(P194) PERCUTANEOUS CORONARY INTERVENTION VERSUS CORONARY ARTERY BYPASS GRAFTING FOR PATIENTS WITH LEFT VENTRICULAR SYSTOLIC DYSFUNCTION: A META-ANALYSIS OF KAPLAN-MEIER DERIVED INDIVIDUAL PATIENT DATA

Background: Coronary artery disease is a common etiology of ischemic left ventricular systolic dysfunction (LVSD), for which the optimal revascularization strategy remains unclear. We aimed to determine whether percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) results in greater survival advantage in patients with LVSD by performing a a) study-level and b) reconstructed individual patient data meta-analysis of Kaplan-Meier (KM) survival curves.

METHODS AND RESULTS: A systematic search of MEDLINE, EMBASE, and Cochrane was conducted for observational studies and randomized trials comparing PCI to CABG in patients with ischemic cardiomyopathy and left ventricular ejection fraction ≤40%. The primary outcome of interest was all-cause mortality at longest follow-up. The secondary outcomes were myocardial infarction (MI), stroke, repeat revascularization, cardiovascular mortality, and major adverse cardiovascular and cerebrovascular events (MACCE). An initial study-level meta-analysis was performed using reported hazard ratios (HR). KM curves were digitized to reconstruct individual patient data. Flexible parametric models with restricted cubic splines were constructed to obtain time-varying HRs due to non-proportional hazards between PCI and CABG. Thirteen studies (10 observational, 3 RCTs) were included with a total of 12,756 patients (PCI=6,229, CABG=6,527) followed for 4.55±1.05 years. Over the entire follow-up, PCI was associated with a higher rate of all-cause mortality (HR 1.35 [95% confidence interval: 1.13-1.62]), MI (HR 2.08 [1.53-2.82]), repeat revascularization (HR 2.45 [1.23-4.64]) and MACCE (HR 1.50 [1.19-1.88]) but not stroke or cardiovascular mortality compared to CABG (Table). Digitized curves from 6 observational studies (CABG=5,151, PCI=4,493) over 5.24±1.73 years were used to reconstruct individual patient data (Table). PCI resulted in an overall increased risk of all-cause mortality (HR 1.56 [1.32-1.85]) and instantaneous HRs calculated 0.1 years apart were statistically significant for the first 6.7 years (Figure). PCI was associated with a much higher rate of repeat revascularization (HR 5.59 [4.38-7.14]), for which instantaneous HRs were significant for 6.1 years. PCI was associated with an overall lower risk of stroke (HR 0.62 [0.40-0.99]) due to fewer events within the first 7 months, after which instantaneous HRs became non-significant.

Conclusion: In patients with LVSD, study-level data indicated PCI was associated with a higher rate of all-cause mortality, MI, repeat revascularization, and MACCE compared to CABG. An individual patient data meta-analysis confirmed a larger effect size for mortality and repeat revascularization. Clinically, PCI appears less beneficial than CABG in the long-term but is associated with almost 40% reduction in stroke hazard during the initial 7 months post-treatment.