Qualitative and Quantitative Coronary Angiography

Irregular Lesions

With the advent of coronary stents, the prognostic importance of irregular lesions has been diminished substantially, although identification of an irregular plaque on angiography suggests an acute coronary syndrome and intracoronary thrombus. Semiquantitative and quantitative measurements of lesion irregularity were developed in the early 1990s to better characterize lesion morphology in patients with acute coronary syndromes, but these methods have not found clinical utility independent of other clinical risk factors.

In 65 patients referred for first ACS, intravascular ultrasound (IVUS) was performed prior to any intervention. IVUS identified 224 lesions, including 115 plaque ruptures, whereas angiography identified 49 suspected plaque ruptures, with a 40% sensitivity and 97% specificity; positive and negative predictive values of angiography for plaque rupture were 96% and 61%, respectively. Proximal coronary location, a wide cavity, and counterflow rupture were strong predictors of correct angiographic diagnoses, enabling four specific angiographic patterns to be identified using three-dimensional (3D) IVUS plaque rupture reconstruction. An aneurysmal or ulcerated lesion was one of several components of a stent thrombosis risk score.

Angulated Lesions

Vessel curvature at the site of maximum stenosis should be measured in the most unforeshortened projection using a length of curvature that approximates the balloon length used for coronary dilation. Although balloon angioplasty of highly angulated lesions is associated with an increased risk of coronary dissection, in the era of coronary stenting, the greatest impediment of angulated lesions is the inability to deliver the stent to the stenosis and straighten the arterial contour after stent placement, which may predispose to late stent fracture. Newer stent designs have leveraged conformability for angulated vessels to provide less vessel straightening after deployment. Prior to the understanding of the deleterious of late scaffold thrombosis, one of the potential advantages of bioresorbable scaffolds had been less vessel straightening than conventional metallic stents, particularly as the scaffold is resorbed over time.

Lesion Calcification

Coronary artery calcium remains an important marker for coronary atherosclerosis. Conventional coronary angiography has limited sensitivity for the detection of smaller amounts of calcium and is only moderately sensitive for the detection of extensive lesion calcium (60% and 85% sensitivity for three- and four-quadrant calcium, respectively). Coronary calcification reduces the compliance of the vessel and may predispose to dissection at the interface between calcified plaque and normal wall after balloon angioplasty. Coronary calcification also reduces the ability to cross chronic total occlusions, and in severely calcified lesions, stent strut expansion is inversely correlated with the circumferential arc of calcium. Patients treated with sirolimus-eluting stents who had lesion calcification had higher TLR rates than those who did not. A pooled analysis of seven trials of first- and second-generation DES identified severe calcification as an independent predictor of poor outcome after PCI.

Rotational atherectomy or other ablative technologies are the preferred pretreatment method in patients with severe lesion calcification, particularly ostial lesions; calcium ablation facilitates the delivery and expansion of coronary stents by creating microdissection planes within the fibrocalcific plaque. Even with these contemporary methods, moderate or severe coronary calcification is associated with reduced procedural success and higher complication rates, including stent dislodgment. A novel device, the Diamondback 360 Orbital Atherectomy System (Cardiovascular Systems, St. Paul, MN), used centrifugal action of a diamond-coated crown to modify calcified lesions in a series of 443 patients with severe calcification to facilitate stent placement; stent delivery was successful in 97.7% of cases. In less severely calcified lesions, no difference in the restenosis rate was found after paclitaxel-eluting stent implantation in calcified or noncalcified vessels. Calcification in SVGs usually occurs within the vessel wall rather than the lesion and is associated with older graft age, insulin-dependent diabetes, and a history of smoking. Calcified lesions were an independent predictor of stent thrombosis in one series.

Degenerated Saphenous Vein Grafts

SVGs degenerate over time, with 25% occluding within the first year after coronary bypass surgery and 50% developing occlusion within 10 years after surgery, often necessitating repeat revascularization, although these rates have been improved with aggressive lipid lowering therapy. SVGs are the most common site for a culprit lesion in patients with acute coronary syndromes after CABG and account for 6% of all PCIs. PCIs for SVG lesions have been associated with a worse late outcome compared with native vessel interventions. The SVG plaques are particularly prone to distal embolization. The risk of embolic complications appears to be related to the degree of overall graft degeneration and the length and bulkiness of the lesion. The risk of embolization may be reduced with the use of excimer laser atherectomy and with the use of undersized balloons.

Embolic protection devices (EPDs) have been associated with improved angiographic and clinical outcomes in percutaneous SVG interventions. In the landmark Saphenous Vein Graft Angioplasty Free of Emboli Randomized (SAFER) trial, use of a distal balloon occlusion EPD was associated with a decrease (9.6% vs. 16.5%) in the 30-day composite outcome of death, myocardial infarction, emergency CABG, or TLR compared with no embolic protection. Subsequent noninferiority comparisons have demonstrated similar benefit with proximal occlusion and distal filter EPD, with the benefit limited to reduction in periprocedural myocardial infarction. When feasible, the use of EPDs during SVG PCI has been given a class I recommendation in the ACC/AHA PCI guidelines. One exception may be in patients treated for in-stent restenosis (ISR), for which embolic protection may not be required.

