ABC | Volume 112, Nº5, May 2019

Original Article Andrade et al Stent versus CABG: a meta-analysis Arq Bras Cardiol. 2019; 112(5):511-523 Figure 4 – Stent versus CABG in left main coronary artery disease: one-year mortality (top) and long-term mortality (bottom). The size of each box is proportional to the number of patients of the trial. The bars represent 95% confidence interval. The diamond represents the syntheses of results. ULMCAD: unprotected left main coronary artery disease. CABG: coronary artery bypass graft. LE MANS: Left Main coronary artery stenting study; SYNTAX LEMANS: subgroup of ULMCAD of SYNTAX (Synergy between PCI with Taxus and Surgery); PRECOMBAT: Premier of Randomized Comparison of Bypass Surgery versusAngioplasty Using Sirolimus-Eluting Stent in Patients with Left Main Coronary Artery Disease; EXCEL: Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization; NOBLE: Nordic-Baltic-British Left Main Revascularization Study. Boldriot: Boldriot et al: J Am Coll Cardiol. 2011; 57: 538-545. –0.2 –0.1 0.1 0.2 0 Favours stent Favours CABG –0.2 –0.1 0.1 0.2 0 Favours stent Favours CABG 2349 2351 100.0% –0.01 [–0.02, 0.00] 2193 2171 100.0% 0.00 [–0.02, 0.02] 2 5 101 4.3% 40.5% 2.2% 25.2% 12.8% 15.0% 957 53 592 300 348 38 4 17 8 15 100 948 52 592 300 357 38 1 9 6 15 71 87 71 53 903 53 592 275 348 16 33 23 50 175 913 52 592 279 357 11 36 17 42 177 –0.03 [–0.08, 0.02] 0.00 [–0.02, 0.02] –0.06 [–0.14, 0.02] –0.01 [–0.03, 0.00] –0.01 [–0.03, 0.02] –0.00 [–0.03, 0.03] 0.02 [–0.00, 0.04] –0.09 [–0.26, 0.08] 0.01 [–0.02, 0.03] –0.02 [–0.07, 0.02] –0.03 [–0.08, 0.02] 41.6% 2.4% 27.1% 12.7% 16.2% Total (95% Cl) Total events Test for overall effect: Z = 1.29 (p = 0.20) Heterogeneity: Chi 2 = 3.64, df = 5 (p = 0.60); I 2 = 0% Total (95% Cl) Total events Test for overall effect: Z = 0.01 (p = 1.00) Heterogeneity: Chi 2 = 5.99, df = 4 (p = 0.20); I 2 = 33% Boudriot et al 2011 EXCEL 2016 SYNTAXLM 2013 LE MANS 2008 NOBLE 2016 PRECOMBAT 2011 EXCEL 2016 SYNTAXLM 2013 LE MANS 2008 NOBLE 2016 PRECOMBAT 2011 One year mortality Late mortality Study or Subgroup Stent CABG Risk Difference Risk Difference Events Total Events Total Weight M-H, Fixed, 95% Cl M-H, Fixed, 95% Cl results (HR 1.16, CI 0.51-264, p = 0.49). Considering this, we believe that PCI indications for LMCAD in AHA/ACC guidelines may be modified in near future. Diabetic patients are a present challenge for PCI. A more diffuse atherosclerotic disease is a possible explanation for the worse comparative results of PCI in this population. Our results suggest that there is still a superiority of surgery over PCI in long-term mortality, even in the DES era. There is a hypothesis that the greater mortality of PCI compared to CABG in diabetic patients may be attributed to the presence of more complex lesions in diabetic patients and, not to the metabolic disturbance. The fact that in the subgroup analysis of MACCE results (Figure 5) the meta-Adjusted value was 0.09 (adjusted for SYNTAX score) supports this hypothesis. This review was not aimed to compare the results of BMS and DES for several reasons: in BMS trials patients had less complex angiographic lesions (average of 2.3 stents per patient in ARTS and SOS trials versus 3.8 stents per patient in SYNTAX, FREEDOM, BEST, PRECOMBAT and CARDia trials) and had a small percentage of diabetic patients. Otherwise, medical adjunctive treatment and results of surgery for patients with failed PCI also evolved. But the good comparative results of PCI in BMS era trials suggest that for patients with less complex lesions, or patients with unstable angina (ERACI II trial) or high surgical risk (AWESOME trial) initial PCI is a good alternative to CABG. In terms of major adverse composite events, the analysis of subgroups showed that diabetes and a high SYNTAX score were the most important factors to influence adversely the results of PCI. Presence of left ventricular dysfunction did not influence the results, but the number of patients with this finding was small. A high SYNTAX score was an independent risk factor for adverse outcomes, even when adjusted for diabetes, but diabetes was not an independent risk factor for adverse outcomes when adjusted for SYNTAX score. In the present review despite the clear superiority of CABG in the outcome of new revascularization, it is possible to notice the progressive improvement of PCI results. This was particularly striking when we consider the outcome “new revascularization by alternative procedures”, in which there was a tendency to superiority of PCI in the DES era. The evidence presented here should be used to inform patients, helping them in choosing the more adequate form of revascularization in multi-vessel and LMCAD. Some patients may prefer having PCI to avoid the higher morbidity and short-term mortality of surgery. Other patients may put greater emphasis on the superiority of surgery regarding long-term mortality. However, PCI using second generation DES may still be considered as an alternative to CABG, 518