The longest follow-up to date of patients with left main coronary artery disease (LMCAD) has found no significant differences in rates of death, myocardial infarction or stroke between patients who underwent percutaneous coronary intervention (PCI) and those who had coronary artery bypass graft (CABG) surgery.
The PRECOMBAT study was unveiled at the American College of Cardiology’s Virtual Scientific Session Together with World Congress of Cardiology (ACC.20/WCC Virtual) by Duk-Woo Park (Asan Medical Center, Seoul, South Korea). It also showed that more patients who had PCI required a repeat procedure than those who underwent surgery. The findings were simultaneously published online in Circulation.
Presenting online, Park told the audience: “The data are still limited on very long-term, especially beyond five years, outcomes of PCI compared to CABG in patients with left main coronary artery disease. Available long-term studies show conflicting results. There remains uncertainty about long-term outcomes, warranting this longer term follow up study.”
PRECOMBAT randomly assigned 600 patients with left main coronary artery disease to either PCI with a sirolimus drug-eluting stent (n=300) or to CABG (n=300) at 13 medical centres in South Korea between 2004 and 2009. The baseline clinical and angiographic characteristics were well balanced between the groups: mean age was 62.3±9.7 years, 76.5% were men, and 32% had medically treated diabetes.
Distal left main bifurcation disease was present in 64.6% of the patients, and the mean SYNTAX score was 24.8±10.3 (low in 42.4%, intermediate in 35.3%, and high in 22.3%).
The primary outcome was a major adverse cardiac or cerebrovascular event (MACCE; death from any cause, myocardial infarction, stroke, or ischaemia-driven target-vessel revascularisation [TVR]).
One-year and five-year results of PRECOMBAT, published in 2011 and 2015, found no significant differences between the two groups for the primary endpoint, or for any of its components, except that patients treated with PCI were more likely to need a repeat procedure.
Complete revascularisation was achieved in 68.3% in the PCI group and 70.3% in the CABG group. In the PCI group, intravascular ultrasound was used in 91.2% of the patients, and a mean of 2.7 stents were implanted per patient.
The median duration of follow-up in all patients was 11.3 years. Ten-year follow-up for all clinical endpoint events was achieved in 288 patients (96%) randomised to PCI, and 288 patients (96%) randomised to CABG. The primary endpoint of MACCE at 10 years occurred in 29.8% of the patients in the PCI group and 24.7% of the patients in the CABG group (hazard ratio with PCI vs. CABG 1.25, 95% CI 0.93–1.69). There was also no significant between-group difference with respect to the secondary composite outcome of death, myocardial infarction, or stroke (18.2 vs. 17.5%; hazard ratio 1, 95% CI 0.7–1.44) and death from any cause (14.5% PCI vs. 13.8% CABG, hazard ratio, 1.13, 95% CI 0.75–1.7) at 10 years.
The latter was described by Park as “one of our interesting findings”.
The incidence of myocardial infarction and stroke at 10 years did not significantly differ between the two groups. However, the 10-year incidences of ischaemia-driven target-vessel revascularisation and any revascularisation were higher after PCI than after CABG.
The 10-year rate of primary outcome between PCI and CABG were consistent across multiple subgroups, except for those stratified by the extent of concomitant coronary artery disease, in which the event rate was higher after PCI than after CABG in patients with left main and three-vessel disease.
There was no notable trend across the ordered SYNTAX score tertiles in the incidence of the primary outcome, composite of death, myocardial infarction, or stroke, and death from any cause. The rate of ischaemia-driven target vessel revascularisation was significantly higher after PCI than after CABG in the high SYNTAX score group.
Listing the limitations, Park said that due to the limited number of patients and the low event rate “our trial did not have sufficient statistical power to detect clinically relevant differences in endpoints; blinding of PCI versus CABG was not possible; some degree of ascertainment bias cannot be excluded.”
He added: “Because we evaluated a first-generation drug-eluting stent, our findings should be refuted or confirmed through longer term follow up trial involving contemporary drug-eluting stents like the EXCEL or NOBLE trials.”
Park described the study as hypothesis generating because it was underpowered.
In an accompanying comment published in Circulation, Fernando Alfonso (Hospital Universitario de La Princesa, Madrid, Spain) and Adnan Kastrati (Technische Universität, and Munich Heart Alliance, Munich, Germany) write that the study provides “useful insights”.
They say: “The 10-year results of PRECOMBAT further expand our knowledge, providing compelling evidence that clinical equipoise, defined as uncertainty regarding the relative benefits of competing revascularisation modalities, persists for patients with LMCAD and low-to-intermediate anatomic complexity. However, the occurrence of revascularisation failure at this critical site remains of concern considering its prognostic implications. The higher need for TVR in the PCI arm consistently seen in all LMCAD studies overshadows the long-term results of this strategy.”