Applicability of the EXCEL Trial Criteria to an All-Comers Real-World Cohort of Unprotected Left Main Percutaneous Coronary Intervention

Abstract

The Evaluation of XIENCE versus EXCEL (Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization) trial is the largest randomized study comparing percutaneous coronary intervention (PCI) with second-generation drug-eluting stents and bypass surgery in unprotected left main disease (ULMD). Our aim was to assess which proportion of patients from a contemporary all-comer population of ULMD PCI would be eligible for the EXCEL trial and whether these patients show different long-term outcomes than the rest of the ULMD population. A total of 246 consecutive patients underwent ULMD PCI between January 2018 and December 2021 and were retrospectively analyzed. After application of the EXCEL trial inclusion and exclusion criteria, 98 patients (39%) were allocated to the "EXCEL-like" cohort and 148 to the "non-EXCEL-like." Baseline characteristics of the 2 groups were substantially comparable. The mean baseline SYNTAX score was lower in the EXCEL-like patients than the non-EXCEL-like group (20.1 ± 7.1 vs 23.0 ± 11.2, p = 0.025), with the latter also showing a higher degree of incomplete revascularization (residual SYNTAX score >8 : 11% vs 26%, p = 0.005). At 1 year, the cumulative major adverse cardiovascular and cerebral events rate was 7% for the EXCEL-like and 17% for the non-EXCEL-like patients, respectively (p = 0.03). At the last follow-up available, the significant difference in major adverse cardiovascular and cerebral events rate persisted (19% vs 37%, p = 0.02) and a significant difference in any unplanned revascularization was also shown (1.5% vs 13%, p = 0.04). In conclusion, in a contemporary real-world cohort of ULMD PCI, only a minority of patients could be enrolled in the EXCEL trial; the EXCEL-like patients, in our population, showed less clinical events at long-term follow-up than the cohort of patients who were ineligible for EXCEL, despite comparable baseline cardiovascular profiles; these findings may challenge a broad external applicability of the results of the EXCEL trial in routine clinical activity.