Restenosis and TLR rates are lower with DES compared with BMS in SVG interventions, even though mortality and stent thrombosis rates are similar. Self-expanding stents made with expanded polytetrafluoroethylene (ePTFE) provide no additional advantage over noncovered balloon-expandable stents in the development of early complications or late restenosis.

Thrombus

Conventional angiography is relatively insensitive for the detection of coronary thrombus, but large thrombus may be visualized in up to 15% to 30% of patients undergoing PCI for acute coronary syndromes. Angiographic thrombus is usually identified by the appearance of discrete, intraluminal filling defects within the arterial lumen, and it is associated with a 6% to 73% incidence of ischemic complications after PCI. A large thrombus burden is an independent predictor of stent thrombosis in patients with STEMI treated with DES, and it may be managed with intensive anticoagulation therapy before PCI to reduce periprocedural complications.

Primary PCI-related complications of thrombus-containing lesions are distal embolization and thrombotic occlusion, and the risk of angiographic thrombus complications is related to the size of the coronary thrombus. Routine rheolytic thrombectomy provides no benefit in patients with AMI, although it may be useful for patients with a large thrombus burden. Several aspiration catheters have been investigated in patients with AMI and large thrombus burden, but a meta-analysis of three randomized trials involving >18,000 patients undergoing PCI found no difference between PCI with routine aspiration thrombectomy and PCI alone for recurrent myocardial infarction, stent thrombosis, heart failure, or target-vessel revascularization.

Traditionally, the extent of coronary thrombus has been determined using the semiquantitative TIMI thrombus grade (TTG). A novel method to assess intracoronary thrombus burden uses the discrepancy of luminal areas assessed with edge detection and video densitometry and measured with the Cardiovascular Angiography Analysis System II. Thrombus remains an important predictor of outcome after PCI.

Ostial Location

Ostial lesions begin within 3 mm of the origin of the coronary artery, and they are classified as aortoostial or non aortoostial lesions. Balloon angioplasty of ostial lesions is limited by suboptimal procedural outcome, primarily due to technical factors such as difficulties with guide catheter support, lesion inelasticity precluding maximal balloon expansion, and early vascular recoil limiting the acute angiographic result. Debulking techniques such as directional and rotational atherectomy improve compliance of the aortoostial lesion but have had limited effect on preventing late restenosis. Ostial lesions have been associated with higher rates of TLR after DES placement.

DES have become the default therapy for most aortoostial lesions, although there are unique challenges of stent placement in the aortoostial location, such as protrusion of the stent into the aorta precluding subsequent injection catheter engagement, stent compression, and avulsion of the stent struts into the aorta when devices such as cutting balloon angioplasty are used to treat ISR.

Aortoostial lesions remain associated with higher failure rates than non aortoostial lesions. Isolated non aortoostial stenoses of the left circumflex and left anterior descending coronary arteries and ostial side branch bifurcation lesions are also effectively treated with DES, but they pose unique challenges regarding vessel wall geometry, adequate ostial branch coverage (particularly for a narrow angle with the adjacent branch), and plaque shifting causing compromise of the parent or adjacent branch vessels. Whereas stent protrusion into the parent vessel of less than 1 mm is usually well tolerated, greater stent protrusion precludes treatment of the parent branches. Stent fractures have been reported, with more advanced stenting techniques used to treat the parent vessel and ostial side branch stenoses.

Long Lesions

Lesion length may be estimated quantitatively as the shoulder-to-shoulder extent of atherosclerotic narrowing greater than 20%, although many clinicians estimate lesion length based on the identification of a normal-to-normal segment, which is usually longer than the length obtained with quantitative methods. Conventional balloon angioplasty of long lesions has been associated with reduced procedural success, particularly when the segment is diffusely diseased (e.g., >20 mm long), primarily because of the more extensive plaque burden in long lesions.

Stents improve late outcomes compared with balloon angioplasty, but stent and lesion length remain the most important predictors of restenosis in the stent era. Coronary stents have been used to treat suboptimal angiographic results (i.e., spot stenting) and dissections after balloon angioplasty of longer lesions, although the “full metal jacket” stent approach to diffuse disease is associated with a higher recurrence rate in the absence of complete stent expansion, particularly in smaller vessels. Overlapping sirolimus-eluting stents provide safe and effective treatment for long coronary lesions. However, stent length >35 mm remains a risk factor for restenosis and the need for revascularization. Longer stented lesions were associated with stent thrombosis in one series.

