Overall, this implies that CABG mainly applies to patients who are expected to have a long longevity (e.g., > 3–4 years).
Near consistent increase in the risk of stroke with CABG (by an absolute ~2%).
Patients with prior stroke were essentially excluded from these trials. In fact, very few patients had carotid disease. Yet, despite this, CABG led to a higher risk of stroke than PCI, which is expected to be even higher in patients with prior stroke or carotid disease.
Randomized patients were relatively young (median age ~65).
Thus, these trials do not necessarily prove the superiority of CABG in a sicker population: older patients, higher prevalence of stroke, lower expected longevity, severe LV dysfunction or HF. In SYNTAX trial, elderly patients (>70) had similar 5- and 10-year mortality with CABG and PCI, and even similar combined cardiovascular events, despite a high mean SYNTAX score of 30, while patients <70 had a reduction in mortality and events with CABG.88,89 Similarly when an older but also sicker population was randomized to CABG vs. PCI in the VA-AWESOME trial, the 6-month survival was better with PCI, and the 5-year survival was equivalent with both strategies.90 In the VA-AWESOME trial, patients had at least one of the following: age >70 (50%), MI <7 days (33%), EF ≤35% (20%), prior CABG (33%). Thus, PCI still has a role in ill patients with multivessel CAD and comorbidities who are deemed at a high surgical risk.
X. High-surgical-risk patients
In a patient with multivessel disease where CABG or PCI is considered, two risk scores are used: (i) SYNTAX score to determine PCI outcomes, and (ii) STS score or EuroSCORE to assess surgical mortality (Table 3.4).91,92 The average CABG mortality is 1.5-2%, but increases to >5% with the combination of several risk factors, especially combined CABG and valve surgery, redo CABG, age >80, prior disabling stroke, severe lung disease, acute MI, HF or hemodynamic compromise. Patients who are at a high surgical risk, much higher than the 2% seen in randomized trials (e.g., > 5%), may be better served with PCI (class IIa for low-SYNTAX left main disease, IIb for low-SYNTAX 3-vessel CAD). Thus, in the era of an aging population with comorbidities, multivessel complex PCI still has a role. A heart team discussion is warranted for these patients.
In addition to high-surgical risk patients, patients with small vessels and diffuse distal disease that is severe or calcified may not have appropriate distal targets for CABG and may not be CABG candidates, especially when the LAD cannot be grafted. They may undergo PCI of focal, critical, proximal disease. A third reason that may preclude CABG is the lack of conduits, in particular venous conduits in patients with large varicose veins and venous insufficiency.
The use of LIMA necessitates surgical dissection through the pleural cavity, with a high risk of pleural effusion and deterioration of pulmonary function in patients with severe lung disease.
XI. Role of complete functional revascularization
Complete revascularization is defined as revascularization of all functionally significant stenoses in vessels ≥1.5 mm supplying viable territories. However, this has been defined differently across studies, and some based it on angiographic disease >50–70% rather than on functionally significant disease, or on achieving revascularization of the three major epicardial vessels (as opposed to all branches).93 CABG generally achieves more complete revascularization than PCI, as it is less affected by lesion complexity (e.g., CTO) (67% vs. 53% in SYNTAX trial).94 In most registries and post-hoc analyses, incomplete revascularization was associated with impaired outcomes only after PCI (New York and ARTS registries).95,96 This adverse outcome may be related to the residual disease itself, or more so, to the fact that residual disease is a marker of more extensive and aggressive atherosclerosis even across the revascularized arteries, which explains why incomplete revascularization is more unfavorable after PCI, which only treats focal disease, than after CABG. In fact, the intense pursuit of complete revascularization may not, by itself, improve outcomes. In two analyses of CABG patients, incomplete revascularization of a small/poor target RCA or LCx in patients receiving LIMA-to-LAD graft did not adversely affect long-term outcomes; in fact, too aggressive revascularization of >1 non-LAD vessel may be associated with worse outcomes.97,98 This is called “reasonable” incomplete revascularization. This reasonable revascularization concept fits with the functional revascularization concept where only large and ischemic territories are revascularized. Incomplete revascularization usually has a worse connotation with PCI than with CABG, as (i) PCI is a suboptimal therapy for extensive disease, and (ii) PCI more frequently omits large, otherwise graftable vessels because of technical challenge, such as CTO.
XII. Hybrid CABG–PCI
The superiority of CABG mainly results from the longevity of the LIMA-to-LAD graft. DES stenting of non-LAD vessels, when feasible (e.g., no CTO, no heavily calcified disease), is likely associated with equal or superior results to placement of venous grafts, ~20% of which occlude by the first year. The so-called hybrid strategy consists of performing LIMA–LAD using off-pump CABG (beating heart), followed by DES PCI of the remaining stenoses during the same hospitalization (hours or days later, with clopidogrel loading after CABG). Occasionally, in patients with critical non-LAD disease, non-LAD disease is stented first, followed by performance of off-pump CABG under clopidogrel therapy.
Table 3.4 Variables analyzed in surgical risk scores (STS and EuroSCORE).
| Underlying patient-related factors and comorbidities (i) age (the risk doubles with every 10 yrs >60); (ii) women (~50% higher risk than men); (iii) moderate or severe COPD; (iv) severe CKD; (v) prior disabling stroke or neurological illness; (vi) carotid disease, PAD Underlying EF–HF (i) EF (<50%: mild score; < 30%: severe score); (ii) HF functional class IVPrior cardiac surgery, especially redo CABGIsolated CABG <isolated valve surgery <CABG + valve surgeryCurrent cardiac status(i) recent MI; (ii) unstable vs. stable angina; (iii) hemodynamic or electrical instability and requirements for IABP and/or inotropes |
The off-pump LIMA-to-LAD surgery offers the advantage of avoiding aortic manipulation, which is necessary during on-pump CABG or during SVG grafting (whether off- or on-pump). Also, the shorter surgical time, the lower blood loss, and the avoidance of cardioplegia and cardiopulmonary bypass are advantages.
Currently, the hybrid approach is particularly applicable to: (i) patients with heavily calcified, porcelain aorta in whom aortic manipula- tion needs to be avoided; and (ii) patients with good LAD target but poor LCx and RCA targets that, nonetheless, have severe proximal disease amenable to PCI.
XIII. Enhanced external counterpulsation (EECP)
This therapy consists of inflating cuffs around the lower extremities during diastole, and deflating them in systole, creating an effect similar to IABP. The systolic cuff depression reduces LV afterload and O2 demands. It may also have a sustained effect on endothelial function, collateral function, and oxidative stress, explaining the sustained benefit. EECP consists of 35 one-hour sessions. Non-randomized and limited randomized data suggest a reduction of angina and a reduction of ECG and nuclear ischemia with this therapy (class IIb for refrac
tory angina).12
XIV. Mortality in CAD (Table 3.5)
Note on left main disease – Medical therapy has not been addressed as a standalone therapy for left main disease in any contemporary trial. Old studies such as the VA-CABG trial and the CASS registry have shown that the mortality of medically treated left main disease, whether symptomatic