A trend toward improved mortality at six months in patients with severe ischemic cardiomyopathy who received elective intra-aortic balloon pump (IABP) support during PCI appears to persist in a long-term analysis of the BCIS-1 trial. The five-year outcome results put in focus the need for longer-term follow-up for these devices, according the authors of an accompanying editorial.

The findings and editorial were published in the Jan. 15 issue of Circulation.

BCIS-1 (Balloon Pump–Assisted Coronary Intervention Study) enrolled 301 patients with left ventricular ejection fraction of less than 30 percent and severe coronary disease from 17 centers in the U.K. between December 2005 and January 2009 (JAMA 2010;304:867-874). Patients were randomized to either receive elective IABP support during PCI (151 patients) or no support during PCI (150 patients). The primary outcome was the occurrence of major adverse cardiac and cerebrovascular events (MACCE)—defined as death, acute MI, cerebrovascular event or urgent further revascularization—at hospital discharge, which was capped at 28 days. Secondary outcomes included all-cause mortality at six months.

The patient groups had similar baseline characteristics, comparable revascularization and similar procedural success. In the initial study, there was no difference in MACCE at hospital discharge but at six months there was a trend toward improved all-cause mortality. Patients in the IABP group also had fewer procedural complications, and 12 percent of patients in the group that was randomized to no support crossed over to receive IABP support.

Divaka Perera, MD, FRCP, of the cardiovascular division at St. Thomas' Hospital Campus, Kings College London, and colleagues designed their study to evaluate the long-term mortality in BCIS-1 by tracking outcomes through databases at the Office of National Statistics in England and Wales and the General Register Office in Scotland. Using those resources, they tracked mortality status for all 301 patients through October 2011.

With a median follow-up of 51 months, they calculated mortality rates of 7.9 per 100 patient-years in the IAPB group and 12.1 per 100 patient-years in the no support group.

“Fewer deaths were observed in the elective IABP group at six months,” they wrote, “and the present study demonstrates that this difference is maintained, with a 34 percent relative reduction in mortality observed in the elective IABP group at long-term follow-up.”

The high overall mortality rate—33 percent at follow-up—reflected the poor prognosis for this patient population. “In this context, a reduction in relative risk by one third would translate to a large, clinically significant treatment effect, although it remains to be established whether these findings will be borne out in larger clinical series.”

Perera and colleagues pointed out that BCIS-1 primarily was designed to track in-hospital MACCE, and did not prospectively look at all-cause mortality. Consequently, the results could reflect a chance finding, they warned. While the two databases allowed them to capture mortality data, they did not include cause of death.

John P. Vavalle, MD, and E. Magnus Ohman, MD, both of the Duke Clinical Research Institute in Raleigh, N.C., highlighted parallels between the long-term BCIS-1 results and a 90-day analysis of PROTECT II. PROTECT II randomized patients undergoing high-risk PCI to either IABP or the Impella 2.5 device (Abiomed). At 90 days, the study also showed a trend toward improved outcomes in the Impella group (Circulation 2012;126:1717-1727).

“On the surface, the primary endpoints of both BCIS-1 and PROTECT-II were neutral, suggesting that these devices did not offer any additional benefit when examined at 30 days. … However, both studies suggest a possible later effect,” they wrote. “The later outcomes data with IABP (at five years) and Impella (at 90 days) suggest that when more complete revascularization is desirable, these devices will offer the potential to improve outcomes with high-risk PCI.”

Follow-up using a death index could provide a method for assessing longer-term safety data for medical devices, as has been advocated by the FDA. This method could allow for prospective collecting of outcomes “without holding up an approval process. This would be of particular value for high-risk interventions where long-term mortality data are not usually available,” Vavalle and Ohman wrote.

Perera et al’s findings underscore the need to include longer-term follow-up in randomized trials for hemodynamic support devices, they concluded.