Intervention for Coronary Chronic Total Occlusions

Chronic Total Coronary Occlusion

Chronic total coronary occlusions (CTOs), along with major bifurcations (including the left main coronary artery) and long lesions, constitute a disease type that remains a major technical challenge for interventional cardiologists individually, and represents a significant limitation for interventional cardiology as a field. Percutaneous coronary intervention (PCI) is still predicated upon a steerable guidewire spanning the target lesion and acting as a rail over which therapies are delivered and as a “safety line” in case of target-vessel complications. The essential challenge of CTO PCI lies in the initial traversing of the lesion with a guidewire. Length of the occlusion, as well as presence of calcification and suture line fibrosis after graft failure, further increase the complexity of CTO PCI.

In the past few years, important advancements have been made in techniques that safely and more predictably overcome this essential challenge. Dissemination of these techniques along with increasing risk profiles that make patients poor candidates for cardiac surgery has led to an accelerated adoption of CTO PCI into mainstream interventional practice. Most major PCI programs now have, or are planning for, specialized competence in contemporary CTO procedures. An understanding of the field has become essential knowledge for all interventionalists and arguably for any cardiologist who advises patients regarding revascularization options. This chapter strives to provide that understanding.

Definition

A CTO is most often defined as a coronary occlusion known to be present for 3 or more months or a newly documented occlusion not attributable to a similarly recent ischemic event. The term occlusion is most accurately reserved for a stenosis within which there is neither a continuous visible lumen nor any visible antegrade flow that cannot be accounted for by collaterals (Thrombolysis in Myocardial Infarction [TIMI] grade 0). Lesions with trace antegrade lumenal flow (TIMI grade 1) are referred to as functional or subtotal occlusions. Lesions with a residual lumen but without antegrade flow because of competing collateral flow are referred to as pseudo-occlusions . Functional and pseudo-occlusions overlap in practice and may be difficult to distinguish from a true CTO unless imaged with dual-catheter angiography or probed with a guidewire. Finally, spontaneously recanalized CTOs are sometimes identified, wherein a long, tortuous microchannel exists within the presumed architecture of a previous occlusion.

Detection of coronary occlusions remains the domain of catheter-based angiography. Current coronary computed tomography (CT) angiography has limited spatial resolution and does not capture the temporal information required to determine flow rate or directionality. As such, it cannot readily be used to distinguish a CTO from a functional or pseudo-occlusion or even reliably from a high-grade stenosis with preserved flow.

Prevalence

Reports that describe the prevalence of CTOs vary widely depending on the cohort studied. Moreover, unless developed specifically for the study of CTOs, angiographic and interventional databases generally do not contain data fields that reliably distinguish subacute from chronic occlusions, true occlusions from subtotal or pseudo-occlusions, or potentially bypassed from nonbypassed occlusions. Thus the literature reveals an inconsistency of methods for determining the true prevalence of CTOs in the population.

A retrospective study from a large U.S. Veterans Administration (VA) center extracted patients with at least one 70% stenosis (by visual estimate), no prior coronary artery bypass grafting (CABG), and no myocardial infarction (MI) within 3 months from 8004 consecutive patients undergoing cardiac catheterization between 1990 and 2000. Within the derived cohort of 3087 patients, 52% (1612) had a CTO.

In a prospective study, Fefer and colleagues reported data on more than 14,000 patients undergoing nonemergent angiography at three tertiary Canadian centers in 2008 and 2009. They found nonacute coronary occlusions present in 14.7%. After excluding patients with prior CABG, as well as those without significant CAD, they reported that 18.4% of the remaining 7680 patients had at least one nonacute occlusion.

A prospectively collected population-based Swedish registry of nonacute general angiography and PCI found a similar CTO prevalence of 16% in over 91,000 patients with significant coronary artery disease (CAD). Like the Canadian registry, this prevalence was calculated after exclusion of those with prior CABG. An unexpected finding was a decline in CTO prevalence over the 8-year period examined, from 17.2% in 2005 to 15.1% in 2012.

pathology

The histopathology of CTOs is an evolving area that has grown from necropsy and ex vivo tissue analyses to include the use of intravascular micro–computed tomography (MCT), micro–magnetic resonance imaging (MMRI), and intravascular ultrasound (IVUS). Studies characterizing the histology of CTOs have led to further insights regarding procedural success and failure.

Postnecropsy samples of CTOs consistently demonstrate preservation of vessel architecture; a multilayered structure in which the intima and neointima (including atherosclerotic plaque) is distinguishable from the muscularis and adventitia. Importantly, the external elastic lamina remains intact. This preservation of architecture is a fundamental feature enabling percutaneous recanalization ( Fig. 26.1 ). Early necropsy studies to examine human CTOs have shown that angiographic characteristics, such as proximal cap morphology, correlate with histology. Tapered-tip occlusions contain areas of luminal recanalization with microchannels and loosely packed fibrous tissue. Such occlusions also tend to be shorter. Conversely, lesions with a blunt cap are typically composed of densely packed fibrous material and fibrocalcific intimal plaque and tend to be longer.

