Author + information
- Published online May 2, 2018.
- Adrian W. Messerli, MD∗ ( and )
- Naoki Misumida, MD
- Gill Heart Institute and Division of Cardiovascular Medicine, University of Kentucky, Lexington, Kentucky
- ↵∗Address for correspondence:
Dr. Adrian W. Messerli, University of Kentucky HealthCare, University of Kentucky Cardiovascular Medicine, CT Wethington Building, Room 324B, 900 South Limestone, Lexington, Kentucky 44122-5037.
Since its introduction in mid-1990s, hybrid coronary revascularization (HCR) has been advocated to be a promising alternative revascularization strategy to conventional coronary artery bypass grafting (CABG) for selected patients with multivessel coronary artery disease (CAD) (1). The aim of HCR is to combine the respective strengths of CABG and percutaneous coronary intervention (PCI), namely, the long-term durability of a left internal mammary artery (LIMA) graft to the left anterior descending coronary artery (LAD) with transcatheter revascularization of non-LAD lesions with drug-eluting stents. Multiple observational studies have been published, but these are restricted by size, dissimilar hybrid approach, limited follow-up length, and inherent selection bias. The long-term efficacy of HCR versus conventional CABG (or multivessel PCI) still needs to be addressed in large-scale randomized clinical trials.
The primary 1-year results of the only randomized trial comparing HCR and conventional CABG, POL-MIDES (Safety and Efficacy of Hybrid Revascularization in Multivessel Coronary Artery Disease), were published in 2014 (2). A total of 200 consecutive patients with multivessel CAD involving the LAD and >70% lesions in at least 1 major non-LAD artery amenable to both PCI and CABG were randomized to either HCR (n = 98) (minimally invasive direct coronary artery bypass + everolimus-eluting stents) or conventional CABG (n = 102). Patients with left main involvement or >1 chronic total occlusion were excluded; the mean SYNTAX (Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery) score was 23. HCR was completed for 93.9% of patients, whereas conversion to standard CABG was required for 6.1%. At 1 year, both groups had similar all-cause mortality (2.0% with HCR vs. 2.9% with CABG; p = 0.10).
In this issue of JACC: Cardiovascular Interventions, Tajstra et al. (3) report the 5-year results of the POL-MIDES trial. Again, no statistically significant difference was found between HCR and conventional CABG in mortality, myocardial infarction (MI), repeat revascularization, or stroke. Several considerations deserve closer scrutiny.
Considering the equivalent patency of the LIMA-LAD graft at 1-year angiographic follow-up (94% with HCR vs. 93% with CABG), comparable 5-year mortality (6.4% with HCR vs. 9.2% with CABG; p = 0.69) is perhaps not surprising. After all, the LIMA-to-LAD graft contributes much of the survival advantage provided by CABG, protecting roughly half of the left ventricular mass. The preserved patency rate of the LIMA-LAD graft in the HCR group is particularly reassuring, especially given the technical challenges with a minimally invasive approach. The largest prior observational study evaluating HCR, which included 306 patients undergoing HCR and 918 propensity score–matched patients undergoing conventional CABG, noted equivalent 3-year mortality (8.8% with HCR vs. 10.2% with CABG, p = 0.72) (4). Although the POL-MIDES trial was not powered to draw a definitive conclusion regarding mortality, these lines of evidence collectively support long-term comparable efficacy of HCR.
The low restenosis and stent thrombosis rates of contemporary drug-eluting stents coupled with a less invasive procedure have been proposed as 1 of the primary factors favoring HCR. By comparison, 1 of the weaknesses of conventional CABG has been the suboptimal long-term patency of saphenous vein grafts (5), which can have 1-year failure rates exceeding 20%, whereas later-generation drug-eluting stents have 1-year restenosis rates <5%. Even so, in the POL-MIDES trial, no significant difference was found in the rate of in-hospital perioperative MI (5.1% with HCR vs. 3.9% with CABG; p = 0.69) or of spontaneous MI at 5-year follow-up (4.3% with HCR vs. 7.2% with CABG; p = 0.30) (2). In fact, zero additional spontaneous MIs or repeat revascularizations were noted in the CABG group within the first year, despite a 21% rate of non-LAD graft occlusion observed at the time of 1-year angiography (the reported 3.9% MIs were all perioperative). Graft occlusion may well remain clinically silent if adequate native coronary flow is maintained, whereas acute stent occlusion will usually result in an MI. So a simple comparison between graft patency and stent failure rate may be misleading vis-à-vis patient outcomes. It is generally accepted that ischemia-driven revascularization occurs more frequently among PCI-treated patients than among CABG-treated patients (6).