In a contemporary analysis of 10,004 patients undergoing surveillance, binary restenosis was detected in 2643 (26.4%) patients. Use of a first-generation DES or BMS (odds ratio [OR] = 0.35) and use of a second-generation DES or first-generation DES (OR = 0.67) were independent predictors of lower rates of restenosis. On multivariate analysis, smaller vessel size (OR = 1.59 for each 0.5-mm decrease), total stented length (OR = 1.27 for each 10-mm increase), complex lesion morphology (OR = 1.35), diabetes mellitus (OR = 1.32), and history of bypass surgery (OR = 1.38) were independently associated with restenosis and were similar across the spectrum of stent devices.

Bifurcation Lesions

Bifurcation lesions are common (up to 20% of all PCIs), and their percutaneous management is associated with higher rates of restenosis and thrombosis compared with nonbifurcation lesions. The risk of side branch occlusion in bifurcation lesions is related to the extent of atherosclerotic involvement of the side branch and geometry of the carina.

Several classification systems have been proposed ( Fig. 64.1 ). The Medina classification is the most commonly used and characterizes stenoses in the proximal parent vessel (0 = no disease; 1 = disease), distal vessel (0 = no disease; 1 = disease), and side branch (0 = no disease; 1 = disease). For example, a bifurcation lesion that involves the proximal and distal parent vessel and a side branch with a greater than 50% diameter stenosis is designated Medina 1,1,1, whereas a bifurcation lesion involving only the proximal parent vessel is designated Medina 1,0,0.

One stent usually is preferable to two stents in the parent vessel and side branch. If two stents are planned for the parent vessel and side branch due to complex bifurcation disease, several stenting techniques are possible, including simultaneous kissing stents, crush and double-kissing (DK) crush, culotte, T, and T-and-protrusion (TAP) stenting. Common to all of these strategies is a final kissing balloon inflation in the parent vessel and side branch, although a sequential two-step dilation of the branch vessel followed by the parent vessel may also be suitable. There is no consensus on which techniques are the optimal treatment strategy for bifurcation lesions.

The origin of the side branch rather than the parent vessel is the most common location of failure (recurrence) after bifurcation stenting, and one study of Medina 0,0,1 bifurcation lesions suggested that a two-stent strategy was associated with lower rates of clinical restenosis than a single-stent strategy. Other studies have suggested that the DK crush technique is superior to Culotte stenting for left main CAD and native vessel bifurcations.

Dedicated bifurcation stents were developed to provide adequate vessel coverage and side branch access, but they have been challenged with their higher profile and difficult delivery. A specifically designed side-branch stent has shown favorable results. Drug-eluting balloons have been used in bifurcation lesions with mixed results.

The angiographic analysis of bifurcation lesions is a challenging task. Although the visual assessment of these lesions is inaccurate, the standard quantitative coronary angiography (QCA) packages designed for single lesions cannot overcome the complexities of the bifurcation lesions. In an effort to address this shortcoming, dedicated bifurcation QCA algorithms have been developed, such as the Cardiovascular Angiography Analysis System (Pie Medical Imaging, Maastricht, The Netherlands) and QAngio XA (Medis, Leiden, The Netherlands). In a survey of experts in the field of bifurcation PCI, accuracy and precision of visual estimates of stenosis severity in phantom bifurcation lesions varied greatly and was less precise compared with the dedicated QCA algorithms, justifying the use of these software packages in clinical and research practice.

Total Occlusion

Total coronary occlusion is identified as an abrupt termination of the epicardial vessel; anterograde and retrograde collaterals may be helpful in quantifying the length of the totally occluded segment. Coronary occlusions are common findings and often lead to the decision to perform coronary bypass surgery rather than PCI in the setting of multivessel disease. The success rate for recanalization depends on the occlusion duration and on certain lesion morphologic features, such as bridging collaterals, occlusion length >15 mm, and the absence of a nipple to guide wire advancement. Although newer technologies have been used to recanalize refractory occlusions, better guidewires and operator and wire techniques have accounted for much of the improvement in crossing success. Simultaneous coronary injections are sometimes useful for identifying the length of the total occlusion ( Fig. 64.2 ). After the occlusion has been crossed, coronary stents have been used to provide the best long-term outcomes. DES usually are preferred to BMS.

The optimal technique for coronary revascularization is determined using four angiographic parameters: location of the proximal cap, occlusion length, existence of branches and size and quality of the target vessel at the distal cap, and suitability of collaterals for retrograde techniques. On the basis of these four characteristics, there has been a substantial improvement in the ability of the operator to secure access to the coronary vessel.

A key component to the assessment of total occlusion is definition of the collateral grades that provide blood flow to the jeopardized myocardium. The Rentrop classification system includes Rentrop grade 0 (no filling), grade 1 (small side branches filled), grade 2 (partial epicardial filling of the occluded artery), and grade 3 (complete epicardial filling of the occluded artery). Anatomic collaterals summarized by the 26 potential pathways were consolidated into four groups: septal, intra arterial (bridging), epicardial with proximal takeoff (atrial branches), and epicardial with distal takeoff. The size of the collateral connection can be quantified as group 0 (no continuous connection between donor and recipient artery), group 1 (continuous threadlike connection ≤0.3 mm), or group 2 (continuous small, branchlike collateral through its course ≥0.4 mm).