Background: Left ventricular (LV) dysfunction was associated with adverse outcome after percutaneous coronary intervention (PCI) in the balloon-angioplasty and bare-metal stent era. Technological advances have reduced complications after PCI. The impact of left ventricular ejection fraction (LVEF) on outcomes in current clinical practice is unknown, with commonly used risk stratification models not consistently incorporating preprocedural LVEF.
Methods: A total of 2328 consecutive patients undergoing PCI in a single centre between April 2005 and July 2009 were analysed. Patients were eligible if LVEF had been categorized before PCI as good (LVEF ≥50%), moderate (LVEF 30-49%) or poor (LVEF <30%). Those in cardiogenic shock were excluded. Mortality data were tracked using the UK Office of National statistics database. Logistic regression analysis was used to predict the risk of mortality at 30-day and long-term follow-up.
Results: Overall all-cause mortality was 1.0% at 30 days and 5% at long-term follow-up. Kaplan-Meier analysis revealed an early divergence in survival curves according to LVEF. Mortality rates stratified by LVEF category were 0.4, 1.3 and 6.3% at 30 days and 3.3, 5.7 and 12.0% in the long term (2.2±1.1 years) (P<0.0001). Multiple regression analysis confirmed that impaired LVEF (≤50%) independently predicts 30-day [hazard ratio 4.20 (confidence interval 2.50-7.04), P=0.001] and long-term all-cause mortality [hazard ratio 1.67 (1.28-2.19), P=0.001].
Conclusion: LV impairment remains a strong predictor of early and late mortality after PCI. LV function assessment is integral in risk stratification and patient optimization and should be recommended, wherever feasible, before PCI.