Advanced age is an important predictor of outcome following PCI (2, 3). Part of this increased risk is secondary to a higher likelihood of multivessel coronary artery disease and increased prevalence of comorbid conditions (4). A major limitation to understanding the impact of age in PCI outcomes is due to the exclusion of patients >75 years of age from most large, randomized studies. In 2003, the National Heart Lung and Blood Institute’s (NHLBI) Dynamic Registry identified that when compared with younger patients (<65 years old), patients older than 80 were more likely to have 3-vessel coronary disease (20% vs. 38%), a higher burden of coronary calcification (20% vs. 40%), and more commonly required multivessel PCI (30% vs. 40%) (5). A recent analysis in 2012 reported significant reductions in target vessel revascularization (TVR) and less repeat revascularization after PCI with drug eluting stents (DES) with no increase in death or MI, regardless of age (6). Further analysis of data involving more than 8,000 patients with a mean age of 83 from the American College of Cardiology-National Cardiovascular Data Registry (ACC-NCDR) who underwent PCI reported good procedural success and acceptable mortality (4). Multisite PCI was performed in 35%, and angiographic success was high (93%), with stents placed in 75% of patients. Indications for PCI in this registry were most commonly acute MI (33%) and unstable angina (56%), rather than elective stable angina (11%).

Elderly patients are more likely to have complications associated with PCI. A prospective registry study of patients >85 years of age reported higher rates of unadjusted mortality (6.93% vs. 1.20%), postprocedural MI (4.46% vs. 2.74%), renal, neurological, and access-site complications in the very elderly compared with patients <85 years of age (7). In this study, death occurred predominantly in very elderly patients undergoing nonelective PCI. Further data from the ACC-NCDR reported that elderly (>65 years) patients undergoing PCI had a higher in-hospital (4.6% vs. 0.6%) and 1-year mortality (11% vs. 2.1%) compared with younger patients (4). In-hospital mortality was further increased if the patients were admitted with acute MI (reaching 14%) and was as high as 43% if they were admitted with cardiogenic shock. Taken together, elderly patients are at higher risk of adverse outcomes after PCI, and decisions to proceed with PCI should take into account all prognostic variables. However, when medical therapy fails to control anginal symptoms, PCI can be performed with a high success rate, resulting in improved quality-of-life and health status (8, 9).

Over 24 million individuals worldwide and 7 million individuals in the United States alone suffer from heart failure with >500,000 new cases diagnosed per year. As a result, approximately 5% to 15% of patients referred for PCI will have a LV ejection fraction (LVEF) <40%. Several studies have reported increased mortality, nonfatal MI, stent thrombosis, and target lesion revascularization after PCI in patients with decreased LVEF (10, 11, 12, 13 and 14). In a large retrospective cohort study of over 55,000 patients, a reduced LVEF was directly associated with a stepwise increase in in-hospital mortality and major adverse cardiac events (MACE) (15). Similarly, O’Keefe et al. reported that among 700 patients followed longitudinally, LVEF <40% is a predictor of both in-hospital mortality and survival at 5-year follow-up compared with subjects with preserved LV function (89% vs. 81%, respectively for survival, p =0.05) (16).

In general, the benefit of revascularization among patients with LV dysfunction is best achieved in patients with documented myocardial viability (17). Furthermore, a more complete revascularization approach appears to be associated with improved clinical outcomes in subjects with low LVEF and multivessel disease (18). Historic studies have reported that in patients with low LVEF and disease involving the left anterior descending (LAD) artery, CABG was associated with more complete revascularization, improved LV function, fewer cardiac events, and fewer TVRs. However, no
effect on mid-term survival was observed (19). The STICH (Surgical Treatment for Ischemic Heart Failure) trial examined patients with a LVEF <35% treated with coronary artery bypass grafting (CABG) or optimal medical therapy and reported similar rates of death from any cause after 5 years of follow-up (20). However, CABG was superior to medical therapy for secondary endpoints, including death from any cause or recurrent hospitalizations. Final analysis of the STICH trial at 10 years follow-up is currently pending. Studies directly comparing PCI and CABG among patients with low LVEF are rare. The Bypass Angioplasty Revascularization Investigation (BARI) study reported that survival among patients with three-vessel disease and reduced LV function (mean LVEF 41 ± 6%) was nonsignificantly different between CABG and PTCA (70% vs. 74%, p = 0.6) (21). When diabetics were excluded from the analysis, survival was not significantly different between CABG and PTCA, irrespective of LVEF.