Another important consideration is the issue of incomplete revascularization, and the increased risk for subsequent revascularization, which are among the main shortcomings of any PCI-based strategy. Incomplete revascularization is associated with unfavorable long-term outcomes (7). In prior landmark randomized trials comparing PCI and CABG, the rates of complete revascularization in the PCI-treated cohort were suboptimal (50% to 60% range) (6,8). In contrast, both cohorts in the POL-MIDES trial achieved very high and comparable rates of complete revascularization (78.6% with HCR vs. 78.4% with CABG; p = 0.84) (2). Despite such, the rates of repeat revascularization at 5-year follow-up were perplexingly high in both groups (37.2% with HCR vs. 45.4% with CABG; p = 0.38). The individual clinical indications for these procedures were not provided in the report; however, the large majority seem to have occurred after 1 year. It is likely that the planned follow-up coronary angiography at 1 year, which was performed in >80% of the patients in both groups, and the resultant “oculostenotic” reflex contributed to this high rate of revascularization. A prior meta-analysis of 6 observational studies comparing HCR with conventional CABG suggested a higher rate of repeat revascularization in the HCR group (9).
Reducing stroke risk is paramount in patients undergoing revascularization. Because minimally invasive direct coronary artery bypass in hybrid strategy is typically performed without manipulation of the aorta, embolic stroke risk reduction may be the greatest appeal of HCR. Completely avoiding aortic manipulation by using in situ internal thoracic arteries for inflow (“no-touch aorta”) has been shown to be associated with a lower stroke risk in comparison with conventional CABG (10). In the POL-MIDES trial, in which off-pump surgery accounted for 85% of conventional CABG, the stroke incidence at 5-year was numerically lower in the HCR group than the conventional CABG group (2.1% vs. 4.1%; p = 0.35). However, all stroke events were late events (>1 year) and thus not attributed to surgical techniques.
Finally, is a minimally invasive surgical approach truly “minimally” invasive? In the POL-MIDES study, minimally invasive direct CABG was performed via a small (6 to 8 cm) anterolateral thoracotomy (11). Interestingly, there was no difference between the HCR and CABG cohorts regarding the total amount of pericardial or chest tube drainage, the rate of blood transfusions, or the overall length of stay (2). Although several studies noted a lower in-hospital need for blood transfusion and shorter length of stay for patients undergoing HCR, these differences were seemingly minor (9). So in truth, the appealing, less invasive promise of HCR also has yet to be realized.
So how should we incorporate the findings of the present report into clinical practice? Is there a sufficient argument for HCR to be the revascularization strategy of choice in certain patients? Although these 5-year data are very welcome, the overall published HCR experience remains small and mostly hypothesis generating. To be fair, all available data suggest that HCR is feasible and safe in skilled hands, with 1- and 5-year outcomes noninferior to conventional CABG in low- to intermediate-risk patients with multivessel CAD. But the use of HCR remains low, in part because of a number of clinical unknowns but also because of practical limitations: logistics (simultaneous in hybrid operation room vs. 2-stage); specific expertise, cooperation, and collaboration of surgical and interventional teams; risk for perioperative bleeding with the use of aggressive antiplatelet and antithrombotic therapy; insecurity about left main cohort outcomes; and uncertain cost-effectiveness (12). Until HCR can prove superiority to conventional CABG in clinical outcomes such as significant morbidity or stroke reduction, adoption of this strategy will likely continue to be modest.
No randomized trial comparing HCR with multivessel PCI has been completed, but there is an ongoing, ∼2,400-patient, multicenter, randomized trial funded by the National Heart, Lung, and Blood Institute (NCT03089398). Pending these results, HCR may yet become a more commonly used third interventional strategy for certain patients with multivessel CAD.
↵∗ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Gąsior M.,
- Zembala M.O.,
- Tajstra M.,
- et al.
- Tajstra M.,
- Hrapkowicz T.,
- Hawranek M.,
- et al.,
- on behalf of the POL-MIDES Study Investigators
- Harskamp R.E.,
- Vassiliades T.A.,
- Mehta R.H.,
- et al.
- Goldman S.,
- Zadina K.,
- Moritz T.,
- et al.
- Mohr F.W.,
- Morice M.-C.,
- Kappetein A.P.,
- et al.
- Farooq V.,
- Serruys P.W.,
- Bourantas C.V.,
- et al.
- Panoulas V.F.,
- Colombo A.,
- Margonato A.,
- Maisano F.