Studies in mice, porcine, and rabbit femoral arterial occlusion models have provided experimental insights into the development of CTOs. Recent thrombotic occlusions display an inflammatory infiltrate dominated by neutrophils and mononuclear cells that penetrate the occlusive thrombus. The density of this infiltrate peaks at 2 weeks after occlusion and declines thereafter. Proteoglycan-enriched extracellular matrix replaces thrombus; deposition occurs early, between weeks 2 and 6 after occlusion. It is concentrated at the proximal and distal ends of the lesion. Over time, these areas are replaced by densely packed collagen. Negative remodeling of the surrounding vessel and variable decay of the internal elastic lamina accompany this process. These experimental observations may in part explain the clinical paradigm of mechanically resistant proximal and distal caps.

Human and animal necropsy studies and advanced imaging studies have confirmed the frequent presence of recanalization microvascular channels (microchannels) within chronically occluded segments. Microchannels in human CTOs were first described by Katsuragawa et al., wherein small vascular channels ranging from 160 to 230 μm in diameter were noted at both proximal and distal lesion segments. This observation led to speculation that microchannels might consistently provide a through-and-through route for fine guidewires to track. Subsequent human and experimental pathologic studies suggested these microchannels are more typically configured as corkscrews or sharply angulated channels that become fragmented and discontinuous as occlusions age and typically do not traverse the full length of an occlusion. Thus, while microchannels are unlikely to provide a continuous channel for guidewire passage, especially in longer CTO segments, they may still render a “path of low resistance” through the CTO segment tissue.

Clinical Profile and Presentation

When compared with a cohort of CAD patients without CTO, patients with CTO display important differences in baseline characteristics that point toward more advanced atherosclerosis. Patients are on average 1 to 2 years older, more likely to be male, and more likely to have diabetes and hypertension. Multivessel CAD is more commonly present, as is peripheral arterial disease and cerebrovascular disease.

A twofold excess in history of prior MI was observed among CTO patients in both Canadian and Swedish registries (40% vs. 23% and 37% vs. 17%, respectively, P < .01 in both). However, the relationship between a CTO, the presence and extent of infarction in the subtended territory, and the degree of left ventricular (LV) dysfunction attributable to the CTO is not well established. Half of the patients in the Canadian registry had a left ventricular ejection fraction (LVEF) less than 50%, but significant Q-waves were present in only 32% of right coronary arteries (RCAs), 13% of left anterior descending (LAD) coronary arteries, and 26% of left circumflex (LCX) artery occlusions. In the VA registry, chronic CAD patients with CTO had a significantly lower LVEF than those without a CTO (LVEF 53% vs. 60%, P < .0001) but also had significantly more multivessel CAD (66% vs. 42%, P < .0001).

Despite the long-term presence of a CTO, an acute coronary syndrome (ACS) arising in other coronary segments is a common trigger for cardiac catheterization that leads to CTO detection. ACS immediately preceded CTO detection in 46% of the Canadian cohort and 40% of the Swedish cohort. Conversely, in a prospective single-center registry of nearly 3300 consecutive primary PCI procedures for ST-elevation MI (STEMI) the prevalence of a preexisting CTO was 12.8%. Patients with a CTO detected at primary PCI had significantly worse prognosis than patients with single-vessel CAD (infarct vessel only) or those with multivessel CAD without a CTO.

Indications for Revascularization

Published indications for CTO revascularization largely mirror the indications for revascularization in otherwise obstructive nonocclusive stable CAD. In this regard, the primary indication for CTO PCI in the setting of single-vessel CAD is the relief of ischemic symptoms that persist despite antiischemic medical therapy. However, the frequency with which PCI is offered to patients with a CTO (but otherwise limited CAD not warranting CABG) is much lower than for obstructive nonocclusive CAD in practice and moreover varies widely among physicians and PCI centers.

Multicenter U.S. registries of patients treated in the 1990s indicate that only 8% to 15% of patients with a CTO at angiography underwent PCI. The subsequent prospective VA registry found that the presence of a CTO at angiography was an independent predictor of a physician recommending CABG or medical therapy as opposed to PCI. In the Canadian CTO registry that enrolled patients in 2008 and 2009, PCI was undertaken in only 10%. Importantly, there is wide practice variation, with as few as 1% and as many as 16% of patients with CTOs treated with PCI ( P < .001), despite similar rates of CABG at all centers (22% to 28%, P = not significant [NS]). The low and variable use of PCI for CTO likely reflects many factors, among them a wide variation in technical expertise of operators and lack of evidence of benefit of CTO PCI on hard outcomes such as death and MI.