Diabetic patients represent 20% to 25% of patients undergoing revascularization, and they are more likely to have diffuse coronary artery disease, multivessel involvement, and LV dysfunction than nondiabetic patients (22, 23). In particular, diabetics more often have multi-vessel disease (MVD) compared with nondiabetic patients (44.7% vs. 25.4% p = 0.002) (23). The BARI trial compared Percutaneous transluminal coronary angioplasty (PTCA) with CABG among patients with MVD and identified better 5-year survival with CABG (with at least one internal mammary conduit) than PTCA among diabetic patients (80.6% vs. 65.5%) (24). This observation led the ACC/AHA committee to designate a Class IIb indication for PCI among patients with MVD with significant proximal LAD and treated diabetes (1). In 2009, BARI-2D studied the effect of prompt revascularization by either PCI or CABG with intensive medical therapy (with or without insulin sensitization or insulin-provision therapy). At 5 years of follow-up, no difference in the rate of death and a composite of death, MI, or stroke (major cardiovascular events), was observed between PCI and medical therapy. MACE was lower after CABG compared with medical therapy (25). However, the ARTS (Arterial Revascularization Therapies Study) trial randomized 208 patients with diabetes to either PCI with stenting or CABG and reported no significant difference in 5-year survival between the two groups (26). Use of DES was associated with reduced restenosis among different cohorts of patients, including diabetic patients. Early data suggest a reduction in revascularization seen with the sirolimus- and paclitaxel-coated stents in diabetic patients with near equivalency to that seen in the surgical arms. It remains unclear whether these findings will hold in randomized trials, and more information is needed before reaching a more definite conclusion. The equivalency of percutaneous versus surgica l revascularization is being re-evaluated in the FREEDOM trial (Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease), which is a multicenter, prospective randomized trial comparing CABG with PCI stenting using sirolimus-eluting stents in diabetic patients with multivessel disease (27).

Among patients with significant renal insufficiency, >70% will have MVD on angiography (28, 29). Patients with severe renal insufficiency are more likely to be older, female, diabetic, and hypertensive. In addition, they more frequently have MVD, vein graft disease, complex lesions, and had been more frequently treated with rotational atherectomy in past registries. In a detailed analysis of renal insufficiency and cardiovascular outcome from the Mayo Clinic PCI registry and the PRESTO (Prevention of REStenosis with Tranilast and its Outcomes) trial, 1-year mortality after PCI was 1.5% among those with a creatinine clearance ≥70 mL/min (n = 2,558), 3.6% for a clearance of 50 to 69 mL/min (n = 1,458), 7.8% for a clearance of 30 to 49 mL/min (n = 828), and 18.3% for a clearance of <30 mL/min (n = 141) (30, 31). In the setting of PCI, chronic renal insufficiency impacts long-term outcome with increased risk of cardiovascular mortality and concern of worsening renal function with contrast-induced nephropathy. Pharmacotherapy dose adjustment is needed with renal insufficiency, especially for several anticoagulants, including small molecule GP IIb/IIIa inhibitors, low-molecular weight heparin, and direct thrombin inhibitors. Unfractionated heparin may be a better choice compared with low-molecular-weight heparin, because of easier monitoring and adjustment using activated partial thromboplastin time or activated clotting time. To avoid contrast-induced nephropathy, patients with chronic renal insufficiency should be considered for staged procedures.

Historically, the Veterans Administration Cooperative study reported improved survival with surgical revascularization versus medical therapy for LM disease (35). Early experience with balloon angioplasty alone was complicated by coronary dissection, abrupt closure, and restenosis, thereby confirming the benefits of surgery (36). The unprotected LM (UPLM) registry (ULTIMA) reported improved outcomes with coronary stenting and atherectomy and further demonstrated higher risk if LM PCI was performed in the presence of a LVEF <30%, severe mitral regurgitation, cardiogenic shock, renal insufficiency, or severe coronary calcification (37). A 1-year mortality rate of 20%-24% may have been due to a selection bias toward “inoperable” patients with severe comorbidities. This observation is supported by a lower mortality rate of 3.4% among low-risk patients characterized by age <65, LVEF > 30%, and no evidence of cardiogenic shock. Several studies have since compared
PCI with CABG in the setting of UPLM disease. The MAIN COMPARE multicenter registry was a nonrandomized study of 2,240 patients with UPLM receiving CABG (n = 1,138) or PCI (n = 1,102) with either a drug-eluting (71%) or bare-metal stent (29%) (38). At 3-years follow-up, mortality (HR, 1.02; 95% CI: 0.74-1.39) and composite death, Q-wave MI, or stroke were comparable between groups. TVR was more common after PCI compared with CABG. Similar 3-year outcomes of death, MI, and composite death/MI/stroke were reported by the SYNTAX randomized trial of 1,800 patients receiving CABG or PCI with a DES (39). In this study, a prespecified subgroup of 705 UPLM patients was analyzed. TVR was again higher after PCI compared with CABG (20% vs. 11.7%, p = 0.004). In 2011, the PRECOMBAT trial randomized 600 patients with UPLM to DES-PCI versus CABG. One-year MACCE was noninferior for PCI compared with CABG (8.7% vs. 6.7%, p = 0.01 for noninferiority) (40). Ischemia-driven revascularization was higher after PCI than CABG (9% vs. 4.2%, p = 0.02). Despite these emerging datasets, the standard of care for patients with LM disease remains surgical revascularization. The ACC/AHA guidelines designate a Class I for bypass surgery for patients with significant LM, Class III for percutaneous revascularization if they are acceptable candidates for surgery, and Class IIb for percutaneous revascularization if they are deemed high risk for surgery (1).