgical therapy in most high-risk patients with acute coronary syndromes unless they have severe multivessel disease or the culprit lesion responsible for the unstable presentation cannot be adequately treated. In STEMI, thrombolysis or PCI (primary PCI) are effective methods to restore coronary blood flow and salvage myocardium within the first 12 hours after onset of chest pain. Because PCI is more effective than thrombolysis, it is preferred if readily available. PCI is also performed following thrombolysis to facilitate adequate reperfusion or as a rescue procedure in those who do not achieve reperfusion from thrombolysis or in those who develop cardiogenic shock.
Other Interventional Techniques
Interventional treatment for structural heart disease (adult congenital heart disease and valvular heart disease) is a significant component of the field of interventional cardiology.
The most common adult congenital lesion to be treated with percutaneous techniques is closure of atrial septal defects (Chap. 236). The procedure is done as in a diagnostic right heart catheterization with the passage of a catheter up the femoral vein into the right atrium. With echo and fluoroscopic guidance the size and location of the defect can be accurately defined, and closure is accomplished using one of several approved devices. All devices use a left atrial and right atrial wire mesh or covered disk that are pulled together to capture the atrial septum around the defect and seal it off. The Amplatzer Septal Occluder device (AGA Medical, Minneapolis, Minnesota) is the most commonly used in the United States. The success rate in selected patients is 85–95%, and the device complications are rare and include device embolization, infection, or erosion. Closure of patent foramen ovale (PFO) is done in a similar way. PFO closure is an approved procedure in patients who have had recurrent paradoxical stroke despite adequate medical therapy including anticoagulation. The use in the treatment of migraine is under clinical investigation and is not an approved indication.
Similar devices can also be used to close patent ductus arteriosus and ventricular septal defects. Other congenital diseases that can be treated percutaneously include coarctation of the aorta, pulmonic stenosis, peripheral pulmonary stenosis, and other abnormal communications between the cardiac chambers or vessels.
The treatment of valvular heart disease is the most rapidly growing area in interventional cardiology. Until recently the only available techniques were balloon valvuoplasty for the treatment of aortic, mitral, or pulmonic stenosis (Chap. 237). Mitral valvuloplasty is the preferred treatment for symptomatic patients with rheumatic mitral stenosis who have favorable anatomy. The outcome in these patients is equal to that of surgical commissurotomy. The success is highly related to the echocardiographic appearance of the valve. The most favorable setting is commissural fusion without calcification or subchordal fusion and the absence of significant mitral regurgitation. Access is obtained from the femoral vein using a transseptal technique where a long metal catheter with a needle tip is advanced from the femoral vein through the right atrium and atrial septum at the level of the foramen ovale into the left atrium. A guidewire is advanced into the left ventricle, and a balloon-dilatation catheter is negotiated across the mitral valve and inflated to a predetermined size to enlarge the valve. The most commonly used dilatation catheter is the Inoue balloon. The technique splits the commissural fusion and commonly results in a doubling of the mitral valve area. The success of the procedure in favorable anatomy is 95% and severe complications are rare (1–2%). The most common complications are tamponade due to puncture into the pericardium and the creation of severe mitral regurgitation.
Similarly, severe aortic stenosis can be treated with balloon valvuloplasty. In this setting, the valvuloplasty balloon catheter is placed retrograde across the aortic valve from the femoral artery and briefly inflated to stretch open the valve. The success is much less favorable with an initial success rate of only 50% and a restenosis rate of 50% after 6 to 12 months. This poor success rate has limited its use to patients who are not surgical candidates or as a bridge to surgery in patients who are expected to improve sufficiently to become surgical candidates. In this setting, the mortality rate of the procedure is high (10%). Repeat aortic valvuloplasty as a treatment for aortic valve restenosis has been reported.
Percutaneous aortic valve replacement has been introduced to treat patients who are not suitable candidates for surgical aortic valve replacement. Currently, two valve models, the Edwards SAPIEN valve (Edwards Lifescience, Irvine, California) and the CoreValve ReValving system (CoreValve Inc., Irvine, California), have been approved for use in Europe. In more than 4000 cases worldwide, follow-up shows no evidence of restenosis or prosthetic valve dysfunction in the midterm. Both are placed either retrograde from the femoral artery or can be placed via the left ventricular apex following surgical exposure. The CoreValve is self-expanding, while the Edwards valve is balloon expanded. Following balloon valvuloplasty the valve is positioned across the valve and deployed with post-deployment balloon inflation to ensure full contact with the aortic annulus. The success rate is 80–90% and the 30-day mortality rate is 10–15%, not unexpectedly as only high-risk patients are undergoing the procedure currently. Both valves are undergoing clinical testing in the United States.
Peripheral Arterial Interventions
The use of percutaneous interventions to treat symptomatic patients with arterial obstruction in the carotid, renal, aortic, and peripheral vessels is also part of the field of interventional cardiology. Randomized clinical trial data already support the use of carotid stenting in patients at high risk of complications from carotid endarterectomy (Fig. 246-5). Ongoing trials will determine whether carotid stenting should be used even more broadly. The success rate of peripheral interventional procedures has been improving, including for long segments of occlusive disease historically treated by peripheral bypass surgery (Fig. 246-6). Peripheral intervention is increasingly part of the training of an interventional cardiologist, and most programs now require an additional year of training after the interventional cardiology training year. The techniques and outcomes are described in detail in the chapter on peripheral vascular disease (Chap. 249).
An example of a high-risk patient who requires carotid revascularization, but who is not a candidate for carotid endarterectomy. Carotid artery stenting resulted in an excellent angiographic result. (From M Belkin, DL Bhatt: Circulation 119:2302, 2009; with permission.)
Peripheral interventional procedures have become highly effective at treating anatomic lesions previously amenable only to bypass surgery. A. Complete occlusion of the left superficial femoral artery. B. Wire and catheter advanced into subintimal space. C. Intravascular ultrasound positioned in the subintimal space to guide retrograde wire placement through the occluded vessel. D. Balloon dilation of the occlusion. E. Stent placement with excellent angiographic result. (From A Al Mahameed, DL Bhatt: Cleve Clin J Med 73:S45, 2006; with permission.)
Circulatory Support Techniques
The use of circulatory support techniques is occasionally needed in order to safely perform PCI on hemodynamically unstable patients. It also can be useful in helping to stabilize patients before surgical interventions. The most commonly used device is the percutaneous intraaortic balloon developed in the early 1960s. A 7-10 French 25- to 40-mL balloon catheter is placed retrograde from the femoral artery into the descending aorta between the aortic arch and the abdominal aortic bifurcation. It is connected to a helium gas inflation system that synchronizes the inflation to coincide with early diastole with deflation by mid-diastole. As a result, it increases early diastolic pressure, lowers systolic pressure, and lowers late diastolic pressure through displacement of blood from the descending aorta (counterpulsation). This results in an increase in coronary blood flow and a decrease in afterload. It is contraindicated in patients with aortic regurgitation, aortic dissection, or severe peripheral vascular disease. The major complications are vascular and thrombotic. Intravenous heparin is given in order to reduce thrombotic complications.
Another useful tool is the Impella device (Abiomed, Danvers, Massachusetts). The catheter is placed percutaneously from the femoral artery into the left ventricle. The catheter has a small microaxial pump at its tip that can pump up to 2.5 liters per minute at a speed of 50,000 rpm from the left ventricle to the aorta. Other support devices include the hemopump and percutaneous cardiopulmonary bypass.