Neither American nor European guidelines distinguish between CTOs and obstructive nonocclusive lesions with respect to the threshold for undertaking revascularization, whether by PCI or CABG. The European guidelines consider CTO PCI a complex procedure that demands experienced operators at centers with specialized CTO equipment and access to circulatory support and cardiac surgery. Ad hoc CTO PCI (coincident with diagnostic catheterization) is discouraged. The American College of Cardiology (ACC)/American Hospital Association (AHA)/Society for Cardiovascular Angiography and Interventions (SCAI) guidelines state that “(CTO PCI) in patients with appropriate clinical indications and suitable anatomy is reasonable when performed by operators with appropriate expertise” (class IIa, level of evidence B).

The updated 2017 Appropriate Use Criteria for Coronary Revascularization in Patients With Stable Ischemic Heart Disease from the American Societies no longer distinguish between appropriate indications for CTO PCI and PCI of obstructive nonocclusive lesions. Asymptomatic patients are classified as “inappropriate” in many scenarios, but may be appropriate in the setting of high risk findings on noninvasive testing, including severe LV dysfunction. Given these recommendations, we advise caution in pursuing CTO PCI in asymptomatic patients, with careful review for these special indications.

Ischemia and Left Ventricle Function

The burden of ischemia in a territory subtended by a CTO is presumably determined by the left ventricular mass supplied by the occluded segment, the degree of infarction in that zone, the degree of collateralization, and the severity of coronary obstruction, if any, in vessels that supply those collaterals. Because the intracoronary pressure beyond an occlusion is inherently low, the myocardium subtended may be especially vulnerable to subendocardial ischemia that is exacerbated by elevation of end-diastolic pressure. No systematic study has been done to quantify the degree of ischemia in an unselected cohort of patients with CTO. However, in a study of 569 consecutive patients with multivessel disease undergoing myocardial scintigraphy after incomplete revascularization by PCI, 126 had at least one residual CTO. Of these, 64 patients (50.8%) had a severely abnormal scan (with summed stress score >8). There was a linear relationship between the global summed stress score and the summed stress score of the territory subtended by the CTO. At a median follow-up period of 44 months, summed stress score was incrementally and significantly associated with hard (cardiac death and MI) and soft (unstable angina and repeat PCI procedures) end points.

The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial failed to show a difference in death or MI in patients with CAD who had stable angina and myocardial ischemia randomized to an initial strategy of PCI versus medical therapy. While CTO patients were excluded from the study, angina was improved in the PCI group. There was also significant crossover in the medical therapy arm, with 32.6% of patients undergoing revascularization with PCI or CABG at 4.6 years. A substudy that evaluated serial nuclear imaging at baseline and 1-year postrandomization revealed that PCI with optimal medical therapy (OMT) led to greater ischemia reductions compared with OMT alone (33% vs. 19%; P = .0004), especially patients with moderate to severe pretreatment ischemia (78% vs. 52%; P = .007).

The International Study of Comparative Health Effectiveness with Medical and Invasive Approaches Study (ISCHEMIA) trial is currently underway, and randomizing patients with moderate to severe ischemia to revascularization and OMT versus OMT alone may contribute to our understanding of ischemia-guided revascularization including patients with CTOs.

The relationship between CTO, ischemia, regional and global LV dysfunction, and post-PCI LV recovery is complex and incompletely characterized. Patients undergoing primary PCI for STEMI who are also found to have a CTO in a noninfarct-related vessel are significantly more likely to have moderate or severe LV impairment (28% vs. 17%, P < .01) and are more than twice as likely to suffer progressive LV impairment at follow-up (39% vs. 17%, P < .01) compared with patients who do not have a CTO.

The available reports exploring the relationship between CTO revascularization and LV function do not include medically treated comparators. With this important limitation, patients with regional LV dysfunction in whom the abnormal wall segment is subtended by the CTO commonly demonstrate regional LV improvement after CTO PCI. Improvement in global LVEF following CTO PCI is less easily demonstrated, in part because more than half of CTO patients have normal baseline ejection fraction (EF). In the Total Occlusion Study of Canada (TOSCA), 244 patients had paired baseline and 6-month core laboratory evaluated LV angiograms, and 106 of these (43%) had an impaired EF at baseline (EF <60%, mean 46% ± 9%). EF did not change at follow-up among those with normal baseline EF but rose 3.8% (± 8.4%) in the cohort with impaired baseline EF ( P < .001). The direction and magnitude of these EF changes are consistent with those observed in smaller series.

Further, in patients with akinetic segments or significant LV dysfunction and remodeling, myocardial viability should be considered prior to CTO PCI. Multiple studies have linked viability to improvement in regional function following PCI. While its use is controversial in the face of two negative trials, they are limited by significant crossover and lack of adherence to viability results.