Real-World LVAD Use Promising in Patients Otherwise Considered Too Sick for PCI


Elderly patients with comorbid conditions and extensive coronary disease previously thought to put them at extreme risk for PCI can achieve favorable outcomes with low rates of adverse events when a left ventricular assist device (LVAD) is employed during revascularization, according to results published online August 15, 2015 ahead of print in the American Heart Journal.

Take Home: Real-World LVAD Use Promising in Patients Otherwise Considered Too Sick for PCI

“Patients with the highest probability of having severe CAD (3-vessel disease and/or left main disease), such as those with more advanced age, renal dysfunction, and [congestive heart failure], are the least likely to undergo coronary angiography and revascularization,” say Mauricio G. Cohen, MD, of University of Miami Hospital (Miami, FL), and colleagues. Yet these patients also stand to benefit the most from revascularization, they note.

The researchers looked at data from 637 patients at high risk for PCI who were treated at 47 US and 2 Canadian sites between June 2007 and September 2013. All underwent PCI assisted by the Impella 2.5 LVAD device (Abiomed; Danvers, MA) while enrolled in the USpella registry. Outcomes from this cohort were compared with those of 216 patients treated in the LVAD arm of the PROTECT II trial, which randomized patients with complex 3-vessel or unprotected left main disease and severely impaired LV function to PCI with an LVAD or intra-aortic balloon pump.

Among registry patients, 53.2% would have met the enrollment criteria for PROTECT II. Most had 2-vessel disease, including 16% with left main disease.

Compared with trial participants, registry patients were older and more likely to have chronic kidney disease, prior MI, and prior CABG. However, the groups had similar rates of diabetes, peripheral vascular disease, and prior stroke, as well as STS predicted risk of mortality. Trial patients had a greater number of treated lesions and stents likely due to a specific requirement in PROTECT II to perform the most complete revascularization possible in a single procedure. Approximately one-fifth of trial and registry patients underwent rotational atherectomy, but the average number of passes per lesion was lower in the registry.

Trend Toward Lower Mortality in Registry Patients

Assessment of NYHA class showed a 42.2% reduction from baseline to discharge in class III-IV symptoms in registry patients with available data (P < .0001) and a 27.8% reduction in PROTECT II trial participants (P = .008). Left ventricular function also increased substantially from baseline to discharge in the registry group (21.4% to 28.4%; P < .0001).

While most major in-hospital outcomes were similar between PROTECT II patients and registry patients, the latter group had lower rates of MI and repeat revascularization and a trend toward higher survival (table 1). There were no cases of stroke or TIA, emergency CABG, acute aortic regurgitation, or valve injury.

 Table  1. In-Hospital Outcomes

Importantly, Dr. Cohen and colleagues also point out that blood transfusions decreased over time in the registry, from 12.2% in 2009 to 6.1% in 2011, “possibly reflecting a learning curve effect.” While they attribute the lower rates of mortality and adverse events in the registry population to proficiency, the auhors say “inconsistent documentation in registry patients, in particular for those events defined by temporal changes in biomarker levels, such as periprocedural myocardial infarction,” cannot be ruled out.

Hope For Sickest of the Sick

In a telephone interview with TCTMD, Jeffrey W. Moses, MD, of Columbia University Medical Center (New York, NY), a PROTECT II investigator, agreed that the in-hospital death rate “bespeaks either to patient selection or the learning curve.” Additionally, Dr. Moses said the LVAD devices are evolving to become more powerful and effective “even in the face of real collapse.”

The fact that the registry patients had comparable outcomes despite being even more ill overall than the PROTECT II patients, who “were the sickest patients ever enrolled in an interventional cardiology trial,” provides reassurance that revascularization in this population is clearly beneficial, he commented.

“It shows that it is feasible to treat these patients, and in a way that has meaningful effects on their quality of life,” Dr. Moses observed. “This is a technology that we are only just now getting our arms around, and it gives us an opportunity to treat these very sick patients… who are frequently just not treated. It’s very hard to argue based on the trial and registry data that these patients are not benefiting from these procedures.”

Partially on the basis of the strength of the USpella Registry data, Abiomed announced this week in a press release that it is now seeking supplemental premarket approval for 2 other LVADs—Impella 5.0/LD and Impella CP. The Impella 2.5 device used in the registry and PROTECT II received premarket approval for high-risk PCI in March 2015.


Cohen MG, Matthews R, Maini B, et al. Percutaneous left ventricular assist device for high risk percutaneous coronary interventions: real world versus clinical trial experience. Am Heart J. 2015;Epub ahead of print. 


  • Dr. Cohen reports receiving speaker honoraria from and serving as a consultant to Abiomed. 
  • Dr. Moses reports serving as a consultant to Abbott and Boston Scientific. 

Related Stories:


We Recommend