Elective Intra-aortic Balloon Pump Placement in High-Risk Percutaneous Coronary Intervention

To the Editor: In their randomized controlled trial comparing elective intra-aortic balloon pump (IABP) counterpulsation vs no planned IABP support in high-risk percutaneous coronary intervention (PCI), Dr Perera and colleagues¹ concluded that elective IABP did not reduce the incidence of major adverse cardiac and cardiovascular events (MACCE). We agree with the study design, which used composite end points analyzed over a short period, because most patients enrolled in the Balloon Pump–Assisted Coronary Intervention Study (BCIS-1) had severely impaired left ventricular systolic function and ischemic heart failure, which conferred a high risk of death within 5 years that would probably blunt any beneficial effect of IABP intervention at long-term follow-up.^{2 - 3}

Although the authors reached conclusions about negative results in BCIS-1, we have 2 concerns that may contradict this. First, differences in the low incidence rates of cerebrovascular accidents and further revascularization in the primary end points and mortality in the secondary end points might have been statistically significant if the sample size were larger. Although overall associated benefits and hazards were in balance, the sample size effect on mortality should be emphasized because reduction of death rate during high-risk PCI is the major goal of prophylactic IABP insertion for the BCIS-1 cohort.

Second, a confidence bias for the procedure may have existed between these 2 groups, with the physician possibly spending more time dealing with higher-risk coronary lesions in the group undergoing prophylactic IABP insertion, including left main artery disease and left anterior descending coronary artery ostial stenosis, which could have resulted in increased risk of death.^{4 - 5} Accordingly, the authors should not only describe the severity scores of the coronary anatomy but also compare the procedure time and which target coronary vessels underwent PCI between groups.